Go/No Go Decisions Based on Early Phase Oncology Trials
Dr Li Yan: Good morning, everyone in China, and good morning from the US. Welcome to the second webinar of Pre-ASCO China Summit 2022 Series. I’m Li Yan, Chief Medical Officer, Brii Biosciences, also the Managing Director of US Chinese Anti-Cancer Association. I’ll be the modulator for today’s meeting. First of all, let me introduce the distinguished conference chair, Dr Jin Li, of Tongii East Hospital, and a former Chinese Society of Clinical Oncology president to give a welcome remark. Professor Li, please.
Dr Jin Li: Thank you, Dr Yan. ASCO annual meeting is one of the most influential oncology conferences in the world. I’m so glad that USCACA and East China Health continue to organise the Pre-ASCO China Summit this year for webinar in May showcasing China’s progress in the oncology. And finally, the in-person China dinner in Chicago on Sunday June 5. Unfortunately, because of COVID-19, we cannot go to attending the ASCO meeting in Chicago. However, we are again attending ASCO virtual meeting from the comfort of our homes and offices, especially for me in hospital for the third consecutive year. Today’s webinar, Go/No Go Decision is based on early phase oncology trials. It’s a very fitting topic for current China oncology clinical trial space. As you know, I have been publicly criticised in the (INDISTINCT) (3:54) and the repetitive clinical trials in China. So, as you know, the PD-1 antibodies (INDISTINCT (4:02) related to that product so they waste our valuable clinical resources (not only the pressures are in the findings) (04:18) but also in clinical doctors and the staff and, most importantly, they waste so many patients. Innovative biotech company CROS, principal investigators as well as investors, should have the courage to say no to the advancing (4:43) to phase trials based on ambiguous or failed early phase trial results. Today when will discuss this important go/no go decision-making process. We are all here from seasoned executives of top successful Chinese pharmaceutical companies, including the RemeGen and HengRui USA and from global and Chinese CROs who provide clinical services to biotech and pharma companies. Now I hope that this webinar will benefit attendees in your decision-making for go/no go decisions based on early phase clinical results and it will help promote efficient and productive use of the limited and valuable clinical trial resources in China and in the world for the better future for our cancer patients. Welcome and thank you. Thank you, Dr Yan.
Dr Li Yan: Thank you, Professor Li, always, for your insightful work and remarks. And I also want to say for folks who are in Shanghai and Beijing, my sincere feelings for you and I do hope that you will regain your freedom very soon. Before we start the session, I would like to express our appreciation to the sponsor, Novotech, for their generous support makes this webinar possible. And please feel free to ask questions via the Q and A window. The speakers or panellists can either reply to you via text message in the Q and A window or answer your questions live. And we also have Chinese language real-time interpretation for this webinar, so for the audience, please choose the language icon and then you can choose Chinese if you prefer to watch and listen to the webinar in Chinese language. So now let’s get started. First let me introduce the first speaker, Dr Sharma, who’s Senior Medical Monitor, Novotech. Dr Sharma.
Dr Ramandeep Sharma: Thank you so much. And I’m sharing my screen. Let me know once you can see that.
Dr Li Yan: Yes.
Dr Ramandeep Sharma: Great. And I hope the audience is getting my audio very clear. So first of all, allow me to thank everyone in the organising team for this opportunity. These are very important decisions, go/no go decisions. And let’s go to the next slide. And the key message at the end of my presentation - I want to convey here that a robust early phase design will help to lay the groundwork for successful Phase III. Phase III trials may not be even warranted. And we have seen that we have now a lot of adaptive Phase II, III design coming in the way. So, for certain indications, if we plan our Phase II (hash) (8:09) Phase III, they are very helpful. From a go/no go decision perspective I’d like to go back upstream and couple of next - of my slides. I'm trying to highlight all the early decisions in the Phase I and Phase II help to lay the groundwork for the Phase III. It's important and couple of points which I - it's very important while designing the studies, while designing the clinical development programs, to put your primary objectives very clearly. And it's important to limit the number of objectives. For example, in Phase I, safety is paramount importance, but if we put too many objectives here, it will limit the trial success. So, these objectives should be very thoroughly thought out right in the Phase I, because these help in what - when we go for our next phases of the design, and they are critical. Now, it's also important that we define and use our secondary endpoints to generate hypotheses to be explored in the future research. For example, even in the early phase, when we are looking at safety, we can define secondary objectives around PFS and how we are looking at the efficacy endpoints. At the same time, the exploratory endpoints from getting data on the biomarkers is also very, very critical. Again, learning from the lesson because we have handled studies right from immune-oncology studies to CAR-T therapies, personalised vaccines and small molecules. The lab objectives also correlate - they help the organisation with translational research. Now, second important point, again going upstream in the early phase studies, who should enter the trial? Typically, we say all-comers when we are evaluating the safety, but this is very, very important and its inclusion-exclusion criteria should be study directed, and they should mimic what we have seen in our preclinical models. So, the selection of the patients can be based on what you have seen in the cell lines, what you have seen in the xenograft models. Very clear-cut vision around the disease time, prior treatments and other variables, right. So, here, proper patient selection plays an important role in the successful completion and the scientific impact. This is kind of a summary. Again, the definitions early on, what treatment we are giving, what is the dose and what dose and scheduled variations are planned. If we are looking at multiple schedules here, how long the treatment, what are the specific steps in the drug administration, and what should be done to prepare, create and monitor the patient prior, during and after the treatment? So, if these are very well defined, we get a robust data in the early phase data or early phase trials to lay ground for our subsequent studies. Again, these are some of the nitty gritties and nuances. Where are the primary lesions to be measured? And imaging is playing an important role in go/no go decision. So, people have debated on whether tumour burden or survival should be taken into account. But if we define these things very clearly, what happens during the treatment, location, size, density, what are the criteria for the response and define the deviations well in advance. They help a long way in putting a very scientifically just background for Phase III studies. At the same time, TTP, TTF, overall survival, duration of response, functional imaging. These are the important markers which give us information. We know very well in oncology space, the Phase I and the Phase II cumulative data gives us surrogate endpoints. When we move into Phase III, it is largely based on overall survival. So, this kind of data garnered earlier during the studies will be helpful later on.
Also very important, again, these are the nuances what we have seen. When will you do the primary, secondary and translational endpoint? Timing of the assessments? Take into account from operational perspective, the dose delays or modification - how will they impact the assessment schedule? And at the end of it, do we have sufficient patients to get the adequate information?
Again, safety and toxicity are important in the early phases, so defining the DLTs in alliance with what we have seen in the preclinical data. (Is the - my protocols) (14:24) have adequate scope to take into account those reductions, those delays and those omissions? And what kind of supportive care, whether prophylactic intervention or secondary prevention has been taken into my protocol? Also important to take into account the discontinuation of individual treatments and termination of study. So, these are the criteria which help.
Now I'll summarise it in three phases. Phase I, objective is to establish the safety in R2PD. Now (INDISTINCT) (15:03) that from design perspective, classical 3 + 3. If you have a very good wide margin of safety observed in your preclinical it works well. Now, whether should I go for a 3 + 3 or should I go for a BOIN-based design? Again, the decision is based on because when we come to BOIN we know the advantages of BOIN designs. BOIN designs give us more robust R2PD but they are also associated with the statistical considerations taken into account early on. So, either way we want to go again should be based on my preclinical safety analysis. If I have a narrow safety of margin, I think recommendation is to go for a BOIN design. Now, there is also a middle way, what we call modified toxicity probability interval design. That's a bit of a bridge between classical 3 + 3 and BOIN design. Purists from academic point of view, (INDISTINCT) (16:15) MTPI are part of our subset of (INDISTINCT) (16:18). But what is the advantage here, when we look at BOIN or MTPI it allows us to accrue the patients at a particular dose level so that we have got more informed decision on the R2PD. And a very important point here is many of the Phase I studies we have seen, now they are combining Phase IA with a IIA phase. So are we looking at a monotherapy or also we are looking to garner information of a combination therapy. So, again, the design should be based on the preclinical safety as well as proof of concept what we have done in our preclinical stages here. Now, if we are planning for a combination therapy, the pathway is - have a monotherapy escalation, go for the combination escalation and dose expansion. On my right side, what I'm trying to highlight here is even in the early phase we have the opportunity to garner biomarker and surrogate efficacy data and there is an opportunity for those (expansion cohorts) (17:31), too, for robust RTPD decisions. All these should be guided. Our protocols should have adequate safety mechanism, SRC or SMC, whatever we call that in the early phase, sorry. And we need to take regulatory inputs if required in the grey area of the DLTs.
Now, we have done a very robust Phase I, we have an idea about the R2PD, we have safety data available. Phase II, the objective is to garner efficacy data. Now, here, it's important to narrow down on the target population. Sometimes we have multiple populations which can be addressed by the investigational product. So, are we doing it sequentially or are we doing it (INDISTINCT) (18:32)? So, this, again, can be taken in because it has financial as well as timeline inputs on. Are we looking only at monotherapy data, or we're looking at the parallel combination with standard of care? So, basically, Phase II is all about planning a proof-of-concept trial. Another important point here is that we can talk about the patient enrichment strategies. Broadly, designs can target the targeted strategy or all-comers. Just an example, with KEYTRUDA they started with all-comers. But as we got a lot of experience with the PD-1 inhibitors, they started working on a patient enrichment strategy, where PD-1 expression more than initially 10% and 25% And some of the studies have said more than 50% expression to be included into the study. Now, the advantage of enrichment strategy is you know that where your drug is working the best and you can have multiple cohorts around that, which gives you a robust data. So, after Phase II, the key point is do I have enough data to go into Phase III? There are statistical tools available which take into account either tumour mutational burden or they have data on the safety. But the key point here when you take a Phase III, is we always look for null or alternative null hypotheses to be validated. When going into Phase III, it is important to plan futility or efficacy analysis in the protocol with the robust input from the stats team. And it's highly recommended when we are going into the Phase III, our independent data monitoring committee should be incorporated to facilitate decisions. All what I've presented, it is based on our experience. We have footprints across various geographies, largely focused on APAC. Novotech is called an APAC specialist. But we do have footprints in US as well. And Novotech provides clinical services from first in human studies to Phase IV studies. We have got very robust different segments and departments who address the various questions, right from biometrics to the safety services, (INDISTINCT) (21:34) management. In terms of experience, where we are coming from, we have experience in oncology in solid tumours, right to gastric and rectal carcinoma. And we have experience with various therapies, right from IO therapies, oncolytic viruses, CAR-Ts and personalised (INDISTINCT) (21:55). And we have been partnering with the biotech companies, right from the study design from Phase I to Phase III. And this is kind of a breakup. A large part of our experience is oncology studies but we have handled other therapeutic areas as well. So, a large part of our clients, if you see from the right side, they are in APAC region. And we do have presence in Europe as well as US. I'll take a stop here and if there are any questions, otherwise, we can take the questions in the panel discussion. Thank you.
Dr Li Yan: Thank you, Dr Sharma, for sharing your thoughts and kicking off the webinar. We’ll park the questions because we do have two veterans who actually worked on KEYTRUDA at Merck and then they moved to BMS. I’m sure we could talk a lot about anti-PD-1 antibody drug developments, which are the first, the second approved in the world, during our panel discussion. Now, let me turn to Dr Xiaoxia Yan who is the Vice-President of Highthink. Highthink is also the co-organiser of our webinar today, so thank you, and please go ahead, Dr Yan.
Dr Xiaoxia Yan: Distinguished guests, dear colleagues and moderator. Good morning. I am from Highthink. My name is Xiaoxia. I’m very happy to have this opportunity to share with you this topic on strategy and cases of innovative tumour drug research and development. As is known to all because there are a lot of many unmet needs in oncology, the regulators they have (INDISTINCT) (24:04) for oncology drugs. There are pathways in (China ) (24:09) accelerated approvals, for example, there is breakthrough therapies and conditional approvals (INDISTINCT) (24:18) so for the single-arm studies that could go into (INDISTINCT (24:22) phase and even launched so there are many policies that could be used because oncology tumour is a very complicated and heterogenic disease and different patients might have different manifestations. And also, the tumours in different organs will respond differently to the drugs and also the genetic mutations will also respond differently to different drugs. And also in the same patient in a different stage of the disease, the characteristics might change a lot. So, for global development programs, we also need to consider the heterogeneity of tumour. So, when we are deciding on the strategy and making the no or no go decisions, it is more complicated (INDISTINCT) (25:17) in tumour. So, go/no go decisions is not only evaluating the safety or the single data. Although there are some early-stage trials, we still need to give a lot of defining data, we need to take comprehensive considerations including safety and efficacy, but they are not the only (language) (25:39) that we do. We need to do a lot of background check, as well as a lot of factors, for example different routes of administration, different doses and dose schedules, we need to understand what would be the impact, the toxicity, the PK parameters, as well as the PD indicators of the on-target effect. And also, is the monotherapy effective or shall we consider a combination therapy? All of these should be decided based on clinical data and also is the drug in combination with other drugs in (INDISTINCT) (26:19) populations, not only indication in terms of population or special population with a specific mutation. We need data to support that and also need to focus on the progresses of the same class drug. Especially if you would like to go to the accelerated approval channel, this would be very important. If there's already another drug in this class that have already been approved, then it will impact your accelerated approval or Green Channel or breakthrough designation - maybe you do not have advantage because the other drug in the same class already approved. And also another consideration is racial differences. As we know that there are a lot of drugs being developed globally, there’s a need to consider the racial differences because race might mean that there are differences in aging for that gene expression in the prognosis, in also the manifestation, and also the data from overseas and China, we can extrapolate. And we can do some data analysis. The early-stage data might come from a small sample size. Although the results might not be ideal maybe it's because of the sample size. Another thing to be considered is the feasibility of clinical operations, especially for genetic mutations. We encounter this a lot. For example, the (27:58) failure rate is high and in this situation, it would be very challenging for us to enrol patients and the development would need a long time before the launch is finally made. Then it will require a long time and a lot of efforts and it would also impact our go or no go decisions. And also, we would need to consider time and cost and financial considerations. And apart from that, in order to make go/no go decisions, a multi team, multidisciplinary decision, it’s not a decision made by one person, one team. So, now we have this regulatory platform to help you with the data evaluation and assessment, including the pharmacology and the mechanism action experts, because they need have a good understanding of the mechanism of the trials and the targets so that it’s easier to identify the target population or indication and also we would need (statistician) (29:09). Just like the previous speaker has mentioned that statistician needs be involved as soon as possible because they are not only important in sample size calculation or the threshold calculation but also, most importantly, when we acquire the data during the study, they need to help us to do the data analysis, especially subgroup analysis. As we know, the launch of many drugs start from subgroups’ efficacy and results. So many of the time we discover the good dose and population from the subgroup analysis. And also clinical experts are very important because they can help us to do the assessment of the current therapies. What is the position of this new therapy, this new technology, what is advantage if there's new technologies to launched, and how do we leverage that advantage?
So that would be very beneficial to us, would be more accurate population targeting. And then we will need to involve the registration experts. The reason is that the whole process of registration needs to be compliant and needs to be legal. For example, after monotherapy, there might be some combination therapy, (and there are many other ways) (30:30) that combination therapy needs to apply for registration and approval. But there are many companies, they started a combination therapy studies directly, without submit for the registration. However, this is against the regulation. Then, also, when we are making a decision, we need to seek communication with the government, especially for the accelerated approval pathway and the design pathways, for example. There are some companies, they have some ideas, they have some (INDISTINCT) (31:01) studies planned and some real-world study method. Whether these studies could support our future launch or future studies, you need to confirm with the regulatory bodies, the government. It's not a (INDISTINCT) (31:15) yourself, because the regulatory bodies, they might not support your studies for future launch. So you need to confirm with them.
And here I would like to share with you some of the cases. So because many of the data are confidential, so sometimes just block the data. So this is the immunotherapy anti-tumour drug. The design is in a first human study based on preclinical data, and also based on some same class target, the drugs were the same target. We did the three doses, we (INDISTINCT) (31:54) the three doses and then we had the expanded study. We designed two groups. The first group is combination therapy. And then we did another illustration, and we were approved for this combination therapy study in the second cohort. Actually, in the previous dose escalation we had a whole population, but we did genetic testing for each patient. And then we discovered that there's one genetic loci that might have a predictive value for the efficacy response though we did not have data in other countries’ studies. So we're considering doing a single arm study in these patients with this genetic mutation. So, this is the patient selection. And also compared to (where) (32:40) the current study protocols. And another very important factor is our prediction for future efficacy, how much efficacy we want to improve before we can get recognition of advantage and also when statistician involvement, because if the expectation is that we might apply for the pivotal study, because this is an expansion study. And through a thorough evaluation we entered into a study in this process because it’s a single arm, open label study. We followed that closely with safety and efficacy study and finally the result was very good.
We have the results that we have expected, and also even better than our expectation, our (INDISTINCT) (33:31) has been improved by a further 0.5%. So now we have already gained the authorised approval in the future, so we are conditional approval. This is an alternative indicator as an endpoint. So this is single arm study, we still have a lot of things to go, for example, we need to dynamically improve the data in the later stage and also in the single arm study there are many documents required if you want to launch this product for registration, for example RIC and PMC and also your communication record and IMT. And very important thing is for efficacy and safety dose and exposure dose required. So we need to submit a safety report. This report is not only for this particular study, but for all the studies because this one study is only part of the studies for the launch of this product. Apart from the study we also carried out the studies under other indications. So the data that we submit are from multiple studies. When I designed our strategy, we started multiple perspectives at the same time and also in that particular population we wanted to get accelerated or prioritised approval. So the result was good. We had what we wanted to achieve. But does it mean that we're very successful in this case? So we have our concerns, because right now we used an alternative indicator. We still don't have an (OS) (35:16) data. I can show you the paper. So could this ORR data be transferred into an OS benefit? We continue to study to the end that we want it to be. So as we can see that our data is good but, in the end, OS, we haven’t seen the OS benefit.
So the result of the study, is it a no go? Not necessarily, because this study failed because we know (that greater than three AE) (35:56) is twice that of the control group and also, it's combination therapy. Maybe next step, the dose optimisation will be the space for development for next step. And like the next case, I'm showing you the data from the paper in 2000. This product was approved in a prioritised approval pathway. But in the 10 years after that, we found that there are a lot of toxicity, liver toxicity and pulmonary toxicity. But they optimised the dose, they used the low dose, more frequency, in terms of the (INDISTINCT) (36:39) and in 2017 - they acquired the approval in 2017, re-approved so we can see that dose, in the past, was 9 milligrams and now it was 2 milligrams per square metre. And we can see the induction 6 and then 3 and then 2.
So we could see the time that is spent for the previous approval and the second approval, more than 10 years has been spent on those. How many of the companies would have the perseverance and determination to continue doing that? Not all of them, I think. Go/no go decision - there's one principle that we need to stick to, no matter you decide to go or no go. We need to make the decision based on the clinical needs and to be oriented on the patient's benefit. This is our first principle. And then after we launch, we'll have a good market potential. And also, we need to have communication with the regulatory bodies. Sometimes the company would ignore this, but it's very important, this communication. And in the consideration of the pathway and designer for clinical trials, as well as the methodology or the protocol of the studies, this will all need communication with the regulatory bodies. They might not give us the answer, the precise answer, but they can help us to correct our directions to guarantee that the direction is not deviated from what we want. So before (many people do studies) (38:18), the regulatory bodies will ask us to have communication with them. Actually, our program, we already started communication in the interim study. So in a single arm study registration, this is already a very good foundation because we started earlier communication with them. And also our utilisation of the scientific tools will also be very (INDISTINCT) (38:43) AI technology. And it will be very beneficial for the selection of biomarkers and compounds and also patient selection, (INDISTINCT) (38:53) selection. So we can use some technologies to help us to clarify on the direction and also for the constitutive pharmacology. In many innovative drugs, not only oncology drugs, we use this method. Especially in the (edification) (39:12) of those and extrapolation of population, this technology is very beneficial, very helpful. And with accumulation of the data, you have a more robust tool to support you. So you could contact us if you have certain needs in the development. This is what I want to share with you. Thank you very much.
Dr Li Yan: Let’s move on to the next speaker. Let me introduce the third speaker, Dr Jianmin Fang who’s a co-founder, CEO and CSO of RemeGen. Dr Fang.
Dr Jianmin Fang: Hi. Thank you very much. Thank you, Professor Li Jin and Dr Yan Li for inviting me to join this meeting. It’s a pleasure. So I’m going to talk about the (INDISTINCT) (40:34) RC48 in urothelial cancer. (INDISTINCT) (40:41) our decision was made from Phase I to the clinical study and also highlighting (INDISTINCT) (40:53 – (41:10) that can release the toxin more efficiently and also the (INDISTINCT) (41:14) of the Cytoxan is (MME) (41:14) which has a bystander effect, meaning it kills the neighbourhood tumour cell, even the cell is not the (INDISTINCT) (41:26) so it's more in the tumour with the heterogenous expression of antigen or the (INDISTINCT) (41:34) will be more efficacious. So that’s the purpose or design of the original structure. And based on our early study in the (INDISTINCT) model, we’ve shown that this (INDISTINCT) (41:47) actually is high affinity for (antibody part) (41:51) and also high efficiency in (INDISTINCT) (41:52) for the cell. And also really can, after injecting into animal, you can see the accumulation of this toxin micro drug in the tumour site and compared to the normal tissue or the plasma so it really can deliver the (INDISTINCT) (42:15) to the tumour site based on antibody or ADC mechanism. So this is a HER2 targeting ADC so in a HER2 expansion, especially in the HER2 low expression tumour - we see that in this xenograft model showing that, you know, it’s more efficacious compared to another ADC HER2 ADC drug T-DM1 so DM1 is good in high expression tumour but not really doing well in HER2 low tumour. But this new ADC actually still can achieve significant anti-tumour activity, even in the low concentration (INDISTINCT) (43:05) concentration.
So this gives us the confidence that move on to the clinical and apply for R&D in China. So actually this was the first ADC drug R&D in China, so (INDISTINCT) (43:24) required time to evaluate this application and finally we sent the application in in 2014 and got approval in 2015. For that time, actually, the overall regulatory landscape is quite different from now and it was considered still good that we took one year to get R&D approval. So then, in 2015, we started the Phase III study and so actually we did two Phase I for those escalation (INDISTINCT) (44:07) and one is traditional breast cancer study. Of course the HER2, of course the breast cancer is obvious indication, and in this study we have different dose, this 3 + 3 kind of design and for the HER2 it’s the traditional HER2 definition. First part was defined by (INDISTINCT) (44:37-44:42) so this is a traditional HER2 (INDISTINCT) (44:46) definition. Actually, this study was led by Professor (INDISTINCT) (44:51). Then next, we initiated a second Phase I study (INDISTINCT) (44:51) for the broader cancer type, so solid tumour, mostly gastric cancer but we also included other tumour types. So this is actually, as a result, turned out to be this study, 002, we started first dose, 0.1 is our first in-patient dose. Then also we expanded the HER2 definition in this study so not just to include a conventional HER2 (INDISTINCT) (45:40) definition IHC 3 or IHC 2 (INDISTINCT) (45:45) but we also included the (INDISTINCT) (45:50) but this FISH negative. This was previously defined as FISH negative but now we define it as a part of HER2 low expression tumour.
So during this study, actually, besides the gastro cancer and the other cancer types, we were also interested in the dose escalation in order
(INDISTINCT) (46:13 - 46:28)
Then she got a patient from Dr (INDISTINCT) (46:32) also in that Beijing cancer hospital and this patient actually responded very well, partial response. Then it becomes very interesting, because the urothelial cancer (INDISTINCT) (46:52) in that time and HER2 therapy actually was now successful (INDISTINCT) (46:58) so we think that this may be interesting phenomena so we tried to enrol more patients. So, as a result, during the dose escalation in the expansion stage, we enrolled four patients for the different dose. And finally, in these four patients, two of them got a partial response. So it’s (INDISTINCT) (47:30) really good. So then, based on this limited information we decided to initiate a Phase II study so our team member discuss with Professor (INDISTINCT) (47:45) in his office and decide (INDISTINCT) (47:50) go to trial. So then we studied a Phase II study for the urothelial cancer.
So actually, by that time, the treatment for the urothelial cancer, especially the (metastised) (48:10) you will see was limited and most treatment is first in line chemotherapy, more common (INDISTINCT) (48:23) and after (INDISTINCT) (48:24) the chemo, then there’s not much option. It’s only after 2017, then the PD-1 and PD-L1, this checkpoint inhibitor came around and become (INDISTINCT) (48:44) and initially approved in the US and then more recently in China. But even this second and third (INDISTINCT) (48:53) immunotherapy, ORR or the PFS is also limited so still there’s lots of medical need for this so that’s why we’re really interested to move on for the (UC) (49:12) .
And also in terms of HER2 expression, actually UC is a tumour that was really good HER2 expression and now if you look at this graph you can see that if you also include the IHC 1+ or 2+ then the number of percentage of HER2 expression actually is very high, it’s almost like 60% of the patients, so this is a good opportunity for this drug development. Also the HER2 expression is associated with the patient survival, the poor patient survival with the high level expression. But previously, before our study, a number of HER2 targeting (INDISTINCT) (50:11) had failed in the UC, including, for example, the Transtuzumab, TKI, even T-DMI also failed in the second therapy, so it was very challenging and (INDISTINCT) (50:29) HER2 (INDISTINCT) (50:31) for the target therapy for UC.
So then we did a Phase II study based on the (INDISTINCT) (50:45) study we had a dose reaction with 2.0 mg and also Q2W the frequency and also because this dose and this interval, the safety and efficacy we see overall is good. So based on all this early information, we decide to plan for a 40-patient Phase II study. So then the result actually was quite good. You can see the result here and here and actually (ORR) (51:25) was 51.2%, the OPR, and also many patients (they’re also SD) and mPFS is 6.9 months, OS is 13.9 months. So this is actually pretty good and we did not finish this study. During the middle of this study, we already saw that data looks pretty good so then we communicated with the (CD) (52:02) and had a meeting with CD and presented our data. (INDISTINCT) (52:09) agree that we can convert this Phase II as a pivotal study for the single-arm (INDISTINCT) (52:19) potential conditional approval. But CD did not agree our request and they ask us to initiate a separate trial for conditional approval. So basically, yes, you can do a single arm study as a pivotal study but you’ve got to start it as a separate trial, not this trial. So then we agreed with that and then (INDISTINCT) (52:55) for the second trial, second Phase II trial. And also in this meeting, CD required us that – for the prior therapy, not just the standard (INDISTINCT) (53:13) and also have to include (INDISTINCT). (53:17) So this is a little bit different from the common therapy but we think it’s fine, it’s okay, so then we did this second Phase II trial as a pivotal trial. This was based on our statistical (INDISTINCT) (53:43) actually the patient number was quite aggressive, about 60,000 patients for a (INDISTINCT) (53:52) trial.
So actually the result was also quite similar – also there were additional drug and the chemo but overall, result is still similar, so cORR is 50% and mPFS 5.1 and OS 14.2. So the trial was successful. Then in 2020, based on the result that we had pre BOA meeting with CDE and CDE agreed that we can (fire) (54:33) BOA and so we have this result from two trials, two Phase II trials, and applied for BOA and it was approved in June last year, become the first ADC drug by a company approved by the Chinese CDE. So that drug now launched in the market and actually – sorry, the UC was approved the last day of December last year. The first (INDISTINCT) (55:15) was gastro cancer approval and then the UC in December so, yeah.
So basically now this drug has been approved in China and also for UC now have two other ADC drugs approved by FDA and not in China yet. (INDISTINCT (55:42 - 55:48) the HER2 targeting ADC (INDISTINCT) (55:52) actually is better than the other two ADCs for the UC.
So after that, actually we are also exploring the combination study with PD-1 antibody and we did a little dose exploration, the 1.5 and the 2.0, and the PD-1 antibody is (INDISTINCT) (56:25) 001 and then the – we call the C0 4 study and from the result actually it’s very amazing that most patients got very good response, (INDISTINCT) ((56:51) and if we look at the HER2 expression level, in the IHC 3 all three patients got (PR, OCR) (57:04), then the IHC 2 + is 77. 8 OR and even in the IHC 1 + was 66.7. In the 0 expression, we also see (the half patient) (57:23) got PR. So basically, except the one patient, the (INDISTINCT) (57:27) patient, most patients, they’ve got the good tumour reduction. So this is very encouraging and based on this data also we initiate a Phase III study for this combination therapy with PD-1.
So right now, there’s a number of studies ongoing (INDISTINCT) (57:27) UC will also have breast cancer, we have other gastro cancer on the program but in UC now in China we have the C016 study, which is this ADC + PD-1 antibody in the first line and (INDISTINCT (58:09) UC as a Phase III study and we also have a C017 which is also combination with PD-1 but with neo-adjuvant (INDISTINCT) (58:19).
In the global part, this drug now has been (INDISTINCT) (58:25). Previously, Seattle Genetics, a leading ADC company, (INDISTINCT) (58:31) take over China, Asia study and now together we have G001 study. It’s an ADC drug plus – it’s a single-arm monotherapy in the second (INDISTINCT) (58:31) so it’s ongoing. And (INDISTINCT) (58:59) therapy designation to this second UC for this drug.
And also there’s another trial called the DV001. It’s a combination with the PD-1 antibody, it’s a first- line UC and a Phase III study. This is now in the stage of communicating with FDA to finalise the protocol. So basically, we have lots going in China and also outside China. So this is really a process, (amazing process) (59:38) that we did not expect that HER2 will work for the UC (INDISTINCT) (59:44) the clinical studies (so when we find the early sign for that) (59:52) and also overall good safety profile then we decide to explore this indication and then move on quickly to the pivotal and then approval in China and now it’s not just in China patient, also for the global patient. So we hope this is example how the innovation China works, that initially benefit patient China, then can go to global and also take advantage of rapid development and patient enrolment in China. So this was a great, a fantastic journey for us. So thank you very much.
Dr Li Yan: Thank you, Dr Fang, for sharing the very successful RemeGen story with us. Congratulations, again, with the big deal with Seagen and really great to see the drug is moving very well in the US. Now, let me welcome my dear friend and our last speaker for today’s webinar, Dr Joe Eid, Chief Medical Officer of HengRui USA. Joe.
Dr Joseph Eid: Thank you, Li, and thank you for the organising committee for the opportunity to share with you, you know, some of the thoughts on early development in oncology. And I want to also apologise if there are some repetitive themes that we have seen in the prior presentation. Let me put this in presentation mode. Okay. So drug development, you know, is a long process, is a complex process. The overall aim is to create medicines that are effective and safe and to provide also meaningful benefit with an appropriate benefit-risk ratio. The process is highly innovative. It's cost intense. Cross-functional effort is conducted. And there's also a regulatory framework that is evolving in a highly competitive environment. The process itself is iterative, going over many years with defined decision gates. The sponsor and the development team aim to make optimal choices between several alternatives along the way, anticipating changes in the therapeutic landscape. So the bottom line is the development sponsor must take timely decision at different levels in order for this process to be successful. So we've seen in the first presentation, the various stages, going from I to IV. Each one has a specific goal and objective. The evolution of the drug development, starting with discovery and the early phase where we have the least amount of certainty, the questions that are being asked are do we have a drug? Which compound? Whether it's a lead or backup to accelerate or to kill? And in the later stage, you have the following questions. Is the optimal design for the confirmatory trial the right one? What is the probability of success? And, as you can see, moving from left to right, there's an increasing level of evidence that has been accumulated, that helps make a decision. Now, in order to maximise development success, you have to have different ways of planning a trial and designing a trial, starting with the umbrella design, and the basket trial design - those are different. Starting with the umbrella design. the battle study, which was conducted in lung cancer, had the same histology, lung cancer, with different genetic mutations and each of those cohorts was randomised to a drug or a combination. On the right side is the basket trial design where you have multiple tumours that carry the same gene mutation or molecular biomarker. I actually have, you know, the privilege to have led work on the Keytruda program where we had a successful basket trial, and that was the KEYNOTE-16 followed by the KEYNOTE-158 in (MSI high) (64:52). That was a positive and it led to the first FDA approval of histology agnostic cancer led by a biomarker. The second trial that we tried the same strategy and that was KEYNOTE-28 using PD-L1 one as the biomarker in 20 tumours. That trial was, you know, considered by the FDA as not appropriate for approval. But this is the difference between an umbrella and a basket trial where you are asking the same question in one protocol targeting different diseases, different biomarkers and different hypotheses. The other trials that I will detail on the subsequent slides are the adaptive designs, biomarker enrichment designs.
Trials can be fixed, where you have the patient population, the disease assessment, the treatment, the sample size, the hypothesis, the primary outcome all fixed. There's typically no change in the design features during the study. The adaptive design, on the other hand, includes a prospectively planned opportunity for modification of one or more specified aspects of the study design and hypotheses based on analysis of data from subjects in the study. So the data coming into the study is being reviewed and informing the next step on the study. It also allows for planned design modification. And the modifications are based on data accrued in the trial up to the interim analysis. This could be blinded or unblinded. The control probability or false positive rate for multiple options and control operational bias. And it assumes independent increments of information. The other aspect of, again, informing a go/no go or a decision milestone is biomarkers. Those are characteristics that are objectively measured and evaluated as indicators of normal biological process, pathogenic processes or pharmacological responses to a therapeutic intervention. Biomarkers could be used for screening, prognostic or predictive outcomes. Biomarkers can be any of the following: it could be clinical, could be physiological, could be morphological, or could be molecular biological parameters. The key factor in biomarkers is that it tells us which patients not to treat and by definition or by consequence, it tells us who will benefit the most.
The enrichment designs, you know we have different categories of enrichments. You have the untargeted or all-comers designs, post-hoc enrichment, prospective and retrospective designs. Marker evaluation is done after randomisation. One example comes to mind is the KRAS in cetuximab colorectal cancer. Marker stratified design where the marker evaluation is done before randomisation and is applied to all patients randomised. You stratify by marker and the example here is the histology in pemetrexed lung trials. And marker strategy design where you have randomisation to two strategies, marker based and not marker based. In the marker-based strategy you assign marker positives to new product and negative to the standard of care. And the original design where the market evaluation is done before randomisation, and only marker positive patients are randomised, such as in the BRAF mutation Vemurafenib studies of melanoma. And the adaptive enrichment design where all patients are randomised and the marker evaluation before the interim analysis is done on all these patients. Based on interim results, you stop accruing the marker-negative patients if found either futile or unsafe to treat patients with the new product.
And the principles of decision making follow the following steps. There's the challenge of making a future decision where the outcome is not known. You build on the science and the data and you have to acknowledge that there's uncertainty in the process, and you apply good judgment. There's a clear link between a scientific question, a method, the results and the decisions and actions that you take. The good decision is an informed choice made in a transparent and consistent way. And you have to be clear upfront. This cannot happen at the end where you will have different opinion.
And it has to be based on valid, albeit incomplete, data and applying judgment. And in clinical development, a good decision can be moving forward or terminating a program, because that can save, as Dr Jin Li said, patients, physicians’ time and resources.
Some factors to consider and this has been used in different, you know, configuration, but there are five key factors to consider in clinical development. Factor number one is the novelty and importance of target and mechanism of action which we can be informed in the preclinical models (that activity) (71:19). The second factor is the pharmacokinetic properties and the route of administration. The third factor is the activity and toxicity in the early Phase I and II stages. The fourth factor is the potential role in treating a particular disease alone or in combination. And the fifth factor is the strategic role or the market position of the of the asset. All these obviously have to be aligned with the regulatory and the clinical environment in order to be successful. And this is a proposed classification matrix that, you know, is used for novel molecularly targeted therapies. And if you move from left to right, you have biological activity, strongest on the left and diminishing to the right, and you have safety or toxicity from top to bottom with more toxicity on top and modest to lower toxicity on the left. The green is where you want to be. And you can see that's the minority where you have a very effective drug with a very safe profile. If you move to the extreme right, you have drugs that are the least effective or no activity and higher or significant toxicity. And you have the in-between. That's where, again, good judgment comes into play because if you're in the red or the green, those are the easy decisions. It's that orange middle ground where how do you select the best population that is going to respond? Is it a biomarker? Is it a Bayesian analysis? Is it a basket trial? You know, it's going to depend on what information you collect at what stage on the program so you can select the winner in the study design. And that's where, you know, the money is made or lost, in those groups. And before we conclude, a case study, which incorporate many of what I just talked about. The case is the BRAF inhibitor and the MEK inhibitor in melanoma. We knew early on with the success of the BRAF inhibitor that there is an emerging resistance driven by the MAPK kinase pathway. And to address this problem a Phase I and II trial of combined treatment with a BRAF inhibitor, dabrafenib, and a MEK inhibitor, trametinib, was designed. This trial was an open label dose escalation Phase I, II study looking at safety pharmacokinetics, pharmacodynamic, and clinical activity of BRAF inhibitor and a MEK inhibitor in subjects with the BRAF mutant, metastatic melanoma, a total of 247 patients.
The study design is shown here. The key eligibility criteria are standard for this population. There were three parts of this study. So you can see already how Part A informed B and C. The Part A was in BRAF naive, looking at PK and drug-drug interactions. Part B was an expansion cohort looking at a group that was BRAF naive and a group that was B AF failure to, again, inform the next step, which was the part C, where there was a randomised cohort in three equal groups - a monotherapy group, a combination of a lower dose group, and a combination with a higher MEK dose group and allowed patients to cross over from the monotherapy to the combination if they failed. So this trial combines biomarker, combines an adaptive design, combines a gate and decision making along the process, and that actually led to an approval from this trial. So there was no need to do a Phase III trial over many years and hundreds of patients. So a total program of 247 patients led to an approval in a very truly unmet need in patients who benefited immensely from this innovation. So, in conclusion, clinical trial decision making should start when planning a study, not in the middle, not at the end. Decision making in drug development is a cross-functional effort. It requires from all team members general understanding of principles for good decision making and a mindset embracing constructive debate. The statistician has a key role to guide the cross-functional team. Along with the statistician, the clinical scientists must work closely together to develop the criteria to make the go/no go decision. The decision-making requires a transparent and consistent approach to define that go/no go criteria and this includes the supporting evidence and underlying assumptions. Teams should be clear on the question they aim to address and the action that is related to each of the possible outcomes. Decision frameworks provide decision recommendation but the final decision must be made considering the totality of the data. So even, you know, when you have a successful program, you have to take into consideration what disease, what indication, what is the competitive landscape, what is the business priority. So it's a very complex decision that leads eventually to taking a drug into full development and eventually to approvability. So thank you for this opportunity and I hope this was informative and I hope we can move to the next stage early.
Dr Li Yan: Thank you. Thank you, Dr Eid. Your presentation essentially summarised the whole session. So great, thank you. Let's now move on to our panel discussion. Dr Jin Li, we welcome you to co-modulate this panel discussion. Before we start the panel discussion, I would first like to welcome two panellists, Dr Vivian Gu, Head of Clinical Department and Regulatory Affairs, Chief Medical Officer, Novotech, China. And our other panellist is my dear friend, Dr John Wigginton, Senior Advisor and the Chairman of Scientific Advisory Board Cullinan Oncology, who's also a former head of Immune-oncology, early clinical development at BMS.
So, welcome all the panellists. If all the panellists could just - as well as the speakers so you're all welcome to join the panel discussion, the go/no go decision-making. Dr Jin Li, welcome back to this session, to this webinar. See, Professor Li is a very popular chairman so I think you're co chairing two different webinars. So let's back to the early go/no go for oncology. So during the four discussions, I think one thing just jumped out. And I already mentioned, you know, both Jon and Joe, you both worked at Merck and at BMS. At that time, if we still remember right, Merck was four years behind BMS in terms of the anti PD-1. Five years behind BMS in terms of anti PD-1 development, but Merck was able to surpass and caught up with BMS, and actually got the US approval ahead of BMS. So both of you were involved in Keytruda development and later on, you went to BMS, so you could actually provide insights from both sides. Could you just share with the audience from your viewpoint what did Merck do right, and what BMS didn't do that really resulted in this five-year gap that was (INDISTINCT) (80:29) and then leading to Keytruda got approval first.
Dr Joseph Eid: Jon, do you want to talk about Merck and I’ll talk about BMS?
Dr Jon Wigginton: Yeah. Well, I think to be honest, I mean, again, you know, I was Head of Clinical for BMS and we did the Phase I, large Phase I studies with Nivo, and presented those in 2012. The two trials with PD-1 and PD-L1, they were presented at ASCO and published in the New England Journal. And those were very important studies. And, really, I think ignited a lot of interest in the field. I think what Merck did, you know, Merck came back the very next year, and presented their very large Phase I studies as well. And I think that it's hard to retrospectively (quarterback) (81:31) some of the key things, but I think a couple of things that stand out to me is that Merck bought into very early on a biomarker strategy, which I think has, you know, over time served them well. And I think Merck really got out front in using Phase I data, actually, to support filings, to support the actual filing, which I think was an important strategic move. So I think a lot of interesting things from both perspectives to learn from. Joe?
Dr Joe Eid: Yeah, thanks. Thanks, Jon. And what I would say that BMS did was pioneer this space, and thanks to Jon and his colleagues, actually presented data that showed that this target works. And that gave a lot of encouragement and support to the other companies, including, you know, my company where I was at the time, otherwise that program would have not survived. And, you know, what BMS also did well was generate the science on biomarker. Now, later on, the focus on biomarker was less, I would say, at BMS than what Merck did. And, again, when you're behind, you take two approaches, either repeat what the leader did and come back 2, 3, 4, 5 years after them but repeat essentially the same course or forge a different course. There were hints from the programs that BMS did that, you know, there is something to the story of PD-L1 one. So we did an analysis of tissue and targeted the tumours with the highest expression of PD-L1. The other component, which was more logistical, and Jon alluded to it - every protocol that goes to an IRB, that goes to an execution, you have contract negotiation, you have time approvals, etc. So instead of going the classical course of repeating a sequence of trials, we started building on the same trial and amending it to add cohorts but minimise the time between the next cohort and the approval and the reviews. And also the costs, I would add. Now, the good part and the bad part in that is at the time, the FDA didn't have a point of view. After we accumulated hundreds of patients on the Phase I and got the approval of two indication, melanoma and lung, and a biomarker (off) (84:39) the same Phase I, the FDA said no more because they need to have the separation between the different cohorts. And also, the database was not built to be a regulatory level database. Again, we got it past because it was a breakthrough, it was innovative, and it was helping patients. So those are the lessons that could be applicable in in different mechanism of action and diseases
Dr Jon Wigginton: Yeah. Joe, let me add an add something to that as well that I saw as a real positive (if I'm not correct) (85:18). You know, the Bristol trials that we came forward with were I think about 300 and 200 patients respectively for PD-1 and PD-L1. As I recall, your trial at Merck was upwards of 1,200, 1,300 patients, as well. So very rapid accumulation of an even more dramatic dataset. And then I think the other thing, correct me if I'm wrong, didn't Merck also very early on go to (INDISTINCT) (85:46) review on their scans?
Dr Joseph Eid: We did.
Dr Jon Wigginton: Which allowed them to, when they went to talk to the FDA about response rate, talk with a different level of credibility, right, then (some can) (85:57) if you don't do that.
Dr Joseph Eid: Yeah, I mean, we did a lot to accelerate, improve our chances, etc, including monitoring Congress deliberation of a breakthrough designation to see when that would happen. And we filed first with the (FDA) (86:18) on that ground.
Dr Li Yan: So a couple of things, I guess, just summarise both of your discussions. One is really, (Eid) (86:28) to know what you're doing because KEYNOTE-001, as what Dr Eid alluded to, is not something that a small biotech company could actually take on. Even Merck, Merck had to pour in essentially a large part of their biostats programming data manager resources just to get that data clean from each cohort and to present (discuss) (86:52) with FDA. This is not something that a biotech will be able to do. The second thing is, I remember again Joe alluded to is KEYNOTE-001 actually was the first trial really fully leveraging the breakthrough designation. And FDA was really -their agenda to really push and make a good example on the breakthrough destination. So both sides, you know, aligned, and it will not happen again I would say. It will be very hard to make a KEYNOTE-001 repeat that trial and repeat the outcome again. But history was made. So congratulations to both of you. Professor Jin Li, I’ll turn to you to ask questions and the next one.
Dr Jin Li: Sure. Thank you. I think that you just gave us a very good talk and showed us how to make decision during the launch of the early stage of a clinical study. So I just have a question, because -you are coming from a very big company, you don't have to communicate with the investor, but for a small company, they will, (INDISTINCT) (88:19) by investor during the investigation period, so if the results is not so good, how can you give them some suggestion for the methods to communicate with the investor?
Dr Joseph Eid: I mean, I would say that science and data is king. So first, you have to have the conviction of the work and the development you're involved with. And that has to be present in the work you do as well as in the communication, number one. Number two, it's also good to take lessons from other programs. Not every program is a breakthrough designation. As I showed on the table, it's a rarity to have a winner, a star. More often you are in that middle ground where having the sense of the field, the drug development in the field, and providing and educating and informing the investors of what were the successful programs and why. And, you know, it's not something that happens overnight. It's a long process, and it has to be a deliberate progress and clarity in the communication. And building the right expectation is another factor in speaking to investors and making sure that they stay strong and committed to Investing in your company or your assets.
Dr Jin Li: Yeah, thank you. Sometimes the staff from medical department, they communicate with me. They want to know exactly how efficacy of the drug at the very, very beginning. So say “You have to give me some results, because the investor, they are very eager to know the results. If we want their support further, financial support, we have to give them the results.” So I call it the capital driven study. So I think some of the companies, they have that experience. Thank you very much.
Dr Vivian Gu: This is Vivian. Probably I can add a few comments because we work mainly with a lot of local biotech companies. So, I will say even before the study starts, probably we need to have a go/no go decision in-house. I mean, we will compare the historical data to have the go/no go decision in-house or something in between (INDISTINCT) (91:15) to be decided. So if the result from the early phase study is something in between or even not that good as we expected, probably we can use. Just even in the early phase study we have adaptive (INDISTINCT) (91:30) right now so we can (backfill) (91:32) a bit more patients to accumulate more data to give more evidence for the further decision. So that's my comment. Thank you.
Dr Li Yan: Absolutely. Actually, Dr Vivian Gu, I do have a follow-up. Not a question, but a topic for you to comment on. In China, a lot of biotech companies work with CROs, especially during the early stage as well as late-stage drug development, including CRO such as Novotech. So could you comment - during such interactions how could a CRO, who actually sees the data before the sponsor, right? You see the data before the sponsor? And how could a CRO make – is it possible, is it reasonable for CRO to actually play a more proactive role rather than just a passive role, whatever sponsor decides, and you will just execute it. And even though you see the data and you don't believe it's a good drug in such a situation, what would you do? Do you just take the money (on the run) (92:46) of Phase III you know it's going to fail? Would you do something different?
Dr Vivian Gu: I think it's a very good question. I believe, as the doctor, we need to make the decision as you just mentioned, based on the size. So we will, I think, work together with the sponsor, not only the sponsor, but also the investigator to make the right decision because, at the end, are all for the patient. And so that's what we will do.
Dr Li Yan: Okay, great. I think just as Professor Li Jin mentioned at the beginning of this webinar, the reason why we're here together to discuss is to make sure that we ring fence and protect the very precious resources, right, the funding, the physicians, especially the patient resources, to only advance good drugs, but not to advance drugs because of capital driven reasons. Back to you, Professor Li.
Dr Jin Li: I have a question for Joe. Because sometimes at a very early stage of study, the efficacy. I mean, after the (RP2) (94:02) is decided but efficacy is not so good. But after they analyse the data, they find out some subgroup of the patients are still good. So how can you give the projection to the sponsor how to go further to put a (INDISTINCT) (94:28) to get the indication?
Dr Joseph Eid: I think Dr Fang presented a case today of they were going in one direction, you know, and then they found something that helped broaden the horizon of the drug development and the scientist and the dedicated developer will pick up on these little hints. One of my experiences, again, going back to the Keytruda experience is, you know, we were very attentive to what we hear from our advisors, patients’ experiences. And that helped to enhance our knowledge and help us, you know, go in directions that we weren't thinking of going. The MSI high, for example, that made sense biologically, right? I had the opportunity in January 2013, to visit the Hopkins team. And I was convinced that there would be an opportunity to test Keytruda on that population. And, you know, it turned out that that was a goldmine. Again, you have to be attentive, because, you know, MSI high may not be a big target at the point. But it became a bigger target when you had all the MSI high in one group, you know, agnostic to histology. So those are the things that the investigator or the developer has to be very attentive to emerging data, both internally and externally.
Dr Jon Wigginton: But Joe, you know, with that point. I mean, you know, that started with attentiveness to your point, to one patient. There was one patient that started that whole story.
Dr Joseph Eid: In your study. BMS study.
Dr Jon Wigginton: Good example!
Dr Joseph Eid: Yeah, so that's very true, Jon, and then the conviction that the Hopkins group had about the science, it made it work. But that's the little hints, the little hints.
Dr Jin Li: After you confirmed that for subgroup patients, they can definitely get the benefit, after that, the one arm study, do you think you have to still compare with standard treatment to confirm that?
Dr Joseph Eid: I mean, the standard in regulatory is the answer is yes. In some places, like with FDA, you get accelerated approval on the basis of a non-randomised data but they still would want you to confirm that. Now in some diseases where it's very rare, the standard of care is unethical, etc., they may give you a pass, but, in general, there's always an expectation that you come back with a confirmatory trial.
Dr Jin Li: Yeah, that's the question. (INDISTINCT) Dr Fang, because he introduced us to the RC-48 strategy for urothelial cancer who expressed the HER2 but, actually, those kinds of patients is very, very rare. So if the FDA or CD asked you to do the confirmatory comparative trial, where can you get so many patients for your confirming trial?
Dr Jianmin Fang: Thank you, Professor Li Jin. So, actually, just as I showed in the slide, HER2 expression in the UC is a traditional HER2 definition, about maybe 15% or 10 to 15%. But now we actually expanded the HER2 expression definition from conventional to include not just a IHC one plus, also two plus was (efficient) (99:14) negative. So that's what dramatically increase the patient number. So it's about 50% to 60% of the entire UC population. So for the patient size (INDISTINCT) (99:26 - 99:29) to bring this medicine to the patient. And the secondary (intel) (99:35) for how to do confirm total yes, this was - approval was conditional but we (INDISTINCT) (99:41) discuss with the CD and actually, this PD-1 combination study would serve as a complementary study. So that's what - actually this was included from the IHC one plus through three plus without (INDISTINCT) (99:59) So it would be easy to enrol the patient.
Dr Jin Li: Okay, so I hope that you will succeed.
Dr Jianmin Fang: Yeah, thank you! Based on our Phase II, the data now it looks – overall response is very high and we (INDISTINCT) (100:19) good PFS. It’s too early but PS is really good so we hope that will work well and (INDISTINCT) (100:32) there’s (data presented) (100:33) to update all these trials.
Dr Jin Li: Yeah, although you just expressed the strategy how you (INDISTINCT) (100:44) but actually for the market. I mean, for the market in the future after you got the approval it’s very rare for the patients, so the market is very limited. So your investor should have very strong heart in the future, I think.
Dr Jianmin Fang: Yeah, so that's why, of course, besides this UC we also have other Program B, C, gastric cancer (INDISTINCT) (101:12), hopefully other things, but each indication is different and so we need to test. And also, we actually will move for the early line. So that's why we’re interested in the first line therapy for UC, also even a new adjuvant, so that's maybe we’ll get more patients. That will be the direction we will trial.
Dr Jin Li: Oh, by the way, I have a question for Jon. Do we have the experience in the previous study the investor or the sponsor say, “Okay, we just want to start the trial.” But actually, after you analyse the data, you will suggest to do further to successfully put (it that the approved) (101:59)? Do we have that experience?
Dr Jon Wigginton: Let me reflect back. I think that question, if I can, is from prior experience, an example where a trial was overtly negative, right, but there were subsets right, that - So I don't think I have examples I can point to from my personal journey but I certainly, you know, I think we're all aware, right? I mean, I think there are examples where that has happened, right, where people have, again, back to Joe's point, done their diligence, either at the patient level, or at the subset level, looking for a hypothesis that supports a patient selection strategy. I mean, that's why we all do careful, diligent translational medicine in early studies, right, to help inform those hypotheses that emerge in the early studies.
Dr Jin Li: So when you decide go or no, what’s the - the factor is very important? I mean, when you make the decision to go or no go what aspect do you consider most?
Dr Jon Wigginton: Yeah, well, you know, there's a temptation sometimes for - and this is where, you know, I think we've heard some really important points over the course of the speakers. Dr Sharma talked about writing high-quality protocols, right, where we have clear goals and clear objectives. And I think I heard over the course of the discussion, Dr Yan also talk about some adaptive strategies, if I recall correctly. I sort of like it both ways, which is I think we need to write very good trials, right, that are very well written but that we need to leave ourselves the flexibility to be drug developers and not write things in a way that's so rigid that we can't use our judgment. And coming back to your question directly, I think it's really a matrix of thing. Sure, there's a response rate cut-off that we ought to use, if that's appropriate for the indication we're thinking about, but you know, if we're looking for 25% response rate and we get 24 and everybody else had, you know, 29% shrinkage of their tumour with bulky disease, you know, that's meaningful too so my point is it's a matrix of considerations - response rate, response quality, and certainly safety and tolerability as well. So it's really a matrix of things.
Dr Jin Li: Yeah, I know that 24% or 25% off the response rate, it's a quite high, but actually sometimes, as you know, (INDISTINCT) (104:48). So it's only one, just one, so give a talk just to our audience, how do you think when the efficacy is only in one or two patients how can you develop that drug? Your suggestion?
Dr Jon Wigginton: Yeah, well, so I'm going to give you two sides of the answer, if you will. One, I would say is, look, there does come a time when we're developing trials, despite our belief in the science and the biology, that the clinical data is just negative. and we need to accept that because back to the issue of resource use and trials in patients, exposing patients, sometimes we just need to say no. That said, you know, to me, certainly the colorectal cancer example that Joe highlighted is a great example, where an N of 1 could be transformative. But there, I would say, there was a very clear mechanistic story, a very clear scientific rationale, and a line of sight for clinical development. And when all those things line up, then that's the time I think you need to double down and dig harder, even if it is a patient or two, but sometimes you also just need to say no, and stop.
Dr Jin Li: Yeah, thank you.
Dr Li Yan: We’re getting questions from the audience or from online. So there's one question - does a PD cohort, I guess a pharmacodynamic cohort, and the biomarker data facilitate go/no go decision apart from patient positive outcomes? Maybe we could ask Dr Vivian Gu to answer this question.
Dr Vivian Gu: Yeah, I think so. And because, you know, in the past, we always decided (INDISTINCT) (106:47) the dose mainly based on the safety, because at that time, most drugs are chemo drugs. But nowadays, actually, most drugs are the targets therapy, (and I was happy) (106:59). We will make the (INDISTINCT) (107:01) decision. It's an overall assessment. I believe Dr Yan also mentioned during the presentation this morning, that it's the overall assessment. We will make the decision based on the overall efficacy data, safety data, and the PK PD data. So even sometimes, we will consider the late toxicity when we make the decision. So that definitely will be included into the consideration. And we can see the pharmacology discipline play more and more important role in the drug development area. That's all. Thank you.
Dr Li Yan: Okay, great, thank you so much. Dr Gu, you answer the recommended Phase II dose question. Do you want to comment further about the pharmacodynamic and biomarker question? So I guess if you don't see patient responses, for example, but you do see clear modulation of a PD target based on biomarker results, what would you do in such a situation? Are you going to advance the drug or are you going to go back to the lab and figure out what will be - maybe the combination will be the best way to move forward?
Dr Vivian Gu: Oh, okay. I think sometimes that's a complicated decision, because we know that the preclinical data or the lab test in vivo and in vitro may not always translate to the clinical response. So if we didn't see the signal, like in the clinical data, probably we can further investigate. I mean, to have more patients in the clinical or we can also go back to the preclinical. But I would also like to hear other panellists to comment for this question, if any.
Dr Jin Li: Yeah, actually, sometimes the clinical study just go (too rush) (109:03), so there's no very efficient data from preclinical study. So they just took clinical study, but actually find out the efficacy, the result is not so good. And also recently I have experience for a study - I just suggest them to go back to PDX model, do further, so try to find out which the biomarker could predict the efficacy. So if you pick the patients for all-comer, definitely you cannot succeed. So I think in Chinese way (SPEAKS CHINESE) (109:47). I don't know how to translate into English. So it just means do your preparation well, that will make you work efficiently, right? I don't know.
Dr Vivian Gu: Very good comment. Thank you.
Dr Jin Li: So I just asked them to go back to the PDX model to try and find out very good biomarker. So that's in the future. I mean, don't just go to the clinical study too rush, I mean.
Dr Jon Wigginton: Yeah. You know, I think you raise an important point, actually. I think that the challenge - fortunately, even biotech companies, I think, many do a pretty good job at translational medicine in a thoughtful way. But the reality is smaller companies are under even more pressure to execute quickly to keep up with the larger companies and have less resources. So it's a vicious cycle sometimes, but you raise important points.
Dr Jin Li: Yeah, for (INDISTINCT) (110:56) study you don't have to do a lot of preclinical study because you have the way - you know how to go to their different target. But for the brand-new target, you have to do the preclinical study very well. Yeah. Thank you. Yeah. Totally agree.
Dr Li Yan: That’s great. Thank you. Dr Jianmin Fang, there’s a question to you.
Translator: This question is for Dr Fang. Based on NPA’s guideline on oncology drugs, in the future, for the same class ADC drug, shall we show a stronger efficacy than DS-8201 in order to get approved?
Dr Jianmin Fang; So, just for my personal opinion, I think that depends on – you’re talking about which indication and which line. If in this particular indication and the same line then HER2 or any other ADC has become standard therapy recognised by CD or our medical community as standard therapy then probably yes, you have to, but if it’s different indication, different cancer type or different line then you have options. For example, in HER2 now it’s probably, I don’t know, (the BOA) (112:34), in China it’s the third line or second line of the BC, then, if your drug happened to be the same line then you have to think about the trial. Of course, you have to discuss (INDISTINCT) (112:48) what is the control, right? But it’s possible that they will ask you (INDISTINCT) (112:54). But if you have a first line or the line after this drug, then you don’t have (INDISTINCT) (113:02) control. If this is different, for example, UC, actually HER2 is not as important as in the BC (113:10) then it’s a different story. So it depends on your indication and line of therapy, so that’s what you depend on in the individual situation. That’s my answer.
Dr Li Yan: Right. So still just based on the standard of care, the current standard of care, I think that's really important. It really fits with the current guideline policy in the US and in China. So this is really great. Professor Li, maybe we ought to turn you to summarise today's webinar before we close it.
Dr Jin Li: I think that you are the organiser and also moderator. I just invite you to give the brief comments for this.
Dr Li Yan: Happy to do that. Thank you. Thank you, Professor Li, again, for your support, and kicking off the webinar. Again, I want to thank all the speakers and the panellists for a very informative, interactive and productive discussion. And really, specially my special thanks to dear friends on the East Coast. I know it's pretty late. It's almost 11pm at night for your time. So thank you for staying up. And also, I’d like to thank all the people joining from Asia Pacific as well as in China. It's your Saturday, and really appreciate it. We had a great discussion. Really just to summarise, early go/no go decision is really critical for drug development, especially in Phase I to Phase II, as well as, of course, from Phase II to Phase III but in oncology, we tend to try to move things as fast as we can, faster than other therapeutic areas. Sometimes we rely on our judgments based on limited number of patient data. Sometimes we are very optimistic, based our decisions, PD and a biomarker result. So I think based on today's discussion, unless you see really unambiguous data, such as HER2 ADC and RAS + BRAF, a combination, as well as anti PD-1 in early phase development, unless you see those really clear results, I think, you know, everybody needs to be really calm down and be careful in terms of jumping to fast. As Dr Li Jin said (SPEAKS CHINESE) (115:46). So you need to really work thoroughly before you make a decision as jumping into the larger, bigger and the longer trials because they're very, very big investment in the funding and it's going to tie up a lot of physicians in the clinical trial, and especially patient resources. So, this is really the take-home message that I think we want to share with our audience.
So let's thank the sponsor, Novotech, again, to make this webinar possible. And I also would like to advertise our next two seminars. Actually, in May, in the month of May, we had four seminars (of a pre ASCO seminar) (116:31) so we still have two really outstanding seminars to be chaired by Professor (INDISTINCT) (116:37) and Professor (INDISTINCT) (116:40) in the following next two Saturdays.
And most importantly, very exciting, I hope all the Chinese attendees could be there next year but this year we’re going to have an in-person gathering, China dinner at ASCO Chicago. I heard that the food menu is fantastic so Jon and Joe, if you will be in Chicago, please do take your team there and meanwhile eChina Health will be there to welcome you in person. So this is a great event, to see that we’re back to normal in the US. With that, I want to just close this session, thank all the audience and speakers and the panellists. Thank you very much. Have a great weekend. Take care.