The Prognostic Value of PSA Levels Post-Radiation in Prostate Cancer - Praful Ravi
June 15, 2023
Alicia Morgans interviews Praful Ravi about his abstract presentation concerning PSA nadir within the ICECaP data set. ICECaP, a comprehensive data repository of individual patient data from more than 40 randomized trials conducted in localized prostate cancer, reveals that metastasis-free survival (MFS) is a proven surrogate for overall survival. In their study, Ravi and his team leveraged the ICECaP data to investigate the role of PSA nadir in patients receiving radiation, with or without ADT, for prostate cancer. The findings show that a lower PSA at six months post-radiation treatment is highly prognostic for MFS, overall survival, and prostate cancer-specific mortality. The results also raise questions about intensifying or de-intensifying therapy based on PSA levels, and they offer useful information for patient counseling and future trial designs. Ravi emphasizes, however, that PSA levels are strongly prognostic but not proven surrogate markers for MFS or overall survival.
Biographies:
Praful Ravi, MB, BChir, MRCP, Medical Oncologist, Dana Farber Cancer Institute, Boston, MA
Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, MA
Biographies:
Praful Ravi, MB, BChir, MRCP, Medical Oncologist, Dana Farber Cancer Institute, Boston, MA
Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, MA
Related Content:
ASCO 2023: Prognostic Impact of PSA Nadir (N) ≥0.1 ng/mL Within 6 Months After Completion of Radiotherapy for Localized Prostate Cancer: An Individual Patient-Data Analysis of Randomized Trials from the ICECaP Collaborative
ASCO 2023: Discussant: Hitting the Sweet Spot with Radiation and Androgen Deprivation
ASCO 2023: Prognostic Impact of PSA Nadir (N) ≥0.1 ng/mL Within 6 Months After Completion of Radiotherapy for Localized Prostate Cancer: An Individual Patient-Data Analysis of Randomized Trials from the ICECaP Collaborative
ASCO 2023: Discussant: Hitting the Sweet Spot with Radiation and Androgen Deprivation
Read the Full Video Transcript
Alicia Morgans: Hi. I'm so excited to be here with Dr. Praful Ravi, who is a GU medical oncologist at Dana-Farber Cancer Institute in Boston. Thank you so much for being here with me today.
Praful Ravi: Thank you, Alicia.
Alicia Morgans: Wonderful. Congratulations, on a fantastic oral abstract presentation at ASCO 2023. You gave us some insights on PSA nadir within the ICECaP data set, which is a huge data set. I'd love to hear about that data set, and what you guys investigated.
Praful Ravi: Yeah. ICECaP stands for Intermediate Clinical Endpoints in Cancer of the Prostate. It was set up about 10 years ago and it has collated individual patient data from more than 40 randomized trials conducted in localized prostate cancer, so patients getting prostatectomy and radiation, between 1987 and 2016. So it's a big data repository. And it has previously shown that MFS or metastasis-free survival is a proven surrogate for overall survival in the localized setting.
So what we did here, was to use and leverage this repository to look at this phenomenon or feature called PSA nadir, or sort of the landmark PSA in patients who are getting radiation with or without ADT for prostate cancer. By way of background, people have looked at this in the past, several single institution series, some pooled analyses of randomized trials as well have looked at the PSA level at a certain point after radiation or after neoadjuvant ADT and prior to radiation, and have found that a PSA less than 0.1 or less than 0.5 is highly prognostic. The largest such analysis has been about 2,400 patients and it was just radiation and short-term ADT. So in this study, we looked at the entire ICECaP one data repository to look at this particular question in a larger data set.
Alicia Morgans: Wonderful. What did you find?
Praful Ravi: We had about 16 trials included encompassing 10,400 patients. Median follow-up was 10 years. These patients had radiation alone, radiation and short-term ADT, so 3 to 6 months, radiation and long-term ADT, so 24 to 36 months. And we basically looked at the PSA at 6 months after completion of radiotherapy. We used a cut point of 0.1, A, because it was used in prior studies and shown to have value, and B, it's sort of the lower limit of assay detection or sensitivity in the era during which these studies were done. Most of these patients were treated before the year 2000 because of the nature of the data repository. Essentially, we looked at that landmark PSA at 6 months after radiation completion and examined its prognostic value in terms of MFS, OS, as well as prostate cancer-specific mortality.
Now, the first thing to note is that relatively few patients achieved that favorable PSA less than 0.1. It was only 2% of patients in the radiation arm, 16% in the radiation and short-term ADT arm, and 23% in the radiation and long-term ADT arm. But what we did show is that it was quite convincing that that achievement in that low PSA at six months post RT was highly prognostic for MFS at 5 years, 10 years, overall survival, and 10-year prostate cancer-specific mortality. For example, those in the radiation and long-term ADT group, their 5-year MFS was 87% if they had the PSA less than 0.1, compared to 81% in the greater than equal to 0.1. And this association really held true regardless whether it was just the radiation alone patients, radiation and short-term ADT, or radiation and long-term ADT, and it was robust with all three endpoints of interest.
Alicia Morgans: That's so interesting. A little bit concerning that the rates were so low to get to that level. And so how do you think that through, because they were also mixing populations that may have ongoing T suppression, And so how do you think that through?
Praful Ravi: It's a very good point, and we looked at this because we were a bit surprised too because the median of the lowest PSA achieved at any point in the radiation and ADT groups, either short or long, was 0.1, and the time to that median was about 6 months. But, when we actually looked at the dataset, a lot of the patients were actually at 0.1 exactly. So that's why the less than 0.1 is comparatively lower than what you might expect, because greater than or equal to encompasses a lot of the patients at that 0.1 threshold.
Alicia Morgans: So interesting, and that's not a limitation of the assay used to measure?
Praful Ravi: It could well be. That is certainly one limitation. But we had extensive statistical rigor and data cleaning to make sure that these results really held true when looking at different permutations of that.
Alicia Morgans: Interesting. In your mind, does this mean that we should be intensifying therapy more effectively for patients? Should we give them, maybe, additional AR-signaling inhibitors to get more patients down to that less than 0.1 or is it something about the innate biology that we really can only judge based with ADT? How do you really think about this?
Praful Ravi: I think that raises a very interesting question. I think the first thing to say is that this kind of confirms what we knew. When we treat patients in clinic, we know the patients who get to that low PSA, they're going to do well. And this sort of validates that and sort of really proves that in a big data set.
I think the points about using it for intensification or intensification are very interesting. One could certainly think about designing a trial where you intensify based on if you don't get to that PSA less than 0.1, as in the people who are PSA greater than or equal to 0.1 at 6 months after RT. Perhaps you intensify at that point. Or you deintensify, potentially, in patients getting long-term ADT if you get to that favorable less than 0.1. Those kinds of studies are trying to be done, it's hard to accrue to. But that's certainly one thing to do. I don't know whether we could really use this to say everyone should be getting up upfront intensification, because it's really based on that landmark value. It's really using that landmark value at 6 months to say, well can we do more or potentially do less, peel back, depending on where you are at that PSA at that point.
Alicia Morgans: Yeah, I think that's a very great point because we need that 6 month data actually to understand what the direction's going to be. In your practice, are you going to be using this information as you counsel patients whoare finishing their radiation or their radiation plus hormonal therapy?
Praful Ravi: Exactly. I think I sort of use it already in the sense that the lower the PSA goes the better, but this really shows that if you have a patient who's 6 months post radiation, either on short or on long-term ADT, we can use that as sort of a barometer of how things are going to go in terms of prognostically. And that's useful in routine day-to-day practice and counseling patients, as you say.
Alicia Morgans: Absolutely. So what would your main message be? What's the summary of this data from the fantastic presentation that you and the team gave?
Praful Ravi: I think the summary is that this PSA less than 0.1 or greater than equal to 0.1 is very strongly prognostic for all long-term endpoints in prostate cancer. So metastasis survival, prostate cancer mortality, overall survival. And it's true regardless of whether the patients are just getting radiation alone or radiation with short-term or long-term ADT.
So as we were just saying, it really aids and counseling patients treated in routine practice and has implications for trial designs in the future. Intensification or intensification approaches, but also you could envision when we are looking at, say, novel systemic therapies in combination with ADT, probably long-term ADT and radiation, perhaps the highest risk patients, we could have a multi-year, multi-stage design where we have several agents being tested and we could move forward or terminate arms based on the ability of that particular agent to get that low PSA number at 6 months post completion, and then take that forward into a larger trial, perhaps using the more valid, the longer term or accepted endpoints of MFS and OS. Because the one thing I want to stress is that we haven't proven that this is a surrogate marker for either MFS or OS. We've shown it's very strongly prognostic. So that's another message that I would want to relay.
Alicia Morgans: Absolutely. So important as we design these future trials. And as you said, they are hard to design, they're hard to accrue to and really run to completion, so it is wonderful that we have this ICECaP data to give us a sense at least what the prognostic implications might be of achieving this level and I really look forward to what you and the team continue to do within this data set and continue to teach us about what we can learn from prostate cancer, from the ICECaP information. So thank you so much.
Praful Ravi: Thank you.
Alicia Morgans: Hi. I'm so excited to be here with Dr. Praful Ravi, who is a GU medical oncologist at Dana-Farber Cancer Institute in Boston. Thank you so much for being here with me today.
Praful Ravi: Thank you, Alicia.
Alicia Morgans: Wonderful. Congratulations, on a fantastic oral abstract presentation at ASCO 2023. You gave us some insights on PSA nadir within the ICECaP data set, which is a huge data set. I'd love to hear about that data set, and what you guys investigated.
Praful Ravi: Yeah. ICECaP stands for Intermediate Clinical Endpoints in Cancer of the Prostate. It was set up about 10 years ago and it has collated individual patient data from more than 40 randomized trials conducted in localized prostate cancer, so patients getting prostatectomy and radiation, between 1987 and 2016. So it's a big data repository. And it has previously shown that MFS or metastasis-free survival is a proven surrogate for overall survival in the localized setting.
So what we did here, was to use and leverage this repository to look at this phenomenon or feature called PSA nadir, or sort of the landmark PSA in patients who are getting radiation with or without ADT for prostate cancer. By way of background, people have looked at this in the past, several single institution series, some pooled analyses of randomized trials as well have looked at the PSA level at a certain point after radiation or after neoadjuvant ADT and prior to radiation, and have found that a PSA less than 0.1 or less than 0.5 is highly prognostic. The largest such analysis has been about 2,400 patients and it was just radiation and short-term ADT. So in this study, we looked at the entire ICECaP one data repository to look at this particular question in a larger data set.
Alicia Morgans: Wonderful. What did you find?
Praful Ravi: We had about 16 trials included encompassing 10,400 patients. Median follow-up was 10 years. These patients had radiation alone, radiation and short-term ADT, so 3 to 6 months, radiation and long-term ADT, so 24 to 36 months. And we basically looked at the PSA at 6 months after completion of radiotherapy. We used a cut point of 0.1, A, because it was used in prior studies and shown to have value, and B, it's sort of the lower limit of assay detection or sensitivity in the era during which these studies were done. Most of these patients were treated before the year 2000 because of the nature of the data repository. Essentially, we looked at that landmark PSA at 6 months after radiation completion and examined its prognostic value in terms of MFS, OS, as well as prostate cancer-specific mortality.
Now, the first thing to note is that relatively few patients achieved that favorable PSA less than 0.1. It was only 2% of patients in the radiation arm, 16% in the radiation and short-term ADT arm, and 23% in the radiation and long-term ADT arm. But what we did show is that it was quite convincing that that achievement in that low PSA at six months post RT was highly prognostic for MFS at 5 years, 10 years, overall survival, and 10-year prostate cancer-specific mortality. For example, those in the radiation and long-term ADT group, their 5-year MFS was 87% if they had the PSA less than 0.1, compared to 81% in the greater than equal to 0.1. And this association really held true regardless whether it was just the radiation alone patients, radiation and short-term ADT, or radiation and long-term ADT, and it was robust with all three endpoints of interest.
Alicia Morgans: That's so interesting. A little bit concerning that the rates were so low to get to that level. And so how do you think that through, because they were also mixing populations that may have ongoing T suppression, And so how do you think that through?
Praful Ravi: It's a very good point, and we looked at this because we were a bit surprised too because the median of the lowest PSA achieved at any point in the radiation and ADT groups, either short or long, was 0.1, and the time to that median was about 6 months. But, when we actually looked at the dataset, a lot of the patients were actually at 0.1 exactly. So that's why the less than 0.1 is comparatively lower than what you might expect, because greater than or equal to encompasses a lot of the patients at that 0.1 threshold.
Alicia Morgans: So interesting, and that's not a limitation of the assay used to measure?
Praful Ravi: It could well be. That is certainly one limitation. But we had extensive statistical rigor and data cleaning to make sure that these results really held true when looking at different permutations of that.
Alicia Morgans: Interesting. In your mind, does this mean that we should be intensifying therapy more effectively for patients? Should we give them, maybe, additional AR-signaling inhibitors to get more patients down to that less than 0.1 or is it something about the innate biology that we really can only judge based with ADT? How do you really think about this?
Praful Ravi: I think that raises a very interesting question. I think the first thing to say is that this kind of confirms what we knew. When we treat patients in clinic, we know the patients who get to that low PSA, they're going to do well. And this sort of validates that and sort of really proves that in a big data set.
I think the points about using it for intensification or intensification are very interesting. One could certainly think about designing a trial where you intensify based on if you don't get to that PSA less than 0.1, as in the people who are PSA greater than or equal to 0.1 at 6 months after RT. Perhaps you intensify at that point. Or you deintensify, potentially, in patients getting long-term ADT if you get to that favorable less than 0.1. Those kinds of studies are trying to be done, it's hard to accrue to. But that's certainly one thing to do. I don't know whether we could really use this to say everyone should be getting up upfront intensification, because it's really based on that landmark value. It's really using that landmark value at 6 months to say, well can we do more or potentially do less, peel back, depending on where you are at that PSA at that point.
Alicia Morgans: Yeah, I think that's a very great point because we need that 6 month data actually to understand what the direction's going to be. In your practice, are you going to be using this information as you counsel patients whoare finishing their radiation or their radiation plus hormonal therapy?
Praful Ravi: Exactly. I think I sort of use it already in the sense that the lower the PSA goes the better, but this really shows that if you have a patient who's 6 months post radiation, either on short or on long-term ADT, we can use that as sort of a barometer of how things are going to go in terms of prognostically. And that's useful in routine day-to-day practice and counseling patients, as you say.
Alicia Morgans: Absolutely. So what would your main message be? What's the summary of this data from the fantastic presentation that you and the team gave?
Praful Ravi: I think the summary is that this PSA less than 0.1 or greater than equal to 0.1 is very strongly prognostic for all long-term endpoints in prostate cancer. So metastasis survival, prostate cancer mortality, overall survival. And it's true regardless of whether the patients are just getting radiation alone or radiation with short-term or long-term ADT.
So as we were just saying, it really aids and counseling patients treated in routine practice and has implications for trial designs in the future. Intensification or intensification approaches, but also you could envision when we are looking at, say, novel systemic therapies in combination with ADT, probably long-term ADT and radiation, perhaps the highest risk patients, we could have a multi-year, multi-stage design where we have several agents being tested and we could move forward or terminate arms based on the ability of that particular agent to get that low PSA number at 6 months post completion, and then take that forward into a larger trial, perhaps using the more valid, the longer term or accepted endpoints of MFS and OS. Because the one thing I want to stress is that we haven't proven that this is a surrogate marker for either MFS or OS. We've shown it's very strongly prognostic. So that's another message that I would want to relay.
Alicia Morgans: Absolutely. So important as we design these future trials. And as you said, they are hard to design, they're hard to accrue to and really run to completion, so it is wonderful that we have this ICECaP data to give us a sense at least what the prognostic implications might be of achieving this level and I really look forward to what you and the team continue to do within this data set and continue to teach us about what we can learn from prostate cancer, from the ICECaP information. So thank you so much.
Praful Ravi: Thank you.