Agios Pharmaceuticals Inc (AGIO) 2018 Q3 法說會逐字稿

Good morning, and welcome to Agios' Third Quarter 2018 Conference Call. (Operator Instructions) Please be advised that this call is being recorded at Agios' request.

I would now like to turn the call over to Renee Leck, Associate Director, Investor Relations.

Thanks, Nicole. Good morning, everyone, and welcome to Agios' Third Quarter 2018 Conference Call. You can access slides for today's call by going to the Investors Section of our website, agios.com.

With me on the call today are Dr. David Schenkein, our Chief Executive Officer, who will review key business updates and milestones for 2018; Dr. Chris Bowden, our Chief Medical Officer, who will provide an update on our clinical development activities; Steve Hoerter, our Chief Commercial Officer, who will provide an update on the launch of TIBSOVO; and Andrew Hirsch, our Chief Financial Officer, who will summarize our third quarter 2018 financial results. Dr. Scott Biller, our Chief Scientific Officer, will also be available for Q&A.

Before we get started, I'd like to remind everyone that statements we make on this call will include forward-looking statements. Actual events and results could differ materially from those expressed or implied by any forward-looking statements as a result of various risks, uncertainties and other factors, including those set forth in the Risk Factors section of our most recent Form 10-Q filed with the SEC and any other filings we may make with the SEC. In addition, any forward-looking statements made on this call represent our views only as of today and should not be relied upon as representing our views as of any subsequent date. We specifically disclaim any obligation to update or revise any forward-looking statements.

With that, I'll turn the call over to David.

Thanks, Renee, and good morning, everybody. And welcome to our third quarter earnings call. The third quarter of 2018 was an exciting time for Agios as the hard work and dedication of our team culminated in the achievement of important clinical and regulatory milestones and put us on track to complete the remaining milestones that we laid out at the beginning of the year. In addition, we began to lay the groundwork to support our next phase of growth in 2019 and beyond, focused on expanding indications for our wholly-owned precision medicine, TIBSOVO, also known as ivosidenib, rapidly advancing both our cancer and rare genetic disease clinical development programs and continuing to foster our rich research pipeline. Notably, the July approval and launch of TIBSOVO, the first and only treatment for IDH1 mutant relapsed or refractory acute myeloid leukemia, was a remarkable achievement for our organization. We're pleased with the progress of the U.S. launch, and Steve will share additional details later in the call.

With both of our IDH inhibitors now available to patients, we are rewriting the textbooks on treating IDH mutant AML. Just last week, out of the field of 20 nominees, IDHIFA won the U.S. Prix Galien Award for Best Pharmaceutical Product of 2018, which recognizes outstanding achievements in improving patients' lives through the development of innovative therapies. We're incredibly honored and humbled by the recognition of the practice-changing impact in medicine like IDHIFA can have for patients and are committed to continuing the work that makes an achievement like this possible.

For TIBSOVO, we know that our work has just begun to provide access to more patients with an IDH1 mutation. Today, we announced 2 key updates to our ambitious clinical development strategy in newly diagnosed AML. Most importantly, we plan to submit a supplemental new drug application for single-agent TIBSOVO in AML patients who are not eligible for standard therapies, which Chris will cover momentarily.

Turning to solid tumors. In September, Agios gained worldwide rights to our pan-IDH inhibitor, AG-881, from our partner, Celgene. We now have global rights for 2 IDH1 mutant inhibitors for the potential treatment of low-grade glioma where the IDH1 mutation is present in approximately 80% of patients and for whom there are currently no approved curative or targeted therapies. Beyond IDH, our team is conducting important patient identification and disease education work to help enroll our Mitapivat pivotal trials in pyruvate kinase deficiency, which we expect to complete enrollment in 2019. We are also exploring the role of pyruvate kinase activation in other anemias and remain on track to start a Phase II proof-of-concept study in thalassemia by the end of the year.

Our early oncology development pipeline continues to advance as we progress AG-270 through dose escalation. In addition, we have submitted the IND for AG-636, our DHODH inhibitor for the treatment of hematologic malignancies. This will be the company's seventh IND over the past 10 years. As we head into year-end, we're also focused on the upcoming ASH meeting, where we'll be presenting several clinical updates, including 2 oral presentations in newly diagnosed AML.

I'll now turn the call over to Chris to discuss our clinical development activities.

Thank you, David. Over the course of the third quarter, we advanced numerous clinical initiatives across our oncology and rare genetic disease programs that keep us on track to achieve our remaining 2018 milestones and expand the scope of our clinical programs over the next several years. The most significant update is that we've been able to accelerate our AML frontline strategy

(technical difficulty)

in 2 patient populations.

First, the FDA has confirmed that event-free survival is an acceptable primary endpoint for our Phase III AGILE study, evaluating ivosidenib in combination with azacitidine in frontline AML patients ineligible for intensive chemotherapy. The study design is being revised to reduce the number of patients required to appropriately power the study for the EFS endpoint. And as a result, we now expect to complete enrollment in 2020, a year earlier than our prior guidance.

Second, after discussions with the agency, we now plan to submit a supplemental NDA by the end of January 2019 for ivosidenib in newly diagnosed AML patients who are not eligible for standard therapies. The FDA has accepted our application into their real-time oncology review pilot program, which has enabled us to engage with the agency prior to the planned sNDA submission and may result in a more efficient review process.

As background, there are a number of factors that can impact a newly diagnosed AML patient's ability to tolerate standard therapies, including age, comorbid conditions, poor performance status and disease-related adverse prognostic factors. Many of these patients also have secondary AML, a condition that evolves out of treatment for myelodysplastic syndrome and is often resistant to chemotherapy. Based on AML patient claims analysis we have conducted, we believe approximately 25% to 30% of all newly diagnosed patients may not receive any standard therapies.

The sNDA for ivosidenib will be based on safety and efficacy data from a cohort in our single-agent Phase I trial, where we enrolled 34 newly diagnosed AML patients with IDH1 mutations who were not candidates for standard therapy due to comorbid conditions, ECOG performance status and/or adverse risk factors. Later this morning, our ASH abstract will be available, showing updated data from this cohort, including a CR+ CRh rate of 41%, a complete response rate of 27% and an overall response rate of 59%. I would like to note that these data were generated from a patient population that was older with a median age of 76.5 years. Approximately 80% of these patients have secondary AML and 41% had received a hypomethylating agent for a prior hematologic disorder. Additional data will be included as part of an oral presentation at ASH.

I'll now turn to our frontline AML combination strategy, which includes Phase I and Phase III trials of our IDH inhibitors with both intensive induction and consolidation chemotherapy or azacitidine. With the FDA agreement of event-free survival as an acceptable endpoint for the Phase III AGILE trial, we are now in the process of amending the study protocol. Utilizing event-free survival as a primary endpoint captures clinical benefit in AML and removes the effects of post-trial therapies that can potentially confound overall survival. Overall survival will remain a key secondary endpoint in the trial.

We continue to make progress on our trials in the frontline AML population eligible for intensive induction and consolidation chemotherapy, also referred to as 7+3. Our Phase I study combining either ivosidenib or enasidenib with 7+3 has completed enrollment and the data continues to be compelling and supportive of the Phase III strategy. Updated data from this study has been accepted for an oral presentation at ASH.

HOVON 150, a Phase III trial sponsored by European cooperative groups, combining both IDH inhibitors with 7+3, remains on track to initiate this quarter. The trial includes a 2-year maintenance phase, where patients will continue to receive either IDH inhibitor as a single agent even if they are in complete remission. In Europe, we are on track to submit a Marketing Authorization Application for ivosidenib in relapsed or refractory AML by the end of 2018, assuming that regulatory package based on the Phase I trial and our clinical development plan is acceptable to the European Medicines Agency.

Beyond AML, we continue to evaluate ivosidenib as a single agent in patients with high and intermediate risk myelodysplastic syndrome. Our Phase I data from a limited number of patients demonstrated durable remissions in this population. Follow-up data will be shown in the poster presentation at ASH, and we are actively exploring clinical development opportunities in this indication.

Turning now to the solid tumor setting. Our Phase III CLARITY trial evaluating ivosidenib in previously treated IDH1 mutant cholangiocarcinoma continues to enroll patients, and we are on track to complete enrollment in the first half of 2019. Ivosidenib has the potential to be the first targeted therapy approved in this terrible disease.

In glioma, we are continuing to evaluate 2 molecules: ivosidenib and AG-881. We are randomizing patients to each drug in our perioperative window study to better understand their relative uptake in brain tumor tissue and reduction of 2HG levels. Updated Phase I glioma data for AG-881 will be featured in an oral presentation at the Society for Neuro-Oncology Annual Meeting later this month. The focus of this presentation will be on longer follow-up from the data presented at ASCO in addition to new volumetric data. Similar data from the ivosidenib trial were presented at last year's SNO meeting, which demonstrated a reduction of tumor growth rates, a clinical measure that may provide a clearer picture of an IDH inhibitor's activity in these slower-growing tumors. Our clinical data in this setting, together with input from regulatory authorities, will inform an internal decision on our pivotal strategy by year-end.

I will wrap up oncology with an update on AG-270, our first-in-class MAT2A inhibitor currently in a Phase I dose escalation trial. This is a Phase I basket trial enrolling patients with several different tumor types with an MTAP deletion. We are currently in the dose escalation portion of the trial, and accrual is on target. The goal for this portion of the Phase I trial is to evaluate the pharmacokinetics and pharmacodynamic markers of MAT2A inhibition and establish the recommended dose for potential expansion cohorts. To date, with AG-270, we are seeing the expected effects on the pharmacodynamic markers of MAT2A inhibition in accordance with drug exposure, which we will continue to monitor as we proceed with dose escalation. Consistent with all of our clinical programs, we will submit an abstract to a major medical meeting once we have enough data to begin to tell the AG-270 clinical story.

In rare genetic diseases, we continue to make progress in our pivotal program for Mitapivat in pyruvate kinase deficiency, the first of its kind for this rare chronic anemia. We are opening new sites and enrolling patients in both the ACTIVATE and ACTIVATE-T trials while expanding on patient finding efforts across the globe as we seek to complete enrollment in these studies. Based on the operational work we have completed thus far, we anticipate completing enrollment in both trials in 2019. Our global PEAK Registry, aimed at building a more comprehensive picture of disease burden among children and adults with pyruvate kinase deficiency, is now active across 16 sites in 6 countries. The potential for activating a wild-type PKR enzyme provides a rationale to study Mitapivat in other anemias, and we have designed a Phase II proof-of-concept trial in thalassemia, which is on track to initiate by the end of the year.

2018 has been a year of execution across the clinical organization, and we look forward to carrying that momentum into the close of the year.

I'll now hand it over to Steve for a commercial update.

Thanks, Chris. I'm pleased to provide the first quarterly TIBSOVO launch update where we generated net U.S. sales of $4.5 million for the first partial quarter of launch. Since our launch in July, our team of senior hematology sales consultants has had approximately 2,000 customer interactions, raising awareness of TIBSOVO and educating physicians on how this novel, first-in-class precision medicine can be used to benefit their patients with relapsed or refractory AML with an IDH1 mutation.

Increasing testing rates of the IDH mutation is a key strategic imperative for the adoption of both of our IDH inhibitors, and we are encouraged with the steady growth driven by our field team throughout this year. According to market research from August, just after TIBSOVO launched, approximately 80% of physicians were testing their patients for the IDH1 and IDH2 mutations. We believe academic centers are screening the vast majority of their patients for actionable mutations such as IDH. The community is typically behind academic centers in this respect, and our focus is on improving IDH testing rates among these prescribers.

With respect to physician awareness, we are encouraged that awareness of TIBSOVO is already above 90%, which compares favorably to a 50% awareness of IDHIFA at a similar time in its launch. During the first 2 months of the TIBSOVO launch, we estimate over 100 unique prescribers have written a prescription for our medicine, and we are pleased with the uptake we are seeing. Consistent with our early IDHIFA experience, we expect that the initial patients treated with TIBSOVO will be more advanced in their disease, and as the launch progresses and the patient mix evolves, we expect treatment duration to increase.

On the reimbursement front, we are pleased with the payer response to TIBSOVO since launch, and we are not aware of any payer-related coverage issues. The payer mix for the first partial quarter on the market was approximately 60% Medicare, 35% commercial insurers and the remaining 5% split among Medicaid, cash and other payers. This payer split is largely consistent with what we have seen with IDHIFA and what we expected prior to launch.

The team is off to a great start with the launch of TIBSOVO, and I look forward to updating you on our progress in future quarters.

I'll now turn the call over to Andrew.

Thanks, Steve. Our third quarter financial results can be found in the press release we issued this morning, which I'll summarize. More detail will be included in our 10-Q filing later today.

Total revenue for the third quarter was $15 million compared to revenue of $11 million for the third quarter of 2017. The year-over-year increase in revenue was driven by $4.5 million of net U.S. sales of TIBSOVO, a $1.3 million increase in the IDHIFA royalty, which was offset by a decrease in collaboration revenue recognized during the quarter.

The gross to net discount for TIBSOVO in its initial quarter of sales was in line with our expectations of low to mid-teens as a percentage of sales. We expect the gross to net percentage to increase slightly within this range as TIBSOVO utilization increases in government channels subject to mandated discounts. We ended the third quarter with what we believe is a reasonable level of TIBSOVO inventory to expect going forward.

Cost of sales for the quarter were $700,000. Based on our policy to expense cost associated with the manufacture of our products prior to regulatory approval, certain TIBSOVO manufacturing costs were expensed prior to July 20 and, therefore, are not included in cost of sales during the current period. As a result, the majority of our cost of sales for the third quarter were fixed period expenses. And therefore, we expect cost of sales as a percentage of product revenues to decrease over time until we deplete these inventories.

Now turning to our operating expenses. We reported $82 million of research and development expense for the third quarter of 2018, compared to $73 million for the same period in 2017. This increase was largely driven by increased investments in the Mitapivat PK deficiency pivotal program, IND-enabling activities for AG-636, our DHODH inhibitor, and ongoing research efforts across our discovery platform programs. Third quarter selling, general and administrative expenses were $31 million, an increase of $13.5 million from 2017, driven by increased investment in our infrastructure and commercial capabilities with the launch of TIBSOVO. During the third quarter, the performance criteria for our performance stock units held by certain employees were met upon the approval of TIBSOVO in July. As a result, R&D and SG&A stock compensation expenses for the quarter were higher by $3.9 million and $3.7 million, respectively.

We ended the quarter in a strong financial position with cash, cash equivalents and marketable securities of $878 million. We expect that this cash balance, in addition to expected TIBSOVO revenue, will fund our current operating plan through at least the end of 2020.

With that, operator, please open the line for questions.

(Operator Instructions) Our first question comes from the line of Anupam Rama of JPMorgan.

Maybe 2 quick ones from me. I'm wondering if you could help us understand what to expect for the ivosidenib newly diagnosed AML not eligible for standard of care dataset at ASH. And how much of that data set has been shared with regulators such that you could proceed with this sort of accelerated frontline AML strategy? And then second question, I know you mentioned physician awareness on a time-adjusted basis for IDHIFA and TIBSOVO. Wondering if you could comment on genetic testing rates for IDH1 with TIBSOVO relative to IDH2 and IDHIFA on a similar kind of adjusted basis.

Thanks, Anupam. Chris, you take that first one; and Steve, the second.

So the data that we'll show at ASH will be on those 34 patients that were in cohort 2, Anupam, and we presented some of the efficacy data previously, but now you'll get a greater sense of the safety as well as an update on efficacy and some of the other important variables around describing the -- both the efficacy and safety of the drug as well as some of the molecular data that we published in the relapsed/refractory space in the past. So I can't provide more details at this point because we're putting the presentation together, but I think it'll give people a good picture of the activity of the drug and the safety in this patient population. With regards to the FDA component, so this -- what we submitted for the relapsed/refractory AML was the entirety of that patient experience in that one 001 study. And so that included not just relapsed/refractory patient population but also this group that I was more speaking to this morning as well as patients with other hematologic disorders. So the FDA has seen these data as we've been moving through the submission. I think the other component of this that I mentioned is the RTOR -- that is Real-Time Oncology Review, and this is part of the pilot. The FDA has provided some information around this on their website that you can take a look at. And there have been 2 approvals that they've spoken to or that have occurred through the pilot. And that gives you some sense of the range of time lines that one could expect. One was relatively rapid and the other one was, I would say, closer to a standard review process. So from our perspective, we are focused on pulling the submission together with the time line that I spoke to earlier, getting the data sets and working with FDA to make this happen as quickly as possible.

Yes. Anupam, it's Steve. So as to your second question, you may recall from our Investor Day that we held earlier this year, we disclosed and talked about 2 different data points for IDH2 testing rates. The first was from the time point of October of last year, where we said that the testing rate at that time was about 50%. By January, that climbed to approximately 70%. And as I said in the prepared remarks, we now are seeing testing rates of around 80%. So this is what we expected going into the launch of TIBSOVO that the IDHIFA experience, both with respect to testing rates as well as with respect to class awareness, was going to put us at a starting point that was going to be higher than we had with IDHIFA. And that certainly is what we're now seeing with TIBSOVO. So we're really pleased with the fact that we're seeing such good progress and rapid acceleration of testing rates in the U.S.

Our next question comes from the line of Kennen MacKay of RBC Capital Markets.

Maybe one for Chris on the MTAP program. Was just wondering if we should sort of anticipate -- you mentioned of seeing pharmacodynamic effects as expected but sort of lack of language on anticancer activity or efficacy. There is a signal that the anticancer effect is not being seen yet. Or is that just reading too much into it? And then also on that, wondering when you mentioned that this program is on track, if you could elaborate a little bit sort of as to what internally got attractive and -- in terms of dosing and then potentially where you might expect to start seeing anticancer activity as it relates to either dose or the pharmacodynamic markers that you saw in your preclinical in vitro data and your patient-derived xenograft data.

Kennen, this is David. Maybe I'll start there. And I don't think you should infer anything because as you know, our practice is that we're going to present any clinical data until we have a complete story and we bring it to a major medical meeting. So one should not infer in either direction by our comments today. We just wanted to give an update on the progress of the study, which obviously, as Chris said, we're very pleased with the progress, both enrollment and some of the early PK data, but it's just still early days. So internally, as you know, we're going to continue dose escalation in the basket part of the trial, and then when there is enough data to tell a story, as Chris mentioned, we'll bring it to a major medical meeting, so important not to infer anything in either direction.

Fair enough. Maybe I can squeeze one additional question in then. Just wondering from either IDHIFA or from TIBSOVO sort of what kind of durations we're currently seeing in the commercial market in the relapsed/refractory patients and then how we should maybe be thinking about duration in the frontline patients who are ineligible for other therapies. And also, on that definition, just wanted to clarify that -- is that ineligible for all therapy sort of including Venclexta given that wasn't yet out when these patients were being enrolled? Or how does that definition sort of work?

We'll have Steve start; and then we'll come to Chris.

Yes. So thanks. Ken, it's Steve. So just with respect to what we're seeing currently in the market on IDHIFA duration, I don't have any real update for you on IDHIFA duration. It's very consistent with what we've previously reported. As we said, it's going to take some time for us to get to a duration that's similar to what was reported in the clinical trial setting. And so we continue to see some movement there, but I don't have a substance of update for you with respect to duration on IDHIFA. For TIBSOVO, it's just too early for us in the launch to provide any additional commentary with respect to duration. I think you should expect, for modeling purposes, to see a similar kind of set of facts play out for TIBSOVO, as we've seen with IDHIFA, where, as I said in my prepared remarks, the initial patients that go on to therapy are going to be patients that may have worse performance status, they have progressed further on the course of their disease and, therefore, have shorter duration. And over time, we fully expect that, that duration will continue to increase.

Kennen, it's Chris here. So with regards to the eligibility for standard therapy, that was largely a clinical -- a clinician call for those who were involved in the trial. The study sites for its centers of excellence. So they would have included in their thought process eligibility for approved standards or other standards of care, low-dose Ara-C, azacitidine, et cetera. And they would have also considered protocol treatments that they could have given patients. So it was a clinical call, and we didn't specifically say venetoclax, yes, no, et cetera. The study wasn't designed that way. We were really thinking of getting an understanding of the efficacy and safety of single-agent ivosidenib in a group of patients that, in general, wouldn't get any anti-leukemia therapy, if you will. Because we want to understand -- we thought the -- and from working with the investigators in our experience in relapsed/refractory given the overall side effect profile and the efficacy of the drug that this could be an option for people that previously wouldn't -- have been told to go home or have no therapy.

Our next question comes from the line of John Newman of Canaccord.

And certainly, I'll add congrats on all the clinical progress going on and improving quickly. I have another question on duration regarding to TIBSOVO versus IDHIFA. My question is a little more general and that is given that physicians are aware of the mechanism of action now for IDHIFA, do you think that the increase in duration of therapy for TIBSOVO could be a little bit quicker? Or should we expect probably the same type of rate there?

John, it's Steve. It's a good question. I mean, I think what we should expect to see is similar to IDHIFA. I think this has less to do with familiarity with the class and experience with the class and more to do with the distinct patient populations and the fact that patients with the IDH1 mutation would have been waiting for TIBSOVO to get approved. And therefore, as I mentioned, we would expect patients that are further along in the course of their disease who have been the initial patients to be treated. I do think we'll benefit downstream from other measures related to awareness of the brand related to testing, but with respect to duration, I would expect a very similar trajectory.

And then another question is just from a mechanistic standpoint. You happen to have patients on either IDHIFA or TIBSOVO that were treated either now or in the future with combination in the frontline, be it 7+3 or azacitidine. Mechanistically speaking, would there be a potential opportunity for those patients to receive monotherapy after that point?

Monotherapy, after they get combination, yes. So I think that's something that will be tested. So, for instance, in the HOVON study, patients get randomized, such a randomized trial, John, 7 and 3 plus or minus your IDH inhibitor depending on whether you're 1 or 2 positive. And then basically, it starts with the initiation of induction and then assuming and hoping that you make it through your induction, your consolidation and there's this -- a maintenance phase of up to 2 years. And that's for patients who achieved CR if they go to a transplant and so on and so forth. So that will be very important data for us to understand the outcomes, and we're looking forward to getting a wealth of molecular data relative to see how these things track. And if you think about the AGILE study, that's a combination study with azacitidine. But like any combination oncology trial, there are components build-in where investigators may have to make adjustments for toxicity. And you could see a situation where after a period of time, if patients are in a situation where the clinician feels like the tolerance for azacitidine is such they're able to reduce the drug or they may stop and then patients can stay on either TIBSOVO, ivosidenib or placebo. So yes, we see that happening. And then I think there is the other aspect as how the drug might be used in the community, and I wouldn't be surprised at all if we saw clinical strategies along the lines of what you just suggested. And that's something that Steve and his team and our team in the field as well as all of us here are doing that interact with investigators. We'll get more color on it as time goes on.

Our next question comes from the line of Peter Lawson of SunTrust.

Just a question for Andrew or perhaps Steve just on thoughts around TIBSOVO revenues, how they should be tracking versus IDHIFA around number of salespeople you think you need after the initial launch and how we should be thinking about the SG&A for the rest of the year.

Yes. Peter, it's Steve. So maybe I'll take the first 2 parts of your question and then turn it over to Andrew on the SG&A question. So with respect to field sales organization, as we previously disclosed, we have in the mid-30s in terms of field sales personnel out there promoting the brand. We're well sized to address nearly all of the potential prescribers for the medicines. So physicians who treat AML, clearly, are our targets for that. So we don't have any immediate plans to scale up that organization because we believe we already go deep enough into the prescriber base to cover all of the relevant targets. And Peter, sorry, the first part of your question had to do with trajectory of TIBSOVO revenues versus IDHIFA. Did I get that right?

Yes, yes, exactly.

Yes, yes. So I think you can expect a similar trajectory for TIBSOVO relative to IDHIFA. I'd -- I say that but should also note that this is a treatment landscape that is changing pretty rapidly. This has been fantastic news for patients and for the physicians who treat these patients, who now have multiple new therapies available to them to treat patients. So anything from drugs that target FLT3, certainly the IDH inhibitors, other new medicines that are available. So there is an element of -- there's a dynamic, I guess, in the treatment landscape that is in flux. And so it's going to take some time, I think, for us to see how things settle out. I don't want to put a time course on that because I think it's going to be an evolution over the next few years to see how physicians decide to sequence medicines, to combine these different medicines and to learn how best to use them based on new clinical trial data that's going to be emerging. So it's tough to predict exactly how this landscape shakes up, but I think for the timing being, I think it's relevant -- or reasonable rather to use IDHIFA as a good analogue for TIBSOVO.

Yes, Peter, and then just on the SG&A, I think you can expect that, given what Steve said, to continue on the run rate that we've sort of guided to from Q1 and Q2. I think just important to note the comment I made in my prepared remarks that Q3 was impacted by the increase in stock comp expense due to the performance stock units triggering the performance target of the approval of TIBSOVO. So that's a onetime event that I would not project going forward.

Our next question comes from the line of Mohit Bansal of Citi.

Just one specific question on the duration. So we'll see the ASH data later today but just trying to understand in newly diagnosed patients who are ineligible for therapy, how should we think about the duration of the treatment in that patient segment?

Chris?

Well, I think that we'll provide some data in terms of the -- some information around the duration of responses and that will be helpful in terms of thinking about some of those aspects. Overall, I think the approach in this indication is similar to how you approach a patient with relapsed/refractory disease. That is you start the drug. And then if you think about our label in the relapsed/refractory AML, the guidance is in the absence of unacceptable toxicity or over progression that continue for up to 6 months. Because when -- at least in the relapsed/refractory space, you can see there is a range of time to achieve the best response. There's associations of benefits from a transfusion perspective even in the setting of patients who achieve stable disease. So we're working through all those same features as we think about newly diagnosed patients much like we do with relapsed/refractory AML. And those are the factors that will come in, in terms of determining our labeling guidance that FDA works out with us as well as how clinicians actually use the drug.

Got it, very helpful. And then if I can squeeze one more on your Phase III trial, HOVON. So you're doing an interesting thing, which is maintenance treatment after the induction. Just for the comp purposes, a patient who get transplant versus not get transplant, how should we think about the control arm here? And wouldn't like getting transplant versus not getting transplant confound the result in terms of EFS as the primary endpoint? How should we think about that?

In terms of the -- an event occurring relapsed whether that's in the setting of transplant or not, it's certainly something that we'll have to look at and see what aspects -- how we want to perform additional secondary analysis within that data set. I think you're right when you talk about the potential for transplant to confound and present some challenges in terms of doing an analysis. I think the key feature is that transplant is a treatment option that's very important in this disease. So we think it's important that patients are going to go to transplant, and the other important aspect is they are able to stay on maintenance drug that's either drug or placebo because we think that assuming that things are balanced overall that the contribution of the IDH inhibitor component will win and demonstrate clinical benefit using an event-free survival endpoint. I think where transplant can really raise some problems is in a survival study. And so that's one of the good things about having an event-free survival endpoint in the sense that we haven't completely eliminated that compounding variable. We've certainly reduced it by using that endpoint, which has been demonstrated -- which has been verified as an endpoint for clinical benefit and as an approvable endpoint in both the U.S. and Europe in this patient population. That is induction and consolidation eligible patients.

Our next question comes from the line of Terence Flynn of Goldman Sachs.

Maybe just a follow-up on the AG-270 program. I was wondering if you can comment if the PK effect that you noted were seen across types of tumor? And then anything on safety and tolerability at this point that you can provide? And then Andrew, as we head into 2019, any commentary you can provide with respect to total spend and thinking about it versus 2018?

Terence, this is David again. Yes, as I mentioned before, we're not going to present any additional information from the 270 trial. We're really pleased to put the way that it's going, but we're going to save any data, like we always do, for a medical meeting. Andrew?

Yes. So in terms of the spend, I think what the guidance we're giving is that our -- the cash balance that we end in Q3 with is going to provide runway through the end of -- at least the end of 2020. I think you could expect probably R&D to increase over time. I don't think it's going to be a step change, but as we continue to enroll patients in our many global Phase III trials, that should drive that up.

Our next question comes from the line of Chris Shibutani of Cowen.

Congratulations on the strong start for TIBSOVO. One very quick housekeeping question and then a bigger picture question. On the very quick housekeeping side, initial launch inventory, any aspect of that net sales number that you just reported for the third calendar quarter should we think about as being inventory at all? I know you're using specialty distributors. Can you just respond to that quickly?

Yes, I mean, we're not certain to disclose exact percentages. There obviously was inventory because there's inventory in the channel, but I would just characterize that the majority of revenue was demand.

And then on the gross to net side, as far as maybe, in particular, as a proxy IDHIFA?

Yes, that's reasonable. I mean, there is certainly differences in the distribution channel between what Celgene does and ours. We have a fairly enclosed distribution channel compared to theirs, but it's not an unreasonable comparison.

Great. And then bigger picture, you did mention that the FDA has a pilot program, which enables maybe more facility and access to enable your discussions with the FDA. Can you be a little bit more concrete in terms of what you think that could do in terms of time lines and your ability to make smart decisions?

No, I can't. What I can tell you is that there are 2 examples of drugs that have been approved using the Real-Time Oncology Review program. And one of them was relatively rapidly after the submission went in and one of them was comparable to, I would say, a normal approval time line. The pilot really allows us to exchange some data sets earlier to start interacting with them as they're thinking about the submission. And so that's the most guidance we can give. We're pleased that we were asked to participate in it, and I think it will be a good experience for us. And we hope that -- to have a positive outcome, of course. Further guidance around time lines, we just don't have that.

Got it. And then the last would be a big picture question for David. As an organization, you made the decision to make the announcement of transitioning that's going to happen as far as at the CEO level and in your own role. Can you just share with us a little bit what the messaging has been internally to the company? Any transition represents some degree of uncertainty. Obviously, you've made a choice with a familiar entity with Jackie being on your board. But can you just share with us what the messaging has been since the transition won't officially occur until quite a few months from now in March?

Yes. Thanks, Chris. I mean, I think it's been essentially the same thing that we've said when we announced externally, and Jackie is very well-known to the organization through her involvement in Celgene and then being on our board. And we have the opportunity for Jackie and I to spend a fair amount of time in front of the whole company right after the announcement. And so the message internally is exactly the same one that we've used externally that it's an exciting time for the organization, it's the right time for me personally, and Jackie is very excited to come onboard to continue the great work that's here, so nothing different. And so far things are going really well in the transition, which will happen on February 1.

Our next question comes from the line of Michael Schmidt of Guggenheim.

I just had a couple more on future development plans and maybe first on TIBSOVO and MDS. What are some of the factors that could influence your plan here to more formally pursue this program from a regulatory point of view?

Yes, thanks for the question. It's Chris Bowden here. Well, there's the frequency of IDH mutations in the disease. There's the spectrum of myelodysplastic syndrome. And so we've got -- and then thinking about our data in that context. So we've got to think about the risk populations that we'll be going after, intermediate versus high risk. We also -- one also has to consider not just what's approved in a disease but what the overall standard of care is and what the appropriate endpoints are. Those are all things that are in the mix now, and we're considering a number of different options. Our team is looking through that both in terms of thinking about our data and discussing this with the great group of investigators we've been working with. And so that's something we look forward to thinking through further in terms of whether and how would we do a registration program.

Okay. And then great to see the enrollment update on the -- guidance update on the AGILE trial. I was just wondering when you might be in a position to guide on enrollment time line for the HOVON study as well.

Well, we're excited that, that study is moving forward, and we're looking forward to initiating it. We've had a great collaboration with the -- with our -- the cooperative groups in Europe, the AML Study Group in HOVON. And right now, we've got to get the study launched and get things up and running. And so I can't provide any guidance at this point, but certainly in the future, once we get a better sense of how things are going, we'll be able to communicate that to you.

Understood. And then last one on -- regarding the DHODH inhibitor, you said you're filing an IND by year-end. I know there is a couple other DHODH inhibitors in development. Just wondering -- it may be too early, but in terms of differentiation, are there any features of the molecule that unique compared to others?

Maybe, Michael, let me just start by just -- we did announce this morning that we actually submitted the IND. And maybe Chris can tackle the second question or Scott.

Well, I think -- it's Chris here. The -- it's too early. There is a number of different molecules in the clinic now. And we look forward to testing our molecule and understanding it, both the safety and efficacy in this early -- in the early development setting. And I think that will be a very interesting story to play out now over the ensuing months to years as -- both for this class of drugs, but I would also reference back to Steve Hoerter's comment that this is an exciting time for those of us in drug development who are developing new drugs in AML because it's been a very, very, very long period where there has really been a lot of failure and not a lot of innovation. And it looks like that's changing now, and I hope that DHODH inhibitors will also contribute to that.

Scott, anything to add there?

Nothing really except for we have a lot of confidence in our molecules, and I think we have some novel insights that haven't developed, and so we think we have a great path forward. And I think although there are other molecules in the space, we're really pleased with our position.

Our next question comes from the line of Tyler Van Buren of Piper Jaffray.

I also am pleased to see positive updates with respect to the frontline programs. Specifically with respect to AGILE, I was hoping you guys could help clarify some of the time lines. Assuming the trial completes enrollment in 2020, could you give us some sort of indication as to when we could see top line data on an EFS endpoint and then when that would be followed up with overall survival? I see on ClinicalTrials.gov, even though it hasn't been updated, it says, 8 months on average for EFS, 12 months for overall survival and also 392 patients being enrolled. Maybe also some sense as to the reduction in patients on the trial given this update.

Yes. So it's Chris here, Tyler, and -- so we will update that information on ClinicalTrials.gov and we use some of the other mechanisms that we have used in the past trials and progress, et cetera, at ASCO and other major meetings when we have the amendment finalized and completed, and it's going out to sites. So we are in the process of doing that now. I think that -- and then the follow-up is what type of guidance can we give in terms of when the study would start to read out for those endpoints that you talked about. I think the key feature is it's an event-driven analysis. So once you complete accrual, then you have to follow patients and wait for events to come in. And some of those numbers that we've provided are based on some estimates based on historical controls. And so those are -- that's the information that's required to put into ClinicalTrials.gov. It doesn't really give us -- give you a lot of information in terms of when these -- when trials will actually start to read out. So it's just a little early for us to be able to provide guidance around that. The key feature is it's an event-driven analysis, and depending on how much activity you have in your control, in your experimental, I mean, how accurate you are in those predictions has the biggest effect on when the trials can read out.

Great. That's helpful. And just the second question would be with respect to the upcoming update with 881 at SNO. Obviously, we saw an update at ASCO. Just curious to get any thoughts there on what we should expect as well as what data we could see on additional solid tumors.

Well, for SNO with 881, we'll be providing an update both in terms of efficacy and safety. And as you know, we've spent a lot of time working with our investigators in terms of understanding the dynamics around changes and volume over time. We -- and we think that could be complementary to some of the standard ways of imaging patients who have low-grade glioma in this challenging space. As far as how other solid tumors go, the other area that I commented on in my remarks is with CLARITY. And so we're looking forward to the completion of accrual in the first half of next year.

And I'm showing no further questions at this time. I'd like to hand the call back over to David Schenkein for any closing remarks.

Thank you. Our 2018 progress validates Agios as an organization built to bring medicines all the way from the lab to patients. And we're confident in our ability to do this again and again. As you know and as I mentioned earlier, on February 1, 2019, I will transition to the role of Executive Chairman of the Board of Directors and Jackie Fouse, a member of our board, will become Agios' next Chief Executive Officer. I'm confident that as CEO, Jackie will maintain the innovative core of the company while also helping Agios reach new possibilities. We are poised to expand its impact on the lives of patients with cancer and rare genetic diseases, and I look forward to celebrating many of its future successes. I'd like to thank all of the tremendous employees at Agios for their dedication and passion to make a difference for patients. And, of course, I want to thank all of the patients, the caregivers and the physicians who participated in our clinical trials. Thanks for joining us today, and enjoy the rest of the day.

Ladies and gentlemen, that does conclude today's program. You may all disconnect. Everyone, have a great day.