We sat down with Prof. Jacques Cadranel, International Coordinator of the DEEP-Lung-IV study and Head of the Pneumology Department at the Hospital Group University Hospitals of Eastern Paris, who shared his experience in this collaboration with SOPHiA GENETICS, and the importance of the integration of multimodal data in clinical practice to advance personalized medicine in lung cancer.

Watch the spotlight:

Hello Professor Cadranel, thank you for welcoming us today at the Tenon Hospital. Could you explain to us what DEEP-Lung-IV is and the genesis of such a project?

The genesis of this project is 2020, so it's been a while. With the desire to move from a slightly comparative medicine of treatment arm to arm, to a more individualized medicine.

Taking into account the fact that we have what we call artificial intelligence (AI), which allows us to accumulate a lot of data, and ultimately be able to develop signatures that we cannot develop in our heads, or even with, a usual statistical approach. And also to have the impression that the patient cannot be reduced to a sex, a weight, a performance status, an imaging test, or a molecular test.

The patient is a whole, and as long as we haven't integrated this whole, I think we will still be a long way from individualized medicine.

Could you explain the DEEP-Lung-IV project as it is today?

At the time, and what it has become, was first to say: “Let’s do something from the actual standard of care as it exists today”. That is to say, not to build a project that won't be applicable in real life. That was point number one.

Point number two was to accumulate the usual clinical data we have in our medical records, the usual biological data, blood counts, creatinine levels, liver tests, and imaging data while avoiding focusing solely on what we call the targets that we measure in interventional trials, but rather taking into account the patient as a whole. So, having a radiological phenotype. And then also integrating molecular parameters. Creating an initial profile of the patient, treating them as they were treated, and having their (outcome) information at the first assessment – are they stable or responders? - so that we can subsequently create signatures that would allow us to define their treatment outcomes before exposure.

From that, we can predict not only what treatment the patient received but what they would have been like if they had received another treatment. So that's the first step, that's the basis of DEEP-Lung-IV.

We talk about predictors, we talk about making the signature available so that it can be deployed in the clinical routine. What does such a platform look like to you?

The platform would be similar to what SOPHiA GENETICS was kind enough to show us. It's a platform that is, first and foremost, very tactile, very easy, very pleasant.

That's essential in the Multidisciplinary team meetings (MTBs). We can't get carried away, it musn’t be complex and should be extremely user-friendly. What we envision is a simple interface that provides some kind of a detailed patient profile, including characteristics such as pathology, performance status, and offering clear probabilities for treatment response, progression, overall survival, and for each therapeutic option.

 Why choose SOPHiA GENETICS as a partner for such a project?

I believe that’s not how you find a partner. A partner, you look him in the eye and you say, "I want to see". And then you start exchanging, and you want to see even more. And little by little, that's how you build a bond. I wouldn't have been able to choose SOPHiA GENETICS if I hadn't felt, from the start, the idea that SOPHiA GENETICS was coming towards us with was really important, because that's not what usually happens. Meaning that we have private partners who do not listen to the creativity of the doctors and of their understanding of the complexity of the patients, so that they can provide us with tools that respond to this complexity. Usually, these partners come with their prototypes and expect us to agree with it.

This is what I really appreciated about SOPHiA GENETICS. It was a real partnership from the beginning. Together, we created something very original, something that neither of us could have done alone. And that's what's so enjoyable about this collaboration.

Professor Cadranel, you are the international coordinator of the study and also the scientific committee's chair. Could you tell us what this represents for you, in terms of challenges and opportunities, and why you accepted such an appointment?

Why did I accept? Because I share this baby with SOPHiA GENETICS. It's also my baby, and it's a really meaningful project for me. I believe that conceptually, we can completely change the paradigm and move on from a Newtonian medicine, that is: we observe A, which does better than B in interventional trials. And then we happily apply this, trusting the hazard ratio to double the probability of response, etc. But ultimately, that's not how it works. It's useless, and we should rather lean towards what I call a kind of quantum medicine. That's why I find this project extremely exhilarating and exciting, and I was given some chance to be part of it. That's also unique. That's why I don't want to miss out.

So the challenge is getting others on board. Right now, SOPHiA GENETICS and I, we need to accelerate the project a little more to produce results. And we are close. We hope that this new year will be the one of results and of raising awareness of this project. And SOPHiA GENETICS may not yet be aware that doctors aren't ready to hear what we're going to tell them, and so there will be some challenges ahead of us. We need to educate the doctors currently conducting the investigation because I think they didn't fully understand what they were getting into. They provided data for this study, they trusted SOPHiA GENETICS and also me, but I think they didn't understand what we were going to offer them at the end. And this is an exceptional challenge because it truly represents a complete paradigm shift for precision medicine.

Professor Cadranel, could you explain to me what you expect from the multimodal approach and the use of AI in clinical routine?

When we discussed DEEP-Lung-IV early on, I represented a little slide with my brain, which was composed of the performance status, the burden of the disease, molecular biology, organ pathologies, extent and type of metastases, and so on. And I told you, in fact, I integrate all of this unconsciously, but I also integrate it consciously.

There's still a need for us, first of all, to ultimately try to mimic our approach, without us realizing it, is what I call the black box, to make things more objective and also a little more reproducible, in particular with some “grey” situations where we don't really know what the best strategy is for the patient. If we resubmit the same patient, changing their name, changing their sex, or gaining one kilo, we realize that halve the times, we don't give the same answer. That's still a problem.

It's a concern that we're not reproducible within a MTB, but between MTBs. So that also means that there's an inequality when it comes to patient care, which we hope will be mitigated once we have that AI signature.

What other applications would you see in the context of multimodality and artificial intelligence?

There are. One part that we don't control is the patient's side. Having a perspective from the patient’s and caregiver’s perspectives. Integrating their perspectives into these signatures would be extremely valuable. Even with the greatest desire to do well as a doctor, when you make an announcement to a patient, when you offer them a treatment, you project an extraordinary amount of information on what the patient will receive and how they will experience it.

We did some sociological work to see how they felt about it. The most enthusiastic, compassionate doctor, the one who wants to do good the most, is wrong in 99% of cases by the patient. Patients are an essential element to consider.

What we should also integrate into this therapeutic signature is how it integrates into people's emotional, social, and personal lives.

We thank Prof. Cadranel for his time and for sharing his experience. To learn more about the DEEP-Lung-IV Study, visit the dedicated page .

SOPHiA GENETICS products are for Research Use Only, not for use in diagnostic procedures unless otherwise specified.

We met with Dr. Sébastien Couraud, DEEP-Lung-IV Scientific Committee member and Head of the Pulmonology and Thoracic Oncology Department at Hospices Civils de Lyon, to talk about his participation in SOPHiA GENETICS’s DEEP-Lung-IV study and reflect on the benefits of multimodal approaches to transform precision medicine and improve patient outcomes.

Watch the spotlight:

Hello Sébastien, thank you for receiving us here at Lyon Civil Hospitals (HCL). You are the biggest recruiter of the DEEP-Lung-IV study that we launched a few years ago and also a member of the scientific committee. I would like to have your perspective on why you joined the study, and then your vision on the project in general.

Hello Marion, thank you very much for the invitation. I am very happy to welcome you here and to have a discussion on the DEEP-Lung-IV project.

We were immediately won over by the ambition of the project and by the fact that this project was very multimodal, precisely.

And I think that is really what we are going to discuss together today. This very multimodal, very ambitious side immediately won us over. In addition, it is true that we have had quite a few strong relationships with the members of your team from the start. We already knew each other before, so it was quite logical for us to finally support you on this new project.

Could you tell us more about the objective of the DEEP-Lung-IV study and how this study will meet the objectives of precision medicine in the future?

The objective of the DEEP-Lung-IV study is really to go, collect quite a massive amount of data on patients who are treated in different investigation centers and who are treated for lung cancer.

And in fact, the principle is really the multimodal collection of massive data, to then be able to create decision support tools that will help us on a daily basis.

Concretely, it manifests itself in a quite simple way - in reality, when we take care of a new patient, these are patients where we will ultimately integrate all of the data that we have generated for this patient. And collected in a database.

The multimodality here comes from the fact that we will collect radiology, pathology, and molecular biology data, and connect it with clinical data.

All this data will ultimately make a very large database, with centers from all over the world, and will then allow us to ultimately generate decision-making tools.

For the future, how do you see the next steps of the study and the collaboration with SOPHiA GENETICS?

For the future, obviously, the first step is the results. We were talking about it earlier, we need to have these results, see precisely the type of results that we have, the tools that have been generated, and we will then have to ask ourselves the question of whether we can use and integrate these generated tools into practice, how to do it and evaluate it.

We are finally at the beginning of a collaboration, and it would be a shame to stop on such a good trajectory. The objective is to continue the collaboration with SOPHiA GENETICS because it is indeed really important, from now on, to enter a partnership a little more operational, if I dare say.

You previously told us that artificial intelligence (AI) was expected in real routine practice. In your opinion, what would allow us to bring this to routine?

This step towards the clinical routine of integrating AI, that really is a very good question and I think it's really, if I may say, the golden question to which ultimately no one really has an answer today. We all think that AI will have a strong impact on medicine, in several dimensions of medicine. Obviously, when we talk about lung cancer, the decision of which today is very multimodal, and is a field of knowledge that is expanding almost hour by hour.

In reality, we imagine that AI will be a very important decision-making tool, daily. Nevertheless, we work on humans and we work with patients, with lives. So everything we do must necessarily be evaluated and we must be certain of what we do.

And that's the important element and what we're missing today. And what we're missing today are studies that will allow us to show that compared to the absence of artificial intelligence, the addition of it improves patient care on very specific events, such as survival, progression-free survival, treatment tolerance, the choice of a more suitable treatment, etc. So really the next step is prospective evaluation studies that will allow us to integrate these tools into real life and compare them to the outcome and current care.

Sébastien, let's think about the first day of the launch of the project and, with hindsight over these 4 years. If we had to do it again, would you join in?

That's a good question. Yeah, I think so. Yes, I think there was a bit of a crazy side indeed when you came to see me and you told me “We are going to take all the data from your patients, integrate them, and you're going to send everything to us”. It still required a lot of organization for us, but finally, I think we would do it again the same way. The collaboration was really pleasant in reality, that is to say, that it was done in a fairly simple, fairly flexible way. And in reality, with a little bit of organization on our side. Objectively, it went well, so I think that I would do it again. Yeah, I would sign again.

We have been working together, HCL and SOPHiA GENETICS, for several years. Could you tell us about the perception you have of SOPHiA GENETICS as a company?

Yes, so it's true that we've been working together for years.

I think that SOPHiA GENETICS is one of those companies that has entered the health ecosystem through one end and has succeeded in developing this multimodal aspect precisely.

That is to say that when we met at the very beginning, you were really in biology, in genetics. And I remember conversations that date back a very long time. You managed to really open up your field of possibilities by integrating this notion of multimodality and by very quickly understanding the interest of multiple modalities, instead of staying in a single field, in which you were nevertheless quite an expert, but to take a risk by exploring other fields and opening up to other possibilities. And I find that this risk-taking is interesting.

I find it interesting because it ultimately makes it a company that was able to understand a little in advance the interest of multimodal, to bet on it, and today, to open up to it very widely.

So, there is really a fairly innovative side to your company.

Sébastien, today, the treatment decisions for a patient are processed in Multidisciplinary team meetings (MTBs). What do you expect from an algorithm and the machine learning tool in general?

In fact, at the risk of surprising you, the idea is to perhaps be a little less human. You have to understand that today when we take care of a patient, we take care of them based on our instinct, we take care of them based on their story - the patient's story - we know them, we know their story. And then we have our experience - the experience of whether we have other similar patients or not, that we have taken care of. And so that's what ultimately constitutes the decision that we're going to make in the MTB.

It's obviously science, right? There's no doubt about it. We have guidelines, and we rely on these guidelines, but then it's ultimately a collective of clinical experiences, good and bad, that will allow us all together to make the decision that we think is most appropriate for the patient.

Certainly, it's good since it's been working like that for years, and today, we're still practicing good medicine.

However, having a tool that is completely dehumanized will allow us to humanize this question less.

Now, what I'm saying is going to seem very odd, but in fact, it will allow us to tell ourselves that science in this situation, specifically for this group of patients, is telling you that. The human will then come and modulate the scientific decision and will say: the machine's decision is this. I will adapt it with the knowledge of my patient, but at least I will start from something very current, very factual, very scientific, and I will articulate based on it. This is perhaps where it will change things a little.

In practice, this machine support, how do you view it in MTBs? What does it look like?

It looks like a very intuitive interface. In fact, we must not forget that we still have a lot of work. We are increasingly solicited, we are asked for more and more things. The MTBs are becoming very, very heavy. I think all colleagues see that the MTBs are becoming more and more complicated, there are more and more patients. Patients are surviving more and more, which is an excellent thing, but it obviously increases the volume of MTBs.

All this requires us to be very simple and very pragmatic.

The idea is really to have a tool at the service of the clinician, a very intuitive, very easy-to-handle tool, that will give very visible results, very simple for the whole room in a few seconds - I enter the clinical characteristics of my patient, I immediately visualize the data in the database and it will very easily help me to be able to make decisions.

What is also important is that we can vary the parameters a little because sometimes we will hesitate between two strategies. We will say to ourselves: “Here, do I use this molecule or that molecule? Or do I use this diagram or that diagram?” and that finally can be done in one click. Add or remove a therapeutic option and immediately visualize the effect it can have in a cohort. That will help us greatly in our decision making.

We thank Dr. Couraud for his time and for sharing his experience. To learn more about the DEEP-Lung-IV Study, visit the dedicated page .

SOPHiA GENETICS products are for Research Use Only, not for use in diagnostic procedures unless otherwise specified.

We sat down with Prof. Jean-Christophe Bernhard, UroCCR Coordinator, and Dr. Gaëlle Margue, Urology Fellow, at University Hospital Bordeaux, to discuss the collaboration between UroCCR – the French Kidney Cancer Research Network - and SOPHiA GENETICS, and get their insights on the use of AI-powered multimodal approaches to improve patient care.

Watch the spotlight:

Gaëlle, Jean-Christophe, hello. It's a real pleasure to be with you today to discuss the collaboration between SOPHiA GENETICS and UroCCR.

Before we go into more details about this collaboration, I would like you to tell us a little about your background, your life as a surgeon at the hospital, and also introduce the UroCCR network and the role you play within this network.

Gaëlle Margue, I am a junior doctor in urology at the Bordeaux University Hospital. I arrived in 2018 to start my internship and now, I have been a PhD student for two years and my science thesis focuses on kidney cancer, surgical and oncological themes, within the framework of the UroCCR network, the French Kidney Cancer Research Network, and the I.CaRe team, a Kidney Cancer Research team in Bordeaux.

Jean-Christophe Bernard, I am a professor of urology at the Bordeaux University Hospital, and I coordinate the I.CaRe team, the Kidney Cancer Research and Innovation Program in Bordeaux. I am also the national coordinator of the French Kidney Cancer Research Network, UroCCR.

Gaëlle, we worked together on your medical thesis around UroPredict. Could you describe UroPredict, and what it does?

Yes, absolutely. That was the subject of my medical thesis. The initial idea was to better characterize the risk of recurrence after kidney cancer surgery. Therefore, for patients with localized kidney tumors that are operated and are considered cured or in remission after surgery, we want to better determine what factors lead to patient relapse or not. We have prognostic scores to try to determine that, but they are not very effective, and so we wanted to better characterize that, to propose a follow-up schedule tailored to the risk of recurrence, or additional treatments for patients who have a high risk of recurrence.

In UroCCR, we have a lot of data that relates to these kidney cancer surgeries - clinical, biological, imaging, surgical, and monitoring data -, which we were not able to leverage, with traditional statistics, to better characterize the risk of recurrence. So the goal was to create a new machine learning score, a tool to predict recurrence in these patients, based on all this data that we have in UroCCR.

It's quite fascinating, from an outside perspective, to see surgeons saying that we need Artificial Intelligence and Machine Learning, to be able to advance our care. Jean-Christophe, from a strategic point of view, how do you see the collaboration with SOPHiA GENETICS, in particular? And then more broadly, the role these precision medicine tools are expected to play in the future?

We see it as something that will be decisive, and I would say, necessary to maximize the value of all the data we gathered over time.

UroCCR, to go back a little into the history, is a project that dates back to 2006, which was certified by the National Cancer Institute in 2011. It was initially deployed as a multicenter project in 2013 across 11 centers. And today, we’ve grown from 11 centers to 54 and soon 58. And so, the positive excitement of the system means that we are constantly collecting highly qualified data on the pathology of patients who are diagnosed with a urological tumor. We collect this data regardless of the treatment method, whether the patient is in active surveillance, whether he is operated on, whether he is treated systemically with medication, or whether he is treated with interventional radiology.

In doing so, coupled with the increase in the number of centers, and the increase in the number of patients, we reached the 20,000 patients included in the UroCCR, this dataset became even more considerable since we are linked to the SNDS, the National Health Data System, and therefore we can also do medico-economic evaluation, representing nearly 10 million data points available on patients treated contemporaneously by French teams for kidney cancer.

It actually becomes a need, not so much a vision but ultimately a necessity to have new statistical methods to process this data, and derive insights to enhance our ability to provide patients with care that is as personalized and individualized as possible.

This is, I believe, the foundation of our collaboration with you and your team at SOPHiA GENETICS, to explore how we can take advantage of all the data patients have entrusted us with, regarding their illness, since everything is done with the patients’ consent - and I think it’s also important to point this out. Being able to produce new tools, and what Gaëlle said is that her thesis work, which initially is a scientific and fairly general work, I would say, has nevertheless led to the online publication of a calculator, which can now be used in a pragmatic way to answer a question that we may ask ourselves for a given patient.

Ok. So, beyond the research aspect that we were able to carry out together, which I think we are all very happy with, there is this role that an industrial company can play afterward, which will be the deployment of the tools, their validation, their improvement, the entire life cycle of the software.

Are these the things that are beginning to resonate in the minds of the medical profession, or are we still at the beginning and still at a time when everyone needs to find their role?

Yes, it is... It is very timely, because the UroCCR network has just been certified by the ANR, the National Research Agency, as a clinical investigation network on medical devices. These are also themes that we are addressing within the framework of the I.CaRe program, within the framework of the RHU (University Hospital Research in Health) where there is this desire for collaboration between academics and manufacturers for carrying out projects and arriving at outlets that are tools, products that can be used in everyday life and in the routine of patient care.

The UroPredict tool that we co-developed during your thesis, Gaëlle, it is now deployed and accessible. How is it being used by you?

It is a tool that is not yet a certified medical device, so it cannot yet be used to change patient care strategies.

However, it is very useful for discussing with the patient. We can present it as a research tool that helps us better assess their risk of recurrence and concretely show the patient their risk factors for recurrence and their personalized risk. And this allows us to develop the discussion around monitoring or potential complementary treatments to surgery, for the continuation of their care.

Do you see among your colleagues or among the pharmaceutical companies, a little reluctance to apply them, to trust them, at the level of clinical routine?

So, reluctance… I don’t think it’s really a reluctance. I think that at each time there is something new, that potentially can bring a change in practices. There is always an observation phase and we have to assimilate what this novelty can be and what it can bring. We experienced the same thing, for example, for the introduction of robotic assistance in surgery, which today everyone is convinced on its benefit, both for health professionals, for surgeons, but also for patients. There was a whole phase where the community asked itself the question of whether introducing robotics would actually make it possible to do better than conventional techniques that were, I would say, well known, and had been validated for many years.

So there is always this moment where we ask the question of whether it is a real innovation, whether it will really bring added value. There is always this observation phase. Afterwards, I think that more and more, the medical and surgical community is convinced that with technological progress, we are finally able to improve what our daily routine is, what our practice is and obtain additional information that will enrich our practice and patient care.

What do you think the next algorithm we should develop is?

We have several ideas. We could further enrich this one by perhaps incorporating radiomics or pathomics, as we discussed, adding imaging data to increase the precision of the current tool. Then, we could explore many other algorithms with different objectives, such as predicting kidney function loss after kidney cancer surgery, or predicting the risk of morbidity, and intra or post-operative complications.

Objectives and things that we are trying to better characterize and there are several that we could look into and probably we have all the necessary data from UroCCR, we need to be able to exploit it.

And I think if I can complete.

There is also a question that we should address. Every time, we evaluate outcomes that are very objective, focused rather on the practitioner, the surgeon, the quality of his surgery, the outcome, the evolution of the disease, etc.

But I think that we should also be able to work on outcomes that are reported to us by the patient.

Quality of life, satisfaction, anxiety, which are things that we capture in UroCCR and in particular thanks to the UroConnect application. Today, in UroCCR, we have what we call PROMS and PREMS, which are patient-reported outcomes, and I believe this is another field worth investigating, moving beyond purely scientific outcome predictions, but also to take into consideration what the patient's experience is, and to judge what we do or to predict how we can improve what we do and propose to patients, based on their feedback about the quality of their care. I think this is one of the objectives we should collectively aim for.

We thank Prof. Bernhard and Dr. Margue for their time and for sharing their experience. Visit the UroPredict page to learn more about this machine learning model on real-world data for the prediction of kidney cancer recurrence.

SOPHiA GENETICS products are for Research Use Only, not for use in diagnostic procedures unless otherwise specified.

Context & Study Overview

Radiation pneumonitis (RP) is a significant and relatively common complication of (chemo) radiotherapy (RT) in the treatment of locally advanced non-small cell lung cancer (LA-NSCLC).

This study investigates clinical, dosimetric, and radiomic features predictive of lung toxicity, specifically grade (G)≥2 RP, in patients undergoing (chemo)RT for LA-NSCLC, with the aim to build a predictive model to estimate its occurrence.

The researchers conducted a retrospective multicenter analysis of 153 patients treated with (chemo)RT, between 2015 and 2019, to identify key risk factors. Baseline CT scans were segmented to extract radiomic features from the lungs and the tumor, and integrate them with clinical and dosimetric features.

The study employed a machine learning (ML) approach using logistic regression and random forest models to develop predictive models for RP occurrence in this patient population.

Key Findings

The clinical and dosimetric risk factors linked to an increased RP risk included high initial hemoglobin levels, older age, low Tiffeneau ratio (FEV1/VC), decreased initial platelet count, dosimetric factors (mean dose to lungs, lung V20Gy and V13Gy), and the use of adjuvant durvalumab.

Seven radiomic features related to intensity distribution and texture were significantly associated with RP risk.

The developed ML-based model (random forest) integrating clinical, dosimetric and radiomic data achieved the best performance with an AUC = 0.72 (95% CI [0.63-0.80]), outperforming models based on combined clinical and dosimetric data (AUC = 0.64), or on radiomic data alone (AUC=0.64).

Conclusion and Clinical Implications

Integrating radiomic features with clinical and dosimetric ones improves the prediction of RP, providing a more comprehensive tool for risk stratification in lung cancer patients undergoing radiotherapy.

This study showed that identifying high-risk patients for RP could allow for a more personalized treatment planning to reduce their risk, such as adjusting radiation dose constraints, introducing protective measures, or enhancing follow-up care.

Explore this infographic to learn more about this project and the predictive model developed by Evin et al’s.

Evin. C, et al. Clin Lung Cancer. 2024 Nov 20:S1525-7304(24)00248-1. doi: 10.1016/j.cllc.2024.11.003
Principal Investigator: Eleonor Rivin del Campo, MD, PhD

SOPHiA DDM™ for Radiomics and SOPHiA DDM™ for Multimodal are concepts in development. May not be available for sale.

In the era of precision oncology, it has become increasingly common for patients diagnosed with cancer to undergo tumor sequencing. Identifying the mutations that make up a tumor’s genomic landscape can help guide selection of targeted therapies and inform prognosis. Despite the recognized value of tumor-only sequencing, labs performing this type of testing face a number of technical challenges that, if not properly addressed, can render the results uninformative or even misleading.

Although there are a variety of inherent challenges in tumor-only sequencing, all ultimately impact the ability to accurately distinguish somatic mutations driving tumorigenesis from germline variants associated with cancer predisposition. In fact, it has been estimated that as many as one third of mutations identified by tumor-only sequencing may be false-positive germline changes, including in potentially actionable genes¹. Having an accurate picture of a tumor’s genomic makeup and contextual genetic environment is crucial to an accurate clinical assessment, which impacts therapeutic recommendations and represents the patient’s best chance for successful treatment.

In this blog we explore different strategies for enriching tumor analysis for somatic mutations and discuss why matched tumor-normal sequencing has become the preferred method.

Filtration by variants in large population databases

One approach is to use variants present in large population databases as a filter to remove likely germline variants from a tumor sample². While this practice is generally effective, it will also remove true somatic variants that happen to be identical to germline variants, resulting in a false negative. Database-driven approaches can also overlook any rare germline variants missing from large population databases due to underrepresentation of non-White individuals. These variants will remain in the sequencing data and can result in false-positive germline findings.

Focusing on variants with low allele frequency

Taking allele frequency into consideration can help. This strategy is based on the premise that an allele frequency of 50% is consistent with a heterozygous germline variant, and an allele frequency of ~100% is consistent with a homozygous germline variant¹. It then stands to reason that focusing on variants with a lower allele frequency increases the likelihood of somatic origin.

While this is true, such an approach can be complicated by many factors including contamination of the tumor sample with normal tissue, tumor heterogeneity, sequencing artifacts, difficulty mapping reads in regions of high homology, high level mosaic variants that arose early in differentiation, as well as changes in allele fraction due to copy number changes or loss of heterozygosity (LOH). Any of which can lead to inaccurate attribution of origin.

Matched tumor-normal sequencing retains somatic variants

Matched tumor-normal sequencing that pairs analysis of a tumor sample with that of a comparable, normal sample – most often from the same individual – has been shown to be a more effective strategy, yielding more reliable identification of the somatic changes specific to a tumor^1,3,4. As the name suggests, variants in the matched normal sample are determined to be germline in origin, or of alternate origin unrelated to the current tumorigenicity. When used as a filter against the tumor sample, somatic variants relevant to the cancer at hand can be identified with a high degree of confidence. Variants found at low frequencies in the normal sample can be confidently classified as false positives if they are not significantly enriched in the tumor.

While the most important function of matched tumor-normal sequencing is to identify and retain somatic mutations, it also serves other important functions.

Reducing false positives due to sample variability and sequencing artifacts

At the most simplistic level, biological samples can exhibit variability due to factors such as environmental influences, biological processes and sample handling. Matched-tumor normal sequencing provides a built-in baseline of background noise resulting from these factors, or from introduction of sequencing artifacts, that can be filtered out.

In the case of FFPE samples, extracted DNA is often fragmented and of a lower quality than fresh tissue samples. Matched tumor-normal sequencing provides a comparison that helps distinguish true alterations from noise resulting from degradation of the DNA, enhancing sensitivity.

Reducing false positives originating from CHIP variants

Cell-free DNA (cfDNA) samples, also known as liquid biopsy samples, contain DNA from tumor cells, but they also contain a significant amount of DNA from white blood cells. In many individuals, especially those who are older, these phenotypically normal blood cells contain acquired mutations subsequently increased in relative frequency due to clonal expansion. These clonal hematopoiesis of indeterminate potential (CHIP) variants often, but not always, occur in the same genes associated with blood cancers like leukemia. However, while they are indicative of an increased risk of developing a blood cancer in the future, they are not likely to be relevant to the tumor being analyzed.

Simultaneously sequencing matched white blood cells as a normal control can successfully distinguish somatic mutations that are relevant to driving tumorigenesis from somatic mutations arising from the normal process of clonal hematopoiesis⁴. This is such an important consideration that both ESMO and AMP guidelines specify that matched white blood cell sequencing should be used for interpretation of somatic variants in cfDNA testing^5,6.

Removal of false positives arising from CHIP variants is not only important for accurate cfDNA analysis, but also FFPE analysis. In a study by Memorial Sloan Kettering Cancer Center (MSK) investigators, matched tumor-normal sequencing results showed that 5.2% (912/17,469) of patients with advanced cancer would have had at least 1 clonal hematopoietic (CH)-associated mutation erroneously called as tumor-derived in the absence of matched blood sequencing⁷. Of these CH variants, 49.7% of them were classified as oncogenic or likely oncogenic based on OncoKB™, and 3.2% were associated with approved or investigational therapies (e.g. mutations in IDH1/2). Failure to recognize such mutations as blood-derived may result in inaccurate precision therapy recommendations.

Streamlining germline variant analysis

The ability to distinguish between somatic and germline variants has the additional benefit of streamlining analysis of germline variants which have additional implication for a patient’s clinical care. Notably providing information about future disease risk which can be managed in part through surveillance as well as allowing for testing of family members who may also be at risk for disease.

It is for the reasons discussed here that MSK-ACCESS^® powered with SOPHiA DDM™ for liquid biopsy and MSK-IMPACT^® powered with SOPHiA DDM™ for comprehensive genomic profiling (CGP) utilize the matched tumor-normal analysis strategy to accurately delineate somatic variants from germline and CHIP variants.

Contact us to learn more about adopting advanced liquid biopsy and CGP technology in your laboratory.

References

1. Jones, S, et al. Sci Transl Med. 2015. 7(283):283ra53.
2. Mandelker, D, & Ceyhan-Birsoy, O. Trends Cancer. 2020;6(1):31-39.
3. Cheng, D.T, et al. J Mol Diagn. 2015;17(3):251–264.
4. Brannon, A.R, et al. Nat Commun. 2021;12:3770
5. Pascual, J, et al. Ann Oncol. 2022;33(8):750-768.
6. Lockwood, C.M, et al. J Mol Diagn. 2023;25(12):876-897.
7. Ptashkin, R.N, et al. JAMA Oncol. 2018;4(11):1589–1593.

The ESMO Precision Medicine Working Group (PMWG) first published its recommendations for when to use next-generation sequencing (NGS) in routine practice for patients with metastatic cancers in 2020¹. At that time, based on identification of recurrent genomic alterations in the eight most deadly cancers and their ranking on the ESMO Scale for Clinical Actionability of molecular Targets (ESCAT), NGS was recommended for advanced non-squamous non-small-cell lung cancer (NSCLC), prostate cancer, ovarian cancer and cholangiocarcinoma. It was additionally noted as an alternative to PCR for advanced colorectal cancers.

Advances in precision medicine during the past four years have resulted in revised ESCAT rankings for a number of biomarkers, leading the PMWG to reconfirm their previous recommendations and to expand the set of metastatic cancers recommended for NGS testing. The recommendations now include NGS testing for advanced breast cancer as well as the following advanced rare cancers: gastrointestinal stromal tumor, soft-tissue sarcomas, thyroid cancer².

Due to the potential of NGS to help identify the primary tumor, plus its positive impact on patient outcomes, the PMWG additionally recommends that NGS testing be performed for cancers of unknown primary.

In countries where targeted therapies are accessible, recommendations for NGS testing now further extend to detection of the following tumor-agnostic biomarkers:

• NTRK1,2,3 fusions
• RET fusions
• FGFR1/2/3 fusions and mutations
• BRAF V600E mutations
• MSI-H (microsatellite instability-high)
• TMB-H (tumor mutational burden-high).

The PMWG stresses the importance of carrying out NGS testing in clinical research centers as well as ensuring that the selected test includes the actionable biomarkers of interest. They specifically flag the importance of assessing the chosen assay for its ability to detect fusions and homologous recombination deficiency (HRD), when relevant, as not all assays have these capabilities. The PMWG also highlight the assay’s ability to identify clonal hematopoiesis (CH) as an important consideration. High-risk CH can be found in patients with solid tumors, especially when plasma cell-free DNA sequencing is performed. To be considered as CH of indeterminate potential (CHIP), the somatic variants of haematological malignancy-associated genes should be with a variant allele fraction (VAF) of ≥2%.

See the full set of guidelines here: ESMO Recommendations for the use of NGS for patients with advanced cancer.

Our SOPHiA DDM™ applications support the guideline-recommended use of NGS

At SOPHiA GENETICS, we offer solutions that cover the major biomarkers highlighted in the ESMO recommendations, including the key fusions noted above. Moreover, the PMWG recognizes SOPHiA DDM™ Dx HRD Solution as one of only a few methods validated for HRD detection in advanced ovarian cancer.

Additionally, we offer liquid biopsy and solid tumor applications that leverage a matched tumor-normal sequencing approach to identify CHIP and germline variants, revealing genomic alterations of true somatic origin. Recent ESMO guidelines on reporting genomic test results for solid cancers recommend that, when feasible and with proper consent, the NGS report should specify whether alterations are of germline versus somatic origin³. With matched tumor-normal sequencing, the germline origin of any variant can be determined with certainty³.

Read our flyer for further information on how SOPHiA DDM™ for Solid Tumors advances clinical research by aligning with guideline recommendations: View the flyer

References

1. Mosele F, et al. Annal Oncol. 2020 31(11):1491-1505.
2. Mosele F, et al. Annal Oncol. 2024 S0923-7534(24)00111-X.
3. van de Haar J, et al. Annal Oncol. 2024 doi.org/10.1016/j.annonc.2024.06.018.

Tell us a bit about your laboratory.

The Genetics and Genomics Laboratory in which we operate belongs to a Local Health Authority in the Sardinia region. We provide molecular evaluation of gene alterations in many solid neoplasms to identify patients eligible for personalized treatments (i.e. targeted therapy). This service is provided to medical oncology departments in the city of Cagliari and many peripheral areas in the region.

What are the roots of your collaboration with SOPHiA GENETICS, and which application are you using?

The decision to collaborate with SOPHiA GENETICS was made in 2019, as our request to activate our NGS service was subject to CE-IVD certification of the library preparation and sequencing path. Furthermore, the process seemed quite linear to us.

We are currently using the SOPHiA DDM™ RNAtarget Oncology Solution (ROS) for RNA sequencing.

Can you describe your experience of using SOPHiA DDM™ ROS in routine for your RNA sequencing?

It has an excellent ability to identify fusion gene events involved in many oncological diseases. The application allowed us to identify several known and unknown molecular alterations (single nucleotide variants, Indels, gene fusion and exon skipping events) involved in various cancers, which represent biomarkers for approved and agnostic target drugs. In this way, it is possible to perform a combined analysis of different tumor pathologies in a single analysis session.

“It has an excellent ability to identify fusion gene events involved in many oncological diseases.”

Can you describe your experience with automating the workflow?

The automation of the workflow for library preparation has allowed us to standardize the quality (e.g. coverage of control genes, average size of transcripts, coverage of target regions).

In addition, we have solved the problem of complexity in library preparation. The three days of work required for preparation of libraries, up to the loading of the pools, have been reduced to only a few steps. The manual procedure now involves the set-up of the automation tool, loading of the pools into the sequencer, and approx. 2–3 days for the analysis of the data (for 16–24 samples).

“[With automation], we have solved the problem of complexity in library preparation. The three days of work required for preparation of libraries have been reduced to only a few steps.”

What are the greatest benefits of using SOPHiA DDM™ ROS?

The greatest benefit of this application is the advanced variant filtering capability and pathogenicity pre-classification that optimizes the data analysis flow and facilitates data interpretation.

In addition, the creation of a particularly exhaustive report for clinical information and ongoing trials relating to the pathologies under examination, make the results usable by the clinician for the therapeutic choice and prognosis of the disease.

We’d like to thank Manuela Badiali, Rita Congiu, and Stefania Murru for their time and for sharing their experience. Click here to learn more about SOPHiA DDM™ RNAtarget Technology and request a demo!

The Laboratory of Genetics and Genomics use a CE-IVD version of SOPHiA DDM™ ROS called SOPHiA DDM™ Dx ROS, available in in the European Economic Area (EEA), the United Kingdom and Switzerland. The CE-IVD application is not designed for use with automation – The Laboratory of Genetics and Genomics have validated the automated workflow for clinical use. SOPHiA GENETICS products are for Research Use Only, not for use in diagnostic procedures unless otherwise specified.

We were delighted to meet with the two presenters of this poster, Elizabeth Ratsma, Pre-Registration Clinical Scientist - Cancer Genomics and Charlotte Flanagan, PhD, Innovation Lead at The Royal Marsden NHS Foundation Trust. The poster was recently presented at the ESMO Gynaecological Cancers Congress 2024 in Florence, Italy.

We would like to warmly thank them for their insightful presentation and for sharing the key takeaways of their study and experience with us through this following interview.

Dear Elizabeth and Charlotte, could you please share with us more information about the scope of this study? What was the aim of it?

Homologous Recombination Deficiency (HRD) testing is available to patients in the NHS who have been newly diagnosed with high-grade epithelial ovarian cancer.

Prior to December 1, 2023, eligible patient samples were sent to the United States for testing via Myriad. The aim of this study was for the North Thames Genomic Laboratory Hub to establish an alternative in-house method for testing ovarian FFPE samples for genomic instability and BRCA mutation status. This would provide an HRD score to enable patients to access PARP inhibitors.

How was your experience with implementing the SOPHiA DDM™ GIInger Genomic Integrity Solution?

We had experience using the SOPHiA DDM™ Platform from previous demos trialing the SOPHiA DDM™ Dx HRD Solution. The SOPHiA DDM™ GIInger Genomic Integrity Solution (or GIInger™) is accessed in a similar process, and it was not difficult to navigate this new workflow in the software. Our bioinformatics team developed a CLC script with support from SOPHiA GENETICS to upload our low-copy WGS sample FASTQ files into the SOPHiA DDM™ Platform, where they are analyzed and genomic instability scores are generated.

SOPHiA GENETICS were receptive to our suggestions to help optimize functionality for use in a clinical diagnostic setting. For example, updating the format and accessibility of reports to facilitate pairing the results with our in-house tBRCA reporting (RMH200, Roche analyzed using DRAGEN, Illumina) and LIMs system.

We have received good technical support when needed through the SOPHiA GENETICS JIRA system.

"We are looking forward to future updates, including a web-based portal and automated download of result files for our clinical scientists to access with ease."

In the scope of this study, you compared GIInger™ and the SOPHiA DDM™ Dx HRD Solution. Could you please summarize the purpose and outcomes of this comparison?

The SOPHiA DDM™ Dx HRD Solution is a CE-marked HRD solution that utilizes SOPHiA GENETICS’ preferred chemistry for whole genome and capture library preparations. At Royal Marsden Hospital (RMH), we had previously implemented a robust NGS protocol (RMH200) capable of producing BRCA capture and low-copy whole genome libraries. To optimize operational efficiency, we decided to explore using our existing automated chemistry and library preparation workflow and opt for a bioinformatics solution to analyze the whole genome data for genomic instability to provide a HRD status.

To this end, 23 samples (previously tested via an orthogonal method) were sequenced using our in-house NGS chemistry and analyzed using GIInger™ paired with our in-house tBRCA calling and compared to the SOPHiA DDM™ Dx HRD Solution, which utilizes different chemistry and a full bioinformatics solution. Reassuringly, we found 100% concordance.

Thank you for sharing these insights with us! To conclude this spotlight, we would like you to share the key takeaways of this study.

Utilizing the SOPHiA GENETICS GIInger™ bioinformatics solution paired with our in-house RMH200 panel for tBRCA status, RMH successfully launched in-house HRD testing in December 2023. By April 1, 2024, we had tested 106 samples internally, achieving recent average turnaround times of less than 21 days.

"The support provided by SOPHiA GENETICS has been sufficient and rapid, which has been invaluable during the first six months of this new service."

We would like to thank Elisabeth and Charlotte for their participation in this spotlight.

Learn more about:

SOPHiA DDM™ GIInger Genomic Integrity Solution

SOPHiA DDM™ Dx Homologous Recombination Deficiency (HRD) Solution

Disclaimer

SOPHiA GENETICS products are for Research Use Only and not for use in diagnostic procedures unless specified otherwise. The SOPHiA DDM™ Dx Homologous Recombination Deficiency Solution is available as CE-IVD product for In Vitro Diagnostic Use in European Economic Area (EEA), the United Kingdom and Switzerland.

Assessment of Homologous Recombination Deficiency (HRD) and BRCA mutational status in high-grade serous ovarian carcinoma (HGSOC) samples has become crucial for personalized medicine, guiding treatment decisions, and predicting response to specific therapies, such as PARP inhibitors.

The aim of this multicentric study, published at this year’s European Society of Gynaecological Oncology (ESGO) conference was to assess the reliability and reproducibility of SOPHIA DDM™ Dx HRD Solution across three different hospitals (Hospital del Mar, Hospital Vall d’Hebrón and Hospital Clinic de Barcelona) to consider its implementation as a decentralized solution in routine molecular diagnostics.

Our Spanish team led by Giovanni Velotta, CS Manager at SOPHiA GENETICS, was happy to meet Dr. Gardenia Vargas, Molecular Geneticist and Responsible for Hereditary Cancer and Rare Disease Molecular Diagnosis at Hospital del Mar in Barcelona, Spain, who joined us for an insightful interview, sharing her experience in implementing SOPHiA DDM™ Dx Homologous Recombination Deficiency Solution.

Watch or read the full interview with Dr. Vargas below:

  
Dr. Vargas thank you for joining us today! Could you please share with us more information about the scope of this multicentric study? What was the aim of it?

Before I answer I would like to thank you and thank AstraZeneca for their support in this study. I must mention that all my answers represent my own perspectives and not necessarily the official stance of the hospitals involved in this project. 
 
The study aimed to investigate the feasibility and effectiveness of implementing the SOPHiA DDM™ Dx HRD (Homologous Recombination Deficiency) Solution within the clinical setting of three hospitals. This includes evaluating various aspects such as the practicality of integrating the test into existing diagnostic workflows, the accuracy and reliability of test results, the impact on patient care and outcomes, and the potential benefits and challenges associated with in-house testing. The study seeks to provide valuable insights into the utility and suitability of adopting the SOPHiA DDM™ Dx HRD (Homologous Recombination Deficiency) Solution as a routine diagnostic tool for identifying HRD in ovarian cancer patients.

What has been the role of Hospital del Mar in this study and how was your experience being a part of this initiative? 

First of all, I must say that all three hospitals worked equally on this project, and the idea of doing it was born before I was incorporated to it.  Beatriz Bellosillo from Hospital del Mar was the principal investigator, and we had different responsibilities such as protocol writing, ethics approval coordination, securing funding to support the study's activities, and management and allocation of some resources like sequencing supplies. 
And of course, it was truly a rewarding experience collaborating with renowned hospitals and esteemed colleagues. 

What is your experience in implementing the SOPHiA DDM™ HRD Solution in your routine testing? 

Currently, while we have conducted a multicenter study to evaluate the feasibility of implementing the SOPHiA DDM™ Dx HRD Solution, we have not yet implemented it into routine diagnostic practice at my hospital.

The implementation process requires significant financial support, particularly for acquiring the necessary equipment, reagents, and personnel training. Additionally, to ensure cost-efficiency and timely responses to patients, we need to conduct sequencing runs with a minimum number of samples per run. 

Unfortunately, our hospital alone does not have sufficient sample volume to meet this requirement. Therefore, we are actively seeking financial support from government agencies to fund the implementation of the study. Additionally, we are exploring collaborative efforts with other hospitals or healthcare institutions to pool together an adequate number of samples for sequencing runs, thus reducing costs per sample and ensuring a rapid response to patients. 

By securing both financial support and an adequate sample volume, we aim to overcome these logistical challenges and proceed with the implementation of the SOPHiA DDM™ Dx HRD Solution into our diagnostic routine, ultimately enhancing our ability to provide efficient and effective patient care. 

What are the key takeaways of this study? 

It highlights the importance of collaboration among hospitals, government agencies, and other stakeholders to support the implementation of advanced diagnostic technologies into routine clinical practice. 
  
Overall, the study provides valuable insights into the challenges and opportunities associated with integrating advanced genomic testing into routine diagnostics, with the aim of improving patient care and outcomes. 

Would you consider the implementation of the SOPHiA DDM™ HRD Dx Solution as an efficient decentralized solution in routine molecular diagnostics? 

This collaborative study concludes that SOPHIA DDM™ Dx HRD Solution provides reliable and consistent results across different hospitals and sequencing runs. 

These findings contribute to the growing body of evidence supporting the use of the decentralized SOPHiA DDM™ Dx HRD Solution in clinical settings for genomic analysis. 

 
 
DISCLAIMER: SOPHiA GENETICS products are for Research Use Only and not for use in diagnostic procedures unless specified otherwise. SOPHiA DDM™ Dx Homologous Recombination Deficiency Solution is available as CE-IVD product for In Vitro Diagnostic Use in Europe and Turkey.

Explore this infographic summary to learn more about the machine learning model developed by Margue et al. for the prediction of disease-free survival in patients undergoing surgery for renal cell carcinoma.

Margue G, et al. UroPredict: Machine learning model on real-world data for prediction of kidney cancer recurrence (UroCCR-120). NPJ Precis Oncol. 2024 Feb 23;8(1):45.

SOPHiA GENETICS multimodal analysis technology and concepts in development. May not be available for sale.