Insight Molecular Diagnostics Inc. (IMDX) Earnings Call Transcript & Summary

Good morning, and welcome to the OncoCyte Expert Webinar. [Operator Instructions] As a reminder, this call is being recorded and a replay will be made available on the OncoCyte website following the conclusion of the event. Before turning the call over to Doug Ross of OncoCyte, I would like to remind you that during this conference call, the company will make projections and forward-looking statements regarding future events. Any statements that are not historical fact are forward-looking statements. We encourage you to review the company's SEC filings, which identify the specific risk factors that may cause actual results or events to differ materially from those described in these forward-looking statements. I'd now like to turn the call over to Dr. Doug Ross, Chief Scientific Officer of OncoCyte. Please go ahead, Doug.

Well, thank you all for joining us. I really appreciate your taking some time out of your day to learn about some of our latest developments of our product DetermaIO. For those of you who are new to OncoCyte, we're a precision diagnostics company that has a pipeline of tests that help inform key decision points along a patient's cancer journey. We're here today to focus on a product called DetermaIO, which is something that classifies, characterizes the tumor immune microenvironment, and we've had some really exciting data released over the course of the past year, but especially 4 weeks ago or so, at the American Society of Clinical Oncology meeting. And I'm very pleased to be able to introduce some of our guests that will be the stars of the show, not me. We're going to start actually with a colleague, Rob Seitz. Rob is going to review -- Rob is the Head of Immune Oncology at OncoCyte and arguably the inventor of the test. He had the insight to recognize that one of the classifiers that was in the academic literature for classifying triple-negative breast cancer brought out elements of the tumor immune microenvironment, and he had the notion that those elements of the tumor immune microenvironment might be a good predictor of immune checkpoint inhibitor response and has worked with us, and now with OncoCyte, to develop a product that's now on the market. In addition to Rob, I'm really pleased to introduce 2 of our guest speakers, Dr. Chiara Cremolini, I just met a few weeks ago at ASCO. She is a specialist in medical oncology focusing on gastrointestinal, GI, malignancies and she is responsible for the clinical trial office in oncology at the Santa Chiara Hospital in Pisa, and she's also the past scientific secretary and now current President of the GONO Foundation. I'll let her pronounce the long version of the GONO Foundation. It's in Italian, and I'll screw it up. But she's a rising star. I don't want to embarrass her, but she's very much a rising star in translational oncology and taking insights from molecular medicine and making them useful to oncology patients. And so she will be following Rob's talk, and then we'll all get together in sort of a fireside chat online, which we've all gotten used to. I'm pleased to introduce Dr. Adam Brufsky who's a Professor of Medicine at the University of Pittsburgh, UPMC, and has been focused on translational research in oncology for the past 25 years or so. He is a Professor of Medicine at the University of Pittsburgh School of Medicine, serves as Co-Director of the Comprehensive Breast Cancer Center and Medical Director of the Women's Cancer Center at the UPMC Magee-Womens Hospital in Pittsburgh and he also serves as the Associate Director for strategic initiatives for the UPMC Hillman Cancer Center in Pittsburgh. And we've gotten to know him that way through talks about partnerships and collaborations. He's truly a pillar. And for those of you who don't know UPMC, it's one of the premier translational centers in oncology in the world. And so pleased to introduce him. So without further ado, I'll turn it over to Rob to give you an introduction to DetermaIO and some of the data that we presented at ASCO recently.

Thank you, Doug. Let's move to the first slide. I want to begin actually not with ASCO but with data we released in December at San Antonio because 2 of the abstracts that we released at ASCO really build on what we did at San Antonio and answer a lot of the critique of what we did at San Antonio. San Antonio, we released our first randomized clinical trial data. You're going to hear about our second randomized clinical trial data when Dr. Cremolini speaks, but this was our first randomized clinical trial data with the Fondazione Michelangelo group out of Milan with the NeoTrip clinical trial. And this was in triple-negative breast cancer, 220 patients randomized to Genentech's Tecentriq, which goes by the generic name of atezolizumab. They got chemotherapy in 1 arm and chemotherapy plus atezo in the other arm. The chemotherapy was 2 drugs, carboplatin and Abraxane; Abraxane which is a form of paclitaxel, a form of taxane. And the endpoint was what's called pathologic complete response, or PCR. You do a biopsy of the patient, you treat the patient and then you do surgery after treatment and you look to see if you can find malignant cancer cells. If you can't find them, you consider that the best initial result for the patient. And you call that a PCR, or pathologic complete response. We were pleased to find out that we validated in this test, but we were particularly excited with simply how clean the results were. Not only did we see all the benefit was in chemotherapy plus atezolizumab if and only if you were positive for DetermaIO. And I think I can show the benefit a little bit later in the next slide. Yes, so here we go. So I want to show you what would have been less clean data. Less clean data can be seen beginning at the top left. And I've kind of gone and altered the color of where I've altered the data of the slides. What would have happened if we've seen a slide like the top left where DetermaIO positive had a great deal of benefit if you got atezolizumab or if you got chemotherapy. That had been saying, "Hey, we're finding patients who are going to do better, but they're going to do better no matter what you give them." That can be useful information, but it's not really useful relative to immune checkpoint inhibitor. We're simply saying we're finding better patients. So it doesn't really help us tell physicians, tell clinicians information about an immune checkpoint inhibitor. If we move over to the next graph, you can see I've altered the color of the third bar. And what you see there is that, yes, the patients who are DetermaIO-positive are doing the best. But patients who are DetermaIO-negative and got atezolizumab, they're still doing better than patients who got chemotherapy alone. That's kind of predictive. That's kind of giving us information about immune checkpoint inhibitors, but it's kind of confusing, too. What does a physician do with that information? Down at the bottom left, I'd sort of just changed this slightly. I've said, "Hey, the patient is getting some benefit from chemotherapy." And again, it's kind of good information, it's helpful information, but it still can be very confusing. It's only in the very clean information that we've got where all of the benefit is seen only in 1 combination. You're either positive for DetermaIO and you've got the immune checkpoint inhibitor. And it was this cleanness of the result that we said we are really developing a tool that is going to help the clinical community. Now there were really 4 criticisms that came out of the NeoTrip study and all of them were fair. One of them was, "This is pathologic complete response. We want to know how this is going to affect overall survival." The good news is the NeoTrip study is ongoing. They're going to have survival data. And DetermaIO is going to be a part of that. It's going to be coming out either at the end of this year or beginning of 2023, but we're going to get that data out of this study. So that's one piece of data that we're going to get out of that. Another part of it was it's only 1 study. Well, we're going to be putting more studies, and that's partly what I'm going to be showing you here that there are even more studies coming down the line. A third factor though is that this is not the drug that is approved on the market. The drug that's approved on the market is Merck's KEYTRUDA, otherwise known as pembrolizumab. So we have data with Genentech's drug that's not approved. We don't have data with Merck's KEYTRUDA. And the fourth criticism is really that this is a retrospective study. We don't have prospective data. This is part of the reason why San Antonio is so exciting for us. This study is the NeoPACT study in neoadjuvant TNBC. This is really the only study we've ever done that we are not authors on. We actually just went and were provided the data and really didn't see the full data until San Antonio, until we walked up and saw the poster. Compared to NeoTrip, there's a lot of similarities. Endpoint is pathologic complete response. You look at the chemotherapy that they got, it's really the same class of drugs. It's carboplatin. And in this case, it's paclitaxel, but it's still a taxane. A couple of key differences, it's pembrolizumab, the FDA-approved drug. Another difference is it's not a randomized trial, we only have patients who are treated. One point that I'll come back to later, I'll just ask you to remember it, is that neither regimen has an anthracycline. Anthracyclines are part of what's approved when giving pembrolizumab, and there are some big toxicity issues for that. So it's actually good that this study didn't have it because a lot of clinicians are looking for reasons not to give an anthracycline. They'd love not to have a part of this. And I think that could be a good question for Dr. Brufsky later on in the talk. Here is the actual result of the NeoPACT study. We got a very high PCR rate. It was actually slightly better than what we got with NeoTrip, I'll show that in the next slide, but 81%. If you were DetermaIO-positive, you're 81% likely to have a pathologic complete response. That outperformed 2 other measures they used in this study where they looked at lymphocytes around the tumor and where they looked to see if the DNA had been damaged. It outperformed both of those. An interesting little study said [ pills ] work together, but we outperformed both of those. They're asking what about the 43% of patients who had a pathologic complete response with DetermaIO-negative. Remember, some of these patients are going to have responded to the chemotherapy portion of the regimen. So this is a pretty robust response. In fact, I've taken now the NeoTrip data, that same slide I've been showing over and over again, and I changed the color so you can compare them head-to-head with the NeoPACT. And you can see the NeoPACT pretty much gave us better results even than the NeoTrip, and we were awfully excited with the NeoTrip data. So we had similar and slightly better numbers. Now it does not have a chemotherapy control arm but it has similar chemotherapy that the first study did. So we're now looking at this and saying, we didn't find responders or predict responders with the chemotherapy in the NeoTrip, which gives us confidence as we go on to other studies, the other randomized trial, that we're not going to find responders in the chemotherapy arms of other studies. It gives us a lot of hope as we move on to randomized trials that involve pembrolizumab that we're going to be finding responders to the pembrolizumab arm alone. Again, this was not without a control arm. It doesn't have all the data that NeoTrip can have, but it had about the best results we could possibly anticipate for what it was. And again, it is the drug that the FDA has approved. We're also excited about the idea that this didn't have the anthracycline because a lot of the feedback we get from the community is they don't like the fact that anthracycline is a part of this regimen. The toxicity is just a lot. They'd love to know that there was a biomarker out there that could say you don't need to continue to pile on all these drugs. Following up on the same indication and the same drug, we also did a very small study in metastatic triple-negative breast cancer. The first is basically neoadjuvant triple-negative, the first time the patient reports. This is after, unfortunately, the cancer has come back. This, again, is a very small study, just a few patients. However, it's the drug on label. So when we look at this, we saw that we were finding responders within this group. Small study, not nearly as large as either NeoPACT and definitely not as large as NeoTrip, but we got separation of curves and they were significant. What was also refreshing about this and exciting about this study is that because it was a clinical trial, we got access to patients that were PD-L1 negative. Now why is that important? Because these patients are not on the FDA label. According to the FDA label, to get this drug, you have to be PD-L1 positive. So you don't get those by just going to the local doctor's office and say, "Can we test your patients?" They don't exist. You have to get those within a clinical trial. And that's where we've got access to these PD-L1 negative patients, and we showed that we were finding responders. That's exciting because then our value proposition to pharma is, "Hey, we can expand your label. Right now, your label is PD-L1 positive. It could also be PD-L1 negative, IO positive and expand the label that way." So it's an approach. Is pharma going to be interested with this level of evidence? Probably not. But the people who are interested is other groups with clinical trial data and PD-L1 negative patients, and those conversations are ongoing now. And then the final study I want to talk about before turning it over to Dr. Cremolini is this very small study in gastric cancer. Once again, we consider this to be really a proof of concept study. It was 59 patients. Our endpoint was objective response and we simply asked the question, "Does DetermaIO separate the patients into responders and nonresponders?" And the answer was, yes, it definitely does. It separates them into responders and nonresponders. We got a very clean separation of putting the responders in 1 group, putting the nonresponders in another group simply by whether or not you were DetermaIO positive or negative. Why is this important? Because gastric cancer is the third leading cause of cancer worldwide. Most of those deaths are in Asia. We don't really necessarily see this so much as a U.S. opportunity as we see it as a potential licensing opportunity for DetermaIO in Asia. They are using immune checkpoint inhibitors in Asia for this. We think this has a strong potential to be a biomarker. So in conclusion, we did 2 TNBC studies: 1 in the neoadjuvant setting, 1 in the metastatic setting. Both of those show that DetermaIO were finding patients who benefit with KEYTRUDA, pembrolizumab, the FDA-approved immune checkpoint therapy. The NeoPACT study showed strong results, along with NeoTrip, with using the same type of chemotherapy, which gives us hope that we're going to be able to go to a pembrolizumab randomized trial and find out that it's predictive. The result also argues that potentially we could remove some of the more toxic chemotherapy than this. We consider gastric cancer to be a potential licensing arrangement for Asia. DetermaIO also, and this is very exciting for us, is now validated in over 1,100 patients with 11 public studies, 6 different cancer types. We talked publicly about 5 of them. And you're about to hear about the sixth when Dr. Cremolini talks, and I will go ahead and turn it over to her now.

Just before we start with Dr. Cremolini, thank you, Rob. And just to summarize it at a high level for some of the perhaps non-doctors and scientists on the call, what Rob has shared with you is that DetermaIO is a test that looks at the tissue at diagnosis and helps make a decision around whether or not adding immune therapy to the standard chemotherapy regimen is useful in managing a particular clinical situation. And what's been exciting to us, as Rob has relayed, is that over the past year, we've shown data from a randomized clinical trial, the NeoTrip trial, that really demonstrated that the addition of immune therapy to the standard chemotherapy backbone, that DetermaIO is predictive of response to that. And then we followed up with the data released at ASCO that with the drug that is on the market, pembrolizumab, it has seen a result that was very much supportive of the predictive result that we got with atezolizumab. And then Rob shared that at ASCO, we also extended now to 6 tumor types, the number of tumor types that DetermaIO has been shown to be predictive. And Dr. Cremolini, who's now come on the screen, is the author of the sixth tumor type, colon cancer. And so she's going to present the data that she revealed at ASCO 4 weeks ago. So Dr. Cremolini, I'll turn it over to you. Thank you.

So thank you very much, Doug. It's a pleasure to present you the results that we have recently shown at the ASCO annual meeting about our research in the field of immunotherapy in metastatic colorectal cancer. In particular, just to give you a few words of introduction, in the last year or so, we have learned a lot about the cancer immunity cycle. Tumors are able to somehow evade from the control of the host immune system by activating a lot of alternative mechanisms. And in order to restore the anticancer immune response of the host, then some drugs have been developed, including checkpoint inhibitors that act mainly at 2 stages: by increasing the ability of T cells and, in particular, cytotoxic T cells, to kill cancer cells; and by increasing the activation of these cytotoxic cells, in particular, CD8+ T cells, that are able to accept the anti-tumor activity and the need to be activated of a close interaction with antigen presenting cells. So checkpoint inhibitors are monoclonal antibodies working at these 2 levels. And in the field of colorectal cancer, they have already provided very important, I could say revolutionary, and impressive results in a small subgroup of colon cancers that are dose characterized by specific peculiar genomic characteristics, which is a microsatellite instability. Having these microsatellite instability is the result of a malfunction in the system of DNA repair that is a mismatch repair system. So the deficiency of this mismatch repair system is associated with a high tumor mutational burden and to a really impressive sensitivity to immune checkpoint inhibitors. On the other side, in tumors that do not have these characteristics, the landscape is really different. With regard to microsatellite-instable tumors, so we are now using the immunotherapy in advanced lines of therapy. Actually, the first evidence that have been provided by clinical trials were collected in this field, advanced lines of therapy, for metastatic colorectal cancer patients. Then we achieved and collected the results in the first line of therapy of metastatic colorectal cancer patients. And now we also have evidence about the potential usefulness of this approach also in early-stage patients, in particular in the new adjuvant therapy of both colon and the rectal cancers. But as previously mentioned, the situation is completely different when we focus on the very vast majority of metastatic colorectal cancer patients that are bearing microsatellite-stable tumors. That accounts for around 95% of metastatic colorectal cancer. In this case, we do not have an inflamed tumor microenvironment. We do not have a high T cell infiltrate into the tumor. We have most frequently an immune excluded landscape or an immune desert. As a consequence, the checkpoint inhibitors do not work by themselves or accept a really limited antitumoral effect in these cases. And a lot of research efforts are currently ongoing in order to find out new strategies, mainly new combination of strategies, and new molecular markets to make immune checkpoint inhibitor efficacy also for the vast majority of metastatic colorectal cancer patients, those bearing proficient mismatch repair that's also known as microsatellite-stable tumors. So in terms of new strategies, this mainly means the combined immunotherapy with other drugs that are active in the treatment of metastatic colorectal cancer and that may potentially exert an immune-modulatory effect. In terms of new markers, the effort that we are going to do is to identify subgroup of patients within the pMMR subgroup of metastatic colorectal cancer patients that may actually be able to derive benefit from the use of checkpoint inhibitors. And some of these potential markets are listed in this slide. With regard to potential combinations that may make efficacious checkpoint inhibitors in colorectal cancer, bevacizumab is an antiandrogenic agent that is commonly used in the treatment of these patients. And bevacizumab is really actually able to somehow modulate the immune microenvironment of the tumor by potentiating the activation of dendritic cells that work to activate and prime lymphocyte activities. They facilitate the infiltration of lymphocytes into the tumor, and they also decrease the activity of immunosuppressive cells. So it looked like a very good idea to combine bevacizumab with a checkpoint inhibitor. But at the beginning of this story, this combination that was very successful in other disease indications such as, for example, hepatocellular carcinoma did not get to the expected results in metastatic colorectal cancer. We asked whether another help to the immune activation of the tumor could come from an active upfront chemotherapy. So the combination of the checkpoint inhibitor with bevacizumab but also with a cytotoxic combination of chemotherapy agents that by achieving an important treatment activity could also be able to lead to the release of cancer cell antigens that is somehow inducing this cycle of cross-talk between cancer cells and the host immune system. With this background and this biological rationale, we designed this trial that was named AtezoTRIBE that was conducted in Italy. It was an academic nonprofit trial sponsored by the GONO Foundation where 218 metastatic colorectal cancer patients that were not previously treated for their metastatic disease were randomized in a 1:2 ratio to receive a potential upfront therapy for these patients the combination of chemotherapy FOLFOXIRI plus the anti-VEGF bevacizumab or the same regimen with the addition of an immune checkpoint inhibitor, the anti-PD-L1 atezolizumab. The treatment was continued up to 8 cycles and then followed by the potentiated maintenance therapy with 5FU, which is a single cytotoxic agent, plus bevacizumab, plus or minus atezolizumab according to the randomization. The accrual rate of this study was really very high. And this is, in my opinion, a hint about the huge interest of treating physicians that will explore the opportunities of immune checkpoint inhibitors in microsatellite-stable metastatic colorectal cancer patients in order to expand the reach, to extend the reach of immunotherapy in this disease. Of note, patients enrolled in these trials were not selected according to their MMR status. So both microsatellite-stable and microsatellite-instable patients were eligible. The characteristics of these patients were quite well balanced between arms, and most of them were in very good general conditions. And this is expected because they were exposed to quite intensive upfront chemotherapy regimen. Most of them had synchronous metastases, and most of them had negative prognostic features. The 7% and 6% of patients in both treatment arms had a deficient mismatch repair tumor. So those lucky guys where immunotherapy is extremely efficacious were a small percentage of patients included in this trial as expected based on the overall incidence frequency of this molecular alteration. In terms of safety, we did not find any kind of issue. So the use of the checkpoint inhibitor with this upfront therapy did not provide any problem in terms of tolerability and the trial met its primary endpoint of progression-free survival, showing the significant benefit from the addition of atezolizumab to first-line FOLFOXIRI plus bevacizumab in the overall study population. While no difference was reported in terms of tumor shrinkage that is known as objective response rate, that was not significantly different between treatment arms. However, as I previously mentioned, we did not know this at the beginning of the trial because the increase in the amount of knowledge about this topic was very quick in the last years. But we know that patients with deficient mismatch repair tumors would clearly derive benefit from immune checkpoint inhibitors, and we do have these patients enrolled in the trial. And actually, we found out a significant interaction effect between treatment arm and microsatellite status, showing a very, very important and impressive ratios for PFS of 0.10 among the limited, very limited, a small number of patients with deficient mismatch repair tumors. So we clearly found out that these patients do derive a huge benefit from the use of checkpoint inhibitors. But this was not so new and so impressive. What was, in my opinion, more intriguing was that if you look at the dotted lines of these curves, you'll see the curves of the study arms in proficient mismatch repair tumors. That is those tumors where we do not expect efficacy of immune checkpoint inhibitors. Actually, we do not see a clear separation of the curves as in the case of deficient mismatch repair patients that are those depicted by the continuous lines. The continuous red line shows a really impressive result with the addition of atezolizumab to FOLFOXIRI, bevacizumab in microsatellite-instable patients that are around 6% of the overall study population. But a signal of efficacy, of immunotherapy is there also in the much larger subgroup of patients with a proficient mismatch repair tumor. So the take-home message, the interpretation of the results of the AtezoTRIBE study in the future perspectives and in the overall landscape of metastatic colorectal cancer, is that by one side, the combination of immunotherapy with chemo combination could be a good idea in the small subgroup of deficient mismatch repair tumors. But however, we already know, it reacted very well to immunotherapy alone. But for the 95% of metastatic colorectal cancer patients, we could be able to identify some patients that may be optimal candidates to achieve benefit from these drugs. And so we run this in a subgroup analysis in the subgroup of patients with proficient mismatch repair tumors and found out that the 2 molecular markers were of special interest: the tumor mutational burden, which is quite expected, but it's only really a very, very limited percentage of patients; and Immunoscore-IC, which is an immunohistochemical evaluation of CD8+ T cells and PD-L1 positivity and their proximity, showing that the immunohistochemical marker may somehow be a surrogate of the immunogenicity of the tumor microenvironment. So a larger, a higher benefit from the addition of atezolizumab to FOLFOXIRI plus bevacizumab was evident in the subgroup of patients with Immunoscore-IC high. But we also had the chance, thanks to the collaboration with OncoCyte, to evaluate whether DetermaIO's [ 27-gene ] signature could be able to catch the intrinsic immunogenicity of a colorectal cancer's microenvironment and, therefore, to be helpful to select those patients that may derive more benefit from the use of immune checkpoint inhibitors. This is a burning question, especially in the field of proficient mismatch repair tumors but could be useful also to better identify the deficient mismatch repair tumors that do not need any other treatment but just probably a single shot of immunotherapy. We interpreted the results according to 2 different cutoffs: the traditional cut point predefined of 0.09 and another binary classifier that was explored on our series of samples for the first time. So first of all, when adopting the traditional cutoff, we found out that the 73% of samples from patients enrolled in AtezoTRIBE study, we were able to evaluate and to assess DetermaIO in 122 out of 218 patients randomized in the trial, which is a very good percentage. And IO was found positive in the 27% of cases, which is clearly an impactful percentage in the overall scenario of metastatic colorectal cancer patients. And IO positive versus IO negative tumors did not defer according to specific clinical and molecular characteristics that are known in the general clinical practice. In other words, the common characteristics that we already know about the tumor are not able to suggest whether it will be an IO positive or IO negative cancer. We did not find out any prognostic impact for IO both in the overall population and in the proficient mismatch repair group. But interestingly, we found out a close to significant p for interaction while looking at the predictive impact of IO with regard to the efficacy of immune checkpoint inhibitor and in particular with efficacy of the addition of atezolizumab to first-line FOLFOXIRI plus bevacizumab. In fact, while completely no effect from the addition of the immune checkpoint inhibitor was seen in the IO negative population, this was not the case for IO positive tumors where we had another ratio of 0.39, with a significant p value of 0.018. We ran the same analysis in the proficient mismatch repair subgroup that clearly is slightly smaller than the overall patient population in this study. But still, it's 110 patients. No difference again from the addition of immune checkpoint inhibitor to first-line therapy in IO negative. But on the other side, a significant effect of the use of the immune checkpoint inhibitor in the IO positive subgroup with a ratio of 0.47 and close to a significant p of [ 0.07 ]. Again, taking into account the lower power of this analysis as compared with the analysis in the general population, still close to a significant p for interaction. We also explored, as previously mentioned, a new cut point that could be more specific for colorectal cancer and, in particular, for metastatic colorectal cancer. And we identified that this new cut point could be able to better describe between those patients that may derive the benefit and/or not from the use of immunotherapy. With regard to the new data cutoff, here I am presenting the results in the proficient mismatch repair subgroup, IO positive tumors are seen to be associated with a slightly better prognosis in terms of progression-free survival. Clearly, when increasing the cut point, we also reduce the slice of IO positive tumors, that is now the 11% of the overall pMMR population. But again, we see a really significant p for interaction when looking at the impact of immune checkpoint inhibitors in the IO positive versus IO negative subgroup. So in conclusion, the AtezoTRIBE trial met its primary endpoint of progression-free survival but for sure, the immediate benefit from the addition of immune checkpoint inhibitor to first-line therapy in the overall patient population and, in particular, in the overall population of proficient mismatch repair metastatic colorectal cancer is not satisfactory by a clinical point of view. So we need to individualize our choice and to search for molecular markets that are predictors of benefit from these drugs. We have achieved interesting results with some of the markers that we have explored and, for sure, also with IO. The investigated IO score signature using the predefined cut point is able to predict the benefit from immune checkpoint inhibitors also in the field of colorectal cancer and also in the subgroup of proficient mismatch repair tumors that are those where the demand for the efficacious use of immunotherapy is more pressing. On the other side, the exploratory IO optimal cut point should be validated in independent cohorts of metastatic colorectal cancer patients to really test and assess its reliability. And I thank you so much for your kind attention. I am more than happy to answer any questions.

Well, thank you very much, Dr. Cremolini. I'm actually extremely excited about this data. And just to summarize for our lay audience, the opportunity here is in colorectal cancer and metastatic colorectal cancer. Immune checkpoint inhibitors have shown really fantastic response in a very small subset of patients, those patients that have genetically unstable tumors, those that are deficient in mismatch repair and/or called MSI unstable. So the real opportunity is to identify a subset of patients beyond those genetically unstable patients, which are only on the order of 5% of metastatic colorectal cancer patients. And so thank you for partnering with us to look at DetermaIO as a potential marker. It identified 27% of patients, which included the mismatch repair ones and showed similar to what Rob showed with triple-negative breast cancer the potential for prediction of response with the addition of immunotherapy to the chemotherapy backbone. And so that's now our second study in a randomized clinical trial cohort where we got a predictive result. So with that high-level summary, it's time for our fireside chat, if you will. And Dr. Brufsky, maybe I'll start with you. Can you take us down to the -- as translational scientist, we look at populations of patients and look at Kaplan-Meier curves. But really what this is about is helping individual patients and especially in triple-negative breast cancer and neoadjuvant triple-negative breast cancer. Can you put on your translational hat but also your clinician hat and explain to us what the choices are the physicians, oncologists and their patients face that could be potentially addressed by a test like DetermaIO?

Well, I think the interesting thing is that everybody now for triple-negative breast cancer over 2 centimeters will get neoadjuvant chemotherapy with pembrolizumab. It's now the standard of care based on a very large trial called KEYNOTE-522, which we can talk about using [ your testing ], if you want. I think that the issue with KEYNOTE-522 is whether to give the pembrolizumab for 12 months or the 6 months of chemotherapy. That's 1 issue. Number two, not everybody responds with pathologic complete response to pembrolizumab. Right now, the current estimate is about 64% of all-comers. PD-1/L1 does not predict. It's more prognostic. That is that you still get a benefit in the PD-1/L1 population both for chemo by itself and pembrolizumab. So there is a need, I think, to better select the patients for pembrolizumab, number one; and number two, whether to give them pembrolizumab for 6 months or 12 months. The reason for that is, a, the expense of the drug. It's about $100,000 a year. And if you could only give half, that saves $50,000 per patient. Way over -- I don't know what your test costs, but way over the cost of the test. It's number one. And number two, especially whether to give someone pembrolizumab, I think that a substantial fraction of women get hypothyroidism. That seems to be what happens quite a bit, and it's about 10% to 15%. And if you could avoid pembrolizumab in those women, and so think about some other therapy other than pembro, that would be a good option. So I think that there is a real need for something like this in the neoadjuvant space in triple-negative breast cancer.

Fantastic. And maybe Dr. Cremolini, the same answer -- or the same question to you. Can you put on your clinician's hat and talk about some of the struggles you may have with managing colorectal cancer patients and where you might see this test, like DetermaIO, playing a role?

Well, as I mentioned during my presentation, we clearly need biomarkers that benefit from immune checkpoint inhibitors in the proficient mismatch repair tumors. And here, we are speaking about the very vast majority of our patients that today actually do not derive benefit from immune checkpoint inhibitors and the use of these drugs is not indicated for their tumor. So if we want to have a chance of being able to exploit immune checkpoint inhibitors for these patients, we needed to individualize our treatment choices and probably to somehow develop targeted strategies for each individual patient based on the characteristics of their tumor. Up to date, precision medicine in colorectal cancer has been made for the vast majority of comprehensive genomic profiling to identify molecular alterations that were actionable by some specific targeted drugs. Also, with regard to immunotherapy in colorectal cancer, differently than in other disease. We need to individualize our treatment choices based on the characteristics of the tumor. But at this stage, comprehensive genomic profiling is not the answer, probably is not the only answer. We need to learn more about the microenvironment of the tumor. And mRNA signatures, such as DetermaIO, could stand as a really precious tool to do that in the frame, for sure, of the clinical trial but hopefully also for the clinical practice.

I have a question.

Yes, please.

I don't want to hijack your thing, but I have a question for Dr. Cremolini. I did a lot of work with bevacizumab in breast cancer about 15 years ago. And obviously, it was approved and disapproved by the U.S. FDA, and we were always trying to find some sort of biomarker for it. I was fascinated to see your slide where apparently, bevacizumab has some sort of immune effect, allowing T cells to infiltrate the tumor microenvironment and do some other things, mature dendritic cell function or whatever. And I'm wondering whether you guys would think about doing DetermaIO on some of the prior studies that got bevacizumab approved in colon as well as the ones for [ lung and ] breast. That seems like a really interesting idea because benefit in terms of immune checkpoint, you're probably not going to see that much of a difference, right? If you're adding immune stimulant to immune stimulant on top of it, I'm just curious about your thoughts about that. And again, Doug, I'm sorry. I don't mean to hijack your presentation here, but I'm just curious to see whether people have thought about that.

In your criticism is that we need the pre and the posttreatment samples in order to find out whether actually the treatment that we have administrated has been able to change the tumor microenvironment. And often this is not so feasible, except if you have it prospectively planned. But we have a small series of patients that were treated with chemotherapy plus the targeted agents and then underwent a secondary resection of their liver metastasis in colorectal. And we were actually able to see a change in the immune landscape of these tumors pre versus the posttreatment of the new properties of the tumor microenvironment after the exposure to the active drugs. Now I do not know if the question is specifically to bev. We do not have patients treated only with bev because bev does not work by itself. It needs to be combined with chemotherapy in breast cancer as well as in colorectal. But still, I think this is the way to move things forward to evaluate whether some combinations may be able to prevent and counteract intrinsic resistance mechanisms.

Yes. So thank you guys for that. So Dr. Brufsky, let me bring you back to that subject and DetermaIO. So one of the things in lung cancer that we've been trying to provide some evidence for is that potentially, DetermaIO can be useful not just for the question of whether or not an immune checkpoint inhibitor should be used, but also whether or not it should be used with cytotoxic chemotherapy. And perhaps, and I'm throwing this back to you in the case of breast cancer, whether or not the aggressiveness of the cytotoxic regimen is a consideration that DetermaIO, given that we understand that it's specific to the addition of pembro plus the cytotoxic arm, would this influence your choice of chemotherapy backbone in any way in triple-negative?

Substantial number of women still got adjuvant anthracyclines after having a pCR. But the question really is if you have a pCR, do you really need the adjuvant anthracyclines as well? The nice thing about it was only 6 months of bev during the chemo. Just to tell you, kind of things are -- what's old is new again. I developed [ carbo/taxel tier ] in 2000 actually. Part of it is the pembro. And I think that, sure, not yet. But I mean, I think one of the things that came out of that abstract is that when you add the IO, you do identify patients with higher pCR rates of 81%. And the theory is when you have a pCR, you probably don't need the adjuvant therapy, especially adjuvant anthracycline. But right now, the standard of care from KEYNOTE-522 is to give everybody [ carbo/taxel ] for 12 weeks, followed by AC every 3 weeks for 4 times with pembro every 3 weeks and then a year of pembro after or 6 months of pembro after. That isn't going to change upfront. I think that, obviously, if there's a biomarker that predicts for pCR, I think there will be a randomized trial where we use the biomarker to stratify those patients maybe to only give them [ carbo/taxel tier ] and pembro for 6 months. Maybe in specific patients, I would try using very specific ones. But right now, in the absence of a large-scale clinical trial, it could be tough to eliminate the AC at this point.

Yes. So that brings up the subject of the path forward both in colon cancer and triple-negative breast cancer and across [ the 6 ] different tumor types that we've demonstrated in association with response. So perhaps, Dr. Cremolini, you can put on your translational hat now and talk a little bit about the path forward and how we layer in evidence that really teaches oncologists how DetermaIO might be useful. How do you see that path forward?

Well, I am a big fan of clinical trials, and I definitely believe that we will need to have more evidence from prospective clinical trials. For sure, having an evidence that what we have observed in a clinical trial will work also with other drugs and in other trials would be useful -- will be useful, but also evaluating whether specific immunotherapies may work in the selected population of IO positive tumors will be very important, to me, to provide a good level of evidence to move things forward towards the daily clinical practice.

So Bob, you should guide me here and maybe trying to take some questions from the audience. And Rob -- while Bob comes onboard, maybe, Rob, you could answer the question -- I think one of the questions we often get is there's a lot of biomarker development going on. PD-L1 and TMB are already out there. Can you talk a little bit about what you think might make DetermaIO different from some of the other things that are in development?

Yes, I can definitely talk about that, and it actually ties in with a question we've already been supplied as well. So it actually says why do we think this test has a pan-cancer classification. And it all goes back to the fact that DetermaIO is meant to measure really multiple components of the tumor immune microenvironment. So when you look at PD-L1, I [ countered ] it up with 4 different antibodies, 3 different scoring methods and 4 different cutoffs. I think that comes out to be 60 different definitions of PD-L1 positive. And so there's a great -- when somebody says how do you do -- how does it compare to PD-L1, I always have to go, "Which one?" and I usually choose what is -- whatever is in the FDA label. But it's a very confusing biomarker, and the number keeps changing. Similarly, TMB, they have an FDA [ positivity ] of 10. And the moment they said 10, everyone came along and pointed out 10 is maybe useful here, maybe not useful here. One of the things is we've always kept our algorithm the same. We kept our cutoff the same. The reason why we feel we can do that is because we -- 2 things. Number one, we really are not looking at the tumor. One of our measures does look at the tumor after it's undergone a change called epithelial to mesenchymal, but that's something that all certain types of tumors do. The rest of them are really just looking at the microenvironment. And my analogy is if immune checkpoint inhibitors are calling in the calvary of your immune system, DetermaIO is telling you whether that calvary is going to be running over smooth grasslands or trying to go through swampland. We really are telling you, is the environment ready for an immune checkpoint inhibitor? And so I think that's really both our pan-cancer application and what makes us different from these other types of markers. So I think that really gets to the differences between why we don't have to change our cut point; why we're not changing our algorithm; why we're pan-cancer; why we're different from PD-L1, which is constantly changing its scoring system and its methodology; why TMB -- why we're different from TMB, which is constantly trying [ to decide ] whether it's TMB positive or negative. And I'll touch very quickly on these other points. When is DetermaIO going to be used within pharma trials? We're working with -- this week alone, we've actually reached -- working with 2 different pharma groups in terms of putting out -- just basically setting up what a pharma trial would look like for them. And then it actually talks about clinical trials. And in clinical trials, we are actually -- have got in, I would say, a clinical trial group, and OncoCyte have put together a proposal for the National Cancer Institute on how to use DetermaIO in a clinical trial.

Exciting. So my apologies to Tara because Tara's going to come in and moderate the question and answer, not Bob. So Tara, do you want to join us?

Yes. No worries, Doug. [Operator Instructions] So our first question comes from Mark Massaro from BTIG.

Can you hear me?

Yes.

Okay. Great. I wanted to start with questions to Dr. Cremolini and Dr. Brufsky. DetermaIO is a tissue-based test. Obviously, there are a lot of questions, at least in the investor circles, about tissue versus blood testing. So I wanted to get a sense maybe for what percentage of the time is it difficult to find the FFPE tumor tissue for your patients? Is it still a minority of the time? Is it a logistical challenge? Because I know a number of companies, including OncoCyte, are contemplating launching a blood-based test that does similar types of activity. I'm just curious to what extent tissue testing poses a logistical challenge versus the utility of a blood test.

I can start if you want. Definitely, I think that the answer is really different according to the different disease characteristics, I mean, the different histology and also according to the different health systems. For example, with regard to colorectal cancer, having a tissue sample from the tumor is very easy. There is no problem of the quantity and the quality of the sample. And it's very different in other situations, such as lung cancer, where often, having an adequate in terms of both quality and the quantity of material is not so easy. So by a technical perspective, it's easy. There is no problem. The problem may be more logistic because it may happen that this tissue is stored in a clinic or a lab that is different from the lab or the clinic where the patient is treated for the cancer disease. So in this case, it is -- there is a logistic issue in collecting the tissue in order to make it analyzed for each kind of biomarker. But for the vast majority of tumors, now we really need this tissue and the assessment of these biomarkers to inform our treatment choices independently of the use of immunotherapy. In metastatic colorectal cancer patients, we cannot choose the best first-line therapy without having some biomarkers analyzed in order to choose the best path for each individual patient. So to this end, collecting the tissue is a need not only in the frame or with the perspective of DetermaIO test, but for every kind of biomarker. And so in my opinion, it will not be a big issue to have the tissue to perform also DetermaIO on top of other markers that are absolutely needed for the daily clinical practice.

So in breast, it's a little bit different. For neoadjuvant therapy, everybody has FFPE, a core biopsy FFPE. So that's not an issue. So early-stage breast cancer, everybody has. For late-stage breast cancer, I'd say 50% of the time, we'll do a biopsy for a variety of reasons for tumor markers like HER2 and ER, make sure they're retained. We'll also -- some people will do it for next-gen sequencing. With triple-negative breast cancers, not everybody gets. So I'm not worried about that. I think I'm curious, given the fact that, as I think you're aware, a number of companies have developed whole exome/RNA sequencing now, Caris is one of them and there's a bunch of other ones that are doing it, like Tempus and some other ones, it'd be really interesting to overlay this assay on top of that. And I think that probably is coming -- I mean, I don't know what the DetermaIO guys are doing. But on the other hand, that's kind of, I think, where a lot of this is going to go for blood-borne testing. But nonetheless, I think at least -- I would -- if I had to estimate right off the top of my head, I'd say at least 40%, 50% of triple-negative breast cancer will get another biopsy, just to be sure the markers are still negative as well as getting next-gen sequencing on the biopsy. Early stages, [ it's 100% ].

I can actually answer that question because a lot of our data, we have 2 different ways we perform the assay. One is a very convenient real-time PCR assay that's most friendly to the lab, if you will. But a lot of our data has been performed on whole transcriptome RNA-seq data. And so a lot of the data is very much similar to the whole exome/transcriptome assays that are being run by Caris and others as well as ourselves. So it can be done both ways. And so as we -- as the field moves the whole genome analysis, that's probably not next year. It's probably a few years from now. We're in a very good position to run DetermaIO as part of that broader look at the tumor.

So our next question comes from Mike Matson from Needham.

So I have kind of a higher level question, I guess. I mean we've seen more and more of this data around DetermaIO, and it [ definitely ] seems to work in terms of identifying the patients that will benefit from these immunotherapies. But what really has to happen from here to -- in order for this to become more of the kind of standard of care for these patients to -- that this really is used to screen out the patients that would benefit from these drugs? And then how do you see this kind of playing out? Do you think -- it seems like, I guess, it probably happened kind of one cancer at a time, not, say, it can be a parallel, but like you need -- it wouldn't necessarily happen across multiple cancers at the same time in terms of guidelines and data and things like that. But I don't know if you can answer that, but that's my question.

Maybe, Dr. Brufsky, we can start with you from a breast cancer perspective but also a broader perspective as an extremely experienced translational scientist.

So from the work that's been done with Oncotype and MammaPrint in breast cancer that I'm sure a lot of you are familiar with, the early-stage multiparameter test, they were initially approved based on 2 reasons. One, it had a predictive effect for chemotherapy; and two, there were large amounts of retrospective data, like hundreds of patients that were from the ATAC trial, from NSABP B-20. There's all these -- and so that allowed the payers -- it gave enough evidence for the payers to pay for it. And so for that reason, that's why it took off. I think the same thing will have to happen here. I think that there is now in breast really I think -- probably it looks like 500 to 1,000 patients already where they have enough data on this. I suspect if they were able to do some of the larger clinical trials like -- just not saying which ones they're doing, but I'm just giving examples because I don't know, KEYNOTE-522 would be a great one. I think NSABP B-59. These are trials with 1,500 women in them. And if they're able to take a subset of those women and show predictive benefit similar to what they've shown in the smaller neoadjuvant trials, I think there may be enough evidence for the payers to pay for it. Again, I can't predict what they're going to do, but I suspect that's what's going to happen. I mean, clearly, we want a prospective trial. We all want prospective data. That's what happened with TAILORx with Oncotype. But the issue is that, that takes a while, 2 to 3 to 4 years, to actually generate the data. So I suspect -- if it were me, I suspect that they're going to go after some of the bigger trials. I don't know and I don't know [ if we'll have ] to tell you, but I think that is where I would go. And that will probably get approved from the payers if it does show benefit in that setting.

Yes. And maybe, Dr. Cremolini, I'll give you a chance to answer as well. But to prompt you a little bit that a lot of those clinical studies are controlled by big pharma. And in some cases, for instance, in the case that you presented, we're talking about market expansion, if you will, in colorectal cancer. Right now, immunotherapy drugs are used for 5% of that metastatic population, and we're talking about a marker expansion to 27%. So perhaps you can answer it with the hat of -- how the pharmaceutical companies see this and what's the potential for impacting them.

Yes, I definitely believe that the scenario may be quite different according to the different disease indications where immunotherapy is already used vastly for the treatment of these patients or not. But I cannot agree more with what Dr. Brufsky was saying about the need of prospective validation and the development in the frame of clinical trials. And so I don't know how long it will take. But definitely, to provide a high level of evidence, we need more prospective information.

Rob, do you want to comment a little bit on some of the internal plans?

Well, I did a little bit right there. We are already being written into some prospective trials right now. So we are being written into some prospective trials right now, and we are working very much to get access to retrospective trials that will be released -- that are randomized that are going to be released soon. So we are working extensively. So not so much further we can go with that on this call. The internal plans are access to prospective trials and also then access to other randomized trials that are already out there. So we are planning on doing both of those things. I do think when it comes to colorectal cancer, we're going to find a little bit more cooperation from the pharma groups because that is going to be a market expansion for them. So I do think there's going to be somebody who wants to really work with us there much more so than being scared that we're going to cut their market.

Yes. And the genomic health experience with Oncotype is very much on our mind. I'm old enough to have lived through that. Actually, Rob and I competed with them with an alternative product way back when. And the TAILORx trial took, what, 10, 12 years to report out, and there was plenty of use of the test as they continued to build layers of evidence that taught oncologists how to use it. And so we very much have that model in mind as we're pursuing [ evidence ].

And [ don't ] predict -- just as an example, [indiscernible] predict test. [ They still only got ] retrospective data, and it's approved by most of the major payers. And it's on the NCCN guidelines, just to let you guys know.

Yes.

Our next question comes from Thomas Flaten from Lake Street.

A question for the group, and this is a bit apropos Dr. Brufsky's question about bev. Is there a reason to think that multiple tests would be required if there's some remodeling of the tumor microenvironment depending on which intervention has been given and that you might see an IO negative become an IO positive over time?

Yes. Dr. Cremolini, do you want to start on this one?

Yes, I definitely think that it may be possible also because now we are starting to think about immunotherapy also in terms of PD-1, PD-L1, CTLA-4 because these checkpoint inhibitors are those that we are currently using in our daily clinical practice. But a lot of novel immunotherapy agents are currently on their way towards clinical practice or at least towards prospective controlled clinical trials. And so I definitely -- we'll definitely [ back ] that in the next future. Immunotherapy will not mean only PD-1, PD-L1. And it may be possible that having more information about the immune microenvironment may be useful to individualize the choice among several options. Clearly, being spoiled for choice is probably the future for those disease where immunotherapy sets a high magnitude of effect, for example, the case of melanoma or some kind of non-small cell lung cancer. But it will be more difficult to the scenario for metastatic colorectal cancer. Still, I do believe that having different angles or perspective towards the same thing, which is the immune microenvironment, may be useful to find out the best strategy to exploit the different treatment options to be efficacious in individual patients.

Dr. Brufsky, any comments?

No. I mean I think that I would agree. I think it's just -- I'd just leave it alone. I think I would agree with my colleague.

Okay.

Let me just say one thing. I mean I think that -- I think it's a fascinating idea, I think, that you would give sequential therapies. Maybe in the adjuvant setting, you could see what -- if IO -- if bev gave some like [ set ] and then pembro or whatever gave another [ set ]. So no, I think it's an interesting idea.

Our next question comes from Bruce Jackson from Benchmark.

Can you hear me okay?

Yes, we can.

Okay. Great. There's been a fair amount of discussion this morning about the importance of assessing the tumor microenvironment and some of the earlier tests like tumor mutational burden. We also discussed briefly whole exome sequencing. How -- what does a complete assessment of the tumor microenvironment look like for you? And how well does DetermaIO cover that assessment?

Yes. So Rob, maybe you want to chime in on this one, some of the work we've done really evaluating what is the biology that DetermaIO and how that potentially differs from some of the test, for instance, that Dr. Cremolini talked about in terms of looking at particular cell types in the tumor. Rob, do you want to comment on that?

Yes. So be a little careful with the word complete because I don't ever know that we'll ever know the complete tumor immune microenvironment. But what we consider to be an assessment at least as far as immune checkpoint inhibitors, we are really looking at 3 components. And that really is the ability of the tumor to attract the immune system, the ability of -- and that's, again, not the tumor itself. The tumor is being surrounded by inflammatory cells, which are attacking the tumor. That tends to evolve in 2 ways. It tends to evolve into immune cells whose job it is to suppress the immune system. And it's kind of like where you just are telling the immune system -- where the immune system is being told, lay off the tumor, leave it alone. It's just us. It's actually a very negative thing for the patient and a very positive thing for the cancer because the cancer continues to grow. So that's another component of what DetermaIO measures. We measure immunosuppression when they have kind of transitioned beyond this inflammation point to a point where the immune system is being told to lay off. And then Dr. Cremolini had a great slide up there, what's called the immune desert, and that is when the tumor itself has evolved to such a point that it just no longer looks like anything to the immune system. One of the ways it does that is it just mutates out all of the things that look like an antigen to the immune system. It just gets rid of them. That's one of the key ways it becomes immune desert. It just becomes nothing. So the tumor itself just stops looking like anything. It's still very dangerous because it's still cycling through. It's still growing. It's still being able to spread and metastasize. But relative to the immune system, it looks like nothing. So those 3 components, that inflammatory component we're measuring, that is a good sign that an ICI is going to work. But we're simultaneously measuring the fact that it's also suppressing the immune system or trying to suppress the immune system and also become an immune dessert. And my analogy is most of these other tests, PD-L1, TMB and numerous genetic signatures, are like bragging about how strong the jail cell is without ever asking the question, does the prisoner have a hacksaw or a file or has stolen the warden's key? Tumors are constantly trying to escape the immune system. And what really makes us different is we're not only measuring how strong the jail cell is, we're measuring whether the tumor is trying to escape and how well is it doing that.

Yes. Let me just add a little bit to that, that Dr. Brufsky and Dr. Cremolini haven't seen all the work that we're doing internally to characterize the biology that we're recognizing very efficiently with these 27 genes, and it's not published yet. But what I can tell you is that when we talk about a cancer-associated fibroblast signature, it's not just cancer-associated fibroblast. There are hematopoietic inflammatory type cells that are essentially sorting with those cancer-associated fibroblasts. So it's really quite complex biology. Same with the inflammatory side of the signature. It's not just CD8 cells that you can recognize with an immunohistochemistry marker. But there's a complex interaction between T cells, B cells and inflammatory type cells, macrophages, antigen-presenting cells. So it's really quite remarkable biology, frankly, that's been recognized by this relatively simple test, and we're starting to discover that's perhaps the key to why it works. It is assessing the tumor microenvironment in a way that seems to be very, very specific to whether or not you're going to respond to an immune checkpoint inhibitor. It's not just characterizing it. It's characterizing the biology that, as Rob says, is related to both the positive side, the ability to [ attack ] the tumor. But sometimes, the pathologic negative side, the inhibition that's come in that prevents you from attacking the tumor. We seem to have both of that, and it's very complex biology. So that's an exciting component of it. Dr. Brufsky or Dr. Cremolini, any comments there?

Yes. I think it'd be really fascinating. I mean there's a lot of other immune inhibitors and immune stimulators that are currently in development, as I think you all know. There's TIM-3 inhibitors. There's IDO inhibitors than prevent tumor exhaustion. Adenosine analogs that prevent tumor exhaustion -- or T cell exhaustion. OX40 monoclonal antibodies. There's all the stuff in there. And I think that it will be kind of neat to kind of use this to see if it predicts those benefits or not. And that will give you a hint at the immune microenvironment because all of these affect the different parts of the immune microenvironment. So I think guys like me are waiting to kind of see what happens, what these guys do and OncoCyte to kind of get an idea of what this is really predicting in the immune microenvironment.

Yes. And many of those targets are actually brought in by some of the biology that we're looking at. Many of those targets fall into the cancer-associated fibroblast type signal. And so it's a potential biomarker for response to those second-generation agents is kind of the terminology I would use. Other comments? Or should we go back to Tara? Tara?

Great. This concludes our verbal portion of the Q&A. I'll now turn the call over to Troy Williams of LifeSci Advisors to read the remainder of the questions.

Thanks, Tara. So just if you could touch on when will DetermaIO be available for pharma trials and eventually, clinical use?

Rob?

So again, pharma trials, like hopefully very soon. We are in discussions right now with 2 pharmas about that. Clinical use, it is being used on an early access program right now in specific use for non-small cell lung cancer, triple-negative and not yet, but we hope in bladder cancer very soon. So it is out there, and we're being very slow and very controlled about this. Early access means the physicians have to go through an education program, understand what the evidence says, understand what it doesn't say, but it is currently out there right now. As we continue to get more evidence, we hope to continue to expand.

Yes. The only other thing I'd add to that is that we've been aggressive about creating a kit form of this. We believe that, obviously, the use of this is very closely linked to the use of immune checkpoint inhibitors. And so if that's going to be useful to pharmaceutical companies, they need to be able to reach a worldwide audience. We obviously have launched this out of our CLIA LDT lab in the U.S. We can reach the U.S. market, but it really requires a kit to reach the worldwide market. And so we've gone ahead and invested in that. We've been public about that because that solves the problem for the drug companies that need their compounds to reach a worldwide market. And so that's also, we believe, will accelerate the time line with which we can work with our pharma and smaller biopharma partners.

Great. Thanks, Doug. So that concludes the written portion of the Q&A. Doug, I'll turn it back to you for closing remarks.

Well, so I really appreciate the clinical perspective that Dr. Brufsky and Dr. Cremolini have brought to us. I don't want to end without bringing it back to the patients. Dr. Brufsky pointed out that a neoadjuvant treatment of triple-negative breast cancer, tradition has been to use chemotherapy. Now immunotherapy has been added, but it brings with it some pretty serious autoimmune side effects. I think, Dr. Brufsky, you said 15% of folks come up with some thyroid disease. And so they are really critical questions for patients. The early-stage triple-negative breast cancer is a curable disease, and so that's a lifelong autoimmune disease that comes with the drug. So we are very committed to helping physicians make informed choices as to when that's going to be beneficial and when that might not be beneficial. And similarly, with what Dr. Cremolini presented in colon cancer, it's really a tragedy that only 5% of advanced colon cancer patients get access to immune therapy when now we have evidence that perhaps 25% or more of patients may actually have an opportunity for a strong response. And so that's the potential to really offer a life-saving therapy to a broader set of patients, and that's dear to our heart, both of these. I'm not a clinician. I'm a pathologist. But I have spent enough of my time sitting with patients that doing what's right for the patient is what we have in mind here. And so with that, bringing it back to the patients. I'll thank Dr. Brufsky, Dr. Cremolini and Rob for a fascinating session and thank the audience for your attention for this hour and 25 minutes. Thank you.

Doug, I think Mark Massaro from BTIG had another follow-up question. Mark, you may go ahead.

Sure.

Yes. So I guess, Doug, I think you've now shown data on -- I believe it's 6 different indications with metastatic colorectal cancer. My understanding is that you're planning to submit your dossier for CMS reimbursement sometime in the next, I don't know, month or so, maybe quarter or so. Can you just remind us what your reimbursement plans are for DetermaIO? And when you -- like when do you think you can get reimbursement for this for clinical use? And should -- are we still -- are you still thinking that -- of a pan-cancer or multi-cancer claim? Or do you think it's more realistic to kind of go cancer by cancer?

Well, so that's a great question, Mark. And yes, we do plan on -- we've already started a conversation with CMS, and CMS has come up with this very innovative approach to looking at reimbursement of tests where they have -- they issued what's called a local coverage decision where, for instance, a class of tests, in this case, it would be evaluating a diagnostic tissue specimen for likelihood of response to immunotherapy. They issued that LCD, and then they give reimbursement in an indication-by-indication way. So we will be having that conversation, as you say, this next quarter with CMS saying -- suggesting and working with them to create a local coverage decision that would cover tests that evaluate the tumor immune microenvironment. And then we will be talking to them about indication by indication. Our most advanced data is in neoadjuvant treatment of triple-negative breast cancer. I think some in the audience will know that we have some really nice data in non-small cell lung cancer. And I can tell you that we'll be showing more data on non-small lung cancer this year. In colon cancer, we, to date, have the one study done by Dr. Cremolini. My suspicion is they'll want that confirmed by at least another study. But in that mode of getting a local coverage decision that says, hey, CMS is very interested in reimbursing 4 tests that evaluate the tumor immune microenvironment and have data to support the use of immune therapy drugs either by themselves or in combination with adding chemotherapy, and we will be progressing indication by indication using that mechanism. And so that's the plan.

Doug, any further concluding remarks?

No. I think I did it. Hopefully, brought back to patients, which I think is always important. And once again, thank you for your attention, to the folks in the audience, and thank docs for talking about the patient issues and offering us your perspective as very sophisticated translational scientists. So thanks.