Washington, D.C. 20549


Form 10-K 


(Mark One)


For the fiscal year ended December 31, 2017



For the transition period from            to           .

Commission file number: 001-36740 



(Exact name of registrant as specified in its charter) 






(State or other jurisdiction of incorporation or organization)


(I.R.S. Employer Identification No.)

409 Illinois Street

San Francisco, CA



(Address of principal executive offices)


(zip code)


Registrant’s telephone number, including area code:

(415) 978-1200 

Securities registered pursuant to Section 12(b) of the Act: 


Title of Each Class


Name of Exchange on Which Registered

Common Stock, $0.01 par value


The NASDAQ Global Select Market


Securities registered pursuant to Section 12(g) of the Act:



Indicate by check mark if the registrant is a well-known seasoned issuer, as defined in Rule 405 of the Securities Act.    Yes      No  

Indicate by check mark if the registrant is not required to file reports pursuant to Section 13 or Section 15(d) of the Act.    Yes       No  

Indicate by check mark whether the registrant: (1) has filed all reports required to be filed by Section 13 or 15(d) of the Securities Exchange Act of 1934 during the preceding 12 months (or for such shorter period that the registrant was required to file such reports), and (2) has been subject to such filing requirements for the past 90 days.    Yes       No  

Indicate by check mark whether the registrant has submitted electronically and posted on its corporate Web site, if any, every Interactive Data File required to be submitted and posted pursuant to Rule 405 of Regulation S-T during the preceding 12 months (or for such shorter period that the registrant was required to submit and post such files).    Yes       No  

Indicate by check mark if disclosure of delinquent filers pursuant to Item 405 of Regulation S-K is not contained herein, and will not be contained, to the best of registrant’s knowledge, in definitive proxy or information statements incorporated by reference in Part III of this Form 10-K or any amendment to this Form 10-K.  

Indicate by check mark whether the registrant is a large accelerated filer, an accelerated filer, a non-accelerated filer, or a smaller reporting company. See the definitions of “large accelerated filer,” “accelerated filer” and “smaller reporting company” in Rule 12b-2 of the Exchange Act:


Large accelerated filer



Accelerated filer


Non-accelerated filer


  (Do not check if a smaller reporting company)


Smaller reporting company


Emerging growth company






If an emerging growth company, indicate by check mark if the registrant has elected not to use the extended transition period for complying with any new or revised financial accounting standards provided pursuant to Section 13(a) of the Exchange Act.         

Indicate by check mark whether the registrant is a shell company (as defined in Exchange Act Rule 12b-2).    Yes      No  

The aggregate market value of the voting and non-voting common equity held by non-affiliates of the registrant, computed by reference to the closing price as of the last business day of the registrant’s most recently completed second fiscal quarter, June 30, 2017, was approximately $1,674.4 million. Shares of Common Stock held by each executive officer and director and stockholders known by the registrant to own 10% or more of the outstanding stock based on public filings and other information known to the registrant have been excluded since such persons may be deemed affiliates. This determination of affiliate status is not necessarily a conclusive determination for other purposes.

The number of shares of common stock outstanding as of January 31, 2018 was 82,666,979.


Items 10, 11, 12, 13 and 14 of Part III of this Annual Report on Form 10-K incorporate information by reference from the definitive proxy statement for the registrant’s 2018 Annual Meeting of Stockholders to be filed with the Securities and Exchange Commission pursuant to Regulation 14A not later than after 120 days after the end of the fiscal year covered by this Annual Report on Form 10-K.
























Item 1.





Item 1A.


Risk Factors



Item 1B.


Unresolved Staff Comments



Item 2.





Item 3.


Legal Proceedings



Item 4.


Mine Safety Disclosures


















Item 5.


Market for Registrant’s Common Equity, Related Stockholder Matters and Issuer Purchases of Equity Securities



Item 6.


Selected Financial Data



Item 7.


Management’s Discussion and Analysis of Financial Condition and Results of Operations



Item 7A.


Quantitative and Qualitative Disclosure About Market Risk



Item 8.


Consolidated Financial Statements and Supplementary Data



Item 9.


Changes in and Disagreements with Accountants on Accounting and Financial Disclosures



Item 9A.


Controls and Procedures



Item 9B.


Other Information


















Item 10.


Directors, Executive Officers and Corporate Governance



Item 11.


Executive Compensation



Item 12.


Security Ownership of Certain Beneficial Owners and Management and Related Stockholder Matters



Item 13.


Certain Relationships and Related Transactions, and Director Independence



Item 14.


Principal Accounting Fees and Services


















Item 15.


Exhibits and Financial Statement Schedules













This Annual Report filed on Form 10-K and the information incorporated herein by reference, particularly in the sections captioned “Risk Factors,” “Management’s Discussion and Analysis of Financial Condition and Results of Operations” and “Business,” contains forward-looking statements, which involve substantial risks and uncertainties. In this Annual Report, all statements other than statements of historical or present facts contained in this Annual Report, including statements regarding our future financial condition, business strategy and plans and objectives of management for future operations, are forward-looking statements. In some cases, you can identify forward-looking statements by terminology such as “believe,” “will,” “may,” “estimate,” “continue,” “anticipate,” “contemplate,” “intend,” “target,” “project,” “should,” “plan,” “expect,” “predict,” “could,” “potentially” or the negative of these terms or other similar terms or expressions that concern our expectations, strategy, plans or intentions. Forward-looking statements appear in a number of places throughout this Annual Report and include statements regarding our intentions, beliefs, projections, outlook, analyses or current expectations concerning, among other things, our ongoing and planned preclinical development and clinical trials, the timing of and our ability to make regulatory filings and obtain and maintain regulatory approvals for roxadustat, pamrevlumab and our other product candidates, our intellectual property position, the potential safety, efficacy, reimbursement, convenience clinical and pharmaco-economic benefits of our product candidates, the potential markets for any of our product candidates, our ability to develop commercial functions, our ability to operate in China, expectations regarding clinical trial data, our results of operations, cash needs, spending of the proceeds from our initial public offering and the concurrent private placement, financial condition, liquidity, prospects, growth and strategies, the industry in which we operate and the trends that may affect the industry or us. We have based these forward-looking statements largely on our current expectations and projections about future events and financial trends that we believe may affect our financial condition, results of operations, business strategy and financial needs. These forward-looking statements are subject to a number of risks, uncertainties and assumptions described in the section of this Annual Report captioned “Risk Factors” and elsewhere in this Annual Report.

These risks are not exhaustive. Other sections of this Annual Report may include additional factors that could adversely impact our business and financial performance. Moreover, we operate in a very competitive and rapidly changing environment. New risk factors emerge from time to time, and it is not possible for our management to predict all risk factors nor can we assess the impact of all factors on our business or the extent to which any factor, or combination of factors, may cause actual results to differ materially from those contained in, or implied by, any forward-looking statements.

You should not rely upon forward-looking statements as predictions of future events. We cannot assure you that the events and circumstances reflected in the forward-looking statements will be achieved or occur. Although we believe that the expectations reflected in the forward-looking statements are reasonable, we cannot guarantee future results, levels of activity, performance or achievements. The forward-looking statements made in this Annual Report are based on circumstances as of the date on which the statements are made. Except as required by law, we undertake no obligation to update publicly any forward-looking statements for any reason after the date of this Annual Report or to conform these statements to actual results or to changes in our expectations.

This Annual Report also contains market data, research, industry forecasts and other similar information obtained from or based on industry reports and publications, including information concerning our industry, our business, and the potential markets for our product candidates, including data regarding the estimated size and patient populations of those and related markets, their projected growth rates and the incidence of certain medical conditions, as well as physician and patient practices within the related markets. Such data and information involve a number of assumptions and limitations, and you are cautioned not to give undue weight to such estimates.

You should read this Annual Report with the understanding that our actual future results, levels of activity, performance and achievements may be materially different from what we expect. We qualify all of our forward-looking statements by these cautionary statements.







We are a science-based biopharmaceutical company discovering and developing first-in-class therapeutics. Roxadustat (FG-4592), our most advanced product candidate, is an oral small molecule inhibitor of HIF prolyl hydroxylase (“HIF-PH”) activity in Phase 3 clinical development for the treatment of anemia in chronic kidney disease (“CKD”). Pamrevlumab (FG-3019), a fully-human monoclonal antibody that inhibits the activity of connective tissue growth factor (“CTGF”), is in Phase 2 clinical development for the treatment of idiopathic pulmonary fibrosis (“IPF”), pancreatic cancer, and Duchenne muscular dystrophy (“DMD”). We have taken a global approach to the development and future commercialization of our product candidates, which includes development and commercialization in the People’s Republic of China (“China”).

We are capitalizing on our extensive experience in fibrosis and hypoxia inducible factor (“HIF”) biology and clinical development to advance a pipeline of innovative medicines for the treatment of anemia, fibrotic disease, cancer, corneal blindness, and other serious unmet medical needs. The chart below is a summary of our most advanced product candidates.



Roxadustat is an internally discovered HIF-PH inhibitor that acts by stimulating the body’s natural pathway of erythropoiesis, or red blood cell production. Roxadustat, the first HIF-PH inhibitor to enter Phase 3 clinical development, represents a new paradigm for the treatment of anemia with the potential to offer a safer, more effective, more convenient, and more accessible therapy than the current therapies available for anemia in CKD, such as injectable erythropoiesis stimulating agents (“ESAs”).

Roxadustat is currently in Phase 3 development and we and our collaboration partners have enrolled more than 8,000 patients in our Phase 3 global program for the treatment of anemia in patients with CKD. More than 1,400 subjects have participated in 26 completed Phase 1 and Phase 2 clinical studies for roxadustat in North America, Europe, and Asia. These studies have demonstrated roxadustat’s potential for a favorable safety and efficacy profile in anemic CKD patients for those who are dialysis-dependent (“DD-CKD’), including difficult-to-treat patients who are hyporesponsive to ESAs, and those who are not dialysis-dependent (“NDD-CKD”). According to IMS Health, 2013 global ESA sales in all anemia indications totaled $8.6 billion. While the use of ESAs to treat anemia in CKD has largely been limited to the DD-CKD population, we and our partners believe that, as an oral agent with a potentially more favorable safety profile, roxadustat could increase accessibility and expand the market for anemia treatment by penetrating the NDD-CKD market. In the long term, we believe roxadustat has the potential to address non-CKD anemia markets, including chemotherapy-induced anemia, anemia related to inflammation (such as inflammatory bowel disease, lupus, and rheumatoid arthritis), myelodysplastic syndromes (“MDS”), and surgical procedures requiring red blood cell transfusions.

We, along with our collaboration partners Astellas Pharma Inc. (“Astellas”) and AstraZeneca AB (“AstraZeneca”), have designed a major global Phase 3 program to support regulatory approval of roxadustat for both NDD-CKD and DD-CKD patients in the United States (“U.S.”), the European Union (“EU”), Japan, and China. Our U.S. and EU Phase 3 program has an aggregate target enrollment of more than 8,000 patients worldwide and is the largest Phase 3 clinical program ever conducted for an anemia product candidate. In addition, we have completed a 456-patient Phase 3 program in China, and our partner, Astellas, has completed three of its six Phase 3 trials in Japan, studying approximately 1,000 additional patients. Our U.S. Phase 3 program is designed to incorporate major adverse cardiac event (“MACE”) composite safety endpoints that we believe will be required for approval in the U.S. for all new anemia therapies. These Phase 3 programs are evaluating the use of roxadustat in multiple populations, including patients within the first four months of initiating dialysis, or incident dialysis, and non-incident, or stable dialysis patients and include numerous placebo-controlled NDD-CKD studies of roxadustat.

Background of Anemia in CKD

Anemia is a serious medical condition in which patients have insufficient red blood cells and low levels of hemoglobin (“Hb”), a protein in red blood cells that carries oxygen to cells throughout the body. Anemia is associated with increased risk of hospitalization, cardiovascular complications, need for blood transfusion, exacerbation of other serious medical conditions, and death. In addition, anemia frequently causes significant fatigue, cognitive dysfunction, and decreased quality of life. The more severe the anemia, as measured by lower Hb levels, the greater the health impact on patients. Severe anemia is common in patients with CKD, cancer, MDS, inflammatory diseases, and other serious illnesses. Even when it accompanies prevalent and serious diseases, anemia is often not effectively treated.

Anemia is particularly prevalent in patients with CKD, which is a critical healthcare problem in the U.S. and Europe, and most commonly caused by diabetes and hypertension. CKD affects more than 200 million people worldwide and significantly increases healthcare costs for those patients. CKD is generally a progressive disease characterized by gradual loss of kidney function that may eventually lead to kidney failure, which is also known as end-stage renal disease (“ESRD”). Patients with ESRD require renal replacement therapy — either dialysis treatment or kidney transplantation. CKD accompanied by anemia is associated with worse health outcomes than CKD alone, including more rapid disease progression and an increased death rate. There are five stages of CKD that are primarily defined by a measure of the filtration function of the kidney (GFR).


Stages of CKD and Prevalence in the U.S.



U.S. prevalence is estimated for adults 20 years of age or older

GFR: Glomerular Filtration Rate (ml/min/1.73 m2)

Sources: The prevalence of stage 1 through stage 4 CKD was calculated based on 2014 estimates by the United States Renal Data System (“USRDS”) presented in the 2016 USRDS annual data report: Epidemiology of kidney disease in the United States (“2016 USRDS ADR”), using data from the National Health and Nutrition Examination Survey (“NHANES”) 2011-2014 and 2014 data from the U.S. Census Bureau. The prevalence of stage 5 CKD was calculated based on 2014 data from the 2016 USRDS ADR using data from the U.S. National ESRD database, NHANES 2011-2014 and 2014 data from the U.S. Census Bureau.

The prevalence rate of anemia in patients with Hb<12 g/dL is set forth below.

Sources: The prevalence of anemia in stage 1 through stage 4 CKD and stage 5 NDD-CKD was derived from Stauffer and Fan, Prevalence of Anemia in Chronic Kidney Disease in the United States, PLoS ONE (2014). The prevalence of anemia in patients undergoing dialysis was derived from Goodkin et al., Naturally Occurring Higher Hemoglobin Concentration Does Not Increase Mortality among Hemodialysis Patients, J Am Soc Nephrol (2011).

In the U.S., according to the USRDS, a majority of dialysis eligible CKD patients are currently on dialysis. Of the approximately 475,000 patients receiving dialysis in the U.S., approximately 83% were being treated with ESAs for anemia. Despite the presence of anemia in stages 3 and 4 CKD patients, in clinical practice, patients typically do not receive ESA treatment for their anemia until they initiate dialysis. As of 2014, approximately 14% of U.S. NDD-CKD patients were being treated with ESAs prior to initiation of dialysis (2016 USRDS ADR). In many CKD patients, the disease progresses gradually over decades and patients can spend years suffering from the symptoms and negative health effects of anemia before they receive treatment. Many of these patients die from cardiovascular events before they initiate dialysis.


Limitations of the Current Standard of Care for Anemia in CKD

Current therapies to treat anemia in CKD include injectable ESAs, intravenous (“IV”) iron, oral iron, and blood transfusions. ESAs have been used in the treatment of anemia in CKD for more than 20 years and are administered intravenously or subcutaneously, typically in conjunction with IV iron. NDD-CKD patients who are not under the care of nephrologists, including those with diabetes and hypertension, do not typically receive ESAs and are often left untreated. ESAs currently on the market are all synthetic recombinant versions of human erythropoietin (“EPO”), a hormone that stimulates erythropoiesis and increases Hb levels by binding to receptors on red blood cell precursors in the bone marrow.

The market introduction of the first ESA in 1989 was viewed as a major advance in the treatment of anemia in CKD because it significantly decreased the need for blood transfusions. Since then, ESAs have become one of the most commercially successful drug classes. Because ESAs were never studied relative to placebo in large randomized clinical trials prior to approval, safety issues were elucidated post commercialization. In particular, studies published during 2006 to 2009 demonstrated the safety risks of higher ESA doses used to reach target Hb levels of 13 to 15 g/dL, prompting physicians to balance serious safety concerns against the efficacy of ESAs. The safety concerns observed with injectable ESAs in these studies included an increased risk of cardiovascular adverse events and death, as well as a potentially increased rate of tumor recurrence in patients with cancer.

These safety issues resulted in several changes to ESA drug labeling. The combination of safety concerns and labeling changes, in addition to the subsequent reimbursement limitations described below, led to a steady decline in ESA sales revenues, beginning in 2007. While we believe the decline in ESA sales is primarily due to complete suspension of the label for use in anemias associated with cancer, and restrictions on use in chemotherapy-induced anemia, we also believe the decline in sales is partly due to the progressive decline in ESA dose administered to CKD patients. Compared to the average ESA dose at the end of 2006, the mean monthly ESA dose in patients on hemodialysis (“HD”) dropped by 18%, 36%, 45%, and 45% by the end of 2010, 2011, 2012, and 2013 respectively (2015 USRDS ADR).

Safety Issues of ESAs

Several large clinical trials were designed to demonstrate that targeting higher as opposed to lower Hb levels results in better outcomes. Instead these studies generated data showing that targeting higher Hb levels with ESAs resulted in an increase in adverse events, including cardiovascular adverse events. These adverse events were initially observed in 1998 in the NHCT (Normal Hematocrit Cardiac Trial) in CKD patients on dialysis, where the high Hb level treatment arm targeted Hb levels of 13 to 15 g/dL. Additional safety concerns emerged following the CHOIR (Correction of Hemoglobin in Outcomes and Renal Insufficiency), CREATE (Cardiovascular Risk Reduction by Early Anemia Treatment with Epoetin Beta), and TREAT (Trial to Reduce Cardiovascular Events with Aranesp Therapy) studies in NDD-CKD patients, that were published between 2006 and 2009.

Secondary analyses of NHCT, CHOIR and TREAT, as well as subsequent observational studies in dialysis patients, suggest that these safety concerns, particularly the increased cardiovascular risk associated with ESAs, may result from the high ESA doses used to target higher Hb levels rather than the Hb levels achieved. For example, a secondary analysis of CHOIR showed that patients who achieved the desired Hb level with the lowest amounts of ESA had the lowest risk of adverse cardiovascular outcomes, as measured by composite endpoints consisting of hospitalization for heart failure, heart attack, stroke, and death. Patients who were treated with the highest ESA doses and, particularly those who achieved the lowest Hb levels, had the greatest risk for these events. In addition, observational studies in patients undergoing dialysis highlighted these risks with high ESA doses and also indicated that higher Hb levels achieved with lower ESA doses were associated with better outcomes.


For example, in an analysis of data from the USRDS of 94,569 HD patients, increased mortality was found in patients with increased epoetin alfa dose. Patients who achieved the highest hematocrit level (which is a measure of the percentage of volume of whole blood made up of red blood cells; under typical conditions, Hb level can be estimated as one-third of the hematocrit level) and received the lowest ESA doses (lowest dose quartile, Q1) had the lowest mortality rate, and, at any particular ESA dose quartile, patients with higher hematocrit levels tended to have lower mortality levels, according to Zhang et al. (Am J Kidney Dis 44:866-876) as illustrated in the chart below.

Unadjusted One-Year Mortality Rates (per 1000)

by Hematocrit and ESA dosing quartile

Warnings about these risks have been incorporated into treatment guidelines and position papers from major kidney societies and thought leaders in the industry. Kidney Disease: Improving Global Outcomes (“KDIGO”), a non-profit foundation established in 2003 and operated by the National Kidney Foundation, committed to improving global clinical guidelines for kidney patients, for example, states that, “[t]here may be toxicity from high doses of ESA, as suggested, though not proven, by recent post-hoc analyses of major ESA randomized controlled trials, especially in conjunction with the achievement of high Hb levels. Therefore, in general ESA dose escalation should be avoided.” In addition, the European Renal Best Practices Group specified in a recent position statement that caution should be used in ESA therapy in patients with specific risk factors.

Limited Effectiveness of ESAs in Certain Patient Populations

Hb responses to ESA doses are on a continuum with some patients responding with a satisfactory Hb increase to a small ESA dose and others responding very poorly to very high doses. In addition, patients’ responsiveness to ESAs can change over time and as a result of circumstances such as acute illness or surgery. In an attempt to reach target Hb level, ESA doses are increased in treatment-resistant patients (hyporesponders), which can result in up to a 40-fold difference in ESA doses between the most ESA-resistant and the most ESA-responsive DD-CKD patients. Even with high doses of ESAs and concomitant IV iron, some of these hyporesponders are unable to reach target Hb levels.

Hyporesponsiveness is a significant problem in incident dialysis patients, for whom ESA doses are typically high, and is associated with a combination of critically low kidney function and accompanying illnesses, such as infections and chronic inflammation. Incident dialysis patients are generally more anemic, and have a higher risk of death, than patients who have been on dialysis for many months.


A major cause of ESA hyporesponsiveness is an underlying chronic inflammatory state that exists in many CKD patients. Chronic inflammation has a suppressive effect on erythropoiesis in CKD via two main mechanisms. Firstly, pro-inflammatory cytokines such as tumor necrosis factor alpha (“TNF-alpha”) and interleukin-6 (“IL-6”) have been implicated in the suppression of erythropoiesis through inhibition of the response of erythroid progenitor cells to EPO. Secondly, pro-inflammatory cytokines such as IL-6 elevate the levels of hepcidin, the major hormone regulating iron metabolism. The consequence of elevated hepcidin levels is a reduction in iron absorption from the gastrointestinal tract (“GI tract”) and the trapping of iron in cellular stores. Together, this leads to inadequate availability of iron to keep pace with the demands of the bone marrow for erythropoiesis, despite adequate total body iron stores. This condition is referred to as functional iron deficiency.

In the presence of inflammation, even high doses of ESAs may be ineffective to achieve target Hb levels, and to the extent Hb levels are raised, the risks associated with the higher ESA doses required may outweigh the benefits.

Requirement for IV Iron to Support ESA Activity and Associated Safety Risks

IV iron supplementation is used to support anemia correction in a majority of HD patients treated with ESAs in the U.S. ESA labeling indicates that physicians should evaluate the iron status in all patients before and during CKD anemia treatment and maintain iron repletion. However, many CKD patients have deficient iron stores, or absolute iron deficiency, and cannot absorb enough iron from diet or oral iron supplements to correct this deficiency. Physicians administer IV iron to ensure these patients are iron replete prior to initiating ESA treatment and continue to receive IV iron to mitigate iron depletion caused by ESA-mediated erythropoiesis.

Additionally, many CKD patients who have adequate iron stores suffer from functional iron deficiency. IV iron is administered to these patients, in an attempt to address this shortage of available iron, resulting in many patients having elevated body iron stores. While IV iron can help correct anemia when used in combination with ESAs, published studies have suggested both acute and chronic risk of morbidity and mortality associated with the use of IV iron. The acute risks of IV iron supplementation include infection and hypersensitivity reactions, which can be life-threatening. The warning of anaphylaxis risk appears in every IV iron product package insert in the U.S. Less severe but more common side effects include skin problems, hypotension, and GI tract symptoms. In addition to these acute side effects, there may be chronic adverse effects resulting from the volume of iron administered and associated cumulative deposits of iron in organ systems.

Increased use of IV iron has been associated with increased risk of hospitalization and death. Using data from 12 countries obtained over the past 12 years, Bailie et al. demonstrated a direct dose risk relationship between the amount of IV iron administered monthly to dialysis patients and the risk of hospitalization and death (Kidney International (2014)). The study identified that, even after controlling for other risk factors and adjusting for different practice patterns globally, dialysis patients receiving greater than 300 mg of IV iron per month had a greater risk of hospitalization or death than those receiving less than 300 mg. Mortality was 13% greater among those receiving between 300 mg and 400 mg of IV iron per month and 18% greater among those receiving greater than 400 mg of IV iron per month. Furthermore, hospitalization risk was 12% greater among those who received greater than 300 mg per month. The current paradigm of administrating greater doses of IV iron to decrease ESA doses in light of this recently described associated risk underscores the significant unmet need in the treatment of anemia. However, new and purportedly safer and more effective iron supplementation therapies are being developed and introduced, and if such new therapies are accepted by patients and physicians as a superior alternative to traditional IV iron supplementation therapies, they may help maintain or increase the attractiveness of ESA therapy.

Elevated Blood Pressure

ESAs have long been associated with increased blood pressure, including new onset hypertension and exacerbation of pre-existing hypertension. As a result, ESA labeling carries a warning for the potential for increased blood pressure with ESA usage. Hypertension has been shown to accelerate CKD progression and significantly increase the risk of death in CKD patients due to the increased risk of heart attack or stroke.

Increased Thromboembolism and Vascular Access Thrombosis

ESA use has been associated with thromboembolic events, including stroke, vascular access thrombosis (where the dialysis access shunt is blocked due to blood-clotting), and blood clots in the leg, which may in part be due to increases in circulating platelet levels. ESA labels now carry a warning for an increased risk of thromboembolic events.


FDA Restrictions on ESA Usage

In response to safety concerns elucidated in the large clinical studies described above, the U.S. Food and Drug Administration (“FDA”), steadily increased restrictions on the use of injectable ESAs from 2007 through 2011. During 2007, following the NHCT, CHOIR, CREATE, and several oncology studies, the FDA mandated the inclusion of a boxed warning, or “Black Box” warning, in the package insert for ESAs. A Black Box warning is the strongest warning that the FDA can mandate for prescription drugs. Further, in June 2011, the FDA required additional modification to the package insert for ESA use. The current label carries a black box warning of increased risk of death, myocardial infarction, or heart attack, stroke, venous thromboembolism, thrombosis of vascular access, and tumor progression or recurrence. In addition, the current package insert recommends more conservative dosing guidelines for the use of injectable ESAs in anemic CKD patients. More specifically, the FDA removed the prior target Hb range of 10 g/dL to 12 g/dL and recommends that physicians initiate treatment of CKD patients when the Hb level is less than 10 g/dL, and reduce or interrupt ESA dosing if the Hb level approaches or exceeds 10 g/dL for NDD-CKD patients and 11 g/dL for DD-CKD patients. In addition, physicians are advised to use only the lowest dose needed to avoid red blood cell transfusions.

Reimbursement Challenges Associated with ESAs

In addition to the safety concerns and labeling changes for ESAs, dialysis reimbursement, including associated drugs such as ESAs, has also changed significantly in recent years, which has made ESAs less economically attractive for healthcare providers to administer. Prior to January 2011, the Centers for Medicare and Medicaid Services (“CMS”) reimbursed dialysis centers and other healthcare providers for use of ESAs at average selling price plus a premium to their cost, which enabled providers to realize a profit on the administration of ESAs, regardless of the quantity dosed. Under the Medicare Improvements for Patients and Providers Act (“MIPPA”), a basic case-mix adjusted composite, or bundled, payment system commenced in January 2011 and transitioned fully by January 2014 to a single reimbursement rate for drugs and all services furnished by renal dialysis centers for Medicare beneficiaries with end-stage renal disease. Specifically, under MIPPA the bundle now covers drugs, services, lab tests, and supplies under a single treatment base rate for reimbursement by CMS based on the average cost per treatment, including the cost of ESAs and IV iron doses, typically without dose adjustment.

ESAs administered to NDD-CKD patients have long been reimbursed under Medicare Part B, which requires providers to purchase and store ESAs in advance of being reimbursed, and in many healthcare practices, the amount reimbursed does not cover the cost of ESA administration. For many of these providers, including in nephrology practices where purchase and storing is most common, due to label changes and related reduction in patients available for treatment, ESA administration in NDD-CKD has become economically unattractive. Furthermore, non-nephrologists generally have elected not to provide ESAs. Accordingly, ESA treatment is limited outside of dialysis centers.

Inconvenience of ESAs

In addition to safety, labeling, reimbursement, and efficacy limitations, ESAs must be administered intravenously or subcutaneously, often with IV iron in order for ESAs to be effective at treating to target Hb levels. ESAs are therefore inconvenient for the NDD-CKD population, the peritoneal dialysis (“PD”) population, for whom treatment is often administered at home, and other non-CKD anemia patients who are not already regularly visiting a hospital or dialysis center.

Our Solution

We believe that there is a significant need for a safer, more effective, more convenient, and more accessible alternative to injectable ESAs for the treatment of anemia in CKD patients. In addition, we believe there is a significant opportunity for treatment of anemia in markets not effectively addressed by ESAs, such as in the NDD-CKD population, DD-CKD in the presence of inflammation, and non-CKD anemia markets.


Roxadustat — A Novel, Orally Administered Treatment for Anemia

Roxadustat is an orally administered small molecule that corrects anemia by a different mechanism of action from that of ESAs. As a HIF-PH inhibitor, roxadustat activates a response that is naturally activated when the body responds to reduced oxygen levels in the blood, such as when a person adapts to high altitude. The response activated by roxadustat involves the regulation of multiple, complementary processes to promote erythropoiesis and increase the blood’s oxygen carrying capacity.

Coordinated erythropoiesis includes both the stimulation of red blood cell progenitors, by increasing the body’s production of EPO, and an increase in iron availability for Hb synthesis. Patients taking roxadustat typically have circulating endogenous EPO levels at peak concentration within or near the physiologic range naturally experienced by people adapting to hypoxic conditions such as at high altitude, following blood donation or impaired lung function, such as pulmonary edema. By contrast, ESAs act only to stimulate red blood cell progenitors without a corresponding increase in iron availability, and are typically dosed at well above the natural physiologic range of EPO. The sudden demand for iron stimulated by ESA-induced erythropoiesis can lead to functional or absolute iron deficiency. We believe these high doses of ESAs are a main cause of the significant safety issues that have been attributed to this class of drugs. In contrast, the differentiated mechanism of action of roxadustat, which involves induction of the body’s own natural pathways to achieve a more complete erythropoiesis, has the potential to provide a safer and more effective treatment of anemia, including in the presence of inflammation, which normally limits iron availability.


Our HIF-PH inhibitor technology relies on the natural mechanism by which the body responds to low oxygen levels. HIF is a transcription factor comprised of a HIF-alpha and a HIF-beta subunit, both of which are required to stimulate erythropoiesis. Under normal oxygen conditions, the HIF-alpha subunit is targeted for rapid degradation through the activity of a family of HIF-PH enzymes. However, under low oxygen conditions, the HIF-PH enzymes cannot function and HIF-alpha accumulates. HIF-alpha then combines with HIF-beta, and the newly formed HIF complex initiates transcription of a number of genes involved in erythropoiesis, which ultimately leads to increased oxygen delivery to tissues. Roxadustat works by reversibly inhibiting the HIF-PH enzymes, thus mimicking this coordinated natural erythropoiesis through genes encoding the proteins shown below involved in iron absorption, mobilization and transport, as well as stimulation of red blood cell progenitors.


Our discovery and development of roxadustat resulted from years of experience working with prolyl hydroxylase enzymes, such as those that regulate HIF, and a deep understanding of the complexities of HIF biology. We have explored therapeutic activation of HIF to treat anemia from an integrated perspective with a focus on applying our HIF-PH inhibitor technology to produce coordinated effects on erythropoiesis, iron homeostasis, and metabolism. As part of these progressive efforts, we have explored the ability of our HIF-PH inhibitor technology to increase sensitivity to endogenous EPO by increasing EPO receptor expression on red blood cell progenitors. We have investigated multiple effects of HIF-PH inhibitors on iron metabolism, including their ability to regulate genes that can increase iron bioavailability. We have also shown that administration of HIF-PH inhibitors can decrease expression of hepcidin, the key hormone that regulates iron metabolism. Hepcidin is elevated under conditions of chronic inflammation, leading to reduced iron availability for erythropoiesis. Based on our gene expression and hepcidin data, we believe HIF-PH inhibitors can increase intestinal iron absorption and enhance the mobilization and uptake of iron. In addition, we have shown that HIF-PH inhibitors can improve transferrin saturation (“TSAT”), a measure of circulating iron available for erythropoiesis, and can correct anemia associated with chronic inflammation by overcoming the hepcidin-mediated sequestration of iron.

We selected roxadustat from our extensive library of compounds from various chemical classes of HIF-PH inhibitors, including heterocyclic carboxamides and 2-oxoglutarate mimetics. Roxadustat was selected based on our belief that stabilizing the two main forms of HIF in the cell, HIF-1 and HIF-2, leads to more complete erythropoiesis.

Although HIF-PH inhibitor programs have been subsequently initiated at several other companies, we expect to remain the leader in the development of HIF-PH inhibitors for anemia, with more patients dosed and more studies conducted with roxadustat than with any other HIF-PH inhibitor.

Potential Advantages of Roxadustat for Treatment of Anemia in CKD

We believe that roxadustat has the potential to offer several safety, efficacy, reimbursement, and convenience advantages over ESAs.

Potential Safety and Efficacy Advantages

Our clinical trials to date have shown that roxadustat can treat anemia in CKD with much lower circulating EPO levels than with treatment by ESAs, mitigate the need for IV iron and treat anemia in the presence of inflammation, thereby offering potential safety and efficacy benefits over ESAs. We have incorporated several endpoints into our Phase 3 studies to further elucidate and demonstrate these and other potential clinical benefits of roxadustat.

Potential Cardiovascular Benefits

The CKD patient population is at high risk for cardiovascular events such as heart attacks and strokes. One known side effect of ESAs is elevation of blood pressure, which is particularly dangerous in this high-risk patient population. In contrast, we did not observe increases in blood pressure in patients treated with roxadustat beyond the background levels observed for the comparable, placebo-treated patients in a NDD-CKD Phase 2 trial. However, these data should be cautiously assessed due to the limited number of patients exposed. In Study 041, the NDD-CKD patients treated with roxadustat three-times weekly for more than 12 weeks had a modest decrease in blood pressure in a subgroup analysis of our Phase 2 NDD-CKD study.

In our Phase 2 studies, we did not observe a safety signal for thromboembolic risk. In contrast to the platelet increase with ESA treatment, platelet counts reported in roxadustat-treated patients did not increase, as those with platelet levels in the top 25th percentile at baseline saw their platelet levels decrease towards normal levels while those with platelet levels in the lower 75th percentile at baseline saw their platelet levels remain stable. This finding supports our belief in a potential safety benefit over ESAs since the platelet increase with ESAs could be a contributing factor in the thromboembolic risk associated with ESAs.

In addition, in our Phase 2 clinical trials, we observed reductions in total cholesterol and an improvement in average high-density lipoprotein (“HDL”) / low-density lipoprotein (“LDL”) ratio. Since many CKD patients have high cholesterol levels, which contribute to cardiovascular-related morbidity and mortality, the improvement in the average HDL / LDL ratio observed with roxadustat treatment could confer a benefit to patients.

Based on our preclinical and clinical data generated to date, we believe roxadustat could offer cardiovascular benefits to a CKD patient population that typically has cardiovascular-related co-morbidities and is at a high risk for cardiovascular events.


Potential for Anemia Correction with Moderate EPO Levels

Randomized trials have suggested that high doses of ESAs administered in an attempt to achieve a target Hb level may cause the safety issues associated with ESA therapy. These high doses result in serum EPO levels much higher than physiological range. In contrast, the level of endogenous EPO elevation among patients treated with roxadustat is typically within or near the range observed when ascending to a higher elevation or giving blood. Treating anemia while maintaining lower circulating EPO levels may mitigate, or even avoid, the risks from ESA therapy, including cardiovascular events and death.

The following graph depicts:



the circulating endogenous EPO levels in natural physiologic adaptations, such as adjustment to high altitude, blood loss, or pulmonary edema [left, ];



transient peak endogenous EPO levels estimated for CKD patients who achieved a Hb response to therapeutic doses of roxadustat in our Phase 2 clinical studies [middle, ];



the estimated peak circulating recombinant EPO levels resulting from IV ESA doses in distributions reported by the Dialysis Outcomes and Practice Patterns Study (“DOPPS”), for the fourth quarter of 2011 in the U.S. (after bundling was initiated and when the Hb target in ESA labeling was in the range of 10-11 g/dL [right, ]).

1 Milledge & Cotes (1985) J Appl Physiol 59:360;2 Goldberg et al. (1993), Clin Biochem 26:183, Maeda et al. (1992), Int J Hematol 55:111; 3Kato et al. (1994) Ren Fail 16:645; 4The transient peak endogenous EPO concentrations (“Cmax”), data for roxadustat was derived from a subset of 243 patients who achieved a Hb response to roxadustat in our Phase 2 studies for whom we believe doses depicted approximated therapeutic doses. Hb target ranges for these patients were above the Hb levels specified in the current ESA package insert for CKD patients. Only doses in those patients whose Hb responded in Phase 2 studies are reflected in the figure. The subset of patients included 134 NDD-CKD patients treated to thrice-weekly, twice-weekly, or weekly doses of roxadustat for >16 weeks. The subset also included 109 DD-CKD patients, including incident dialysis patients whose anemia was corrected with therapeutic doses and stable dialysis patients who received maintenance doses. Cmax of endogenous EPO levels were not measured in all patients; instead the range of EPO Cmax levels were estimated based on data derived from a more limited number of patients in whom EPO levels were measured at various roxadustat doses and among whom there was substantial variation in measured EPO levels. Accordingly, individual patients who received roxadustat may have realized EPO Cmax levels significantly above or below these estimated levels. Moreover, the estimates reflected in the graph may not be reflective or predictive of actual EPO Cmax levels or ranges that will be realized in larger populations of patients receiving roxadustat in our Phase 3 clinical trials. 5EPO Cmax was computed from ESA dose distributions based on Flaherty et al. (1990) Clin Pharmacol Ther 47:557.


Potential for Anemia Correction for Patient Populations that are Hyporesponsive to ESAs

Incident dialysis patients and patients who have chronic inflammation are often hyporesponsive to ESAs, which necessitates the use of higher doses of ESAs to increase Hb levels, thus increasing both safety risk and treatment cost. In contrast, the dose of roxadustat may not need to be increased in incident dialysis patients or to overcome the suppressive effects of inflammation on erythropoiesis, which we believe may confer significant safety and efficacy benefits.

As a result of roxadustat’s different mechanism of action, the ability of roxadustat to stimulate erythropoiesis does not appear to be impaired by chronic inflammation.

Our preclinical studies indicate that roxadustat can overcome the direct suppressive effects of inflammatory cytokines on erythropoiesis. In these studies, we observed roxadustat’s ability to reduce hepcidin expression, thus increasing absorption of iron from the GI tract and the release of iron from intracellular stores and mitigating the functional iron deficiency associated with chronic inflammation.

In our Phase 2 studies, patients’ Hb response to roxadustat was independent of the degree of underlying inflammation, as assessed by circulating levels of C-reactive protein (“CRP”), a well-recognized marker of inflammation. Incident dialysis patients are known to have the highest levels of mortality of all dialysis patients. The incident dialysis period is also the period during which mean ESA doses are generally highest. To the extent the increased levels of mortality are associated with high ESA doses, roxadustat may offer a benefit to incident dialysis patients. The median roxadustat dose in our dialysis Study 053 was 1.3 mg/kg; the Cmax of endogenous EPO levels usually associated with this dose level are comparable to the physiologic range naturally experienced by people adapting to high altitude or following blood donation. Refer to additional information on endogenous EPO levels under the heading “Potential for Anemia Correction with Moderate EPO Levels.

Potential for Reduced Hepcidin Levels and Anemia Correction without IV Iron

An important differentiator of roxadustat from ESAs is that roxadustat is expected to correct anemia and maintain Hb without IV iron supplementation. Patients with chronic illness, such as CKD, often suffer from absolute iron deficiency or functional iron deficiency. We believe that elevated levels of hepcidin, the major hormone that regulates iron metabolism, contributes to both absolute and functional iron deficiency.

Our Phase 2 clinical trials have shown that roxadustat can significantly reduce hepcidin levels in patients with DD-CKD and NDD-CKD. The following figure shows an average reduction in serum hepcidin level of approximately two thirds, observed at week 5, in 52 incident dialysis patients treated with roxadustat. Roxadustat’s ability to reduce hepcidin levels was also confirmed in our Phase 3 studies in China.


Reduction of Serum Hepcidin Levels (Study 053) in Incident Dialysis Patients


In addition, we believe roxadustat increases the levels of proteins involved in iron uptake, release, and transport. Data from our Phase 2 clinical trials indicate that oral iron supplementation alone is adequate to correct anemia during treatment with roxadustat, in contrast to ESAs, which typically require IV iron supplementation. Additionally, our data indicate that unlike ESAs, roxadustat treatment does not require that patients be iron replete before initiating therapy.

Avoiding IV iron helps to avoid the significant safety risks associated with IV iron described above, and, because the cost of oral iron is significantly less than the cost of IV iron, could also confer significant costs savings.

Potential Reimbursement and Convenience Advantages

Potentially Differentiated Reimbursement Framework

ESAs are included in the MIPPA bundled payment system in the DD-CKD setting and reimbursed under Medicare Part B in the NDD-CKD setting. Based on our roxadustat data to date, we believe roxadustat has the potential to correct anemia through a differentiated mechanism of action and that offers different therapeutic effects and the potential to displace multiple drugs in current use (such as ESAs and IV iron), and/or those in development (such as agents for suppression of hepcidin). Although the bundle currently covers ESAs or oral equivalents of ESAs or other IV products encompassed by the bundle, due to the differentiated nature of roxadustat, it is unclear whether roxadustat will be included in or excluded from the bundle. We believe that there may be commercial benefits in either event but are unable to predict the potential benefits until further guidance from CMS becomes available.

In the NDD-CKD setting, we expect that roxadustat, an oral treatment, should be subject to Medicare Part D, which would allow physicians to prescribe roxadustat without the financial and reimbursement risk associated with purchasing and storing injectable ESAs. We believe that this should encourage significantly greater usage outside of the dialysis setting.

Potential Reduction of Other Medications

In addition to potentially eliminating the need for IV iron, based on our Phase 2 clinical trial results to date, we believe that roxadustat has the potential to reduce the use of other medications frequently required in some CKD anemia patients, such as anti-hypertensives, anti-coagulants, and statins.

Oral Administration

Many physicians that treat CKD patients, particularly cardiologists, endocrinologists, and internists, do not typically stock or administer ESAs. An easily accessible oral agent that is dispensed by pharmacies could significantly increase the number of physicians treating anemia in patients with CKD, and therefore, the number of patients receiving treatment.

In addition, the oral administration of roxadustat potentially offers a significant convenience advantage for CKD patients who have yet to initiate dialysis and are therefore not regularly visiting a dialysis center. Patients can more easily self-administer medicine in any setting, rather than being subject to the inconvenience and restrictions of regular visits to physicians’ offices or infusion centers to receive treatment with ESAs.

Potential Pharmacoeconomic Advantages

Based on our Phase 2 clinical trial results to date, we believe that roxadustat’s potential pharmacoeconomic advantages over ESA therapy may include safety (with a potential decrease in cardiovascular events and consequently lower associated treatment costs), lower administrative cost, reduction or elimination of IV iron and potentially other medications. If demonstrated in our Phase 3 studies, these pharmacoeconomic advantages may support reimbursement worldwide, including in Europe and China.

The Market Opportunity for Roxadustat

We believe that there is a significant opportunity for roxadustat to address markets currently served by injectable ESAs. According to IMS Health, 2013 global ESA sales in all indications totaled $8.6 billion, driven primarily by $6.2 billion sold in the U.S. and Europe. We believe that a substantial portion of ESA sales are for CKD anemia. For example, in the U.S., EPOGEN, which is primarily used in the DD-CKD patient population, had 2014 sales of approximately $2 billion. We further believe that the number of patients requiring anemia therapy will grow steadily as the global CKD population and access to dialysis care continue to expand, particularly in China and other emerging markets including the rest of Asia, Latin America, Eastern Europe, the Middle East, and the Commonwealth of Independent States.


Furthermore, we believe that there is a significant opportunity for roxadustat to address patient segments that are currently not effectively served by ESAs, such as anemia in the NDD-CKD patient population, which is substantially larger than the DD-CKD patient population. Diabetes and hypertension are the leading causes of secondary CKD. Although we estimate approximately 36% of diabetic and 20% of hypertensive CKD patients are anemic (Hb<12 g/dL), we believe the majority of these patients are currently untreated for anemia since they are under the care of non-nephrology specialists, such as endocrinologists, diabetologists, cardiologists, and internists, where ESA therapies are not readily available.

We also believe that roxadustat may provide a safer option to re-establish the chemotherapy-induced anemia market, which was once a market of comparable size to the DD-CKD anemia market. Other non-CKD anemias, including anemia related to inflammatory diseases, MDS and surgical procedures requiring transfusions, which are not addressed adequately with currently available therapies, could form another opportunity.


We along with our partners, Astellas and AstraZeneca, have designed our global Phase 3 programs to support regulatory approval of roxadustat in both NDD-CKD and DD-CKD patients in the U.S., EU, Japan, and China. These Phase 3 programs are studying multiple patient populations, including incident dialysis patients and stable dialysis patients being compared to epoetin alfa as an active comparator, and include multiple NDD-CKD studies comparing roxadustat against placebo controls.

For our U.S. program, we have agreed with our partner, AstraZeneca, on the timing to complete our Phase 3 studies. We believe at that time we will have accrued sufficient MACE events. We now plan to complete enrollment in the second quarter of 2018, report topline results in the fourth quarter of 2018, and file the NDA for roxadustat in CKD anemia during the first half of 2019.

In 2017, we and our China subsidiary, FibroGen (China) Medical Technology Development Co., Ltd. (“FibroGen Beijing”), reported topline results from our two Phase 3 CKD anemia studies in China. Primary efficacy endpoints were met in both the NDD-CKD and the DD-CKD trials. Results are included below in the section titled “Roxadustat for the Treatment of Anemia in Chronic Kidney Disease in China”. We completed the 52-week safety exposure in the China Phase 3 studies, and in October 2017 the China Food and Drug Administration (“CFDA”) accepted for review our New Drug Application (“NDA”) submission for the registration of roxadustat to treat anemia for DD-CKD patients and NDD-CKD patients. We currently anticipate a market approval decision for our CKD anemia NDA in China by the end of 2018.

In Japan, Astellas is conducting six Phase 3 anemia studies, four in DD-CKD and two in NDD-CKD, of which three have been completed. These studies include conversion studies, studies in ESA-naïve patients, studies in HD and PD, and studies comparing roxadustat to active control.


The table below summarizes our ongoing and completed Phase 3 clinical trials of roxadustat for the treatment of anemia associated with CKD, all of which include Hb level maintenance as a study objective, once correction or conversion is achieved. The chart emphasizes the differences, by regulatory approval region in estimated patient enrollment numbers.

Roxadustat Phase 3 Clinical Trials






Estimated or Completed # of Patients Enrolled


Study Sponsor, Number


































FibroGen - FGCL-4592-060




-------- 900 --------










Astellas - 1517-CL-0608




-------- 597 --------










AstraZeneca - D5740C00001



















Astellas - 1517-CL-0610


Darbepoetin alfa

















FibroGen - FGCL-4592-808



















Astellas - 1517-CL-0310*


Darbepoetin alfa

















Astellas - 1517-CL-0314*



















NDD-CKD Sub Total by Region




































Incident Dialysis



















FibroGen - FGCL-4592-063*


Epoetin alfa


-------- 900 --------





























Stable and Incident Dialysis



















AstraZeneca - D5740C00002*


Epoetin alfa




































Stable Dialysis



















FibroGen - FGCL-4592-064*


Epoetin alfa


-------- 820 --------










Astellas - 1517-CL-0613


Epoetin alfa or Darbepoetin alfa


-------- 838 --------










FibroGen - FGCL-4592-806


Epoetin alfa

















Astellas - 1517-CL-0302



















Astellas - 1517-CL-0307


Darbepoetin alfa

















Astellas - 1517-CL-0308



















Astellas - 1517-CL-0312



















DD-CKD Sub Total by Region




































Total by Approval Region















Combined U.S. and EU total




~ 9,500












Currently recruiting.


Mandatory post-approval safety study of approximately 2,000 patients expected to be required in China.

To maximize the commercial potential for roxadustat, we have incorporated several unique elements into our Phase 3 program. We are performing the first placebo-controlled Phase 3 studies in NDD-CKD patients to potentially demonstrate the benefits of anemia therapy and safety of roxadustat compared to placebo. We are also performing the largest Phase 3 study in incident dialysis anemia patients, who have the highest risk for death, and are the most difficult patients to stabilize and treat for anemia in CKD. Based on data from our Phase 2 studies, we believe that roxadustat may offer a safer alternative to ESAs for this particularly vulnerable patient population. We are also evaluating the cardiovascular safety of roxadustat compared to placebo in NDD-CKD patients to first demonstrate a lack of increased risk to qualify for marketing approval by the FDA, and in these patients we will have an opportunity to measure improvements in patient outcomes with anemia therapy. Separately, we are evaluating cardiovascular safety of roxadustat compared to ESA in DD-CKD patients.

Our U.S. and European Phase 3 Program

Our U.S. and European Phase 3 program has an aggregate target enrollment of more than 8,000 patients worldwide. Our U.S. Phase 3 program is also designed and sized for to demonstrate non-inferiority to comparators for the MACE composite safety endpoints in separate patient pools of NDD-CKD and DD-CKD. Five of the six Phase 3 studies supporting approval in the EU will also support approval in the U.S.


Primary and Secondary Endpoints of Our U.S. and European Phase 3 Program

With our partners, we have designed our Phase 3 studies to evaluate the following endpoints, most of which were evaluated in our Phase 2 studies.


Primary efficacy endpoints for anemia correction studies:



U.S.: Hb change from baseline to the average Hb level during weeks 28-52.



EU: Cumulative % patients with Hb response by week 24. Hb response is defined as Hb of 11 g/dL and an increase of at least 1 g/dL from baseline.


Primary efficacy endpoints for conversion and maintenance studies:



U.S.: Hb change from baseline to the average Hb level during weeks 28-52.



EU: Hb change from baseline to the average Hb level during weeks 28-36.


The primary safety endpoints for U.S. approval will be MACE, which is a composite endpoint designed to identify major safety concerns, in particular relating to cardiovascular events such as cardiovascular death, myocardial infarction and stroke. These results will be pooled across multiple studies and evaluated separately in our NDD-CKD and our DD-CKD trials.


We expect that our Phase 3 clinical trials supporting approval in Europe will be required to include MACE+ as a safety endpoint which, in addition to the MACE endpoints, also incorporates measurements of hospitalization rates due to heart failure or unstable angina.


We also plan to evaluate secondary endpoints, including the following:



IV iron usage in roxadustat-treated patients relative to ESA-treated patients with DD-CKD.



Red blood cell transfusion rate in roxadustat-treated relative to placebo treated patients with NDD-CKD.



Hypertension adverse events in roxadustat-treated patients relative to ESA-treated patients with DD-CKD, and blood pressure in roxadustat-treated patients relative to placebo-treated patients with NDD-CKD.



Total cholesterol, LDL-cholesterol, and very low-density-cholesterol levels in roxadustat-treated patients relative to placebo-treated patients with NDD-CKD and relative to ESA-treated patients in all three anemic CKD patient populations.



Quality of life in roxadustat-treated patients relative to placebo-treated patients with NDD-CKD.



CKD progression in roxadustat-treated patients relative to placebo-treated patients with NDD-CKD.



Hospitalization rate in roxadustat-treated patients relative to placebo-treated patients with NDD-CKD and relative to ESA-treated patients in all three anemic CKD patient populations.



Rate of vascular access thrombosis in roxadustat-treated patients relative to ESA-treated patients in DD-CKD.

Dosing Regimen

Our Phase 3 studies incorporate dosing regimens that were extensively tested in our six Phase 2 studies.


Identified Dosing Regimen. The dosing regimens for our Phase 3 studies are designed to achieve an appropriate rate and magnitude of Hb rise. In our Phase 2 studies, we explored ranges of therapeutic doses under several dosing regimens, including both tier-weight and fixed starting doses and conversion doses. Our Phase 3 program is using two tier-weight starting doses for ESA-naive patients (70 mg for patients between 45 and 70 kg, and 100 mg for patients between 70 and 160 kg). Our Phase 3 dosing strategies are based on our understanding of effective approaches, derived from our Phase 2 studies, tested in modeling and simulation, and designed to achieve Hb correction for patients with varying dose requirements in a manner that is optimal for both patients and physicians.


Dose Titration. Our Phase 3 program is using a predetermined sequence of dose steps to titrate to a patient’s particular response to roxadustat, which had we found simple to implement and sufficient to correct anemia in our Phase 2 studies. In our Phase 2 anemia correction studies, only one or two cycles of dose titration were necessary to achieve Hb correction in at least 80% of patients on average.


Dose Conversion for Dialysis Patients Previously Treated with ESAs. In our Phase 2 conversion studies, we tested a variety of starting doses and developed a mathematical relationship between baseline ESA dose and roxadustat dose required to maintain Hb levels. We use dose conversion tables derived from these Phase 2 studies to formulate starting roxadustat doses in our Phase 3 trials for patients who switch to roxadustat from ESAs.



Dose Frequency. In preclinical and Phase 1 studies, we observed that intermittent dosing yielded optimal responses to roxadustat. Our Phase 2 studies indicated that three-times weekly, twice weekly and once weekly dosing regimens achieved Hb maintenance. In our Phase 3 program we are dosing three-times weekly for all studies except two (060 and 0608), which are dosing some patients twice per week and some patients once per week. We believe that intermittent dosing may help ensure a consistent and durable treatment effect for several reasons:



Greater Hb Response While Minimizing Total Drug Exposure. Early preclinical studies in rodents with a HIF-PH inhibitor (that was not roxadustat) indicated that a greater Hb response could be achieved using a lower total weekly dose with intermittent dosing compared to daily dosing. In the studies shown below, rats were dosed with HIF-PH inhibitor using either a daily or twice weekly dosing regimen. Both a higher Hb response and a better dose response were observed in animals dosed with HIF-PH inhibitor twice weekly compared to animals that were dosed daily. Furthermore, the total weekly dose required to achieve this greater Hb response was lower than daily dosing exposure.

In addition, our previous preclinical studies suggested that a wider therapeutic window was achieved with intermittent dosing as compared to daily dosing. Preclinical observations such as these led us to conclude that intermittent dosing could enable a better Hb response with a lower overall drug exposure and offer a potentially wider therapeutic window.



Reduce the Risk of Changing the HIF Set Point. The HIF system has a built-in negative feedback mechanism. Genes for two of the prolyl hydroxylase domain (“PHD”) enzymes that are responsible for degrading HIF under normal oxygen conditions are actually HIF target genes. Thus, while these PHD enzymes are inhibited by hypoxia, or by a HIF-PH inhibitor, the resulting HIF activation leads to an increase in the very enzymes that are responsible for its degradation following the re-oxygenation, or potential removal of the HIF-PH inhibitor. This negative feedback mechanism is important in enabling the HIF system to reset. However, under chronically hypoxic conditions, it has been shown that the elevation in PHD enzyme levels is maintained, leading to a change in the HIF set point. Based on this knowledge of HIF biology, it is our belief that prolonged HIF activation by a HIF-PH inhibitor drug could similarly lead to a change in the HIF set point, which we believe may require an increased HIF-PH inhibitor dose to elicit the same HIF response. In an effort to avoid this potential outcome, and to potentially prolong drug effectiveness, we have undertaken an intermittent dosing regimen.



Increase Intervals Between HIF Activation. The kinetics of HIF target gene induction (including genes encoding PHD enzymes) are variable, with some HIF target genes being induced very quickly after HIF activation and others requiring longer periods of HIF activation for significant induction. We believe that increasing the intervals between HIF activation using an intermittent dosing regimen has the potential to limit the HIF target gene response.



Potential Commercial Advantages. We expect that a dosing regimen that enables dosing concurrently with HD treatment, typically administered three-times weekly, will be more commercially attractive in the dialysis market.


Our Phase 2 studies indicate that intermittent dosing enabled anemia correction up to 24 weeks and Hb maintenance up to 19 weeks when converting a patient from ESA.

Clinical Trial Eligibility, Iron Status, and Iron Supplementation During Treatment

Unlike ESA clinical trials where patient study eligibility criteria included a requirement of adequate iron availability (measured by ferritin 100 ng/mL and TSAT 20%) and encouraged IV iron use, roxadustat Phase 2 studies included anemic NDD-CKD patients with ferritin 30 ng/mL and TSAT 5% and anemic DD-CKD patients with ferritin 50 ng/mL and TSAT 10%, which permitted the inclusion of iron-deficient patients. Hb response was generally achieved in iron deficient NDD-CKD and DD-CKD patients (ferritin <100 ng/mL and TSAT<20%), despite the fact that IV iron was not allowed during roxadustat treatment.

Our placebo-controlled Phase 3 NDD-CKD studies are using the same iron eligibility criteria employed in our Phase 2 studies, which allow oral iron, but prohibit the use of IV iron (except as a rescue medication). In our Phase 3 DD-CKD studies, since ESA serves as the comparator and similar treatment conditions are required for roxadustat and ESA, study eligibility criteria include ferritin 100 ng/mL and TSAT 20%. Patients are randomized to roxadustat or ESA, and are encouraged to take oral iron as a first-line supplemental agent. IV iron is permitted if there is inadequate Hb response to treatment and if the patient is iron deficient (ferritin <100 ng/mL and TSAT<20%).

Our Phase 2 Program

We and our partner have completed eight roxadustat Phase 2 studies, four in NDD-CKD patients and four in DD-CKD patients, to assess the efficacy of roxadustat to both correct anemia (correction) and maintain the Hb response (maintenance). Two of the six completed Phase 2 studies were conducted in China and are discussed in the section below titled “Roxadustat for the Treatment of Anemia in Chronic Kidney Disease in China”. The efficacy and safety data generated from our China studies were consistent with our U.S. Phase 2 studies and further contributed to the promising efficacy and safety results reported to date. In addition, we announced positive Hb correction and maintenance data from Astellas’ Phase 2 DD-CKD and NDD-CKD studies in Japan in July of 2016 and these studies are discussed in the section below titled “Roxadustat for the Treatment of Anemia in Chronic Kidney Disease in Japan.

Of the remaining four studies, data have been published and presented at various medical conferences and in medical journals. The data from our completed Phase 2 studies demonstrated that roxadustat achieved a clinically meaningful increase in Hb levels in anemic NDD-CKD and DD-CKD patients and maintained Hb levels in DD-CKD patients who were converted from ESA therapy. Roxadustat corrected anemia without the need for IV iron supplementation and exhibited an acceptable safety profile. Specifically, our Phase 2 studies achieved the following objectives:


Identified optimal roxadustat dosing regimens for anemia correction and maintenance of Hb response.


Demonstrated roxadustat’s potential to treat anemia in both NDD-CKD and DD-CKD patients, including incident dialysis patients, the most unstable and high-risk CKD patient population.


Generated substantial safety data, indicating that roxadustat is well tolerated, appears safe, and could offer an improved cardiovascular profile relative to ESAs.


Demonstrated that roxadustat may be able to treat anemia without the need for IV iron supplementation.


Demonstrated that roxadustat can reduce hepcidin levels and potentially treat anemia in a significant subset of patients with inflammation.


The following chart summarizes the design of our completed Phase 2 studies outside of Japan and China (discussed in their respective sections below) and the primary objectives of each study.

Completed Phase 2 Studies










Number of







Study Number,






Number of











CKD Patient








Total Number of

















Patients in Study




Dose Frequencies

FGCL-4592-017 US




Correction, Pharmacokinetics













FGCL-4592-041 US




Correction & Maintenance













FGCL-4592-040 US


Stable Dialysis


Conversion & Maintenance













FGCL-4592-053 Russia, US, Hong Kong


Incident Dialysis















FGCL- 4592-059 US*


Non-dialysis & Dialysis


Long-Term Safety & Maintenance










Up to 5 years






















Study conducted by Astellas


5 patients remain in ongoing study


QW = weekly; BIW = twice weekly; TIW = three-times weekly

Study 017: Dose-Escalating Study in NDD-CKD patients

Study 017 established proof of concept for roxadustat by showing a significant increase in Hb in a dose-dependent manner, and providing data on the relationship between roxadustat dose and Hb response. This formed the basis for the dosing rules that we applied in subsequent studies of longer duration and in a larger number of patients.

This study, a randomized, single-blind, placebo-controlled, dose-escalation study, was the first Phase 2 study to assess the safety and efficacy of a range of roxadustat doses in the correction of anemia in NDD-CKD stage 3 and 4 patients, over four weeks of treatment, and a 12-week safety follow-up period. A total of 117 patients (of which 96 were evaluable) were randomized sequentially into four weight-based dose cohorts: 1 mg/kg, 1.5 mg/kg, 2 mg/kg, and 0.7 mg/kg, respectively. Roxadustat was administered either twice weekly or three-times weekly.

Weight-Based, Three-Times Weekly and Twice Weekly Dosing Leads to Hb Improvement. We tested four different roxadustat weight-based doses administered for four weeks with Hb measurements over a six-week period. As shown in the table below, all of the patients in the highest weight-based dose cohort met the criteria for response in that they achieved Hb rise >1 g/dL in four weeks. As roxadustat achieved 100% Hb response at the 2 mg/kg dose, higher doses were not pursued in this study despite the absence of dose-limiting toxicities. Roxadustat was well tolerated without any safety concerns.

Significant, Dose-Dependent Increases in Hb. As shown in the table below, the dose-dependent change in Hb from baseline in roxadustat patients was statistically significant from placebo by day eight (p=0.025) and remained so at each assessment through week six (p=0.0001 at Day 22; p<0.0001 at Day 26–29/end of treatment).


A p-value is a statistical measure of the probability that the difference in two values could have occurred by chance. The smaller the p-value, the greater the statistical significance and confidence in the result. Typically, results are considered statistically significant if they have a p-value less than 0.05, meaning that there is less than a one-in-twenty likelihood that the observed results occurred by chance. The FDA requires that sponsors demonstrate the effectiveness and safety of their product candidates through the conduct of adequate and well-controlled studies in order to obtain marketing approval. Generally, the FDA requires a p-value of less than 0.05 to establish the statistical significance of a clinical trial, although there are no laws or regulations requiring that clinical data be statistically significant, or that require a specific p-value, in order for the FDA to grant approval.

Hb Responses to a Range of Roxadustat Doses in FGCL-4592-017








0.7 mg/kg



1 mg/kg



1.5 mg/kg





















































Mean Maximum Change in Hb





























Standard Error of the Mean





























% Hb Responder





























Median Time to Response (Days)



























BIW = twice weekly; TIW = three-times weekly

Standard error of the mean (“SE”), is a statistical measure of the amount that an observed mean may be expected to differ by chance from the true mean. For a population that follows a normal distribution, 68% of observed means will be within one standard error of the mean.

Dose-Dependent Reduction in Hepcidin Levels. Roxadustat reduced serum hepcidin levels in a dose-dependent fashion.

Study 041: Study for Optimization of Starting Dose and Dose Titration in NDD-CKD Patients

Study 041 demonstrated that both tier-weight and fixed starting doses can initiate anemia correction. In tier-weight based dosing for this study, we used starting doses based on the patient’s body weight category: high, middle or low. This randomized, open-label Phase 2 study was designed to evaluate the efficacy and safety of roxadustat over 16 to 24 weeks in 145 NDD-CKD patients (of which 143 were efficacy evaluable), and to evaluate the effects of dosing regimens in order to determine an optimized approach to anemia correction. In this trial, we tested six different starting dose regimens: three fixed doses and three tier-weight doses. In fixed dosing, all patients in the same cohort were given the same starting dose. Results from this study were published in the June 2016 Clinical Journal of the American Society of Nephrology.

We tested both three-times weekly and twice weekly dosing frequencies for anemia correction, similar to Study 017, and further demonstrated that Hb levels can be maintained using three dosing frequencies (three-times weekly, twice weekly, and weekly) once target Hb 11 g/dL was achieved. We also studied various dose adjustment rules, with dose adjustment decisions made from five weeks onward, and every four weeks thereafter, to seek the best dose titration scheme.


Hb Correction. We met the primary efficacy endpoint of cumulative number (%) of patients with a Hb response, defined as an increase in Hb 1.0 g/dL from baseline and Hb 11.0 g/dL at the end of treatment. Regardless of the starting dose or dose titration scheme, 92% of patients collectively from all cohorts achieved an Hb increase of at least 1 g/dL from baseline. These data suggest the doses studied are of adequate range for anemia correction. The following figure shows mean Hb levels for the six dose groups.

FGCL-4592-041 Hb Response over Various Dosing Regimens


n at baseline

TIW = three-times weekly; BIW = twice weekly; QW= once weekly

Hb Correction was Independent of Inflammation Status. In this study, in a post-hoc analysis, we observed that the magnitude of increases in Hb in response to roxadustat treatment was comparable for both patients with inflammation (elevated CRP levels) and without inflammation (normal CRP levels).

FGCL-4592-041 Mean (± SE) Maximum Change in Hb (g/dL) in 12 Weeks


This stands in contrast to treatments with ESAs, where elevated CRP is frequently associated with lower Hb response to ESAs. We observed a 38% reduction in mean hepcidin level from baseline with eight weeks of roxadustat treatment (p≤0.0001), which supports our belief in roxadustat’s ability to overcome inflammation and to maintain iron availability for erythropoiesis.

FGCL-4592-041 Mean (± SE) Serum Hepcidin Level (ng/mL)

Hb Correction Without IV Iron and in Patients Who Have Low Iron Levels at Study Initiation. We also evaluated several iron parameters to assess roxadustat’s ability to improve Hb without the use of IV iron. At baseline, 49% of the efficacy evaluable patients did not have sufficient iron levels in the body to qualify for initiation of ESA treatment under current practice guidelines and would have been excluded from participation in all prior ESA Phase 3 trials. These patients would not be considered iron replete and are typically first treated with IV iron prior to ESA treatment initiation in an effort to ensure an adequate response to ESA and to minimize the risk of iron depletion. Of all patients in this study receiving roxadustat, only 38% were taking oral iron supplements. A mean Hb increase of 1.8 g/dL was achieved in the first 16 weeks of treatment without IV iron supplementation. There was no evidence for iron depletion as reticulocyte Hb content (“CHr” or the amount of Hb in newly formed red blood cells) was maintained. On the other hand, there was evidence for improved iron utilization with increases in the MCV and increase in mean corpuscular Hb concentration (MCHC) over the first 16 weeks of treatment with roxadustat from baseline (p=0.0018 and p<0.0001, respectively); both MCV and MCHC typically decrease when there is iron deficiency.

Despite the minimal use of oral iron and lack of IV iron usage, patients who were not iron replete had similar Hb responses at week 16 as patients who were iron replete.

Reduction in Cholesterol Levels. In a post-hoc analysis of all cohorts, total cholesterol decreased during treatment with roxadustat by a mean of 26 (SD+/- 30) mg/dL after eight weeks of therapy. Mean reductions in total cholesterol were greater for patients with abnormally high cholesterol levels (>200 mg/dL). Decreases in cholesterol levels were independent of whether patients were taking statins or other lipid lowering agents. Furthermore, the HDL/LDL ratio improved with roxadustat treatment in the subgroup of patients in which lipid profiles were conducted.

Improvement in Quality of Life. Finally, in an analysis of exploratory endpoints we observed improved quality of life in patients treated with roxadustat using a standard questionnaire called the SF-36 HRQOL. The largest positive changes from baseline occurred in the Vitality subscale (>4 points, p<0.0001) and Physical Component (>1.6 points, p<0.005) subscales of the questionnaire. We believe these data demonstrate that roxadustat may improve quality of life by correcting patients’ anemia.

Study 040: ESA Conversion Study in DD-CKD Patients

Study 040 was designed to evaluate the short- and long-term dosing of roxadustat in patients on HD treatment. These results established a conversion dose relationship between ESAs and roxadustat that will be used for Phase 3 trials. Roxadustat maintained Hb without the use of IV iron, which is generally required for the treatment of anemia by ESAs.


This randomized, single-blind study was the first roxadustat study in patients on HD treatment. Part 1 was a six-week, open-label Phase 2 dose ranging study in 54 patients (of whom 42 were efficacy evaluable) to evaluate the impact of four sequential doses of roxadustat on dialysis patients’ Hb levels over six weeks upon switching from epoetin alfa, in comparison to those continuing prior epoetin alfa doses. Part 2 was a 19-week treatment study in 90 patients (of whom 83 were efficacy evaluable) to establish optimal conversion doses and dose adjustments. Patients included had previously demonstrated a wide range of ESA-responsiveness. In Part 1, study 040 met its primary endpoint of maintaining Hb in patients previously treated with epoetin alfa at week 6, indicating that roxadustat can replace ESAs in DD-CKD. In Part 2, study 040 also met its primary endpoint of maintaining Hb at week 19, indicating that roxadustat may be effective at long-term maintenance of Hb. IV iron was prohibited in both roxadustat-treated patients and ESA-treated control patients during this study.

Maintenance of Hb Levels Following Conversion from ESAs. In Part 1 of this study (six-week treatment), 41 patients were randomized to one of four roxadustat dose cohorts, and 13 were randomized to continue on epoetin alfa treatment. The primary endpoint of this study was to maintain an Hb level equal to or above 0.5 g/dL below baseline Hb by the end of six weeks. As shown in the figure below, roxadustat had a dose-response effect for maintaining Hb levels. The lowest roxadustat dose cohort of 1.0 mg/kg was comparable to epoetin alfa with maintenance in 44% of roxadustat patients and 33% of the control arm, patients who continued treatment with epoetin alfa (but who were required to stop concomitant treatment with IV iron). Roxadustat doses of 1.5 mg/kg or higher performed better than epoetin alfa in maintaining Hb, with 79.2% overall maintenance and 80% maintenance at the 1.5 mg/kg roxadustat dose, 80% maintenance at the 1.8 mg/kg roxadustat dose and 77.8% maintenance at 2 mg/kg roxadustat dose.

In Part 2 of the study (19-week treatment), 67 patients (with baseline ESA dose requirements ranging from 7 to 164.5 U/kg three-times weekly) were randomized to seven cohorts of roxadustat (with various starting doses) and 23 patients were randomized to continue on epoetin alfa. Hb correction in the roxadustat treated patients pooled across all treatment cohorts was maintained over the 19-week treatment period and was comparable to epoetin alfa. The average roxadustat dose requirement for Hb maintenance was approximately 1.70 mg/kg three-times weekly.

In Part 1, which was dose ranging, we observed an increase in Hb level at doses of 1.5 to 2.0 mg/kg TIW as shown in the figures below. In Part 2, which was to establish the optimal conversion dose, we observed similar Hb maintenance between roxadustat and epoetin alfa.

FGCL-4592-040 Mean: (± SE) Hb over Time During Anemia Treatment with Roxadustat or Epoetin Alfa in Dialysis Patients


Part 1 (6-Weeks Dosing)


Part 2 (19-Weeks Dosing)


In addition, in an exploratory analysis of this study we observed a dose-dependent decrease in hepcidin in Part 1 of this study.

FGCL-4592-040: Change in Hepcidin Level from Baseline (ng/mL)


n at baseline


p<0.05 (comparing hepcidin change from baseline between the 2.0 mg/kg roxadustat group and the epoetin alfa group).

DD-CKD patients who switched from ESA treatment to treatment with 2.0 mg/kg roxadustat had significantly greater reduction in serum hepcidin level than those who continued ESA treatment (p=0.038).

FGCL-4592-040 Mean (± SE) Serum Hepcidin Level (ng/mL)


Roxadustat Doses are Associated with Lower Circulating EPO Levels than Epoetin Alfa. The following chart shows the result of six patients who were highly responsive to epoetin alfa and participated in a sub study in which their EPO levels during treatment with roxadustat were compared to EPO levels when the patients were receiving epoetin alfa prior to randomization. Their mean peak EPO concentration after an average dose of 44 U/kg was significantly higher when patients were receiving epoetin alfa relative to when they were receiving a mean roxadustat dose of 1.3 mg/kg as illustrated below. This observation is consistent with the mechanisms of action of ESA and roxadustat, respectively, and we believe the lower EPO exposure observed with roxadustat offers potential safety benefits.

FGCL-4592-040: Mean (+SE) Plasma EPO Levels During Treatment with Roxadustat Compared with Prior Epoetin Alfa Dosing in the Same Patients (n=6)

Maintenance of Adequate Iron Supply. The concentration of Hb within newly formed red blood cells (“CHr”) is a measure of iron availability for erythropoiesis. In an exploratory analysis of this study, without IV iron supplementation (which was prohibited in this study), CHr was maintained during roxadustat treatment but declined in patients who continued treatment with epoetin alfa. This finding indicates that unlike epoetin alfa, roxadustat allows endogenous stores of iron to provide an adequate supply to newly forming red blood cells without IV iron supplementation.


FGCL-4592-040: Mean Reticulocyte Hb Content (CHr) over Time in Subjects Treated with Roxadustat and Epoetin Alfa


n at baseline

Reduction in Total Cholesterol. Consistent with our Phase 2 studies in NDD-CKD patients, we observed in a post-hoc analysis that roxadustat reduced total cholesterol levels in stable dialysis patients, and this effect appeared durable throughout the 19-week treatment period as depicted below.

FGCL-4592-040: Mean (±SE) Total Cholesterol over Time During Treatment of Dialysis Patients with Roxadustat or Epoetin Alfa-Treated


Study 053: Correction of Anemia in Incident Dialysis Patients

Results from this study were published in the April 2016 Journal of the American Society of Nephrology. Incident dialysis patients are at increased risk of serious cardiovascular events and death as compared to stable dialysis patients. The mortality rate among dialysis patients is highest during the first few months of dialysis initiation, and on average, patients also require the highest doses of ESA in this period. These patients typically have high levels of systemic inflammation and require IV iron supplementation for ESA to be effective.

This randomized, open-label study was designed to evaluate the safety and efficacy of roxadustat for correction of anemia in 60 incident dialysis patients (of whom 55 were efficacy evaluable) who were on dialysis for at least two weeks and not more than four months and had not been treated with ESAs, and to compare the treatment responses to roxadustat under the different iron supplementation conditions. All treatment groups in Study 053 met their primary endpoint of increasing Hb level during treatment: each cohort achieved maximum mean Hb increases from baseline, ranging between 2.8 g/dL to 3.5 g/dL, resulting from 12 weeks of roxadustat treatment. We observed that at week 12 in excess of 90% of the patients achieved a greater than 1 g/dL increase in Hb from baseline. In addition, while roxadustat corrected anemia without iron supplementation, oral iron enabled an optimal Hb response. More importantly, oral iron was as effective as IV iron for Hb correction by roxadustat. In contrast, ESA therapy requires IV iron supplementation in this patient population.

This study also showed that roxadustat can correct anemia regardless of the patient’s level of inflammation as measured by CRP. At week 12, the median weekly dose of roxadustat was 4.0 mg/kg in this trial of incident dialysis patients and is similar to the median weekly dose of 4.45 mg/kg at week 12 in Study 040, our trial of roxadustat in stable dialysis patients. In contrast, ESA therapy typically involves higher doses at the time of dialysis initiation.

The 48 HD patients were randomized to one of the three iron supplementation options: oral iron, IV iron, or no iron. Included in the 60 patients were 12 PD patients who received oral iron. This study incorporated the same tier-weight based dosing regimen utilized in Study 041.

Hb Correction in Incident Dialysis Patients without IV Iron Administration. All three cohorts of roxadustat-treated HD patients (no iron, oral iron or IV iron supplementation) and PD patients (oral iron) achieved a significant increase in the maximum Hb change from baseline, which was the primary efficacy endpoint. Most importantly, the maximum increase in Hb was not significantly different between roxadustat-treated HD patients supplemented with oral iron (3.5 g/dL) and those supplemented with IV iron (3.5 g/dL). In contrast, a published study of ESAs in this patient population showed that patients supplemented with oral iron achieved an Hb response comparable to no iron supplementation and significantly lower Hb response than those supplemented with IV iron. These Phase 2 data demonstrate that roxadustat, unlike ESAs, may eliminate the need for IV iron and thus avoid the side effects of IV iron in DD-CKD patients.


FGCL-4592-053: Hb over Time During Anemia Correction with Roxadustat in Incident Dialysis Patients, with No Iron, Oral Iron, or IV Iron Supplementation

Note: Hb = hemoglobin; HD = hemodialysis; PD = peritoneal dialysis; n = number of patients

Note: *p<0.05 compared to IV iron and oral iron

Maintenance of Iron Stores. In an exploratory analysis of this study, TSAT, a marker of iron stores, was well maintained during this period of intensive production of red blood cells with oral iron alone, indicating that iron stores can be maintained without IV iron.

FGCL-4592-053: TSAT over Time During Anemia Correction with Roxadustat in Incident Dialysis Patients, with No Iron, Oral Iron, or IV Iron Supplementation

Hb Correction Independent of Inflammation Status. As is typical of incident dialysis patients, about half of all patients had elevated CRP levels at baseline. In a post-hoc analysis of this study, we observed that Hb responses following roxadustat treatment were independent of baseline CRP levels. These data demonstrate that, unlike the ESAs, roxadustat has the potential to overcome the suppressive effects of inflammation on Hb responsiveness to treatment.


Significant Reduction in Hepcidin. Consistent with our other studies, in an exploratory analysis of this study we observed that patients’ hepcidin levels were significantly reduced, most notably in the no iron and oral iron cohorts, by >50% from baseline, and to a lesser extent in the IV iron cohort. At follow up (four weeks after stopping roxadustat), hepcidin levels approached baseline values. Hepcidin reduction may be one of the mechanisms for overcoming the Hb suppressive effects of inflammation by making iron more available for roxadustat-induced erythropoiesis.

Safety Summary

In addition to the more than 1,100 subjects who have been exposed to roxadustat in Phase 1 and Phase 2 clinical studies, including treatment of some Phase 2 study patients for 24 weeks and several patients for approximately four years in a safety extension study, our ongoing Phase 3 program provides significant additional long-term safety data.

A range of roxadustat doses, up to 3.0 mg/kg in DD-CKD patients and up to 5.0 mg/kg in healthy volunteers, have been administered and all roxadustat doses have been well-tolerated. In December 2017, the roxadustat data safety monitoring board completed its scheduled review of the data from all active Phase 3 roxadustat clinical trials and recommended that the program proceed with no protocol changes. The following summarizes the safety findings of our preclinical, Phase 1 and Phase 2 studies:

No Overall Safety Signals. An independent data monitoring committee consisting of external experts in nephrology, hepatology, and biostatistics reviewed safety data from all U.S. and Europe Phase 2 studies, and determined there were no safety signals. The overall frequency and type of treatment-emergent adverse events (“TEAEs”) and serious adverse events (“SAEs”) observed in these clinical studies reflect events that would be expected to occur in each of the NDD-CKD and DD-CKD patient populations. Safety analyses did not reveal any association between the rates of occurrence of cardiovascular events with roxadustat dose, rate of Hb rise or Hb level. The SAEs experienced in our studies identified by the principal investigator as possibly related to roxadustat were a stroke in a patient with a prior history of multiple strokes, one incident of vomiting, and one incident of deep venous thrombosis. The most commonly reported TEAE in the Phase 2 studies were diarrhea, nausea, urinary tract infection, nasopharyngitis, peripheral edema, hyperkalemia, headache, hypertension, and upper respiratory tract infection.

Of our completed Phase 2 clinical studies outside of Japan and China (discussed in their respective sections below), two (Studies 017 and 040) were controlled, and two were not, one with placebo and one with ESA.

For Study 017, which had a treatment period of four weeks, for 88 subjects on roxadustat, and 28 subjects on placebo, we observed treatment-emergent SAEs (“TSAEs”), in four patients (4.5%) on roxadustat, with zero cardiovascular SAEs and zero SAEs for the composite safety endpoint. There were also TSAEs in one patient (3.6%) in the placebo arm of the study, including one cardiovascular SAE and zero SAEs for the composite safety endpoint. The composite safety endpoint (exploratory analysis) includes death, myocardial infarction, congestive heart failure, subendocardial ischemia, cerebrovascular accident, thrombosis (fistula), arteriovenous fistula occlusion, angina pectoris, and vascular graft thrombosis. A patient may experience more than one SAE, in which case a patient is only counted once in this analysis. TSAEs observed in patients treated with roxadustat were arteriovenous fistula site complications, dyspnea, femoral neck fracture, and non-cardiac chest pain. SAEs observed in patients treated with placebo were acute renal failure and pericarditis.

For Study 040, for those who had a treatment period of 19 weeks, for 66 subjects on roxadustat, and 23 subjects on ESAs, we observed TSAEs in 15 patients on roxadustat (22.7%), including one cardiovascular SAEs (1.5%), and eight SAEs for the composite safety endpoint (12.1%), and TSAEs in four patients on ESAs (17.4%), including two cardiovascular SAEs (8.7%), and four SAEs (17.4%) for the composite safety endpoint. TSAEs categorized by System Organ Class, a standard event classification, observed in patients treated with roxadustat were infections and infestations (5), metabolism and nutrition disorders (2), cardiac disorders (1), gastrointestinal disorders (1), nervous system disorders (2), respiratory, thoracic and mediastinal disorders (2), skin and subcutaneous tissue disorders (1), injury, poisoning, and procedural complications (2), and psychiatric disorders (1). TSAEs categorized by System Organ Class observed in patients treated with ESA were infections and infestations (3), metabolism and nutrition disorders (3), cardiac disorders (1), respiratory, thoracic and mediastinal disorders (1), blood and lymphatic system disorders (1), and vascular disorders (1).

The differences in the SAE percentages described are not considered statistically significant.

The three SAEs described above that were considered by the principal investigator to be possibly related to roxadustat did not occur in these four other studies.

No Liver Enzyme Safety Signal. Liver enzymes were monitored closely in the roxadustat Phase 2 clinical development program. No evidence of hepatotoxicity was observed in any of the roxadustat clinical trials, and the independent data monitoring committee concluded that there was no concern for hepatotoxicity to date. Liver enzymes are being monitored in Phase 3 according to current FDA guidelines, without any special requirements.


Extensive Evaluation of Cancer Risk. Furthermore, to assess the potential cancer risk of roxadustat, we conducted 12 tumor studies in rodents. These studies included xenograft, syngeneic, or spontaneous tumors of lung, colon, breast, pancreas, melanoma, ovarian, renal, prostate, and leukemic origin, several of which are reported to be dependent on vascular endothelial growth factor (“VEGF”), a protein that can be regulated by HIF for which increased levels have potentially been linked to increased tumor growth. No effect on tumor promotion was observed with roxadustat in any of the studies. In addition, roxadustat had no effect on tumor initiation or metastasis in the studies in which these end-points were also measured. Five other HIF-PH inhibitors from our library have been evaluated in many of the same rodent tumor models as roxadustat, as well as some additional ones (35 studies of six HIF-PH inhibitors in 18 models total), with no observed effect on tumor initiation, promotion or metastasis. Finally, no significant increases in plasma VEGF levels have been observed in any of our nonclinical studies at clinically relevant doses of roxadustat.

In March 2015, we received final reports for two-year rat and mouse carcinogenicity studies of roxadustat. Roxadustat treatment had no adverse effect on survival and did not cause carcinogenic effects in either species. Two-year rodent carcinogenicity studies that were conducted with one of the other HIF-PH inhibitors evaluated in the tumor models showed no effect on mortality or incidence of tumors.

In clinical studies to date, we and our independent data monitoring committee have not identified any evidence to suggest tumor risk in the use of roxadustat.

No QT Prolongation. We conducted a thorough QT study evaluating roxadustat doses up to 5 mg/kg (approximately four times the average maintenance dose studied in the NDD-CKD patient population). A lengthened QT interval is a biomarker for certain ventricular arrhythmias and a risk factor for sudden death. Our results demonstrate that roxadustat did not affect the QT interval in this study. Based on the extensive safety data collected to date, we believe that roxadustat has a favorable safety profile that supports its further development in Phase 3 clinical studies.


We believe there is a particularly significant unmet medical need for the treatment of anemia in CKD in China. Specifically, anemia is undertreated in the rapidly growing population of dialysis patients. In the non-dialysis population, only a small percentage receive any anemia treatment, and those who do, do so only at a minimal level, including patients who are eligible for dialysis and are severely anemic. In the context of the rapidly growing Chinese pharmaceutical market, we believe that the demand for anemia therapy will continue to grow as a result of an expanding CKD population, as well as the central government’s mandate to make dialysis, which is still in the early stages of infrastructure development, more available through expansion of government reimbursement and build-out of dialysis facilities. We believe that roxadustat is a particularly promising product candidate for this market. China’s approved generic drug name for roxadustat (also referred to as FG-4592) is 罗沙司他.

Phase 3 Studies in China

In January 2017, we reported topline results from our two China Phase 3 studies of roxadustat in CKD anemia. In October 2017, the CFDA accepted for review our NDA submission for the registration of roxadustat to treat anemia in DD-CKD and NDD-CKD patients. We currently anticipate a market approval decision for our CKD anemia NDA in China by the end of 2018.

Study 4592-808: Eight-Week Placebo-Controlled Portion of 26-Week Correction in China NDD-CKD Patients

In the double-blind, placebo-controlled eight-week portion of the 26-week NDD-CKD clinical trial in China, 151 anemia patients were randomized 2:1 to receive roxadustat (n=101) or placebo (n=50). Roxadustat met its primary efficacy endpoint of correcting anemia, by achieving a statistically significant increase in Hb levels compared to placebo over eight weeks. Furthermore, the secondary endpoint of Hb response was met as Hb response was achieved by a higher proportion of patients in the roxadustat arm than in the placebo arm.

Roxadustat-treated patients achieved a mean Hb increase of 1.9g/dL from baseline (8.9g/dL) over eight weeks of treatment vs. a mean change in Hb of -0.4g/dL (from 8.9 g/dL baseline) in the placebo arm; the least square mean (“LS Mean”) difference of the two arms is significant, p<0.000000000000001.

A significantly higher proportion of roxadustat patients achieved Hb response (an increase ≥1g/dL from baseline) after eight weeks vs. placebo patients, 84.2% compared to 0.0% (p<0.000000000000001). 67% of the roxadustat treated patients vs 6% of placebo- patients achieved Hb correction to be at or above the desired range of Hb≥10 g/dL within eight weeks, p=0.000000000000077.


There was a significant increase in Hb level from baseline at every weekly measurement between weeks two to eight per figure below.

FGCL-4592-808: Mean (+/- SE) Change from Baseline in Hemoglobin (Hb)

After week eight, placebo patients were converted to roxadustat treatment and patients originally in the roxadustat arm continued treatment through week 26. Anemia correction and hemoglobin maintenance were observed up to week 27.

FGCL-4592-808: Mean Change in Hemoglobin over Time in Phase 3 China Non-Dialysis CKD Patients (26 Weeks) Including Placebo Crossover to Roxadustat

In the 26-week portion of this China Phase 3 non-dialysis study, 97.6 % of patients who received up to 26 weeks of roxadustat achieved anemia correction with Hb ≥10.0g/dL. For patients who crossed over from placebo to roxadustat, there was an increase in mean hemoglobin levels over 18 weeks of roxadustat treatment, with mean hemoglobin increasing from 8.6 g/dL (averaged over weeks seven to nine) to 10.8 g/dL (averaged over weeks 23 to 27); a statistically significant increase (p <0.0001). Hemoglobin levels declined after week 27 when patients were no longer receiving roxadustat, as illustrated by the figure above.

In the 26-week portion of this non-dialysis study, roxadustat was shown to increase hemoglobin regardless of baseline inflammation status: both in patients with inflammation (CRP >4.9 mg/L) and patients without inflammation (CRP ≤4.9 mg/L).

The durability of roxadustat’s effect on hemoglobin levels was further supported by data from the subset of patients (n=23) who participated in the 52-week safety extension of this non-dialysis China Phase 3 study. Approximately 95% of the non-dialysis patients who completed the 52-week safety extension period maintained Hb ≥10.0g/dL at the end of treatment.

In addition, in the eight-week portion of this non-dialysis study, roxadustat led to significant reduction in serum hepcidin levels (-56.1 ng/mL for roxadustat patients vs -15.1 ng/mL for placebo, p=0.00000005). Anemia treatment with roxadustat was effective without the use of IV iron and there was no iron parameter (ferritin or TSAT) requirement for patients at study entry.


Study 4592-806: 26-Week Maintenance in China DD-CKD Patients

In the Phase 3 DD-CKD study, 304 patients (271 HD and 33 PD patients) previously on epoetin alfa were randomized to and treated with roxadustat (n=204) or epoetin alfa (n=100) for 26 weeks. Prior to randomization, patients in this study were previously treated with stable doses of EPO: 7% patients were treated with 利血宝® (“Li Xue Bao”) epoetin alfa, manufactured in Japan and marketed in China by Kyowa Hakko Kirin China Pharmaceutical Co., Ltd. (“Kirin EPO”) and 93% of patients were treated with one of eight other brands of EPO commercially available in China. All the patients randomized to the active comparator arm were treated with Kirin EPO.

The primary endpoint was mean change in Hb from baseline to the average level during the final five weeks of the 26-week treatment period. Roxadustat met the predefined non-inferiority criterion for this primary endpoint in comparison to Kirin EPO in both full analysis set and per protocol set (“PPS”) analysis. In a pre-specified sequential analysis, roxadustat also showed statistical superiority over Kirin EPO for the primary endpoint, the mean Hb increase observed in the roxadustat arm was higher than in the Kirin EPO arm, 0.75g/dL vs. 0.46g/dL, with a significant difference in the LS Mean Hb change in the two treatment arms, p=0.037, in PPS analysis; baseline Hb was 10.4 g/dL in both treatment arms.

Anemia in patients treated with roxadustat was corrected earlier than in Kirin EPO patients and roxadustat maintained Hb levels at a higher level (between 10-12 g/dL) than those receiving Kirin EPO despite the increase in average dose of Kirin EPO (relative to average baseline EPO dose) received by patients in the comparator arm (as shown in the inflammation figures below).

FGCL-4592-806: Mean (+/- SE) Change from Baseline in Hemoglobin (Hb)

We performed a subgroup analysis based on patients’ levels of inflammation, a common co-morbidity with CKD patients. Roxadustat raised and maintained hemoglobin levels in patients with inflammation (defined as having baseline CRP levels > ULN 4.9 µg/L) at doses that were equal to or lower than those received by the patients without inflammation. In contrast, patients with inflammation in the Kirin EPO arm realized lower Hb levels than patients with normal CRP levels (despite receiving higher doses of EPO than patients without inflammation), reflecting roxadustat’s potential to overcome the hyporesponsiveness seen in ESA treatment, as discussed in the section above titled “Limitations of the Current Standard of Care for Anemia in CKD”. The two figures below show mean change in Hb with mean patient dose, and mean Hb levels with mean patient dose, in patients with and without inflammation.


FGCL-4592-806: Mean (+/- SE) Change in Hemoglobin (Hb) from Baseline and

Mean (+/- SE) Dose for Patients with and without Inflammation


FGCL-4592-806: Mean (+/- SE) Hemoglobin (Hb) and

Mean (+/- SE) Dose for Patients with and without Inflammation

112 roxadustat patients continued treatment in a safety extension study for a total of 52 weeks. Approximately 96% of the dialysis patients who completed the 52-week safety extension period maintained Hb ≥10.0 g/dL at the end of treatment.

Hepcidin, the key hormone that regulates iron metabolism, is generally elevated and contributory to EPO hyporesponsiveness in patients with inflammation. Consistent with previously reported Phase 2 CKD data from the U.S. and China, a reduction of serum hepcidin levels was observed in our two China Phase 3 studies of roxadustat in CKD. In the Phase 3 dialysis study, the mean decrease in serum hepcidin levels from baseline to the end of 26 weeks of treatment was 30.2 ng/mL in the roxadustat arm vs 2.2 ng/mL in the EPO comparator arm. In the Phase 3 non-dialysis study, the mean decrease in hepcidin levels at the end of 8-week double-blind treatment period was 56.1 ng/mL in roxadustat arm vs 15.1 ng/mL in the placebo arm (p=0.00000005).

Roxadustat was generally well tolerated and there were no safety signals observed in the China Phase 3 clinical trials, including through the 52-week safety extension periods. There were no study drug-related deaths. The AEs and SAEs reported in the Phase 3 studies were generally representative of the underlying patient population and associated co-morbidities. Treatment of anemia with roxadustat in these Phase 3 clinical trials did not lead to an increase in blood pressure.

China Phase 2 Studies

We performed two Phase 2 studies in China, one trial in NDD-CKD patients, and another trial in DD-CKD patients. In these trials, Hb correction in NDD-CKD patients and Hb maintenance in DD-CKD patients replicated the results seen in the U.S. trials. SAEs were progression of CKD, infection and high potassium levels and the most common adverse events were infections, high potassium levels, nausea and dizziness. There were no dose-related trends or imbalances in the nature of adverse events between patients treated with roxadustat compared to patients treated with either placebo (study 047) or epoetin alfa control (study 048) groups.


Study 047: Eight-Week Placebo-Controlled Correction in China NDD-CKD Patients

In this multi-center, double-blind, placebo-controlled study, 91 anemic CKD patients were randomized 2:1 to roxadustat or placebo treatment groups, respectively, in two sequential dose cohorts or placebo. Results from this study were published in the August 2015 Journal of Nephrology Dialysis Transplantation. Iron repletion at baseline was not required and IV iron supplementation was prohibited during the trial; oral iron supplementation was allowed during the trial, similar to the corresponding U.S. Study 041. The study used tier-weight starting dose for four weeks after which the roxadustat dose was adjusted, depending upon the initial response to treatment. Study 047 met its primary endpoint of a mean maximum increase from baseline Hb at the end of week 8. The mean Hb increases at the end of eight weeks of treatment were 1.6 g/dL and 2.4 g/dL in the low-dose and the high dose cohort, respectively, compared to 0.4 g/dL for placebo, p <0.0001 for each cohort compared to placebo.

FGCL-4592-047: Hb over Time (g/dL) in Chinese NDD-CKD Patients

n at baseline

For Study 047, which had a treatment period of eight weeks, for 61 subjects on roxadustat, and 30 subjects on placebo, we observed TSAEs in eight patients on roxadustat (13.1%), with zero cardiovascular SAEs, and zero SAEs for the composite safety endpoint, and TSAEs in four patients on placebo (13.3%), including one cardiovascular SAE (3.3%), and one SAE (3.3%) for the composite safety endpoint. TSAEs observed in patients treated with roxadustat were chronic renal failure (4), upper respiratory tract infection (1), hyperkalaemia (2) and urinary tract infection (1). TSAEs observed in patients treated with placebo were unstable angina (1), anemia (1), retinal detachment (1), pneumonia (1) and gastritis (1).

Study 048: Six-Week Conversion in China DD-CKD Patients

In this multi-center, open-label, ESA-controlled study, 87 HD patients (of which 82 were efficacy evaluable) with Hb 9 to 12 g/dL previously maintained with ESAs were randomized 3:1 to roxadustat or epoetin alfa treatment groups, respectively, in three sequential dose cohorts of increasing starting doses of roxadustat. Results from this study were published in the February 2016 American Journal of Kidney Disease. This study design was similar to Part 1 of Study 040. Study 048, an exploratory study, achieved its objective of number (%) of patients with successful dose conversion whose Hb levels are maintained at no lower than 0.5 g/dL below their mean baseline value at the end of weeks five and six (59.1% for the low-dose, 88.9% for the mid-dose, and 100% for the high dose). The Hb responses to the roxadustat treatment of Chinese dialysis patients, with the low dose cohort were numerically similar to epoetin alfa, while the mid-dose and the high-dose cohorts each had a statistically significantly higher Hb response rate than epoetin alfa. Hb responses to the roxadustat treatment of Chinese dialysis patients (as shown in the figure below) were similar to Part 1 of Study 040 in the U.S.


FGCL-4592-048: Hb over Time in Chinese Stable Dialysis Patients

For Study 048 which had a treatment period of six weeks, for 74 subjects on roxadustat, and 22 subjects on ESAs, we observed zero TSAEs in patients on roxadustat, including cardiovascular SAEs and for the composite safety endpoint. There were also zero TSAEs in the patients taking ESAs.

Phase 1 Trials

We completed Phase 1 trials of single and multiple ascending doses of roxadustat. Key findings were:


Roxadustat pharmacokinetic parameters in Chinese are similar to those in Caucasians and Japanese.


Stimulation of endogenous EPO, a marker of roxadustat pharmacodynamics, in Chinese is similar to stimulation in Caucasians and Japanese. 


Roxadustat was well tolerated and there were no negative safety signals.

Addressable Patient Populations in China

Based on a large-scale cross-sectional survey performed between September 2009 and September 2010 published in the Lancet (Zhang, et al. Lancet (2012)), there are an estimated 119.5 million CKD patients in China. Based on the prevalence ratios, there were approximately 19 million patients in CKD stage 3, stage 4, and stage 5 which we have grouped into three categories: DD-CKD patients; Dialysis Eligible patients who need dialysis under treatment guidelines but are not dialyzed (“Dialysis Eligible NDD-CKD”); and stages 3 and 4 patients as well as stage 5 patients who are not eligible for dialysis (“Other NDD-CKD”).

DD-CKD (Dialysis)

Dialysis can be delivered in the form of HD or PD. In China, HD is mostly performed at dialysis clinics within hospitals, not at freestanding dialysis centers outside of hospitals which is the common practice in the U.S. PD is self-administered at home by patients, and they visit their nephrologists on a monthly basis at the hospital for monitoring and follow-up.

Dialysis Eligible NDD-CKD

Dialysis Eligible NDD-CKD refers to patients who need dialysis under Chinese treatment guidelines but are not dialyzed. The Chinese treatment guidelines recommend initiation of dialysis at eGFR<10 mL/min/1.73 m 2 (and eGFR<15 mL/min/1.73m 2 for diabetic nephropathy patients). The then Minister of Health estimated that one to two million people in China were eligible for dialysis in 2011. Of the Chinese population potentially eligible for dialysis, however, we believe that only 500,000 were on dialysis as of 2016. While the size of the dialysis population in China is similar to that of the U.S. (the second largest dialysis population in the world), it nevertheless falls far short of the number who require dialysis treatment. We believe that this Dialysis Eligible NDD-CKD population is characteristic of developing markets like China and is at risk for severe anemia.


Other NDD-CKD refers to the other sub-groups of CKD patients within non-dialysis who are earlier stage: CKD patients in stage 3 and stage 4, as well as stage 5 who are not eligible for dialysis. Some of these patients receive medical care in endocrinology, cardiology or internal medicine clinics outside of nephrology where they are treated for their primary disease.


Unmet Medical Need

DD-CKD Patients are Under-Treated for Anemia

We believe there is chronic under-treatment for anemia within the DD-CKD patient population, as many patients do not reach target Hb levels despite ESA therapy. The consensus opinion of the expert panel assembled by the Chinese Journal of Nephrology in 2013 advocated treating to Hb 11.0 g/dL to 13.0 g/dL, whereas we believe, based on our discussions with key opinion leaders, that in clinical practice, nephrologists generally target Hb 10.0 g/dL. However, according to the 2012 Shanghai Dialysis Registry, approximately 50% of patients in Shanghai did not exceed a Hb level of 10.0 g/dL and approximately 75% did not exceed Hb 11.0 g/dL. Over 19% of dialysis patients failed to reach a severely low Hb level of 8.0 g/dL. The Chinese Renal Data System reported that in 2015, the most recently reported data, the average Hb level of DD-CKD patients in the registry was approximately 10.0 g/dL.

We believe there are a number of factors that have led to under-treatment of anemia in the dialysis population, including:


The ESA doses used are generally not sufficient to treat to target Hb levels for certain patient populations. We believe that the reasons include constraints on reimbursement for anemia treatment and fixed hospital pharmacy budgets, as well as safety and efficacy limitations of these drugs. Lower dose levels are particularly ineffective in the hypo-responsive patient population.


The use of IV iron, which is often needed to correct Hb to target levels with ESAs, is limited due to limited reimbursement and perceived clinical risk.


The PD population who receive dialysis treatment at home face the same logistical issues that impede ESA use in the NDD-CKD population discussed below.

Dialysis Eligible NDD-CKD and Other NDD-CKD Patients are Largely Un-Treated for Anemia

Apart from the ESAs used by the dialysis patients in China, we believe that there is a low level of use of ESAs in the non-dialysis population. Based on our clinical trial experience in China, we believe use of ESAs in this population is generally limited to “CKD Clinics” at major research hospitals in top cities where CKD patients are admitted into programs for academic research purposes. We believe there are a number of significant impediments that inhibit the use of ESAs in the outpatient setting, for patients who are not already visiting the hospital for dialysis treatment on a regular basis.


Generally, under the Chinese healthcare system, patients do not have a personal physician but rather are seen by the physician on the schedule on the day of the visit. This absence of continuity of care makes managing the potential risks of ESAs and the titration of ESA treatment (needed to maintain Hb within target range) particularly difficult.


Hypertension and associated cardiovascular co-morbidities are top risk factors for the CKD population. Many physicians in China believe that for the outpatient NDD-CKD population, the risk of developing new or exacerbating existing hypertension from ESA with the attendant risk of worsening renal failure outweigh the potential benefits of ESA use.


Injectable drugs like ESAs present a challenge in China because even subcutaneous administration is performed at hospitals and not in the home. Frequent hospital visits for injections, for the sole purpose of receiving ESA treatment, can present a substantial logistical and financial burden on patients.


Nephrologists are the primary prescribers of ESAs. CKD patients with hypertension or diabetes who are treated by other physicians, such as cardiologists and endocrinologists, are generally not treated with ESAs.


Non-dialysis patients are covered under outpatient reimbursement, unlike dialysis patients who are covered under Severe Disease reimbursement. The lower level of reimbursement coverage means a higher patient co-pay, which further limits ESA use and compliance.

We believe that these impediments have contributed to a low rate of ESA use in the NDD-CKD population in China, and that roxadustat, as an oral agent triggering the HIF mechanism of action, has the potential to make this population accessible for effective anemia treatment in CKD.

Growing Market Opportunity

China is the second largest pharmaceutical market after the U.S. Healthcare expenditures in China have been estimated to be approximately $640 billion in 2015. We believe several factors will continue to drive the growth of the overall pharmaceutical market in China as well as the market for the treatment of anemia in CKD. These factors include continuing urbanization, an aging population and the increasing prevalence of chronic diseases (particularly diabetes and hypertension which are common causes of CKD), and income growth. We also believe that the increasing standard of living will drive higher rates of disease awareness, leading to greater rates of diagnosis and treatment.


Current ESA Market Size and Drivers of Market Growth in China

Total ESA sales in China were approximately $330 million in 2017, of which an estimated 80%, or approximately $275 million, is derived from CKD sales, based on data from IMS Health China. The ESA market in China has grown at a 12% compound annual growth rate between 2013 and 2017 based on data from IMS Health China.

We believe that given the limited availability of dialysis in China, the dialysis market is still in the early stages of development relative to the U.S., and has the potential for sustained long-term growth. We believe growth of dialysis will be driven by the expansion of reimbursement and expansion of dialysis facilities. We further believe that the growing pipeline of CKD patients and expansion of reimbursement will drive growth in demand for anemia treatment in CKD patients.


Expansion of Reimbursement. Reimbursement exists for the use of ESAs in the treatment of anemia in CKD and the coverage levels are expanding. Under Basic Medical Insurance, the reimbursement program for the urban population, coverage for healthcare and drugs is categorized into one of three categories: outpatient, inpatient, and Severe Disease. Both the Dialysis Eligible and Other NDD-CKD patients are reimbursed under outpatient coverage. As an example, coverage levels for outpatient are in the 60-85% range in Shanghai, depending on level of hospital visited and patient age. Dialysis patients, on the other hand, receive reimbursement under the more generous Severe Disease coverage, which is reimbursement for catastrophic healthcare expenditures. Coverage levels are set at a minimum level of 50% by policy and are as high as 85% for employees and 92% for retirees in Shanghai. We expect the availability of Severe Disease reimbursement to significantly drive the utilization of dialysis services and ESAs in the coming years.


Expansion of Dialysis Infrastructure. We believe the number of DD-CKD patients increased from approximately 275,000 in 2011 to approximately 500,000 in 2016 and has grown at a compound annual growth rate of 13% per year from 2011 to 2016. Despite this substantial rate of growth, the Ministry of Health and the Chinese Society of Nephrology have publicly recognized the need for further investment in dialysis infrastructure to accommodate the expected continued growth of the patient population requiring dialysis. PD is an alternative to HD and does not require the level of capital investment in facilities and equipment that is necessary to enable HD. At the end of 2015, PD was estimated to account for 14% of the current dialysis population.


Demographics-Driven Growth. Diabetes and hypertension are common causes of CKD, the rates of which have been growing in China over past two decades. China is experiencing epidemiological changes in metabolic diseases due to economic development, urbanization and an aging population. We believe the increase in diabetes and hypertension prevalence will result in increasing numbers of patients with CKD in the future.

Our China Solution

We believe that roxadustat, if approved, has the potential to address the unmet medical need for the treatment of anemia in each of the three categories of CKD patients in China. Several of the safety, efficacy, reimbursement and convenience advantages that roxadustat, our oral therapeutic, potentially offers over ESAs (refer to “— Our Solution — Roxadustat — A Novel, Orally Administered Treatment for Anemia”) are particularly applicable in the China market.

Roxadustat May Address Chronic Under-Treatment in DD-CKD Patients

We expect roxadustat to be viewed as more attractive than ESAs, and particularly attractive within certain categories of the dialysis population — patients who are not treated to target Hb levels for any reason, patients who are hyporesponsive to ESAs, patients on PD, which is home-based, and DD-CKD patients who have not previously received ESA treatment.


Roxadustat May Increase Rate of Successful Anemia Treatment. We believe that the level of ESA dosing generally used in China is not adequate to achieve target Hb levels for many dialysis patients, especially with minimal use of IV iron. The dose levels used are within a very narrow range due to clinical concerns over ESA safety at higher doses. Moreover, reimbursement limits may cap ESA dose. In contrast, assuming roxadustat is approved, we believe we can price roxadustat so that reimbursable doses of roxadustat will be sufficient to treat most patients to target Hb levels.


Roxadustat May Address Hyporesponsiveness. Hyporesponsive patients, who often fail to respond to ESA treatment, in particular are often inadequately treated due to need for significantly higher doses of ESAs. Our data suggest that roxadustat may be safe and effective in this patient population without the use of high doses.


Roxadustat May Reduce Requirements for IV Iron. ESAs generally require IV iron for effective anemia treatment, and IV iron use is limited in China due to limited reimbursement and perceived clinical risk. Roxadustat potentially eliminates the need for IV iron to reach treatment target.


Roxadustat May Address Lack of Access of ESA Treatment in NDD-CKD Patients

We view NDD-CKD as the segment where roxadustat, with the benefits of the HIF mechanism of action and being an orally administered small molecule, could potentially represent the only viable treatment solution for this patient population.


Roxadustat May Make Treatment Accessible and Feasible. As an oral agent, roxadustat eliminates the need for frequent hospital visits which are needed for ESA administration, decreasing the overall cost and inconvenience of treatment, particularly for DD-CKD patients undergoing PD who are otherwise treated in the home, as well as Dialysis Eligible NDD-CKD and Other NDD-CKD patients.


Roxadustat May Have an Improved Safety Profile. ESA treatment is associated with an increased risk of severe adverse events including hypertension, stroke, myocardial infarction and death. Our data suggest that roxadustat may not increase the risk of these events and therefore may be safer than ESAs thereby potentially removing a significant deterrent to anemia therapy in China.

Roxadustat May Add Value in Both the NDD-CKD and DD-CKD Patient Populations


Roxadustat May Reduce Overall Cost of Treatment Associated With Anemia. For the equivalent reimbursement cost to the government, we believe that roxadustat may deliver a higher potential clinical benefit compared to ESAs. Roxadustat, if approved, could treat patients to target Hb level. Roxadustat could also potentially lower the use of IV iron and anti-hypertensives. Moreover, the total cost of care would be reduced by lowering loss of time and cost of hospital-based ESA injections, and eliminating the infrastructure costs necessary to store ESAs in a cold storage environment. Finally, patients would benefit by reducing the cost of travel to the hospital and the potential lost wages for hospital visits.


Regulatory Strategy

We are committed to bringing first-in-class innovative medicines to China on an accelerated basis, and consistent with this commitment, roxadustat is being advanced in China as a Domestic Class 1 applicant under the CFDA designations. Innovation by domestic companies has become a top priority for the Chinese government. “Domestic” means that all clinical data for approval is generated from within China, and manufacturing for both drug substance and drug product is conducted in China. Whereas any new chemical entity (“NCE”) that has not yet been approved anywhere in the world can be designated as Class 1, roxadustat is also first-in-class, defined as the first NCE for a brand new mechanism of action.

The NDA that was accepted by the CFDA in October 2017 is the first NDA filing of any HIF-PHI worldwide. China could become the first in-world approval country for a first-in-class drug. Upon NDA approval, we expect to be granted a license through the Marketing Authorization Holder (“MAH”) pilot program, described further in the Manufacturing Certification section below.

Manufacturing Certification

As a Domestic Class 1 applicant, FibroGen will be performing commercial manufacturing for both drug substance and drug product in China. FibroGen Beijing has been operating a 4,800 square meter manufacturing facility in Beijing after we received a Pharmaceutical Production Permit (“PPP”) from the Beijing CFDA in August 2014. The PPP is a general certification by the CFDA that the facility is deemed ready for current good manufacturing practices (“cGMP”) production. We expect to be granted the Manufacturing License for Drug Substance and Drug Product for roxadustat at the time of NDA approval, and will become the sole licensed manufacturer for each in China.

Under the current regulatory system in China, it is the manufacturer, not the sponsor, who has the right to sell the manufactured product to the market. A recently implemented CFDA regulation has opened up manufacturing and supply chain options that were previously not possible. In November 2015, China announced the three-year MAH pilot program in certain regions, and implemented the program in August 2016. Under this system, a sponsor of a compound such as roxadustat has the right to obtain additional drug supply from contract manufacturing organizations without giving up its manufacturing license. We intend to participate in this program and if accepted, we may be able to outsource drug product or active pharmaceutical ingredient (“API”) manufacturing to third parties to support commercialization.

We submitted an application to be designated MAH at NDA approval. Considering that FibroGen is both the sponsor and the manufacturer, FibroGen would have the right to market and sell even without MAH. However, under MAH, we may also have the ability to contract for additional supply (for redundancy and to meet capacity needs) from contract manufacturing organizations without giving up our manufacturing license.


There are evolving environmental and manufacturing regulations in China which would limit chemical manufacturing in large cities such as Beijing. In order to prevent or mitigate delay in commercialization, we are establishing a 5,500 square meter commercial API manufacturing facility in Cangzhou, China, and expect it to be operational shortly after NDA approval.

Market Segmentation

We believe DD-CKD market in China is readily addressable in the near term, and we believe roxadustat has the potential to deliver a compelling value proposition in particular to certain subgroups within DD-CKD: patients who are not treated to target Hb levels for any reason, patients who are hypo-responsive to ESAs, and patients on PD, which is performed at home. In addition, we believe that roxadustat, if approved, would have the potential to be the preferred anemia treatment for newly-initiated dialysis patients who have not been previously treated with ESA. With the expected expansion of Severe Disease reimbursement, we believe that the number of DD-CKD patients will increase steadily. We believe that it could require more than a decade for China to address the treatment gap between patients who need dialysis and those who are actually dialyzed.

If roxadustat is approved, we believe the Dialysis Eligible NDD-CKD population could represent another readily accessible and potentially new market segment for anemia therapy. There is an urgent and severe unmet medical need for these very sick patients, and the current low rate of treatment within this patient group could be addressed by an approved anemia treatment such as roxadustat. We view the Other NDD-CKD population as a longer term market opportunity where the potential number of patients could be substantial.

We believe the hospital-based nature of the China healthcare system is a very attractive feature of this market as it lends itself to rapid adoption of roxadustat within nephrology practices and across specialties, unlike in the U.S. where dialysis is performed separately at freestanding dialysis centers and CKD is treated at widely dispersed clinics and primary care offices across the country. In China, within nephrology, the same physicians care for dialysis, Dialysis Eligible NDD-CKD and Other NDD-CKD patients. Moreover, cardiologists and endocrinologists are located at the same hospitals as nephrologists, and prescriptions from all specialties are often filled at the same hospital pharmacy; as a result, the points of sale are highly concentrated.


As roxadustat is potentially a chronic use drug that addresses an unmet medical need and is intended to benefit large numbers of Chinese patients, we intend to apply for reimbursement by the Chinese government. Pricing for drugs sold without reimbursement is determined by the drug manufacturer, whereas pricing for drugs under reimbursement is negotiated with the government. We believe the compelling pharmaco-economic value proposition will support fair pricing for roxadustat.


We have entered into an agreement with AstraZeneca relating to roxadustat in China. Under the agreement, FibroGen Beijing will hold all of the regulatory licenses issued by China regulatory authorities at NDA approval, including the New Drug License, the Drug Approval Code, and the GMP License. During development, FibroGen is primarily responsible for regulatory, clinical and development-stage manufacturing activities. After NDA approval, FibroGen will be responsible for commercial manufacturing, pharmacovigilance, and medical affairs.

AstraZeneca will conduct commercialization activities as well as serve as the national distributor for roxadustat, sourcing the distribution of roxadustat to a network of regional and local distributors.

We believe that the collaboration will not only help to accelerate market access and patient adoption, but also reduce our risks associated with roxadustat launch in China, as AstraZeneca has significant experience with the China market and will be paying for launch-related commercialization costs in advance and recouping 50% of these expenses from initial roxadustat profits.

Planned Phase 4 Studies

The CFDA imposes a five-year monitoring surveillance period after NDA approval on all Class 1 innovative drugs like roxadustat. Based on current CFDA guidelines, we believe we will need to conduct a 2,000 subject post-approval safety study to demonstrate the long-term safety of roxadustat, as well as provide additional information related to the quality of the manufacturing process for roxadustat. The study design and patient size will be determined after Phase 3 data become available, in consultation with the CFDA as part of NDA review.



In Japan, Astellas has completed three of its six Phase 3 studies of roxadustat for the treatment of anemia in DD-CKD and NDD-CKD patients. Astellas anticipates topline results from its long-term ESA conversion Phase 3 study in patients on HD and its Phase 3 correction (ESA-naïve) study in the first quarter of 2018.

Phase 3 Study 1517-CL-0302: 24-Week Trial in Peritoneal Dialysis (PD) Patients in Japan

In 2017, Astellas and FibroGen reported results from Astellas’ multi-center, open-label Phase 3 study of roxadustat in PD CKD patients with anemia. This 24-week study enrolled a total of 56 PD patients, of whom 43 patients had previously received ESAs (ESA-conversion patients), and 13 patients had not previously received ESAs (ESA-naïve patients). Roxadustat was well tolerated and shown to correct hemoglobin levels in ESA-naïve patients and maintain hemoglobin levels within the target range in both ESA-conversion patients and ESA-naïve patients.

The Hb maintenance rate, as measured by the proportion of subjects with average Hb levels within the target Hb range of 10.0 g/dL to 12.0 g/dL for weeks 18 to 24, was 92% in ESA-naïve patients who were corrected from baseline hemoglobin levels and 74% in ESA-conversion patients. The preliminary safety analysis for this trial is consistent with the safety profile of roxadustat in previous clinical trials.

Japan Phase 2 Studies

In 2016 we, along with our partner, announced positive data from Astellas’ two Phase 2 studies in Japan of DD-CKD and NDD-CKD patients. Roxadustat was well tolerated and met the primary objective of demonstrating dose-related rates of Hb increase measured over the first six weeks of treatment, as well as anemia correction and Hb maintenance over the 24-week treatment period.

Study 1517-CL-0303: 24-Week Placebo-Controlled Correction in Japan NDD-CKD Patients

In this multi-center, randomized, parallel-group, placebo-controlled, double-blind study over 24 weeks, 107 NDD-CKD patients were randomized to one of three roxadustat treatment arms (50 mg, 70 mg, 100 mg) or to a placebo arm, with roxadustat orally administered TIW for the first six weeks of the study to evaluate dose response of efficacy and safety. This was followed by dose titration every four weeks until Hb response was achieved, at which point Hb was maintained with patients randomized to one of two dosing regimens (continuation of TIW dosing or a change to weekly dosing). Results showed achievement in the full analysis set of dose response in the three roxadustat treatment arms, with a mean rate of Hb increase of 0.200, 0.453, and 0.570 g/dL per week (50 mg, 70 mg, 100 mg, respectively), as measured over the first six weeks of the study, compared to a mean Hb decrease of 0.052 g/dL per week in subjects receiving placebo. Of note, 93.8% of roxadustat-treated subjects achieved Hb correction as measured by Hb response defined as Hb more than or equal to 10 g/dL and Hb increase of at least 1 g/dL from baseline. In the placebo arm, Hb response was achieved in 14.8% of the subjects. Results were presented at the American Society of Nephrology Kidney Week in November 2016.

Study 1517-CL-0304: 24-Week Trial in Japan DD-CKD Patients

This was a multi-center, randomized, parallel-group, darbepoetin-controlled, double-blind (roxadustat arms)/open-label (darbepoetin) study over 24 weeks in DD-CKD patients on chronic stable dialysis. The 120 subjects discontinued previous standard-of-care therapy (erythropoiesis-stimulating agents) to reach Hb levels of <9.5 g/dL and were then randomized to one of three roxadustat arms (administered orally TIW at a fixed dose) or to the darbepoetin arm (darbepoetin administered intravenously QW) over the first six weeks of the study to evaluate dose response of efficacy and safety, followed by dose titration to the desired Hb level every four weeks. During weeks 18 to 24, average Hb levels achieved (a secondary endpoint) in the full analysis set were 10.31 g/dL (1.33 g/dL Hb increase), 10.20 g/dL (1.37 g/dL Hb increase), and 10.53 g/dL (1.57 g/dL Hb increase), respectively, in the roxadustat treatment arms, compared to 10.25 g/dL (1.42 g/dL Hb increase) in the darbepoetin arm.

Status with Regulatory Agencies

In the last five years, we and our collaboration partners have had interactions with regulatory agencies in multiple territories regarding the planned development and potential path to approval of roxadustat.

We met with the FDA in May, June and July of 2014 to discuss the overall scope of our Phase 3 development program. In order to comply with FDA’s recommendation, we have designed and sized our Phase 3 program for, and have included MACE composite safety endpoints that we believe will be required for approval in the U.S. for all new anemia therapies.


We have also discussed our Phase 3 clinical development program with three National Health Authorities in the EU and obtained scientific advice from the European Medicines Agency, which was confirmed in writing in January 2014 with respect to the adequacy of our current clinical development program to support the indication for the treatment of anemia in NDD-CKD and DD-CKD patients. We expect the Marketing Authorization Application submission in Europe to precede our NDA filing in the U.S.

Between 2014 and 2016, we and our partner Astellas also met with the Pharmaceuticals and Medical Devices Agency and reached agreement on the Phase 3 program required for roxadustat for the treatment of DD-CKD anemia and NDD-CKD in Japan.

Investigational New Drug and Clinical Trial Applications

Roxadustat is being studied under one Investigational New Drug Application (“IND”), and several Clinical Trial Applications (“CTAs”), all with a specified indication of treatment of anemia in CKD. We originally submitted the IND in the U.S. to the FDA in April 2006. Our collaboration partner, Astellas, submitted the CTA in Japan to the Pharmaceuticals and Medical Devices Agency in June 2009. We and our collaboration partners Astellas and AstraZeneca have also submitted CTAs in Europe, Latin America, Canada, Russia, and Asia, beginning in 2013.


Based on roxadustat’s safety and efficacy profile to date and its mechanism of action, we believe that in addition to treating anemia in CKD, roxadustat has the potential to treat anemia associated with many other conditions, including MDS.

Background of Anemia in MDS

MDS are a group of disorders characterized by poorly formed or dysfunctional blood cells, leading to anemia in most cases. Anemia, a serious medical condition, is associated with increased risks of hospitalization, cardiovascular complications, need for blood transfusions, exacerbation of other serious medical conditions, and death, and frequently leads to significant fatigue, cognitive dysfunction, and decreased quality of life.

Incidence and prevalence of MDS are not yet well understood, and may be greatly underestimated. MDS diagnosis became reportable under the World Health Organization oncology classification system only in 2001. According to its latest available data, the National Cancer Institute estimates approximately 4.9/100,000 people in the U.S. general population were diagnosed with MDS annually for the period of 2007 to 2011, increased from 3.3/100,000 annually for 2001 to 2003. The incident rate increases with older populations, to 30.2/100,000 people among those 70 and 79 years of age, and further to 59.8/100,000 among those 80 years of age and older. The population-based registries are believed to have underestimated the incidence of MDS due to under diagnosis. It has been reported, using Medicare billing claims data, that the incidence of MDS in patients aged 65 years and older was approximately 162/100,000 as of 2003. The prevalence of MDS in the U.S. is estimated to be between 60,000 and 170,000, and continues to rise as more MDS therapies become available and patients are living longer with MDS.

In China, the incidence of MDS has been estimated to be 1.51/100,000 in the adult population. This is lower than those seen in other major markets such as the U.S., Western Europe, and Japan, and we believe that under diagnosis may be a contributing factor to the lower estimate in China.

Anemia is the most common clinical presentation in MDS, which results in red blood cell transfusions and related risks such as iron overload and significant impairment of the quality of life in affected patients. Dependency on red blood cell transfusions has also been associated with shorter life expectancy in patients with MDS.

Limitations of the Current Standard of Care for MDS and Anemia Associated with MDS

As a bone marrow disorder, anemia treatment options are limited and MDS patients often rely on repeated blood transfusions. Currently, there is no drug approved for the treatment of anemia in MDS patients in China or in the U.S., and transfusions are not readily accessible in China due to limited blood supply.

Stem cell transplantation is the only treatment that can cure MDS, but is available to only a small fraction of higher risk young MDS patients who are eligible for bone marrow transplant, for whom the treatment is often delayed until the disease progresses because of the known risks associated with transplantation itself and low success rates. This treatment option is unavailable to the majority of MDS patients who are older or who are deemed low risk.


There are very limited approved pharmacologic treatments for MDS. The FDA-approved treatments for MDS include the azanucleosides (HMA) 5-azacitidine and decitabine which are typically used for treating intermediate or higher risk patients (based on the International Prognostic Scoring System). These therapies are not approved for lower risk MDS patients because of their significant undesirable effects on bone marrow. These hypomethylating agents can achieve remission in a minority of the treated patients for a short duration of time before progression to acute myeloid leukemia, and are associated with significant levels of neutropenia and thrombocytopenia on top of those caused by MDS. Revlimid ® (lenalidomide) is approved in the U.S. and in the EU only for treating MDS patients with 5q (del) (a condition present in only 7% to 15% of MDS patients in whom a region of DNA has been deleted on one of the pair of chromosome 5 in the patient’s immature red blood cells), and has a 61% to 67% responder rate in this sub-population, along with significant side effects such as neutropenia (55% to 75%) and thrombocytopenia (41% to 44%).

While there are no approved therapies for anemia of MDS in the U.S., treatment guidelines recommend the use of ESAs to address anemia in lower risk MDS patients that have a low EPO level. ESA doses used for MDS anemia are generally five times the doses typically used for treating anemia in CKD patients, and the response rates are as low as 20% to 32% in lower risk MDS patients. Reasons for the high ESA dose requirement in MDS include inflammation contributing to ESA-resistance, which is associated with elevated hepcidin levels and functional iron deficiency, and the underlying progressive dysfunction of the bone marrow. Even patients who initially respond to ESAs generally will develop resistance to ESAs as their MDS progresses and become dependent on blood transfusions once they stop responding to ESAs.

Red blood cell transfusion is generally reserved for severe anemia in MDS patients. Red blood cell transfusions are usually used when Hb <9.0 g/dL or lower in the U.S. The hemoglobin threshold for red blood cell transfusion is as low as <6.0 g/dL in MDS patients in China, where transfusions are not readily accessible due to limited blood supply. Frequent red blood cell transfusions impose both financial and time burdens on patients and payors since frequent visits to the hospital for blood tests and transfusions are required. More importantly, transfusions are associated with risks of development of alloantibodies, which is related to the number of prior transfusions, and the transmission of infectious agents, which is a significant concern especially in MDS patients, some of whom are already immunocompromised due to neutropenia as part of their bone marrow dysplasia. Most notably, chronic red blood cell transfusions result in iron overload where iron can damage the heart, liver and other organs, as well as have a negative impact on clonal evolution and on hematopoietic stem cell therapy outcome. Iron overload from red blood cell transfusions is thought to be inhibitory on erythropoiesis and excess iron impacts hepcidin regulation causing a self-reinforcing feedback loop on erythropoiesis. For these reasons, patients with iron overload are treated with iron chelating agent in an attempt to reduce iron toxicity, but gastrointestinal and renal side effects lead to 1‑year discontinuation rate of iron chelator as high as 49%. Given these risk factors, red blood cell transfusion dependency is a strong non-favorable prognostic factor for survival of MDS patients.

The disease burden of anemia in MDS is high. Severe anemia interferes with MDS patients’ quality of life and their ability to work, in addition to imparting a negative effect on the function of other organ systems due to insufficient oxygen delivery to tissues. When red blood cell transfusions become necessary to sustain bodily functions, the risk of transfusion-related complications can further threaten MDS patients’ lives and well-being.

Our Solution

We believe there is a significant need for a safer, more effective, and more convenient approach to address anemia in patients with lower-risk MDS. Roxadustat, our orally administered small molecule HIF-PH inhibitor, stimulates the body’s natural mechanism of red blood cell production and iron hemostasis based on cellular-level oxygen-sensing and iron-regulation mechanisms. Roxadustat activates a coordinated erythropoietic response in the body that includes the stimulation of red blood cell progenitors, an increase in the body’s production of endogenous EPO, and an increase in iron availability for hemoglobin synthesis. Moreover, in anemia of CKD, roxadustat has demonstrated the ability in clinical trials to increase and maintain hemoglobin levels in the presence of inflammation as measured by C-reactive protein, where ESAs have shown limited effect. We believe that we may be able to replicate this result in MDS anemia patients, where inflammation is a significant contributing factor.

Clinical Development of Roxadustat in MDS

For the U.S. and Europe, we have begun enrolling patients for our Phase 3 clinical trial to evaluate the safety and efficacy of roxadustat for treatment of anemia in MDS. This is a global multi-center Phase 3 study in transfusion-dependent, lower risk MDS patients with up to 24 patients in the open-label, lead-in portion which precedes the 160 patient randomized, double-blind, placebo-controlled part of the study, in which subjects will be randomized 3:2 to receive roxadustat or placebo three-times-weekly for 28 weeks, with safety extension to one year. The primary endpoint is the proportion of patients who achieve transfusion independence.


In March 2017, we received approval from the CFDA on our Phase 2/3 MDS clinical trial application to evaluate roxadustat for the treatment of anemia in patients with MDS. This Phase 2/3 clinical trial will evaluate the safety and efficacy of roxadustat in non-transfusion dependent, lower risk MDS patients with anemia. The initial open-label portion of the study is expected to enroll up to 40 patients, with 135 patients planned for the randomized, double-blind, placebo-controlled Phase 3 portion of the study, in which subjects will be randomized 2:1 to receive roxadustat or placebo three-times weekly for 26 weeks. The primary endpoint for this study is percentage of patients with Hb response. This Phase 2/3 study is expected to initiate in the first half of 2018.

HIF-PH Inhibitor Platform

We have been a world leader in prolyl hydroxylase inhibition since the mid-nineties. Over the past two decades, we have built a robust drug discovery platform based on our deep understanding of the inhibition of prolyl hydroxylase enzymes using small molecules. Our platform is supported not only by internal research but also by numerous academic collaborations, including a long-standing funded collaboration with a research group at the University of Oulu, Finland, headed for many years by our scientific co-founder, Dr. Kari I. Kivirikko. Dr. Kivirikko is one of the world’s leading experts in collagen prolyl hydroxylases, and he remains an advisor to us.

Prior to the discovery of HIF regulation by prolyl hydroxylase activity, we had acquired compound collections from several pharmaceutical companies and assembled a diverse library of prolyl hydroxylase inhibitors to target collagen prolyl hydroxylase enzymes for fibrosis. Consequently, we were particularly well positioned to rapidly generate proof-of-concept for a number of aspects of HIF biology, and to direct medicinal chemistry efforts towards increasing potency and selectivity for the newly identified HIF-PH enzymes.

We have applied our expertise in the field of HIF-PH inhibition to develop an understanding, not only of the role of HIF in erythropoiesis, but also of other areas of HIF biology with important therapeutic implications. This consistent progression of discovery has led to findings relating to HIF-mediated effects associated with inflammatory pathways, various aspects of iron metabolism, insulin sensitivity and glucose and fat metabolism, neurological disease, and stroke. The extensive patent portfolio covering our discoveries represents an important competitive advantage.

The strength of our platform capitalizes on these internal discoveries, as well as some of the complexities of HIF biology that we and the scientific community have uncovered over the past decade. There are at least three different HIF-PH enzymes that are known to regulate the stability of HIF — these enzymes are commonly referred to in the scientific literature as PHD1, PHD2 and PHD3. Studies of genetically modified mice, in which the individual HIF-PH enzymes have been deleted, have revealed that PHD2 plays a major role in the regulation of erythropoiesis by HIF. In contrast, PHD1 and PHD3 appear to play less important roles in HIF-mediated erythropoiesis, but instead have been implicated in other important biological pathways.

We believe that inhibitors selectively targeting certain prolyl hydroxylases could have important therapeutic applications beyond anemia. For example, as PHD1 has been implicated in ischemic tissue injury, it has been proposed that PHD1 inhibitors may provide a novel therapeutic approach to protect organs and tissues from ischemic damage. PHD3 on the other hand has been implicated in insulin signaling, raising the possibility that PHD3 inhibitors may have therapeutic utility in the treatment of diabetes. Despite the challenges associated with selectively inhibiting just one enzyme from a closely related family, we have made important advances in the identification of selective HIF-PH inhibitors.

We currently have active research programs focused on exploring the therapeutic utility of selective prolyl hydroxylase inhibitors.


We were founded to discover and develop therapeutics for fibrosis. We began studying CTGF shortly after its discovery. Our ongoing internal research, efforts with collaboration partners and the work of other investigators have consistently demonstrated elevated CTGF levels in pathologic fibrotic conditions characterized by sustained production of extracellular matrix (“ECM”), elements that are key molecular components of fibrosis. Our accumulated discovery research efforts indicate that CTGF is a critical common element in the progression of serious diseases associated with fibrosis.

From our library of fully-human monoclonal antibodies that bind to different parts of the CTGF protein and block various aspects of CTGF biological activity, we selected pamrevlumab, for which we have exclusive worldwide rights. We believe that pamrevlumab blocks CTGF and inhibits its central role in causing diseases associated with fibrosis. Our data to date indicate that pamrevlumab is a promising and highly differentiated product with broad potential to treat a number of fibrotic diseases and cancers. We are currently conducting Phase 2 trials in pancreatic cancer and DMD. In 2017, we completed the double-blind and comparator portions of our Phase 2 trial in IPF. Pamrevlumab has received orphan drug designation in IPF in the U.S.


Based on its ability to block CTGF, pamrevlumab may be a treatment for a broad array of fibrotic disorders of nearly every organ system. In animal studies, such as radiation-induced pulmonary fibrosis in mice, we have demonstrated that pamrevlumab is capable of reversing fibrosis. In clinical trials, we have used advanced medical imaging technology to quantify changes in fibrosis throughout the lungs. Our data to date using these measures demonstrate that pamrevlumab may stabilize and in some instances reverse pulmonary fibrosis and improve pulmonary function in IPF patients.

Certain cancers have a prominent ECM component that contributes to metastasis and progressive disease. Specifically, ECM is the connective tissue framework of an organ or tissue; all tumors have ECM. In the case of fibrotic tumors, ECM is more pronounced and there is more fibrosis than in other tumor types. In mouse models of pancreatic cancer, pamrevlumab treatment has demonstrated reduction of tumor mass, slowing of metastasis and improvement in survival. In an open-label Phase 2 study of pamrevlumab plus gemcitabine and erlotinib, pamrevlumab demonstrated a dose-dependent improvement in one year survival rate. We are also currently conducting a randomized, active-control, neoadjuvant clinical trial combining pamrevlumab with nab-paclitaxel plus gemcitabine in patients with locally advanced pancreatic cancer.

DMD is an inherited disorder of the dystrophin gene that leads to progressive muscle loss and results in early death due to pulmonary or cardiac failure. Numerous pre-clinical studies including those in the mdx model of DMD suggest that CTGF contributes to the process by which muscle is replaced by fibrosis and fat and that CTGF may also impair muscle cell differentiation during muscle repair after injury. pamrevlumab treatment has improved muscle strength and exercise endurance in the mdx model of DMD. In December 2015, we began an open label single arm trial in non-ambulatory boys with DMD.

Results to date indicate that pamrevlumab has broad potential to address unmet needs for the treatment of fibrotic diseases and cancers. Specifically, given our preclinical and clinical data to date, our primary focus for clinical development of pamrevlumab is in IPF, pancreatic cancer and DMD.

Overview of Fibrosis

Fibrosis is an aberrant response of the body to tissue injury that may be caused by trauma, inflammation, infection, cell injury, or cancer. The normal response to injury involves the activation of cells that produce collagen and other components of the ECM that are part of the healing process. This healing process helps to fill in tissue voids created by the injury or damage, segregate infections or cancer, and provide strength to the recovering tissue. Under normal circumstances, where the cause of the tissue injury is limited, the scarring process is self-limited and the scar resolves to approximate normal tissue architecture. However, in certain disease states, this process is prolonged and excessive and results in progressive tissue scarring, or fibrosis, which can cause organ dysfunction and failure as well as, in the case of certain cancers, promote cancer progression.

Excess CTGF Causes Fibrosis. Pamrevlumab Blocks CTGF and Can Reverse Fibrosis


Excess CTGF levels are associated with fibrosis. CTGF increases the abundance of myofibroblasts, a cell type that drives wound healing, and stimulates them to deposit ECM proteins such as collagen at the site of tissue injury. In the case of normal healing of a limited tissue injury, myofibroblasts eventually die by programmed cell death, or apoptosis, and the fibrous scarring process recedes. In fibrotic conditions, excess CTGF results in chronic activation of myofibroblasts, which leads to chronic ECM deposition and fibrosis (refer to figure above).

Multiple biological agents and pathways have been implicated in the fibrotic process (Wynn J Pathol (2008)). Many fibrosis pathways converge on CTGF (refer to figure below), which the scientific literature demonstrates to be a central mediator of fibrosis (Oliver et al., J Inv Derm (2010)). In the case of cancer, the sustained tumor-associated fibrotic tissue promotes tumor cell survival and metastasis. The figure below shows the commonality of cellular mechanisms that may result in fibrosis and cancer.

Most Biological Factors Implicated in Fibrosis Work Through CTGF

CTGF is a secreted glycoprotein produced by fibroblasts, endothelium, mesangial cells and other cell types, including cancers, and is induced by a variety of regulatory modulators, including TGF-ß and VEGF. CTGF expression has been demonstrated to be up-regulated in fibrotic tissues. Thus, we believe that targeting CTGF to block or inhibit its activity could stop or reverse tissue fibrosis. In addition, since CTGF is implicated in nearly all forms of fibrosis, we believe pamrevlumab has the potential to provide clinical benefit in a wide range of clinical indications that are characterized by fibrosis.

Until recently, it was believed that fibrosis was an irreversible process. It is now generally understood that the process is dynamic and potentially amenable to reversal. Based on studies in animal models of fibrosis of the liver, kidney, muscle and cardiovascular system, it has been shown that fibrosis can be reversed. It has also been demonstrated in humans that fibrosis caused by hepatitis virus can be reversed (Chang et al. Hepatology (2010)). Additionally, we have generated data in human and animal studies that lung fibrosis can be reversed in some instances upon treatment with pamrevlumab. We do not believe that there is clinical evidence that therapies currently on the market directly prevent or reverse fibrosis in IPF. While certain other companies are working on topical inhibition of CTGF, we are not aware of other products in development that target CTGF inhibition for deep organ fibrosis and cancer.


Clinical Development of Pamrevlumab — Overview

We have performed clinical trials of pamrevlumab in IPF, pancreatic cancer, liver fibrosis and diabetic kidney disease. In eleven Phase 1 and Phase 2 clinical studies involving pamrevlumab to date, including more than 450 patients who were treated with pamrevlumab (about half of patients dosed for more than six months), pamrevlumab has been well-tolerated across the range of doses studied, and there have been no dose-limiting toxicities seen thus far.

In IPF, we completed the double-blind portion and the comparator sub-study of our randomized, placebo-controlled Phase 2 trial of pamrevlumab for first-line treatment of IPF in patients with mild-to-moderate disease. The sub-study examined the safety of pamrevlumab in combination with approved therapies. We reported topline pulmonary function and safety data from the study in the third quarter of 2017 and presented these results at the 2017 European Respiratory Society meeting.

In pancreatic cancer, we are currently conducting a randomized, active-control, neoadjuvant Phase 2 trial combining pamrevlumab with nab-paclitaxel plus gemcitabine in 37 patients with locally advanced pancreatic cancer. Interim results were reported at ASCO 2017 Gastrointestinal Cancers Symposium in San Francisco, showing an improvement in survival among patients in the combination arm, as compared to chemotherapy alone. In addition, a greater proportion of subjects treated on the combination arm containing pamrevlumab were converted from unresectable to fully resectable status. We completed enrollment in the first half of 2017 and completed the six-month treatment period at the end of 2017. Previously we performed an open-label, dose-finding Phase 2 trial in a total of 75 patients with advanced pancreatic cancer.

We are continuing to enroll patients in an exploratory single arm trial of the safety and efficacy of pamrevlumab in non-ambulatory subjects with DMD. The primary endpoint is change in forced vital capacity; other endpoints include changes in arm function and in muscle and heart fibrosis.

Actual dates depend on a variety of factors and are subject to numerous risks and uncertainties, including with respect to patient enrollment, safety results, manufacturing, third party contractors, and government regulators, some of which are out of our control. Also refer to “Risk Factors,” and particularly those risk factors under the heading “Risks Related to the Development and Commercialization of Our Product Candidates.


The table below provides a summary of our clinical trials involving pamrevlumab.

Completed and Ongoing Pamrevlumab Clinical Trials


Study, Study #















Phase 1—IPF, FGCL-MC3019-002


Open-label, dose-



1, 3, or 10







Phase 2—IPF, FGCL-3019-049


Open-label, dose-



15 or 30


Every 3 weeks


45 weeks



Phase 2—IPF, FGCL-3019-067







30 mg/kg


Every 3 weeks


45 weeks



                         '067 Sub-study







30 mg/kg


Every 3 weeks


24 weeks



Phase 1/2 —Pancreatic Cancer, FGCL-MC3019-



Open-label, dose-



3, 10, 15,

25, 35, or


17.5 or



Every other week weekly


Until disease


1 to 89




Phase 1/2—Pancreatic Cancer, FGCL-3019-069



active control





Cycle 1 = Days 1, 8

and 15


Cycles =

Every other week


24 weeks



Phase 2—Liver Fibrosis, due to HBV, FGCL-3019-








15 or 45


Every 3 weeks


45 weeks



Phase 2 – Duchenne muscular dystrophy, non-ambulatory FGCL-3019-079


Open-label, single arm




Every 2 weeks


45 weeks


Target 22*

Phase 1—Diabetic Kidney Disease, FGCL-



Open-label, dose-



3 or 10


Days 0, 14, 28 and 42


6 weeks



Phase 2—Diabetic Kidney Disease, FGCL-








5 or 10


Every 2 weeks Every 4 weeks


12 weeks

12 weeks



Phase 2—Diabetic Kidney Disease, FGCL-







3 or 10


Every 2 weeks


26 weeks



Phase 1—Focal Segmental Glomerular Sclerosis, FGCL-



Open-label, single arm




Every 2 weeks


8 weeks




Currently enrolling.


Idiopathic Pulmonary Fibrosis

Understanding IPF and the Limitations of Current Therapies

IPF is a form of progressive pulmonary fibrosis, or abnormal scarring, which destroys the structure and function of the lungs. As tissue scarring progresses in the lungs, transfer of oxygen into the bloodstream is increasingly impaired. Average life expectancy at the time of confirmed diagnosis of IPF is estimated to be between three to five years, with approximately two-thirds of patients dying within five years of diagnosis. Thus, the survival rates are comparable to some of the most deadly cancers. The cause of IPF is unknown but is believed to be related to unregulated cycles of injury, inflammation and fibrosis.

Patients with IPF experience debilitating symptoms, including shortness of breath and difficulty performing routine functions, such as walking and talking. Other symptoms include chronic dry, hacking cough, fatigue, weakness, discomfort in the chest, loss of appetite, and weight loss. Over the last decade, refinements in diagnosis criteria and enhancements in high-resolution computed tomography imaging technology (“quantitative HRCT”) have enabled more reliable diagnosis of IPF without the need for a lung biopsy more clear distinction from other interstitial lung diseases.

The U.S. prevalence and incidence of IPF are estimated to be 44,000 to 135,000 cases, and 21,000 new cases per year, respectively, based on Raghu et al. (Am J Respir Crit Care Med (2006)) and on data from the United Nations Population Division. We believe that with the availability of technology to enable more accurate diagnoses, the number of individuals diagnosed per year with IPF will continue to increase. In 2011, Decision Resources Group estimated that there will be approximately $4.6 billion in sales of IPF drugs in the U.S. and Europe in 2020.

There are currently two therapies approved to treat IPF in Europe and the U.S., pirfenidone and nintedanib. The approvals and subsequent launches of pirfenidone and nintedanib have clearly shown the commercial potential in IPF. Hoffmann-La Roche (“Roche”) reported worldwide sales of pirfenidone for 2016 of approximately $770 million, and approximately $930 million for 2017. Similarly, Boehringer Ingelheim Pharma GmbH & Co. KG (“Boehringer Ingelheim”) reported total sales of approximately $660 million for nintedanib in 2016, and approximately $527 million in the first half of 2017.

We believe that pamrevlumab has the potential to have a measurable and significant effect on lung fibrosis and if approved, improve the prognosis for patients with IPF. We expect to receive feedback from the FDA on our Phase 3 plan in IPF in mid-2018.

Study 067 – Randomized, Double-Blind, Placebo-Controlled Phase 2 Trial of Pamrevlumab in IPF

In August 2017, we reported positive topline results from our randomized, double-blind, placebo-controlled Phase 2 clinical trial (Study 067) designed to evaluate the safety and efficacy of pamrevlumab in patients with mild-to-moderate IPF (baseline forced vital capacity (“FVC”) percentage predicted between 55% and 90%), as well as topline results from two sub-studies that were added to evaluate the safety of combining pamrevlumab with recently approved IPF therapies.

In the double-blind, placebo-controlled 48-week portion of this study, one hundred-three (103) patients were randomized (1:1) to receive either 30mg/kg of pamrevlumab or placebo intravenously every three weeks. Lung function assessments were conducted at baseline and at weeks 12, 24, 36 and 48. Quantitative HRCT assessments were performed at baseline and on weeks 24 and 48.

Pamrevlumab met the primary efficacy endpoint of change of FVC percent predicted, a measure of a patient’s lung volume as a percentage of what would be expected for such patient’s age, race, sex and height. The average decline (least squares mean) in FVC percent predicted from baseline to week 48 was 2.85 in the pamrevlumab arm (n=50) as compared to an average decline of 7.17 in the placebo arm (n=51), a statistically significant difference of 4.33 (p=0.0331, using a linear slope analysis in the Intent to Treat (“ITT”) population).

Pamrevlumab-treated patients had an average decrease (least squares mean) in FVC of 129 ml at week 48 compared to an average decrease of 308 ml in patients receiving placebo, a statistically significant difference of 178 ml (p=0.0249, using a linear slope analysis in the ITT population). This represents a 57.9% relative difference. In addition, the pamrevlumab-treated arm had a lower proportion of patients (10%) who experienced disease progression (defined by a decline in FVC percent predicted of greater than or equal to 10%) or death, than did the placebo arm (31.4%) at week 48 (p=0.0103). The percentage of pamrevlumab patients who experienced disease progression and discontinued therapy was less than 15% of that in the placebo arm.

In this study, we measured change in quantitative lung fibrosis from baseline to week 24 and week 48 using quantitative HRCT. This analysis showed a statistically significant attenuation of lung fibrosis in the pamrevlumab-treated group as compared to the placebo-treated group at week 24 and week 48. Missing or unreadable 24 or 48-week data were imputed using the multiple imputation method. As in our previous open label Phase 2 study, a correlation between FVC percent predicted and quantitative lung fibrosis was confirmed at both week 24 and 48 in this study.


We are not aware of any IPF therapies that have shown a statistically significant effect on lung fibrosis as measured by quantitative HRCT analysis.

The treatment effects of pamrevlumab were demonstrated not only on change in FVC, a measure of pulmonary function and IPF disease progression, and change in fibrosis using quantitative HRCT, but pamrevlumab treated patients also showed a trend of clinically meaningful improvement in a measure of health-related quality of life using the St. George’s Respiratory Questionnaire (SGRQ) vs a reduction in quality of life seen in placebo patients over the 48 weeks of treatment. The SGRQ quality of life measurement has been validated in chronic obstructive pulmonary disease. In the patients that were evaluated by the UCSD Shortness of Breath Questionnaire, pamrevlumab treated patients had a significant attenuation of their worsening dyspnea in comparison to placebo.

Pamrevlumab was well tolerated in the placebo-controlled study. The TEAEs were comparable between the pamrevlumab and placebo arms and the adverse events in the pamrevlumab arm were consistent with the known safety profile of pamrevlumab. In this study, as compared with the placebo group, fewer pamrevlumab patients were hospitalized, following an IPF-related or respiratory TEAE, or died for any reason.

The double-blind, active-controlled combination sub-studies were designed to assess the safety of combining pamrevlumab with standard of care medication in IPF patients. Study subjects were on stable doses of pirfenidone or nintedanib for at least three months and were randomized 2:1 to receive 30 mg/kg of pamrevlumab or placebo every three weeks for 24 weeks. Thirty-six patients were enrolled in the pirfenidone sub-study and 21 patients were enrolled in the nintedanib sub-study. Pamrevlumab appeared to be well tolerated when given in combination with either pirfenidone or nintedanib.

Study 049 – Open-Label Phase 2 Trial of Pamrevlumab in IPF

We completed the open-label extension of Study 049, a Phase 2 open-label, dose-escalation study to evaluate the safety, tolerability, and efficacy of pamrevlumab in 89 patients with IPF. During the initial one-year treatment period, pamrevlumab was administered at a dose of 15 mg/kg in Cohort 1 (53 patients) and 30 mg/kg in Cohort 2 (36 patients) by IV infusion every three weeks for 45 weeks. After 45 weeks of dosing, subjects whose FVC declined less than predicted were allowed to continue dosing in an extension study until they had disease progression. Nineteen patients from Cohort 1 (35.8%) and 18 patients from Cohort 2 (50.0%) entered the extension study. Efficacy endpoints were pulmonary function assessments, extent of pulmonary fibrosis as measured by quantitative imaging and measures of health-related quality of life. We presented data from our open-label Phase 2 IPF extension study (049) at the International Colloquium on Lung and Airway Fibrosis in November 2016, reporting that no safety issues were observed during prolonged treatment with pamrevlumab. Some of the 37 patients who enrolled in the extension study were treated with pamrevlumab for up to five years. Trends regarding improved or stable pulmonary function and stable fibrosis observed during the initial one-year study were also observed in the extension study.

HRCT is typically used to diagnose IPF based on visual assessments of computed tomography (“CT”), images of lung fibrosis. We used quantitative HRCT to measure changes in fibrosis in this Study 049 using software to quantify whole lung fibrosis from the compilation of 1 mm HRCT sections of the entire lung. The computer algorithm, which has been validated by the clinical research organization used for the study, provides an overall determination of the percentage of the lung that contains individually the three characteristic forms of IPF fibrosis, including reticular IPF fibrosis which is expected to make the most dynamic contribution to overall lung fibrosis.

The extent of lung fibrosis as measured by quantitative HRCT has been shown to be accurate and reproducible (Kim et al. Eur Radiol (2011)). Recent publications based on similar quantitative HRCT methods have identified an association between worsening pulmonary fibrosis and mortality in IPF (Maldonado et al. Eur Resp J (2014); Oda et al. Respiratory Research (2014)). However, HRCT has not been used by the FDA to establish efficacy in IPF.

Eighty-nine patients in this Phase 2 open label study received at least one dose of pamrevlumab. We defined disease severity in terms of baseline pulmonary function, measured as the FVC percent of the predicted value for a healthy matched person of the same age, or FVC percent predicted. Severe disease was FVC percent predicted < 55%, moderate disease was FVC percent predicted between 55% and 80%, and mild disease was FVC percent predicted >80%.

In Cohort 1, we enrolled patients with a wide range of disease severity to assess safety and efficacy. Baseline FVC percent predicted for Cohort 1 was 43% to 90%, with a mean of 62.8%. In contrast, other IPF clinical trials, such as those for pirfenidone and nintedanib, have enrolled patients who on average had mild to moderate disease (mean FVC percent predicted 73.1% to 85.5%). Fourteen patients in Cohort 1 withdrew, and ten of the 14 had severe disease.


In order to enroll IPF patients similar to those in other IPF trials, we amended the protocol for Cohort 2 to include only patients with mild to moderate disease (FVC 55% predicted). Baseline FVC percent predicted for Cohort 2 was 53% to 112%, with a mean of 72.7%. Based on this definition of disease severity, 37 patients in Cohort 1 and 32 patients in Cohort 2 had mild to moderate disease.

Disease Severity in Enrolled and Evaluated Patients Treated with Pamrevlumab in FGCL-3019-049






Cohort 1







Cohort 2





































FVC % Predicted





55% to 80%













55% to 80%





































































































































































The table below provides a summary of the observed quantitative change in fibrosis for mild to moderate patients in Cohorts 1 and 2 as measured by quantitative HRCT. Twenty-four percent of these patients had improved fibrosis at week 48. We believe that this is the first trial to demonstrate reversal of fibrosis in a subset of IPF patients. Stable fibrosis has been considered the only achievable favorable outcome in IPF. The table below sets forth the number of patients who showed stable or improved fibrosis at weeks 24 and 48 compared to the amount of fibrosis at the start of the trial.

Changes in Fibrosis in Patients with Mild to Moderate IPF Treated with Pamrevlumab in FGCL-3019-049




Stable or Improved

Compared to Baseline


Improved Compared to



Improved Compared

to Week 24



Week 24


Week 48


Week 24


Week 48


Week 48

Cohort 1











Cohort 2























Fibrosis improvement or stabilization in patients with mild to moderate disease as measured as reticular fibrosis by quantitative HRCT correlated with improvement or stabilization of pulmonary function measured by FVC (p<0.0001; r=-0.59 Cohorts 1 and 2 combined). The figure below shows FVC changes up to week 48 for mild to moderate patients with stable or improved fibrosis versus patients with worsening fibrosis. Patients with stable or improved fibrosis showed improved pulmonary function, on average, which was significantly different or better than patients with worsening fibrosis who showed a substantial decline in FVC (p= 0.0001, Cohorts 1 and 2 combined). Patients with worsening fibrosis had pulmonary function that was similar to the annual decline in pulmonary function for typical IPF patients.

Categorical Analysis of FVC Change from Baseline (BL) (mean ±SE) in FGCL-3019-049


Eighty-nine patients had at least one adverse event. The most common reported events were cough, fatigue, shortness of breath, upper respiratory tract infection, sore throat, bronchitis, nausea, dizziness, and urinary tract infection. To date, including the open-label extension, there have been 45 SAEs in 31 patients, four of which were considered possibly related by the principal investigator to study treatment. During the first year of treatment there were 38 TSAEs in 24 patients. Adverse events observed to date are consistent with typical conditions observed in this patient population.

In aggregate, the data from the Phase 2 open-label, dose-escalation study indicate that a subset of pamrevlumab treated IPF patients experienced improvements in lung fibrosis with commensurate improvement in pulmonary function and a potential for prolonged benefit with continued treatment. These results are consistent with the mouse disease model results which showed that pamrevlumab treatment can reverse lung fibrosis and result in improved pulmonary function. We believe that our patient data showing correlated improvements in both fibrosis and lung function in some patients have not been seen in previously published IPF clinical studies.

Open-Label Phase 1 Trial of Pamrevlumab in IPF

Study 002 was a Phase 1 open-label study to determine the safety and pharmacokinetics of escalating single doses of pamrevlumab. Patients with a diagnosis of IPF by clinical features and surgical lung biopsy received a single IV dose of pamrevlumab at 1, 3, or 10 mg/kg. A total of 21 patients were enrolled in the study; six patients received a dose of 1 mg/kg, nine patients received 3 mg/kg, and six patients received 10 mg/kg. Pamrevlumab was well tolerated across the range of doses studied; and there were no dose-limiting toxicities. TEAE that were considered to be possibly related by the principal investigator to pamrevlumab were mild and self-limited, consisting of pyrexia, cough and headache.

Pancreatic Cancer

Understanding Pancreatic Cancer and the Limitations of Current Therapies

Pancreatic ductal adenocarcinoma, or pancreatic cancer, is the fourth leading cause of cancer deaths in the U.S. According to the World Health Organization, and based on data from the United Nations Population Division, there were approximately 79,000 new cases of pancreatic cancer and approximately 78,000 deaths in the EU in 2012. The National Cancer Center of Japan estimated that in 2010 (latest year available) there were 32,330 new cases of pancreatic cancer. In 2013, Decision Resources Group estimated that there will be approximately $1.3 billion in sales of pancreatic cancer drugs in 2022. According to the U.S. National Cancer Institute, in 2016, there were approximately 53,000 new cases of pancreatic cancer projected in the U.S. Fifty percent of new cases are metastatic. Another 15-20% have localized resectable tumors. The remaining 30-35% have localized but unresectable tumors. For those with non-resectable tumors, median survival is eight to 12 months post-diagnosis, and about 7% realize five years of survival; similar to metastatic cases. For those with resectable tumors, 50% survive 17 to 27 months post-diagnosis and ~20% report five-year survival.

Pancreatic cancer is aggressive and typically not diagnosed until it is largely incurable. Most patients are diagnosed after the age of 45, and according to the American Cancer Society, 94% of patients die within five years from diagnosis. The majority of patients are treated with chemotherapy, but pancreatic cancer is highly resistant to chemotherapy. Approximately 15% to 20% of patients are treated with surgery; however, even for those with successful surgical resection, the median survival is approximately two years, with a five year survival rate of 15% to 20% (Neesse et al. Gut (2011)). Radiation treatment may be used for locally advanced diseases, but it is not curative.

The duration of effect of approved anti-cancer agents to treat pancreatic cancer is limited. Gemcitabine demonstrated improvement in median overall survival from approximately four to six months, and erlotinib in combination with gemcitabine demonstrated an additional ten days of survival. Nab-paclitaxel in combination with gemcitabine was approved by the FDA in 2013 for the treatment of pancreatic cancer, having demonstrated median survival of 8.5 months. The combination of folinic acid, 5-fluorouracil, irinotecan and oxaliplatin (FOLFIRINOX) was reported to increase survival to 11.1 months from 6.8 months with gemcitabine. These drugs illustrate that progress in treatment for pancreatic cancer has been modest, and there remains a need for substantial improvement in patient survival and quality of life.

The approved chemotherapeutic treatments for pancreatic cancer target the cancer cells themselves. Tumors are composed of cancer cells and associated non-cancer tissue, or stroma, of which ECM is a major component. In certain cancers such as pancreatic cancer, both the stroma and tumor cells produce CTGF which in turn promotes the proliferation and survival of stromal and tumor cells. CTGF also induces ECM deposition that provides advantageous conditions for tumor cell adherence and proliferation, promotes blood vessel formation, or angiogenesis, and promotes metastasis, or tumor cell migration, to other parts of the body.

Pancreatic cancers are generally resistant to powerful chemotherapeutic agents, and there is now growing interest in the use of an anti-fibrotic agent to diminish the supportive role of stroma in tumor cell growth and metastasis. The anti-tumor effects observed with pamrevlumab in preclinical models indicate that it has the potential to inhibit tumor expansion through effects on tumor cell proliferation and apoptosis as well as reduce metastasis.


Clinical Development of Pamrevlumab in Pancreatic Cancer

We continue to follow patients in our ongoing open-label, randomized (2:1) Phase 1/2 trial (FGC004C-3019-069) of pamrevlumab combined with gemcitabine plus nab-paclitaxel chemotherapy versus the chemotherapy regimen alone in patients with inoperable locally advanced pancreatic cancer that has not been previously treated. We enrolled 37 patients in this study and completed the six-month treatment period and surgical assessment at the end of 2017. The overall goal of the trial is to determine whether the pamrevlumab combination can convert inoperable pancreatic cancer to operable, or resectable, cancer. Tumor removal is the only chance for cure of pancreatic cancer, but only approximately 15% to 20% of patients are eligible for surgery.

We have continued to see a majority of pamrevlumab treated patients converted from unresectable to resectable cancer, consistent with what we reported at the 2017 American Society of Clinical Oncology GastroIntestinal Cancer Meeting (“ASCO-GI”). The patients continue to be followed for disease progression and overall survival. Our goal is to reach agreement with the FDA on a pivotal trial design by mid-2018.

We reported on 23 evaluable patients at the 2017 ASCO-GI. Of the 22 patients randomized to pamrevlumab plus standard of care (gemcitabine and nab-paclitaxel), 10 continue on treatment, three discontinued therapy due to SAEs unrelated to study drug and seven were reassessed as eligible for resection based on standard scoring criteria set forth in the protocol; three having complete resection (R0) and one having an R1 resection (microscopic evidence of residual tumor cells at the resection margins), while the remaining patients’ tumors were not resected due to the presence of metastatic disease observed during surgery. Of the eleven patients randomized to gemcitabine and nab-paclitaxel alone, five experienced progressive disease, as defined in the protocol, prior to completing treatment, five remained inoperable and one was converted to operable cancer having an R0 resection. After six cycles of treatment, plasma levels of CA19.9, a non-specific tumor marker, showed a mean reduction of 78.3% with pamrevlumab plus chemotherapy compared to 48.7% with chemotherapy alone. In addition, this interim data showed improvement in survival among patients in the combination arm, as compared to chemotherapy alone. Patients with locally advanced unresectable pancreatic cancer have median survival of less than 12 months, only slightly better than patients with metastatic pancreatic cancer, whereas patients with resectable pancreatic cancer have a much better prognosis with median survival of approximately 23 months and some patients being cured. If pamrevlumab in combination with chemotherapy continues to demonstrate an enhanced rate of conversion from unresectable cancer to resectable cancer, it may support the possibility that pamrevlumab could provide a substantial survival benefit for locally advanced pancreatic cancer patients.

Completed Clinical Trials of Pamrevlumab in Pancreatic Cancer

We completed an open-label Phase 1/2 (FGCL-MC3019-028) dose finding trial of pamrevlumab combined with gemcitabine plus erlotinib in patients with previously untreated locally advanced (stage 3) or metastatic (stage 4) pancreatic cancer. These study results were published in the Journal of Cancer Clinical Trials (Picozzi et al., J Cancer Clin Trials 2017, 2:123). The trial tested pamrevlumab doses of 3 mg/kg, 10 mg/kg, 15 mg/kg, 25 mg/kg, 35 mg/kg and 45 mg/kg administered every two weeks, and pamrevlumab doses of 17.5 mg/kg and 22.5 mg/kg administered weekly after a double loading dose. On Day 15, treatment began with gemcitabine 1000 mg/m 2 weekly for three weeks in a four week cycle and erlotinib 100 mg daily. Treatment continued until progression of the cancer or the patient withdrew for other reasons. Patients were then followed until death. Tumor status was evaluated by CT imaging every eight weeks until disease progression to assess changes in tumor mass.


Seventy-five patients were enrolled in this study with 66 (88%) having stage 4 metastatic cancer. The study demonstrated a dose-related increase in survival, as described in the figure below. At the lowest doses, no patients survived for even one year while at the highest doses up to 31% of patients survived one year.

Effect of Pamrevlumab Dose on One Year Survival in Pancreatic Cancer


QW = weekly; Q2W = twice weekly

A post-hoc analysis found that there was a significant relationship between survival and trough levels of plasma pamrevlumab measured immediately before the second dose (Cmin), as illustrated below. Cmin greater than or equal to 150 µg/mL was associated with significantly improved progression-free survival (p=0.01) and overall survival (p=0.03) versus those patients with Cmin less than 150 µg/mL. For patients with Cmin >150 µg/mL median survival was 9.0 months compared to median survival of 4.4 months for patients with Cmin <150 µg/mL. Similarly, 34.2% of patients with Cmin >150 µg/mL survived for longer than one year compared to 10.8% for patients with Cmin <150 µg/mL. These data suggest that sufficient blockade of CTGF requires pamrevlumab threshold blood levels of approximately 150 µg/mL in order to improve survival in patients with advanced pancreatic cancer.

Increased Pancreatic Cancer Survival Associated with Increased Plasma Levels of Pamrevlumab


The Kaplan-Meier plot provides a representation of survival of all patients in the clinical trial. Each vertical drop in the curve represents a recorded event (death) of one or more patients. When a patient’s event cannot be determined either because he or she has withdrawn from the study or because the analysis is completed before the event has occurred, that patient is “censored” and denoted by a symbol () on the curve at the time of the last reliable assessment of that patient.

In the study, the majority of adverse events were mild to moderate, and were consistent with those observed for erlotinib plus gemcitabine treatment without pamrevlumab. There were 99 TSAEs; six of which were assessed as possibly related by the principal investigator, and 93 as not related to study treatment. We did not identify any evolving dose-dependent pattern, and higher doses of pamrevlumab were not associated with higher numbers of SAEs or greater severity of the SAEs observed.

In both the KPC mouse study and in this clinical trial, pamrevlumab treatment had a substantial effect on survival with no apparent increase to the toxicity of the chemotherapeutic regimen.

Pamrevlumab for Duchenne Muscular Dystrophy

Understanding DMD and the Limitations of Current Therapies

In the U.S., one in 3,500 boys have DMD, and there are currently no approved disease-modifying treatments. Most children, despite taking steroids to mitigate progressive muscle loss, are wheelchair bound by age 12, and median survival is age 25. DMD is caused by absence of the dystrophin protein resulting in abnormal muscle structure and function and buildup of fibrosis in muscle, leading to diminished mobility, pulmonary function and cardiac function. Constant myofiber breakdown results in persistent activation of myofibroblasts and altered production of ECM resulting in extensive fibrosis in skeletal muscles of DMD patients. Desguerre et al. (2009) showed that muscle fibrosis was the only myo-pathologic parameter that significantly correlated with poor motor outcome as assessed by quadriceps muscle strength, manual muscle testing of upper and lower limbs, and age at ambulation loss.

Clinical Development of Pamrevlumab for DMD

We continue to enroll patients in our Phase 2 open-label trial of pamrevlumab in non-ambulatory DMD patients. We have enrolled 18 of the 22 patients targeted for enrollment in this study. The primary endpoint is change in pulmonary function compared to each individual subject’s historical decline in lung function. Other endpoints include assessments of cardiac fibrosis and function assessed by magnetic resonance imaging (“MRI”), arm muscle fibrosis and fat assessed by MRI and upper body strength. We have amended our protocol and inclusion criteria with the aim of increasing enrollment.

Other Potential Indications for Pamrevlumab

We believe that pamrevlumab has potential to be a treatment for cancers and a broad array of fibrotic disorders, including:


Cancers — melanoma, breast cancer, hepatoma


Liver — non-alcoholic steatohepatitis


Lung — scleroderma lung disease


Radiation induced fibrosis


Muscular dystrophies other than DMD


Kidney — diabetic nephropathy, focal segmental glomerular sclerosis


Cardiovascular system — congestive heart failure, pulmonary arterial hypertension

Investigational New Drug and Clinical Trial Applications

Pamrevlumab is being studied in the U.S. for the treatment of IPF under an IND that we submitted to the FDA in August 2003. Pamrevlumab is being studied in the U.S. for the treatment of locally advanced or metastatic pancreatic cancer under an IND that we submitted to the FDA in September 2004. Pamrevlumab is being studied in the U.S. for the treatment of DMD under an IND that we submitted to the FDA in June 2015.


Commercialization Strategy for Pamrevlumab

Our goal, if pamrevlumab is successful, is to be a leader in the development and commercialization of novel approaches for inhibiting deep organ fibrosis and treating some forms of cancer. To date, we have retained exclusive worldwide rights for pamrevlumab. We plan to retain commercial rights to pamrevlumab in North America and will also continue to evaluate the opportunities to establish co-development partnerships for pamrevlumab as well as commercialization collaborations for territories outside of North America.


Corneal blindness, defined as visual acuity of 3/60 or less, is caused by various factors, including scarring resulting from infections, such as herpes simplex, physical trauma, chemical injury and genetic diseases affecting the function of the cornea. In countries with sufficient tissue banks and skilled surgeons, the treatment for corneal blindness is the replacement of the damaged cornea with a corneal graft from donor corneas from human cadavers. Despite use of immunosuppressive drugs, graft rejection remains a serious problem, resulting in graft failure within five years in approximately 35% of cases in the U.S. We are developing FG-5200 for the treatment of corneal blindness resulting from partial thickness corneal damage.

In China, there are ethical or religious beliefs, cultural norms and significant infrastructure barriers that limit organ donation or tissue banking possibilities, resulting in an extreme shortage of cadaver corneas. In September 2017, the Chinese State Council issued a regulation banning the importation of human tissue which is expected to further diminish the availability of cadaver corneas for implantation in China. In April 2015, a subsidiary of China Regenerative Medicine International Limited received approval for their acellular porcine cornea stroma medical device for the indication of repair of corneal ulcers in China. However, alternatives to cadaver corneas, such as synthetic corneas using collagen derived from porcine tissue or fish scales, are either experimental or to our knowledge, have not yielded satisfactory results for restoration of vision in patients with corneal blindness. In many cases of corneal blindness, infection and other factors lead to serious risks to the patient.

Market Opportunity

Approximately 40,000 corneal grafts were performed in the U.S. in 2011 using tissue from human cadavers. In contrast, while there are approximately four to five million patients in China with corneal blindness and an incidence of 100,000 cases of corneal blindness each year, there were only about 3,000 corneal grafts performed in China in 2007 using tissue from human cadavers. We believe the number of corneal grafts using cadaver tissue in China may decrease significantly due to recent changes in government policy.

FG-5200 as a Potential Solution to This Unmet Medical Need

FG-5200 Corneal Implant

Our expertise in fibrosis and ECM proteins has allowed us to develop processes for producing human collagen types I, II and III, as well as coordinate expression of several enzymes involved in assembly of collagen. We have successfully produced a proprietary version of recombinant human collagen III that is suitable for use in cornea repair.

FG-5200, a corneal implant medical device we are developing in China, is designed to serve as an immediately functional replacement cornea as well as a scaffold to allow for regeneration of the native corneal tissue for the primary purpose of restoration of vision. In contrast, cadaver graft tissue is never “turned over”; in fact, only limited integration occurs over the life of the graft. Our FG-5200 implant is made of recombinant human collagen that has been formed into a highly concentrated fibrillar matrix to provide physical characteristics optimal for corneal implantation.

In animal models, FG-5200 allows for native tissue to completely regrow in less than one year, including both epithelium (the outer cell layer of the cornea) and stroma. The stroma in these animal models is seen to be infiltrated with nerve fibers, leading to the reacquisition of the touch response critical to the avoidance of additional corneal damage.

Corneal implants using human donor tissue are currently being reimbursed by the government, and similar to many other implantable Class III devices in China (including stents and bone grafts), we would expect that FG-5200 could be added to the reimbursement list for medical devices, if approved.


Clinical Testing of FG-5200

An initial clinical study outside of China has been conducted to test the safety and feasibility of using a biosynthetic implant composed of our recombinant human collagen, and substantially similar to FG-5200, for the treatment of severe corneal damage as an alternative to human donor tissue. Ten patients with advanced keratoconus, or severe corneal scarring, were implanted with the recombinant collagen implants and have been followed for more than five years. Two-year follow-up data were reported in Science Translational Medicine (Fagerholm et al., (2010)) and four-year follow-up data were reported in Biomaterials (Fagerholm et al., Biomaterials (2014)). Key clinical findings include the following:


Patients with biosynthetic implants had a four-year mean corrected visual acuity of 20/54 and gained on average more than five Snellen lines of vision on an eye chart.


Nerve re-growth and touch sensitivity was closer to that of healthy corneas and significantly better in corneas with biosynthetic implants than in human donor corneas.


Corneas with biosynthetic implants maintained a stable shape and thickness without any need for a long course of immunosuppression therapy.


There has been no recruitment of inflammatory dendritic cells into the biosynthetic implant area and no episodes of rejection, in contrast to the control arm of human donor cornea transplantation, where a rejection episode was observed.

FG-5200 Strategy

In January 2016, our subsidiary FibroGen Beijing received CFDA’s written notice of classification of our FG-5200 corneal implant as a Domestic Class III medical device. This allows FibroGen to develop, and if approved, to market FG-5200 corneal implants fabricated in China without any prior reference approval outside of China.

We currently plan to manufacture FG-5200 preclinical and clinical trial material in our aseptic good manufacturing practices production suite located at our Beijing manufacturing plant. We completed the process technology transfer and the registration campaign in 2017. Materials from this campaign are being used in preclinical studies that we expect to complete in 2018 or early 2019. We expect to file a CTA for a pivotal clinical trial after final results of the preclinical studies and discussion with the CFDA.

We plan to develop FG-5200 in China first. If FG-5200 is successful in China, we believe there is a future opportunity to develop FG-5200 in other Asian countries where cadaver materials are in short supply, in part because cultural norms and infrastructure and other challenges in tissue banking limit tissue donations. We also believe there is an opportunity to obtain CE Marking to facilitate entry into other markets, such as Latin America. We may develop FG-5200 in the U.S. and Europe as well, where cadaver corneas are available but the required immunosuppressive therapy may make FG-5200 a potentially attractive alternative.


Our Collaboration Partnerships for Roxadustat


We have two agreements with Astellas for the development and commercialization of roxadustat, one for Japan, and one for Europe, the Commonwealth of Independent States, the Middle East and South Africa. Under these agreements we provided Astellas the right to develop and commercialize roxadustat for anemia in these territories.

We share responsibility with Astellas for clinical development activities required for U.S. and EU regulatory approval of roxadustat, and share equally those development costs under the agreed development plan for such activities. Astellas will be responsible for clinical development activities and all associated costs required for regulatory approval in all other countries in the Astellas territories. Astellas will own and have responsibility for regulatory filings in its territories. We are responsible, either directly or through our contract manufacturers, for the manufacture and supply of all quantities of roxadustat to be used in development and commercialization under the agreements.

The Astellas agreements will continue in effect until terminated. Either party may terminate the agreements for certain material breaches by the other party. In addition, Astellas will have the right to terminate the agreements for certain specified technical product failures, upon generic sales reaching a particular threshold, upon certain regulatory actions, or upon our entering into a settlement admitting the invalidity or unenforceability of our licensed patents. Astellas may also terminate the agreements for convenience upon advance written notice to us. In the event of any termination of the agreements, Astellas will transfer and assign to us the regulatory filings for roxadustat and will assign or license us the relevant trademarks used with the products in the Astellas territories. Under certain terminations, Astellas is also obligated to pay us a termination fee.


Consideration under these agreements includes a total of $360.1 million in upfront and non-contingent payments, and milestone payments totaling $557.5 million, of which $542.5 million are development and regulatory milestones, and $15.0 million are commercial-based milestones. Total consideration, excluding development cost reimbursement and product sales-related payments, could reach $917.6 million. During the second quarter of 2016, we recognized $10.0 million revenue as a result of the initiation by Astellas of the first Phase 3 clinical study in Japan of roxadustat for treatment of anemia associated with CKD in patients on dialysis. The amount was received in early July 2016. The aggregate amount of such consideration received through December 31, 2017 totals $472.6 million.

Additionally, under these agreements, Astellas pays 100% of the commercialization costs in their territories. Astellas will pay us a transfer price for our manufacture and delivery of roxadustat based on a calculation based on net sales of roxadustat in the low 20% range.

In addition, Astellas has separately invested $80.5 million in the equity of FibroGen, Inc. to date.


We also have two agreements with AstraZeneca for the development and commercialization of roxadustat for anemia, one for China (the “China Agreement”), and one for the U.S. and all other countries not previously licensed to Astellas (the “U.S./RoW Agreement”). Under these agreements we provided AstraZeneca the right to develop and commercialize roxadustat for anemia in these territories. We share responsibility with AstraZeneca for clinical development activities required for U.S. regulatory approval of roxadustat.

Now that we have reached the $116.5 million cap on our initial funding obligations (under which we shared 50% of the initial development costs), all future development and commercialization costs for roxadustat for the treatment of anemia in CKD in the U.S., Europe, Japan and all other markets outside of China will be paid by Astellas and AstraZeneca.

In China, our subsidiary FibroGen Beijing will conduct the development work for CKD anemia and will hold all of the regulatory licenses issued by China regulatory authorities and be primarily responsible for regulatory, clinical and manufacturing. China development costs are shared 50/50. AstraZeneca is also responsible for 100% of development expenses in all other licensed territories outside of China. We are responsible, through our contract manufacturers, for the manufacture and supply of all quantities of roxadustat to be used in development and commercialization under the agreements.

Under the AstraZeneca agreements, we receive upfront and subsequent non-contingent payments totaling $402.2 million. Potential milestone payments under the agreements total $1.2 billion, of which $571.0 million are development and regulatory milestones, and $652.5 million are commercial-based milestones. Total consideration under the agreements, excluding development cost reimbursement, transfer price payments, royalties and profit share, could reach $1.6 billion. During the second quarter of 2016, we received an upfront payment of $62.0 million time based development milestone. In October 2017, the CFDA accepted our recently submitted NDA for registration of roxadustat for anemia in DD CKD and NDD-CKD patients. This NDA submission triggered a $15.0 million milestone payment to FibroGen by AstraZeneca, which was received and fully recognized under our revenue recognition policy as license and milestone revenue in the fourth quarter of 2017. The aggregate amount of such consideration received through December 31, 2017 totals 432.2 million.

Payments under these agreements include over $500 million in upfront, non-contingent and other payments received or expected to be received prior to the first U.S. approval, excluding development expense reimbursement.

AstraZeneca purchased 1,111,111 shares of our common stock at the initial public offering (“IPO”) price for an aggregate purchase price of $20.0 million in a private placement concurrent with our IPO. In connection with the purchase of our shares of common stock in the private placement, AstraZeneca has also entered into a standstill agreement which provides that, until November 2019, neither AstraZeneca nor its representatives will, directly or indirectly, among other things, acquire any additional securities or assets of ours, solicit proxies for our securities, participate in a business combination involving us, or seek to influence our management or policies, except with the prior consent of our board of directors and in certain other specified circumstances involving a change of control of our company. In addition, AstraZeneca has agreed to vote its shares in favor of nominees to our board of directors, increases in the authorized capital stock of the company and amendments to our equity plans approved by the board of directors, in each case as recommended by a majority of our board of directors. AstraZeneca has also agreed, subject to specified exceptions, not to sell shares purchased by it in the private placement for the two-year period following such purchase and to limitations on the volume of its sales of such shares thereafter.


Under the U.S./RoW Agreement, AstraZeneca will pay for all commercialization costs in the U.S. and RoW, AstraZeneca will be responsible for the U.S. commercialization of roxadustat, with FibroGen undertaking specified promotional activities in the ESRD segment in the U.S. In addition, we will receive a transfer price for delivery of commercial product based on a percentage of net sales in the low- to mid-single digit range and AstraZeneca will pay us a tiered royalty on net sales of roxadustat in the low 20% range.

Under the China Agreement, which is conducted through FibroGen China Anemia Holdings, Ltd. (“FibroGen China”), the commercial collaboration is structured as a 50/50 profit share. AstraZeneca will conduct commercialization activities in China as well as serve as the master distributor for roxadustat and will fund roxadustat launch costs in China until FibroGen Beijing has achieved profitability. At that time, AstraZeneca will recoup 50% of their historical launch costs out of initial roxadustat profits in China.

AstraZeneca may terminate the U.S./RoW Agreement upon specified events, including our bankruptcy or insolvency, our uncured material breach, technical product failure, or upon 180 days prior written notice at will. If AstraZeneca terminates the U.S/RoW Agreement at will, in addition to any unpaid non-contingent payments, it will be responsible to pay for a substantial portion of the post-termination development costs under the agreed development plan until regulatory approval.

AstraZeneca may terminate the China Agreement upon specified events, including our bankruptcy or insolvency, our uncured material breach, technical product failure, or upon advance prior written notice at will. If AstraZeneca terminates our China Agreement at will, it will be responsible to pay for transition costs as well as make a specified payment to FibroGen China.

In the event of any termination of the agreements, but subject to modification upon termination for technical product failure, AstraZeneca will transfer and assign to us any regulatory filings and approvals for roxadustat in the affected territories that they may hold under our agreements, grant us licenses and conduct certain transition activities.

Additional Information Related to Collaboration Agreements

Of the $1,113.5 million in development and regulatory milestones payable in the aggregate under our Astellas and AstraZeneca collaboration agreements, $425.0 million is payable upon achievement of milestones relating to the submission and approval of roxadustat in DD-CKD and NDD-CKD in the U.S. and Europe.

Information about collaboration partners that accounted for more than 10% of our total revenue or accounts receivable for the last three fiscal years is set forth in Note 14 to our consolidated financial statements under Item 8 of this Annual Report.


The pharmaceutical and biotechnology industries are highly competitive, particularly in some of the indications we are developing drug candidates, including anemia in CKD, IPF, pancreatic cancer, and DMD. We face competition from multiple other pharmaceutical and biotechnology companies, many of which have significantly greater financial, technical and human resources and experience in product development, manufacturing and marketing. These potential advantages of our competitors are particularly a risk in IPF, pancreatic cancer, and DMD, where we do not currently have a development or commercialization partner.

We expect any products that we develop and commercialize to compete on the basis of, among other things, efficacy, safety, convenience of administration and delivery, price, the level of generic competition and the availability of reimbursement from government and other third party payors.

If either of our lead product candidates is approved, they will compete with currently marketed products, and product candidates that may be approved for marketing in the future, for treatment of the following indications:

Roxadustat — Anemia in CKD

If roxadustat is approved for the treatment of anemia in patients with CKD and launched commercially, competing drugs are expected to include ESAs, particularly in those patient segments where ESAs are used. Currently available ESAs include epoetin alfa (EPOGEN ® marketed by Amgen Inc. in the U.S., Procrit ® and Erypo ®/Eprex ®, marketed by Johnson & Johnson, Inc. and Espo ® marketed by Kyowa Hakko Kirin in Japan and China), darbepoetin (Amgen/Kyowa Hakko Kirin’s Aranesp ® and NESP ® ) and Mircera ® marketed by Roche outside the U.S. and by Vifor Pharma (formerly a company of Galenica Group (“Vifor”)), a Roche licensee, in the U.S. and Puerto Rico, as well as biosimilar versions of these currently marketed ESA products. ESAs have been used in the treatment of anemia in CKD for more than 20 years, serving a significant majority of dialysis patients. While NDD-CKD patients who are not under the care of nephrologists, including those with diabetes and hypertension, do not typically receive ESAs and are often left untreated, some patients under nephrology care may be receiving ESA therapy. It may be difficult to encourage healthcare providers and patients to switch to roxadustat from products with which they have become familiar.


We may also face competition from potential new anemia therapies currently in clinical development, including in those patients segments not currently addressed by ESAs. Companies such