EAU 2018: Tumor-Associated Macrophages in Prostate Biopsy Determined by Automated Image Analysis - Independent Prognostic Marker for Biochemical Recurrence in Prostate Cancer Patients
Prostate biopsies from 38 PCa patients who underwent subsequent prostatectomy were selected for this study, 19 of them had biochemical recurrence (BCR). Tissue sections were immunohistochemically stained using the duplex stains CD68/CD163 for TAM and CK18/p63 to identify and characterize prostate glands. All sections were geometrically aligned per case (virtual multiplexing) to enable co-analysis of stains, and quantified within relevant regions-of-interest, using fully automated computational methods (Tissue Phenomics). In particular, the authors determined region-specific densities and average distances of TAM populations, as well as ratios of all measures. The optimal cut-off value for each parameter was calculated using receiver operating characteristic (ROC) analysis. The potential impact of each parameter on recurrence-free survival (RFS) was analysed using log-rank test and Cox regression models.
The median follow-up time was 70 months (IQR 27-80 months). The median number of positive prostate biopsy cores was four (IQR 2-6). There was a significantly higher density of TAM in the BCR group compared with the non-BCR group (p=0.013). The M1 (CD68+) and M2 (CD163+) macrophages were significantly enriched in tissue samples from the BCR group (p=0.016 and p=0.033, respectively). Patients with high TAM density (above the cut-off) in the biopsy had a significantly impaired outcome (median RFS 43 months vs. 98 months; p=0.007). In multivariable analysis (including PSA and Gleason score), high TAM density was an independent prognostic parameter (hazard ratio = 3.53, 95% confidence interval = 1.23-10.10, p=0.019).
These results indicate that a high density of tumour-associated macrophages in tumor tissue from prostate biopsy samples is an independent prognostic biomarker for early biochemical recurrence. The biomarker can be determined by automated image analysis. The authors present a promising approach for standardized and valid risk stratification of prostate cancer patients at an early time, and can support optimal personalization of treatment.
Presented by: Buchner A, Ludwig-Maximilians-University Munich, Dept. of Urology, Munich, Germany
Written by: Hanan Goldberg, MD, Urologic Oncology Fellow (SUO), University of Toronto, Princess Margaret Cancer Centre, twitter: @GoldbergHanan at the 2018 European Association of Urology Meeting EAU18, 16-20 March, 2018 Copenhagen, Denmark