Associations of Adverse Pathology With 20 Year Risk of Distant Metastasis and Prostate Cancer Specific Mortality - Michael Brooks
March 16, 2022
Michael Brooks, MD, Urologic Oncologist, Houston Methodist Hospital, Houston, TX
Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, Massachusetts
Alicia Morgans: Hi, my name is Alicia Morgans and I'm a GU Medical Oncologist at Dana-Farber Cancer Institute. I'm so pleased to have here with me today, Dr. Michael Brooks, who's a Urologic Oncologist at Houston Methodist in Houston. Thank you so much for being here with me today.
Michael Brooks: Thank you for inviting me and it's my pleasure to discuss our recent manuscript that was accepted in Seminars Urologic Oncology. So this paper titled Validating the Association of Adverse Pathology with Distant Metastasis and Prostate Cancer Mortality 20-Years After Radical Prostatectomy, this was a group effort. First and foremost I'd like to thank Eric Klein and his team as well. So the background for this project, adverse pathology at radical prostatectomy has been used as a proxy for long term prostate cancer outcomes, but the goals were to evaluate the associations of adverse pathology with 20 year risk of distant metastasis and prostate cancer specific mortality.
The goal was to assess the association between adverse pathology at radical prostatectomy defined as high grade, Gleason group three, and or non organ confined disease, pT3 extra prostatic extension seminal vesicle involvement or lymph node metastasis with distant metastasis and prostate cancer death. So this was a stratified cohort sample of 428 patients from our institutional database of over 2,600 patients that were treated with radical prostatectomy between 1987 and 2004. AP, again, was defined as high stage pT3 or higher disease or primary Gleason grade pattern four or pattern five. The absolute risk of distant metastasis and prostate cancer specific mortality over 20 years after radical prostatectomy was estimated using Cox regression of cause specific hazards, treating death due to the other causes as a competing risk. We additionally did a subgroup analysis of all low risk and intermediate risk patients that could potentially have been included in an active surveillance protocol.
So this is the clinical information. So this cohort had a baseline PSA or median baseline PSA of 6.0, which is consistent with contemporary cohorts. The majority of the patients were AUA risk group low or intermediate, and only 10% were high risk. Again, this is reflective of a modern, radical prostatectomy cohort as well. On radical prostatectomy, 29.8% of patients had Gleason grade group three or greater. 42.3% had non organ confined disease. And 19.3% had both. Thus, 52.8% had adverse pathology on their radical prostatectomy specimen. So of the 428 patients included in the overall cohort, 105 experienced distant metastasis. And the risk of distant metastasis at 20 years was significantly higher for patients with adverse pathology at the time of radical prostatectomy, compared to those without, corresponding to a hazard ratio of 12.3. In the subgroup even, with low and intermediate risk patients, this again held true with a hazard ratio of 10.48 at 20 years. 53 patients in the overall cohort and 30 patients in the low and intermediate risk cohort experienced prostate cancer specific mortality.
Similar to distant metastasis, adverse pathology was significantly associated with this outcome as well, with hazard ratios of 10.03 and 8.60 in the overall and low and intermediate risk cohorts respectively. Interestingly, on bootstrap comparison, adverse pathology had a stronger association with distant metastasis than either grade or high stage. Primarily because patients without adverse pathology, both low stage and low grade disease, had low risk of distant metastasis at 1.5% at 20 years. And they fared better than patients with low grade and any stage disease or low stage and any grade disease. For prostate cancer specific mortality, the estimated strength of association with adverse pathology was higher than high grade or high stage, but the differences were not statistically significant. So in conclusion, a finding of adverse pathology is strongly associated with these long term outcomes of distant metastasis and prostate cancer specific mortality. After 20 years of follow up, adverse pathology had stronger association with distant metastasis than either of the individual components, high grade or high stage disease.
Alicia Morgans: Thank you. That was really, really helpful. And I think so important in clinical practice because on the majority, if not all of our patients, we should have the factors that we need or the components to really understand if a patient has adverse pathology. This is not typically something we have to send away for an extra test to get this information. So can you tell us from your perspective how this influences your clinical decision making in patients who have maybe Gleason three plus three disease as you're seeing them in clinic and thinking about things like surveillance. Does it come into play?
Michael Brooks: Yeah, so really this comes into play, like you suggested, on whether or not to make a treatment decision. So a patient that you're continuing to survey, if you have a test, either a genomic panel or an MRI that suggests they may have adverse pathology, that is not a patient that you should continue to survey. They have extraordinarily higher risk of distant metastasis and prostate cancer specific mortality compared to a patient without adverse pathology.
Alicia Morgans: Yeah. And that's super important, as you said, as you're making that decision. And I wonder, in order to do studies like this, you have to use cohorts who have been initially diagnosed many decades ago because you're looking at long term outcomes. Do you feel like the patients included in this cohort represent the patients that you see in clinic today? Or are there differences that we should think about as we're trying to apply this data?
Michael Brooks: Right. Certainly, that could be considered. One of the limitations is that perhaps this cohort does not reflect what a modern either active surveillance cohort looks like, or even a recently diagnosed man sitting in your clinic that's had potentially an MRI in the past. However, I would counter that for a few reasons. It is a fairly robust analysis. One is that the pathology specimen were all re-reviewed by an expert GU Onc pathologist. And the review occurred after the majority of the histopathologic changes to Gleason grading had occurred. So that should be reflective of modern standards. Secondly, perhaps some of the low and intermediate risk patients were harboring higher risk disease that just wasn't diagnosed with a systematic biopsy and could have been diagnosed with an MRI. That being said, the risk associated with adverse pathology at radical prostatectomy remains. So that's clinical staging tool, and I think in a way it sort of reinforces the reasons why these findings are important.
Alicia Morgans: Thanks. That's really helpful. So I just wonder, is there anything else that you want to add before we summarize at the end?
Michael Brooks: Yeah, so I think one of the strengths of this study obviously is the long term follow up. So I think this is particularly applicable to your younger, healthier men that have recently been diagnosed with prostate cancer, knowing that a 20 year risk for distant metastasis and prostate cancer specific mortality really does hinge on whether or not they have adverse pathology.
Alicia Morgans: Great. Well thank you for clarifying that for us. And as we think about this data and the work that you and your team have done, both in this area as well as in the genomic profile area for patients, what would your message be or your summary of recommendations to those folks who are trying to apply this data in their practice?
Michael Brooks: Well, we know for a fact from historical series that patients with very low risk disease are going to do great on an active surveillance protocol that does not include any of these adjunct tests. However, if you're considering including a man that may be at slightly higher risk because of a variety of clinical factors that would not have fit into a historical active surveillance protocol, then perhaps adjunct testing with either genomic testing or with an MRI may be useful in finding patients that would be at higher risk for adverse pathology.
Alicia Morgans: Great. Well, thank you. And I think, certainly, this is an area in evolution, an area where we are trying to find our way. And I sincerely appreciate you and your team giving us some extra information as we try to make those really important decisions. Particularly since this information gives us such a long follow up and really quite a clear understanding, I think, of what these features mean over time for our patients. So thank you so much for your work and, of course, for your expertise. We appreciate it.
Michael Brooks: Thank you for your time.