Prostate Cancer Diversity Revealed: AR and Neuroendocrine Signatures - Yu-Wei Chen
June 21, 2024
Yu-Wei Chen discusses his research focusing on androgen receptor (AR) and neuroendocrine signatures in prostate cancer using the Caris database. This study classifies prostate tumors into four molecular subgroups, revealing significant insights into their distinct behaviors and treatment responses. Dr. Chen's findings challenge the conventional histological classifications by demonstrating variations in survival rates and metastatic patterns linked to these molecular profiles. This research underscores the potential of molecular diagnostics to tailor treatment more precisely and improve outcomes for prostate cancer patients, emphasizing the need for ongoing investigation and clinical trials to integrate these findings into practice.
Biographies:
Yu-Wei Chen, MD, MS, Oncologist, Assistant Professor, Department of Medicine, UC San Diego School of Medicine, San Diego, CA
Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, MA
Biographies:
Yu-Wei Chen, MD, MS, Oncologist, Assistant Professor, Department of Medicine, UC San Diego School of Medicine, San Diego, CA
Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, MA
Read the Full Video Transcript
Alicia Morgans: Hi. I'm so excited to be here today with Dr. Yu-Wei Chen, who is joining me from University of California San Diego. Thank you so much for being here with me today, and congratulations on a phenomenal presentation at ASCO 2024.
Yu-Wei Chen: Thank you. Thank you, Dr. Morgans. It's my pleasure to be here.
Alicia Morgans: Wonderful. Well, you presented on AR and neuroendocrine signatures in patients with prostate cancer to help us understand the clinical significance of these signatures using a big Caris dataset. Can you tell us a little bit about that dataset, how you designed the trial, and what you looked for?
Yu-Wei Chen: Yeah, no. As you know, most prostate cancers, histology-wise, are prostate adenocarcinoma; however, the morphological characteristics pertaining to prostate adenocarcinoma and neuroendocrine prostate cancer are not exactly clear cut, and actually prostate adenocarcinoma can have mixed histologies with neuroendocrine differentiation agents.
In this study, we utilized the commercial database, the Caris database. Basically, we looked into the DNA profiling as well as the RNA profilings, and we used the two RNA-based signatures, the neuroendocrine prostate cancer signatures and also the AR signaling signatures, to further categorize the prostate tumors into four categories. So we have two signatures, so we have four subcategories. Most of the tumors are in the AR-positive, neuroendocrine-negative subgroup, followed by the AR-negative, neuroendocrine-negative subgroup. Both negative subgroups have approximately 30% of patients within this category. And then we have the neuroendocrine-positive, AR-negative subgroup, and we also have a subset with both signatures positive subgroups.
Alicia Morgans: Wow. Well, that's so interesting because at present we mostly break it into just things that look like adenocarcinomas that we believe are driven by AR, of course, by histologic classification, and then a neuroendocrine or small-cell subtype that we believe are driven by alternate pathways. So having four categories is really interesting. And of course, it's interesting that you've done this by molecular testing rather than histologic group alone. So tell me, did you find different behavior or clinical patterns associated with these different histologies?
Yu-Wei Chen: Yeah, no. In our dataset, 99% of the histology type is prostate adenocarcinoma. But we're still being able to find a subset of neuroendocrine-positive, but AR-negative in these cohorts. So what we found is that, actually, even though the histology type reads as prostate adenocarcinoma, generally speaking, the overall survival is worse among those who are neuroendocrine positive irrespective of their AR status. That is a key interesting finding in our dataset.
We also noticed that the distribution of the four subgroups is actually different when comparing with the primary prostate tumors versus the metastatic sites. For example, the AR-positive, neuroendocrine-negative subgroups, they have higher representation within the primary prostate tumors, the lymph node metastasis, and also the lung metastasis. And for neuroendocrine-positive, but AR-negative subgroups, they have higher representation with the bone metastasis and also lung metastasis. Last, for the double negative subgroup, they have higher expression in the CNS metastasis.
Alicia Morgans: Wow. So many interesting things to think about there. First, when you talked about all of these looking histologically like adenocarcinoma, then of course some having the signature that would suggest AR-negative, neuroendocrine-positive, I wonder if that has anything to do with when our pathologists say, "Actually, you know what, there's some chromogranin staining or synaptophysin staining in that adenocarcinoma." And that can be tricky and difficult to understand because they're not calling it neuroendocrine, but perhaps it's these molecular differences that allow for that staining. Of course, this is a hypothesis, this is not something you investigated, but I think it's fascinating what you said.
Yu-Wei Chen: Yeah, no, I think that's a great point. I think for future directions that if we can really use those signatures in a real-time setting and further categorize within the prostate adenocarcinoma histology categories, to further categorize those subgroups who actually behave more like a neuroendocrine tumor and to think about treatment justifications in those subgroups. Actually, I would like to highlight a clinical trial here, an Alliance trial led by Dr. McKay and also Dr. Misha Beltran, the PREDICT trial, which is a biomarker-driven trial that is going to use those two signatures to allocate treatment options in castration-resistant prostate cancer.
Alicia Morgans: That's fantastic. And really, I think the PREDICT trial is going to help us by looking at multiple options for treatment and using these kinds of molecular pathways or characterizations to help choose that treatment is really forward-thinking and will absolutely help our patients. So as you're thinking about these different categories and next steps beyond that clinical trial, what else can we imagine that we might be able to do with something like this in clinical practice at some point?
Yu-Wei Chen: Yeah, no. As we know, the current NCCN guideline recommendation for RB1, PTEN loss, and p53 mutations, if you have two out of three alterations, the current NCCN guideline recommends you can do doublet chemotherapy for those patients. So I think maybe if we apply those signatures in the prostate adenocarcinoma histologies, and also, if the patient is neuroendocrine-positive, we might consider a similar approach treatment intensification to chemotherapy double agents for those patients.
Alicia Morgans: Wonderful. To that point, is this a signature that we could order right now clinically from Caris, or is this something that is still investigational?
Yu-Wei Chen: No. It's still under investigation, yeah.
Alicia Morgans: This is really fantastic work. I wonder if you could give a summary to listeners as they're trying to think about this in the context of their own day-to-day.
Yu-Wei Chen: Yeah, no. I think even within the histology of prostate adenocarcinoma, prostate cancer is actually quite diverse, and nowadays, we hopefully can further develop with those two signatures into diagnostic tools to further categorize patients with neuroendocrine-positive signatures as having more aggressive biology and to consider further personalized treatments in those patients.
Alicia Morgans: Wonderful. It's always exciting for us to think about, I think, breaking what looks like a single entity into a more heterogeneous group that might respond differently to treatment. Certainly, this work is the first of many steps along that road, but with the trial you mentioned and this work and the work you'll do in the future, I think we're getting there. So thank you so much for your time and your expertise.
Yu-Wei Chen: Thank you, Dr. Morgans.
Alicia Morgans: Hi. I'm so excited to be here today with Dr. Yu-Wei Chen, who is joining me from University of California San Diego. Thank you so much for being here with me today, and congratulations on a phenomenal presentation at ASCO 2024.
Yu-Wei Chen: Thank you. Thank you, Dr. Morgans. It's my pleasure to be here.
Alicia Morgans: Wonderful. Well, you presented on AR and neuroendocrine signatures in patients with prostate cancer to help us understand the clinical significance of these signatures using a big Caris dataset. Can you tell us a little bit about that dataset, how you designed the trial, and what you looked for?
Yu-Wei Chen: Yeah, no. As you know, most prostate cancers, histology-wise, are prostate adenocarcinoma; however, the morphological characteristics pertaining to prostate adenocarcinoma and neuroendocrine prostate cancer are not exactly clear cut, and actually prostate adenocarcinoma can have mixed histologies with neuroendocrine differentiation agents.
In this study, we utilized the commercial database, the Caris database. Basically, we looked into the DNA profiling as well as the RNA profilings, and we used the two RNA-based signatures, the neuroendocrine prostate cancer signatures and also the AR signaling signatures, to further categorize the prostate tumors into four categories. So we have two signatures, so we have four subcategories. Most of the tumors are in the AR-positive, neuroendocrine-negative subgroup, followed by the AR-negative, neuroendocrine-negative subgroup. Both negative subgroups have approximately 30% of patients within this category. And then we have the neuroendocrine-positive, AR-negative subgroup, and we also have a subset with both signatures positive subgroups.
Alicia Morgans: Wow. Well, that's so interesting because at present we mostly break it into just things that look like adenocarcinomas that we believe are driven by AR, of course, by histologic classification, and then a neuroendocrine or small-cell subtype that we believe are driven by alternate pathways. So having four categories is really interesting. And of course, it's interesting that you've done this by molecular testing rather than histologic group alone. So tell me, did you find different behavior or clinical patterns associated with these different histologies?
Yu-Wei Chen: Yeah, no. In our dataset, 99% of the histology type is prostate adenocarcinoma. But we're still being able to find a subset of neuroendocrine-positive, but AR-negative in these cohorts. So what we found is that, actually, even though the histology type reads as prostate adenocarcinoma, generally speaking, the overall survival is worse among those who are neuroendocrine positive irrespective of their AR status. That is a key interesting finding in our dataset.
We also noticed that the distribution of the four subgroups is actually different when comparing with the primary prostate tumors versus the metastatic sites. For example, the AR-positive, neuroendocrine-negative subgroups, they have higher representation within the primary prostate tumors, the lymph node metastasis, and also the lung metastasis. And for neuroendocrine-positive, but AR-negative subgroups, they have higher representation with the bone metastasis and also lung metastasis. Last, for the double negative subgroup, they have higher expression in the CNS metastasis.
Alicia Morgans: Wow. So many interesting things to think about there. First, when you talked about all of these looking histologically like adenocarcinoma, then of course some having the signature that would suggest AR-negative, neuroendocrine-positive, I wonder if that has anything to do with when our pathologists say, "Actually, you know what, there's some chromogranin staining or synaptophysin staining in that adenocarcinoma." And that can be tricky and difficult to understand because they're not calling it neuroendocrine, but perhaps it's these molecular differences that allow for that staining. Of course, this is a hypothesis, this is not something you investigated, but I think it's fascinating what you said.
Yu-Wei Chen: Yeah, no, I think that's a great point. I think for future directions that if we can really use those signatures in a real-time setting and further categorize within the prostate adenocarcinoma histology categories, to further categorize those subgroups who actually behave more like a neuroendocrine tumor and to think about treatment justifications in those subgroups. Actually, I would like to highlight a clinical trial here, an Alliance trial led by Dr. McKay and also Dr. Misha Beltran, the PREDICT trial, which is a biomarker-driven trial that is going to use those two signatures to allocate treatment options in castration-resistant prostate cancer.
Alicia Morgans: That's fantastic. And really, I think the PREDICT trial is going to help us by looking at multiple options for treatment and using these kinds of molecular pathways or characterizations to help choose that treatment is really forward-thinking and will absolutely help our patients. So as you're thinking about these different categories and next steps beyond that clinical trial, what else can we imagine that we might be able to do with something like this in clinical practice at some point?
Yu-Wei Chen: Yeah, no. As we know, the current NCCN guideline recommendation for RB1, PTEN loss, and p53 mutations, if you have two out of three alterations, the current NCCN guideline recommends you can do doublet chemotherapy for those patients. So I think maybe if we apply those signatures in the prostate adenocarcinoma histologies, and also, if the patient is neuroendocrine-positive, we might consider a similar approach treatment intensification to chemotherapy double agents for those patients.
Alicia Morgans: Wonderful. To that point, is this a signature that we could order right now clinically from Caris, or is this something that is still investigational?
Yu-Wei Chen: No. It's still under investigation, yeah.
Alicia Morgans: This is really fantastic work. I wonder if you could give a summary to listeners as they're trying to think about this in the context of their own day-to-day.
Yu-Wei Chen: Yeah, no. I think even within the histology of prostate adenocarcinoma, prostate cancer is actually quite diverse, and nowadays, we hopefully can further develop with those two signatures into diagnostic tools to further categorize patients with neuroendocrine-positive signatures as having more aggressive biology and to consider further personalized treatments in those patients.
Alicia Morgans: Wonderful. It's always exciting for us to think about, I think, breaking what looks like a single entity into a more heterogeneous group that might respond differently to treatment. Certainly, this work is the first of many steps along that road, but with the trial you mentioned and this work and the work you'll do in the future, I think we're getting there. So thank you so much for your time and your expertise.
Yu-Wei Chen: Thank you, Dr. Morgans.