FibroGen Inc (FGEN) 2018 Q3 法說會逐字稿

Hello, and welcome to the FibroGen Third Quarter 2018 Financial Results Teleconference. My name is Michelle, and I will be your operator for today's call. (Operator Instructions) Please note that this conference is being recorded.

I will now turn the call over to Ms. Karen Bergman. Ma'am, you may begin.

Thank you, Michelle, and thank you, everyone, for joining us this afternoon.

Today we are reporting financial results and corporate updates for the third quarter of 2018. Joining me on the call today are Tom Neff; Chairman and Chief Executive Officer; Dr. Peony Yu, Chief Medical Officer; Ms. Chris Young -- Chung, Senior Vice President China operations; Dr. Elias Kouchakji, Senior Vice President Clinical Development, Drug Safety and Pharmacovigilance; and Mr. Pat Cotroneo, Chief Financial Officer.

Following our prepared remarks, Tom will discuss upcoming milestones, and we'll 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, 2017, and our quarterly report on Form 10-Q for the fiscal quarter ended September 30, 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 new information, future developments 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 the FibroGen website at www.fibrogen.com. The webcast will be available for 2 weeks from today's date.

And with that, it's my pleasure to turn the call over to our CEO, Tom Neff.

Good afternoon. Thank you, for joining us.

Let's begin with the exciting progress in our roxadustat program. We now have 9 of the 15 Phase III studies data locked and completed, 2 in China, 5 in Japan and 2 in Europe. One study in Japan is still ongoing. We are now in the process of carefully planning out the unblinding and release of 5 Phase III studies conducted by FibroGen and AstraZeneca, representing the core of the U.S. NDA. Our timeline includes collection of relevant data, completion of adjudication, unblinding, and finally, the MACE pooling analysis for each of the dialysis and non-dialysis populations. Increasingly, the focus of our project teams is to have MACE analyses completed by March-April next year. In order to complete -- March-April 2019 in order to complete the NDA filing by the end of June of 2019.

Later on this call, Dr. Peony Yu will discuss the process for reporting on these Phase III trials and the adjudication and pooled analysis.

I would like to take a moment to highlight patient populations that I am excited about that are coming to the point of reviewing results, incident dialysis population that has never been systematically study in CKD anemia as well as the placebo study in patients that have hard-to-treat anemia and stage 4 hard-to-treat CKD anemia.

First, we will report results on over 1,000 patients in incident dialysis, the first-ever well-controlled comparison of anemia therapies in dialysis. Prior studies have relied on fully stabilized hemo or peritoneal dialysis patients who have been treated with ESA to achieve stable hemoglobin and anemia correction for many months, if not years, before entering a trial. Thus, these studies included the best responders to recombinant EPO, which we believe is a narrow population, probably only addressing about 1/2 of the total dialysis initiation pool in the real world. Thus, best responders to EPO are over-represented in the control pool as a function of study design. We believe that a significant portion of the general dialysis population has, to date, not been studied. Now we have the opportunity of properly comparing 2 conceptually distinct approaches to anemia therapy as well as having the opportunity to demonstrate the treatment benefit of roxadustat in this broader population.

Second, our the U.S. non-dialysis study versus placebo has been conducted in a population after broad ECUs in non-dialysis settings had been restricted and hemoglobin levels at entry are much lower than prior studies would've allowed. Our study is unique in that there are no well-controlled, large CKD patient studies, where large components of the patient population is in stage 4 and stage 5, including a large number of single-digit eGFR patients at baseline. In effect, this study measures the benefit of anemia therapy in what is now called the peridialysis population.

With both of these study areas, I know there is a large audience of physicians, regulators, patients that are interested in seeing results from these study designs as the information will greatly increase the understanding of these patients with later-stage renal disease.

In October, we and our partners presented 4 studies at ASN Kidney Week 2018 in San Diego. There were 3 posters for -- and an oral presentation for a fourth, all presented roxadustat Phase III data from China and Japan. It happened that our finding meetings with our Chinese regulators, the CDE, were scheduled on very short notice in face-to-face sessions in Beijing. Between travel, the briefing documents and the meetings in Beijing, the core anemia team could not be at ASN. As a result, there was not much time for the team to address the follow up to the poster sessions.

As I assessed the situation, I saw a lot of very smart people getting the wrong message on certain data. Normally, I view this as the bricks of the game in the U.S., but in this situation, the real issue for investors and analysts is whether there's any data from China or Japan representing real safety signal relevant to the upcoming global program readouts. Thus, I instructed our teams to provide clear and concise explanations to help investors and analysts to understand where the rationale is based, in fact, are not going to support a safety signal in the global program.

Dr. Peony Yu will address these assessments and explanations later on this call.

As you know, roxadustat is the most-advanced candidate in the HIF-PHI class worldwide, and we expect approval for this first-in-class oral anemia therapeutic in dialysis with or without prior ESA use by year-end in China. We are completing final steps in a rigorous review process, including multiple regulatory agencies, and covering all of our clinical, non-clinical and manufacturing data.

We also expect NDD, or non-dialysis, to be approved as part of the level. However, the timing is pending a firm scheduling for clinical-site inspections by the CFDI. Many of you may know CFDI is experiencing back up in their scheduling, so we're not quite sure at what point in the first half of the year we'll get inspected.

Chris Chung, our Managing Director for FibroGen in China, will provide updates on where we are in completing the final steps and the approval process for roxadustat in China.

Equally exciting for the FibroGen anemia team, the roxadustat NDA was filed in Japan in September by our partner, Astellas, for the treatment of patients on dialysis who are suffering from anemia associated with CKD.

Peony will speak more about the new data presented at ASN on 2 of the 4 Japan Phase III studies in dialysis-dependent patients. These studies were sponsored by Astellas.

I would now like to turn the update to pamrevlumab. Our pamrevlumab program, which centers on the understanding of the biology of fibrosis and the role of CTGF in various diseases and systems, is proprietary to FibroGen. Our first-in-class anti-CTGF agent, pamrevlumab targets mechanisms involved in fibrosis progression-related diseases, such as idiopathic pulmonary fibrosis, pancreatic cancer and Duchenne muscular dystrophy. In each area, we are seeing potential for unique therapeutic benefit for patients otherwise facing high mortality risks.

We completed our end of Phase II meetings with the FDA where we reviewed and came to agreement on Phase III study designs for pamrevlumab, both in IPF and locally advanced unresectable pancreatic cancer, or LAPC. Our teams are planning on commencing enrollment in our Phase III LAPC study in the first quarter of 2019, followed shortly by our Phase III IPF study.

Pamrevlumab has demonstrated anti-fibrotic activity by reversing or slowing fibrosis in IPF and improving the resectability and resection rates in LAPC. No safety signals have been identified in preclinical and clinical evolution to date. Both of these programs have been granted fast track designation by FDA.

I must also note our program in non-ambulatory Duchenne muscular dystrophy. We believe there may be an important and much-needed role for pamrevlumab in DMD, which is the most common and severe form of muscular dystrophy, and where there is significant ongoing fibrosis that causes impairment of cardiovascular and pulmonary systems as well as severe decline of muscle strength. All 21 patients in our Phase II pilot study will complete 1 year of treatment in March, and we will be able to announce data from that study at that time.

Dr. Elias Kouchakji will talk about these 2 Phase III pamrevlumab studies in our DMD program later on the call.

Let me finish here by addressing finance. As of September 30, 2018, FibroGen had $722 million in cash. The NDA for roxadustat was submitted in Japan by our partner, Astellas, for anemia associated with dialysis-dependent CKD in September, triggering a $15 million milestone payment to FibroGen from Astellas. For accounting purpose this is largely recognized as revenue in the last quarter based on current GAAP rules.

In addition, we have additional cash on hand being reported as a result of sales of roxadustat API to Astellas for commericial use in the Japanese market. Pat will provide more details on this development as well as our cash position and our year-end projections later on this call.

I would now like to turn the mic over to Dr. Peony Yu for updates on the anemia program. Peony, please go ahead.

Thank you, Tom.

As Tom said, it's a busy and exciting time for the roxadustat program. And I'd like to first talk about our progress in China and Japan, highlighting the 4 Phase III studies recently presented at the ASN Kidney Week, then our progress in the U.S. and EU program.

As you know, we are pursuing 4 parallel regulatory pathways. China NDA reviewed by CDE have been going very well, as Christine will discuss further, and the Japan NDA was filed with the PMDA in September. At the ASN Kidney Week in October, there were 4 roxadustat Phase III studies presented, 2 from the China Phase III studies designed to meet China's domestic approval requirements, and 2 from the 4 Phase III studies which are part of the Japan NDA package.

For clarity, these studies are not part of the MACE analyses for the U.S. and EU packages and were not designed nor power to draw conclusions about cardiovascular outcomes.

We have come across a few questions around these presentations, particularly whether certain data from the China or Japan might be a safety signal for the global program. I would like to provide clarification in 3 areas, and then I shall discuss key takeaways from these 4 positive Phase III trials.

First was actual mortality rates seen in the 2 studies. To provide context, patients with end-stage renal disease requiring dialysis are known to have significant co-morbidities and a high mortality risk. The considerable data available on dialysis patients from government-sponsored registries and academic publications provide helpful background rates. Dialysis registries in Japan and China have shown the annual mortality rate in dialysis patients to be 9% to 10% and 6.8% to 8.2%, respectively. And in the U.S., it is approximately 17%. According to USRDS, it was 10.3% annually in the Omontys Phase III study. The annualized mortality rate in our China and Japan dialysis studies was only 2%, which was much lower than the reported rates in these publications.

The extremely low rate of 2% reported in these small studies should not be seen as a signal as it is much lower than the historical rates with ESA in the similar patient population, nor is it predictive for the much larger U.S./EU outcome trials in which many hundreds of events are required for statistical powering.

Second, in the China studies, while there was an apparent imbalance in adverse event reporting by investigators of hyperkalemia and metabolic acidosis, as we reported at ASN, no imbalance of potassium or bicarb abnormality between treatment arms was found in the central lab data for these studies. Rather, these lab values for the study are consistent with, and characteristic of, their chronic kidney disease patient population where the diseased kidney were no longer able to properly regulate potassium levels and acid-based balance, even before study participation. These data have been reviewed with the Phase III lead investigators, CFDI inspectors and CDE as part of the NDA review. Based on our analysis of the data, we conclude there was no causal relationship between roxadustat use and these -- and abnormal laboratory values in potassium or bicarbonate levels. Moreover, a similar imbalance of these events have not been reported in our other completed studies. We, therefore, believe there is no basis to view this as a signal or to expect this to be an issue in the global program.

Finally, I would like to explain the higher rate of early-study discontinuation in roxadustat arm compared to EPO in the open-label China dialysis study. We believe this results largely from the inherent bias against an experimental agent when compared with standard of care in an open-label study design. This often leads to asymmetric dropout rate in the experimental arm where investigators have more experience and patients are more familiar with the standard of care.

In contrast, neither is the Japan dialysis study nor the China non-dialysis study, both of which are double-blinded, show this kind of differential dropout rate.

In addition, in this study, patients in the comparator arm were provided with the premium Kirin-Amgen EPO, which costs 3x the price of the usual knockoff EPO that they typically receive under standard reimbursement for dialysis patients. And this factor would perceive value in the -- higher value in the study comparator drug could have further contributed to the difference in discontinuation rates between the treatment arms.

We hope this information proves helpful.

Now let me turn to the major results. Overall, we have had several positive findings in the China and Japan studies. Primary efficacy endpoints were met with acceptable safety profile, and the results are part of China NDA as well as Japan NDA. We are excited about the progress in China NDA review, on which Chris will provide more detail in a few minutes.

Of note, it is exciting to see roxadustat's superior efficacy over EPO in the subset of inflamed patients in the China dialysis study without incremental dose requirement in the subgroup of patients with elevated CRP. This differentiation from ESA in overcoming EPO hyper-responsiveness, which is ESA's Achilles heel, is expected to be translated into safety benefit over EPO in larger studies, powerful cardiovascular outcome evaluation. We are eager to see U.S./EU data for this thesis.

In Japan, Astellas has submitted NDA for CKD anemia in dialysis patients that included the 4 positive Phase III dialysis studies and has announced completion of 1 of the 2 Phase III CKD non-dialysis studies with plans to submit supplemental NDA upon completion of the second non-dialysis Phase III study in Japan.

Turning to the much larger U.S./EU roxadustat program, which is designed and powered for cardiovascular safety evaluation. There are over 9,000 patients, of whom over 6,000 had already been on the study for over a year. These studies have been undergoing periodic safety review by treatments by the independent DSMB for over 4 years. The most recent DSMB review was on August 1, 2019, with recommendation of continuing per protocol, just like the assessments in all prior reviews.

We and our partners have completed patient treatment in the 7 Phase III roxadustat CKD anemia studies supporting U.S. NDA and European MAA. Astellas has reported positive top line results on ALPS, the first global Phase III placebo-controlled study in CKD patients not on dialysis, and has also completed PYRENEES, a Phase III dialysis study in Europe.

I'd like to take a moment to outline the data readout and submission timeline of the U.S./EU roxadustat program.

We are currently planning to report the rest of the individual Phase III top line study results in CKD anemia in the fourth quarter of this year. This is for both dialysis- and non-dialysis-dependent patients. We are on track to reporting adjudicated pooled MACE results and submission of the U.S. NDA in the first half of 2019 in collaboration with AstraZeneca, our U.S. partner. Working with Astellas, we are also preparing to work together on the MAA submission in Europe in 2019.

I'd like to now turn the call back to Tom.

Thank you, Peony.

Chris Chung, our Head of China Operations, will now share updates on our current activities and regulatory progress. Chris, please go ahead.

Thank you, Tom.

We are progressing towards market approval in China. In the last quarter, we successfully completed an expert review meeting, which is a critical step in the approval process. Efficacy and safety data, the label and post-approval commitments for both dialysis and non-dialysis were addressed with the Center for Drug Evaluation, or the CDE, the technical review arm of the NMPA, in the presence of external experts. We've successfully completed pre-approval manufacturing inspections by the China Food and Drug Inspections, CFDI, of both API and drug product. We've successfully completed analytical testing of samples produced by FibroGen China at the national testing agency, the National institute of Food and Drug Control, also known as the NIFDC.

We believe we're very close to roxadustat approval by the NMPA in the treatment of anemia and chronic kidney disease in both dialysis and non-dialysis. For dialysis-dependent patients, we completed the clinical-site inspections in the second quarter of this year.

While there is no equivalent of the PDUFA action date in China, we anticipate approval for dialysis by year-end. The dialysis label will cover both hemodialysis and peritoneal dialysis, and regardless of whether they have been ESA-treated or are ESA-naïve.

For non-dialysis, we are waiting in queue for clinical-site inspections to be completed. We expect to be able to expand the label to include non-dialysis in the first half of 2019.

Thank you again for your time. Tom?

Thank you, Chris.

Turning to pamrevlumab in our IPF Phase III study. We are targeting a significant untreated population of patients who are currently not on non-standard of care.

I would like to ask Dr. Elias Kouchakji to discuss our 3 pamrevlumab programs in IPF pancreatic cancer and DMD.

Elias, please go ahead.

Thank you, Tom.

Starting with IPF. We have presented clinical and preclinical data this quarter at the European Respiratory Society International Congress, ERS, in the 20th International Colloquium on Lung and Airway Fibrosis, ICLAF. The data highlights the positive safety and efficacy results from our Phase IIb study, PRAISE. Pamrevlumab demonstrated statistically significant attenuation of fibrosis as measured by FVC and quantitive HRCT and statistically significant advantage in disease progression defined as decline in FVC percent predicted of 10% or death.

Following our end of Phase II meeting with the FDA, we are planning a placebo-controlled double-blind Phase III study, similar in design to our PRAISE Phase IIb study. We have plan to enroll approximately 500 patients who are currently not on standard of care. The primary endpoint for this trial will be change in FVC from baseline.

Moving on to LAPC. In the first quarter of 2019, we will initiate a randomized double-blind, placebo-controlled Phase III study, evaluating pamrevlumab therapy in combination with gemcitabine and nab-paclitaxel as neoadjuvant treatment. We plan to enroll approximately 260 patients in this study.

This study is of similar design to our Phase II trial in which we'll assess resectability and resection with the primary endpoint of overall survival. At the end of the treatment period, we will evaluate the resectability and resection rate. And should the recession rate favor the pamrevlumab combination arm, we will request a meeting with the FDA to discuss a marketing application under the provision of accelerated approval. In this study, we will also continue to follow patient for overall survival.

Now I'd like to move to the DMD program. This is another important program for pamrevlumab. Our Phase II study in Duchenne muscular dystrophy is progressing very well.

In September, during the muscle -- the World Muscle Society in Mendoza, Argentina, there was considerable interest in pamrevlumab from the scientific and the clinical community and from the patient advocacy groups.

Our Phase II study is evaluating 21 non-ambulatory young patients over a 3-year time frame. As Tom mentioned, all patients will have completed 1 full year of pamrevlumab treatment by the end of the first quarter of 2019. We plan to update on the progress of this study in the second quarter.

Finally, I would like to say in summary we are excited to be advancing pamrevlumab into late-stage clinical development and to be working on a first-in-class therapeutic candidate with potential to help patients with these serious and life-threatening conditions.

Thank you for your time today, and Tom, I will turn the call to you.

Thank you, Elias.

Pat Cotroneo, our Financial Officer, will now discuss financial highlights for the third quarter of 2018. Pat, could you go ahead please?

Thank you, Tom.

As announced today, total revenue for the quarter ended September 30, 2018, was $29 million as compared to $40.6 million for the third quarter of 2017. For the same period, operating expenses were $71.8 million, and net loss was $42.6 million or $0.50 per basic and diluted share as compared to operating expenses of $63.3 million, a net loss of $24.5 million or $0.32 per basic and diluted share for the same period in the prior year.

Included in operating expenses for the quarter ended September 30, 2018, was an aggregate noncash portion totaling $16 million, of which $14.3 million was a result of stock-based compensation expense as compared to an aggregate noncash portion totaling $11.5 million, of which $9.6 million was a result of stock-based compensation expense for the same period in the prior year.

As of September 30, 2018, we had $722.6 million in cash as compared to $733.7 million at the end of Q2 2018.

For these purposes, total cash refers to cash as well as cash equivalents, receivables, investments consisting primarily of investment-grade corporate debt and restricted time deposits related to our building lease.

We are currently projecting a year-end cash balance in the range of $720 million to $730 million. This has increased from our previously disclosed range of $650 million to $670 million due primarily to an anticipated shipment to Astellas of drug supply totaling approximately $42 million in support of anticipated 2019 pre-launch activities in Japan.

Thank you, and I would now like to turn the call back over to Tom.

Thank you, Pat.

It has been incredibly busy and exciting year for FibroGen people year-to-date. We have had 15 Phase III studies in play, and we're launching 2 major Phase III programs.

I'd like to run through important upcoming developments and milestones for the company and try to keep it simple as much as possible.

Starting with roxadustat. We plan to report Phase III top line study results for roxadustat, Q4. And we're really focusing on trying to get the MACE reported out by March-April time period in 2019. We are expecting to submit U.S. NDA for roxadustat for the treatment of anemia and dialysis-dependent and non-dialysis-dependent CKD in the first half of 2019. And we will also be supporting the filing of the MAA by Astellas in Europe in the same time periods.

In China, we anticipate a market approval decision for roxadustat in dialysis. This is dialysis without a restriction on prior EPO use. And we expect the second part of label NDD to be approved upon completion of inspections, which would be probably Q2 of '19. So the first half, we're looking right at year-end, dialysis, and NDD a few months end of 2019.

In Japan, for the dialysis filing, we expect a regulatory review within a year.

We are entering into a period of significant upscaling for our pamrevlumab program related in part to the Phase IIIs and in part to the scaling production technology for commercial activity.

And locally advanced pancreatic cancer, we'll be commencing enrollment in our Phase III study in the first quarter of 2019. Sometime -- very close to that, we will also be initiating our Phase III IPF study. These are on parallel tracks with different teams.

And DMD, which I think will become a very important narrative as time goes on, we look forward to completing 1 year -- full year -- 1 full-year treatment at the end of the first quarter 2019.

With that, I'd like to turn this call back over to Karen for Q&A. Karen, please.

Thank you, Tom. And Michelle, if you could kindly open up the line for questions.

(Operator Instructions) The first question in the queue comes from Difei Yang with Mizuho Securities.

So just a couple, first one on MACE. Just trying to help us to frame what we should be expecting on the upcoming data? We looked at the Japanese patient population, so the mortality rate was roughly about 1.3. Sounded like per literature and per previous clinical trials, mortality rate, especially for dialysis patients we -- for the EU and U.S., I think we should be expecting a number that is much bigger than 1.3. Just wanted to confirm if that's the right way to think.

Let me try to clarify this a little bit. In Japan, we filed 4 studies for the NDA. I don't know if you're counting all of them in the 1.3 number or not. But nevertheless, the 1.3 is very similar to what we've seen for combined Japan and China. So what we know from the Japan dialysis society website that tracks mortality is that annual mortality rate, which is a rate of mortality per 100 year patient exposure, that's how they express it, is between 9% and 10% annually for many years going back in time. And so I think that, that's a very useful data point. Second, it is true the U.S. RDS is much higher, 16% or 17%. I think you may have mentioned this in your recent research report in some way. You also provide some other citations in this area. We look at China, there's a few different surveys. And we've also had our sites provide real-world data. And the range is low of 6.5% and up to 11% or 12%, depending on which study you're looking at. The Beijing dialysis registry is 8% -- 6.5% for -- up to the first 6 years in dialysis, as I recall. The way we look at this issue is that our responsibility and accountability right now is to clarify if the data in China or Japan, which is proving to be acceptable in China and we hope it's acceptable in Japan, whether it reads out on the global program for safety signal. And so I think it's very clear that we don't have a basis to say that at 1.3 or in the totality, at right around 2% in this rate over 100-year PEY -- 100 year to PEY exposure denominator. So that's one answer. And then the second part is that because we have a lower rate, we don't quite know what that means in the longer run. It might mean, as we enroll more patients and gain more power and look at the U.S., EU data, that we have an outstanding rate. But we don't know that right now. So we're not able to use it as a predictive tool at 2%. We can't just simply say we've had 600 patients in studies, and we're looking at a mortality rates at 2%. This goes directly to investor concern as whether each stage of data would read out as a signal on our MACE safety endpoints because death is 70% of the MACE composite. So it is right on the money in terms of being relevant.

Okay. Thank you, Tom, for the additional color. And then just turning to pamrevlumab. So the IPF study and pancreatic cancer study will both start early next year. Would you guide us on roughly what's the time frame we'll see readouts?

Your question is when we think we'd get the Phase III readouts?

Yes.

Yes, I think it's a bit early right now for us to do that. We will have a better sense of this next year. We do not expect it's 1 or 2 years. It'll be longer than that from the time of enrollments. So -- but how much longer remains to be calibrated carefully.

The next question in the queue comes from Terence Flynn with Goldman Sachs.

This is Gavin on for Terence. I just had a quick question regarding MACE. When the readout comes, should we expect both the DD and NDD populations to readout? And will you provide any clarity on secondary endpoints? If so, which secondary endpoints are you guys focused on?

Okay. So let's do this step by step. We first have to unblind studies. So there's 5 between ourselves and AstraZeneca that we're hoping can get done in December. And then after that, we have certain procedures to work through to get the MACE data from these studies properly booked into software to evaluate comparison. One comparison is versus placebo in non-dialysis and the other one is versus ESA's in dialysis. We do not know now whether we will be trying to announce any of these results of the unblind 5 studies or the 2 MACE pools on a tight, same day-type of disclosure. Right now, we are assuming that's going to be probably too difficult to do, and so we may -- day-on-day rather than all at once. However, the planning for this is scheduled in meetings later this month and early next month with AstraZeneca. So we'll probably have more information and more color on exactly how we'd do the unblinds and the pool reporting. There is considerable other data that's being generated. And on this point I would like to ask Peony to assess what she expects in terms of when anything would read out. Peony, could you address that, please?

Tom, thank you. Yes. Gavin, you asked an excellent question. Other than the primary efficacy endpoint of changing hemoglobin from baseline in non-dialysis as compared to placebo and in dialysis it's in comparison with active control using epoetin alfa. We are -- now you're aware of our pooled cardiovascular analyses that we expect to read out for MACE and additional or secondary endpoints, which are also very important would be MACE plus. And when the rubber hits the road will be for hospitalization rate in both pools of patients. I am particularly excited about the opportunity to make a big difference and be able to show such difference in a placebo-controlled setting in non-dialysis. And we -- when we are combining the study, we will have quite a bit of power. And my hope for the Holy Grail will be to slow down renal progression by measuring eGFR changes. And we would also like to demonstrate an improvement in patients' quality of life. And so there are number of others, but I'm not going to go through the entire list. But these are the ones that are most important in these patient populations.

Yes, correct. We'll report it simultaneously. We'll be able to report it.

So Tom wanted me to state whether the readout on the secondary endpoint will be the same time with individual study or MACE data. We are still coordinating the data analyses with our partners because it is from -- these results will be coming from studies conducted by us, by Astellas and AstraZeneca. So in our next call, we will be able to provide more granularity on timeline of announcing these exciting results.

Gavin, do you have any other questions please?

And maybe one more on pamrevlumab, just the IPF trial design. When should we expect additional details for that in terms of maybe exclusion/inclusion criteria?

So let me turn this question over to Elias. Dr. Kouchakji, could you explain the status here?

We are -- in currently finalizing the protocol, and we will be submitting this to the agency to get their final blessing very soon. Soon after, we will be putting this into ClinicalTrials.gov with all these additional inclusion and exclusion criteria. I just want to repeat that this study will be closely similar to our PRAISE study. So there will be not major differences between our inclusion/exclusion criteria than what we have enrolled in our 067 study, PRAISE.

The next question in the queue comes from Geoffrey Porges from Leerink.

Two quick questions. First, Peony, it sounds like you have some effective unblinding in the placebo-controlled study in China. I'm just wondering if you could tell us what procedures you had in place to ensure that you won't have any such unblinding in the U.S. study that's going to read out shortly. And then secondly, on pamrevlumab, the total study size that you were discussing is substantially smaller than that -- what was completed for the other drugs that have been approved in the indication. Can you help us understand whether -- how we should be confident that the 500-patient study will be sufficient when the incumbents have studied in the range of 1,000 to 1,500 patients in 2 different trials in most cases?

So yes. So Peony, go ahead and explain in China which study was blinded and which one was unblinded to clarify the issue.

Yes. So in China, the dialysis study was open label. That's a 304-patient study 2:1 randomization and with comparator arm treated for 26 weeks. Now for the non-dialysis study, the blinded period was only limited to first 8 weeks because the Chinese regulator felt that it was not ethical to maintain patients on placebo too long. And the patients in the placebo arm were crossover to active for -- and were treated for 18 weeks so that both arm patients will have an opportunity to be treated for a total of 26 weeks. And then a subset of these patient who enrolled first enter extension to have 52 weeks total. Now the difference from this, in the U.S.-European global program, there are 3 non-dialysis studies conducted, 1 by us and the other 2 by each of our 2 partners, where the blind as well as the placebo controlled were maintained throughout the entirety of the study. And it was an event-driven trial, so -- and with variable treatment duration so the patients were treated anywhere from 1 year, for 2, 3 years.

Thank you. Elias, maybe you can address the rationale for the IPF trial design that we developed.

Geoff, as I understand the question is about the study -- the size of the study. Just like to remind that we run already 2 Phase II studies. The Phase IIa and Phase IIb. The Phase IIb, specifically, which is a placebo controlled and the result was statistically significant. And the size of the study and the data from the study provided us enough data to have a statistical plan to power the study to be able to meet statistical analyses that could meet the regulatory requirements for approval. This statistical was reviewed by the agency, and the number of subjects was agreed by the agency. So comparing to other programs that they did not have a sizable study similar to the Phase IIb as our PRAISE study, that has led to our understanding how this can remind everyone that this is in pirfenidone. One of the studies have failed that statistical analysis. I think we are, at least at this time, in a position to understand our data.

Elias, that's helpful. And perhaps you could also -- I think he's asking a question about, given somebody else is doing 3,000 or 4,000 patients in Phase III now, why did the rationale for us result in the 500-patient size.

The other company was planning that is large 1,500 size of patient plus. They don't have a Phase II, which is statistical positive results, similar to our PRAISE study. That is by itself it is the most valuable tool that we have. At the same time we have this, not only we've seen this in the FVC percent predicted and we see volume, we have seen it in our [QSRTC] analysis. At the same time, we have

(technical difficulty)

progression. So by taking all this totality of evidence, we are able to build this power -- the power that's sufficient with this number of patients.

Yes. So I think the other aspect that needs to be mentioned is that in our discussions with the FDA, we had several expert advisers who have spoken. It's very clear that a minority of the population [U. S.], for reasons that Elias can explain, simply won't to use the existing on-market medications. And as a result, the statement could be made that there is on the order of 30% to 40% (inaudible) of patients who have compelling unmet medical need that is urgent in character. And there is presently no way for them, given the circumstances, to be treated. And as it relates to any scenario where we combine with the currently approved products, we inherit their safety profile. And the problem with those safety profiles is that there are lots of issues that cause -- a portion of this minority, the patients that are unable to use those other medicines. Elias, perhaps you could add just a little bit more there?

Thank you, Tom. This is an important point to emphasize. Pamrevlumab has been -- we have over 500 -- close to 600 patients exposed on this product, some of these patients up to 5 years. We have not identified safety signals. So this is compared to other approved product. The safety signals are totally different. There is a good number of patient population who are being treated the standard of care and they are dropping off the treatment. Additionally, there are -- some of them are not responding to the treatment. The quality of life that is affected, either from the skin reaction, the photosensitivity, or the liver enzymes changes which is serious and fatal. So all these are a -- they're big differences that is added to our pool of population. Additionally, we know that there is a population that are rejecting the treatment. Although they -- not only the population themselves, some of prescribers who are not eager to prescribe based on the safety profile of this molecule. As you said, there is 40%-plus are not treated and they need a therapy. This is why this study will provide this tool to prescribers to be able to treat this population who have -- that will have nothing else to be treated with.

I think for now, we'll leave it at that. We can talk more next quarter, maybe. Next is, Joel Beatty.

Yes, Joel Beatty with Citigroup.

On the post the Kidney Week, I had a question about the different trends in hemoglobin in the dialysis patients. Generally, it looks like in the Japan study, hemoglobin and roxadustat and the ESA control group was relatively close throughout the study. However, in the China study, there is a wider gap of around 1.5 grams, especially in the early part of the study. So could you describe what caused the differences between the 2 studies? And which would you expect the U.S.-EU program to be closer to?

Okay. So Peony, go ahead and address those issues.

Joel, your -- that's a very good question. The China study and the Japan 0307 study are very similar in study design. In both studies, patients who were previously treated with EPO were randomized into receiving either roxadustat or a ESA. In the Japan, there is a very small difference, but a very important difference, which I shall highlight. And that explains why we see this difference in hemoglobin comparison. In the Japan study, patients who were highly responsive to EPO, in other words, they were on stable doses of ESA prior to randomization and their achieved hemoglobin were already between 10 to 12, were -- have -- were randomized. However, in the China study, we took in patients with baseline hemoglobin between 9 to 12. So there lies the difference of having more hyper responders and patients with elevated CRP in the China study. And that is the sub-population where our drug is seen as superior to EPO in the China study.

So you're just in effect saying that in Japan, you have double-polar study patients, very stable that is a source of randomization?

Yes.

And the range is very narrow as to what you're treating to with comparison to darbo? Whereas in China, you have a much broader range of patients. Inflammation is a factor that's allowed to very across the natural population. Whereas, in Japan, it's restricted in the enrollment?

Tom, that is correct. In Japan, 38% of the patients on dialysis have achieved hemoglobin below 10. But for the purpose of the study, the most enriched population was selected. And in the U.S.-EU program, the entry criteria is more -- is closer to the China study criteria in the conversion study. And furthermore, as you mentioned, in Tom's part of the script that we have over 1,000 patients in an incident dialysis study where baseline hemoglobin is even lower. We did not set a bottom. So we expect to be able to have enough patients who are inflamed for us to make such comparison.

Michael?

Michael Yee.

Tom, two questions. One is, AstraZeneca said this morning in -- overall that they expect to give, obviously, an update by the end of the quarter, but I think suggested along with some non-MI or just overall pooled safety data. Maybe you could appreciate that The Street does want to know about overall safety when you report on some of this data. So what can you say as to what you'll be able to report or at least comfort The Street on as regards to pooled safety overall when you report on it in December? And the second question is related to a question that Peony just emphasized relating to hyper responders. Do you believe that you'll have enough patients in the dialysis hyper responder pool that you'd be able to make significant claims as to either MACE or other cardiac benefits in that population? And how important is that, either to labeling or being out of the bundle, all of those types of things? How important is that? What could you see there?

Okay. Peony, do you want to try to take on the pooled safety first? We'll come back to the hypo responders after.

Sure, Tom. Mike, yes. We agreed that the pooled safety analysis is very important. And in December, we are expecting to report out the results from the individual study data. However, it's only after individual studies have been -- the data has been locked and the data becomes available for pooling and -- before we can provide the pooled analysis. So it will be -- the pooled safety data will be provided as a second step.

Yea, I think that's right.

Yes. I think our partner AstraZeneca and Astellas had given the similar message recently.

Okay. So when you report out the individual studies, would you be able to at least make comments either individually or overall, tell The Street that you feel very good on safety overall, whether individually or not?

Yes, sure. So in the case of Astellas, they have completed a couple of studies and they have stated the efficacy readout and a generalized statement about safety. And that's been their approach. AstraZeneca and FibroGen are meeting later this month to formulate how we might want to do this. And I don't know to what degree it might vary from what Astellas is doing. We do not want to provide a forum that could conceivably create bias for adjudicators that are working on data. And so we will be very cautious about the timing when the unblinding is happening to make sure that we are at a point with the adjudication where there's no bias then that can be asserted. And so that is a factor. In addition, because MACE itself is blinded, it's a blinded exercise. Anything that causes any understanding or readout, we'll probably -- just for the sake of conservatism, we'll probably wait till the MACE unblinding to report. That leaves some other things open to talk about, but as I said, we have not made decisions yet on that front. So I hope that settles it.

Okay. The more you can say, probably the better. Even if it's not about MACE, obviously.

Sure. It's the tension of various factors, and we're studying it very closely. The hypo response idea and questions are very well thought out. So I would offer that we do expect that there's going to be a portion of the U.S. population that has the characteristics of hypo response, which is to say, ordinarily due to inflammation. They would be using much more EPO and much more iron to achieve any incremental change in hemoglobin. There has been an effort to -- in the incident population approvals in the U.S. to include patients of that type. Our audience for that work is really with CMS. So they're not going to be concerned about the statistical readouts and whether or not we have adequate powering. They're much more in the zone of do you have this mechanism around hepcidin and inflammation? Do you have incremental capability with the medicine that enables us to treat what they consider is about 30% of the hemodialysis population that in their eyes does not have adequate standard of care with current therapies. And so the conversation is more along the lines of CMS encouraging us to continue on the path of working on a hypo response based on the data from Phase III. To me, that's the core of what matters in terms of what's coming out of the Phase III. There's a team working on this with CMS and they are aware of the study design and so on. So my comments sort of take that all into account.

Next question comes from Andy with William Blair.

My question has to do with a policy move. It seems like in the dialysis space, we're kind of moving towards integrated care. Just wondering, Tom, what's your view on that. And how does that impact your original strategy in terms of moving roxadustat out of the bundle as a differentiator?

Yes. So -- I mean this is a very dynamic situation. And it's going to be something people have to watch very closely. Let me start with the principal used in our trial design. With CMS, we have briefed the ability of roxadustat to regulate hepcidin to down regulate it and conditions of inflammation which allowed creating red blood cells. And the notion of that is that it allows CMS to make a decision that this is new technology which gets into a zone where they can talk about separate reimbursement if the data justifies that conversation. And so that dialogue has been going on, on a path that's not really in the public domain right now. But we expect sometime next year that there will be some decision-making by CMS in relation to that type of concept as it relates to our program. In addition, we know that both of the companies that are large dialysis organizations, or LDOs, those being Fresenius and DaVita, are encountering increasing the environment where many of their sites cannot achieve cash flow-positive results. And so it's causing a great deal of thinking and rethinking about how to define the populations differently. And both companies are clearly experimenting with, in effect, offering insurance prior to dialysis for patients that are hard-to-treat CKD so that they can get the patients on board much earlier. And so to me, that's a real important arena to monitor as time goes on here. And I know that with the people that we are in contact with in these organizations, there is a really high degree of demand to understand our data because they are seeing that it is potentially a fulcrum where there's leverageability into a strategy. And so we are -- we've been duty-bound and promised to be able to share data as soon as we can. So I think it's a really interesting time. As to the bundle and what happens with the bundle, it's, in my eyes, certainly an issue with EPO and EPO knockoffs or biosimilar EPOs or whatever people are talking about. But I also think that in our circumstance, there was this encouragement from the outset that because you're a different mechanism, CMS has the capability, if warranted by data, to address the beneficiaries that do not have adequate standard of care under current therapy. And so their -- the lense that the prism through which they do that evaluation starts with decision that is a portion of the patients that don't have adequate care and then goes on to evaluate what we have. And so how that fits into the larger CMS agenda is still not easy to see because there's a lot of things that are being considered by CMS as they sort of reinvigorate regulatory framework after several years of sort of post-bundle. Just -- so everybody's being very exhausted by what caused all the things that led to the bundle. We saw a big change in viewpoint a couple of years ago. I don't know if that's any help, but that's what we see.

There are no final question in the queue, sir, so I'll turn the call back over to Mr. Neff.

To everyone on the call, thank you very much for joining today.

We are preparing in the U.S. and China for transformational clinical and regulatory events. With our partners for roxadustat, we have 2 new drug applications on file in China and Japan, and hope of a third in the U.S. in 2019. In fibrotic disease, pamrevlumab is continuing to show great potential in treating conditions where there are limited to no options for patients and high degree of unmet need. And we are very excited to be able to move into Phase III trials in IPF and pancreatic cancer. And I look forward to sharing DMD Day. There was very much excitement about that as well.

I would like to thank all FibroGen team members for what we achieve every single day in these programs. This year has been a torrid pace, lots of people who've been working right up to the limit. We really appreciate across the board the efforts made in our company. We are all focused on trying to develop these new technologies as medicines that have the ability to improve lives for patients who face, in many cases, very bleak circumstances.

In addition, I'd like to express my gratitude to our partners, AstraZeneca and Astellas, for their ongoing support and assistance and motivation and incentivization and so on. Has many, many different aspects and for all the high-quality people we work with. And likewise with our investor community, I am most privileged to be able to work on these projects in my life. And this happens because investors support them. And we endeavor to be as transparent and clear as possible in our reporting, which is going very fast right now. So we'll do our best to get it up to date as fast as we can as a lot of data coming.

We look forward to keeping everyone updated on our progress.

I would like to wish everyone a good afternoon, and good Thanksgiving holiday coming up. Thank you all for being here.

Ladies and gentlemen, this concludes today's teleconference. Thank you for participating. You may now disconnect.