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

Hello, and welcome to today's webinar. My name is Sara Riordan. I'm a Genetic Counselor and the Director of Medical Education at Oncocyte. We are very fortunate today to be joined by leading lung cancer experts to discuss brand-new data regarding molecular risk stratification in the context of EGFR testing in early-stage lung cancer. This webinar is being recorded, and we'll make the recording available after the webinar. Now I'd like to introduce our speakers and panelists. Our first speaker is Dr. Gavitt Woodard, a thoracic surgeon and Assistant Professor at Yale University. She received her medical degree from Stanford University and completed her general surgery residency and cardiothoracic surgery fellowship at UCSF. Dr. Woodard is the recipient of an American Association of Thoracic Surgeons Robotics Fellowship Award as well as the chief resident achievement award from UCSF. Her research focuses on the genetic drivers of early-stage lung adenocarcinoma, the role of systems genetics in cancer development, small molecule drug development and preclinical testing of novel therapeutics. Our next speaker is Dr. David Gandara, Professor of Medicine Emeritus at the University of California, Davis. Dr. Gandara is a world-renowned medical oncologist and recognized thought leader in lung cancer and has authored over 700 articles, book chapters, abstracts and editorials. He is a past President of the IASLC and a prior board member of ASCO. His research interests focus on developmental therapeutics of new anticancer agents as well as preclinical modeling and clinical research in lung cancer. He is the principal investigator on an Early Therapeutics award from the National Cancer Institute, where he leads an interdisciplinary team of clinical oncologists, pharmacologists, molecular biologists and statisticians in developing new anticancer agents in a variety of novel drug classes. He also leads a multispecialty team in the Southwest Oncology Group, an NCI-funded national clinical research organization, in studies related to improving therapies for lung cancer and developing predictive biomarkers of therapeutic efficacy. We are also fortunate to be joined by Dr. Edgardo Santos, who will be joining in the panel discussion at the end of the presentations. Dr. Santos is a medical oncologist with Florida Precision Oncology and specializes in thoracic oncology and head and neck cancer. He is also a Clinical Affiliate Associate Professor at the Charles E. Schmidt College of Medicine at Florida Atlantic University. Welcome to those of you just joining us. For today's webinar, all attendees are muted, and only our speakers and panelists are on video. [Operator Instructions] We encourage you to submit questions at any time during the presentation, and we'll address as many questions as we can during the panel discussion at the end. Dr. Woodard is going to start us off by giving some background about the DetermaRx test and presenting some new data that was just featured at the IASLC North America Conference on Lung Cancer. Take it away, Dr. Woodard.

Hello. I'm Gavitt Woodard. I'm an Assistant Professor at Yale and a new thoracic surgeon here. Next slide. I have no financial relationship with OncoCyte, and I don't have any financial relationship to this technology. I participated in all of this research while I was at UCSF, and I believe in this assay's ability to predict recurring lung cancer and to cure some of our early-stage patients. And that's why I'm happy to be giving this talk to you today. Early-stage lung cancer remains a very deadly disease. We know that there are over 40,000 patients who are diagnosed with early-stage lung cancer in the United States each year. And despite having the best of care and a complete surgical resection, 30% to 50% of these very early-stage patients will recur after surgery and ultimately die of lung cancer. And this is very different from a disease like breast cancer or colon cancer, where we think of the stage 1 patient population as a very curable group of people. And so in lung cancer, we still struggle to cure even our stage 1 patients. Next slide. Well, an analysis of the IASLC database using the eighth staging edition of the TNM classification shows better overall survival rates than prior analyses and giving early-stage lung cancer survival rates from 80% to 90%. When we try to externally validate this data using the National Cancer Database from the American College of Surgeons, the eighth edition TNM stages shows only a 66% to 74% overall survival at 60 months for stage 1 lung cancer. Next slide. There's been a lot of data to show that the cause of death in early-stage lung cancer is typically due to occult metastatic disease at the time of surgery or diagnosis. Next slide. That one. Okay, forward. Adjuvant therapy has been studied in the early-stage patient population with randomized trials. We've shown that adjuvant cisplatin-based chemotherapy results in increased disease-free survival and overall survival in certain patient populations. However, when we look specifically at the stage 1 patients, we tend not to see this benefit. This is probably because a lot of patients are cured with surgery alone, and it's hard to select the stage 1 patients who are most in need of adjuvant chemotherapy and who will stand to benefit from it. Conventional staging criteria are based mainly on tumor size and not tumor biology, and it doesn't tell us which of these patients with the smallest tumors are at the highest risk. And so when we treat all-comers with chemotherapy, we don't see the benefit in these randomized trials. Next slide. The NCCN knows that data don't support giving therapy -- giving adjuvant chemotherapy to all early-stage lung cancer patients, but they do know that some of these patients are likely to recur and die. So some of them must harbor occult distant metastases. But the question is, how do we determine who is at most risk? The NCCN leaves it up to the clinician to decide which patients are at risk and if there's some examples of features that could be consistent with patients who harbor metastatic disease. And these are things such as poorly differentiated tumors, pleural invasion, lymphovascular invasion. But none of these have been extensively proven, and their prognostic features have never been shown to be predictive of chemotherapy benefit. Next slide. DetermaRx is a genomic test that's performed on formalin-fixed, paraffin-embedded tissue, the same blocks that all patients get when their tumors are sent to pathology. It comes from patients who have been recently surgically resected for early-stage non-squamous non-small cell lung cancer. The results would stratify patients into low-, intermediate- or high-risk of recurrence. And using those results, surgeons and oncologists can use them to tailor the treatment and surveillance strategies of these patients. And the thing that's important to note is that the test is performed on surgical specimens, not from biopsy. And it has been approved for Medicare for patients with stage 1 to 2a non-squamous non-small cell lung cancer. Next slide. The test employs highly accurate, clinically reproducible PCR technology to measure the expression of 14 genes. 11 of these are cancer-related genes, and 3 of them are reference genes. And the 11 cancer genes were identified from a large pool of cancer genes. These were -- 200 cancer genes were selected for their known association with lung cancer or other malignancies, and a large number of tumors from UCSF's tumor banks were analyzed to select, with the computer 11, genes that were the most important drivers of cancer metastasis. And these genes were selected by algorithms only, and no bias was used when we selected the 11 cancer genes to be used in the [indiscernible] assay -- the DetermaRx assay. Next slide. The assay is developed analyzing the RNA expression of more than these 200 cancer genes in a 361-patient training cohort. The cohort included all stages of disease. So it was inclusive of all types of biology with the goal of training a highly robust discriminant assay. The 14-gene assay algorithm that was developed from this training cohort was unlocked, including the cutoff points. And the algorithm has been validated on 2 subsequent data sets, 1 from Kaiser in Northern California that included 433 patients with stage 1 disease and a second cohort from the Chinese clinical trials consortium that consisted of over 1,000 patients with stage 1, 2 or 3 disease, and this was all independently blinded validations. Next slide. This is the Kaiser cohort that included only stage 1a and 1b patients. The curve on the very right shows the subset of patients in the Kaiser cohort who would have been characterized as low risk based on NCCN prognostic features. This group included all patients with stage 1a disease and a subset of patients with stage 1b disease. Over 1/3 of these patients had high-risk biology with only 50% overall survival at 5 years. Next slide. The Chinese clinical trials consortium cohort included patients with stage 1, 2 and 3 disease. Their overall survival decreases with increasing stage. But with each of these staging risks, the molecular assay is clearly able to distinguish the high-risk patients who have the poorest survival within each stage group. Next slide. We also performed a subset analysis of the same Kaiser and clinical -- and the Chinese clinical trials consortium cohorts that was published in JAMA. This subset analysis looked at patients with the very smallest tumors, less than 2 centimeters, which corresponds to stage 1a1 and 1a2 by the eighth edition criteria. And in the Kaplan-Meier curve on the right, there's still a very significant separation in overall survival between low-, intermediate- and high-risk groups, meaning that DetermaRx can identify patients with high risk of recurrence and death even in the very earliest stage of the disease and in the smallest node-negative tumors. Next slide. What I'm going to present on now is data from the poster that we just had presented at the Northern American IASLC conference, all virtual now. But the important part of this poster is that we then took our assay and used it in a real-world clinical setting at UCSF prospectively to determine which patients would benefit from adjuvant chemotherapy in our patient population. Next slide. We have the publication and submission. And the important part of this study that we want to address are validating and demonstrating that the DetermaRx assay can work prospectively to identify these patients and not just as a retrospective validation. Secondly, we wanted to show the ability of DetermaRx to identify the high-risk patients who will benefit from adjuvant chemotherapy. And thirdly, given the increasing use of EGFR-tyrosine kinase inhibitors in the adjuvant setting, we wanted to look at the impact of EGFR mutation setting on DetermaRx risk stratification and how that might interplay with adjuvant chemotherapy benefit. Next slide. So this is our study design. We took 250 patients from the UCSF. They were stage 1 to 2a, and they all underline a complete surgical resection. After surgery, their blocks were sent to pathology. And from pathology, they then went to be tested with the 14-gene biomarker assay. Those results were back to us in the clinic in real time so that we can discuss them with the patients during their postoperative visits. Low-risk patients were thought to be at low risk of recurrence and were -- just underwent surveillance. But the high-risk patients and the intermediate risk patients were sent for discussions with our medical oncologists as to whether or not they might benefit from adjuvant chemotherapy. And it's important to note that this was a nonrandomized study. So not all of our intermediate- and high-risk patients were randomized at therapy. However, a good number of them did agree to undergo adjuvant chemotherapy despite the fact that they may not have otherwise qualified for it based on NCCN guidelines. Next slide. This is on our paper that we published a few years ago, looking at the difference between DetermaRx risk stratification and NCCN guidelines criteria. And the main takeaway from it is that there are differences between who would be recommended to undergo adjuvant chemotherapy if you just use the NCCN guidelines or if you were to use DetermaRx risk stratification. So there's not direct overlap between these tests. And the other takeaway is that DetermaRx does a better job at separating out patients who are at risk for recurrence. So the Kaplan-Meier curves on the left show risk stratification using DetermaRx, and the Kaplan-Meier curves on the right show risk stratification using NCCN criteria. Next slide. This is their data demonstrating the survival of our early-stage patients. And you can see that in the blue line, patients who are molecular low risk of recurrence had very low recurrence rates, about 5%. However, the patients that were determined to be high risk of recurrence, those who did not receive any adjuvant chemotherapy, had survival rates around 70% in 5 years. And for the same group of high-risk patients, patients who agreed to undergo adjuvant chemotherapy, their survival rates were excellent, around 95%, showing a huge benefit to giving adjuvant chemotherapy treatment to molecular high-risk patients. Next slide. This is a focus chart looking at just stage 1a patients alone. And you can see here that these differences persisted. So low-risk stage 1a patients have very low rates of recurrence, but we still see very high number of recurrences in stage 1a patients who are molecular high risk. And the survival of these patients is not very good. However, if you take these patients and treat them with adjuvant chemotherapy, we have had no recurrences in this group of patients to date. Next slide. This is the recently published ADAURA study from the New England Journal. Many of you are probably familiar with this study. But to quickly recap it, for those who aren't, they're just patients who were stage 1b, stage 2 and 3a and treated them with either 3 years of adjuvant osimertinib or placebo after surgical resection. All these patients had an EGFR mutation. This patient population is a little bit different from the patients in our studies, but I'll highlight some of these differences just because there have been questions about this. About -- only 1/3 of their patients were stage 1 in the study. So it really focused on a more advanced patient population. 60% of their patients had nodal disease at the time of surgery. Again, this is a group that we don't focus on. Our assay is used for node-negative patients. And then 60% of their patients also received adjuvant chemotherapy. Next slide. So even if you were to apply the study, we still need ways of selecting which patients we are going to treat with adjuvant chemotherapy, even if you're thinking about an EGFR mutant patient population. Next slide. We were able to obtain next-generation sequencing data on 150 of the patients in our 250-patient data set. We then took this data, and we were able to retrospectively evaluate the impact of EGFR mutation status on being molecular high risk or low risk and determine the interplay between these 2 things and then evaluate the benefit of patients who were treated with adjuvant chemotherapy in our study. Next slide. On the left of the Kaplan-Meier curve are patients who were EGFR wild-type, and on the right is the Kaplan-Meier curve of the patients who had an EGFR mutation. And though the numbers are small, you can see that clearly, we see the same sorts of Kaplan-Meier curves and the same sort of curve separation for both EGFR wild-type and EGFR tumors. And the take-home message is really that EGFR mutation status and DetermaRx are completely different risk stratifiers, and they're completely independent of each other. And that EGFR mutant patients who are DetermaRx high risk will still benefit from adjuvant chemotherapy, and EGFR wild-type patients who are DetermaRx high risk will still benefit from adjuvant therapy. Next slide. We have a prospective randomized trial that has about 20 clinical sites that have been selected, and patient enrollment has begun. The data that I presented is all from a UCSF trial that was prospective but was not randomized. And so we're hopeful that having this randomized trial data will provide additional support for the use of our assay, though we find the results from UCSF very convincing. Next slide. So in summary, there are a lot of patients who have early-stage surgically resected lung cancer who are not cured with surgery alone. And identifying the patients who stand to benefit from adjuvant chemotherapy remains a question for surgeons and oncologists. Using the DetermaRx risk stratifier can help identify patients who are very low risk of recurrence and don't need any adjuvant treatments. And it can also identify the patients who are intermediate and high risk who do stand the benefit from adjuvant chemotherapy. Thank you.

Thank you so much, Dr. Woodard. We're passing the proverbial mic on to you, Dr. Gandara.

Okay. Thank you, Sara, and thanks, Gavitt, for a very nice presentation. I am tasked today with updating very recent information regarding adjuvant and neoadjuvant therapy for early-stage non-small lung cancer because the DetermaRx data that you just saw, so nicely presented, did not occur in a vacuum. And today, in October 2020, if you treat lung cancer, early-stage disease, there are other things you need to consider. So that's what I will address today. All of this really pertains to personalized or precision therapy. And if you think about DetermaRx, for 25 years, and I've done several studies in this space, we've been trying to find the markers that tell us who should benefit most from chemotherapy, both in advanced-stage disease and early stage. But what else is on the table now? And if you can advance. These are my disclosures. I will just point out that Sara inadvertently had an older bio sketch of mine. So I am the past Chair of the SWOG Lung Committee. I'm afraid if Dr. Karen Kelly sees this presentation, she will think that there's been a coup and I've come back and taken over. But that's not the case. She is now the Chair of SWOG, and I am a consultant for OncoCyte. Next slide. And I'd start with this figure from a publication a few years ago by Mary Redman, our Head Statistician for the SWOG Lung Group -- for the Lung Committee, where we talked about the paradox and the application of predictive biomarkers. And this is highly relevant to the data that Gavitt just presented. So as you can see at the top, in terms of how often and what sort of application there is with predictive biomarkers, it is the highest in a stage 4 disease. And if you can click once. On the other hand, the potential for cure is greatest in those patients with early-stage disease. And yet, we had essentially, what we've had until recently, no biomarkers to tell us how to personalize therapy for stage 1. Advance and advance again. Now I'll just follow up to Gavitt's presentation because this will be pertinent in terms of my discussion of neoadjuvant and adjuvant therapies that although there have been only a few direct comparisons of neoadjuvant chemotherapy versus adjuvant, there have been several meta analyses. And these have shown that the results overall are fairly equivalent. But you can see when given in an empiric manner, so not using a biomarker such as DetermaRx, that the hazard ratios for benefit in survival are modest at best. Advance. And I think this is best illustrated in terms of our current dilemmas by quickly just showing you a case. This is a patient with about a 4-centimeter right upper lobe mass. It's an adenocarcinoma of the lung. There's no other sites of disease. There's no enlarged hilar and mediastinal lymphadenopathy. The patient undergoes a right upper lobectomy, mediastinal dissection, and all of the lymph nodes are negative. Advance. Well, by the staging system that was then in place from 2013 to 2016, which is the seventh system, this patient is stage 1b. The reason this is important is the staging information that Gavitt showed you was largely collected in either the sixth staging system or the seventh, and all of these are derived by IASLC and adopted by staging committees all around the world. So when we talk about, would you give chemotherapy in a stage 1B, and you know the data were generated in a different generation of staging, it makes all the difference in the world. Advance once more. Because this same patient, this identical patient in the eighth staging system, which is what we use today, would not be 1b. This patient would be a stage 2a. So when I present this in our medical conferences and I present it as a stage 2a, everybody says, "Well, of course, I would give adjuvant chemotherapy." If I presented as a stage 1b, they said, "Well, maybe I wouldn't." All of this is what we'd call stage migration. Advance once more. And just to say that during the time when most of the clinical trials were done that Gavitt showed you, we were actually using the sixth edition. And this patient would revert back to stage 1b. Next slide. And Gavitt has already emphasized this, so I don't want to dwell on it. Other than to say that this is the publication from IASLC when they began to introduce the eighth system to show what the stage changes were. If you can advance once. So you can see this stage migration with the patient now becoming stage 2a. And advance once more. And as Gavitt emphasized, even using the eighth system, you can see the long-term survival, 5-year survival in stage 2a, the same patient we talked about, only 60% if chemotherapy is given empirically. Next slide. All right. Well, what's new? Let's do chemotherapy. You've heard about the DetermaRx. What's new and targeted therapy? What's new in immunotherapy? Please advance. Well, one master protocol being done in the United States is this one, it's ALCHEMIST. So this is taking early-stage surgically resected patients. You can see the stage distribution, complete resection. It's doing molecular testing. And if the patient has an EGFR mutation in a cancer, they are randomized to erlotinib or placebo, same sort of randomization for ALK, the crizotinib or placebo. And initially, there was nothing for immunotherapy. If you can advance once. But now we have an almost completed trial called ANVIL, in which patients who do not match for EGFR and ALK are randomized to adjuvant nivolumab or placebo, no biomarker required. Advance. But what else is new? So Gavitt already introduced the ADAURA study, a landmark study published in the New England Journal. And you know the randomization here in patients who had surgically resected disease. They have 1 of the 2 most sensitizing radiation -- I mean, EGFR mutations, randomized to osimertinib or placebo. One key difference from this study to all the other adjuvant trials in the EGFR space is that osimertinib was given for 3 years, not 2 years. This is important in interpretation of the data. The primary end point, disease-free survival. It was the investigator choice whether to give adjuvant chemotherapy. Now some of this is based on stage. Some of this is likely based on comorbidities that would have precluded chemotherapy or patient choice, but it is not uniform throughout the study. I think this is a global study, and that's explainable. Next slide. And of course, these are the data that were presented at the plenary session at ASCO, updated at ESMO, published in the New England Journal, dramatic improvement in disease-free survival with adjuvant osimertinib. You can see we almost nadir in our oncology hazard ratios like this, 0.17. And you can see essentially all of the groups benefited from adjuvant osimertinib. And you can see stage 1b, a little less. And if you can go to the next slide. We see here the disease-free survival by stage, greatest benefit, as might have been predicted in stage 3a. Intermediate benefit in stage 2, still very dramatic, though. And stage 1b, statistically also positive for disease-free survival. And you can see the 2-year rates in the table below. You can see the hazard ratios. Again, all of these, 0.5 for stage 1b. Next slide. Well, the question, of course, is overall survival. And as shown here, these data are very immature. In fact, it was my recommendation that they not even show these data, but they did. And so I don't think we can make much of knowing whether osimertinib will actually cure patients. One thing in particular, remember the drug is given for 3 years. That may be important. If it's not cured, then an extending delay of recurrence. Now can you go to the next slide? Okay. Go back. I didn't put this slide in, I should have. The study was updated at the ESMO meeting in regard to brain metastases because that's very common in patients who had surgical resection. Early-stage EGFR-mutated lung cancer patients are terrified of getting brain metastases. And the osimertinib in this trial, ADAURA, markedly decreased the incidence of brain metastases. To me, that's an additional advantage. So this is not FDA approved yet. My impression is that it likely will be. And especially the data on brain metastases, even though we don't have the OS information yet, would lead me to use this in most patients. And I think as Gavitt has explained to you, we now have a way to interface the DetermaRx with EGFR testing. And maybe we'll come back to this in the discussion. Next slide. Well, what else is happening? What's happening in adjuvant immunotherapy? Here, we have no data yet. But as you can see here, multiple randomized trials, a lot of these trials not using any sort of biomarker, not even PD-L1, for entry into the trials. And you can see that by and large, the primary end points are disease-free survival. So no data yet. Next slide. Well, I'm going to close just by talking a bit about neoadjuvant for the data from much earlier. But there's a lot of fanfare, and some of this came from this very nice publication by Forde et al. now 2 years ago, giving a small group of patients, only about 20, 2 doses of nivolumab prior to surgery. What they saw, as you can see in some of the examples, is there was a poor correlation of radiographic response to subsequent major pathologic response. So in many of these cases, no chemo, 2 doses of nivolumab, the lesion didn't shrink, but there was a major pathologic response. And you can see that occurred in almost half of these patients, where only 10%, 2 patients, had a radiographic response. So this meant that biologically, these drugs, checkpoint inhibitors, can induce major biologic effects. Next slide. Well, since then, there have been several other trials, and they have not been as positive. This is just some of them. And in fact, some had shown that you lose about 10% of patients to surgery who are highly curable because they either progress, they have an immune-related side effect, they have something that makes them not able to go to surgery. And you can see the major pathologic response, much, much lower than what was reported in the initial nivolumab study. Next slide. And here, I'm thinking I lost a slide or 2. And what I was going to show you was more recent data regarding new adjuvant chemotherapy, plus checkpoint immunotherapy, several small trials, some of them showing higher-than-expected pathological response and only published last week, the data from the Spanish group. The first of these studies, neoadjuvant, checkpoint immunotherapy, atezolizumab and paclitaxel carboplatin, showing 77% of patients with stage 3a resectable disease who have not progressed at 2 years, 24 months. So all of this is still research in my book. It's not standard of care, but there's a lot of other things going on. And I think we'll stop here, and we'll go to the panel discussion. Thank you.

Wonderful. Thank you for you fantastic presentations, Dr. Gandara and Dr. Woodard. I'm excited to open up the panel discussion at this point. We've seen some great questions come in. [Operator Instructions]

And I'll start by asking some questions that have already come in during the presentation to our panelists. So considering that about half of the patients in the ADAURA trial received chemotherapy, how can EGFR and DetermaRx results be assimilated in terms of treatment decisions? If a patient has both EGFR mutation and is classified as high risk by DetermaRx, then what is the treatment algorithm by which a clinician should decide whether to administer osimertinib first, chemotherapy first or both? So Dr. Santos, maybe I'll throw this one to you as a medical oncologist and then would love for Dr. Gandara and Dr. Woodard to weigh in as well. You're on mute.

Thank you, Sara. Can you listen? Yes. Yes. We appreciate that question and great presentation by Dr. Woodard and Professor Gandara. Thank you both of you. So honestly, when we think what is available right now and what is in the guidelines or what has been approved by the FDA or by our Medicare and CMS because everything is complex, I will say the following. This DetermaRx, as shown by Dr. Woodard and her group, that when you are on the high-risk DetermaRx, that they show clearly has a better outcome in terms of the disease-free survival, which at the end of the day, was also the primary end point of the ADAURA trial. And I remember the presenter of the ADAURA trial, Dr. Herbst, I mean, that for him, there is still, for early-stage lung cancer, adjuvant chemo is still a standard and an important part of the treatment. So if I have a patient with DetermaRx high, I will offer that patient chemotherapy. The issue of the EGFR mutation is a separate issue. If I found that the patient has an EGFR mutation positive, I will offer the patient osimertinib, as Dr. Gandara once presented. And I agree with him that this is going to be a clinical practice change. If the patient has a positivity, we call, improve the disease-free survival, you see a hazard ratio of 0.17, very striking result and then the data of also keeping those patients away from having a recurrence in the CNS. So that's what -- how I will approach a patient in that sense. So I don't know, Dr. Gandara, you have something to add on these or Gavitt?

Thank you, Edgardo. I agree with you 100%. And just remember, the way ADAURA was done and the way the ALCHEMIST trial is being done in EGFR mutated cancers is if that patient is a candidate for chemotherapy, they get the chemotherapy first and they complete it. Typically, that's 4 cycles. Then they go on to the EGFR-TKI. So again, now we have the DetermaRx, so that may factor into who you give chemotherapy to for very early-stage disease. But for some stages, like stage 3a resected, we're going to do it in everyone. So I would sequence it. I wouldn't move the EGFR-TKI concurrently with the chemotherapy because there can be some interactions there that may decrease efficacy. I do it afterwards because that's what the trials did. And again, like you said, I think the brain recurrence data are important. And Gavitt, did you have comments?

I agree with both of you that I think they're completely separate issues to treat with adjuvant chemotherapy first. And then if the patient is a good candidate for EGFR-tyrosine kinase inhibitors, meaning more advanced stage and they have an EGFR mutation, that those patients could go on to a long-term TKI adjuvant therapy.

I also would like to add something, Sara and colleagues, is that if we go by the book and depending on how the FDA is going to approve the use of osimertinib in the adjuvant setting, if they go by the data, it will be for a stage 1b plus surgery. But I think that David and Gavitt, we can disclose a little bit more about the stage 1a because the prospective data that Dr. Woodard presented on stage 1a for many years, and Dr. Woodard has more experience than me by far, we only have this kind of stage 1a disease, in which we believe that those patients who had tumors, certainly, the data that Dr. Woodard and her colleagues have developed is very striking because while we have work to do with 1a, most likely, we may not be able to give osimertinib to those patients. But certainly, we need to rectify the stage [indiscernible] with the stage 1a or high risk, that they should get benefit from adjuvant chemotherapy.

I think our imaging base has become so good, and we're catching more aggressive tumors at an early stage. So that might explain some of the reason why some of our stage 1 tumors are so aggressive and recur postoperatively. I think that you can't think of a stage 1a tumor as a benign, curable tumor. You need to approach them very aggressively. These patients, in general, don't do as well as we expect them to.

I'll just add that -- and I don't think you showed this, Gavitt. Correct me if I'm wrong, but the guidelines from organizations like NCCN have been, I don't know what the proper word is, but they have said, if patients have stage 1 disease, you should consider giving adjuvant chemotherapy if they have high-risk factors.

For stage 1a, they say observe only. And then it's 1b and the 2as where they gave some vague guidelines.

They only changed their guidelines 2 months ago, Gavitt. Some of us are still living in 2020 August.

This summer.

And they changed it not based on any data. They made an arbitrary change. But they still say at the end of the bar for stage 1a, they see adjuvant therapy. That's what's in the guideline. When they do that, they say, consider adjuvant therapy -- and let's even just say for stage 1b because you remember, the staging system has already migrated to later stage, what they say is, consider if the patient has vascular invasion. Okay, that seems reasonable. They also say, consider if it's poorly differentiated. There is no data to support that in early-stage disease, that you should give adjuvant chemotherapy. They may have a worse prognosis. That doesn't mean chemotherapy is going to work better. They say if the patient didn't have a good mediastinal dissection, give it. Well, the answer there is the patient should have had a good mediastinal dissection. You don't throw blindly chemotherapy at them. And then the other thing, the most egregious, is they say, if the patient had a wedge resection, give chemotherapy. Well, a lot of times, the patients had wedge resection, you know, Gavitt, you're a surgeon because they had some semisolid ground-glass thing. It's probably the least likely to kill them from distant disease. So DetermaRx has got to be better than what the current guidelines say. Anyway, that's my opinion.

Well, David, I completely with you. But again, the data has been collected for so many years. But Dr. Woodard and many others at University of San Francisco and then validated, as you well said, Gavitt, by Kaiser, a cohort, the big trial from the China consortium, more than 1,000 patients, retrospective and then prospective, maybe I don't know about you, Dr. Woodard. Again, clearly there, the data is very compelling and consistent. So I agree with Dr. Gandara.

All of you guys have mentioned the idea of really sequencing -- for EGFR positive patients, sequencing chemotherapy prior to osimertinib therapy. Do you see a role in giving adjuvant chemo prior potentially preventing patients developing osimertinib resistance by eliminating minimum residual disease at an earlier point?

Well, I guess I'll start. I don't know. That's a great question, Sara. But unless we had some trial to sort that out, I mean, I certainly wouldn't -- well, let's say, number one, we have to see if ADAURA gets approved. I'm talking about the United States. If it's approved, is it approved for all 3 stages? I neglected to mention that the primary end point in ADAURA was disease-free survival in stage 2 and stage 3, not all 3. But who knows? They may get approved for all 3. But I think the point is that we would not think that chemotherapy necessarily would have eliminated an EGFR mutant clone a priori. We'd still sequence them, I would at least, and give platinum chemotherapy. And remember, unless the patient absolutely can't take it, it's cisplatin. It is not carboplatin. We don't have one study in the entire literature that shows improved survival with carboplatin. We had a CALGB trial, which was published, said OS is better. Followed up for another 18 months, statistically no improvement, no long-term improvement, cisplatin.

I agree. I would agree also with that, and that will require a clinical trial to answer that question. And the question was that also with crizotinib, to eliminate that clone will require the use of a [ story like CPC ], looking for those circulating tumor cells prior to surgery and then postsurgery and then try to prove that principal concept, yes.

Great. Thank you. We had an interesting question come in. How does the DetermaRx risk category correlate with postsurgery minimal residual disease by ctDNA of EGFR positive or EGFR wild-type in the population? Any thoughts there?

Gavitt, do you want to take that one first?

Yes. Sure. We have not studied that in relation to circulating DNA. The thing, I think, that we do know, though, from studies of it and so on, who is more expert in this can correct me if I'm wrong, but circulating tumor DNA doesn't pick up micrometastatic disease. I think there's more data for its use in terms of detecting a recurrence prior to the recurrence being visible on cross-sectional imaging. And so we don't have data on that in the post-op setting. And so I mean, it really hasn't been studied in this context. But I think that no one at this point is using circulating DNA as the reason for selecting patients for adjuvant therapy because I don't think it's very accurate in that setting. And that would be my main response to that. I don't know if Dr. Gandara or Dr. Santos wants to weigh in.

So tune in on October 30 for the answer to this question. Seriously, I'll be presenting at the International Society for Liquid Biopsy on this concept on October 30. And there are data in bladder cancer, non-small cell lung cancer, head and neck cancer, colorectal cancer, breast cancer now, that circulating tumor DNA postsurgery and totally resected cancers is prognostic in all of those settings. In other words, if you know what the mutation was at baseline before surgery, either in tissue or in blood, and then that mutation or mutations are found, let's say, 30 days after a surgery, that portends a poor prognosis. But you're right, Gavitt. What we don't know is if that means giving therapy or giving more intensive therapy or giving a different kind of therapy, whether that will improve outcomes, in other words, not -- we need to know predictive, or wasn't it predictive for a certain therapy, and that's what we don't know. I will also add that, that's the big news which I think we will hear at IASLC World Lung coming up in January, which we will hear the data for circulating tumor DNA from the ADAURA trial, I think, because they did collect blood. And they collected blood at baseline and blood afterwards. So we'll see how it played out with ADAURA.

That will be exciting. We had a question about checkpoint inhibitor therapy. And Dr. Gandara, you presented some -- the trial results about checkpoint inhibitor therapy potentially in the adjuvant and neoadjuvant setting. How do you think that plays in to molecular -- other molecular testing at early stage?

Well, as I said, for the adjuvant setting, we actually don't have any of the Phase III results yet. So I would consider that approach to be investigational. In other words, I don't know of anyone in the lung cancer community, Edgardo can comment on this as well, I don't know of anyone who's giving adjuvant immunotherapy to surgically resected patients post-op. I think the same thing is true for neoadjuvant. There is no FDA-approved approach, and we don't know -- this may not be like breast cancer. We don't know that gaining a complete pathologic response by getting something prior to surgery, neoadjuvant, is going to translate to better survival. So I don't actually -- I would not recommend this to be done. I've already seen a couple of cases. And I'll just tell you, you always hear about these disasters. One patient who had clinically and radiographically what would be early stage 2 with a positive in 1 node on scan, who was treated with a neoadjuvant checkpoint inhibitor and platinum chemotherapy, and then they were restaged prior to surgery, and they had metastatic disease, that patient lost their chance for a cure. Patient came to see me for a second opinion when they had stage 4 disease. And I said, "Well, we have things we can treat you with for stage 4. Unfortunately, you've already gotten immunotherapy. You've already gotten chemotherapy. You don't have an oncogene in your cancer." So what we're left with was not much for that patient. And there -- anyway, what I'm saying is there's research and then there's clinical practice. So Edgardo, what are your thoughts here?

Yes. Yes. I think at this moment, checkpoint inhibitor are completely out of any kind of argument for neoadjuvant or adjuvant because as you pointed out, those studies continue recruiting, and we continued with that. So as soon as the clinical trial, I would not comment on that either. And also in ESMO, one of the many presentation, the adjuvant setting using checkpoint inhibitor was discontinued because all support the therapy's mortality. So at this moment, I think that we need to just wait for all those trials that you will be receiving the same, which is a long list of trial in both setting, neoadjuvant and adjuvant, and wait for their result because either of the patient will have the opportunity to offer surgery or also corrective complication could also happen. So I think that we need to wait.

Mr. Woodard, since DetermaRx can identify patients that have likely been cured by surgery in the early-stage setting, do you see a role for this assay in pharma trials for checkpoint inhibitors or other adjuvant therapy in helping to identify the patients that are likely cured versus would need additional treatment?

Yes. I think that's a great question. And definitely, there would be a real role for this. If you were wanting to really identify the patients who are at risk for recurrence, it makes perfect sense to eliminate the 50% of all these surgical patients who we can identify as low risk from your study pool and to remove them. And then you would see a greater effect of any treatment that you might do on the remaining patient population. I think that's why these studies struggle to find significance in the early early-stage patient cohorts is that there aren't a lot of recurrences. And so if you can remove these patients who we know are cured from your denominator, you're more likely to see the curve separation between your treated and untreated groups.

Yes. Very interesting potential there. I think we've gone through the questions that have come in from the audience. Any final thoughts from any of our panelists?

It's an exciting time for lung cancer patients. I think some of the audience know that during the month of May, there were 7 FDA approvals in non-small cell lung cancer, but none of them in early-stage disease. So early-stage disease still is an unmet need. I think that DetermaRx is a step in the right direction there. And then within the next year, we're going to have the results of some of the other studies, like the Phase III adjuvant checkpoint immunotherapies. And just like in advanced stage disease, if you're a practicing oncologist, you have to keep your running shoes on because new data is coming out all the time. And it will become more complex. But we'll factor in algorithms. And these will be available to practicing oncologists so they can sort through who should get, if it gets approved, immunotherapy, if they still all get chemotherapy, who maybe gets a TKI, and do you use a test like DetermaRx to help sort out the chemotherapy. So I think it's exciting times.

I also think so as a thoracic oncology. And seeing how aggressive is lung cancer, certainly, this DetermaRx is coming in a time in which we have a gap for those early early-stage lung cancer. Dr. Gandara, you speak about the migration of the staging. But when we see about stage 1a, very early stage, we know that there is 1/3 of those patients that we should not leave them without therapy. And regarding Dr. Woodard and her team, show this prospective data. We also need to remind our audience that these tests is also approved by Medicare. So for all those patients, this is an available test that can change and they are most likely to improve their survival. So as you said, we slowly are making progress in something that has been a really big gap for us as a thoracic oncology and to avoid that kind of poor survival for our earlier-stage lung cancer patient.

Great. Yes, it's definitely exciting times for patients with early-stage lung cancer and having more treatment options available to them. Thank you to all of you for this invigorating discussion. Thank you, Dr. Woodard and Dr. Gandara, for your presentations. And thank you to all of you for joining today's webinar. You'll be receiving an e-mail with the link to today's recording. And of course, please pass that on to any of your colleagues who you think might be interested in listening in. This concludes today's webinar. Thank you again for joining us.

Thank you.

Bye now.

Bye.

Thank you. Bye-bye.

Goodbye.