Improving PSA Doubling Time Calculation for High Risk Biochemical Recurrence - Alicia Morgans
March 31, 2025
Zachary Klaassen speaks with Alicia Morgans about examining the impact of PSA doubling time awareness on treatment decisions for high-risk biochemical recurrence. Dr. Morgans discusses findings from a retrospective chart review showing that approximately two-thirds of patients had no documented PSA doubling time, despite it being a crucial risk stratification factor. Among physicians who did document doubling times, 88% overestimated the values (suggesting less aggressive disease) compared to standardized calculator results. Patients with known doubling times received treatment significantly faster and at higher rates than those without. Dr. Morgans emphasizes that accurately calculating and documenting PSA doubling time is essential for proper risk assessment and informed treatment decisions, especially now that treatments like enzalutamide are available that can prolong metastasis-free survival. She suggests consistent use of standardized calculators rather than mental estimations to improve accuracy and ensure appropriate patient care.
Biographies:
Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, MA
Zachary Klaassen, MD, MSc, Urologic Oncologist, Assistant Professor Surgery/Urology at the Medical College of Georgia at Augusta University, Wellstar MCG, Georgia Cancer Center, Augusta, GA
Biographies:
Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, MA
Zachary Klaassen, MD, MSc, Urologic Oncologist, Assistant Professor Surgery/Urology at the Medical College of Georgia at Augusta University, Wellstar MCG, Georgia Cancer Center, Augusta, GA
Read the Full Video Transcript
Zachary Klaassen: Hello, my name is Zach Klaassen, urologic oncologist at the Georgia Cancer Center. I'm delighted to be joined on UroToday by Dr. Alicia Morgans, who is a medical oncologist at the Dana-Farber Cancer Institute. Alicia, always great to have you on UroToday. How are you?
Alicia Morgans: Thank you so much for having me here, Zach. It’s great.
Zachary Klaassen: Absolutely. We’re talking about some of your data presented at ASCO, specifically looking at PSA doubling time and high risk biochemical recurrence. And this is important because, are we doing it? Are we doing it, well? Your study really dove into these questions. So I’d love for you to walk us through study design and some of the important findings.
Alicia Morgans: Great. Thank you so much. So this study was really looking at the impact of physicians’ awareness of prostate-specific antigen doubling time, or PSADT, on treatment decision making in patients with high risk biochemical recurrence (BCR). So really, I think trying to understand if we’re calculating and documenting PSA doubling time, which we know is really one of the main factors that we use to risk stratify our patients with biochemical recurrence.
In terms of background, the PSA doubling time is really one of the strongest predictors of outcome in patients with biochemical recurrence. And it’s one of the criteria that we use to define high risk biochemical recurrence. This is important because we actually have a new treatment, enzalutamide with or without androgen deprivation therapy, that’s available for patients with high risk biochemical recurrence. And it prolongs metastasis-free survival, studied in the EMBARK trial.
Importantly, this is part of our guidelines through the NCCN and others. And so because of that, it’s just critical that we as physicians know and understand what a patient’s PSA doubling time is, and of course document that so we can use that information in our treatment decision making. But until this work, I don’t believe there has been any study that really looked at whether physicians were routinely calculating and documenting PSA doubling time in their charts. And so we really undertook this study to understand if that was happening and how that might influence treatment decision making in patients.
We were just interested in understanding how that workflow happened. So this was a retrospective, observational, multi-site, physician-abstracted medical chart review, which is a mouthful. Essentially, physicians abstracted from their charts from the Cardinal Health Oncology Provider Extended Network. And this is a US network, and they were looking between 2018 and 2022.
Importantly, we pulled out patients with high risk biochemical recurrence, and we had identified patients who had a known or documented PSA doubling time. And in those patients, physicians reported the PSA doubling time in the report forms that they used in this retrospective chart review. There were also patients in the population who did not have a PSA doubling time reported in those CRFs or those report forms, and so that was categorized as an unknown PSA doubling time group.
For all patients, we could actually retrospectively calculate a PSA doubling time using the Memorial Sloan-Kettering PSA doubling time calculator. And so for known and unknown patients, we could understand what their true or calculated PSA doubling time was based on that calculation. And so we try to understand and compare these two groups, patients with a known PSA doubling time and an unknown PSA doubling time, to understand their baseline characteristics and their time to first treatment within the first 60 days after their identification.
And for that known PSA doubling time group, we also, as I said, retrospectively calculated the PSA doubling time. In that case, we were doing it to understand if the doubling time that was recorded in the charts was actually correct, or was in line with what we calculated when using that standardized calculator.
So in terms of the baseline characteristics, you can see there are about 280 patients. In general, the patients with a known PSA doubling time were older than those with an unknown PSA doubling time. There were a higher proportion of patients with a known PSA doubling time who had more aggressive disease features or were older as well. Also, these patients often had a shorter time to recurrence of disease. And importantly, about 2/3 of patients had an unknown PSA doubling time, and then the rest had a known PSA doubling time. So many of the patients did not have a known PSA doubling time.
When we looked at the time to first treatment, the time to first treatment for the known group was shorter than the unknown group. And on the curve on the right, we can see the probability of receiving treatment, which was a lot higher in those patients who had a known PSA doubling time than an unknown PSA doubling time. And so I think this is also really important. The median time from the index to the end of follow up, you could see, was similar between the groups, and the proportion of patients that receive treatment within 60 days in the known group was about three times higher than the unknown group, and also a much shorter time to that first treatment. So in any event, knowing the PSA doubling time seemed to be important in terms of identifying those patients ready for treatment.
The other piece that I mentioned is among patients with a known PSA doubling time, we calculated to see on a standard calculator whether that was correct, more or less, or was kind of off. And what we found is that 88% of the physician-reported known PSA doubling times were actually overestimated. So they were longer than when we put them into the calculator. And so what this does is it tends to underestimate our understanding as physicians of the aggressiveness of the cancer, because a short PSA doubling time means a more aggressive cancer, and a long PSA doubling time usually means a less aggressive cancer.
On this figure on the right, you can see that a majority of these PSA doubling times are in the purple area, which shows that they were actually overestimating the PSA doubling time as compared to our standardized calculator. It may be that they were just kind of guessing, doing mental math, maybe not putting them into a standard calculator. I don’t know the reasons for this. Those are my guesses. But they clearly were predominantly overestimating. Really, patients were not having an underestimated PSA doubling time. They were typically overestimated by one to three months in 36% of patients, and then over three months in about 33% of patients.
So of course, limitations should be acknowledged. Unreported PSA doubling time values were assumed to be unknown by the physician based on that input, because somebody not only had to calculate it in the chart, but they could have if they wanted to calculate it for the ECRF; they could have calculated it and still put it in there, but they did not. So we assumed that they were unknown by the physician. But PSA doubling times may have been calculated but just not included in the ECRS. Discrepancies in the reported PSA doubling time values could exist due to the variations of the method used— as I said, sort of gestalt or guess versus actual calculation using one of those calculators. And this analysis was generally based on a small sample size of physicians, was more descriptive, and we didn’t adjust for baseline characteristics when we talk about the time to first treatment. But it is really a starting point to start the conversation around PSA doubling time, if we’re using it, if we know what it is, and where we might need to go in terms of optimizing care.
So our take-home message here is that most patients’ PSA doubling time was unknown by the treating physician at the time of the high risk biochemical recurrence, a non-metastatic hormone-sensitive prostate cancer diagnosis. And a greater proportion of patients received treatment within a shorter time frame when the PSA doubling time was known, so they could accurately have those conversations and share decisions with their doctors around whether it was the correct time to start treatment with the information that might inform that decision. And physicians that knew their PSA doubling time often overestimated it compared to our calculation on the back end, which was an underestimation of the tumor’s aggressiveness, potentially, and the risk of progression. Many patients with high risk BCR could remain unidentified in our clinical practices when they may benefit from treatment or at least benefit from the shared decisions around whether treatment is right for them or not. And so in order to ensure that we have those conversations, we share options with patients, I think it’s important for us to recognize that PSA doubling time is something we should probably be including in our considerations for these patients. Just like scans are used in patients with advanced disease, metastatic disease, this is something we need to use in the setting of biochemical recurrence. So thank you for listening, Zach. Happy to take questions.
Zachary Klaassen: Yeah, great presentation. I really like this study because it’s simple, but it’s effective in sort of showing that we’re really not identifying these patients. So if we look at this at a high level, we have 2/3 of patients who don’t have a known PSA doubling time. Among the ones that we are calculating doubling time, we’re not doing a very good job at it. So we are— if I’m correct, we’re undertreating probably a large proportion of these patients. Is that fair?
Alicia Morgans: I think that’s fair, with the caveat that we might be undertreating some people who want to be undertreated.
Zachary Klaassen: Sure.
Alicia Morgans: But I think it’s important that we at least have the information and then have those shared decisions and let patients decide what’s right for them with all the information that would be necessary to inform that decision.
Zachary Klaassen: Yeah, and I think too, I mean, we’re not looking to identify the PSA doubling times of 24 months or the elderly and co-morbid patients, but really trying to pinpoint those patients that are going to benefit, at least from a discussion about treatment, right?
Alicia Morgans: Absolutely. And I would just add that I think it is important that we have finally, in the biochemical recurrent population, treatments that can really prolong time to metastasis, which for many people is an endpoint that’s clinically relevant, important to them, and something that they would like to at least consider treatment for.
Zachary Klaassen: And have an understanding of what the options are for sure. You mentioned the memorial calculator. Is that one that you use? Is there other ways to help improve these PSA doubling time calculations?
Alicia Morgans: Yeah, so I can’t claim to know all of the PSA doubling time calculators. I have, since fellowship, always used the memorial calculator, which is probably why we used it in this study. But I think no matter what calculator you use, if it’s one that’s used consistently, you trust it, you’re comfortable with it, I think it’s helpful. I think some systems have an EMR where physicians can actually ask the teams that support the EMR to put calculators for structured data like PSAs and dates into programs that might be able to spit out an automatic PSA doubling time or something like it. So I would say maybe teams could even think about doing that in their EMRs. And that might be a strategy. Whatever it is though, it’s not something we can calculate in our heads, because it’s interesting that we often saw these physicians overestimating the PSA doubling time. That’s what happens when I try to do it in my head. And every time I plug it in, I’m like oh, shoot, it’s actually a lot shorter than I thought.
Zachary Klaassen: Yeah.
Alicia Morgans: Yeah.
Zachary Klaassen: No, I think you’re absolutely right. I think there’s a lot of ways to improve this. And one of them is having the calculator, but the main one is just bringing awareness to this. And this is what the discussion is about. Great study doing that. Anything we didn’t hit on? Any closing remarks?
Alicia Morgans: I would just say that this is an important piece of the decision making for this population. And as I mentioned, we use scans for metastatic disease. We should really be using something like PSA doubling time in this setting. Obviously, we want to try to help our patients have the information they need to make informed decisions. And this is really a key piece of the puzzle.
Zachary Klaassen: Yeah, absolutely. Well said, Alicia. Thanks so much as always, for joining us on UroToday.
Alicia Morgans: Thank you so much.
Zachary Klaassen: Hello, my name is Zach Klaassen, urologic oncologist at the Georgia Cancer Center. I'm delighted to be joined on UroToday by Dr. Alicia Morgans, who is a medical oncologist at the Dana-Farber Cancer Institute. Alicia, always great to have you on UroToday. How are you?
Alicia Morgans: Thank you so much for having me here, Zach. It’s great.
Zachary Klaassen: Absolutely. We’re talking about some of your data presented at ASCO, specifically looking at PSA doubling time and high risk biochemical recurrence. And this is important because, are we doing it? Are we doing it, well? Your study really dove into these questions. So I’d love for you to walk us through study design and some of the important findings.
Alicia Morgans: Great. Thank you so much. So this study was really looking at the impact of physicians’ awareness of prostate-specific antigen doubling time, or PSADT, on treatment decision making in patients with high risk biochemical recurrence (BCR). So really, I think trying to understand if we’re calculating and documenting PSA doubling time, which we know is really one of the main factors that we use to risk stratify our patients with biochemical recurrence.
In terms of background, the PSA doubling time is really one of the strongest predictors of outcome in patients with biochemical recurrence. And it’s one of the criteria that we use to define high risk biochemical recurrence. This is important because we actually have a new treatment, enzalutamide with or without androgen deprivation therapy, that’s available for patients with high risk biochemical recurrence. And it prolongs metastasis-free survival, studied in the EMBARK trial.
Importantly, this is part of our guidelines through the NCCN and others. And so because of that, it’s just critical that we as physicians know and understand what a patient’s PSA doubling time is, and of course document that so we can use that information in our treatment decision making. But until this work, I don’t believe there has been any study that really looked at whether physicians were routinely calculating and documenting PSA doubling time in their charts. And so we really undertook this study to understand if that was happening and how that might influence treatment decision making in patients.
We were just interested in understanding how that workflow happened. So this was a retrospective, observational, multi-site, physician-abstracted medical chart review, which is a mouthful. Essentially, physicians abstracted from their charts from the Cardinal Health Oncology Provider Extended Network. And this is a US network, and they were looking between 2018 and 2022.
Importantly, we pulled out patients with high risk biochemical recurrence, and we had identified patients who had a known or documented PSA doubling time. And in those patients, physicians reported the PSA doubling time in the report forms that they used in this retrospective chart review. There were also patients in the population who did not have a PSA doubling time reported in those CRFs or those report forms, and so that was categorized as an unknown PSA doubling time group.
For all patients, we could actually retrospectively calculate a PSA doubling time using the Memorial Sloan-Kettering PSA doubling time calculator. And so for known and unknown patients, we could understand what their true or calculated PSA doubling time was based on that calculation. And so we try to understand and compare these two groups, patients with a known PSA doubling time and an unknown PSA doubling time, to understand their baseline characteristics and their time to first treatment within the first 60 days after their identification.
And for that known PSA doubling time group, we also, as I said, retrospectively calculated the PSA doubling time. In that case, we were doing it to understand if the doubling time that was recorded in the charts was actually correct, or was in line with what we calculated when using that standardized calculator.
So in terms of the baseline characteristics, you can see there are about 280 patients. In general, the patients with a known PSA doubling time were older than those with an unknown PSA doubling time. There were a higher proportion of patients with a known PSA doubling time who had more aggressive disease features or were older as well. Also, these patients often had a shorter time to recurrence of disease. And importantly, about 2/3 of patients had an unknown PSA doubling time, and then the rest had a known PSA doubling time. So many of the patients did not have a known PSA doubling time.
When we looked at the time to first treatment, the time to first treatment for the known group was shorter than the unknown group. And on the curve on the right, we can see the probability of receiving treatment, which was a lot higher in those patients who had a known PSA doubling time than an unknown PSA doubling time. And so I think this is also really important. The median time from the index to the end of follow up, you could see, was similar between the groups, and the proportion of patients that receive treatment within 60 days in the known group was about three times higher than the unknown group, and also a much shorter time to that first treatment. So in any event, knowing the PSA doubling time seemed to be important in terms of identifying those patients ready for treatment.
The other piece that I mentioned is among patients with a known PSA doubling time, we calculated to see on a standard calculator whether that was correct, more or less, or was kind of off. And what we found is that 88% of the physician-reported known PSA doubling times were actually overestimated. So they were longer than when we put them into the calculator. And so what this does is it tends to underestimate our understanding as physicians of the aggressiveness of the cancer, because a short PSA doubling time means a more aggressive cancer, and a long PSA doubling time usually means a less aggressive cancer.
On this figure on the right, you can see that a majority of these PSA doubling times are in the purple area, which shows that they were actually overestimating the PSA doubling time as compared to our standardized calculator. It may be that they were just kind of guessing, doing mental math, maybe not putting them into a standard calculator. I don’t know the reasons for this. Those are my guesses. But they clearly were predominantly overestimating. Really, patients were not having an underestimated PSA doubling time. They were typically overestimated by one to three months in 36% of patients, and then over three months in about 33% of patients.
So of course, limitations should be acknowledged. Unreported PSA doubling time values were assumed to be unknown by the physician based on that input, because somebody not only had to calculate it in the chart, but they could have if they wanted to calculate it for the ECRF; they could have calculated it and still put it in there, but they did not. So we assumed that they were unknown by the physician. But PSA doubling times may have been calculated but just not included in the ECRS. Discrepancies in the reported PSA doubling time values could exist due to the variations of the method used— as I said, sort of gestalt or guess versus actual calculation using one of those calculators. And this analysis was generally based on a small sample size of physicians, was more descriptive, and we didn’t adjust for baseline characteristics when we talk about the time to first treatment. But it is really a starting point to start the conversation around PSA doubling time, if we’re using it, if we know what it is, and where we might need to go in terms of optimizing care.
So our take-home message here is that most patients’ PSA doubling time was unknown by the treating physician at the time of the high risk biochemical recurrence, a non-metastatic hormone-sensitive prostate cancer diagnosis. And a greater proportion of patients received treatment within a shorter time frame when the PSA doubling time was known, so they could accurately have those conversations and share decisions with their doctors around whether it was the correct time to start treatment with the information that might inform that decision. And physicians that knew their PSA doubling time often overestimated it compared to our calculation on the back end, which was an underestimation of the tumor’s aggressiveness, potentially, and the risk of progression. Many patients with high risk BCR could remain unidentified in our clinical practices when they may benefit from treatment or at least benefit from the shared decisions around whether treatment is right for them or not. And so in order to ensure that we have those conversations, we share options with patients, I think it’s important for us to recognize that PSA doubling time is something we should probably be including in our considerations for these patients. Just like scans are used in patients with advanced disease, metastatic disease, this is something we need to use in the setting of biochemical recurrence. So thank you for listening, Zach. Happy to take questions.
Zachary Klaassen: Yeah, great presentation. I really like this study because it’s simple, but it’s effective in sort of showing that we’re really not identifying these patients. So if we look at this at a high level, we have 2/3 of patients who don’t have a known PSA doubling time. Among the ones that we are calculating doubling time, we’re not doing a very good job at it. So we are— if I’m correct, we’re undertreating probably a large proportion of these patients. Is that fair?
Alicia Morgans: I think that’s fair, with the caveat that we might be undertreating some people who want to be undertreated.
Zachary Klaassen: Sure.
Alicia Morgans: But I think it’s important that we at least have the information and then have those shared decisions and let patients decide what’s right for them with all the information that would be necessary to inform that decision.
Zachary Klaassen: Yeah, and I think too, I mean, we’re not looking to identify the PSA doubling times of 24 months or the elderly and co-morbid patients, but really trying to pinpoint those patients that are going to benefit, at least from a discussion about treatment, right?
Alicia Morgans: Absolutely. And I would just add that I think it is important that we have finally, in the biochemical recurrent population, treatments that can really prolong time to metastasis, which for many people is an endpoint that’s clinically relevant, important to them, and something that they would like to at least consider treatment for.
Zachary Klaassen: And have an understanding of what the options are for sure. You mentioned the memorial calculator. Is that one that you use? Is there other ways to help improve these PSA doubling time calculations?
Alicia Morgans: Yeah, so I can’t claim to know all of the PSA doubling time calculators. I have, since fellowship, always used the memorial calculator, which is probably why we used it in this study. But I think no matter what calculator you use, if it’s one that’s used consistently, you trust it, you’re comfortable with it, I think it’s helpful. I think some systems have an EMR where physicians can actually ask the teams that support the EMR to put calculators for structured data like PSAs and dates into programs that might be able to spit out an automatic PSA doubling time or something like it. So I would say maybe teams could even think about doing that in their EMRs. And that might be a strategy. Whatever it is though, it’s not something we can calculate in our heads, because it’s interesting that we often saw these physicians overestimating the PSA doubling time. That’s what happens when I try to do it in my head. And every time I plug it in, I’m like oh, shoot, it’s actually a lot shorter than I thought.
Zachary Klaassen: Yeah.
Alicia Morgans: Yeah.
Zachary Klaassen: No, I think you’re absolutely right. I think there’s a lot of ways to improve this. And one of them is having the calculator, but the main one is just bringing awareness to this. And this is what the discussion is about. Great study doing that. Anything we didn’t hit on? Any closing remarks?
Alicia Morgans: I would just say that this is an important piece of the decision making for this population. And as I mentioned, we use scans for metastatic disease. We should really be using something like PSA doubling time in this setting. Obviously, we want to try to help our patients have the information they need to make informed decisions. And this is really a key piece of the puzzle.
Zachary Klaassen: Yeah, absolutely. Well said, Alicia. Thanks so much as always, for joining us on UroToday.
Alicia Morgans: Thank you so much.