AUA 2017: Reflex Testing with Serum, Urine, and Tissue Biomarkers – Which, When and How Valuable?
Dr. Loeb started her talk by noting that the consequences of using PSA alone for considering which men to biopsy are that men may undergo unnecessary prostate biopsies (potentially leading to complications), as well as over-diagnosing and potentially over-treating men, noting that we are still over-treating >50% of prostate cancer patients. In the 2016 NCCN guidelines describing when to recommend an initial biopsy, Dr. Loeb notes that percent free PSA, 4Kscore or Phi are all recommended as possible options.
Free PSA has been FDA approved since 1998 and allows improved specificity over total PSA. The 4Kscore and Phi biomarkers take into account total PSA, free PSA, intact PSA, and hK2 (4Kscore), or total PSA, total PSA, and -2proPSA (PHI). As Dr. Loeb notes, the 4Kscore and Phi performed similarly in a head-to-head study (AUC 0.71), specifically if patients have 4Kscore >10% or a Phi >39, we can avoid 29% of biopsies, but miss 10% of high grade cancers. Furthermore, new results using ‘Phi density’ (Phi/prostate volume) at a cutoff of 0.43 indicate that we may be able to avoid 38% of unnecessary biopsies and only miss 2% of clinically significant prostate cancers (AUC 0.84).
Dr. Loeb then shifted gears to discuss what the role is of biomarkers in the setting of prior negative biopsy, specifically the role of multiparametric MRI (mpMRI). Dr. Loeb advocates for mpMRI in the setting of a prior negative biopsy prior to repeat biopsy, but noting that whether to perform MRI in this setting should also depend on other biomarkers to identify patients that may warrant repeat biopsy. These may include PCA3, Phi, 4Kscore, etc. One of the most important questions is whether we can exclude biopsy if the MRI is negative. Based a recent study demonstrating that the NPV of any prostate cancer is 82% and that of significant prostate cancer is 88%, Dr. Loeb feels like we are not quite at the point where we can replace biopsy, given that the accuracy is variable and may also be influenced by prostate cancer risk
Dr. Loeb then presented models that may appropriately incorporate biomarkers with MRI. For instance, if the 4Kscore for risk of high grade prostate cancer is <5%, no biopsy needs to be performed. If the 4Kscore is 5-32%, the patient could undergo an MRI and if a PIRADS lesion is ≥3 then a biopsy should be performed. Finally, if a 4Kscore is >32%, no MRI needs to be performed and the patient should proceed straight to prostate biopsy. Similar models have been studied with Phi are awaiting external validation.
In summary, Dr. Loeb states that there are a number of ‘reflex testing options available’ and that these tests outperform PSA and have the potential to reduce downstream harms. Importantly, markers remain crucial in the ever-expanding world of mpMRI.
Presenter: Stacey Loeb, New York University, New York, NY, USA
Written By: Zachary Klaassen, MD, Urologic Oncology Fellow, University of Toronto, Princess Margaret Cancer Centre
Twitter: @zklaassen_md
at the 2017 AUA Annual Meeting - May 12 - 16, 2017 – Boston, Massachusetts, USA