Okay. Good afternoon, everyone, and welcome back to the Cantor conference. We're on time here, so we're going to get started. My name is Eric Schmidt. I'm one of the analysts at the firm, and it is my pleasure to welcome Tango Therapeutics today.
And Barbara Weber, the company's Chief Executive Officer, will be sharing this fireside chat with me. We also have in the audience Adam Christel, the company's President and Head of R and D, and Liz Hicklin as well, who runs IR and Communications. Thanks, team, for coming.
Thank you. Barb, maybe you could just start with a kind of quick two minute summary of state of affairs at Tango. So, I think things are pretty exciting and pretty busy right now at Tango. Where we are is really moving rapidly forward with our lead program, which is TNG462, an MTA cooperative PRMT5 inhibitor. I think the data that we've released some of in the past and our competitors, particularly at Bristol Myers Squibb, have shown that this is really an important target moving forward in cancer in multiple histologies, we feel excited to be part of that and release our data later this year.
So, yeah, big next few months for the company. Maybe for those who are a little bit less familiar, let's start at square one. What is an MTA cooperative PRMT5 inhibitor, and what's kind of unique and novel about these compounds that we're about to discuss?
Yeah, that's a biochemically complex answer, so I'll try to be as high level as possible. PRMT5 is a methylase that is important for regulating many different proteins with many different functions. And it is an essential gene, so if you fully inhibit PRMT5, any cell will die. The key is finding something that will give you differential activity in tumor cells versus normal cells, so that you have a therapeutic index. And that turns out to be what's called MTAP deletion, which is a genetic alteration that's present in a large percentage of human cancers across different histologies, and that creates an environment where you can make a molecule, an MTA cooperative molecule, that is differentially much more active up to a certain high exposures in tumor cells with an MTAP deletion versus wild type.
Okay. High level overview on MTAP deletions, how common they are, which histologies?
So, MTAP deletions occur in fifteen to twenty percent of all solid tumors, but they're not of equal incidence or prevalence across histology. So the biggest population is lung cancer, both because of the frequency, which is in the fifteen to twenty percent range, but because the number of lung cancer patients is so high. After that, pancreatic cancer and glioblastoma are also big patient populations with a significant number of MTAP deleted patients, and then there's a number of other indications below that.
And how routine is screening for MTAP deletions these days? Do patients know that they have them?
Unfortunately, generally not. To be honest, screening in general, even in places where you think it should be happening in every single patient, isn't always routine. But the short answer is that MTAB deletion is included on all of the big commercial panels and on many of the academic panels. So, patients who are being screened, they will often be screened for MTAP. And I think because of the recognition that's increasing that PRMT5 inhibitors are very active in MTAP deleted patients, at least in academic centers, patients are increasingly getting screened to see if they'll be eligible.
Okay. And there are different ways to inhibit this pathway, in general, in MTAP deleted tumors. PRMT5 is just one, but we've heard about MAT2A inhibitors. I don't think there's I think it's all that we know about. But any thoughts on the two different strategies?
Well, the direct way to inhibit PRMT5 is to inhibit PRMT5, right? Inhibiting it through a molecule that fits into the enzymatic pocket of PRMT5, and that's what the MTA cooperative inhibitors do. MAT2A inhibitors, as you say, is the other approach. What they do is modulate a cofactor of PRMT5 that increases the ability to inhibit PRMT5 in MTAP deleted cells, but is not specifically targeting PRMT5. I think, as our former founding CSO, Alan Wong, always said, inhibiting PRMT5 with a MATA2A inhibitor is sort of like trying to inhibit a specific kinase by lowering ATP levels.
It's an indirect way. It has some activity, but it's hard biologically to imagine how a good PRMT5, a direct PRMT5 inhibitor wouldn't always be superior to an indirect inhibitor.
And I guess what we're starting to see from this class we'll get into all the compounds players in a moment but we're starting to see a somewhat gradual onset of action in terms of the activity of these inhibitors. Why is that?
I don't think there's a good biologic explanation, but observation and it's in contrast to what you tend to see with kinase inhibitors that inhibit oncogenic drivers. With a traditional kinase inhibitor, you'll see the first scan will pick up the full ORR. It'll usually get the maximum response benefit for any individual patient, and then those patients will sit there for some period of time and then the tumor will start to regrow. What you see across PRMT5 inhibitors, and it's true with what we're seeing as well, is that tumor starts to shrink and it just keeps shrinking over time. And the second thing is, it's important, is even in those patients who don't get all the way to full shrinkage as measured by a resist PR, those patients benefit significantly.
And in the case, I would say, of kinase inhibitors, if you don't have a PR, you can expect your benefit to be fairly short lived. So, there's two big differences there.
Okay. So, that durable tumor control is a feature of the class. What do we know about resistance? It implies that we have difficulty generating resistance, but what can we do?
Yeah, we know very little about it. We nobody, to my knowledge, including us, has been able to generate resistance mutations in the lab. There's not much known about when patients recur, why they're recurring. And I think that's probably for a couple different reasons, but the biggest one is that it's probably for molecular reasons difficult to generate drug resistance mutations, specifically in the enzymatic pocket of PRMT5, but in addition to that, because PRMT5 modulates, you know, 90 something proteins in a number of different pathways, my guess is that most of the resistance occurs down those pathways and is very pleiotropic and it's going to be hard to sort through.
What you're describing mechanistically, it's almost like an IO oriented mechanism, where you have a slower onset but a great tail of the effect and good tumor control and durability. But is there anything that could be happening in the immune system here?
I mean, I think that the immune system, as much as I hate to say it as a geneticist, is probably always involved in the response to tumors, but this does have the hallmarks of something that's requiring that as part of it. There is an interferon gene cluster deletion that is part of the same deletion sometimes that includes MTAP, so whether that plays partially a role in this as well, I don't know.
But we don't understand it.
We don't. If it is. We really don't.
All right, let's get to the landscape and your lead candidate, TNG462. First, just maybe orient us to who are the players and what stage of development they're at.
So I think the leading players in terms of time are us, BMS and Amgen. I think the way we look at the data that are out there, we really see Bristol as our biggest competitor. I think that molecule, the former Mirati molecule, is a good PRMT5 inhibitor and I think the data that they showed, particularly at ASCO this year, is consistent with our expectations for what we should see with a good molecule. We think four sixty two, based on the data we have shown publicly, particularly around PK data, tolerability data, and the knowledge that BMS is limited by off target effects as opposed to being able to fully inhibit PRMT5 to the extent that they could without those off target toxicities has the potential to be at least as good, if not better, than BMS from an efficacy standpoint as well, but over time, we'll see.
And who else is kind of in the next tier of competitors?
So, first of all, of those three, I would say the Amgen molecule, I think, has a number of characteristics that make it less attractive compared to Tango or to Bristol Myers. After that is AstraZeneca. We haven't seen their data yet, really, in any clinical forum, and don't know yet if that'll happen this year. They are probably about a year behind four sixty two with their molecule, and then behind that is BeiGene and a few others.
Okay. What makes TNG well, what makes Bristol a good molecule and TNG462 a better molecule?
I think the key for really a good PRMT5 inhibitor is, in fact, the MTAP selectivity, because the more drug you can give, the more you can inhibit PRMT5, the more response you get, and most importantly, the bigger the more the durability that you get with that is. That's clear from sort of both preclinical and clinical data at this point. And so, to start out with a preclinical characteristic, you want a very MTAP selective molecule. Then, you want a molecule that's free of off target toxicity so that you can fully utilize that MTAP selectivity. I think that's where we start to see an advantage with four sixty two.
And then you want a tolerability profile that keeps you from having to dose reduce beyond where you really ultimately dose. And again, someplace where we think we have a slight advantage.
If we do compare the Bristol and Tango molecules, MTAP selectivity you've already mentioned, you think you're a little bit superior on. Any off target effects on either of these molecules that you would call out potentially?
For four sixty two, I would say it's very clean. For BMS, we know based on the data that they've disclosed that their dose limiting toxicity is off target, which is seizure and, to some extent, rash, GI, tolerability and fatigue. The known on target toxicity of a PRMT5 inhibitor is anemia and thrombocytopenia, and we can dose escalate fully up until where we start getting dose limiting anemia and thrombocytopenia. BMS doesn't get there because they can't use their full range.
Okay, so your dose expanding at two fifty mg, and we don't know their final dose, but it's going be higher than that. And what kind of relative target coverage do you think you're getting versus them?
If you normalize this by saying how much what is the how much more drug over the IC90 in preclinical cells do you get with each of the molecules, four sixty two if BMS picks their highest possible dose, we're hitting the target about one and a half times harder. And if they went with four hundred milligrams, which is also possible, it would be about twice as hard. Okay.
So, at the end of last year, you did provide a glimpse of your clinical experience on the molecule. Maybe just remind us what you said then.
Sure. The we disclosed the median PFS from the dose escalation data set, and that was a range of 13 or 14 different histologies. That number was, to me, a remarkable twenty four weeks of median PFS in a dose escalation cohort in the active doses of patients with late line, difficult to treat cancers. And I think that alone was to me sort of a striking indication that this was going be a very active drug. We also released the data on the only cohort of tumor type that we had enough patients to comment on, which was cholangiocarcinoma, and showed that the ORR was roughly twice what BMS and Amgen had disclosed and almost four times what's standard of care second line chemotherapy in cholangiocarcinoma.
That is not to say that we're going to develop in cholangiocarcinoma as an isolated indication, it's a very small group of patients, but I think it was a way to, again, benchmark against the competition.
Okay. And what has Bristol shown that got folks even more excited at ASCO this year?
So what they did was they finally, and it's been great for us as well, have shown what we've been saying, and they've been saying as well for a while, which is that the real metric to evaluate PRMT5 inhibitors is the durability of response that you see both in the patients with stable disease and in the partial responders. And I think, you know, there was good reason, I think, for skepticism of people saying, Yeah, you can tell me that it's all about durability, but you haven't shown me the durability data and the indications we want to see. And BMS did that. They showed their spider plots for pancreatic cancer and lung cancer and really showed that PRMT5 was hitting PRMT-five as a target, I think, was going to be really important in thinking about changing the face of pancreatic cancer treatment away from chemotherapy and to something that's both more active and more tolerable.
Let's start with pancreatic cancer. What is good response rate data and good durability data? Or good PFS data, however you want to define it.
Yeah, the PFS for second line chemotherapy in pancreatic cancer is a pretty dismal ten to twelve weeks. And I think it's difficult to you have to extrapolate to come up with a number from their data. But what we're looking at is saying, If you can at least double that, if not more, then that's a straightforward registration strategy, And that's an important advance for patients. And so I think that was in the range that they showed.
Okay. And maybe it's time to talk about your update from at the back half of this year. What how would you characterize the data set?
So, we will disclose the data on the full tumor on the full trial, which is a pretty big trial now at this point, but the focus will really be on pancreatic cancer, and the goal of that data disclosure will be to provide evidence that TNG462 has a successful path forward in pancreatic cancer, both as a second line monotherapy and then with plans going forward for first line.
Okay. So, how many patients and when?
So, the overall trial is well over one hundred patients. What we've said in pancreatic cancer is that we'll have, again, well more than twenty patients that have each of them at least six months of follow-up. And then, in addition to that, we've now accrued so well in that cohort that we've enrolled a number of second line patients, so we'll be able to show data for second line versus third line plus pancreatic cancer patients as well.
Will you have 20 of each?
Yeah, I think we'll be probably in that range.
You have six months follow-up? It's a very meaningful Adam, data
shake your head yes or no. Adam says yes.
Okay. So, we could see 20 line, twenty third line patients, each with follow-up six months, and we would get a good sense of that benchmark that you reported earlier, ten to twelve weeks, hopefully doubling that in the second plus line.
That's the key for the registration strategy. I think there are other things beyond that that will be important going forward. But that's the first and most important piece of deciding on empowering the registration study.
Okay. And with regard to the trial readout or the timing of the readout, when and what form?
There will be a company event. We will then follow that by the next feasible medical conference. We're not timing the data update to a medical conference, so just keep that in mind. And sorry, what was the other question?
Time roughly when in the year?
Oh, we haven't given any further guidance other than second half. But, you know, as we've been talking, the because the data because the tumor response to a PRMT5 inhibitor continues to increase over time, That makes ORR a time dependent variable with PRMT5 inhibitors, which is not the case, right, with kinase inhibitors. And just so, in general, the longer you hold on to the data, potentially more attractive the data becomes. So, it's a balance between
Would you show us patients treated less than six months, or is that disadvantageous to the drugs data set?
I mean, I think we will disclose the whole data set, but the focus will be on the second line pancreatic cancer patients that will inform the Phase III study.
And relative as you know, Bart, we're always interested in comparing and contrasting across data sets, however much hazard that may bring us. We saw with Bristol response rates in the seventeen percent to twenty five percent range. You mentioned that their PFS looked good, but it wasn't called out as a reported variable, but maybe it was in the six months range, it seems, from the spider plots. So, thoughts on whether those are the appropriate benchmarks to be looking at your data?
I think they are, right? And I think that this sort of seventeen percent to twenty five percent, depending on the doses that you look at, with the Bristol data for pancreatic cancer is reasonable. It's not very different from the RevMed data, which is, you know, again high, twenty percent, but it's about the durability. So, I think as the BMS data are a reasonable bar, and as we've said, we expect to be in the ballpark for sure.
And how should we think about your view that this is a best in class inhibitor? What might we see from your data to support that view at this early stage?
Well, I think the data to support it that we've already shown, right, is both the PK showing that we can hit the target harder, and the tolerability showing that we have a better safety profile, will or tolerability profile, which will keep patients on drug other than, you know, in an important way. We I'm going to step back from promising you that our efficacy data at this early time point are going to be clearly better, but I would definitely continue to say in the ballpark with the opportunity over time for this to be as good or better than the BMS. We do think that it has the potential to be there.
And on tolerability, what do you think it is about the Bristol drug that, you know, can in particular be improved upon in a way that benefits patients?
It's the Well, it's really about the ability for the target coverage, right? It's what I mentioned before, that because BMS is dose limited by off target toxicity, they're just not able to hit the target as hard as they otherwise would be able to, and as hard as we can. And I think in the end, that has in the past always translated into efficacy in oncology, and I think it will in this case as well.
But beyond even the pharmacology clinical tolerability, do you think that it's tolerable enough at the lower doses that patients are not coming
Yeah, I don't want to overstate that, right? The BMS molecule is a good molecule and it's very well tolerated. But with the dose limitation that they have, they're just not able to get the target coverage that they could if they didn't have that.
Okay. So, what are the next steps in PANC?
So, our next steps are our first registration trial, which we're planning next year. And in addition to that, we've already started our study in combination with the two RevMed RAAS inhibitors, doroxonrasib and zoldonrasib, which we're really excited about because we think that may be the preferable path to first line registration is PRMT5 plus RAAS, four sixty two plus a RAAS inhibitor. And for us in pancreatic cancer, that's really important because essentially all MTAP deleted pancreatic cancers have a RAAS mutation.
Okay. So, I guess we can't talk about one without the other in terms of your strategy for a pivotal study as well as these combinations with the KRAS and the multi RAS inhibitors. Just remind us when the multi RAS trials started in combination with four sixty two?
In late June of this year, we dosed the first patients on that study, and there's both the combination of four sixty two plus Diraxon and four sixty two plus Zoldon are in that trial and are accruing at sort of equal paces.
How are they accruing, and when could we see data?
Well, they're accruing very well, but it is a dose escalation study, so it still has the inherent, you know, slowdowns to get through
DLTIP Nine is where you started the
All three of the molecules are starting within active dose ranges.
About half your target dose for four sixty two?
For four sixty two, it's at two hundred, which And is an active
for the others?
It's one hundred of Duraxon and I think six hundred of Zoldon.
So we're talking one or two dose
At the most.
Okay. So there's not much dose escalation here. That's right. We thought it was a dose escalation trial.
That's right.
Okay. So, we'll quickly hit an expansion cohort. And then sorry, we're nodding. So, for those on the webcast, we're nodding. And then, we will flush out that cohort and have data. When do we think?
We haven't really given guidance on that. We just need to see where we are with the dose escalation, then we sit down and talk to RevMed, then we figure out where we are.
So, does the plan today, as per this study, include a dose expansion cohort, or is it just, Let's see what we get from a dose escalation cohort?
There are plans for the dose expansion cohort, but there is a discussion that needs to happen at the end of the dose escalation.
Remind us who's paying for the trial.
We are.
Okay.
We hold the IND and we are paying for the study.
Okay. And you would report the dose escalation portion of the trial at some point in the not distant future?
Presumably, but it's again, it's a discussion and, yeah.
You don't own the data?
We jointly own the data.
Okay.
So, how does this all fit in with your potential thoughts on pivotal study? And I guess, we can talk about second and first line.
So, my unilateral thoughts on this are that we move forward quickly with a second line monotherapy, that assuming that the data from the clinical study for the combination with RAS inhibitors is roughly in line with the preclinical data, then we will move quickly to registration planning in first line with RAS plus four sixty two and get that study started as quickly as possible as well. And I think it's a complicated space. It's a good thing for pancreatic cancer patients, but both PRMT five inhibitors and RAS inhibitors are going to be really important moving forward, and there's going to be some amount of needle threading over the next three to five years of single agent combinations, which one first, and that we think we'll have a role in both of those.
Okay. So, unilateral view, and I'm not sure who else's opinion matters, but your Doctor. View Goldsmith. Okay. But it's to start a second line study monotherapy against chemo. Well, that's your your view. That's what you would do. Yes. And then, try and follow that as quickly as possible
with With front first line study. Now, whether that's going be plus RAS or plus chemo, the data will tell, but that is our plan, yes.
But the thought is not to skip the second line monotherapy trial at
think there will be value in second line after RAS and after chemo for some period of time that's valuable, yes.
Okay. So, are you doing to prepare for the second line study with chemo in terms of exposure with chemo?
So, that would be a first line. So, let me
be Oh, sorry.
Second line is monotherapy. That's okay.
My bad.
It would only be if there was some reason not to move forward with the RAS combination. Okay.
So, you've got what you need, or you will have what you need shortly to go forward.
Once we have the efficacy data from the combination sorry.
We believe On the therapy second line.
We are already preparing at risk for the monotherapy second line study. And as soon as we have the data from the combination of with RAS, we'll have what we need to plan the first line study.
So, why what is rate limiting to starting the second line monotherapy study? What else do you need to see from your ongoing cohort?
I think we're preparing at risk, so I think we feel pretty confident about that. We there is the slight matter of the money to pay for the study, so that will be an important piece of what's gonna happen this fall.
Okay. But there's is there anything else that you need to see from the data to be convinced that Nope.
Not to me.
Okay. So, it's a monetary question.
Well, it's a timing question, right? There's all sorts of things that have to happen to get a phase III study going.
That's huge issue.
We are moving And forward with
then, in the front line, you can go in one of two different directions. My bad again. You go with chemo or you can go with the RAS inhibitors, and you need, obviously, the data from the RAS inhibitors to make that decision, or the data from a combination study with chemotherapy to make that decision, if you choose not to go with RAS inhibitors.
I think in terms of the strategy around the combination with chemo, what we will be doing is the tolerability studies for the two most important combination regimens in pancreatic cancer that are used in Europe and in The US. In terms of an efficacy study, that probably wouldn't happen. It's more a matter of the efficacy study from the RAS combination to decide whether we think based on historical data that would be superior.
Okay.
It's pretty clear what the activity of chemotherapy is in first line pancreatic cancer. If we think the RAS combination data will beat that, then we would just go straight forward with that.
And what is the benefit of having the second line indication if you're also going to the first line with either of these two strategies?
Well, to some extent it's timing, right? There will be some number of years where both, one is, you have a second line approval and then you go to a first line. But also, as practice changes, and as practice changes in Europe versus The U. S, right, at some there will still be value for some period of time to having a therapy for post chemo or post RAS first line treatments.
And when you look at the RAS landscape, you have a partner, obviously, that has a very strong position there, but there are multiple other players. How do you think about the tensions across this landscape where, you know, you may have the best PRMT5 inhibitor and others may have relatively equal, potentially, RAS inhibitors?
My personal feeling, based on the data that I've seen from RevMed and from others is that RevMed has the best RAS molecules. If that were to change, we could make a different decision, as could they on PRMT5. These are non exclusive, non binding collaborative agreements. But I feel pretty good about RevMed as having the best RAS inhibitors right now.
And we should touch on your other PRMT5 inhibitor, fifty four fifty six. You. Recently entered the clinic. Why? What's special about this drug?
So, what's different about four fifty six than four sixty two is that it's brain penetrant. And because almost half of glioblastoma is MTAP deleted, it's really important to have that brain penetrance in a PRMT5 inhibitor to hope to be effective in glioblastoma. So, that is just a different molecule that allows us with similar MTAP selectivity to be able to get into the brain and ask the question, is it active in GBM? I believe it should be if we can get enough drug in. There's nothing magical about GBM genetically.
It's just that it's hard, it's A, been difficult to find the right target in GBM, and then to get enough of the right drug that hits that target into the brain.
How quickly can you enroll a cultivated population of GBM patients?
Well, enrolling GBM patients is sadly not difficult at all, right? We were flooded with GBM patients when we did the initial nine zero eight trial. Unfortunately, we weren't getting enough nine zero eight into the brain, but I think that's not going to be a much of a limiting factor.
The phase one is all comers, or is it just going be all GBM because it's so prevalent?
The initial dose escalation steps are in non GBM patients, but we will move to adding GBM patients as we get into the active dose range.
And are you getting are you excluding GBM patients from the Phase one?
From the early dose escalation because there's suboptimal dosing and
So, how quickly can you get to a cohort of GBM patients?
I mean, I think that's something that still remains to be seen, but accrual is going well with that study, and PK is as expected or better, so I feel good about that.
And remind us what dose you could start that trial
That study was that was a standard dose escalation study. I think we expect something like four to five dose escalations into the active dose range.
Okay. So, not too long. And the CoREST program, your last program to touch on as well. We should see some data from that at the end of the year?
Yep, totally different program, totally different patient population, totally different mechanism, but the idea based on both our preclinical data and the clinical retrospective data that patients with lung cancer and who have STK11 mutations are not responsive to checkpoint inhibitors, and that adding two sixty to that combination in patients who have already failed a checkpoint inhibitor is able to reverse the effects of the STK11 mutation transcriptionally and make those tumors responsive.
Should we be optimistic for good data?
I'm not sure how I can answer that question. I think all the data that we've shown so far is consistent with the preclinical data that it is hitting the target, that it's well tolerated, and it has the mechanistic effects that you would expect to be able to reverse the resistance to checkpoint inhibitors.
Do you have aspects of the trial in place to measure good PD biomarker activity? Yes. So, you'll know whether it's hitting the target?
We already know that. The answer is yes. And in fact, we've already disclosed the proof of mechanism to show that you can market the upregulate PD L1 expression on those cells, change the influx of cytotoxic T cells, all the things that are required to be able to produce an effect.
Terrific. We will hold tight and be patient with those results. We're out of time. Thank you, Barb. Appreciate the update.
Very much.
Thanks, everyone. Always fun.