Генетический анализ опухолей поможет определить тактику лечения метастатического рака простаты

Q&A: Tumor DNA classification a ‘valuable’ prognostic tool for metastatic prostate cancer Key takeaways: - Men with metastatic prostate cancer who had favorable genetic profiles had significantly longer survival. - Tumor classification could help identify patients needing aggressive or de-escalated care. Tumor DNA profiling could be used to identify which men with metastatic prostate cancer need aggressive treatment and those who may benefit from de-escalation. A model, trained and tested on nearly 2,500 veterans with metastatic hormone-sensitive prostate cancer, found those with favorable germline profiles had a median OS more than 4.5 years longer than those with unfavorable classifications. “All men with metastatic prostate cancer should be getting a biopsy of a metastatic lesion. It clearly makes prognostic difference,” Kara N. Maxwell, MD, PhD, assistant professor of medicine in hematology-oncology and genetics at Penn Medicine, told Healio. The Somatic Tumor Risk Assessment for OS-Prostate (STRATOS-P) classification, which used 16 gene mutations associated with prostate cancer, showed men with favorable profiles (median OS = 87 months) had significantly lower risk for death compared with those with intermediate (median OS = 52 months; adjusted HR = 1.54; 95% CI, 1.33-1.78) and unfavorable (median OS = 31 months; aHR = 2.37; 95% CI, 1.97-2.49) classifications. In a validation cohort of non-veterans, patients with favorable profiles also had significantly improved OS compared with those with intermediate (aHR = 2.45; 95% CI, 1.87-3.21) and unfavorable (aHR = 4.37; 95% CI, 3.06-6.22) classifications. Healio spoke with Maxwell about the impact tumor DNA profiling could have for prostate cancer, future research and more. Healio: Why did you choose to investigate comprehensive genomic profiling in prostate cancer? Maxwell: Prostate cancer is a number of different diseases. Some men have really aggressive prostate cancer and can quickly progress through treatments, even a lot of our newer treatments. They are the minority, though. Most men, even those with metastatic prostate cancer, can live many years. Androgen deprivation therapy is really hard on patients, the symptomology that occurs, the cardiovascular risk. We want to be able to stratify which men really need aggressive treatment for the rest of their lives for their metastatic castrate-sensitive prostate cancer, and which men could we potentially de-escalate therapy. That’s the underlying goal. There are a lot of studies that have looked at this using RNA-based profiling, but now that we do so much DNA sequencing, we wanted to see if we could leverage what we’re already doing in the metastatic setting for potential targeted therapies to then also use it as a prognostic marker. Healio: What was your reaction to the results? Maxwell: It’s really exciting. When you learn how to use statistics to interpret genomic data, you can see between the lines and understand more about patients and their cancer. Patients with favorable risk cancers had a median survival of 87 months compared with 31 months in the unfavorable group. That’s many years difference. Healio: How can this information be used to make treatment decisions? Maxwell: All men are getting androgen deprivation therapy. That’s how we treat prostate cancer. For the unfavorable group, we probably need to give more. We already know this from a lot of other studies, and we have options. We have fancy new treatments like lutetium Lu-177 vipivotide tetraxetan (Pluvicto, Novartis) for prostate-specific membrane antigen-positive disease. On the flip side, we also want to know who we can de-escalate and maybe give a shorter course, or give holidays. There are a lot of studies of men who are on androgen deprivation therapy that feel tremendously better with breaks. It’s really exciting to look and say, this is a group of men who are already metastatic, who are being told this is an incurable disease, but we could figure out with your genomics that you might have really responsive disease. It’s exciting to think about. Healio: Was the model more prognostic than you expected? Maxwell: What was unexpected was how well the individuals did that don’t have any genomic alterations. Not only is it important for how we think about treatment, but when designing a study of a new prognostic or predictive biomarker, you want to make sure your comparator group is appropriate genomically in comparison to whatever group you’re working on. I know that sounds obvious, but this shows how much heterogeneity there is. If you’re doing a study with a small number of patients with castrate-sensitive prostate cancer for some new therapy, there would be huge heterogeneity if you didn’t subdivide your control group by their genomic profiles. It’s going to be important when people are designing future prognostic and predictive biomarker studies to first do the genomic classification of their patients so that they’re matched from the beginning. That’ll give way more accurate results. Healio: Do these results suggest everyone with metastatic prostate cancer have their tumor biopsied? Maxwell: I was on service recently and we had this question come up. We don’t always biopsy metastatic prostate cancer. If you have a man who comes in who had prostate cancer in his past, and he now has metastases on an imaging scan, and his PSA is high, you know he has metastatic prostate cancer. Biopsy is not our standard recommendation for this individual. We need to move into this concept and socialize it generally, that patients with metastatic prostate cancer biopsy a metastatic site, and if you have to biopsy the prostate, that would also be impactful. This happened in breast cancer, too. We weren’t doing metastatic biopsies in breast cancer, and now it’s standard of care. This is part of the genomics shift. Healio: Will this have any impact on non-metastatic disease? Maxwell: That is work that some of my colleagues within the VA are looking at doing. Could STRATOS-P score at the time of definitive therapy with either a prostatectomy or radiation help us predict who will end up with either local recurrence or metastatic spread? Could it help guide adjuvant therapy? That’s a much larger patient population that is also faced with the difficult choices of whether to go on a course of androgen deprivation therapy or radiation, both of which have significant side effects? If we could show that the favorable group really never metastasizes, but they otherwise have features that would suggest they need radiation or androgen deprivation therapy, perhaps we could save them from those adjuvant treatments. The flip side would be patients who look like they have benign looking disease with their Gleason score, but they already have this genomic classification as being very unfavorable, pushing those patients to doing the adjuvant therapy. That’s the next goal. Healio: What is the big message you want clinicians to get from your study? Maxwell: If you’re taking care of patients with metastatic prostate cancer, there’s valuable information in doing a biopsy at the time of metastasis. There’s value in doing a biopsy in the castrate-sensitive setting and not just the castrate-resistant setting, which is where we currently are doing most of our biopsies. Healio: What are the next steps in research? What future questions must be answered? Maxwell: The next goal is opening a clinical trial to de-escalate therapy in men who have this favorable genomic profile. Can we use the genomic classifier to choose different therapy types? There are other studies like this that are going on, so all these data will be extremely helpful. Then in the localized setting, using STRATOS-P to help direct adjuvant therapy. Healio: Where do you think genomic classification will be in 5 years? Maxwell: I think we’re going to reach a point where the majority of patients with cancer will benefit from knowing their germline testing. That’s a controversial topic, and it’s a tough one from a healthcare utilization perspective because that’s a lot of patients. When we started doing genomics, we thought every time we do it, we’d find a different way to treat the patient. We were thinking about it from a targeted therapy standpoint, like BRAF inhibitors for melanoma, but there’s so much more rich information besides just one target, one therapy. Solid tumor oncologists have always been jealous of our liquid colleagues who had all these really useful classifications using genomics, because their cancer cells are in the blood, and they could assay them quickly and easily. We’re definitely behind in this space, but I think in 5 years, there is going to be a genomic base, whether it’s a small set of genes or a whole genome signature, that can at least prognosticate all these types of tumors, and then, hopefully have very robust data that tell you how to treat, not just with targeted therapy, but who needs aggressive vs. non-aggressive treatment. Healio: Is this approach scalable? Can it be used throughout the country? Maxwell: In the United States, it doesn’t matter rural or not, we can sequence these tumors and should be doing this testing in as many patients as possible. The cost is really small in comparison to much of what we do. You can sequence a whole genome in a research setting for $300 now. That is something that all patients, regardless of where they are, would have access to. For more information: Kara N. Maxwell, MD, PhD, can be reached at [email protected].