0:00
We will be switching topics this afternoon and I will cover our ultra-sensitive MRD assay that is utilised to overcome challenges in early stage cancer recurrence and disease monitoring.
0:16
I'll start by giving a quick overview of who we are as a company.
0:20
Personalis is based in Fremont, CA in the US We have a large QSR grade facility that has been built intentionally to be FDA compliant.
0:32
We have a large office and lab space.
0:34
We currently process 100,000 samples per year and we'll be expanding that capacity to 300,000 run capacity and we operate 24/7.
0:45
We have ISO certification.
0:47
We are a CLIA and CAP accredited lab and have received New York State permit to utilise our assays in immunotherapy clinical trials all at exome scale.
0:57
We also have a device master filed with the FDA to support personalised cancer vaccine trials where we are at the front end of that work for and that process.
1:07
Of course, we have premier NGS partners that would work with you to support your studies.
1:14
Alliance managers, project managers and FESs and have IP in the space specifically for original sequencing based assays.
1:23
We partnered with multiple biopharma pharmaceutical companies as well as government entities and renowned cancer centres globally.
1:35
We offer our, we have two different products that are that they're applied to different, they have different applications.
1:42
The first application is for tissue immunogenomic profiling using the ImmunoID NeXT platform.
1:47
This is a combined exome and transcriptome assay that provides a comprehensive suite of DNA and RNA biomarkers from the tumour microenvironment.
1:56
And we have the diagnostic counterpart of that acid that is offered in the US and ordered by clinicians.
2:02
We return a clinical diagnostic report back to those clinicians where they can make some actionable decisions for patients.
2:09
Our flagship product is NeXT Personal which is a liquid biopsy tumour and formed MRD assay with an industry leading sensitivity.
2:18
It is a whole genome sequencing based assay that boosts that sensitivity.
2:24
We offer it in the for research applications for IO applications as well as a diagnostic test LDT in the US for which we are seeking reimbursement.
2:35
For the remaining of the presentation, I will focus on our NeXT Personal assay and give a quick overview of the design of it.
2:42
It starts by performing whole genome sequencing on the tumour and the normal.
2:46
The whole genome sequencing data from those samples informed the selection of up to 1800 targets.
2:53
These 1800 targets are of high quality and high allelic frequency and then now they make up the MRD signature that is personalised for that individual patient.
3:02
We manufacture a panel with these probes and that panel is then used to profile the plasma time points sampled throughout the treatment protocol of that patient to look for ctDNA signal.
3:15
So the key features are the whole genome sequencing based approach and the number of targets that are included in the panel that boost the sensitivity of the assay.
3:24
We have analytically validated NeXT Personal and have published a validation an encore target back in 2024.
3:30
This analytical validation helped us take the assay as an LDT to that to the diagnostic space.
3:37
We have demonstrated an analytical specificity of over 99.9% and sensitivity of 100% in through this validation and down to one parts per million, meaning that we can identify one mutant molecule per million molecule sequenced in the plasma.
3:56
We made sure to abide by guidelines from Bloodpac, from CAP and CLIA as well as FDA guidelines for the design of the validation as well as analysis and interpretation of the results.
4:07
We also made sure to include cancer indications that fall on the full spectrum of shedding potential.
4:14
So we know that different cancer indications have different shedding capacity into the plasma.
4:19
We have included different cancer indications and cell lines that are representative of that range on the spectrum of shedding and the results are shown here in this slide where we have determined our limit of blank.
4:32
The detection threshold down to 1.67 PPM, LOD50 and LOD95 of 3.45 PPM with a wide and dynamic analytical range that is that starts at 10 PPM up to 300,000 parts per million.
4:50
So in addition to the analytical sensitivity and the high specificity of the assay, we maintain a sample sparing approach.
4:56
We work with as low as 4ml of plasma needed for the assay and one millimetre cube of FFPE tissue.
5:03
This is roughly about 6 FFPE slides for the assay.
5:06
And another feature is that we can combine NeXT Personal with the tissue based assay with ImmunoID NeXT.
5:13
So in addition to looking at the MRD signature in the plasma of that patient, we can also profile and characterise the tissue itself, the tumour microenvironment and provide you with the DNA and RNA biomarkers such as RNA sequencing, TMV, MSI and a number of other biomarkers that are important to investigate.
5:31
We have a growing list of clinical data and clinical studies with our with key opinion leaders in their respective fields as well as affiliations with renowned again cancer centres globally and these range from early stage disease to late stage and metastatic disease in multiple different indications.
5:52
I'll provide a quick highlight of our published data and non-small cell lung cancer as well as early breast cancer in the next few slides.
6:00
But this list keeps on growing with different and different indications, different disease settings and different KOLs in the space.
6:09
The first study I'd like to focus on is the TRACERx study.
6:12
This is in collaboration with Doctor Charles Swanton, who is leading the TRACERx effort and looking at ultra-sensitive ctDNA detection for preoperative disease stratification in early stage adenocarcinoma.
6:25
This work has been presented at ESMO 2023 and is now published in Nature Medicine.
6:30
We're very excited about that.
6:33
The cohort for this study included 171 patients with over 2000 plasma time points profiled for this cohort also included stages one to three non-small cell lung cancer and two subtypes adenocarcinoma and non-adenocarcinoma.
6:51
This was a prospectively collected sample cohort, but analysed retrospectively and the patients were monitored for a period of five years, which is important because if there is recurrence, we're able to see it during that time.
7:04
The first figure here shows the NeXT Personal detection rate at baseline and we're looking at, I'm just trying to see if I have a laser here.
7:17
If you look at the green bars, the first green bar shows an 81% detection rate with NeXT Personal compared to the grey and blue bars.
7:26
And these are technologies that they were previously evaluated by the TRACERx group.
7:30
These are non-personalised technologies and also are exome sequencing based, not WGS based.
7:36
Importantly, when you look at stage 1 now you see really a prominent detection rate of 52% and that's drastically different from 14 and 13% with other technologies.
7:46
And this is important because for these patient in stage 1, now we're opening up opportunities for them to get intervention and benefit from clinical management right at early on in the stage of their disease in this case.
8:01
So we see a two to four time better sensitivity in this adenocarcinoma subtype.
8:10
Next, we wanted to look at the correlation of ctDNA signal with clinical outcome.
8:15
And here we see that patients who are ctDNA negative had 100% five year overall survival and 94% five year relapse free survival compared to patients who are ctDNA positive.
8:27
So really a great segregation and stratification of cancer patient population with this test and with the signal that we can detect.
8:36
Another important factor we wanted to look at is we wanted to look at the ultra-sensitive range.
8:40
And for this cohort, this was defined as below 80 PPM, 80 parts per million is the limit of detection of one of the other technologies that was tested with TRACERx.
8:51
And what we see here now is that we are kind of exposing a new population of patients in that ultra-sensitive range, the grey bar here, this intermediate risk population that could also benefit from intervention.
9:02
These patients would have deemed negative, ctDNA negative by other tests and received no intervention and no treatment.
9:08
So with this stratification, with this capability with NeXT Personal, we're now providing these patients with an opportunity to get intervention and therapy.
9:18
But just to conclude this part here the green is NeXT Personal.
9:22
We have a limit of LOD95 demonstrated at 4 parts per million compared to 80 and 100 with the other technologies.
9:29
We were able to track more variants up to 1800 variants.
9:34
And this indeed helped with the lead time.
9:36
So we can detect signal six months prior to imaging or routine clinical surveillance, which is very important and critical for patients.
9:46
The study has now expanded and was the next part of the study was presented at ESMO just last year and this part focuses on the landmark time point.
9:56
We expanded patient population to up to over 400 patients with 3000 plasma time points sequence.
10:03
We're looking here at swimmer plots of these patients with the two subtypes adenocarcinoma and non-adenocarcinoma.
10:09
And what they wanted, what our collaborator wanted to focus on in this case is that landmark time point, that window after surgery where a physician can't have, has to make critical decisions for patients.
10:23
The first thing we've measured is the sensitivity at landmark and we demonstrate the 61% sensitivity at the landmark time point.
10:31
One Part 1 factor why we were able to achieve the sensitivity is because 42% of the ctDNA positive signal fell in that ultra-sensitive range where next person really performs the best.
10:44
And this is a key performance of the assay.
10:49
We also looked at correlation with relapsed free survival and overall survival.
10:54
This is the Kaplan-Meiyer plot showing ctDNA negative in purple.
10:58
And then we segregated the ctDNA positive signal to those that were less than 80 PPM and then above 80 PPM.
11:05
We also see distinct, you know, differences between these two populations as well.
11:10
So the assay itself is helping stratify patient populations based on high, intermediate and low risk for the for this cohort, which is a novel a new use case for NeXT Personal or for liquid biopsy assays in general in the space.
11:29
Then we looked at combining the landmark time point with a baseline time point, you get a more powerful readout and predictive readout of how patients will behave or the prognosis for these patients.
11:42
So when you combine preoperative negative with landmark negative, these patients are ctDNA negative throughout their journey have preoperative positive and landmark negative tend to be in that intermediate risk cohort and then pre op positive, landmark positive have the worst prognosis.
11:58
And this is holding true with different other indications.
12:01
And this is also aligning with what the Friends of Cancer Research are trying to push forward as guidelines to look to combine time points to get a more powerful readout for patients.
12:11
So the key takeaway from the TRACERx study, this is again the largest TRACERx cohort that has been ever evaluated and studied.
12:19
We looked here at landmark time point and we see 61% detection rate for relapsing patients.
12:26
This is unprecedented.
12:27
And TRACERx cohort 42% of the signal is below the ultra-sensitive range is below ADP is in the ultra-sensitive range below 80 PPM.
12:35
And the data is showing that we can help next person is valuable in stratifying landmark negative patients with and your carcinoma and correlating that with clinical outcome.
12:46
The next study is in early stage breast cancer.
12:49
This is in collaboration with Doctor Nick Turner at the Royal Marston and the Institute of Cancer Research, as well as Doctor Isaac Garcia Murillas.
12:56
This has been also this was presented as a this was a Podium presentation at ASCO 2024 and is now a published manuscript.
13:03
So again, very exciting data coming out.
13:06
The cohort included 78 patients and over 600 plasma time points sequenced and analysed.
13:12
The different types are shown here.
13:14
We had triple negative population, HER2 positive and ER positive population and we had multiple time points for these patients and they were followed for a period of six years.
13:25
The first thing to look at here is this what we call a lollipop plot showing all ctDNA positive signals.
13:31
The X axis is showing ctDNA in parts per million.
13:35
So this is a quantitative readout that we provide and we show that 39% of ctDNA, 39%, we have 39% detection rate that fall in the ultra-sensitive range below that green line that you're seeing there.
13:53
And most importantly, the 55% of those signals are on treatment samples, on treatment patients.
14:00
So patients have received treatment where you expect that signal to go down.
14:04
These signals would have not been detected by other technologies.
14:07
They would have been missed.
14:08
So it would be hard to see whether the treatment is efficient or working for or patients are responding to it in an identification manner.
14:17
The other nice thing that was part of the study is that we did have a head to so on the right here we're showing the NeXT Personal assay compared to digital PCR and a whole exome sequencing based technology in the same patient.
14:32
So the same patients had two the two technologies conducted and where we show 100% detection rate for the NeXT Personal compared to 76 and 84% with the other technologies.
14:43
And in this particular cohort we had 98% detection rate at baseline for these samples of 49 out of 50.
14:50
And the tables on the left are showing that breakdown by subtypes.
14:54
The purple there is showing the baseline detection rate with other published assays and other technologies that were also evaluated.
15:03
Then we started looking at longitudinal analysis.
15:06
This year is showing the molecular relapse of patients that had clinically relapsed.
15:11
These patients had one, at least one positive MRD time point throughout their cancer journey, and we have established a lead time of 15 months, meaning that we were able to detect ctDNA signal 15 months prior to clinical relapse measurement.
15:27
Some interesting vignettes and some interesting use cases here that we've identified the vignette on the left for this patient is showing that this patient had a persistent ctDNA positive signal 2.5 years.
15:41
This signal was detected with NeXT Personal 2.5 years before clinical relapse before imaging was able to confirm that there is indeed a lesion, that there is recurrence in that patient.
15:52
This is really critical and significant for these patients.
15:55
And on the right we also see, and it is in the red rectangle there, the red box we're comparing green is next person which is WGS based and grey is the whole exome sequencing based acid that looks at 15 mutations.
16:09
And we can see that NeXT Personal can detect signal three months ahead of the other technologies and ahead of actual clinical relapse occurring.
16:19
So really a great critical difference between.
16:22
Between the two assays and great performance of NeXT Personal in these cancer indications and for these patients.
16:30
And then we looked at longitudinal analysis of patients that did not relapse and indeed we show that 100% of patients that were ctDNA negative longitude and they remained disease free and did not relapse.
16:42
So what this data is telling us is that we can trust the negative and we can trust the positive that we are providing with this assay.
16:50
And lastly here correlation with clinical outcome and relapse free survival and overall survival.
16:55
We do see 100% PPV and NPV in this case and great correlation with ctDNA negative and positive status.
17:04
So the takeaways from this particular study is that next person's ultra-sensitive range enabled early detection of recurrence with a means and 15 months lead time over imaging detection, 100% of patients that recurred were detected with NeXT Personal, 100% of patients that were negative on longitudinal testing did not recur.
17:23
We were able to identify 39% of all detections falling below the 100 PPM range.
17:29
So in the ultra-sensitive range.
17:31
And this really demonstrated better sensitivity and better lead times compared to the other MRG technologies that were assessed on the same cohort on the same patients.
17:40
So I think with that I will, I'll just maybe leave you with this slide here.
17:46
So what I presented is early stage cancer, lung and breast cancer.
17:50
And we have identified that a large proportion of these signals and positive signals fall in the ultra-sensitive range.
17:56
But also when we move to the metastatic status and IO monitoring status, we see the same thing.
18:00
There's an assumption that metastatic disease is a high shutter, there's no challenge to detect signal.
18:05
But what we see is really there is 20% or 37% respectively in gastroesophageal and Melanoma here just as examples.
18:12
These are percentages of ctDNA positive signal that fall below 100 PPM.
18:17
So in that ultra-sensitive range and these would be missed by other technologies.
18:21
So it's something to really keep in mind and think about when you're selecting a liquid biopsy assay for disease monitoring, recurrence monitoring and measuring the dynamics of ctDNA for response.
18:33
You do need a clinical, A clinically validated assay as well as an ultra-sensitive assay to support clinical trials and that journey from phase one to commercialization.
18:44
Thank you for listening, Happy to take questions.
