Using MRI to Safely Prevent Unnecessary Protocol-Advised Repeat Biopsy During Active Surveillance Journal Club - Zachary Klaassen

January 5, 2023

In this UroToday Journal Club, Zach Klaassen highlights a European Urology Oncology publication entitled Multivariable Approach Using Magnetic Resonance Imaging to Avoid a Protocol-Based Biopsy in Men on Active Surveillance for Prostate Cancer -- Data from the Multi-International Multicenter Prospective PRIAS Study. In general, the use of MRI prior to initial biopsy is now relatively widely mandated as it increases the diagnosis of clinically significant disease while decreasing overall biopsy rates. However, the role of MRI to omit prostate biopsy is debated in surveillance. The aim of the study being highlighting in this discussion is to determine predictors for grade group reclassification in patients undergoing an MRI-informed prostate biopsy during active surveillance and to evaluate whether a confirmatory biopsy can be omitted in patients diagnosed with upfront MRI.

Biographies:

Zachary Klaassen, MD, MSc, Urologic Oncologist, Assistant Professor Surgery/Urology at the Medical College of Georgia at Augusta University, Georgia Cancer Center


Read the Full Video Transcript

Chris Wallis: Hello, and thank you for joining us for this Uro Today Journal Club discussion. Today, we're talking about a recent publication entitled A Multivariable Approach Using Magnetic Resonance Imaging to Avoid a Protocol-based Prostate Biopsy in Men on Active Surveillance for Prostate Cancer-Data from the International Multicenter Prospective PRIAS Study. I'm Chris Wallis, an assistant professor in the division of urology at the University of Toronto. With me today is Zach Klaassen, an assistant professor in the division of urology at the Medical College of Georgia. You can see here the citation for this recent publication in European Urology Oncology led by Dr. Roobol and the PRIAS team.

Active surveillance, as most will know, is the standard treatment approach for patients with low-risk prostate cancer. Active surveillance aims to avoid unnecessary active treatment in patients who have overall indolent disease, and in appropriately selected patients, it has excellent longterm cancer-specific survival. In nearly all active surveillance protocols and guidelines, a confirmatory biopsy following initial diagnosis is mandated. However, prostate biopsy may be burdensome and morbid, and so there is interest in finding ways to safely avoid this.

MRI has an increasingly large role in men with prostate cancer, and you can see here citations for the precision and precise studies, as well as the ASIST trial looking at the role of MRI prior to initial biopsy and then during the surveillance protocol. In general, the use of MRI prior to initial biopsy is now relatively widely mandated as it increases the diagnosis of clinically significant disease while decreasing overall biopsy rates. However, the role of MRI to omit prostate biopsy is debated in surveillance.

In 2021, the EAU guidelines removed a recommendation to omit protocol-based biopsy in men who had a negative MRI, and part of this was driven by the fact that there's a variable negative predictive value of MRI. In patients diagnosed with upfront MRI, the EAU guidelines recommend omitting the confirmatory biopsy, however, though it must be noted that the strength of evidence for this recommendation is weak. In the context of the PRIAS study, they started to collect data on MRI results to reduce protocol-based biopsies beginning in 2013. In 2017, an analysis of this cohort demonstrated that no patient with a PIRADS score less than or equal to three, essentially a negative MRI, and a PSA density of less than 0.15 had great progression or reclassification during surveillance. However, the sample size at that time was too small to be definitive.

So this study comprises a multicenter prospective study designed to inform the practice of active surveillance. In the context of the PRIAS study, patients from around the world can be added by their clinicians to the cohort by online submission. For the present analysis, the inclusion criteria allow for the inclusion of patients added to the cohort between 2013 and 2020 must have a diagnosis of prostate cancer at least in grade group one, no evidence of extracapsular extension on digital record examination, a PSA of 10 or less, PSA density of less than 0.2. They must be fit for definitive therapy. And then among those over the age of 70 [inaudible 00:03:38] grade group two is allowed provided that it is a maximum of 10% per core and two or fewer cores are involved.

In terms of followup, if patients did not receive an initial MRI prior to a biopsy, an MRI and targeted biopsy were performed within three months of inclusion. An MRI plus targeted and systematic biopsy were then repeated at years 1, 4, 7, and 10 and then every five years thereafter. In this particular analysis, the author has included patients undergoing MRI informed prostate biopsy during surveillance regardless of the timing of this biopsy. It excluded men with lesion on an MRI who did not undergo targeted biopsy. And if patients underwent multiple targeted biopsies, each was included as a separate event for analysis.

Patients were divided into two subgroups, and we'll see this more clearly in a figure on the coming slide. Group A included patients who were diagnosed without an upfront MRI and group B included those who were receiving an upfront MRI. The primary outcome of the study was a reclassification, which was defined as a higher Gleason Grade after the targeted biopsy and the sensitivity analysis was performed redefining this as a Gleason Grade group three or higher. You can see here an overview of how patients were allocated into groups A and B. Again, with the moment of analysis being the time of their MRI-informed biopsy.

The authors used time-dependent Cox proportional hazards models with repeated measures to assess the time from the last MRI to the current MRI or in those who didn't receive an initial MRI, the time from diagnosis to MRI as their time indicator. Covariates included the digital rectal examination findings, age, PIRADS score, PSA density, and the percentage of positive cores on a systematic biopsy. Logistic regression models were used to assess predictors of reclassification with the same covariates used, as well as a indicator variable regarding whether MRI was performed prior to the first prostate biopsy.

I'm now going to hand it over to Zach to walk us through the results of this analysis of the PRIAS cohort.

Zach Klaassen: Thanks so much, Chris. So this is the table one, characteristics of the patients. As you can see here on the left is the characteristics, to the right of that is the combined cohort, and then subsequently, patients that have confirmatory biopsy after the first MRI and patients who had confirmatory biopsy after the second MRI. All told, there was 1,488 MRI-informed biopsies. The median age of these patients was 67. The PSA at MRI was 6.2 in each of the two groups with a PSA density of 0.13. Not surprisingly, the majority of these patients, nearly 90% were T1CDRE at MRI with the grade group before MRI being grade group one in 95% of patients. The time between diagnosis and MRI was just over one year. And the number of previous MRI scans was zero in 61% and one in 31% of the patients. Once we go to the outcome at first MRI, no lesion was present in 31.4% of patients, PIRADS three at 18.3% of patients, PIRADS four in 36% and PIRADS five lesions in 14.3%. The most common percentage score is positive on the last systemic biopsy, was 33% of patients having 5.1 to 10% of the cores positive. With regards to biopsy in the case of a lesion, targeted biopsy was performed alone in 33% of patients whereas targeted plus systematic biopsy was performed in 67% of the patients.

This looks at the outcomes of biopsy among the PIRAD score in all patients included in this time-dependent Cox model. And to sort of summarize this table, I've made several points at the bottom of this slide, in that 22% of patients were reclassified to Gleason grade group two or higher and 7% to Gleason grade group three or higher at MRI biopsy. Reclassification to grade group two or higher and three or higher included 6% and 1% of patients with a negative MRI, respectively, and 29% and 10% of patients with a PIRADS three or higher lesion, respectively. So we can see here that even in patients with a negative MRI, there is still a percentage of patients that will have reclassification to Gleason grade group two or three.

This is the time-dependent Cox model looking at upgrading to grade group two or higher, and we can see here that with regards to the significant variables in this model, age per 10 years was protective with a hazard ratio of 0.84 and a 95% confidence interval of .071 to 0.99. The strongest predictor of upgrading to grade group two or higher was MRI outcome. So compared to a negative MRI, a PIRADS two score had a hazard ratio of 2.46, a PIRADS four score has a ratio of 3.39, and a PIRADS score has a ratio of 4.95. Additional significant variables in the model included PSA density with a hazard ratio of 1.20 and percentage cores positive on last systemic biopsy with a hazard ratio of 1.16.

This figure looks at the Kaplan Meier for upgrading-free survival over time. We can see that the median upgrading-free survival was approximately just over seven years, as you can see from the figure here.

This figure looks at the percentage of upgrading after MRI biopsy stratified by MRI outcome and PSA density. And so you can see at the bottom is PSA density less than or greater than equal to 0.15, and this is stratified by no lesions, PIRADS three, four, and five. The green boxes are no upgrading, the orange boxes are upgrading the Gleason grade group two, and the red boxes are upgrading to Gleason grade group three or higher. So if we break this down by MRI results, so we see across the board, for all of these MRI breakdowns, we see an increase in upgrading to Gleason grade group two and greater than or equal to three for both no lesion, PIRADS three, four, and five, for patients that had a PSA density more than 0.15. So again, no lesions, we see more upgrading in the higher PSA density. For PIRADS three, again, these two columns here, as well as for four and five. So there's a gradient not only with the MRI results but also significantly favoring a higher PSA density for each of these assessments by MRI.

This looks at the multivariable logistic regression analysis for upgrading to Gleason grade group two. And so this is an odds ratio assessment. If we look at the DRE benign versus suspicious, suspicious having an odds ratio of significant of 1.81, again, we see age at biopsy not being significant in this assessment. We again see a very strong association with MRI outcome. So compared to no lesion, PIRADS three ratio 3.18, PIRADS four odds ratio 4.50, and PIRADS five odds ratio 10.14. The PSA density was also significant. Odds ratio of 1.40 and a 95% confidence interval of 1.17, so 1.69. Interestingly, at the bottom here, we can see that compared to upfront MRI, patients that did not have an upfront MRI, this was not statistically significant in this model.

So, several discussion points from this analysis of the PRIAS study. The available evidence and the resulting guideline recommendations to temporarily forgo a prostate biopsy during active surveillance with the use of MRI remains heavily debated. The results of this analysis from PRIAS showed that upgrading an MRI biopsy during active surveillance is fairly common, as we saw 22% to Gleason grade group two or higher, and 7% to Gleason grade group three or higher. MRI outcomes, as we saw on the previous slide, seemed to be the strongest predictor of grade group reclassification. There is likely significant institutional variation in the diagnostic performance of MRI during active surveillance, and the authors not several causes, which include different risk characteristics of the cohorts and the quality of MRI varying by institution.

A multivariable risk assessment should be used to determine whether a biopsy is indicated. So they have several scenarios and several recommendations. So first, it may be feasible to skip protocol-based biopsy if MRI is negative and the PSA density is less than 0.15. We should use shared decision making to discuss biopsy if there is a PIRADS three lesion on MRI or a PSA density that's greater than 0.15. And finally, we should not skip protocol-based biopsies in all patients with PIRADS greater than or equal to four lesions.

So, in conclusion, age, MRI outcome, PSA density, and percentage positive cores are significant predictors of reclassification at an MRI-informed biopsy. Patients with a negative MRI and a PSA density of less than 0.15 can safely omit a protocol-based prostate biopsy, whereas in other patients, a multivariable approach is advised. And finally, being diagnosed with upfront MRI does not appear to significantly affect reclassification risk; thus, a confirmatory MRI biopsy cannot completely be omitted as of yet.

We thank you very much for your attention. We how you enjoyed this Uro Today Journal Club discussion of the recent analysis of the PRIAS data set in European Urology Oncology.