Greetings and welcome to the Personalis ASCO highlights call. At this time, all participants are in listening-only mode. A brief question-and-answer session will follow the formal presentation. If anyone should require operator assistance during the conference, please press star zero on your telephone keypad. As a reminder, this conference is being recorded. It's now my pleasure to introduce our host, Richard Chen, Chief Medical Officer. Please go ahead.
Good morning, everyone, and welcome to our webinar today. I'm Richard Chen, Chief Medical Officer and EVP of R&D at Personalis, and I'm thrilled to be joined by Dr. Isaac Garcia-Murillas and Dr. Rodrigo Toledo, both experts in liquid biopsy testing in cancer. They're here with me today to review the results they presented just a few weeks ago at ASCO, one of the largest cancer conferences in the world, where the latest advances in cancer are discussed. Before turning it over to them, I wanted to take a second to briefly describe our ultra-sensitive NeXT Personal MRD test, which is the focus of the ASCO studies you will hear about today. With NeXT Personal, we saw the opportunity to design a more sensitive blood test that could detect recurrent or residual cancer in patients and to detect it much earlier.
To do this, we designed NeXT Personal to detect extremely small traces of what we call circulating tumor DNA in the blood, and we can detect that down to a 1 part per million number. This means we can detect as little as one circulating tumor DNA molecule in the blood out of a background of a million other molecules. Just to put that in context, that's roughly 10 to 100 times more analytically sensitive than other approaches. To do this, we have developed NeXT Personal with a combination of whole genome sequencing and our NeXT Sense technology platform. We start with whole genome sequencing at the patient's tumor that allows us to identify a unique and more comprehensive genetic signature consisting of up to 1,800 mutations.
We then use our NeXT Sense platform to design a custom personalized test for that patient with those 1,800 mutations, which is used to detect small traces of tumor in the blood that can point to residual and recurrent cancer in the patient. So with this ultra-sensitive approach, it allows us to detect traces of cancer that might be missed with other tests, especially at a level below 100 parts per million and down to this ultra-low one part per million range. We refer to this as the ultra-sensitive range. Just a few months ago at the ESMO conference, Dr. Charlie Swanton, one of our collaborators at the Institute of Cancer Research in the UK, presented the TRACERx lung cancer study data that showed how clinically impactful an ultra-sensitive test like NeXT Personal can be for lung cancer.
That groundbreaking study showed dramatic sensitivity improvements compared to other technologies, and we also saw that the clinical impact of that higher sensitivity led to identifying patients that are at low or high risk for recurrence and led to detecting the recurrence earlier. Today, we are going to hear from Dr. Garcia-Murillas, who has graciously agreed to walk through new exciting results from their groundbreaking breast cancer study. There they used our NeXT Personal test to detect recurrence of early-stage breast cancer.
Dr. Garcia-Murillas comes from a team at The Institute of Cancer Research, London and The Royal Marsden NHS Foundation Trust in the UK, a team led by Professor Nicholas Turner, renowned for his work in the use of liquid biopsy and circulating tumor DNA in breast cancer. This data was presented at a podium presentation just a few weeks ago at the annual ASCO conference. With that, I'm going to turn it over to Dr. Garcia-Murillas.
Hello, my name is Isaac Garcia-Murillas. I'm a Senior Scientist at the Breast Cancer Now Toby Robins Research Centre at The Institute of Cancer Research, London. Today, I'm going to talk to you about the data we presented at ASCO on ultra-sensitive ctDNA mutation tracking to identify molecular residual disease and predict relapse in early breast cancer patients. Detection of circulating tumor DNA in patients with early-stage breast cancer after completion of curative intent therapy associates strongly with relapse. To enable the detection of clinically occult molecular residual disease, assays with very high sensitivity to detect very low levels of ctDNA are required. Genotyping assays currently used in the advanced stage setting lack the required sensitivity and specificity to detect ctDNA in early-stage settings. Current MRD tumor-informed assays use exome sequencing to identify mutations to track in plasma DNA.
We and others have shown that with the current generation of exome-powered MRD assays, ctDNA detection rates at diagnosis prior to any treatment range between 51%- 84%. When it comes to molecular relapse detection, lead time from molecular relapse to clinical relapse ranges from 8.9- 11.7 months. What is clear is that irrespective of the assay used to track mutations in plasma, either digital PCR that tracks one to two mutations or multi-mutation sequencing NGS-based approaches that track between 16 mutations and 50 mutations, detection of circulating tumor DNA is prognostic of worse relapse-free survival. Here, I will present data with NeXT Personal Dx, a tumor-informed whole genome powered MRD detection assay. Matched tumor FFPE DNA and germline DNA samples were whole genome sequenced to a median depth of 38X. We selected up to 1,800 high-quality variants for each MRD patient-specific signature.
We created tumor-informed personalized panels for each patient and used bioinformatics to enhance signal and suppress noise before they were deployed into ctDNA derived from plasma. cfDNA was extracted from a median volume of 3.3 milliliters of plasma, and panels designed contain a median of 1,421 variants per panel. At a specificity target of 19.9%, at a specificity target of 99.9%, NeXT Personal Dx detection threshold is 1.67 parts per million. The 95% limit of detection was 3.45 parts per million, which roughly will correspond to the ability of detecting 3.45 cancer genomes mixed with 1 million genomes. For this retrospective proof-of-principle study, 89 patients were included. 10 patients had a failed panel design, and one patient had a successful panel design, but the plasma did not meet the pre-specified analysis thresholds. 78 patients had a successful panel design and met the analysis criteria.
We included 18 hormone receptor-positive, HER2-negative patients, 35 HER2-positive patients, and 23 triple-negative breast cancer patients. 619 plasma samples were successfully analyzed, with a median of eight samples per patient and a range of two to 14 plasma samples per patient. The median follow-up from the study for this cohort was 76 months. 39% of positive detections were in the ultra-sensitive range below 100 parts per million, which will roughly correspond to 0.01% tumor fraction. 45% of positive post-surgery detections were in the ultra-sensitive range as well. In this cohort, the median detection was 366 parts per million, and the median ctDNA level at first molecular relapse detection was 13.1 parts per million. Typically, the current whole exome sequencing powered assays report between 10 and 100 parts per million. 64% of patients had an accessible baseline sample.
Out of these, 98% had ctDNA detected at this baseline time point before initiation of any therapy. A single cut of 64% of patients had an accessible baseline sample. Of these, 98% had ctDNA detected at the baseline time point before initiation of any therapy. At a single cut of detection of ctDNA, it was prognostic of worse RFS and worse OS. Continuous Cox models of log-transform baseline ppm values were prognostic of increased RFS with an HR of 1.82 and a Wald P value of 0.05, and OS with an HR of 2.19 and a Wald P value of 0.04. When it came to molecular relapse detection, the lead time in this cohort was 15 months ahead of a clinical relapse with a range of four to 41 months. Detection of circulating tumor DNA during follow-up was prognostic of worse RFS and worse OS.
The median overall survival was 62 months for those patients with ctDNA detected and not reached for those patients with ctDNA undetected. The longitudinal performance of the test was perfect 100% across sensitivity, specificity, PPV, and NPV, taking into consideration that three patients had MRD detected post-surgery but subsequently cleared in all remaining time points and were not included in this analysis. 77% of patients had ctDNA nondetected post-surgery and did not relapse, showcasing the high sensitivity, sorry, the high specificity of this assay. 77% of patients had ctDNA nondetected post-surgery and did not relapse, showcasing the high specificity of the test, while 14% of patients had ctDNA detected post-surgery and went on to relapse during follow-up. The pickup rate in this subgroup was 100%. Interestingly, 4% of patients had ctDNA detected post-surgery or during follow-up and did not relapse.
The biology of this is not currently well understood, although we hypothesize that these tumors might be under the control of adjuvant therapy or under immune surveillance. In a like-for-like comparison in the same patient and the same plasma time points, we observed an increase in baseline detection. When we compared digital PCR to whole genome sequencing, we saw an increase in detection from 76% to 100%. When we compared whole exome sequencing to whole genome sequencing, we saw an increase in baseline detection from 84% to 100%. As an example of an improved lead time, a vignette of a triple-negative breast cancer patient with intraductal carcinoma. This patient received neoadjuvant chemotherapy, did not receive adjuvant therapy, and had 17 months follow-up. We observed an increase of over three months lead time when we compared whole exome sequencing to whole genome sequencing.
To conclude, ultra-sensitive detection with a bespoke whole genome sequencing-based tracking assay improves detection of ctDNA at baseline and during follow-up and increases lead time of a clinical relapse. ctDNA detection during follow-up strongly associates with worse relapse-free and overall survival. An ultra-sensitive detection assay identifies early MRD-positive patients that clear ctDNA and do not relapse during long follow-up, with the biology of this currently not understood. Let me finish by expressing my gratitude to all the patients and their families that participated in the study, as well as the clinical and scientific staff that participated in this work. Thank you for your time.
Thank you, Dr. Garcia-Murillas, for that terrific presentation. Now, I'm going to turn it over to Dr. Rodrigo Toledo, an expert in Cancer Biomarkers from the Vall d'Hebron Institute of Oncology in Barcelona, Spain. He'll be presenting data on how the NeXT Personal test can be used to predict and monitor patient response to immunotherapy. Each year, several hundred thousand cancer patients are put on immunotherapy. And while 40% of patients with cancer are eligible for immunotherapy, approximately 12% of patients respond, which really underscores the need for a blood test that can monitor and predict treatment response for patients, doctors, and for payers. And so, with that, I'd like to turn it over to Dr. Rodrigo Toledo.
Hello, everyone. Thank you so much, Dr. Chen, for the kind introduction and for the invitation for me to speak today about our work that we presented recently at ASCO this year, 2024, in Chicago. I'm Rodrigo Toledo, and I'm a Group Leader at Vall d'Hebron Institute of Oncology in Barcelona. And today, I'll present to you our study, its prognostic and predictive value of ultra-sensitive ctDNA monitoring in a metastatic pan-cancer cohort treated with immune checkpoint inhibitors in the context of phase I clinical trials. So, as main takeaways of our study, is that it demonstrated the potential of ctDNA-based monitoring to improve clinical management of patients treated with immune checkpoint inhibitors in the late-stage pan-cancer setting. As the three outlines of our study, we found that baseline-only ctDNA levels are prognostic for the outcome of immune checkpoint inhibitors in this metastatic pan-cancer setting.
We also found that early ctDNA dynamics, in this case, baseline and post cycle three of the treatment, is an indicator of molecular response and better clinical outcome. Lastly, we saw that ctDNA clearance going to undetectable level in this ultra-sensitive analysis correlates with prolonged radiological responses, including partial responses and complete responses. So, this study, we learned that clinical care can benefit from ctDNA monitoring, in part due to the improvement in the technology sensitivity, but also increase in the lead times and compared with imaging.
So, as a background, this study is a collaboration between my translation laboratory at Vall d'Hebron Institute of Oncology and the phase I clinical unit led by Dr. Elena Garralda in Vall d'Hebron University Hospital in Barcelona. We profiled a pan-cancer metastatic cohort. These are refractory patients highly treated previously before going to phase I clinical trials. We profiled this cohort using NeXT Personal.
So, NeXT Personal is a bespoke tumor-informed whole genome-empowered ultra-sensitive liquid biopsy assay that showed many very good results recently in other settings. So, we tried this approach in our cohort as well. Here, we can see the distribution of the tumor types in our study. So, in total, we included 124 patients, as you can see here, many different tumor types. So, it's a really pan-cancer cohort. The majority is colorectal, but we see here many patients with melanoma, breast, non-GI colorectal, and other tumor types, including some more rare tumor types, as for example, neuroendocrine tumors as well. So, from these patients, we obtained tumor samples coming from fresh tumor biopsies or archive FFPE blocks.
We obtained as well 873 plasma samples, longitudinal plasma samples, including samples at baseline, early time points of the treatment with immune checkpoint inhibitors before each of these treatments, including every second, third, or fourth week of the treatment, depending on the clinical trial. We also had follow-up plasma samples until disease progression. The imaging schedule of the study was every six to eight weeks, depending also on the clinical trials. All the samples were collected using Streck tubes and processed within 24 hours. We maintained the pre-analytical very, very stable and homogenized. Here, we can see the distribution of immune checkpoint inhibitor modalities and the response profile in our study.
So, as you can see, the majority of the patients were treated with a combo immune checkpoint inhibitor, including more than one drug, and majority of them, and also bispecific with one drug with two targets. Regarding the response profile, remembering that this is highly pretreated patients, they are being treated now with immune checkpoint inhibitors. So, we identified seven cases with complete response or partial response, 58 cases with a stable disease, and approximately half of the patients, 59 with a progression disease as the best response by RECIST. So, in this study, we applied NeXT Personal Dx. So, as I mentioned before, this is a tumor-informed bespoke genome-empowered minimal residual disease detection assay developed by Personalis. Personalis is a company based in California. So, they developed this assay that after the genome sequencing of paired blood and tumor samples to identify somatic events.
They designed a specific panel that includes 1,800 up to 1,800 different somatic mutations leading to very high sensitivity to track these mutations in the plasma samples, including baseline and ongoing treatment. This approach led to ultra-sensitive detection down to one part per million. We can detect with this assay down to one molecule circulating tumor DNA among a million circulating free DNA coming from non-tumoral cells with the limit of detection of 3.45 ppm and a specificity of 99.9%. And this is very high and very important when we go to clinical trials. These are the data from our study first regarding detectability. At baseline, we detected circulating tumor DNA in 98% of the cases. This is very, very high. We were very happy with these results.
Our previous data in this cohort at baseline, we had approximately a detection rate of 60%, so a huge increase. We were very happy to really be able to detect and track ctDNA of nearly all the patients that were included in the study. The baseline median ppm ranged from 3.26 to 517,000. During non-treatment, we saw 1.36 to 640,000 ppm. Regarding prognostic value, we saw already that baseline values of ctDNA was prognostic for patient outcome. Patients with low ctDNA value correlated with improved clinical outcome. Here we can see median PFS. It's from patients with low ctDNA levels at baseline were the double, comparing with those with the higher median ctDNA at baseline, and also almost the double OS in these patients as well. The patients here that, interestingly, that who attained durable clinical response exhibited significantly reduced ctDNA levels at baseline.
This is very interesting already, but we went ahead to see for dynamics during the treatment. We observed that early changes in ctDNA were strongly correlated with PFS and OS, again, with more than a double PFS and the double OS here in the cases that had molecular response. In our study, we defined molecular response, a reduction of at least 30% of ctDNA from baseline. So, post cycle three, at least 30%, this we include as a molecular response. These are the patients that really benefited the most for immune checkpoint inhibitors. Remembering that this is highly treated patients. Early molecular response correlated with a longer PFS with hazard ratio of 0.36 and a longer OS with a hazard ratio of 0.45.
Importantly, in addition to the molecular response that was below 30%, we observed that the patients that even when they started the treatment with high levels of cfDNA, but after the early time point, they had a clearance of the ctDNA using this ultra-sensitive assay. They even, with this highly sensitive assay, they really have clearance to undetectable levels. So, these patients had really increased PFS compared to more than three times here. And the OS was not reached in our study. So, the patients with ctDNA clearance at any time point during the treatment had a 3.4 times median PFS compared to those who remained ctDNA positive. And the median OS was 8.3 compared with not reached in our study, as I just mentioned. So, the median OS was not reached for six months, but we are continuing monitoring these patients.
I will show you some interesting data, further data on the clearance now. So, patient 570 was a metastatic breast cancer patient that achieved complete response of peritoneal metastasis. So, here we can see the metastasis at the baseline, shrink at T1, and a complete response later on during the treatment. As I mentioned, these radiological responses were correlated with clearance of ctDNA, as we can see here. So, in this graph, we can see here in green the sum of the targeted lesions by RECIST. And in blue, the ppm ctDNA levels. As you can see here, this patient had high levels of ctDNA in the beginning of the treatment, but with already a huge drop in ctDNA levels, marked drop in T1 in the first imaging. And importantly, this patient had a complete ctDNA clearance at T3, at imaging three.
This very low level, actually undetectable level obtained here in T number three were persistent throughout the treatment. As we can see here, the sum of the targeted lesions were decreasing throughout the treatment, reaching the complete response, as I showed you later on. This patient is still alive after three years of the start of the treatment. We are still monitoring these patients using ultra-sensitive analysis. This is another patient. This is patient 446. This is a metastatic head and neck patient that achieved a partial response and a ctDNA clearance at T1 already. So, as we can see here, another patient with positive levels of circulating tumor DNA at baseline. In this case, at the first imaging already, there was a huge drop, actually clearance to undetectable levels.
Similar to the other patient, the following samples remained undetectable throughout the treatment while the patient had maintained partial response. This patient is alive after 43.4 months after the start of the treatment. We are continuing monitoring with ultra-sensitive assay as well. Lastly, we saw that ctDNA increase precedes the imaging-based progressive disease during longitudinal follow-ups. Here we can see the molecular progression in the black dot and the marks of the RECIST. Progression by RECIST. We have here, for example, in gray, we have here the PD, the progressive disease of these cases. As we can see, in the majority of the cases, great majority of the cases, we have the molecular progression already seen in ctDNA months before. ctDNA molecular progression preceded radiological progression by an average in our cohort of 81 days.
The lead time calculation includes patients where radiographic progressions occur prior to molecular progression as well. In conclusion, our study demonstrated the potential ctDNA-based monitoring to improve clinical management of patients treated with immune checkpoint inhibitors in the late-stage pancreatic cancer setting. We obtained three main results, very interesting results at baseline, early dynamics, and clearance. We saw that at baseline, we saw already a prognostic value related with outcome. Patients with lower levels of ctDNA at baseline had improved outcome to immune checkpoint inhibitors. We also saw that on early dynamics, we see that patients that had a molecular response with a decrease in the ctDNA levels at cycle three and more than 30%, there were also those that improved.
So, even patients in the beginning that had high levels of cell-free DNA, those that later on responded, had this molecular response, they benefited as well. Also, we saw a correlation with radiological responses. Again, this is a very heavily treated patient. So, to obtain a partial response or a complete response here in this setting, it's very interesting to see that this correlated with a complete clearance of ctDNA, early clearance of ctDNA, early and prolonged clearance to undetectable levels of ctDNA throughout the treatment during the time of radiological response.
So, we believe that the clinical care can benefit from ctDNA monitoring, in part due to the technology improved and achieving highly sensitive ctDNA level detections. As we can see in our cohort, we believe that this is very important going from research to the clinic and moving these technologies further on to the clinical setting.
And in addition to the increased sensitivity, we see that there was an increase in lead time compared with imaging. This is, again, highly important as we move liquid biopsy to the clinical setting. So, I'll stop here. Thank you so much for the invitation again. It was a pleasure. And goodbye.
Thank you, Dr. Toledo, for a wonderful talk. And thank you. Thank you, Dr. Toledo, for the terrific presentation. And thank you again, Dr. Murillas. Both excellent talks. And I just want to back up for a second. In summary, with the data you just heard from VHIO, really kind of shows that how a test like this can be used across many different cancer types, in this case, 18 different cancer types, to really monitor and predict treatment response to immunotherapy, which is one of the pillars of cancer treatment for patients today.
And if you look at the breast cancer data that was presented a little bit earlier, it really showed some very dramatic results for using an ultra-sensitive test like ours to track patients with early stage breast cancer. In particular, we saw that all the patients that recurred were detected with our test up to a median of 15 months earlier than imaging. Just as importantly, the patients that were repeatedly testing negative throughout the study remained disease-free. And 39% of all the positive detections during that study were in this ultra-sensitive range. And then finally, the patients that were tested in this study, they actually tested some of those patients with other technologies and showed that our approach had significantly better sensitivity and lead times compared to those other approaches.
So with that, I'd like to again thank Dr. Murillas and Dr. Toledo, and also all the patients that participated in these studies, without which we wouldn't have been able to do this study. So with that, I'd like to turn it over to Q&A and also take all the questions. And I think there's some instructions for how to access the Q&A.
Certainly. We're now beginning to open your question- and- answer session. If you'd like to be placed into question queue, please press star one on your telephone keypad. A confirmation tone will indicate your line is in the question queue. You may press star two if you'd like to remove your question from the queue. For participants using speaker equipment, it may be necessary to pick up your handset before pressing star one. And we ask that you please ask one question and one follow-up, then return to the queue. Our first question is coming from Yuko from Morgan Stanley. Your line is now live.
Hello, this is Yuko. Thank you for taking my question. Maybe starting with the IO response monitoring study, could you provide some color around the patients who progressed despite clearing their ctDNA? While there was a clear separation for patients who saw ctDNA clearance versus those that didn't, was there a common characteristic that could explain why those specific patients progressed?
Yeah. No, that's a great question. One of the interesting things about this study is that these were extremely, extremely sick patients across the board. So if you remember, if you kind of look at that swimmer plot that he showed, one striking thing there was the vast majority of those patients had positive ctDNA through their entire course. These were phase I patients across different phase I studies.
Patients in phase I studies tend to be the sickest of the sick, even compared to other metastatic patients. These were really, really sick patients. I think, and so in the metric that talks about clearance, it was clear that the patients that had some clearance during that treatment period did do substantially better than the ones that didn't. But that metric is based on clearance at some point. Just because they were clear doesn't mean they were completely clear the whole time. If you look at the swimmer plots for some of the patients that did have clearance, some of them did have actually we started to detect ctDNA again in those patients, and they did recur.
Got it. That was helpful color. And then on the early breast cancer study, could you elaborate on how to reconcile the three patients who had ctDNA detected post-surgery or during follow-up but did not relapse? Essentially, those patients who were false positives. What are some of the ways that you could mitigate some of these?
Well, actually, our collaborators, Dr. Turner and Isaac, who gave the talk, strongly feel that these are not false positives. So if you look at it, they're actually really interesting because we actually are highly sensitive here. If you look at those three patients, they start out positive. And then essentially, while they're on treatment at the very beginning, it's adjuvant treatment clear very quickly, and then they're negative through the rest of their course. And so it actually is we're actually very accurately characterizing, it looks like, the biology of what's happening with these patients.
Got it. Thank you.
Thank you. Next question is coming from Dan Brennan from TD Cowen. Your line is now live.
Great. Thank you. Thanks, Rich. And for the presentation. Maybe the first one you point out, I think 39% of the patients that you identified were in the ultra-sensitive zone below 0.1%. Just kind of what's the can you speak to the importance of that kind of stat and how we think about it for the differentiation of the test?
Yeah, absolutely. Yeah. So that was a really striking finding. So it just means, again, that every test that was done, about 39% of those were in this very low range that we are kind of uniquely poised to detect. And these are time points that would have potentially been called incorrectly negative by less sensitive tests.
And so I think if you think about that in terms of the implications for the patients, I think you can see that what it would lead to is incorrectly kind of thinking about the patient as being negative when they're in fact positive and then delayed detection, essentially, because the cancer would continue to progress. And down the line, once the tumor is significantly bigger, then a less sensitive test may eventually pick it up. So what we see is better lead times because of the extra sensitivity. And you kind of saw that with the 15-month lead time in those results, which compares very favorably to what's been reported previously by other assays. Roughly kind of nine to 11 months is what's been reported previously. So it looks like we're doing significantly better in terms of detecting the cancer earlier.
The other thing that was very striking was, and I don't know that this is pointed out, but the first detection, the first time we detected the positive patients, and we detected all the patients that eventually relapsed. But that very first detection, the median level was 13.1 ppm. So if you think about that, again, it means that we were able to pick it up at this very low level at that very first time point. But a less sensitive test might have missed that detection and called it negative.
Got it. Great. And then maybe the second one, just can you remind us kind of what the path forward is from here, obviously, with all this data in hand? Just, how does this? Is this in line with expectations in terms of the outcomes that you saw and kind of how does this influence timing for ultimate Medicare submission?
Yeah. No, our collaborators were really excited about the results. You can see that when they compared it to the other technologies they had used on the same cohort, that we performed significantly better than what they had seen. Our baseline detection rate was 98%, very close to 100%, which compares incredibly favorably to what has been reported before. And then we had these increased lead times for detecting the patients. And so there was a lot of excitement at ASCO, a lot of buzz around the results. I think the interest is across the board for the potential use of our test in the breast cancer patient journey.
For example, I think there's a lot of interest in using this test to detect the cancer earlier and then escalate treatment for breast cancer patients, the ones that need it, that are recurring. The other area is actually in the area of de-escalation. The idea is that there are a lot of patients receiving therapy now. Many of these patients, especially breast cancer patients, may not actually need the therapy, but there's no way to understand whether they're truly clear of disease or not. There's a lot of interest in using a test like ours to help make a decision of whether to de-escalate these patients. And you can only do that with a very sensitive test, right? If you don't have a sensitive test, you can't really trust a negative result.
And then, of course, the other area is using it to monitor treatment response, whether it's neoadjuvant treatment prior to surgery or after surgery during adjuvant treatment.
Great. Thanks.
Thank you. As a reminder, that star one is to be placed into question queue. Our next question is coming from Mark Massaro from BTIG. Your line is now live.
Hey, guys. Hey, Rich. Thank you for taking the questions. Certainly encouraging data readout at ASCO. Maybe my first one is, what level of specificity do you think is needed to have a follow-up monitoring test? I mean, I have to imagine that hopefully we'll be monitoring breast cancer patients repeatedly over maybe five years or so. So would you agree that obviously 100% is ideal? But what would you characterize as the minimum level of specificity needed to ensure that you're calling positives true positives?
Yeah. That's a great question. The specificity is incredibly important. So we feel like a specificity of 99.9% or greater is going to be critical at minimum to achieve good clinical outcomes for these patients. And that's what we design NeXT Personal towards. And as we saw in this case, we had exceptional specificity, and we've demonstrated it over and over in our other studies as well. I think, to be fair, I think there still need to be studies that say, "Show that with that level of specificity, that is what's needed to get to great clinical outcomes for the patients." So those studies are kind of ongoing, not just for us, for other folks. But the hypothesis for us and the folks we've talked to is something greater than 99.9% is going to be important.
Okay. Great. I know on your Q1 call, you guys talked about a goal of getting three submissions or three cancer-type submissions to Medicare by the end of 2024. I'm curious. So the study from Dr. Morillas in breast cancer, I believe that was a retrospective study design. Is it your understanding that you can submit that upon publication, you could submit a retrospective study to Palmetto GBA for consideration for Medicare coverage? Or should we expect other studies, perhaps prospectively designed? Just give us a sense for how we should be thinking about the three, whether or not they're retro or prospective.
Yeah. No, I think there's precedent for retrospective studies to anchor the submissions to MolDX and so on. These are really strong studies. Our feeling is that they will be a key part of supporting our MolDX submissions.
Okay. Great. I'll keep my questions to two. Thanks.
Thank you.
Thank you. Next question today is coming from Thomas Grattan from Main Street Capital. Your line is now live.
Hey, Rich. Thanks for taking the questions. Hey, Eric, with respect to the MRD data, there were 10 panel design failures that were noted in the slides. How comparable can you explain? Is there anything to take away from that number? And then how does that relate to your typical experience that you see in the lab?
Yeah. No, great question. Yeah. Actually, we were and our collaborators were actually quite thrilled with that number. These were actually unusual in that they were very old samples in many cases. And so the fact that we were able to be successful a high percentage of the time was actually very, very positive.
It's actually in terms of our kind of diagnostic testing where experience, we've launched this kind of early access, we're seeing higher success rates than that, for sure.
Great. And then maybe some biology background for me. So Dr. Toledo had the patient 446 that had complete clearance, but also a partial response that was enduring. If you still have tumor tissue there, why is it not shedding?
Yeah. And I think that those patients we're going to be digging into a little bit more. But I think there's a couple of things that are thoughts there. It could be that the biology is that they're not shedding. But actually, in some of these cases, I think one of the powers of the circulating tumor DNA is that it can kind of suss out the difference between, is this tumor or sometimes it could be scar tissue?
Or perhaps there's a case of pseudoprogression where it looks like the tumor is large, but there's actually a response happening because the tumor is enlarged due to immune cells that have actually infiltrated a tumor, not because the tumor is enlarging itself. So I think more investigation is needed. But I would suspect that some of these cases, it may be that it's not shedding, but it also could be that there's a fair amount of scar tissue in some of these cases.
Got it. And then just one quick final one. Do you have a sense of the timing of the next TRACERx data release, the full cohort?
Yeah. Yeah. We're actively working on it now. I don't have a timing for you, but I can tell you it's definitely on the top of our list.
Excellent. Thanks, Rich. Appreciate it.
Thanks.
Thank you. We reached the end of our question- and- answer session. I'd like to turn the floor back over for any further closing comments.
I just wanted to thank everyone for joining us today to review some of the exciting results we presented at ASCO. And I, again, want to thank the speakers that took their valuable time to walk through the results. Thanks again.
Thank you. That does conclude today's webcast. Let me just connect the line at this time and have a wonderful day. We thank you for your participation today.