The Promise of Precision Profiling
A roundtable with Marc Peeters, Özlem Er, and Philip Beer
Michael Schubert | | Longer Read
Precision medicine is an increasingly common buzzword in healthcare – but what does it mean to pathologists and laboratory medicine professionals? How does it affect the patients who seek out molecular profiling – whether for cancer treatment, pharmacogenomics, or other reasons? And what is its outlook in the near future and beyond? Three experts gather to discuss molecular pathology, precision oncology, and the future of diagnostic and prognostic medicine.
What is your background in precision medicine – and how do you use it in your work?
Marc Peeters: Over the years, my profession as an oncologist has changed dramatically. At the beginning of my career, we were happy to have any drugs at all to treat our patients. Nowadays, it’s a question of selecting which of a wide array of drugs will be best for each individual patient.
Certain types of tumors have a limited number of biomarkers that we routinely test for treatment selection. In other cases, we’re searching for a solution for patients who have become resistant to standard lines of treatment. We need biomarker information to locate clinical trials or explore drugs that, although not indicated in the patient’s tumor type, might yield a response in that specific patient based on their biomarker profile. At the moment, we don’t fully profile every patient at my institution (beyond the routine testing); it’s reserved for those who have exhausted the standard options. Smaller sets of biomarkers are tested if they are clearly linked to a registered (and thus reimbursed) drug.
Özlem Er: I am a clinician trained as an internal medicine and medical oncology specialist treating solid tumors. As a medical oncologist whose goal is to improve the therapeutic options of cancer patients, I use several tests to detect actionable targets in tumors. These tests can be performed on either blood or tissue samples. It’s an approach that I hope will expand to many areas of oncology in coming years until it’s common in the clinic. The more reliable biomarkers we have to help select appropriate candidates and match treatments to their molecular profiles, the more this kind of testing will improve the lives of people with cancer.
Philip Beer: I’m a clinician. My initial training was in hematology and I’m dual-accredited in internal medicine and pathology, which means I have both diagnosed and treated patients with leukemia.
Hematology was ahead of the game in using genetics to guide therapies – after all, the first really famous targeted therapy was imatinib for chronic myeloid leukemia. In hematology, clinical trial enrollment rates are 20–40 percent for adult leukemia (versus 5–10 percent for common solid tumors). That said, the solid tumor space is catching up; the genomics are a bit more complex, but the diseases are far more common, so there’s a lot to be gained from better biomarkers and corresponding treatments. And that’s the space I’ve been working in almost exclusively since 2014. I have a deep interest in using genomics and other complex biomarkers to accelerate cancer drug discovery – and I’m pleased to be at the interface between healthcare delivery and drug discovery.
There are historical barriers between academia and the commercial sector – and, indeed, between healthcare delivery and commercial diagnostics and drug discovery. I hope stories like mine, where people overcome those barriers, become more common. In fact, I would like to see those barriers come down altogether. People should be free to move between the different sectors. We have a lot to learn from one another!
What is the current state of precision medicine in Europe – and how does it compare to the rest of the world?
MP: I think precision medicine, for the moment, is still a work in progress. Each country has its own management approach; some have a central organization, whereas in others adoption depends on individual institutions with the resources to integrate precision medicine into daily practice.
The desire to take up precision medicine is there, but we must work to translate that clearly into clinical practice. Although some big companies are working to make information available to physicians for treatment decision-making, we have to be careful about how we use the data. First of all, the information that we get from genomic analysis cannot always be linked to specific tumors or therapies. And, second, it’s possible to rely too much on these analyses to guide our decisions, especially when they are new and unfamiliar. In some cases, it’s better to stick to the standards and integrate new information judiciously.
That said, it’s a fast-moving field. If we talked about this again in three to five years, it would be a totally different conversation!
ÖE: Precision medicine in oncology is relatively new worldwide – and becoming more important every day. By definition, precision medicine is “an emerging approach for disease treatment and prevention that takes into account individual variability in genes, environment, and lifestyle for each person (1).” This approach allows us to decide more accurately which treatment strategy will work for a specific patient by taking into consideration the genomic differences between tumors.
Precision medicine in oncology is more common in North America and Europe than in other parts of the world. As populations age, cancer incidence increases – and so, in turn, does the need for cancer treatment. In the long run, genomic testing will increase demand for tailored cancer therapeutics.
PB: I think we are at an interesting point in Europe. We’ve made some early gains – but I think the low-hanging fruit has been plucked. We have had successes with therapies like EGFR inhibitors for lung cancer, which have made a big impact. The recent success with NTRK inhibitors may well be the last time that we have really powerful results, though. Now, we need to move beyond the single-gene/single-drug model. It’s time for comprehensive genomic profiling to inform targeted therapies.
We’ve had a number of fairly high-profile failures in the targeted therapy realm, too. A recent example is IDO inhibitors, which failed in a fairly spectacular fashion in late-phase clinical development. Although the biological rationale for IDO inhibitors is strong, clinical trials were not backed up by comprehensive biomarker studies to understand how these drugs work in humans or if their activity is limited to a subset of patients. And that’s why I feel we need to change the infrastructure of precision medicine to get to the next level; we need to bring in complex genomic profiling and use the data to make better use of therapies.
In terms of the rest of the world, America generally leads the way when it comes to medical progress – mainly through sheer scale of numbers and clinical trial infrastructure. In the large hospitals, where they pull in enormous numbers of patients, they can run complex clinical trials more easily. But there are disadvantages to taking up genomics through centralized providers. Not all patients have equal access, which means that the studies don’t necessarily include all relevant populations. Europe has the potential to “get ahead” by embedding complex genetic profiling within the publicly delivered healthcare system so that it’s available to everybody. We are still a little bit behind in Europe at the moment – and funding is always an issue – but I think we have a unique opportunity to overcome the infrastructure barriers other systems may face.
I think that the key to unlocking precision oncology lies in better profiling of patients. Genomics is a big part of that and immunotherapy will also bring in other profiling techniques. What we need, though, is to routinely perform these techniques on all cancer patients referred into hospital, rather than waiting until they’re in a clinical trial. The other important part is that we collect outcome data. We need to match complex biomarker profiles not only to the treatments our patients receive, but also to what actually happens as a result. It may be easier to do that in a socialized healthcare system, such as we have in Europe, than to get the big institutions to harmonize and share data in the context of the American model. By ensuring equitable access to therapy and collecting outcome data, we can work out the true significance of complex biomarkers.
How will precision medicine change in the near future?
MP: Genetic profiling is rapidly moving forward and will be a significant part of our daily routine in the next few years. Bioinformatics is also becoming much stronger. The rest is still a work in progress. We need to work on integrating all of the available data into a clinical interpretation. We have yet to fully optimize our methodologies, and there’s still a lot we don’t know.
If we talk about the microbiome, for instance, it’s clear that we’re just at the beginning of our scientific exploration. What information do we get out of the microbiome? Does it help us develop or select treatments? Can we modify the microbiome in therapeutic ways? These types of questions remind me of where we were only a few years ago with genetic profiling – but now, tumor genetic profiling based on tissue or even liquid biopsy is becoming routine. The outlook is promising.
At the moment, we focus our profiling efforts mainly on treatment response and efficacy – but we sometimes forget to consider toxicity. Can we identify patients who should receive modified doses because they are at greater risk of toxicity? I think so – and that’s an area in which we’ll make a lot of progress in the future.
PB: An interesting thing that’s going on in a few different European countries is a restructuring of genomic services. If that’s done properly, it will probably take the better part of five years to get useful systems up and running, let alone big data collection – but at least people are beginning to think about the practicalities of making it happen.
The process of precision oncology is happening hand in hand with an interest in real-world data. There are companies in America set up solely to aggregate real-world patient record data – and, by doing it at scale, they’ve defined patterns of physician behavior, tracked prescription habits, and identified some interesting therapeutic connections. It will take a while for us to learn how to use real-world data effectively, but – if things go well – we could be in a great place in five years to accelerate the process of drug discovery.
We live in uncertain times for a number of reasons, so it can be difficult to look much farther ahead than that. But look at immunotherapy; it hasn’t been long since people working in that field were considered charlatans and outcasts by mainstream scientists. And now – almost out of nowhere – immunotherapy has completely changed the face of cancer treatment. I don’t think anyone, perhaps even the experts, expected quite such a sea change. So I hope that, in the next decade or so, we might be in a similar situation. I’d like to see individualized treatment based on patients’ biomarker information become standard practice.
With the advent of cheaper, better sequencing, our understanding of inherited diseases is improving by leaps and bounds. England is now considering a move to whole genome sequencing as a standard test for suspected inherited disorders. On a wider scale, pharmacogenomics is an interesting space to watch. We’re aware of a few markers with strong predictive effects right now but, when the effects become weaker, we need to use “polygenic risk scores” that combine many genomic features to predict a patient’s risk or response. It all comes down to data generation – and we don’t yet have the enormous amounts of data we need. Nevertheless, I do think how patients’ genomes impact their drug responses – and how we can use that information to target treatment – will be a big story over the next 10 years.
It seems clear to me that bioinformatics will be the skill of the future – and not just pure bioinformatics, but everything that goes along with it: data architecture, natural language processing, and a variety of computational skills.
What are precision medicine’s greatest strengths… and weaknesses?
MP: Its greatest strength at the moment lies in situations where patients have driver mutations with clear links to known therapies. Identifying a patient’s individual mutations can create new treatment options – some that offer dramatic improvements.
Its weakness lies in the challenge of integrating information and increasing access in routine clinical care – not just in big academic institutions, but also in smaller centers that handle more routine cases. I think an important task for key opinion leaders is to ensure that all clinicians have access to this information and all patients receive optimal treatment. Another weakness is the current strong focus on genetics and genomics; to see the full benefit of precision medicine, we need to take a multidisciplinary approach that integrates proteomics, the microbiome, and more. Precision medicine made its first step in cancer, but the future is much more complex.
ÖE: Precision medicine’s greatest strength is increased cancer treatment efficacy accompanied by less toxicity than conventional chemotherapy. It also creates new treatment options for chemorefractory patients. However, because genomic tests are not reimbursed by insurance companies, we need financial solutions to enable routine testing.
PB: I think the greatest strength is that it has the potential to completely change the face of cancer therapy. We hear about amazing breakthroughs all the time in the papers – but the actual path will be slower and more laborious, of course, and it will have ups and downs. Despite all that, the breakthroughs we’ve made so far in immunotherapy are extraordinary. For instance, nearly half of all patients with metastatic melanoma are now being cured by immunotherapy. NTRK inhibitors are now also approved to treat cancers in many different tissue types. We’ve begun to realize the long-held thought that cancer is a molecular disease, rather than a tissue disease.
Unfortunately, we’re still stuck in old paradigms; patients don’t get comprehensive biomarker profiling and, when they do, the data aren’t necessarily good quality. As a result, I still have a slight concern that precision oncology will be written off before we can really ask the important questions. What is the role of genomics? What is the role of complex biomarkers? How can we find the right drug for each patient? We’re beginning to ask these questions, but there’s still a long way to go. For me, the big downside at the moment is that people may become disillusioned with precision medicine – before we’ve understood the full potential.
Can you share any anecdotes from your experience with precision medicine?
MP: The information we got from tissue 20 years ago was relatively limited. If you consider the information available to us today – not only from tissue, but also from new techniques like liquid biopsy – we’ve made dramatic progress. Along with that progress have come changes in mindset; now, the information that we oncologists get from pathologists is becoming much more important. It’s no longer used only for diagnosis – we need the information to guide treatment decisions in certain patients. A few years ago, we couldn’t adequately treat many of our patients with kidney cancer; today, we can link a specific marker to a specific immunotherapy treatment – a sea change in which we now see dramatic responses and achieve disease control we never previously imagined.
We see the same situation with NTRK fusions and the drugs targeting the TRK pathway. These patients, who were previously poor candidates for systemic treatment, now show responses of 60 to 70 percent in some cases. It’s clear that the current diagnostic and prognostic landscape is very different to what it was just a short time ago.
ÖE: Metastatic non-small cell lung cancer (NSCLC) has a poor prognosis; statistics show a median survival of 12 to 18 months. But, thanks to precision medicine, one of my patients with metastatic NSCLC survived for over five years after diagnosis. His tumor was EGFR-positive, confirmed with genomic testing, so he was treated with an EGFR tyrosine kinase inhibitor for almost two years before the disease progressed. While on this oral medication, he was asymptomatic, performance status ECOG 0, and lived life as if he were disease-free. Upon progression, we did another genomic test, which revealed a resistance mutation – EGFR T790M – whose activity can be inhibited by osimertinib, another oral tyrosine kinase inhibitor. The patient was able to continue treatment with oral medication and manageable side effects that didn’t affect his daily life. It wasn’t until his fourth year of metastatic disease that he progressed to the point where chemotherapy was needed. Not only does a survival of greater than five years show a clear advantage to this treatment approach but, for much of that time, the patient and his family were able to enjoy an excellent quality of life.
PB: I’m involved in the molecular profiling of patients for several projects. The more you provide patients with complex molecular genetic profiling as a matter of routine, the more you see things you might not otherwise have noticed. You might identify therapeutic pathways that wouldn’t otherwise have been considered. Recently, we picked up an inherited predisposition with consequences to the patient and their family – but not one that was typically associated with the patient’s (rare) cancer type. You might also unlock access to clinical trials – a frequent occurrence. In perhaps one in four patients who come to us for genetic profiling, we find something interesting that can potentially take them down a different avenue for therapy.
The barriers lie in changing the attitudes of both physicians and patients. It can be difficult to encourage people to consider clinical trials when making decisions about their treatment. In particular, we should be considering them earlier in the therapeutic pathway; clinical trials are still seen by the general public as a bit of a “last-ditch attempt” at treatment. There’s a misconception that the patient is not going to benefit from the trial and that it’s only for the benefit of future patients. But, more often than not, that’s untrue, and the persistence of that false belief means that the number of patients registering for clinical trials is quite small – which, in turn, slows down progress in drug development.
But it’s not only patients who view clinical trials as “the end of the road.” Physicians, especially those who don’t work in large academic centers, often seem to have a similar view. I’ve noticed that the majority of oncologists – certainly in the UK and the US – still very much stick to chemotherapy-led pathways. Immunotherapy is making inroads and, in specific areas, targeted therapies are beginning to expand – but I think we still have a little way to go to persuade oncologists that clinical trials should be thought of early in the treatment selection process. Molecular profiling can be expensive, so patients tend not to be profiled until they’ve run out of traditional treatment options and are in poor health. As you can imagine, this affects their options in terms of selecting effective treatments. I think it would be beneficial to get patients access to genomic profiling earlier, find out what clinical trials are available, and consider them as part of the standard of care for cancer therapy. We shouldn’t wait to explore potentially promising options until there’s nowhere else to go.
If you could make one big change (in your field) to advance precision medicine, what would it be?
MP: I would integrate multidisciplinary decision-making into a model of pathology that goes beyond just precision medicine; that’s where we have to go in routine clinical practice, but we’re not there yet. At the moment, the majority of centers have molecular tumor boards, which are mostly separate from the multidisciplinary tumor boards in which we discuss patients. In the future, we need to bring those two entities together – at least when we’re discussing tumor types that benefit significantly from molecular profiling.
PB: For precision oncology, I would prioritize genomic profiling of all cancer patients (or at least all patients with advanced metastatic disease) at diagnosis. And, following closely on the heels of that, I would begin collecting outcome data at a national level. It’s a lot of work but, without it, I think precision oncology will stall. Despite our early successes, we’ll have an increasing number of clinical trials that have failed because we didn’t select the right patients. And that’s why I think widespread genomic profiling and data collection must be the next challenge we tackle.
What advice do you have for pathologists and laboratory medicine professionals with an interest in precision medicine?
MP: Networking is important. This type of technology is expensive and not everyone has equal access to it. Additionally, knowledge is becoming much more specialized than before, so it’s no longer possible for one person to have all the answers – having contacts with different areas of expertise is vital. Finally, investment is key, especially if you want to do routine molecular pathology. You need technology robust enough to give you the information your patients need. For those working in academic centers, it’s important to focus on new predictive and prognostic markers, and to consider markers for toxicity as well as for efficacy.
Funding is often a sticking point, but the cost of gene profiling has dropped dramatically over the last five to 10 years because we’re using it so much more. I expect that pattern will continue. Most of the major pharmaceutical companies are also investing a lot in companion diagnostics, which should further our goal of getting the right drug to the right patient at the right time and with minimal toxicity.
ÖE: Pathologists and laboratory medicine professionals are important members of the multidisciplinary team in the management of cancer patients – and their importance will only increase as precision medicine evolves. As a clinician, I recommend that they take an active part in clinical tumor boards – especially molecular boards – to help appropriately direct each patient’s treatment options.
Remember that we are not just doctors, but also potential patients. Cancer is a major global health problem – and almost half of all cases can be prevented by good lifestyle choices: not smoking, moderating alcohol consumption, exercising, maintaining a healthy weight, eating a balanced diet, and protecting ourselves from the sun. That said, it’s also important to be aware of family history and take preventative measures against known genetic risks – and, of course, to follow the relevant cancer screening recommendations.
PB: For these efforts to be truly successful, they need to happen at a national level. We are beginning to build those infrastructures, but it’s a little fragmented because bringing everybody together is difficult. As a pathologist, I can say that colleagues of mine have seen this as a threat. They feel that a lot of pathology skills could become less relevant and practitioners will increasingly rely only on molecular profiling.
I disagree, though. I think history has shown us more than once that people don’t just suddenly find themselves out of a job. Things don’t change that quickly. But you do need to “go with the flow” – and right now, to me, that means encouraging pathologists to become more genetically aware. We need to understand these novel technologies and embrace them as a part of our future routine work. As medicine evolves, to protect the future of pathology, we need to evolve along with it.
What’s your key take-home message about precision medicine?
MP: It’s clear that the profile of the pathologist has changed dramatically over the last five to 10 years.
ÖE: Precision medicine is very important for the management of oncology patients, and its impact will only increase in the near future.
PB: Genomics is here to stay. It is going to touch on many aspects of cancer – and beyond, for instance in pharmacogenomics. So what I want to get across is that it is not something to hide from, but something to understand and embrace. Get involved!
Marc Peeters is Professor and Head of Oncology at Antwerp University Hospital, Coordinator of the Multidisciplinary Oncology Center Antwerp, and Chairman of the Department of Oncology at the University of Antwerp, Belgium.
Özlem Er is Professor and Head of the Department of Medical Oncology at Maslak-Acibadem University Hospital in Istanbul, Turkey.
Philip Beer is a specialist in precision oncology and holds leadership roles at various public and private centers in the United Kingdom.
- P Garrido et al., “Proposal for the creation of a national strategy for precision medicine in cancer: a position statement of SEOM, SEAP, and SEFH”, Clin Transl Oncol, 20, 443 (2018). PMID: 28861725.