Read the Full Video Transcript

Neeraj Agarwal: Hi, my name is Dr. Neeraj Agarwal. I'm a Professor of Medicine and a genitourinary medical oncologist at the University of Utah Huntsman Cancer Hospital in Salt Lake City in the United States.

Today, we have Dr. Marta Garcia de Herreros from Hospital Clínic of Barcelona, who presented her data in the context of biomarkers in patients with metastatic hormone-sensitive prostate cancer during the 2024 ESMO annual meeting in Barcelona. So, we'd like to hear from Dr. García de Herreros about her presentation, and then we will discuss how it impacts our practice. So, Marta, would you like to start with your talk first?

Marta Garcia de Herreros: Thank you for the invitation. I'm Marta García de Herreros. I'm a medical oncologist at Hospital Clínic of Barcelona, and today I will discuss my work called "Tumor Suppressor Gene Signature Based on RB1, PTEN, and TP53 Predicts Early Progression in Metastatic Hormone-Sensitive Prostate Cancer Patients."

As background, alterations in tumor suppressor genes RB1, PTEN, and TP53 are associated with decreased survival and aggressive variants of prostate cancer.

TSG alterations have been mostly studied in the castrate-resistant prostate cancer setting, where they seem to benefit from platinum combination. Our group has previously developed and validated a signature of low expression of these three tumor suppressor genes, called the TSG-low signature, that is associated with poor outcomes in metastatic hormone-sensitive prostate cancer patients treated with ADT plus docetaxel.

So, let me explain briefly how we designed this signature. Basically, what we did was to assess RNA expression of each of these genes, and we assessed the score of this normalized expression, and we put a cutoff in the lower tertile of expression. So we call it TSG-low when we have the lower tertile expression of either two of these genes, and TSG wild type for the rest.

In our previous work, we explored this TSG-low signature in a multicentric cohort of metastatic hormone-sensitive prostate cancer patients treated with ADT or ADT plus docetaxel across different Spanish hospitals. We found that our TSG-low signature, either in an exploratory cohort, in a validation cohort, and even in an external validation cohort, was associated with shorter castration-resistant prostate cancer and shorter overall survival.

In the present study, we aim to assess this TSG-low signature in a cohort of metastatic hormone-sensitive prostate cancer patients treated with ADT plus androgen receptor pathway inhibitors, which could be abiraterone, enzalutamide, or apalutamide. We also aim to explore the clinical characteristics and the development of aggressive variants of prostate cancer at progression.

This is a multicenter retrospective study of metastatic hormone-sensitive prostate cancer patients treated with ADT plus ARPI. TSG expression was assessed by nCounter, and TSG-low was considered when two out of these three tumor suppressor genes presented low expression of these previously validated cutoffs, while TSG wild type was considered in the remaining cases. TSG-low expression was correlated with castration-resistant free survival—that was our primary endpoint—and also with overall survival. The presence of aggressive variants of prostate cancer was analyzed according to the established criteria by Aparicio.

Coming to our results, we included 137 patients; 96 of them received abiraterone, 38 enzalutamide, and 3 apalutamide. Nearly 80% of our patients were de novo stage IV, 20% had visceral metastasis, and 33% were considered high-risk disease. 17% of our patients were classified as TSG-low according to our signature, and we found that the TSG-low signature was associated with a higher frequency of high-risk disease.

Regarding patient outcomes, we found that TSG-low expression was correlated with shorter castration-resistant prostate cancer-free survival, which was statistically significant. As you can see here, for patients that were considered TSG-low, the median castration-resistant prostate cancer-free survival was 28 months, while the median wasn't reached for patients that were TSG wild type. We couldn't find significant differences in overall survival.

Next, we assessed the characteristics of these patients that developed castration-resistant prostate cancer according to the aggressive variants of prostate cancer criteria. We found that nearly 30% of patients developed aggressive variants of prostate cancer at castration-resistant prostate cancer progression. These patients that developed aggressive variants of prostate cancer presented a significantly higher frequency of TSG-low, suggesting that maybe the TSG-low signature can predict the development of these aggressive variants. When we looked specifically at the PTEN-low plus RB-low expression, we found that all patients that presented low PTEN and RB expression presented aggressive variants of prostate cancer, and that was, of course, statistically significant.

Our conclusions are that TSG-low expression is associated with early castration-resistant prostate cancer progression in metastatic hormone-sensitive prostate cancer patients treated with ADT plus ARPI. TSG-low was associated with a higher frequency of aggressive variants of prostate cancer, and this may help to predict the development of these aggressive variants at castration-resistant prostate cancer progression. We believe that TSG-low could be a potential biomarker to predict adverse outcomes in metastatic hormone-sensitive prostate cancer patients and may help us to identify these patients that may need more careful monitoring or more intensified treatment. We also want to highlight that there is a need for new therapeutic strategies in this subset of patients with more aggressive disease. Thank you very much.

Neeraj Agarwal: Thank you, Marta. This is great—very pertinent findings. So, to summarize, please correct me if I'm wrong, what you presented. First of all, congratulations for doing a multicenter study encompassing more than a hundred patients with metastatic hormone-sensitive prostate cancer, and showing that patients who had lower expression of tumor suppressor genes at baseline had a higher risk of developing castration-resistant prostate cancer. When they developed castration-resistant prostate cancer, especially in the context of PTEN loss and RB loss, they were more likely to have aggressive variant prostate cancer, which basically means higher incidence of visceral metastasis, less driven by androgen signaling, higher lytic bone lesions. These patients need different modalities of therapies than what we use conventionally for castration-resistant prostate cancer patients. Is that a correct synopsis of this study?

Marta Garcia de Herreros: Yeah, that's perfectly summarized. Yeah, these are our main conclusions.

Neeraj Agarwal: Thank you. So tell me, first of all, how did you analyze the samples? If I want to use these data in my practice—because I would like to have—I directly see three implications for my patients. Number one, counseling of patients. When I see them, when I see that they have low expression of tumor suppressor genes, I want to be able to tell them that they have poor prognosis. The second implication I see is I would like to see these patients more often in my clinic and more frequently do the CT scan and bone scans, because in these patients, I may not rely on PSA level to monitor.

Marta Garcia de Herreros: That's correct.

Neeraj Agarwal: And number three, when they develop castration-resistant prostate cancer, I think we would like to do a biopsy in these patients to make sure they do not have neuroendocrine or what you call aggressive variant prostate cancer—we also call it AR indifferent prostate cancer—that if they have this cancer, we need to know. So we need a biopsy for these patients. These are the three immediate implications for our patients.

So, please tell me, how would you recommend we go ahead and test these patients? What kind of technology did you use to identify that they had low tumor suppressor gene expression?

Marta Garcia de Herreros: Yes, so that's a very interesting question because we assessed this question in our previous study because we wondered what's the best technique to assess not the alterations in these genes—is it a mutation? Is it protein expression by immunochemistry? Is it RNA expression? All of them have certain limitations and strengths. Actually, we compared these three techniques, and we found that RNA expression was the technique that was better correlated with patient outcomes. That's why we followed the path into this RNA expression. So we assessed RNA expression with an nCounter panel. Actually, this panel comprises a set of nearly 100 genes, but here we just focus on these three genes. But we also have analyzed expression of androgen receptor and a lot of important genes in prostate cancer.

So I would say that maybe RNA expression is the way because maybe if you just focus on mutations, you would miss some alterations or epigenetic alterations or post-translational modifications. I think that it's important to focus on expression, and I think that maybe immunochemistry is not very reliable, right? Because you have more heterogeneity or it relies on the pathologist. So I would say explore RNA expression. Here, we used our cutoffs that have been previously validated, but I would say that if you put your own cutoff in this lower tertile or quartile expression, this would also work.

Neeraj Agarwal: So in your study, it was the lowest tertile of RNA expression of the three tumor suppressor genes, which were PTEN, RB1, and TP53. Is that correct?

Marta Garcia de Herreros: Yeah.

Neeraj Agarwal: And you say TSG-low when you have two of these three genes. So you can have RB1 and PTEN, or PTEN and TP53—so combinations of these three genes.

Marta Garcia de Herreros: So at least two have to be low.

Neeraj Agarwal: At least, yeah. Okay, that's very good to know. Especially, we have trials reporting out next year, especially in the context of PTEN loss. We have the CAPItello trial. We look forward to seeing the results in the near future, which is using capivasertib in patients with PTEN loss. So I think great news for our patients that we will have targeted therapies which may be approved soon to address some of these TSG losses.

Now, coming back to the discussion about how to measure these—in the US and many other countries, we have CLIA-certified comprehensive genomic profiling available for the tumor, and many oncologists and urologist colleagues obtain these tests from multiple companies such as Caris, Tempus, Foundation Medicine, and many other CLIA-certified labs. So if you are in the community—you are working in a big hospital, but say somebody is working in a community hospital and does not have access to great scientists like you have—how do they go ahead? Do they have expression data? How do they analyze? Do you have any recommendation for them? Or should we just be asking the CLIA-certified companies to provide the expression data more clearly in their reports?

Marta Garcia de Herreros: That would be the most interesting point, but I think it's not very feasible because these panels have been commercialized and analyzed mostly mutations. So, I would say if you have a panel that analyzes mutations, then go ahead and assess mutations, and just be very careful if your patient presents two of these three mutations. I would say reanalyze it if you have the opportunity to make a biopsy at castration-resistant prostate cancer, but don't miss the opportunity to analyze these mutations early in the hormone-sensitive prostate cancer because what our work wants to highlight is that it's important to know it early in the disease, not to wait until castration-resistant, because you have an impact on prognosis, or you can maybe intensify treatments with these new strategies that are coming, maybe with AKT inhibitors, maybe with platinum combinations, that these patients deserve better treatment strategies and monitoring.

Neeraj Agarwal: So the message here is we should do the comprehensive genomic profiling with RNA expression in the baseline samples before they're treated when samples are fresh when patients are newly diagnosed, so that we get the genomic data or molecular data from the cancer tissue without any effect of the treatment. Is that correct?

Marta Garcia de Herreros: Yeah.

Neeraj Agarwal: And is it sufficient to say if you cannot get hold of expression data, if you are in the community, or if you are in a setting where you do not have expression data available, at least if there is mutation of two of these three genes—PTEN loss, or there's a PTEN loss, RB1 loss, or P53 loss—if you see mutation in two of these three, that should prompt frequent monitoring, patient counseling should be tailored accordingly that it is associated with poor prognosis. And when they develop castration-resistant disease, we ought to be biopsying one of the castration-resistant sites or one of the metastatic tissue sites to make sure there is no evolution of what we call neuroendocrine differentiation or aggressive variant dedifferentiation. Is that correct?

Marta Garcia de Herreros: Yeah, that's correct. Yeah. Actually, in our study, we biopsied five patients, and all of them presented neuroendocrine differentiation.

Neeraj Agarwal: These are very pertinent results, I think, because not only does it have an effect on patient counseling, and obviously there's no question that all of these patients need to be getting intensified androgen deprivation therapy, and there is no role of androgen deprivation therapy alone. Unfortunately, many patients still get androgen deprivation therapy alone in the metastatic HSPC setting. So this is one more reason to not use single-agent ADT.

And the third most important lesson we learned from you today is that when these patients develop mCRPC, they ought to be undergoing biopsy of one of the metastatic sites. Before we conclude, how do we treat these patients when they develop metastatic castration-resistant prostate cancer with aggressive variant prostate cancer? How do you treat those patients?

Marta Garcia de Herreros: So, first of all, we look if we have any clinical trial available. We are lucky to have some of them. So maybe you have a clinical trial where you can give an AKT inhibitor for a PTEN loss or PTEN mutation. That would be interesting. Of course, a platinum combination such as cabazitaxel with carboplatin is a very good strategy. Or if you have a neuroendocrine differentiation, you can combine platinum combination with etoposide, or, of course, clinical trials. EZH2 inhibitors would also be a promising option for these patients.

Neeraj Agarwal: So depending upon what is available, clinical trial, and if a clinical trial is not available, these patients ought to be considered for treatment with platinum-based regimens, either in combination with taxane if there is still the AR differentiation variant present. If it's completely neuroendocrine differentiation or dedifferentiated prostate cancer, then, of course, platinum plus etoposide or small cell regimens. And if there is a component of adenocarcinoma present with neuroendocrine differentiation, then docetaxel or cabazitaxel plus carboplatin. Is that a good summary?

Marta Garcia de Herreros: Yeah, I agree.

Neeraj Agarwal: Yeah, that's fantastic. Well, thank you very much for joining us today, Dr. Marta García de Herreros from Hospital Clínic of Barcelona. Congratulations again for your presentation at the ESMO meeting, and such pertinent data which affect our day-to-day practice. Thank you very much.

Marta Garcia de Herreros: Thank you very much, and nice to discuss our topic with you.