Read the Full Video Transcript

Zach Klaassen: Hi, my name is Zach Klaassen. I'm a urologic oncologist at the Georgia Cancer Center in Augusta, Georgia. I'm delighted to be joined today by Dr. Cédric Pobel, who is a medical oncologist at Gustave Roussy in Paris, France. Cédric, thanks very much for joining us today.

Cédric Pobel: Thank you for having me.

Zach Klaassen: So today we're going to discuss some ESMO data that you presented, looking at an ancillary study of PEACE-1, specifically looking at phenotypic and genomic characterizations of metastatic castrate-sensitive prostate cancer. So we're excited to talk about this. Why don't you share your slides? They’re excellent slides, and just walk our listeners through that data.

Cédric Pobel: Okay. Thank you for having me. Today, I will present to you the results we showed at ESMO this year, 2024, which show phenotypic and genomic characterization of de novo metastatic prostate cancer, an ancillary study of the PEACE-1 Phase III.

As a reminder, the PEACE-1 trial was a paradigm shift in first-line treatment. The design consisted of four arms of treatment: standard of care, with or without abiraterone, and with or without prostate radiotherapy. The results show benefits in overall survival with the addition of abiraterone to castration and docetaxel. And these are the very first results we presented for the ancillary study, for which the main hypothesis was that aggressive or neuroendocrine-like variants could be detected by diagnostics and are associated with prognosis.

For this purpose, we retrieved paraffin-embedded biopsies at diagnosis for almost 80% of patients who had consented to this ancillary study. Up to now, we have performed two main analyses: immunohistochemistry first with 10 markers, allowing us to identify five phenotypes: high-high, and high-low with androgen receptor expression in immunohistochemistry and negative neuroendocrine marker; neuroendocrine phenotype with negative androgen receptor and neuroendocrine marker-positive; amphicrine tumor, which was double positive for androgen receptor and neuroendocrine markers; and, on the opposite end, double-negative tumors. Secondly, we performed genomic analysis with next-generation sequencing using a restricted panel rather than a whole exome.

It's hard to read everything on these slides, but the important thing is that we had almost 400 patients with immunohistochemistry-contributed data and 190 patients with contributive sequencing. Among them, we had 101 patients with both immunohistochemistry and genomic data. And as you can see on this table, all the patient characteristics were similar between the cohorts.

Regarding phenotypic analysis: if you look at this slide, we have some outliers with double-negative tumors on these three patients, so it's hard to draw conclusions about them. But at the bottom of the curves, you have the neuroendocrine phenotype with negative androgen receptor and neuroendocrine marker-positive in IHC. Only five patients in this group, but they clearly had a shorter overall survival. The other phenotypes were AR-high or low with a similar overall survival, and amphicrine patients had a slightly shorter overall survival.

When we look at the same data but focus on each marker, AR-negative patients—10 of them—clearly had a worse prognosis, as well as patients with synaptophysin and chromogranin A-positive tumors. And surprisingly, we had almost one-fourth of all patients with at least one neuroendocrine marker among synaptophysin, CD56, or chromogranin A, and all these patients had the worst prognosis. It's important to note that apart from this prognostic value, we had no biomarker predicting the benefits of abiraterone for survival.

Regarding genomic analysis: all the alterations that you can see here were practically the same as previously published results in mCSPC, with a very low number of androgen receptor alterations as expected in this castration-sensitive setting.

When we look at patients with at least two alterations among TP53, PTEN, and RB1, these patients had overall survival divided by two. But we only had nine patients with this signature, so only a small number of them.

As a conclusion, I would say that combining AR and neuroendocrine marker expression assessments allows the identification of mCSPC patient subgroups with very different outcomes. Both AR-positive and neuroendocrine marker-positive expression were displayed in about 25% of mCSPC patients at baseline and predict poor prognosis. The alteration in at least two tumor suppressor genes, TP53, PTEN, and RB1, also predicts poor prognosis. And finally, a predictive biomarker for abiraterone benefits remains to be established in mCSPC.

Thank you for your attention.

Zach Klaassen: So, Cédric, thank you for that beautiful presentation. Some very interesting data. This is the beauty of having large-volume Phase III data, being able to do these ancillary studies. And I think when I was looking at your slides and listening to you present this, the thing that jumps out to me is that we have all the guidelines suggesting we should be doing genomic and phenotypic testing on all these patients, particularly for metastatic castrate-sensitive prostate cancer patients. So maybe just reinforce for our listeners the importance of getting these tests, so that we can really understand what their risks are based on the data you've shown.

Cédric Pobel: For the data I showed, there is actually no practice-changing information in the sense that we didn't find any predictive biomarker for abiraterone response. And in the PEACE-1 clinical trial design, we can't really assess this for the docetaxel response. But that also means that all patients benefit from abiraterone, regardless of the biomarkers we were assessing. However, it is still important to assess all these genomic alterations in these patients because, as I showed in my presentation, we had a similar number of patients or frequency of patients with BRCA alterations. These could be targeted by a PARP inhibitor at a later stage, as a set of resistance to castration. And probably, these patients should be treated in clinical trials at an earlier stage with PARP inhibitors as well. But it has to be assessed in clinical trials.

Also, we can retrieve alterations of PTEN and PIK3CA. We found this signature of worse prognosis related to neuroendocrine features and other markers. So, it could be useful to test these patients and find more active therapies for these neuroendocrine-like tumors in this early setting.

Zach Klaassen: That kind of leads to my next question. I think, looking at this data, particularly the neuroendocrine tumors, how does this data funnel into future trial design, maybe by neuroendocrine expression, and potentially finding targets for these patients with a poor prognosis?

Cédric Pobel: What we found is very surprising. There are a lot of patients with all these neuroendocrine features at diagnosis in mCSPC—almost 25% of them with positive immunohistochemistry markers. We also got some signals with this genomic signature in TP53, PTEN, and RB1. And we probably need more efficient treatments for them, especially now that we have all these new targets emerging in resistant castration prostate cancer, such as CEACAM5, DLL3, and TROP2, which could be targeted with ADCs, CAR-T cells, specific antibodies, or even BiTEs. It will be really interesting to see the efficacy of these treatments in these patients. As for platinum-based regimens, we don't have really strong efficacy data yet, and they are also very toxic treatments. The question remains about the utility of these treatments. Also, probably targeting BRCA in these patients in the early setting.

Zach Klaassen: Yeah, absolutely. Beautiful work. Congratulations. If you could just leave us with a couple of take-home messages for our listeners.

Cédric Pobel: I think the important message is that in de novo mCSPC, around 25% of patients have neuroendocrine features in immunohistochemistry, and these patients have a worse prognosis and should be treated more intensively. The alteration of at least two genes among TP53, PTEN, and RB1 is also a signature of worse prognosis in these patients. But no biomarker was found to predict abiraterone benefits in these patients.

Zach Klaassen: Wonderful. Cédric, thanks so much for your time and your expertise on UroToday today.

Cédric Pobel: Thank you, Zach. Thanks for having me.

Zach Klaassen: Thanks.