Biochemical Recurrence Risk in Prostate Cancer Patients with Perineural Invasion: A Systematic Review and Meta-Analysis

Abstract

Background: The association between PNI observed in radical prostatectomy (RP) and the risk of BCR in patients with prostate cancer remains unclear. We performed a systematic review and meta-analysis to quantify overall BCR risk in subgroups of individuals with PNI that underwent RP for localized prostate cancer. Methods: Eligible studies were retrieved from PubMed, Embase, Web of SCIENCE, Scopus and Google Scholar. Pooled multivariate hazard ratios (MHR), univariate hazard ratios (UHR) and univariate odds ratios (UOR) were calculated for eligible studies. Sub-analyses were performed for studies with mean follow-up time (MFU) of less than 2 years (y), 2-3 y and more than 3 y. Analyses were performed using random effect (RE) models to assess the pooled relative effects under consideration of heterogeneity among included studies. Results: A total of 28 studies were retained. The MHR analysis (n=22) indicated that PNI is significantly associated with BCR events (HR=1.53, 95%CI=1.33-1.78, p<0.0001). MHR sub-analysis revealed that the PNI~BCR association was not significant for studies with MFU less than 2 y (HR=1.39, 95%CI=0.92-2.08, p=0.115), but was for 2-3 y (HR=2.01, 95%CI=1.48-2.72, p<0.0001) and more than 3 y (HR=1.48, 95%CI=1.25-1.75, p<0.0001). The UHR analysis for 16 studies showed that PNI is significantly associated with BCR events (HR=1.94, 95%CI=1.46-2.57, p<0.0001). UHR sub-analysis revealed that the PNI~BCR association was not significant for studies with MFU 2-3 y (HR=1.35, 95%CI=0.48-3.85, p=0.572), but was for less than 2 y (HR=1.78, 95%CI=1.06-2.99, p=0.028) and more than 3 yr (HR=2.09, 95%CI=1.69-2.58, p<0.0001). The UOR analysis for 3 studies showed that PNI is significantly associated with BCR events (OR=1.59, 95%CI=1.19-2.11, p=0.002). UOR sub-analysis revealed that the PNI~BCR association was significant for studies with MFU 2-3 y (OR=7.59, 95%CI=2.66-21.70, p<0.0001) and more than 3 y (OR=3.70, 95%CI=2.27-6.04, p<0.0001). Conclusions: Although individual studies differed in their findings about the association between PNI and predicted BCR risk, our meta-analyses revealed a significant association between PNI and predicted BCR risk in studies with longer follow-up. This suggests that BCR, if it were to occur, is likely to be detectable 2-3 y post-surgery. Thus, there is an increased risk of BCR in pts with PNI; if expressed, this risk is to show anytime 2-3 y post-surgery. Student’s contributions: Stephanie Fletcher, the student, performed all statistical analyses, drafted the initial manuscript and was involved with all steps of this thesis. Dr. Abraham supervised all steps of this meta-analysis. Jennifer Martin assisted with part of the database searches and critically reviewed the manuscript. Dr. McBride, Dr. Abraham and Dr. Gomez critically reviewed the manuscript. Dr. Abraham and Stephanie Fletcher critically reviewed the statistical analyses and revised the manuscripts after critical review.