Combining the 4Kscore or ExoDx with multiparametric MRI appears to be the optimal strategy for detecting high-grade prostate cancer, a study from the University of California, San Francisco suggests. However, it is reasonable to consider biomarker testing only in patients who want to avoid biopsy, said first author Claire M. de la Calle, MD. In this interview, Dr. de la Calle discusses the study’s clinical implications, its limitations, and what additional research is needed.
What was the rationale for conducting this study?
PSA is a very poorly specific test. It has great sensitivity, but as we know, it lacks specificity. Because of that, many men undergo unnecessary biopsies, and a lot of indolent prostate cancers are diagnosed. Now that more and more men are enrolled into active surveillance protocols, it is even more important to screen for prostate cancer in a smarter way.
For that reason, biomarkers have been developed in an attempt to improve PSA specificity. Some of those biomarkers are 4Kscore, ExoDx, and SelectMDx, which we looked at in this study. Multiparametric MRI of the prostate has also been looked at as a potential tool for prostate cancer screening. However, the NCCN guidelines don't specifically indicate which biomarker is best to order and if MRI adds any value to them, and if so, in which order these tests should be ordered. Those were the main reasons for our study.
How did you conduct the study, and what were the key findings?
We retrospectively identified almost 900 patients who had been referred to UCSF from 2016 to 2019 for consideration of prostate biopsy. Every patient had at least one biomarker test; the vast majority had 4Kscore. Around half of our patients underwent prostate biopsy and about 650 had a prostate MRI. In our biopsy group, approximately 295 had a PSA of 2.5 to 10, which is close to the target PSA with which most of these biomarkers were validated.
We found that the 4Kscore and ExoDx missed the least amount of high-grade prostate cancer, defined as Gleason 3+4 or higher in our study. They missed around 3% to 6% of high-grade cases. But SelectMDx and multiparametric MRI were better at avoiding unnecessary biopsies, defined as negative biopsies or biopsies with only Gleason 3+3 cancer. However, if we combine a biomarker with MRI, 4Kscore and ExoDx missed almost no high-grade cancers—around zero to 1%—but more unnecessary biopsies were performed with the combination.
Our take-home message is that if the goal is to miss almost no high-grade disease, then combining the 4Kscore or ExoDx with multiparametric prostate MRI is likely the best strategy. However, if the patient prefers to avoid biopsy, it is certainly reasonable to consider getting only a biomarker, as only 3% to 6% of high-grade cancers would be missed with just biomarker testing.
What are the limitations of your study?
Not all of our patients underwent prostate biopsy; about half of our cohort had a biopsy. Also, most of our patients had the 4Kscore. We do have a good number of patients who received ExoDx and SelectMDx, but much less than 4Kscore. For that reason, what we'd like to do in the future is a prospective study where we would follow patients after getting at least two and hopefully all three of the biomarkers and a prostate biopsy so we can do a true head-to-head comparison of each biomarker and have a true negative group to further determine what the best strategy is for prostate screening.
Do the findings of your study essentially match up with NCCN guidelines on prostate cancer detection?
Interestingly, the NCCN guidelines mention that although there are multiple validation studies for these biomarkers, there is still insufficient data to recommend biomarker testing upfront with PSA. In addition, they mention that multiparametric MRI can be considered, but again there is no recommendation for upfront testing with MRI. And there is not enough data to determine what the role of MRI is, in addition to those biomarkers.
So as of right now, the NCCN guidelines only mention that both biomarker testing and multiparametric MRI can be considered if the patient and/or provider would like to have further information on the true risk of finding high-grade disease on biopsy. Hopefully, our future study will help make the guidelines a little clearer for providers.