Ladies and gentlemen, good morning or good afternoon. Welcome to the Roche Conference Call for Investors and Analysts from ESMO 2012. I'm Goran, Chorus Call operator. [Operator Instructions] The conference must not be recorded for publication or broadcast. At this time, it's my pleasure to hand over to Mr. Karl Mahler, Head of Investor Relations. Please go ahead, sir.
Yes. Thanks a lot. Good afternoon, and good morning, and welcome to the Roche Analyst Conference Call or ESMO Congress in Vienna. I trust that you had the opportunity by now to download the presentation from our website. To those of you who joined us at our Investor Day in London in September, I want to say welcome back. We had 200 people -- 240 people came together to hear senior management speaking about our strategy and highlight our rich R&D pipeline. Thanks for your interest again and the good feedback you gave -- we got from you and you gave us.
Today's conference calls is the third investor science event at a medical meeting this year that we have been organizing for the analyst and investor community in order to present key results on our late-stage pipeline.
At this year's ESMO, Roche and Genentech did present data from 144 scientific abstracts. 24 of them were given as oral presentations. This shows our strength and the continuing leading role that Roche plays in oncology. Oncology is and will remain the major focus for Roche not only because we are the market leader and have extensive know-how in oncology research and development but also because oncology is expected to remain one of the main areas of high medical need. With about 50% of our investment going to oncology, we also invest most in this therapeutic area.
And as you can see on Slide #5, oncology represents about 60% of new molecules in our R&D pipeline. Among them are early-stage assets such as anti-CD22 antibody–drug conjugates or the Bcl-2 inhibitor, which are being investigated in hematological cancers and our potential success of MabThera rituxim. Late-stage assets are, for example, T-DM1 and pertuzumab, which will report Phase III results from the MARIANNE trial, studying the combination of T-DM1 and pertuzumab in HER2-positive type breast cancer, somewhere in 2013.
Looking on the months ahead, the overall survival data from the CLEOPATRA study will be presented at the San Antonio Breast Cancer Meetings beginning of December. And finally, data on MabThera subcut from the Phase III SABRINA study will be an edge [ph].
Outside oncology, we will present the results of the aleglitazar Phase II renal safety study, ALENEPHRO, at the annual meeting of American Society of Nephrology in November.
Let me now turn to Herceptin. Since it was on the market in 1998, Herceptin has become the foundation of care in HER2-positive breast cancer and as such, has prevented breast cancer recurrence. As you can see on Slide 6, after its approval in early breast cancer, which was based on 4 large Phase III studies, the decline in recurrences became even more pronounced. It is estimated that Herceptin has contributed to preventing breast cancer from coming back in 28,000 women in 5 major EU countries alone.
Roche and the Breast Cancer Initiative both presented at this year's ESMO the final analysis of the HERA study. Data that was shared with the public yesterday are comparing the disease-free survival of women getting Herceptin treatment for 2 years compared to those treated with Herceptin for 1 year. The results confirmed that 1 year of Herceptin remains the standard of care in early breast cancer. Importantly, after a median follow-up of 8 years, the improvements in disease-free survival and overall survival for women who received Herceptin for 1 year remained statistically significant compared to women who underwent observation alone despite heavy crossover of patients.
Results from the PHARE study, which was run by the French Cancer National Institute, investigating 6 months versus 1 year of Herceptin treatment were also presented, and the trial failed to show its objective that 6 months treatment is good as 12 months, so it failed to show non-inferiority.
Last but not least, final overall survival data from the Phase III EMILIA study were shared with the public, with signing overall survival benefit over at standard of care.
Allow me now to thank Sabine Borngraeber reporting from our team for putting all the slides together and for organizing the call. And with this one, I would like to hand over to Stefan Frings, the Global Head of Medical Affairs for Oncology at Roche, to walk us through the presentation. Stefan, please?
Thank you, Karl. A warm welcome from my end as well. I will now spend the next 20 minutes or so to cover the items Karl mentioned already in a more granular level and will mention a few more points.
I'm now on Slide 7, where you have seen what we are going to cover.
Moving to Slide #9, it's the HER2 franchise, and this is really what Roche is about. Roche is about innovation. And the HER2 franchise, we can well position that we're able to innovate with a three-pronged approach. We're innovating via Herceptin Subcutaneous, T-DM1 and Perjeta. The study is structured by line of therapy. In the second-line therapy, we're going to introduce first T-DM1. In the first-line setting, we have already Perjeta approved in the U.S. and Switzerland, and we will then reporting next year already data of the MARIANNE trial and the adjuvant study. In the adjuvant setting, we are having Herceptin Subcutaneous, which was already filed in the European Union. This is not an opportunity for the United States. And trial with Perjeta is already ongoing as we speak in the adjuvant setting.
Moving to Slide #10. This is about EMILIA. The trial results were updated after Kim Blackwell presented them at the ASCO meeting.
On Slide #11, you see the study design. Nothing has changed in this respect. I would like to just to preempt the question of the competitor arm. We have chosen the regulatory standard in the pretreated Herceptin setting, which is capecitabine plus lapatinib. Since this ESMO meeting, data of the CEREBRAL trial were presented Monday afternoon, which compared Xeloda/lapatinib versus Xeloda/Herceptin, which was post-marketing commitment for GSK, where Xeloda/Herceptin was superior in PFS and OS compared to Xeloda/lapatinib. But an important subgroup was shared at the conference, and it's not part of the abstract in the Herceptin pretreated setting so that lapatinib and Xeloda/Herceptin are rather similar in terms of PFS and OS, which gives us confidence that we have chosen not just the regulatory standard of care but also the clinical standard of care option.
Moving to Slide #12, those are the PFS curves as presented at ASCO. And since those PFS data hit the statistical significance level, those were then final at that time and will not be updated.
The next slide, #13, are the overall survival, where we, as presented at ASCO, really missed the significance level by a sliver. Therefore, they remain interim analysis. And in collaboration with the health authorities, we agreed to perform a second interim analysis, predefined, once again, p-value level and with more follow-up than reported at this year's ESMO. Yesterday, the data displays on Slide #14. You see a further nice evolution of the overall survival Kaplan-Meier plot and strong separation of the curve. Strong hazard ratio was 0.682 and separation of the median of 5 to 6 months.
Strong data, and the data triggers now for us the confirmatory significance level. Therefore, this analysis becomes now the final analysis from a statistical point of view, but we will still conduct further follow-ups and run a final analysis in 2014. We have filed in Europe and United States. You just see as presented at ASCO. And certainly, we are supplementing the dossier with those latest filing in order to get this reflected in the package insert.
The conclusion of this slide, as presented by Professor Sunil Verma, with nothing much to add, very strong PFS, strong overall survival benefit now having reached statistical significance. But importantly, T-DM1, our antibody-drug conjugate, it's not just about improving efficacy; the drug has properties which allow as well to improve the tolerability profile. We have significantly less severe adverse events. We have an improved safety profile. And I would like to add, it's typical even that their patient felt side effects more on the Xeloda/lapatinib arm: diarrhea, stomatitis, hand-foot syndrome. This is all associated with Xeloda/lapatinib. Whereas on T-DM1, you have side effects not felt by patients, like trends in liver function test elevations, thrombocytopenia, because we haven't seen any difference on bleeding. But thrombocytopenia on its own really is not felt by the patient and is really not cumbersome. Therefore, as said here on the slide, it is an important new therapeutic option. We're looking forward to work with the FDA to obtain approval either this year or latest, by early next year and the European Union, we expect late in 2013. The discussant, Xavier Hart-Cortez [ph], already mentioned that this should become the new standard of care in Herceptin pretreated patients.
Moving on to the adjuvant setting. On Slide #17, we have displayed the current situation. One year of Herceptin is the regulatory and clinical standard of care, and it's endorsed by multiple guidelines around the globe. One year duration was based on 4 large Phase III trials. There's only one trial which is exploring a longer treatment duration, which was a HERA trial using Herceptin in the sequential session. There are numerous shorter-duration trials currently ongoing, and one was just presented here at the meeting, the PHARE trial. There are some 6 versus 1 year, and there are 2 further evaluating 9 weeks even versus a year.
Moving to Slide #18, covering the HERA trial, which has 2 important aspects. First, the comparison of 2 versus 1 year, and very important for us, a long-term follow-up with 8-year median. The HERA trial design is displayed on Slide #19. I would just like to highlight that after the trial reported positive for DFS for one year duration of Herceptin, patients in the observation arm were allowed to cross over. And HERA is actually the trial with the largest percentage of patient crossover compared to any other Phase III trial done in this perspective. 52% received Herceptin in the observation arm, whereas folks under the Bcl-2 trial, it's a single percentage number.
The key takeaways for the 1- versus 2-year comparison is displayed on Slide #20. You see basically superimposed bell curves, and it has a ratio of 0.99. Therefore, prolonged administration of Herceptin beyond 1 year doesn't improve PFS.
This is summarized on Slide 21. A few more points were made by the presenter, Dr. Gelber, and I will not cover this today.
Slide #22 is displaying the long-term follow-up. In the upper half of this slide, on disease-free survival, initially, we saw a hazard ratio of 0.54. We see a small number of events, and then the event number certainly increased with more follow-ups, and the hazard ratio stayed statistically significant with 0.76. Please keep in mind, 52% crossover to Herceptin. And you see on the left-hand side, 48.5% of the follow-up period of the observation arm was under Herceptin. So therefore, the treatment effect of 0.76, for sure, is underestimating the true effect if there wouldn't have been any crossover.
Looking at overall survival, there was initially an effect with 0.76 at the [indiscernible], where DFS was statistically superior. And now with more events, we have retained the statistical significance with 0.76 in presence of strong crossover. This is a terrific effect, and it ensures and reinforces, really, that Herceptin is the gold standard of care, and women with HER2-positive breast cancer should receive this compound.
This is summarized as well on Slide #23. In the interest of time, I'm going to move to the PHARE trial, starting with Slide 24, where 6 versus 12 months are being investigated. The trial design was straightforward. It is committed -- or sequential Herceptin administration was allowed, new advent cases as well, stratification for concomitant versus sequential and hormonal status. Patients were on Herceptin when the decision was made, and then after a period of 6 months, it was either stopped or they were randomized to receive up to 12 months. Certainly, there were patients who may have been taken a different decision, but this was the design, then follow-up throughout.
On 26, the statistical methods. It is a non-inferiority trial, and this is certainly almost far more difficult to interpret. Most people are used to superiority trials. And non-inferiority was defined as the trial -- the 6-month is as good as 12-month if it is not worse than 15% than the 12-month duration. And it converts into a hazard ratio of 1.15, or in absolute terms, if you look at the Kaplan-Meier plot, you would not want the 6-month more than 2% below the 12-month arm. The trial was amended multiple times. Initially, the group set out to recruit 7,000 patients, and therefore, they amended it to ultimately deliver results that after 4 years of accrual and at least 2 years of follow-up, the data should become available.
If we take a look at Slide #27, it covers basically every fifth French patient with HER2-positive disease. And I would like to point out that the Independent Drug Monitoring Committee in May 2010 concluded that they would not want to have further enrollment in the trial because of the data they had seen at that time but at the same time, would want to see that the endpoint report, as designed, 6 versus 12. Therefore, they recommended in May 2010 no crossover of 6 months to 12 months. The data you would see in a second are based on a data locked at 2 years later, end of July 2012.
The trial endpoint data are displayed on Slide #28. You see the DFS Kaplan-Meier plot in orange, the 12-month and dashed black, the data of the 6 months. If you look across, at almost all endpoint and landmark analysis, you see a delta which exceeds the 2%, exclude difference as defined. The effect becomes even stronger when you take a look at the hazard ratio. The predefined margin was 1.15. Not only the point estimate is exceeding this threshold was 1.28, but please be aware that we'll explain this in a second. In order to declare non-inferiority, the upper confidence interval, here, 1.56, should have been below 1.15. So the trial clearly failed to achieve its objective to identify that 6 months is non-inferior to 12.
Going to Slide #29, because I believe subgroups may be a topic you will be interested in. You can see that every single subgroup analyzed has the upper 95 confidence interval far beyond the non-inferiority margin of 1.15. All but 2 subgroups, even the point estimate exceeds the non-inferiority margin of 1.15. Last but not least, all subgroups are at 1 or beyond 1, always on the side favoring the 12-month duration. Of note on this plot, the parity line at 1 is not marked.
Moving to an overall survival display on Slide #30, it clearly supports the notion that the 12-month duration is the ultimate standard of care. Hence, there were less events in the 12-month duration, and the hazard ratio supports this as well.
I would like now to go to 3 slides, the discussant, Sandra Swain, who's the ASCO President currently, presented at the ESMO meeting, especially in order to specify some confusion which may exist on certain terms. Just on non-inferiority just briefly, there's this null hypothesis that the treatment differ by more than an acceptable level of delta, and the delta was predefined not being worse by more than 15%. The alternative hypothesis is that treatments do not differ more than the delta, which then would allow us to declare non-inferiority.
So the last bullet is the most important one. As I explained earlier, non-inferiority can be achieved and can be pronounced if the upper confidence interval of the DFS would be below 1.15. And as you clearly saw on the previous slide, the trial failed to meet its primary endpoint, therefore, non-inferiority cannot be declared.
Coming to the next slide, I would like to explain why Sandra Swain and Dr. Pivot said that the trial is inconclusive because we all can clearly reach a conclusion that the trial failed to declare non-inferiority. In a non-inferiority setting, in case we have not been able to declare non-inferiority, you are then testing whether the 2 arms are truly different. Since you have accepted that the outcome can be as much as 50% worse, you're now testing whether the experimental arm is inferior to the new threshold, to the new goal post of being 15% worse. And in this case, the confidence interval crossed the line of 1.15. Therefore, since it couldn't be clear non-inferiority, here, the trial failed, means that you can declare non-inferiority by a margin of 15%. Based on that term, it is called to be inconclusive. As I said before, the conclusion of the trial is clear. The trial failed to establish non-inferiority. There, no one disagrees. Sandra Swain pointed out as well, more as an observation, that the lower confidence level of the disease-free survival did not include one. Therefore, one can conclude that for sure, 12 months is better than 6 months.
Looking at the point estimate of 1.05 of the lower bound, you can at least say it's at least 5% better and even as good as maybe 56% better, 12 months versus 6, but certainly, those are then in the superiority setting and it was a non-inferiority trial.
And certainly, the trial had hoped to initially get 1,040 events. This will not be achieved, and more events would potentially be able to tighten the confidence interval and potentially then conclude that 6 months is statistically inferior to 12.
Coming to the last slide, which, I believe, is from the Roche perspective, the key take-home message for today. Two years is not better than 1. One year is better than 6. Clearly, one year of adjuvant trastuzumab Herceptin remains standard of care.
And moving on to a brief summary on HER2. Partly, it was already covered by Karl. Karl and HERA reconfirmed the confidence that one year of Herceptin is the standard of care. Subcutaneous, we have filed. We are currently under regulatory review. We are addressing the questions we received from the EU authorities. We have good arguments to defend our case. I think what was presented here at ESMO was a subcutaneous self-administration device, and the device showed comparable PK parameters compared to syringe, which may allow us in the future to further simplify the administration mode and allow a home-based delivery of Herceptin Subcu.
Perjeta in the adjuvant setting is ongoing with the APHINITY trial, and our program with T-DM1 is shaping up, and we hope to be able to pronounce the start in the near-term future.
In the first line setting, pertuzumab in the United States has received approval, just recently, Mexico as well, and the dosing is currently being evaluated by the EU authorities. But there, we would expect a decision in 2013. You will see final OS data with the statistical survival benefit at the San Antonio Breast Cancer Meeting in December.
And I think we have spoken already enough about T-DM1 that I will move to Slide 37, now covering the skin franchise. And just as a reminder, we have Zelboraf, which has been great success, a breakthrough therapy and civilizing our approach with personal healthcare in the metastatic melanoma setting. It direct just [ph] exclusively in BRAF-mutated patients. Just to remind everyone of the size of the opportunity, it's a very high medical need. But although the patient number are limited, globally, we would expect around less than 25,000 patients with metastatic melanoma.
What do we do with Zelboraf which has, in the year since it has been on the U.S. market, become the unequivocal standard of care in the setting? Our launch in Europe is moving well along. We will address with further data to be shown in this call the combination with the MEK inhibitor. Other combinations as a development trial are the combination with ipilimumab with Bristol-Myers Squibb, and our own anti-PD-L1 of our Genentech pipeline is currently proceeding as well for the Phase I. The adjuvant melanoma trial has already had earned first patient in with Zelboraf single agent. In that setting, we're exploring other tumor types on the development program in papillary thyroid cancer and in the medical setting, signal seeking in other BRAF-mutated tumors, including hairy cell leukemia.
I would now like to take you briefly through the Phase I of Zelboraf with our internal MEK inhibitor on Slide 40 or the info slide, and on Slide 41, the scientific rationale. If the tumor, unfortunately, develops resistance to BRAF inhibition, one can conceptualize MEK-independent resistance and MEK-dependent resistance and the combination with a MEK inhibitor is intuitive to block it further downstream, signaling after a blockade via a BRAF inhibitor.
We have a design on Slide 42 to explore 3 different dosing regimens, either 2 weeks on, 2 weeks off; 3 weeks on, 1 week off; or continuously. In green, on the Slide 42, you see those cohorts which were tolerable. And if you look at Slide 43, the trial -- the combination overall was well tolerated. In our perspective, we had seen the main drug-related adverse events being rash, diarrhea, photosensitivity, fatigue or nausea and severe events rather less prominent, in particular, low incidence of dose reductions and dose interactions. And I would like to point out that we are rather pleased by the safety profile observed; I do know that cross-product comparison are always difficult to conduct. But the combination of the GSK combo, BRAF and MEK inhibitor is certainly different. We don't observe those level of [indiscernible], which is troubling this combination.
If you take a look at efficacy on Slide #44. We are highly encouraged by what we see. This is the strata with BRAF-inhibitor naïve patients. And if you look at the waterfall chart, basically all patients exceed the threshold of a response level or approaching it. So very encouraging data, which are summarized on Slide 45.
We can deliver the combination safely. We have -- importantly, we have been able to dose both agents, Zelboraf and the MEK inhibitor at their respective single agent maximum-tolerated dose and feature which GSK did not do. Our safety profile, we believe, is rather competitive. And we are not -- but a shape at rather even front of the anti-tumor activity in this cohort of BRAF naïve patients.
The Phase III study we are initiating, we are hoping to enroll at a rapid pace, and we are looking forward to a readout of those data.
Just one slide on the last topic, what about Zelboraf activity in brain mets. And you see on Slide 46 a waterfall plot in patient with symptomatic brain metastasis we set up very early on when the drug was just becoming available to patients, and you see clearly that the drug has profound activity in this setting. Since this is particularly relevant since the last proportion of patient with metastatic melanoma ultimately will develop brain mets. I'm closing for now and would be very happy to answer any questions you may have.
Thank you a lot, Stefan. So we are ready for questions. Operator, could you please open the line?
[Operator Instructions] The first question is from Mrs. Hauber, Alexandra from JPMorgan.
Alexandra Hauber - JP Morgan Chase & Co, Research Division
I have 3 questions on breast cancer and 2 on melanoma. So I'm not sure whether I'm allowed to ask them all or should go back after my -- in the queue after the first 3. Firstly, on -- just coming back to the statistics of the HERA study, Sandra's saying this yesterday and you made that point as well that because the lower bound of the confidence interval is 1.05, it is at -- 12 months is actually at least 5% better. But you then qualified it saying, well, that will be in a superiority consideration. So given that this was not a superiority trial, is that still in a very, very pure sense, an accurate statistical message? Because obviously then, that would also mean that survival would also be at least 7% better because of the lower bound of the confidence interval. So that's question 1. And the reason why I'm asking is because of the second question. Basically, what in your opinion is going to happen to the informed consent forms for the PERSEPHONE and the short HERA study? Do they need to incorporate the findings from the PHARE study? And would they have to translate it into plain messages like that 7% better survival on the 12 months? And then the third question is, if you just look -- I mean, yesterday, a big point was made that the study needs to continue and there need to be further follow-up. Well, obviously you were very, very far from the targeted number of event. So -- but just looking at these curves, obviously, we can only speculate, but how realistic is that we look at this data again in 1 year? And I mentioned the number of 1 year specifically yesterday. Is it likely that these numbers will change a great deal, that we will get much narrow confidence interval than just the 6 to 12 months of follow-up?
Alexandra, I start with the first question. I mean, the complexity is really the known inferiority setting. And therefore, we are not really pushing a point, that based on a superiority context that we are able to conclude on the confidence intervals, which don't include one, that 12 months is actually superior. But I think the entire audience got it that there's a strong trend in favor of Herceptin 12 months duration. The standard of care is -- of 12 months is strongly reinforced. But from a stretch terms and then you can talk about prespecified, et cetera, it won't allow you to really derive here a superiority claim. On the overall survival bit, for us, it supported the disease survival message that 12 months is clearly better. But we would not really derive a claim that its -- is superiority. The data are still important because the other trials, 3 are still recruiting, we'll take a look at it. The short duration at the beta Herceptin trial community is well-connected. And we are confident that the independent drug monitoring committees, which are in place for those trials, will take a look at -- a very close look at the PHARE data and will guide those decision on informed consent, continuation, amendment and so forth.
Alexandra Hauber - JP Morgan Chase & Co, Research Division
And within what period would you expect to hear if there are changes? Is that happening quickly now in 3 months, or...
Those trials are independently sponsored by -- from Roche. And those independent drug monitoring committees will receive, I'm very confident, those information very shortly. And we are certainly aware that the IDMC chairs, they have been sitting in the presentation on Monday. So we can't comment on the time lines. It's up to the principal investigators in the IDMC to organize the review of the PHARE data.
Alexandra Hauber - JP Morgan Chase & Co, Research Division
Okay. And then on the majority of the curve?
Yes, so far, there the IDMC said, let's stop enrollment in May 2010. So therefore, they won't get the original target of 7,000 patients. The PERSEPHONE study is still targeting 4,000 but pending the IDMC review. For known inferiority claims, as the study would have been only positive if the upper limit, which is currently at 1.56 would have been below 1.15. I'm not very good at predicting the future, but I believe there is not a shred of a doubt that this can ultimately slide below 1.15. So that is really -- will be able to achieve an known inferiority. I think there is a very low likelihood. But still, the confidence intervals can tighten. It can become an inferiority setting, so that 6 months is going to be statistically inferior by at least 15% to 12 months. And this would certainly be a strong signal but even from the current market dynamic. From Roche perspective, when we see what the medical community is reflecting back to us, the 1 year is reinforced. Because the trial didn't meet its known inferiority, it's good enough that the standard of care for all patients remains 12 months.
The next question is from Mr. Michael Leuchten from Barclays.
Michael Leuchten - Barclays Capital, Research Division
Two questions on melanoma, actually. Firstly, I would like to hear a rationale on Slide 42 to go with 960 twice daily, vemurafenib, and 60 milligrams on the 21-7 schedule, given that the 1C Group, which would have been 720 and 60 milligrams on an ongoing, so nonstop regimen, also appears to be safe. And then secondly, what made the Group, 1D, not acceptable? What was the toxicity that didn't allow you to dose high on vemurafenib and 0973 in the 1D group? And then related to that, isn't there a rationale that you should really, given the mutation status, go for an ongoing dosing schedule rather than an interrupted one?
So I believe you have the fundamental question, how you have to dose targeted agents. The BRAF inhibitor, MEK inhibitor, our target -- you need to dose actually to MTD or not. We rather like an oncology as a principal to dose up to maximum tolerated dose, and the dose intensity we have observed with 960 and 60 was the highest we achieved. The BRAF inhibition is for us the most dominant driver, so we really wanted to go up to the MTD here with 960. I don't believe that the clear rationale that there is a benefit for continuous dosing, and we would have had to compromise on the vemurafenib dose. So the choice was 960 and the BRAF 60 on MEK inhibitor is from our perspective, prudent, the tolerability profile is highly competitive. And as we have seen on the waterfall slide, even across those levels, we achieved good responses.
Michael Leuchten - Barclays Capital, Research Division
And what stopped you? So what was the toxicity that didn't allow you to go for a high dose, high dose, so at the Group 1D?
So those data will be presented actually at the next conference at the SNR [ph], therefore, we would not want to disclose this here. But in a matter of weeks, the data will be in the public domain.
The next question is from Andy Kocen from Redburn Partners.
Andy Kocen - Redburn Partners LLP, Research Division
One is the -- sub-BRAF-MEK inhibitor trial that you're moving into Phase III, the end date on clinical trials that covers is like 2017, which is much lower than your closest competitor. So I was wondering if there's any possibility that, that might end up reading out quicker than that date implies. And then the second one, you mentioned T-DM1 trials, adjuvant trials are moving forward. I was wondering if there's any update on the ability to use PHARE as a surrogate for a full duration like a 12-month adjuvant program in order to get a full approval as quickly as possible. I know you're talking to the FDA about that. Has anything moved there? And I wonder what the status is in Europe of using that surrogate.
Starting with the date of the MEK combo at Phase III, that's an error on the mti.gov [ph], and we will correct this in due time. Certainly, it's earlier, and we will commence this accordingly. On the T-DM1 program, we're having -- done the design phase. We are still working on -- we are still fine-tuning with the steering committees of the trial to fund that last few items. But I think the most specific question you were asking, can we use pCR in the new adjuvant setting as an outcome parameter, a surrogate parameter to obtain accelerated approval, which is then later followed up by full approval when the adjuvant data will report.
Andy Kocen - Redburn Partners LLP, Research Division
So for approved 12 months of course, I think it's interesting.
Yes. But I believe we are still waiting. A little bit FDA guidance on this topic is they have discussed, we have engaged with them and there were a few met analysis and they really have mobilized quite a few study groups to obtain the data to evaluate whether surrogacy can be claimed and how then a regulatory path would be moving forward. I think we are well positioned in this perspective, as we have tried either already conducted or are currently designing to capitalize if pCR would qualify as a surrogate parameter for early breast cancer approval.
Andy Kocen - Redburn Partners LLP, Research Division
Okay. And is the EMA amenable to that sort of idea as well?
I mean, we have in Europe, we have for Herceptin, approval for the new adjuvant setting. And our HannaH data actually is using the surrogate parameter pCR for approval as well. So at least, in this setting, we are working with the authorities. But certainly, it's a formulation change. We go from Herceptin IV to Herceptin Subcu. It is not the first indication ever in this setting.
The next question is from Amit Roy from Nomura.
Amit Roy - Nomura Securities Co. Ltd., Research Division
Amit Roy from Nomura here. Just 3 questions. So firstly, on the HannaH study where you presented some safety data. We saw that there's a little bit of a signal around serious adverse events, I think it was 21% versus about 13%, nearly doubling of serious adverse events. You think they're partly infection related and also a slight into possible cardiac signal with tumor-related deaths and septic shock and MY [ph]. Do you think that may change, just alter slightly the risk and benefits of Subcutaneous Herceptin compared to intravenous Herceptin? So the first question. Second question on T-DM1. After we saw the overall survival data on the median and second one versus multiple for Herceptin, I'm just wondering, what is the updated death table from the first line setting but -- where you have the positive PFF? I know the interim of 4 patients died on the T-DM1 Herceptin, do you have the updates on that, try to get a handle on what's happening in that setting? And lastly, a very simple question on trastuzumab. You said the CLEOPATRA study has overall survival benefit and we'll be getting that in San Antonio. It's really a simple question. You're obviously approved and launching, why do you not release that number? Wouldn't that help you gain further traction with the physicians today if you want to get the drug into America?
I'm sorry, the last question, I didn't catch. What was on Perjeta?
Amit Roy - Nomura Securities Co. Ltd., Research Division
Yes, the last question, I understand you have overall -- an overall survival benefit with Perjeta in the first line metastatic setting, and you will be announcing that figure in San Antonio. My question is, why not just publish that number because the drug is approved already, and wouldn't that help you gain further penetration with the oncologists today after you launch Perjeta in the States?
Maybe I'll start with this one. I mean, we really don't want to jeopardize our relationship with the San Antonio Breast Cancer organizers. They would be very unhappy if we divulged the data ahead of time. We are certainly working, similar to what you see -- have seen on T-DM1, that we organized, synchronized scientific presentation and the publication in a peer-reviewed journal. So we were very pleased that New England Journal of Medicine published the Phase III of T-DM1. We are -- we were pressed to present the T-DM1 data at the ESMO Congress because we had submitted a sub-proof abstract for this Congress. In the meantime, the second interim reported positive overall survival, we certainly disclosed the data here at this meeting. But for us, the most important platform to reach the breast cancer community is San Antonio. And therefore, we go out in San Antonio with the data of Perjeta and overall survival. Covering HannaH safety data, we disclosed the data at that EBC conference in Vienna about 6 months ago. We had highlighted even there proactively that the Grade 3-4 incident absolutely comparable between the subcu and IV arms, and we have observed the signal, as you noted, 21% versus 12% for subcu versus IV respectively on SAEs. We have taken a very granular look, have done numerous analysis on those SAEs, and we believe we have good explanation why those were reported. And sometimes it is just really that investigators took caution and hospitalized a Grade 2 adverse event. And as soon as you are being hospitalized, you know this yourself, I mean, this qualifies as an SAE. So we believe we have significant granularity around those findings and are working with the authorities to explain why just the SAEs but not all other safety parameter, where we saw a very [indiscernible] were an area of concern. So we believe the benefit of risk, as it stands is positive for the subcu bit, but certainly, we have to get this through the heads authority review. The last bit on T-DM1, those items were not yet reported, and I believe not in the New England Journal of Medicine either. And I'm not sure whether -- sorry, I just get the queue. This was for 44, 50. And there, we will have next year the MARIANNEs, essentially a test of whether [indiscernible] T-DM1 versus T-DM1 trastuzumab. And there, you will see far more data in order to make a judgment call whether T-DM1 can replace chemotherapy and Herceptin in the first line testing.
Amit Roy - Nomura Securities Co. Ltd., Research Division
Of course, when we get to Phase III, we'll have the answer. My question was firstly on data death tables from the Phase III. You published the interim one, just the updated one.
I mean, for us, those proof of concept trials allow us to trigger an investment decision. The investment was decided and a further update will not help us because we would not want to reverse the investment decision either. So there would be not a decision coming out of it if we derive the further update.
Okay. So we are waiting just for the data.
The next question is from Mr. Matthew Weston from Crédit Suisse.
Matthew Weston - Crédit Suisse AG, Research Division
It's actually not related to that ESMO data, but with Stefan on the line, I thought it would be a good opportunity to just revisit MetMAb. So ArQule and Daiichi have just announced that they're discontinuing their small molecule TKI because an interim analysis demonstrated that the PFS -- they demonstrated the PFS benefit but illustrated no improvement in overall survival, and that they will not achieve that. Can you just remind us what, if there is a mechanistic reason why you believe an antibody or a biologic approach targeting the external receptor ligand interaction may have more efficacy than a small molecule TKI targeting internal to the cell?
Thank you so much for the question, it's because we believe it may not be really TKI was an antibody, but the active program didn't stratify for any kind of biomarker. And we -- when we record our own MetMAb randomized Phase II, we would not have done a Phase III in an overpopulation because we were only positive in the met high subgroup. So the ArQule investigators chose to run it as unstratified without any kind of personal healthcare approach. It's good to hear that the head [indiscernible] on progression-free survival, and how read the announcement, they must have done a futility analysis already precluding with more follow-up that they would ever be able demonstrate in overall survival benefit. So we are rather pleased that we run, based on a solid scientific hypotheses, the biomarkers stratified program with MetMAb in -- with Tarceva in second-line lung cancer.
The next question is from Mr. Tim Race from Deutsche Bank.
Tim Race - Deutsche Bank AG, Research Division
Tim Race, Deutsche Bank. Just wondering on in terms of the PHARE data, how many patients worldwide -- or have you had any estimate of patients worldwide that are actually using less than 12 months duration generally? And are there actually any upside to current sales of Herceptin from this clearly better data for the 12 months?
I mean, when we look in the areas where we run market research, we are tracking really towards 12 months, like 11.6 plus-minus months of Herceptin ratio we have. There a lot of areas in low-income countries where they may not be able to afford 12 months of Herceptin. And even with the current guidelines, it's better to get some Herceptin than no Herceptin. I mean, we are certainly working across the world to provide further access to Herceptin. We have second brand initiatives. We are looking at how to make it accessible to public payers, and those kind of initiatives may help us to get, first, more patients, and second, get a longer duration. But I mean from the upside perspective, it will be more driven by initiatives like commercial access, second brand, those initiatives we have currently ongoing with Herceptin.
I know that, Stefan, you have to run back to the meeting in a few minutes from now, so maybe you may be able to take one last question.
The last question for today is from Mr. Vincent Meunier from Exane BNP Paribas.
Vincent Meunier - Exane BNP Paribas, Research Division
Indeed -- well, I have 2 questions. The first one is on the analysis of PHARE and HERA. For PHARE, the HERA is talking about the analysis of the subpopulations. What, in your view, could be observed in such an analysis? And also, you are talking about the receptor status of the -- for the patients in the HERA study. And here, what -- can you give us more details on this concept of receptor status? What kind of receptor you think you will be able to find? The second question is with regards to Herceptin subcutes. You were talking about questions from European health authorities. And also Slide 36, you were talking about a device in development. So is that other questions relating to the development of the device?
So I start with the last question. I have got it on the other one. On the subcu, we only have under health authority review the subcu syringe. We have presented here at ESMO an auto-injector, which would allow you, at the press of a button, to inject the same formulation as we have in the syringe. So this would be then a line extension further that the same formulation currently in the syringe would allow then more patient home-based therapy with an auto-injector, but it's currently not part of any heads authority review. The HERA, the subgroup analysis on hormone-receptor positive and negative, we discussed this at length with the current investigator, the steering committee. It's an interesting find that maybe for the negative tumors you may want to treat longer, those patients didn't have any additional therapy like the ER-positive where they had coverage with hormonal therapy. There are some scientific questions to be addressed, even like questions whether in the ER-positive one, maybe the compliance on the Herceptin was not as strong as on the observation arm where they just got hormones, which could have been partly disguising then an effect on the Herceptin longer-treatment duration. But big picture, we say Herceptin 1 year versus 2 year, nothing will change. It reinforces a notion, 1 year is good enough. And compared to PHARE, where they failed to show known inferiority, taking a look at the subgroups were we tried to explain in detail basically -- not basically, all subgroups failed to demonstrate known inferiority, even on the subgroup levels, basically, all point estimates failed to demonstrate known inferiority. When the data will be displayed in San Antonio, I wouldn't exclude, when you look then at [indiscernible], my approach is that you may find subgroups where the treatment arms don't diverge too far. But based on steps, we are currently confident that 1 year Herceptin is the standard of care and will not change based on the steps displayed with further subgroup analysis.
So thanks a lot. I was informed by the operator that he has more questions. I really apologize for those who we couldn't address the questions now. We are available behind the office, so please feel free to contact us. I wanted to thank Stefan. I wanted to thank Sabine for the preparation, all the teams who helped on this call. And I wish everybody a nice day, and thanks for your interest in Roche. Bye-bye.
Ladies and gentlemen, the conference is now over. Thank you for choosing the Chorus Call facility, and thank you participating in the conference. You may now disconnect your lines. Goodbye.
Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.
THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.
If you have any additional questions about our online transcripts, please contact us at: firstname.lastname@example.org. Thank you!