Welcome to the FibroGen’s First Quarter 2019 Financial Results Conference Call. My name is Erin, and I’ll be your operator for today’s call. At this time, all participants are in a listen-only mode. Later, we will conduct a question-and-answer session. [Operator Instructions] Please note this conference is being recorded. I’ll now turn the call over to Karen Bergman. Ms. Bergman, you may begin.
Erin, thank you very much, and good afternoon, everyone. And thank you for joining our call. Today, we are reporting financial results and corporate update for the first quarter of 2019.
Joining today’s call are Mr. Tom Neff, Chairman and Chief Executive Officer; Dr. Peony Yu, Chief Medical Officer; Dr. Elias Kouchakji, Senior Vice President, Clinical Development, Drug Safety and Pharmacovigilance; and Mr. Pat Cotroneo, Chief Financial Officer.
Following prepared remarks, Tom will discuss upcoming milestones and we will open the call to Q&A.
During this call, we may make forward-looking statements regarding our business, including our collaborations with AstraZeneca and Astellas; financial guidance; the initiation, enrollment, design, conduct and results of clinical trials; our regulatory strategies and potential regulatory results; our research and development activities; and certain other business matters.
For risks and uncertainties regarding our business and statements made on the call today as well as factors beyond our control that may cause differences between current expectations and actual results, we refer you to our Annual Report on Form 10-K for the fiscal year ended December 31, 2018, filed with the Securities and Exchange Commission. Copies of these filings can be found in the Investors section of our website. We undertake no obligation to update any forward-looking statement whether as a result of our new information, future development or otherwise. The format for today’s call includes remarks from FibroGen’s management team, and then we’ll open the lines to take your questions. The press release reporting our financial results and business update and a webcast of today’s conference call can be found on the Investors section of FibroGen’s website at www.fibrogen.com. The webcast will be available for two weeks from today’s date. And with that, I’d now like to turn the call over to our CEO, Tom Neff.
Thank you, Karen. Welcome, everyone, and thank you for joining us. Today, we are reporting on topline adjudicated results from the seven Phase 3 roxadustat studies performed by ourselves and our partners AstraZeneca and Astellas including the MACE and MACE+ analyses, which an important part of the overall benefit risk assessment regulators will perform. These results are broken up into two safety pools as agreed upon with FDA and EMA. There are 4,000 patients in the pool of CKD patients who are dialysis dependent where roxadustat is compared to epoetin alfa, and 4,300 CKD patients who are not on dialysis where we are compared to placebo.
We are also reporting results in the subgroup of incident dialysis, representing patients who have recently begun dialysis and are not on ESA. This result is especially important as this setting is the most suitable for the comparison of roxadustat data and the current standard of care. In the U.S., we believe that major adverse cardiac events or MACE endpoint defined as the time to first occurrence of death, myocardial infarction or stroke will be the primary basis upon, which the FDA will assess the safety of roxadustat. In Europe, we believe that MACE+, which consists of the defined MACE components plus hospitalization due to heart failure or unstable angina will be the primary endpoint for safety assessment by European regulators. In the EU, we have agreed with regulators on a non-inferiority margin and in the U.S., we have had extensive discussions on this topic and expect to finalize standards in our pre-NDA meeting. We discussed today some of the most important initial safety results based on the totality of data that we’ve analyzed to-date.
In addition, we want to highlight important results from other clinical measures that we are analyzing such as change in renal function and non-dialysis patients as measured by changing eGFR efficacy of roxadustat as compared to epoetin alfa in the presence of patient inflammation and relevant quality of life measures, non-dialysis compared to placebo. Well, the FDA and EMA will make their ultimate approval decisions on their own analyses of benefit risk profile of roxadustat, the applicable patient populations nonetheless, we and our partners believe the data strongly supportive of the efficacy and safety of roxadustat. I would now like to walk through our MACE and MACE+ plus results from the adjudicated pooled Phase 3 data, which have been reviewed by our – with our U.S. partner AstraZeneca and our EU partner Astellas. And dialysis-dependent CKD and the pooled analysis, roxadustat was shown to be non-inferior to epoetin alfa and MACE+ analysis for the U.S. where there were several amount in analyses. We believe there was no clinically meaningful difference between roxadustat in epoetin alfa in MACE risk. An incident dialysis as noted above, we examined the results from the incident population who are patients and newly initiating dialysis approximately four months time before they initiate anemia therapy. We believe this dialysis subpopulation is the fairest setting for comparison of roxadustat versus ESA, as this period of treatment has substantially increased levels of patient mortality, where the stable dialysis population contains the twin biases of patients who have survived the incident period and are already on stable doses of ESA after dose titration. An incident dialysis, roxadustat demonstrated superiority to epoetin alfa and time to first MACE+ analysis and in the time to MACE analysis there is a trend towards reduced risk for patients on roxadustat as compared to epoetin alfa. We would highlight for you the robust result of superiority came from the 1500 patient pool with one to one randomization versus epoetin alfa. And the NDD-CKD population, 4,300 patients randomized over three studies conducted by AstraZeneca, Astellas and FibroGen. These studies represent the first investigation in the CKD population were meeting in eGFR levels are much lower than prior studies. This is a study that was started after FDA restricted the use of ESAs to below hemoglobin 10 and no retreatment at levels above 10. And non-dialysis roxadustat was shown to be non-inferior to placebo and time to first MACE+. In the MACE safety analysis of this population, we believe there is no clinically meaningful difference between roxadustat and placebo.
Another way to view the NDD population results is to test the ITT analysis with long-term follow-up as a majority of patients in this pool followed through the end of the study. This is a conservative way to evaluate long-term safety in the NDD population. Patients on roxadustat received treatment for longer periods than placebo patients, because some of the placebo patients drop out for lack of efficacy since safety day were also collected in the post treatment period. We were able to add in ITT analysis, one from follow-up as a majority of the patients in this pool followed through the end of the study.
As to the intent to treat results in multiple analyses of both MACE and MACE+, MACE free survival and MACE+ free survival CV MACE and CV MACE+. Roxadustat was comparable to placebo. These results were below the commonly applied non-inferiority margin at 1.3. The ITT long-term follow-up analysis is one of several methods that have been discussed with the FDA to address the differential dropout rate.
As to apart from the MACE data results, correction I have to make, which is that we have not yet spoken with FDA. There is a discussion plan with FDA about these various analyses.
So it is planned to be not has happened.
All of these evaluations help to better understand additional benefits of roxadustat apart from MACE and MACE+ safety analyses. It is important to note that we enrolled a unique population eGFR was from zero to 60, including even the sickest patients with median eGFR on the study pool between 15 and 20 as large prior studies such as treat.
We’re enrolling patients much healthier this first time a patient population the sick has been studied.
We have results in the eGFR versus placebo comparison for change in renal function over time. We evaluated where the treatment with roxadustat could slow the rate of decline of renal function and clinically relevant matter. Evaluations were performed across all NDD-CKD populations.
We also included the more conventional baseline cutoffs for non-dialysis patients, meaning the patients with eGFR above 10, as well as the population eGFR above 15. Since receiving the unblinded data after adjudication, we’ve completed the one year analysis of roxadustat versus placebo data evaluation and the study continues in the one year data.
We have observed statistically significant slower rate of eGFR decline in the roxadustat treated patients versus placebo treated patients in the NDD pool as well as both subgroups and the pooled analysis of eGFR change over time from the three NDD studies in patients with baseline eGFR above 15. They showed a treatment difference from baseline of 1.62 at 12 months. We believe the results observed over time and this analysis suggested the roxadustat treatment may slow renal function decline in a clinically meaningful manner.
We are examining the data in more detail, including stratified by level of disease progression to understand the degree of difference and the impact of different subgroups of baseline over the longer-term. And other measures efficacy of roxadustat, we are pleased to have shown statistically significant improvement in the multiple standard quality of life measurements and to have confirmed roxadustat efficacy in the presence of inflammation as measured by CRP where the pool requires increased doses over time. Dr. Peony Yu will describe these results in more detail later in the call.
We are on track for roxadustat submission for U.S. NDA September, October time period with the European MAA submission to follow.
Let me now turn to updates regarding China.
Our NDA for roxadustat was approved in December, 2018 by the National Medical Products Administration, NMPA for the treatment of anemia caused by CKD and dialysis patients. We were pleased that all clinical site inspections for the Phase 3 non-dialysis study have not been completed by the Food and Drug Administration division or CFDI of NMPA.
We expect the population on non-dialysis CKD patients to be added to the CKD anemia indication and the roxadustat label in mid-2019.
We continue to work closely with our partner AstraZeneca to prepare for the launch of roxadustat in China and have much to report in terms of progress.
Our partner is AstraZeneca has already initiated the aggressive build out of a dedicated roxadustat field salesforce covering five regions in China at launch, which will be fully deployed by mid-2019. FibroGen China’s medical affairs field staff now numbers over 30 professionals. The Central Market access team has been in place in active for over a year. FibroGen has the Marketing Authorization Holder or MAH holder in China is responsible for Pharmacovigilance or Pharmacovigilance infrastructure in China, which includes a Pharmacovigilance database and call center has been active for over a quarter now. Commercial manufacturing readiness for both API and drug product is on schedule.
We are confident about the third quarter China launch timeframe. A question we wanted to address is whether it will be added to NRDL list or National Reimbursement Drug List.
We are hopeful that roxadustat may qualify for consideration in 2019 by virtue of receiving market approval at the end of 2018 to be clear only our dialysis label can be considered as non-dialysis approval has not yet been received. Many factors go into our probability of admission NRDL such as perceived unmet medical need, clinical value, pharmacoeconomic value, pricing, which has to be agreed upon between the State Medical Insurance Agency and the sponsor. We believe NRDL is critical affordability for our patients, so this is a top priority for us. Every effort is being made by the joint AZ and FibroGen team to maximize our chances of success.
We expect to have a sense of whether it would be included what price by sometime in October.
Finally, in Japan, our partner Astellas submitted the NDA for roxadustat treatment of anemia and CKD patients on dialysis to PMDA last fall. The application is currently under review and Astellas anticipates filing a supplemental NDA for the treatment of anemia and non-dialysis-dependent CKD patients in 2019 with respect to expand a platform opportunities for roxadustat for the treatment of anemia and other disease settings. I’d like to take a moment to update you on our work with roxadustat and myelodysplastic syndromes or MDS.
We have completed enrollment of 24 patients in open-label lead in portion of our multicenter multinational Phase 3 study in transfusion dependent lower risk patients with MDS. Encouraged by the positive results from the open-label portion as measured by the proportion of patients who achieved transfusion independence.
We have begun enrolling 160 patient double blind placebo controlled portion of the study. Enrollment for the open-label portion of the China Phase 2/3 study is ongoing.
Turning now to pamrevlumab or anti-CTGF antibody.
We are extremely excited to see data reflection one year of treatment in our ongoing Phase 2 study evaluating treatment of Duchenne muscular dystrophy in non-ambulatory patients. The data is supportive earlier reserve trends and examines multiple parameters of disease progression including lung function, cardiac function, and muscle strength. We believe that these data would show increase in function and certain parameters. We’ll support a pivotal study and we plan to discuss with the FDA in the near future. Dr. Elias Kouchakji will discuss these results in more detail shortly. Pamrevlumab recently received orphan drug designation for treatment of DMD.
Our antibody now has orphan drug designation status in all three of the current indications, IPF, pancreatic cancer, and DMD, and has received fast track designation in IPF and in pancreatic cancer.
Finally, I will briefly address financial matters here and Pat Cotroneo, our CFO will provide more detail later in the call. In the first quarter of 2019, we reported $45.4 million of net loss or $0.53 per basic and diluted share in EPS, so that’s negative $0.53 per share.
As of March 31, 2019, FibroGen had $712 million in cash. I would now like to turn this over to Dr. Peony Yu for a discussion of the results from the CV safety pooled analyses and updates on the anemia program. Dr. Yu, please go ahead.
Thank you, Tom.
For the roxadustat program in the U.S. and ER, we are pleased to share with you the adjudicated cardiovascular pool safety analyses topline results from the Phase 3 global roxadustat program. We see these results supporting safety of roxadustat in CKD patients.
We also see other benefits beyond hemoglobin increase. On MACE and MACE+, to reiterate what Tom outlined earlier and what we are disclosing on the adjudicated safety data, MACE and MACE+ composite endpoints. MACE measures the number of patients with one or more adjudicated positive MACE events, which include death, myocardial infarction and stroke. The MACE+ composite endpoint has two more components, in addition to the MACE, which now in also adding heart failure and unstable angina requiring hospitalization. The MACE composite endpoint is what we and our U.S. partner AstraZeneca discuss what the FDA for safety assessment. MACE+ is what we and our European partner Astellas agree with EMA. MACE+ have also been used extensively in safety endpoint and assessments in CKD anemia trials as well as in some diabetes trial in both Europe and in the U.S.
Now what is adjudication and why is – it’s a part of our process. Adjudication is the process of independently and objectively applying clinical standards to consistently determine individual events, qualify as MACE or MACE+ events, to maintain objectivity. Independent experts in cardiology, neurology – and neurology, who are blinded to treatment assignment reveal the patient data and adjudicate the events relevant to their specialty. To maintain data integrity, the process and documents are managed by an independent third party. We conduct the MACE and MACE+ analyses in the following patient pools dialysis or all dialysis. Incident dialysis, which is a sub population of dialysis patients, who are just starting out to receive chronic dialysis treatment, and as well as in non-dialysis patient population. In our dialysis pool of around 4,000 dialysis patients based on the collective results of the various MACE and MACE+ safety analyses, we believe there’s no clinically meaningful difference in MACE and MACE+ risk between roxadustat and epoetin alfa in the all dialysis patient population and the 95% confidence interval of the hazard ratio is above – I’m sorry, off that 95% confidence interval of the hazard ratio is below 1.3, which is what the conventionally accepted measure in such time to analyses of MACE and MACE+. In the incident dialysis pool consisting of over 1500 patients, a subpopulation of the dialysis, our all dialysis pool, which we believe offers a better setting for comparing roxadustat to epoetin alfa, then the stable dialysis population. Roxadustat demonstrate the superiority to epoetin alfa in the time to first MACE+ in this subpopulation with fewer patients with MACE+ events was noted. In the time to first MACE analysis, there is a trend toward reduced risk for patients on roxadustat, compared to epoetin alfa. In our CKD non-dialysis pool of approximately 4,300 patients in the multiple MACE and MACE+ pool, ITT analyses conducted in non-dialysis-dependent CKD patients. Collectively, we believe there is no clinically meaningful difference in MACE and MACE+ risk between roxadustat and placebo in non-dialysis patients.
Now let’s put these safety findings along with the reported efficacy results in clinical context. In our last earnings call, we reported that roxadustat demonstrated efficacy in meeting the primary efficacy endpoint of change in hemoglobin from baseline to mean hemoglobin average between weeks 28 to 52 in each of the seven Phase 3 studies conducted by FibroGen and our partners.
For the all dialysis patient pool, as reported previously, not only was non-inferiority achieve in primary efficacy end points in all dialysis patients. Superiority in efficacy, as demonstrated by a statistically significant larger increase from baseline hemoglobin level in roxadustat treated patients than EPO patients was achieved in both HIMALAYAS study, which is in on incident dialysis and in SIERRAS conversion dialysis studies. In SIERRAS, roxadustat treated patients achieve a more physiologic hemoglobin level of 10.7 grams per deciliter versus 10.2 in EPO arm, with a 33% reduction in red blood cell transfusion risks. We believe the lower achieve hemoglobin level in EPO comparator arm, results from a combination off the lower target hemoglobin level on ESA label, due to FDA’s concern regarding ESA cardiovascular safety and EPO hyporesponsiveness in patients with inflammation as inflamed patients with elevated c-reactive protein require higher doses of EPO, then patients with normal baseline CRP levels, although, they achieve lower mean hemoglobin level. And as we understand some multiple publications that there’s higher risk with higher target hemoglobin, when one is using ESA, while it higher achieved hemoglobin in EPO has been conferred to better safety, when lower doses of EPO is use.
Given the mechanism of our drug being uniquely different than EPO being more physio has been able to achieve a more physiologic hemoglobin level should translate into benefit for patients. Roxadustat efficacy measured in achieve hemoglobin level and dose requirement are not affected by inflammation status, unlike in EPO. This important differentiation from ESA has been observed in multiple Phase 3 studies, including in the U.S. based CRS study, which we believe is reflective of U.S. dialysis practice undercurrent ESA labeling restriction. And we also seem this differentiation in our China Phase 3 study in dialysis patients.
We have also observe a reduction of risk of a red blood cell transfusion in our Phase 3 program.
Although, the roxadustat arm in conversion study start out at a disadvantage being exposed to patients for the first time and being compared to patients on optimize EPO dosing, the roxadustat treated patients, they’ll show significant reduction in red blood cell transfusion risk, as measured by time to first transfusion compared to patients receiving stable maintenance doses of epoetin alfa. Conversion patients comprise a majority in the dialysis, our all dialysis patient pool. Yes. Based on the key MACE and MACE+ analyses, we assessed no clinically meaningful difference in the risk of MACE, MACE+ with between roxadustat and epoetin alfa.
Next, the incident dialysis population is defined as patients, who enter dialysis studies within four months of starting dialysis treatment. A vast majority of the incident dialysis patients, where either ESA-naïve or had very limited prior exposure to EPO.
During the transition from non-dialysis to dialysis in the first four months of dialysis treatment, patients suffer from mortality and hospitalization rate applies those off dialysis patients survived the first year of dialysis treatment. What’s also relevant to roxadustat is that, initiation of dialysis often also coincides with the initiation of anemia therapy. We and our partners are very happy with the result or superiority to epoetin alfa in the time to first MACE+ and a favor of both trend towards reduce risk for patients. On roxadustat compared to epoetin alfa in time to first MACE analysis in the incident dialysis patient populations. We believe roxadustat favorable results in incident dialysis compared to EPO could enable a safer treatment of anemia in CKD patients initiating and continuing dialysis treatment.
Turning to CKD patients, not on dialysis. In the CKD pool analyses from the three non-dialysis studies, roxadustat both consistently raise hemoglobin levels to mean hemoglobin of 11 gram per deciliter in the roxadustat treated patients and significantly reduced red blood cell transfusion compared to placebo. This is very important for patients to preserve their ability to have kidney transplant later down the road, if they need to.
Importantly, roxadustat has shown the potential to preserve renal function as there was a statistically significantly smaller declines in eGFR in roxa-treated patients than placebo. With a treatment difference of 1.62 in eGFR unit, when measuring change at one year from baseline in patients with baseline eGFR of 15 or higher. P value is 0.0001 or a reduction of declined by 38% in eGFR, relative to placebo arm.
We also observe statistically significant improvements in the various quality of life end points at 12 weeks from baseline, including the SF-36 Vitality subscale with p-value that has 0.0002, SF-36 Physical Functioning subscale p-value 0.0369, FACT-AN Anemia subscale with p-value [Audio Dip] FACT-AN Total score p-value 0.0056, EQ-5D-SL VAS score with p-value that has 0.0005 in CKD patients not on dialysis. Putting together these and other important potential clinical benefit with the safety results that we have seen, in comparison with EPO, which is a gold standard for safety measurement and in comparison to EPO, which is the current standard of care in dialysis, but have some limitations. Along with the convenience of a pill, when treating patients with roxadustat to make treatment much, much more accessible than parenteral route of ESA.
We are excited about the potential of roxadustat as an innovative new therapy for CKD patients. We hope this provides some helpful context for you in understanding the significance of the safety top line results of announce today. Tom has already touch on the status of our U.S./EU submission plans, we and AstraZeneca will be in discussion with FDA and NDA submission plan, which we’re targeting for September, October timeframe.
We are also supporting Astellas, MAA submission to EMA to be submitted thereafter.
For China, in April, CFDI inspected our China Phase 3 CKD non-dialysis study sites, we expect that the roxadustat label on CKD anemia, there will be expanded mid-year to include non-dialysis as well as dialysis patients.
As Tom mentioned, we are excited about roxadustat’s opportunity for market access via this year’s NRDL election process in China. In Japan, the NDA on roxadustat for treatment of anemia in dialysis dependent CKD patients submitted by our partner Astellas in September 2018 is now under review. We anticipate that Japan NDA decision later this year.
For the treatment of anemia in MDS patients, we have an ongoing Phase 3 study in transfusion dependent lower risk MDS patients in the U.S., Europe and Asia plus another study, which is Phase 2/3 in non-transfusion dependent MDS patients in China. Each has an open label run in period, anemia in MDS is notoriously difficult to treat and we are striving to make a difference for MDS patients with roxadustat.
We have completed enrollment of the first 24 patients in the open label portion of the U.S./European study, we and our partners are encouraged by the available results, as there where large proportion of transfusion dependent patients able to achieve transfusion independence endpoint. And we have already started dosing in the double-blind portion of the U.S., European Phase 3 MDS study.
Finally, we are on track to start our first clinical trial in chemotherapy-induced anemia with roxadustat, as the Phase 2 study in the U.S. to be a started shortly in 2019. I believe that all of us are aware and would like to do something about the under treatments of anemia in patients who have undergone chemotherapy. I’d like to now turn the call back over to Tom.
Thank you, Peony. Dr. Elias Kouchakji will now provide an update on pamrevlumab, including additional details on our DMD data and update us on clinical development activities for pancreatic cancer and for IPF. Elias, please go ahead.
Thank you, Tom. I will start with the Duchenne muscular dystrophy, as Tom mentioned. In mid-March of this year, all 21 patients completed one-year of treatment with pamrevlumab in our Phase 2 open label non-ambulatory Duchenne muscular dystrophy study.
Soon after, we initiated an administrative analysis of the data in early second quarter of this year. This is in order to inform our clinical development strategy. And we review this study later with key opinion leaders, who are expert in this study of the critical function we tested in this study. Starting with the pulmonary function test, we complete one-year results indicate a potential reduction in the rate of decline in FVC percent predicted from baseline in our study population. Especially, when compared to the data published in 2016 by Mayor [indiscernible] in 2019.
As for the cardiac function test, the result as measured by LVEF left ventricular ejection fraction, the data suggested a positive mean percent change from baseline.
While on the published data by MacDonald in 2018 showed a mean decline of approximately 1% from baseline in one-year.
Additionally, we measure the cardiac fibrosis scores and we collected the data and then published data by Tandem in 2015 showed a strong correlation between the cardiac fibrosis with LVEF.
Our data from the MRI fibrosis score suggest the similar correlation. Similarly in some of the muscle function test. The result of these test in this upper arm showed the mean change from baseline was a smaller than the published data [indiscernible] in 2019. Based on these result and the advice we received from our expert, we are planning to share these results with FDA to develop our clinical development plan for Duchenne muscular dystrophy.
Moving forward with our other indication, we are planning on beginning enrolling in the Phase 3 double-blind placebo-controlled of pamrevlumab as neoadjuvant therapy for unresectable locally advanced pancreatic cancer in the second quarter of 2019. We intend to enroll approximately 260 patients, randomization one to one to receive either pamrevlumab in combination with gemcitabine and nab-paclitaxel or chemotherapy with placebo. Also in the second quarter of 2019, we are on track to begin enrolling our double-blind, placebo-controlled Phase 3 trial of pamrevlumab and approximately 500 IPF patients. The primary efficacy endpoint as it changed from baseline in forced vital capacity.
We are grateful for the opportunity represented by pamrevlumab and providing a needed therapeutic option for each of these three serious and progressive indications, Duchenne muscular dystrophy, pancreatic cancer and idiopathic pulmonary fibrosis. Thank you for listening. Tom, I’ll turn the call to you.
Thank you, Elias. Pat Cotroneo, our Chief Financial Officer will now discuss financial highlights for the first quarter of 2019. Pat, please go ahead.
Thank you, Tom.
As announced today, total revenue for the quarter ended March 31, 2019 was $23.9 million, as compared to $31.9 million for the first quarter of 2018.
For the same period, operating expenses were $72.7 million and a net loss was $45.4 million or negative $0.53 per basic and diluted share.
As compared to operating expenses of $72.5 million and a net loss of $41.4 million or negative $0.50 per basic diluted share for the first quarter last year. Included in operating expenses for the quarter ended March 31, 2019 was an aggregate non-cash portion totaling $20.2 million, of which $16.4 million was a result of stock-based compensation expense as compared to an aggregate non-cash portion totaling $12.5 million, of which $10.9 million was a result of stock-based compensation expense for the same period in the prior year. At March 31, 2019, FibroGen had $712.7 million in cash, restricted time deposits, cash equivalents, investments, and receivables.
As previously stated, our considered judgment is that roxadustat NDA and MAA will be filed this year, which will trigger approximately $192.5 million in anticipated milestone payments, of which the vast majority are associated with these filings. Thank you and I’ll turn the call back over to Tom.
Thank you, Pat.
With the updates reported to you today, we advanced a number of critical events in the coming months. It is our privilege to shared with you today, roxadustat said, CV pooled safety analysis results that we believe support submitting the NDA for both in NDD and DD indications, treatment of dialysis dependent and non-dialysis dependent CKD for the FDA in September or October 2019. In Europe, our partner Astellas anticipates submission of the MAA for dialysis dependent and non-dialysis dependent CKD later in 2019 of December. In China, we expect to add non-dialysis dependenct CKD to the roxadustat label upon approval anticipated in mid-2000 actually Q3 2019. In Japan, Astellas is expecting a decision on NDA approval for roxadustat in dialysis dependent CKD fourth quarter of 2019.
For pamrevlumab, we look forward to updating you on our advancement to Phase 3 and LAPC and IPF and to further findings from our ongoing Phase 2 study in DMD non-ambulatory patients. With that, let me turn this back to Karen to begin the question-and-answer period.
Thank you, Tom. Erin, please open up the lines for questions. Thank you.
Thank you. [Operator Instructions] Our first question comes from Michael Yee with Jefferies. Michael, your line is open.
Hey, thanks guys. Good afternoon. Tom and Peony, can you be very clear for us? I think there’s some confusion around whether you are statistically non-inferior in dialysis and non-dialysis on the MACE analysis, which is what is required for FDA? Can you confirm that or discuss that? And if you can also give us the hazard ratios for dialysis and non-dialysis on MACE? That would be very helpful. Thank you.
Okay, so Michael, there’s two parts to this answer. One is that in the European market, we are doing MACE+, where we have a statistical non-inferiority margin as single margin identified and we are non-inferior in both measures. In the U.S., there are multiple non-inferiority margin that are under discussion. These are reflecting the fact that was not incident dialysis and with the non-dialysis dependent CKD patients.
We’re essentially addressing new indications that have not been investigated previously, so that incident dialysis and CKD dialysis. In discussions with our partner, they’re very mindful of the phrase of totality of evidence.
And so they encouraged the idea that we addressed this in the form of the evaluation of results versus MACE, where we did not see any clinically meaningful difference means that it met the safety standards that people were looking for. And that’s why people are moving forward.
So to be very specific in dialysis and non-dialysis on MACE, are you trending in the right way? Are you trending positive? What do you mean by not clinically meaningful differences?
Yes, I think, the message there is where we’re trending favorably, but at the same time, we have to yet agree with our regulator on specific analyses to be done. There are back and forth…
It was trending positive, but you just don’t have – or would you just don’t have an agreement on what the definition of which analysis you would like to have? But it was trending positive.
Michael, I think that’s a very fair way to say it.
Okay. Thank you.
And your next question that comes from Andy Hsieh with William Blair. Andy, your line is open.
Before Andy, may I add to Michael’s response to Michael Yee’s question? I just – I don’t know whether we made it very clear that when Tom and I used the number 1.3.
We are talking about, we’re not talking about hazard ratio. I want to make it absolutely clear that the hazard ratio for the MACE+ in the incident dialysis is way below one, and it also is below one in dialysis.
Okay. And the upper bound of the 95% confidence interval, when you’re below 1.3 has been commonly except the statistical standard for non-inferiority. And for us to state that we are superior in time to MACE+ analysis in incident dialysis. What I mean is the upper bound of the 95% competence interval is less than one. And we have when you compare the hazard between roxadustat to that of epoetin alfa.
We have a very significant p-value.
So I have this help, but Michael, does this answer your question?
Andy, perhaps you muted.
Andy, are you still on the call with us? This is Karen.
Your next question comes from Joel Beatty with Citi. Joel, your line is open.
Hi, this question is on the two pooled non-inferiority MACE analysis required by FDA. Was there I pre-specified statistical analysis plan agreed to with FDA?
So Joel, we have had discussions with the FDA on who they – on to some methods for – of statistical evaluation of safety endpoints. We believe that we have collected the data that the FDA would like to see. And but we wanted to happy – conscious as in stating our agreement with the FDA because as we – both of us who have you interacted with FDA, oftentimes, the reviewers would like to have the – what like to look at the totality of evidence looking at both the safety as well as efficacy. And the term that is used very commonly is called it’s a review issue.
So it’s not an issue, but it is – if I were to be reviewer, I would look at like to look at audit data before I would commit to a decision of – on a drug.
And so I hope that answers your question. And then to ensure alignment, we will be discussing with FDA and in our upcoming pre-NDA meeting and to make sure that our preparations of the NDA package, we’ll provide the information that have – way that is efficient for FDA to review.
Okay. And then a question on the non-dialysis MACE analyses, did the event rate compare favorably to the comparator arm and how – in that analysis, how does it take into consideration the differences in the dropout rate?
Yes, we do – so first of all, Joel, that’s a good question.
So we have – because our drug is so efficacious and so well tolerated, patients really liked staying on our drug.
Now we all know that placebo doesn’t work too well.
So however, this is the – I want to go back and remind everyone that this is a double-blinded study.
And so even – so many patient – in other words, patients are not – do not have the treatment assignment information.
We have many patients, who are on placebo and on roxadustat who were remain the studies to enable our long term efficacy and safety evaluation. And even we did see a high somewhat higher dropout rate in placebo treated patients.
However, to have some anticipation of this could happen, we have collected safety data on patients during the post-treatment period and that’s why we are able to conduct the ITT analysis. And this will be of course the final assessment and the statistics will be discussed with the FDA.
Okay. Thank you.
And I just wanted to share that – in the – so what we have mentioned in the ITT analysis in the non-dialysis patient population it will be considered a relatively conservative analysis and the fact that we have, we are able to show non-inferiority to placebo under such conditions, really illustrates the strength of our drugs safety. And I wanted to also remind us that placebo is considered the gold standard for safety.
Okay. Go ahead. Andy, are you there?
Yes, I’m here. Can you guys hear me?
Yes, I’m sorry, Tom and to the team that I pressed the wrong button so it caused a little confusion.
So I apologize.
So my question has to do with just looking at it from a bigger picture perspective.
You have on the MACE perspective non-clinically, well I guess no clinical difference between both arms roxadustat versus EPO roxadustat versus placebo. But this is– from a – I don’t know if this is a correct way to think about it, from transitive property of equality, we know that placebo is not – from a clinical perspective does not equal to EPO, so how do you kind of think about this discordance and what you just disclosed on the call?
Well, let’s do this in a simpler ground work on which is MACE+, and I’m doing this because we have very defined criteria with MACE+. The findings are with incident dialysis pool, you have statistically significant advantage over ESA, with the entire dialysis pool non-inferior and then in the NDD non-dialysis pool you have non-inferior, so you have these comparisons. We think that the non dialysis pool comparison to placebo is essentially similar, the idea there is that it’s hard to argue that there’s no incremental safety risk.
Just placebo is what happens with CKDP’s every day right now, they don’t have medicines.
So it’s not as if you can infer an incremental statistically proven risk in non-dialysis. In the dialysis pool, the broader population non-inferior and then the incident population, which we think is unbiased comparison, we had statistical superiority. We don’t expect that MACE will be particularly different than this. It’s just that with U.S., we have an agreement with our partner Astrazeneca to evaluate on the totality of evidence spaces.
So it makes it harder to sum this up in one sentence or two sentences. Its sort of a Pretzel Logic challenge to try to describe this accurately, but I think you can sort of look at the MACE+ results. And from our point of view, we thought the MACE results are going to be a lot different. We would say so, but it doesn’t seem that way, it seems very similar.
So I would say with the U.S attributed to the fact that we do not have a single agreed end-point.
One of the questions we ask with these newly defined populations in incident dialysis and in CKD was, whether or not the entire analysis, the entire risk benefit analysis, you should work a little differently than in the conversion dialysis patients that have been using EPO forever. And the FDA basically said that’s a review issue, you need to spend some time showing us the benefit risk analysis that you see from your data.
Yes. Tom, I supplement this a little bit, I apologize if we use too much of this statistics mumbo-jumbo, but I just wanted to add, even though we are saying that there’s no clinically meaningful difference in MACE and MACE+ in our dialysis. I’m just going to give one example, even though we are saying that it is a very conservative way of expressing our data in absolute terms we have, for example in our dialysis pool, we have fewer patients, now we look at on treatment analysis and there are fewer patients who are tied on roxadustat compared to EPO, we have fewer, numerically fewer patients with MACE events or with MACE+.
So that’s sort of, I’m trying to give you a little color to what we are saying. I hope this is helpful.
So, just, kind of digging deeper and helping us kind of understand the totality of data, different components, death, MI, stroke, on top of that unstable angina leading to a hospitalization, heart failure. Can you confirm that all of these measures are trending in the right direction? Or maybe there are some that, that’s not may be can you comment on that?
So with MACE+ data, I believe we have numeric advantage in each category.
So there’s five categories….
Each and single one of them, Tom.
Every one of them we have in America advantage over ESA. Is that clear now?
Numeric advantage meaning lower.
Fewer events of roxadustat versus ESA in death. Fewer events of roxadustat versus ESA in myocardial infarction, fewer your strokes in Fewer events of roxadustat than ESA due to unstable anginal hospitalizations, fewer congestive heart failures resulting in hospitalizations.
Yes, that’s actually super helpful. Thank you for that. And just one last question about the quality of life measures, 12 weeks into that – given that this is a chronic condition, is that clinically relevant or should you be looking at a longer time frame?
Peony, you want to hit the quality of life studies.
So the quality of life, that number that I called it was at 12 weeks, quality of life measures is actually as quite as difficult to measure over long term, because you want it to measure in the same patient population that you start out the study with. And also on these subjective measures that, where a patient report to you how they feel, when you carry it over a much longer period of time, there’s a patient’s – not only patients, I may have forgotten how things – how bad things were like a couple of months ago.
So generally, it is reasonable, three months is very reasonable period to record patient reported outcome.
Great. Thanks for answering all my questions.
Okay. And your next question that comes from at Terence Flynn with Goldman Sachs. Terence, your line is open.
Hi, thanks for taking the question. Maybe just a follow-up, I think that your answer to the last question regarding the dialysis population was pretty clear, but in terms of the non-dialysis comparison of roxadustat versus placebo, I guess I’m still a little bit confused in terms of how the – I understand you didn’t have a pre-specified comparison, you’re looking at a number of different ways, but could you maybe just walk through how those event rates compare on MACE for roxa versus placebo across the different analyses whether do they all line-up, whether all they are favorable? Was there anything out of line? And then when you look at the separate studies again, I remember back from the MANTA study, they had one trial, that they showed fewer events and one trial they actually had more events in non-dialysis setting.
So again, was there consistency across the three studies that you pooled as well? Thank you.
Yes, so Terence, we recognize that this is a terribly difficult area to state in a succinct manner in a call like this. Having said that and thinking about how to describe the situation most effectively, we decided to describe the ITT results. This is MACE, MACE+, MACE CV, timed MACE+, time MACE so there’s several different measures and in each case the result of the analysis was at a ratio below 1.3 which is a standard non-inferiority comparison in ITT.
So when I say below 1.3 or I mean like 1.18 or 1.21 or 1.27 or 1.28 not above 1.3 below 1.3. And I know speaking as someone involved in this partnership for a long time that people in each of our partner’s executive management group bound great confidence and strength in seeing these results because these are, even though these are maybe aren’t the measures that will ultimately be the ones that are evaluated, they are an ultimate safety evaluation standard that FDA usually asks for whether you post it or not.
So everybody felt like this is something that’s very descriptive and very informative. I would hesitate to do anything else beyond talking about the ITT results because we do not have a specific agreement with FDA on method of analysis and as such as a little bit presumptuous. And I think the challenge for any of these statistics that would be like OT-7 or OT-28 or whatever, is that you have a placebo dropout rate that’s very different than the roxa, stay on steady rate. There is agreement from regulatory body that we can make statistical adjustments. We’ve gotten that in print from our reviewers and there’s certain things like covariates or IPCW adjustments that have been suggested as ways that there’s an acknowledgement that placebo dropout rate is an issue that needs to be evaluated. Peony, you wanted to go ahead from here?
So FDA specifically suggested for us to do exposure adjusted safety analysis.
So, if you have a more patient exposure time on one arm than another how do you compare the number of patients who have how many events.
And so when we – so that’s why the ITT evaluations is one of the ways that we make such comparison.
Another way will be exposure adjusted to match the follow-up time on the two arms, when we do that we again see a very, very, assuring safety data and we see there, this certainly looks non-inferior in our assessment.
Okay. Can I maybe just ask few follow-up?
You have one – Terence, you have one what is it?
Okay, sure. I just again, so below 1.3, so that was for each of the three studies just to be crystal clear on that or that was on a pool basis when you’re talking about the non-dialysis population?
It’s all non-dialysis and it’s evaluation, MACE, MACE+, MACE CV, Peony you want to add to that?
Terence, the agreement we had with the regulators is that for MACE and MACE+ type of analysis will be based on pool analysis across the three studies. And that is …
The three non-dialysis…
The three non-dialysis studies and with the total sample size of about 4,300 patients.
So you can’t give us any more detail about the individual studies, I guess at this point in terms of what the individual numbers look like. Like if the – I’m just trying to understand if they are consistent across the three studies?
So Terence, there is consistency across the three studies.
In fact, we met primary efficacy end-points in all three studies in hemoglobin end points and that we achieve transfusion, superiority, which is clinically very important and each of the individual – none of the – each of the individual study are overall safety trends were acceptable. And I just wanted to share that for MACE and MACE+ events you need a certain powering status – adequate statistical power to have meaningful comparison.
So this is why we are reporting the pool result rather than individual study results.
Okay. Thanks so much, that was a lot clearer. Thank you.
And your next question comes from Geoffrey Porges with Silicon Valley Bank Leerink. Geoffrey, your line is open.
Thank you very much and appreciate all the color that you’re providing. Peony, could I ask you to pivot to the dialysis population and you’ve broken out the incident dialysis population where you seem to have superiority on MACE+ and trying benefit on MACE. Could you give us a sense of what you see in the stable dialysis patients particularly, provide us reassurance about the number of whatever you, however you could add a measure at the number of deaths, MIs and strokes in this stable dialysis patients who were switched to roxadustat still roxadustat, because obviously we don’t have the same power or same apparent benefit in the pool dialysis as you do in the incident.
Okay. Geoff, yes, I wanted to first of all, I think we agreed that the conversion by [indiscernible] study design is biased against the study drug. But even so when we look at sub-group analysis of –between incident dialysis versus the stable conversion dialysis we are quite comfortable with the safety results, when looking at a MACE and MACE+, does that help?
Could you provide the same – is the rate of deaths MI and stroke in that population lower in the roxa treated patients?
So I don’t have the exact numbers sitting in front of me, but I would think that from memory, they seem to be not that far off from one another.
Okay. Peony, could I just follow-up…
We have lot of numbers here, Jeff.
Yes, I understand. Could you just talk about the overall rate of events that you’ve seen compared to expectations and also whether they’re any geographic variances in the comparison, depending upon geography, North America, Europe, et cetera?
Jeff, sorry, just a second. Jeff, I want to get clarity on the question you just asked, but I think Peony wants to clarify one thing. Peony, go ahead.
So Jeff, yes, so we – when we pass that the – yes when we pass that – the non-bio, I wanted – even though I’m not giving you exact number or patients for each category, right? But we – I am willing to share with you that in the subgroup analysis when we test the time to, for example, time to MACE+ and MACE, roxa reduced that was at least non-inferior to epoetin alfa even in the conversion staples dialysis patients.
So you mean, if you take the subgroup incident patients away…
The remainder pool, it was tested as non-inferior consistently.
That’s perfect. Thank you.
No, Jeff, please state the geography question again, it surprised us, I don’t think I heard the whole thing.
I just want to – could you just comment on the overall rate of events that you’ve observed compared to your expectations? And then whether there were any variations in the comparison between the different arms in the – when you look at the subsets by geography?
Jeff, we are at this – we are still on the – we’re leasing the top line results, the fine granular geographic subgroup analysis and more detail analysis are still needs to be conducted. But we’ll plan to – between now and MDA submission, they will be completed before then.
Okay. And the overall event, right?
Right. Question was, do you observe any difference in overall event rate by geographies?
We have not done that subgroup analysis by geography yet.
Okay. Thanks very much.
Yes. Thank you.
And your next question comes from Difei Yang with Mizuho Securities.
Your line is open.
Hi. Good afternoon, and thanks for taking my question. I apologize upfront, then I may be a little slow, I’m trying to get some of the answers.
So Tom and Peony, maybe you could help me think about three patient populations, there’s the DD, NDD and incident dialysis. Is incident dialysis patient population has a possible indication in the U.S.? I’m focusing on the discussion just for U.S. filing alone. Are we dealing with three patient populations in terms of getting approval or are we dealing with two, really either it’s DD or NDD?
So Difei, thank you, that’s a very good question.
We have already – we have had initial discussion with the FDA regarding the three patient populations and the understanding is bad, the incident dialysis population is one that is, I mean the understanding is that all three patient population for indication are under discussion, that will be – that is ongoing with the FTA. And we believe that the results will help drive the decision.
Okay. Thank you for that clarification.
Now, for the indication in the U.S., if we just think about DD and NDD for now, for the MACE status, now we can forget about the MACE+ situation just for the MACE measurement. Have you reached a statistical non-inferiority against either placebo or EPO?
So Difei, we – so whether you reach statistically significant non-inferiority, it really depends on what the non-inferiority margin is. And in Europe we are more clear on the non-inferiority margin and we believe that we have achieved that. And for the incident dialysis, the nice thing about have achieving superiority is that no matter what the non-inferiority margin it is, once we can demonstrate superiority, we had already crossed it. And we are using the conventional standards or non-inferiority, which is well widely published for assessment of CKD anemia and have previously been used by U.S. regulator for assessment of cardiovascular safety in similar types of composite end points that standard has been 1.34 upper bound off 95% confidence interval. If we use that standard, the answer is yes, we have a chain for non-inferiority. And – so the reason that we are not as expressive in saying that is because we got – we wanted to be transparent when we state what we mean by non-inferiority.
Okay. This is very, very helpful.
So then if I could ask a question on quality of life measurement Do you see superiority or non-superiority on the ID and DD patient population? I think NDD, you said, you’ve achieved the superiority, is that the right understanding?
So Difei, the reason that we focus on testing quality of life in non-dialysis is that we are testing against placebo and we do superiority testing in dialysis, but it’s not as meaningful because when we are comparing against an active comparator that currently does not have a label for quality of life, I mean, even if you’re non-inferior, what does that mean? It’s not going to get as a label. And trying to go for superiority, that doesn’t make sense either.
So that’s why we focus our quality of life measures in the non-dialysis patient population.
Just to be clear, we only tested quality of life in non-dialysis, the 4,300 patients in non-dialysis, not in dialysis, okay?
Okay. Thank you. That’s very helpful.
So my final question is that something like you plan to file NDA September, October timeframe. But before then, there will be a FDA meeting. Would you update us or do you have plans to update investors before the NDA submission, when you have more clarifications on the data analysis, is that true?
So Difei, I’m looking at my colleague, having a regulatory and we normally do not do a press release before or at – necessary, when we go see the FDA.
However, we will certainly update our investors when we submit NDA. And I see Wayne nodding his head.
And I would also say that if there is a meaningful change in the timeline assumptions, which now or September or October submission time, we will let investors know.
Okay. Thank you. This is very helpful.
And your next question comes from Adam Walsh with Stifel. Adam, your line is open.
Hi, thanks for taking my question. This is Edwin Zhang on for Adam. Congrats on the data. A couple of questions for me, first, based on this MACE data, how do we sync off the label language if approved? Are we confident to avoid black box?
Adam, thank you for…
You’re not – you’re Edwin, okay, sorry, Edwin.
So, thanks for the questions.
Now when – what FDA puts on the label is something that they – that we may not have much control over, except that we have developed a package that we’ll target a certain label.
And so FDA has advised us that the evaluation of efficacy – primary efficacy will be based on individual studies and we have checked that box. And the evaluation of safety is FDA may – will look at various aspects of safety. And based on what we have seen, we are pretty comfortable with safety. This adjudicated composite safety endpoint was something that we had discussed with the FDA. And at this time we are quite happy with the result. The fact that we believe that the CV safety end points in our studies in the non-dialysis when compared to the goal standard of placebo, we are now seeing increased breath, that makes me think that we have a chance of avoiding some of the – those terms that is currently on the EPO label.
However, this is what we are targeting when we selected placebo as a comparator. But at the end, the assessment is really depends on the medical reviewer at the FDA. And they will – if there were an advisory committee, then there will be input from the advisory committee that FDA chooses to.
Okay. In term of the – to cut it off these – the additional positive benefits in NDD is certainly impressive, like a slower eGFR decline, the quality of life or even efficacy in the inflamed patient. Are we targeting the EPO treatment, what do you think as the agency, we will look at this benefit from roxa? Do you think this we’ll have to open a larger market for roxa and NDD?
So this is a great question.
We have had discussed with FDA on how to look at the drug and FDA has expressively advise us that benefit will be taken into consideration, as they evaluate each drug base on the combination of benefit risk. And we are hopeful of being able to reach more CKD patients you need with this highly assessable drug that is a pill and not have to go drive to the doctor’s office regularly to get shots.
And when do we expect to see the full MACE data?
We are – that’s very good question. Right now, we are sharing with you some very, very top level result, and we’ll be in discussion with our partners to that – two key places where we’ll be sending the pool MACE data, one is through the health of quality, we try to get the drug approved and to submit to major conferences. And since we just got the data, we are in discussion with partners on which conference and which journal to submit manuscript to.
It will be a lot of fun.
Can I ask you one question on DMD, I don’t want you to overshadowed by the MACE question.
When are we going to see the data, the DMD data? And for the next study, what is your plan dosing regimen and the unit point? And what kind of control arm you will use? Thank you.
So I think there is a lot to talk about your – an important idea is the left ventricular ejection fraction LVEF, we had positive numbers, this is sort of a big deal, right? But in terms of what’s going to happen next, I will let Dr. Elias Kouchakji speak to this plan.
So as we talked and mentioned by Tom, that is we – our process we are looking to go and talk with the FDA about the next step in the – with this data. And at the same time we are looking to collect this additional natural disease history data, which is very important. And then from there we hopefully soon after we will be publishing our data and add this one disease on data that we collected in this patient population.
We are looking for that early as this conference is possible that is quick would be pouring some of our data out, which was most likely will be done in the World Muscle Society.
Great. Thank you.
Okay. And we have no further questions at this time.
So I would like to turn the call back over to Tom Neff for closing remarks.
Thank you all for joining us on this call today. These data represent and culmination of many years of hard work on the part of our team in FibroGen some cases measuring a decade or more. Thank you all for your dedication to and belief our program, we actually believe we have very good data. I also would like to thank all the physicians, investigators and patients who participated in the clinical development programs, our collaboration partners and our investors for patience and continued support. We look forward to keeping you updated on our progress throughout 2019. I’d like to wish everyone a good afternoon and evening. Thank you for joining our call.
Thank you. Ladies and gentlemen, this concludes today’s conference. Thank you for participating.
You may now disconnect.