Immutep (ASX:IMM) SITC 2021 Results – Management Update


Immutep Limited (ASX:IMM) Executive Director and CEO Marc Voigt, Chief Scientific and Medical Officer Professor Frédéric Triebel, and Vice President Strategic Development Christian Mueller present on Immutep’s latest clinical results.

Clive Tomkins: Good morning everyone, and thank you for joining us for the Immutep call to discuss the company’s final clinical results from its phase 2B AIPAC study of its lead product candidate, eftilagimod alpha in metastatic breast cancer, Clive Tomkins your moderator here in Sydney. The format today will be a presentation delivered by Immutep, at the end of which we will open the call for analyst questions. I’ll also relay any other questions that we have received via email prior to this call. I’d now like to hand over to Marc Voigt, Immutep CEO, to start the presentation. Marc over to you.

Marc Voigt: Yeah, thank you, Clive. And here you can see my forward looking statement as usual and of course welcome everyone. And thank you for joining us on the call today. And joining from the Immutep management team is Professor Frederic Triebel, our chief medical and scientific officer, as well as Christian Mueller, our vice president strategic development. So today we are pleased to walk you through the positive final data from our AIPAC trial of eftilagimod in metastatic breast cancer in combination with paclitaxel as well as a few other updates. And we have many ongoing clinical trials of efti, as you know, but this data was just presented at SITC last week. So obviously this will be the key focus of this call. Very briefly here, quick snapshot of the company. I think you are aware of what we are doing, aware of our partners, so we shouldn’t spend too much time.

However, worth mentioning is of course our position around LAG-3, Lymphocyte-Activation Gene 3. And before we go to the results, just a quick overview of the LAG-3 therapeutic landscape, where we saw many exciting changes during 2021 and where I expect more changes also in 2022. So you’ll see here, our programs symbolized by the four green arrows. At the bottom two unique programs in autoimmune diseases. Unique is also eftilagimod on top of this page, the only MHC class two agonist globally. And then you’ll see the blocking antibody space or antagonist antibody space in oncology with our license with Novartis. And of course, I believe most of you are aware of Bristol Myers Squibb data around Relatimab released at ASCO this year, and also that they filed for marketing authorization in Europe and have the meeting with the FDA in March next year.

So the commercial journey of LAG-3 has really just begun and I expect more data to come in the next year. And we will certainly be part of that. So a few other key facts around eftilagimod, as you can see here on the left hand side, we have a different mechanism of action compared to the immune checkpoint inhibitors symbolized here on the right hand side. So we provide a broader push to the immune system. We address MHC class two on the antigen presenting cells, and then leading via the physiological way to more cytotoxic CD8 T-cells, CD4, NK cells, activated dendritic cells, and so on. For us, we are not blocking LAG-3 or PD-1 or CTLA-4 on the T-cell. So it’s a unique mode of action, a broader immune response, I believe that’s actually quite important. The pipeline chart we have been always having the goal of multiple different clinical trials with hopefully good readouts.

This is never guaranteed. And it’s maybe worth mentioning that for instance on TACTI-002 for the time being all different indication have been a success, obviously also in metastatic breast cancer. And we are moving forward with the phase three in metastatic breast cancer. We will come to that later on. So you see overall, the horizontal and vertical rollout of the eftilagimod case. Here on slide nine, a few other updates. We, I believe made a good operational progress for TACTI-002. The expansion stage of part two is actually expected to be fully recruited ahead of time. We are backed by our principal investigators and this is reflected, obviously in the recruitment number. We have, as we speak, more than a hundred patients enrolled, and we are fast approaching the 110 patients in total. Dependent on the cutoff, we may even have one or two patients more. So that’s a great success, especially also considering the COVID environment, which is making it definitely not easier to recruit patients.

Also TACTI-003 has received full approvals for trials in five of the eight countries. And this process has been very smooth with no issues with the feedback from the relevant authorities. The INSIGHT platform trial, it’s an investigator initiated trial platform, is continuing to advance. So INSIGHT-003 has recruited the first patients in the triple combination, so anti PD-1, plus eftilagimod, plus hemotherapy. Obviously nothing to report so far, which is good because we are looking here for most at safety. So it’s actually a good sign that you haven’t seen here any update from us. So we can all be optimistic. INSIGHT-005 preparations are still ongoing. And the performance of Bintrafusp alfa is under assessment by Merck. We saw there actually a few setbacks for Bintrafusp and we are in close contact with Merck KGA and will carefully assess potential next steps in the light of the data from Bintrafusp. So let’s see what happens there next.

And our planned phase three trial, AIPAC-003, as it will be called, has been receiving a positive scientific advice from the EMA. So they had to look at the breast cancer program, obviously, and the discussions with the FDA are ongoing. And of course we will update the market once we achieve the next steps.

So here now we come to AIPAC where we presented final overall survival results at SITC. And this is obviously a process which takes long. The trial started some years ago and we had the data cut off in May. Everything run by external companies and data cleaning needs to happen. And once we received the data, so a few months later, we immediately applied for late-breaking at SITC, got that so that we are in a position to present the data. But I would like to hand over to Frederic to go through some aspects here.

Professor Frédéric Triebel: Thank you, Marc. On slide 11 now. Breast cancer is the most frequently diagnosed cancer in women, with more than 2 million diagnoses per year worldwide. 70% of these are hormone receptor positive HER two negative carcinomas. And this major subgroup was chosen based on regulatory interaction with the EMA to first test the therapeutic potential of efti in breast cancer. Note that, at the moment, there is no other active immunotherapy as advanced as efti in terms of product development in this major subgroup. The flow diagram on this slide, maps out the characterization and treatment journey of these patients. What this highlights is that while many treatment landscapes have moved forward in a pre-chemo setting, like the CDK4/6 inhibitor, almost all metastatic breast cancer patients end up on chemo like weekly Paclitaxel as a standard of care.

A lack of innovation for these patients has left them with high unmet medical need. Turning to slide 12 now. Our AIPAC trial was a multicentre placebo control, double-blind phase 2B trial. It involved 227 hormone receptor positive HER-2 negative metastatic breast cancer patients. They were randomized one to one, to a chemo immunotherapy arm treated with efti plus Paclitaxel or to a comparator arm, which involve placebo plus Paclitaxel. And patients receive a weekly Paclitaxel at day one, eight and 15 with either efti or placebo injected subcutaneously on days two and 16 of each four week cycle. This was repeated for six cycles. Thereafter, patients pass over to the maintenance phase with efti alone. So this trial was a hypothesis generating phase 2B study, meaning that it aimed to explore a set of data to find, for instance, very specified subgroup with a better clinical outcome. Then validation of a given hypothesis can be prospectively tested in a subsequent phase three study.

The trial completed patient’s recruitment in 2019. And the primary analysis of PFS was announced in March, 2020. We then reported the first overall survival in December, 2020, when the trial had reached roughly 60% of events. Now moving to slide 13, here we provide detail about the baseline characteristics of the patients as they enter the trial. While the trial is randomized and the two arms should then be balanced, it was still important to make sure that important baseline characteristics were similar between the two groups. This was achieved. AIPAC enrolled patients with very late stage disease, 92% at progress to visceral disease and 69% at elevated LDH. Indeed, the participants were heavily pre-treated having received a median of two other types of therapy for their metastatic disease before joining the trial. For instance, 84% had become endocrine therapy resistant and 44% received CDK4/6 treatment prior to entering trial. The median age of patients was 60 years. About two thirds of the patients were younger than 65 years. Overall, the patients groups were well balanced and representative of the wider patients landscape.

Turning now to slide 14. This slides gives you an overview of the good safety profile reported from the trial. Overall, the combination of efti and Paclitaxel chemotherapy was safe and well tolerated, further building upon efti’s strong safety profile, as these results were obtained in a randomized setting. Looking at the graph, the green bars indicate an adverse event occurring in the efti group, whereas yellow or an orange occur in the placebo group. It shows all adverse events occurring in at least 15% of the patients in any arm, regardless of relationship. The most common efti-related adverse events were local injection site reactions, which were reported in 75 patients, representing 65% of the efti group. These were mostly local transient erythema due to the local hyper activation of dendritic cells in the skin where there is a high efti concentration for a few hours after injection.

The number of patients discontinuing due to adverse reactions was low and similar between the two groups, about 5% or 6%. There was no worthy disbalance between the arms except for the injection site reactions. We’re then very pleased with the safety profile, which plays an important role in any drug approval process by regulatory agencies who are evaluating risk benefit profile. Having a very well tolerated drug obviously helps a lot and we believe the AIPAC trial provided a very strong basis for a good safety profile. I will now ask Christian to go into further detail for the AIPAC efficacy results and take you through the new TACTI-002 part C results. Christian.

Christian Mueller: Frederic. On slide 15, we will now look at the efficacy results from the overall trial population of 227 patients. In this overall population, we are pleased to report final OS data based on 72.5% of events, which means the final analysis was done with very mature data. Patients in the efti group had a median OS of 20.4 month compared to 17.5 month for the placebo group. This reflects a median survival benefit of 2.9 month, which is an improvement compared to the 2.7 we reported in December last year. The hazard ratio for the trend is 0.88. These results can be seen in the left hand graph where the efti group in red has a higher proportion of patients who are alive compared to the placebo group, which is presented by the black line. Where important for the interpretation of overall survival data, is to compare post study treatments patients received.

Roughly 90% of the patients in AIPAC had at least one more subsequent therapy in both groups and predominantly they received chemotherapy. Looking at the right-hand graph, these are the results of the quality of life survey conducted with the patients and quality of life is a secondary endpoint of the study and will become very important once you talk reimbursement with the payers in the different countries. In the graph, a quality of life preservation was observed at six months in the efti group. And this compares quite favourably to the placebo group where you see a significant deterioration in these measures after six months. Turning to slide 16 now. Immune monitoring of fresh blood samples was conducted in 70 patients across the entire trial population, randomly selected based on specific sites participating in this exercise. These samples were analysed before beginning trial treatment, as well as twice during this study at three and six months.

Very important to note is that these samples were always taken two weeks after the previous efti injection, and immediately prior to the next one. Which means in turns it, it will only show you the minimal residual effect which is still present at the time the patient received the next efti injection.

On the left hand graph, you can see that the numbers of cytotoxic CD8 T-cells increase in the efti group reaching a statistical significant difference compared to the placebo group at three and six months, while the difference at baseline were nonsignificant. More important on the right hand graph, you can see that the number of CD8 T-cells under study co-related significantly with improved overall survival in the efti arm, as indicated by the green trend line. More importantly, there’s no such trend in the placebo group as indicated by the yellow line. That these findings are statistically significant, is a very strong signal as the group size with around 35 patients per group is not very big. It shows a proof of principle and a proof of concept in the randomized double-blinded setting, which I think is very important to keep in mind as neither the assessor of the blood samples, nor the PI, nor the patient knew if they received placebo or efti. Let’s discuss for a second how this data should be interpreted.

I think this is quite an important question. So while OS difference did not reach statistical significance in the overall population, which by the way the study was never powered for, they show a clear trend in favor of efti. And the trial also showed that efti helps to maintain quality of life. Effective, which is very important during potential reimbursement discussions. More importantly, the trial was first able to link efti’s mechanism of action, increasing the number of CD8 T-cells to improved overall survival in a double-blinded setting. All in all, we believe the trial gives a very strong signal that efti is active in this patient population in this combination and a development in the phase three is warranted. And this view is shared by investigators as recently published at SITC and the EMA where Immutep received a positive answer during our scientific advice procedure. The main question is now how to ensure that the phase three trial will be successful from a statistical point of view. We believe that AIPAC as an outcome of phase two study will be a good guide here as we had the total patient population to look at. Let’s have a look into different pre-specified subgroups and then let me share a few thoughts on the potential phase three design.

Moving on to slide number 17. Before we start to dive deeper into the data, I would like to explain what we have done. Based on literature research and available data at the time, certain subgroups have been specified prior to unblinding of the trial, which means neither the statistician nor us knew the data at the time and there was no bias introduced. The idea is to identify two types of factors. The first type of factor is the poor prognostic marker. They describe potential characteristics where the efficacy is worse irrespective of the therapy given. For instance, you could assume that liver and brain metastasis in general have a detrimental effect on overall survival. And you use those poor prognostic markers usually to identify potential stratification factors for the phase three.

And then I think most of us are more interested in potential predictive biomarkers. And they potentially identify a patient population where the drug, in this case, efti, works better. Well-known predictive markers are PD-L1 or HER2/neu, where these markers were identified in earlier studies and then confirmed in a phase three. As we all know, they’re now widely used to drive treatment decisions and achieve a better outcome for the patients.

Let’s give it a go with the poor prognostic marker. In an exploratory multi-variant analysis, it indicates that among others, prior CDK4/6 treatment increases significantly the risk of death irrespective of the therapy by round about 37%. If you then compare the median of the two treatment groups, we observed a drop of about five months in the placebo group while the median overall survival in the efti group remained more or less in line with the overall median of 20.4 months. As there’s not much data out there in the post-CDK4/6 era and how chemotherapy works, we feel it’s quite important finding to design and statistically power the next study.

As shown on Frederic’s slide, the usage of CDK4/6 increased in the recent years and especially also compared to the time AIPAC was recruited. In other words, prior CDK4/6 seemed to decrease the efficacy of weekly paclitaxel, which is not the case if you add efti to paclitaxel. And as we all know, prior CDK4/6 will now be standard in the majority of patients.

After we have discussed the main poor prognostic markers, let’s have a look into potential predictive markers. As outlined earlier, the subgroups have been pre-specified prior to unblinding to avoid any bias. The subgroups shown below are younger patients under age of 65 years, patients with low monocytes and those with luminal B tumors and also had not had prior adjuvant taxane chemotherapy before trial start. We would like to take a few moments to drill down further into the encouraging subgroup data on the next few slides. Before we do so, I would like to highlight that the effect seen here for OS in all subgroups is clinically meaningful, meaning if confirmed in a phase three, it would help tremendously during reimbursement discussions.

Looking at the subgroup of patients under 65 years, these patients have a clinically meaningful and significant improvement in overall survival. Also other efficacy parameters like overall response rate and PFS show the right trend. The graph on the left hand side shows that younger patients reported a median survival benefit of 7.5 month. This benefit was highly significant with a p-value of all 0.017 one-sided and a hazard ratio of all 0.66, meaning a risk reduction of death by 34%. Younger patients represent round about two-third of patients in the efti group and making them a sizeable proportion in the whole study population.

In the right hand graph, you can see that age had almost a linear effect on the hazard ratio for OS, meaning that the younger the patient, the higher the benefit they had from addition of efti to weekly paclitaxel. What could be a potential explanation? Why efti works better on younger patients? As efti works on immune cells, an effectively working immune system is very important. A process known as immunosenescence reduces the activity of the immune system in general and direction to activate in elderly patients. I think everybody’s aware nowadays this can well be observed with the COVID vaccination where you can see that it does not work as well in elderly than it does in younger patients. Hence, younger patients usually with a better functional immune system can be expected to react better to immune therapies like efti.

Moving to slide 19, here we look more closely at a low monocyte subgroup. For this subgroup, we have observed a clinically meaningful absolute and relative improvement in all efficacy parameters, overall survival, progression-free survival and overall response rate. From the graph on the left hand side, we can determine that patients starting with a low monocyte count saw an additional 19.6 month in median overall survival in the efti group indicated by the red line. The hazard ratio is 0.44 and highly statistically significant. While this group presents about 22% of the patients in the efti group, it is still remarkable improvement for a significant portion of breast cancer patients.

On slide 20 now, a statistically significant and clinically meaningful improvement in OS was observed in patients with luminal B tumors. Also other efficacy parameters like overall response rate and PFS show the right trend. This group represents about 50% of the patients. Also, here, you could ask the question, what could be a potential reason for the observed findings? And luminal B is known to be a tumor which proliferates a bit quicker and expresses more neoantigens. Efti activates antigen presenting cells to present more tumor antigens. Hence, if there are more tumor neoantigens available, it theoretically, should be better for the efficacy of efti. As you can see, all these groups are somehow linked to either the immune system or the tumor.

Let’s have a look into slide number 21. These are the subgroup analysis results for those patients who had not received prior taxanes as adjuvant therapy before the trial. Here, we report a clinically meaningful improvement of OS of about 4.8 month. And the differences were found significant in a post hoc, multi-variate model. That completes the results from our phase 2b study, AIPAC. And we would now like to take a moment to talk through some aspects of our planned phase three trial, which is still subject to regulatory interactions as we heard.

For this trial, we expect to evaluate efti in combination with weekly paclitaxel in metastatic cancer patients taking into account the learnings from AIPAC and what has been discussed today. The primary endpoint is likely to be overall survival, which is a preferred endpoint for phase three trials by regulatory agencies in such a patient population. And we also feel it’s the preferred endpoint for immune therapies like efti. The chemotherapy agent will be paclitaxel as per the AIPAC trial, but we will allow continuation beyond six cycles to accommodate for US standards.

The team will ensure that the patient population’s clearly defined and statistical readouts structured in a way to increase likelihood of trial success. Just to assure everyone, we will not simply repeat AIPAC here. As we are in ongoing regulatory interactions, we cannot be any more precise at this time, but we expect to be able to elaborate on the trial design as regulatory engagement progresses. So that concludes the section of the webcast devoted to the AIPAC results. Now let’s turn to the recent results of our phase two TACTI-02 trial, which evaluates efti in combination with pembrolizumab.

I just briefly outlined here on that slide the design of our TACTI-02 study, which is being conducted in collaboration with Merck, Sharp & Dohme, known as MSD. TACTI-02 was testing the combination of efti with MSD’s drug, KEYTRUDA, also known as pembrolizumab, in up to 183 patients around the world. The trial is divided into three different settings. Part A comprises treatment-naive patients with first-line non-small cell lung cancer. Part B, a second-line non-small cell lung cancer patients who are clearly refracted, who have progressed on prior PD-1 or PD-L1 containing therapy. And part C, the part we talk today about, second-line metastatic head and neck cancer after failure of platinum-based therapy.

The primary endpoint of the study is overall response rate measured by iRecist. TACTI-02 is an open-label study with a Simon’s two stage design, which means we have been able to expand the cohorts as positive results were observed and reported. Very important to note is that TACTI-02 is a true PD-L1 all-comer trial and does not pre-select for PD-L1 high expressors as in these PD-1 therapies are likely to work anyway. From our point of view, the main challenge is to increase the proportion of patients responding as usually responses are very long-lasting if you have a response with an anti-PD-1 therapy.

We’re in general very happy how the trial turned out in the end. We have been able to extend all three cohorts and for part A, even beyond this, which I believe you couldn’t expect when you start such a trial. So we have seen signals in three different indications. And as Marc will probably explain later, we are ready to give updates there in the course of 2022. The recruitment has progressed very well with all original stages complete. And after good results as explained, we expanded part A to enrol additional 74 patients. And as reported here today, we are confident to close recruitment ahead of time. So all in all, this trial for us as a company is quite a success, and I’m more than happy to guide you through the latest updates of part C.

Looking at slide 25. Today, we are presenting more mature interim results for part C, the patients with second-line head and neck squamous cell carcinoma. And these results were also reported at SITC just a few days ago. As outlined in earlier webcast, the patient population was as expected, overall response rate of about 30% in the intend-to-treat population and overall response rate of 35% in the evaluable patients have not changed since the last update at ASCO 2021 as shown at the table on the left. We also discussed the quality of the response expected. And you can follow the journey here of each patient in the spider plot, confirming that responses are long-lasting. The waterfall plot on the right shows you the deepness of the responses, and the blue bars are indicate complete responses.

What has not been reported in detail is the split by CPS for overall response rate, progression-free survival and overall survival rates, which is displayed in the table on the right. First, we can note that a percent of patients within the three different CPS scores, so meaning CPS negative, CPS 1 to 19, or having a CPS score of larger or equal to 20 is as expected for head and neck cancer, making the trial a true all-comer trial without enrichment of high PD-L1 expressors.

Let’s focus on the patients with a CPS score for at least 1. Here the overall response rate is 41% and the 12 months overall survival rate is 48%. This increases even further to 64% overall response rate and 64% being alive for 12 months in patients with a CPS score of 20 or greater. As outlined earlier, we are very pleased to report these consistent and more mature data for this cohort as TACTI-03 in first-line just started.

Turning to slide 26 now. The median duration of response cannot be estimated yet as shown in the Kaplan-Meier plot on the left, where the duration of response hasn’t hit the 50% mark. Just to reassure everyone, this is a good sign because the responses are still ongoing. And for all the responses still ongoing, it’s ongoing for more than nine months. The right side of the slide compares the results to historical controls, in this case to KEYNOTE-40. As always, historical comparisons should be interpreted with caution. But having said that, the overall response rate in the intent-to-treat setting from efti plus pembro is more than double the overall response rate in all PD-L1 subgroups and also higher in the overall population. You can see this in the bar graph on the right side.

Not shown on the slide but can be compared, the PFS and OS rates at six and 12 months, respectively, are higher in the efti group and all PD-L1 subgroups, as well as in the overall population if you compare this data to KEYNOTE-40. So all in all, the signal we have seen in this part of the trial, we believe is really, really strong if you compare that to historical data with pembrolizumab alone in a similar patient population.

Let’s move to the next slide. Our results from the ongoing part C of TACTI-02 are very pleasing and contributed to the faster designation we were awarded from the US FDA for first-line head and neck of this particular combination. This is obviously commercially more relevant setting and will be explored in the TACTI-03 trial, which is also conducted in collaboration with MSD. Before we dive here more into the detail, let’s recap a few important aspects. Efti plus pembro is very well-tolerated with an excellent safety profile. It doubled overall response rate compared to historical controls in second-line head and neck. And it seems that the excellent duration of response you have with pembrolizumab is maintained with the combination of efti.

Having said that, let’s start here with having a look into the treatment landscape of first-line head and neck. In first-line head and neck existing therapies such as chemo plus rituximab, pembrolizumab monotherapy for patients with non-PD-L1 negative tumors or the combination of chemo plus pembrolizumab are achieving a median OS of approximately 11, 12 or 13 months, respectively, as shown on the left figure, meaning that pembro alone is not inferior to chemo and chemo plus pembro is superior to chemo alone. The addition of chemo to pembro increases considerably the overall response rate, but as shown on the right figure, it decreases the duration of response dramatically. The duration drops from 23 months for pembro alone in the first-line setting to around about seven months for the chemo plus pembro combination.

In this slide, the combination of efti plus pembrolizumab compares very favorably. We believe that efti plus pembro may increase or we have the hypothesis that it increases the response rate and overall survival while at the same time, it does not compromise on safety or the duration of response leading to a more effective therapy in first-line head and neck. While we must always be careful with these comparison and predictions, we believe efti plus pembro could really fill a niche here of high unmet medical need. And having said that, I would like to hand back to Marc, who will provide you now an overview of the TACTI-03 trial.

Marc Voigt: Yeah. Thank you, Christian. And before we conclude this part of the webcast, I would just like to briefly provide a short overview regarding the trial design of TACTI-003 and then status update I believe I’ve been providing earlier on in terms of study startup phase, additional countries on board. And this is obviously ongoing. So TACTI-003 is a randomized study taking place at the clinical sites across Australia, the United States and Europe. The primary endpoint will be overall response rate. And overall, we will recruit approximately 154 patients in first line, head and neck, and the bulk will go to cohort A, and we have then an experimental satellite cohort B, which is obviously outside the label of Pembrolizumab or Keytruda.

So this brings us to the end of the trial results and updates, and just a few words of what to expect. Obviously in the remainder of 2021, we hope to provide additional updates. We said earlier on that we are about to fully recruit TACTI-002. And given that it’s a total 183 patients, I believe it’s quite important that we were able to make their stop. And that we then will be likely in a position to talk about non-small cell lung cancer in first half of next year.

So we will provide updates from TACTI-002. We have ongoing regulatory interaction, the phase three preparations and metastatic breast cancer are ongoing. We will be in a position to talk more precisely about the trial design in first half of next year, pending regulatory interactions. We hope of course, also from data from INSIGHT-003, TACTI-003 updates should be provided. So there will be a number of things to discuss. We should not forget about IMP761. And of course also our partner programs, for instance, the efforts from our Chinese partner for efti in China, with the study start expected at metastatic breast cancer and the remainder of this year. Also, by the way, looking at an improved schedule, so allowing treatment post six months as we would do also in a phase three, I think this is one of the important learnings from AIPAC. And of course, we have other partners which should provide updates as well.

And this all with the background of having yeah, validated LAG-3 imaging class two interaction, and also that commercial journey of LAG-3 is just beginning. So I believe that’s a pretty good outlook, and we can have high expectations for 22.

On slide 31, here we are. I think you are all aware that we are the only LAG-3 pure-play, that we have more programs than anyone else, a number of established collaborations, but I would not like to reiterate everything in order to leave some room for Q and A, of course.

So this brings us to the end of our presentation and I would like to thank Frederic and Christian for their contributions. And I will now pass back to Clive for questions from the analysts who have kindly joined us from the US, Australia, and Europe. Thank you so much.

Clive Tompkins: Okay. Thanks, Marc. All right. In terms of the questions, just to remind everyone, there’s a raised hand icon at the bottom right hand of your screen, if you click on that, I’ll get to each of you in turn. Okay. Our first question today comes from Shane Storey at Wilson’s. Shane, please go ahead.

Shane Storey: Thanks. [inaudible 00:41:40]… and thanks for the presentation. Look, my first question really is around the age as your pre-specified subgroup, please. I just wanted to make sure that I have the distinction right, that even that the 60 was really immune costs rather than performance status or the effect that might have on the disease in those patients.

Marc Voigt: Yeah, the connection wasn’t best, at least on my end. But first of all, you are right. It has been a pre-specified subgroups, meaning of course, defined before unblinding the data. I think that’s an important point because you have the debate of course, of pre-specified subgroups versus post hoc. And this is pretty clear here, pre-specified, and regarding the medical impact of age versus performance status, which also could be correlated, maybe Christian, maybe if you could comment on that?

Christian Mueller: Thanks for the question. I think in terms of age and ECOG I wouldn’t see them strongly correlated. Usually obviously like if the patient is getting very old the ECOC may go down, but usually the ECOC is more driven by the disease. And we feel that age together with the mechanism of the efti and the immunosenescence we talked about today, that this is probably a little bit different pattern and more relevant for us than simply ECOG.

Shane Storey: Right. I’m just got one final follow up for you, Christian, actually. In your slide presentation, you showed that there was a linear relationship, I think, between hazard ratio and age and [inaudible 00:43:34] the hazard ratio, what that age cohort was. But my question is, is that a strong enough signal? Do you think to specifically exclude older patients in that they might not do so well on Efti?

Christian Mueller: The connection was quite bumpy and intermittent, but I think you’re asking if you would feel confident to exclude elderly patients above 65, right?

Shane Storey: That’s correct, yeah.

Christian Mueller: Yeah. So this linear relationship, in order to stick a bit to the terminology we have used today, this was also obviously done then post hoc. After we had the findings from the less than 65 years group, it was not pre-specified. And if you look at the groups are relatively small, it seems like a very linear trend. What makes sense from a theoretical point of view. And then it comes to, I think we heard it earlier about benefit risk, and this is something we obviously discussed with the regulatory agencies, how they see that. And this is about a safety profile, the potential detrimental effect you may have, or you may not have, which is not, this is just the point estimate. So you also have the 95% confidence intervals, which obviously for small groups are relatively large. So this will be discussed with the agencies, in short. Yeah, it depends largely on the benefit risk assessment we will make in the very end.

Shane Storey: Thanks everyone. Yeah. Sorry. Sorry about the audio. Thank you.

Clive Tompkins: Okay. Next question comes from Andrew Paine at CLSA. Andrew, go ahead please.

Andrew Paine: Yeah. Morning everyone. Thanks for taking my question and, and thanks for the presentation. Just following on from the, you know, the strong results that you saw in the AIPAC trials for subgroup populations, and you explain why this may be the case, but just thinking about other trials that you’re doing, would it be correct to think that we may see similar, stronger results in these subpopulations versus the overall population?

Marc Voigt: Not necessarily, I mean, there you would need to go case by case. So that means combination by combination, indication by indication. Obviously here in the chemo eftilagimod setting in metastatic breast cancer, this is what we see. And specifically strong benefit in patients below 65 years of age, which is some clinically established threshold.

But that doesn’t mean in turn that you would see the same. For instance, let’s say a non-small cell lung cancer. And also there we have of course, a biomarker, which is PDL1 relevant for Keytruda. It’s a combination in TACTI of Keytruda plus eftilagimod, so there might be other subgroups worth looking at. So for instance, low or no PDL1. So in other words, you would need to look here really case by case. And that’s the beauty of eftilagimod, that you have a variety of different settings, different indications you could look at, and you need to be very specific in your trial design, and also in the phase three, or any other subsequent study, let it be for the TACTI series, let it be for the AIPAC series, that you choose the right patient population. This is obviously true for cancer immunotherapy. When you look the way Keytruda is developed and some indications with PDL1 as a strong biomarker, non-small cell lung cancer, head and neck cancer, but does not seem to play a big role in melanoma, for instance.

Andrew Paine: That’s great. Thanks. And just looking at the TACTI-002 results for second line neck cancer, are you able to provide any update or around the median overall survival or median progression free survival versus the interim readout? Yeah, looking at the overall population?

Marc Voigt: Yeah, first of all, medically not a lot has changed. We saw obviously at 12 months, 48% of patients, the results have been confirming what we saw previously. We will have a look on median progression, median overall final look, let’s say on that next year. There is also, of course, and the different subgroups or different PDL1 strata one could look at, for instance.

So in terms of median progression-free survival in general, and to a certain extent, also median overall survival, one has to be a bit careful as this is a rapidly growing cancer, very aggressive. So obviously more than 50% of patients are rapid progressors, as you don’t have an overall response rate, which is above 50% there, you see usually limited results. This is why as a qualitative parameter, of course it’s still important, but the duration of response is so those patients who are responding and if you have more patients responding, which we saw with Eftilagimod, to see that the quality of those patients is okay, or actually superior, this is why we were reflecting here on the duration of response, but there will be a look on all final data sets next year.

Andrew Paine: That’s fantastic. Thanks very much. Yeah, congratulations on the results.

Marc Voigt: Thank you.

Clive Tompkins: Okay. Next question from Naureen Quibria at, oh, sorry, Naureen.

Naureen Quibria: Maxim.

Clive Tompkins: At Maxim. Thank you very much. Thank you.

Naureen Quibria: Sure. So thanks for the presentation and congrats on the data. So I guess I’ll start first with, and all my questions are on AIPAC. I’ll start with either Frederic or Christian. I was just wondering, have you looked at a granular level to see if there are any differences in the time to response between these subgroups in the AIPAC study? Like, do you see some responding faster than others?

Marc Voigt: Christian, would you like to take that one, or Frederic?

Christian Mueller: I can do it. Yeah. Obviously, time to response, duration of response. Yeah. These kind of parameters obviously we had a look into, and there’s no difference from top of my head between the two groups. Time to response is relatively similar in the two groups because it’s probably mostly driven by the chemotherapy.

Naureen Quibria: And do you see, what were the rates of discontinuations? I didn’t see that.

Christian Mueller: It’s on the slide Frederic discussed, on the safety side. And the discontinuation rates are pretty similar between the two groups. You have, usually, a couple of hypersensitivities due to Paclitaxel, and then usually dose reductions because of toxicity due to the chemotherapy. But that’s similar between the two groups.

Professor Frédéric Triebel: Yeah. It’s about 5% in the two groups.

Naureen Quibria: That’s helpful. Thank you. And just one more follow up for me. So Marc, perhaps you can help me with this. I recognize and fully appreciate that you don’t have an idea with regards to the phase three AIPAC design, but you know, given that you have multiple subgroups where you are seeing, you know, OS benefit, and that’s been extended, you’re seeing durability, you know, how do you actually approach regulatory authorities about a potential pivotal study on potentially evaluating multiple subgroups? How does that conversation go? You know, is there like precedence for multiple subgroups that you’re aware of?

Marc Voigt: Yeah. Thank you, Noreen. So first of all, of course we have a good idea and understanding, looking at the data, what we could implement. There are a variety of different sweet spots. We have not been saying that we would, for instance, go to all different, three different subgroups we are seeing here or looking at prior tech sayings. So the reason why we have not yet been precise in terms of the trial design, where we said, we are taking the learnings from AIPAC, for instance, scheduling, allow for longer treatment, but also of course, coming from an unselected, all common patient population, as you would normally do to go then for instance, to biomarker related or other parameters. The trial, which is more on target, as you saw with anti-PD1 therapies, for instance, looking at that time as PDL1 has not been established before is important.

We have the typical process of putting together, for instance, a briefing book. These are large documents where the company position is explained, which you are then discussing with a competent authority. And we would like to of course, bring EMA and FDA in line and to a common view. And this is about presenting, getting the feedback, incorporating that, and then come to a final trial design, as this takes time. And we don’t want to misguide the market so that we have a specific idea, and then in the light of any potential feedback from competent authorities, we need to revise it. So this is why we prefer also not to corner, let’s say, any competent authority with any specifics, as you wouldn’t do that and talk publicly about it. But once we have a good robust basis from the two competent authorities, which are most important for us, so FDA and EMA, we will of course, inform the market.

What we try to do is to design a clinical trial, which is designed to deliver a success based on the findings from AIPAC, of course, there’s always a risk involved. And to do that, of course also taking certain results from the subgroups and other parameters into account in order to make it a trial, which is on target. There are different ideas out there. Look at PD1 development, for instance, hierarchical testing. So there could be different design concepts. Those will be discussed under confidentiality with the competent authorities and then discussed, of course, publicly.

Naureen Quibria: Gotcha. Okay. Thank you. And congrats on the data again.

Marc Voigt: Thank you so much.

Clive Tompkins: Okay. Next question from Ahu Demir at Landenberg.

Ahu Demir: Thank you for taking my questions and congrats on the data. First of all, I think we are all eager to see more details on the AIPAC-003 trial, to follow up on Noreen’s comments. My first question will be on the three biomarkers that you disclosed data for. Are these biomarker populations mutually exclusive, or do you see any correlation between each other?

Marc Voigt: Yeah, so there’s an overlap. So for instance, it’s usually known that Luminal-B is more often in younger patients for instance. So the disease is typically a bit more aggressive in younger patients. There’s also of course, an overlap between low monocytes and younger patients. So there is an overlap of the different groups.

Ahu Demir: Okay. This is helpful. I was most impressed by the data on the low monocyte patient population, of course. 19 months of overall survival benefit is impressive. So could you maybe elaborate more on the patient demographics? You mentioned they are generally younger patients. I noticed 21% of the patients had low monocytes based on AIPAC-002 trial, does this represent maybe percentage rate for the general addressable marketing, the metastatic breast cancer patients with monocyte? How do you think of that? Have you done any research? Just curious if you could maybe comment on the separate, Marc?

Professor Frédéric Triebel: I can comment on this one, Marc?

Marc Voigt: Yeah, sure.

Professor Frédéric Triebel: Yeah. Low monocyte here, it’s not really low monocyte. It’s a lower quartile, meaning that about 22, 25% of patients are in the lower quarter in terms of monocyte counts. And why do they have not monocytopenia, but they are low on monocytes, well, we just don’t know. I mean, that reflects a defect in the innate immune system. And of course, efti can obviously come to the rescue and help these patients overcome this defect. So the low monocyte may be just one sign of a defective innate system and we’re looking for other biomarkers. I mean, of course, there could be some post-oc bio marker analysis. So we’re working on this to better understand what does that mean and how we could maybe define a better subset of patients to be enrolled in the phase three trial.

Ahu Demir Right. This is helpful. And also, 20%, is it a reliable, addressable market or is it still to be determined in the general breast cancer population?

Marc Voigt: I would say it’s an indication, but monocytes are not necessarily taken in breast cancer, so it’s like with PDL1, which has not been established prior to anti-PD1 therapy, so in metastatic breast cancer, this would need to be established, of course. It’s well known. It’s defined by effects, so it’s not rocket science, so to speak, to get monocytes, but for metastatic breast cancer for efti, this specific, and as we had, it seems, a representative patient population, I believe it’s a good working assumption to take this for the overall number, but it could vary from that.

Ahu Demir: That makes sense. And my last question will be about CDK4 inhibitors. 45% of the patients were treated previously with CDK4/6 inhibitors. Do you again see a particular trend in this patient population in terms of response rates or monocyte counts or any other biomarkers you have assessed?

Marc Voigt: Christian, would you like to take that one? Or, Frederic?

Professor Frédéric Triebel: No, there are no specifics for these patients treated with CDK4/6. We expect in the phase three to have maybe 90% of patients pretreated with CDK4/6 and not only 50% like here. It’s just that when we started the trial, no patients was treated with CDK4/6 years ago and at the end of the trial, most patients were treated with CDK4/6.

Ahu Demir: Thank you very much for taking my questions.

Clive Tompkins: Okay. Thank you. Next question from Tanu Jain at Bell Potter. Please go ahead.

Tanu Jain: Thanks, Clive. Thanks for taking my questions, guys. Just the first one on clinical practice. In terms of the subgroup for age is self explanatory and I would assume identifying luminal B cancers is also part of clinical practice, but how many patients actually get identified with having a low monocytes? Is that something which is normally done?

Professor Frédéric Triebel: Usually, clinicians look at the ratio between myeloid cells and lymphocytes like notified counts, lymphocyte counts, ratio. That’s what’s called NAR. It’s commonly used by clinicians in some cancer studies. Here, we prespecify, because of past results in the phase one, we prespecify the lower quartile and less than 250 monocytes per microliter of blood. And that’s what we call low monocyte, in short. But of course, again, it’s not, maybe not only the monocyte counts, maybe it’s the ratio between myeloid cells versus lymphocytes and if this ratio is altered leading to less monocyte and then to less generic cells, then cells are derived from monocytes, then, you have a defective innate system and the efti could come to the rescue. That’s the explanation we have at the moment, but possibly, there are other defect than just low monocytes.

Tanu Jain: Right. And just in terms of low monocyte counts, does any existing standard of care at the moment have a relationship to decrease this?

Professor Frédéric Triebel: As I said, there are many studies in cancer where the notified counts, lymphocyte count ratio has shown that clearly, it’s a predictive marker in some cases.

Tanu Jain: No, sorry. I guess my question was more to do with the class of drugs. For example, does chemotherapy or your CDK4/6 inhibitors, are any of them linked to cause this?

Professor Frédéric Triebel: No. No. Our patients have not received chemotherapy in the last, the preceding month or years and CDK4/6 is not known to induce, decreased monocytes count.

Tanu Jain: Right. Okay, great. And just in terms of Ahu’s questions on the overlap, within the less than 65 years old, do you have a number of how many would have presented with low monocyte counts?

Professor Frédéric Triebel: As I’ve said, there is an overlap between low monocytes and younger patients this overlap is not very strong.

Christian Mueller: I think what you can say is you can expect 20% of the patients have, 22% in efti group generally speaking, but around about 20% to 25% have a low monocyte count at start, and they have around about two third of the patients being less than 65 years old. So what you can expect, purely from a statistical point of view, is that you have a similar proportion of younger patients in the low monocyte group all the other way around. So there is an overlap, but if you ask for enrichment, for instance, if you put the two things together, I don’t think that’s what we have seen so far.

Tanu Jain: Great. And then, just moving on to the TACTI-002 result, more on a broader competitive point of view, the results that we’ve seen in the head and neck cancer, second line, they’re really great and they compare extremely well with historical KEYTRUDA monotherapies. Now, generally, we know that KEYTRUDA monotherapy response rates between second line patients and first line patients isn’t that different, so I guess in terms of having a read through from, especially on the PFS and OS rates that you guys are seeing at 12 months and six months, if that could perhaps have any translation to the first line patients, and I guess, comparison with the historical KEYTRUDA data in that first line setting?

Marc Voigt: Yeah, in general, we have here, as you said, I believe, very exciting results and we have been confident earlier on based on what we saw in an interim stage to move to first line head and neck and I think it’s also known that the second line patient population is PDX naive, so there could be a translation, or at least, let’s say a comparison between second to first line or translation from that into first line.

And of course, we hope, in terms of overall response rate, duration of response, progression free survival, overall survival, in combination with safety, that we are able to beat KEYTRUDA monotherapy. Currently, it looks quite nice. The study has just been starting to recruit, so we have to see what the results will tell us, but the expectations are relatively high and this is why we embarked on this commercially much more relevant setting in first line head an neck versus second line head and neck, especially in the PDX naive population as most of the patients in the coming years will be pretreated with PDX. So that could be a good indication. I mean, our complete response rate, very nice overall response rate and this will be the primary, or is the primary endpoint of TACTI 003. And of course, we have then also implemented the additional secondary endpoints to have a look at the quality of the responses so I believe, by and large, one could say yes to your question.

Tanu Jain: Great. And then, just one more last thing. With the 19th March PDUFA date for relatlimab, you’re probably getting very close to have the first LAG-3 product being approved in the market. And obviously, their first indication is melanoma, as we know. I think the last conversation we had, Marc, you did say that a LAG-3 antagonist can never be combined with efti, but I guess in terms of your efti plus KEYTRUDA combo, do you see, in non-small cell lung cancer or in head and neck cancer, potential relatlimab combo as potentially being competing with efti plus KEYTRUDA? And any thoughts around that?

Marc Voigt: Yeah. So first of all, you are right, it wouldn’t be of course, right, to give anti-LAG-3 and efti, which is a natural form of LAG-3 fused to human IgG1 together. So it could be given in alternate scheduling or so, so it wouldn’t exclude it altogether. And we compete with everything out there, which is active in first line non-small cell lung cancer, second line non-small cell lung cancer, head and neck cancer, metastatic breast cancer, so we do not have the narrow view on LAG-3. We are developing drugs towards registration if the data is right, so we will have a view on everything.

If you look at the latest data, for instance, from SITC, regarding anti-LAG-3 and pembro, I believe there was a trial from Merck, if I’m not mistaken was overall response rate of 25%. We had, by blinded independent central review, more than 40%. So we do not fear, currently, other concepts. It’s good for the patients that there is competition, we should never forget about that. There’s, of course, anti-LAG-3, there is anti-TIGIT, but I believe our data, if you compare it, it’s looking quite promising and I believe, besides us, today, also, Bristol-Myers Squibb gave an update in terms of what they plan to do. There were two LAG-3 talking points today. On the one hand side, our webcast here, and then BMS, and they intend to initiate first line non-small cell lung cancer study in first half of ’22, so exactly when we see our results from our phase two, and I believe they plan for phase three in colorectal cancer, and they have also, they are excited even though they didn’t share any specific in terms of overall survival and melanoma. And for us, it’s great.

I mean, anti-LAG-3 is hopefully successful for the patients with eftilagimod we have something unique where the data is very encouraging and it doesn’t need to be tweaked also in a certain way. I think all people can look at the posters, at the data, at the plans we are doing. In other words, if you do the benchmarking, I think we have actually a pretty good position, with all due respect for historical comparisons, but also we have, as we all know a randomized setting with anti PD1 ongoing and just saw the data from another double blind randomized setting here in metastatic breast cancer.

Tanu Jain: Right. And in terms of safety, what you’ve seen with the relatlimab PD1 combo, just on, as it currently stands, do you see efti having a safety advantage in combination?

Marc Voigt: Yes, indeed. And maybe should highlight that more. We rarely have safety slides. Safety is a strong point, one can actually say tolerability is a strong point of eftilagimod. This is explained by the fact that we have an agonist concept with low dose, 30 milligram per patient subcutaneously, so easy route of administration, low dose, so patient does not need to be a exposed to DLT or MTD. Also, in the anti-PD1 combination, we have a strong point, of course, there are injection site reactions. This is known for efti, but especially that we have that, we are looking here at the combination world. Anti-PD1, chemo therapy, targeted therapies combination is obviously the way to go, so to have the advantage of good tolerability is absolutely key and this is where we, I believe, can say that we delivered for the time being.

Tanu Jain: Great. Thank you.

Clive Tompkins: Sorry, Frederic, go ahead.

Professor Frédéric Triebel: Sorry, yes. These two approach going in different direction, of course, relatlimab plus anti-PD1 or efti plus anti-PD1 could work in patients with a high PDL1 expression at the tumor site. But now, if you look at cold tumors where there is no active T-cells at the tumor sites, how could an anti-PD1 plus an anti-LAG-3 work at the end of the day we know that pembro alone, for instance, works only in inflamed tumors with high PDL1? So by just adding an anti-LAG-3 to the anti-PD1 is not going to work in cold tumors while we have seen, in TACTI-002, responses, good responses in low PDL1 tumors, cold tumors. So we’re addressing different population of patients.

Tanu Jain: Great. That’s really helpful. Thank you.

Clive Tompkins: Okay. Look, Clive, here. I think we need to wrap it up. Unfortunately, we can’t get to all the questions within the allotted time. Look, I’d like to hand back to you now, Marc, for us in concluding remarks, please.

Marc Voigt: Yeah, thank you. And thank you for the discussion. My apologies that we have been running a little bit over time here. We received also some other questions which we will answer in writing then. So this is great to see that discussion around efti. We are very thankful for that and on behalf of the Immutep team, we would like to thank everyone for listening today and the analysts, of course, for the questions and discussions. And just as a reminder, a replay of the webcast will be available shortly on the Immutep website. So thank you again, appreciate that you took the time and goodbye.

Ends