To evaluate whether prediction of biochemical recurrence (BCR) after radical prostatectomy (RP) is enhanced by any of six parameters: prostate volume, total tumor volume (TV), high grade TV, ratio of high grade TV to TV, ratio of TV to prostate volume and/or ratio of high grade TV to prostate volume.
1261 patients who underwent RP over a 3 year period had tumor maps constructed with Gleason pattern denoted as either low (3) or high (4 or 5) and volumetric data generated using commercially available software. Univariate Cox regression models were used to assess whether each volume-related parameters was associated with BCR after RP. A multivariable Cox regression "base model" (age, PSA, Gleason score/grade group, pathologic stage and margin status) was compared with 6 additional models (base model + each volume-related parameter) to evaluate enhancement in predictive accuracy. Decision curve analysis was performed to determine the clinical utility of parameters that enhanced predictive accuracy.
On univariate analysis, each parameter was significantly associated with BCR except prostate volume. Predictive accuracy of multivariable base model was high (c-index=0.861). Adding volume-related parameters marginally enhanced discrimination. Decision curve analysis failed to show added benefit even for high grade TV/TV (parameter with highest discriminative improvement).
Tumor volume-related parameters are significantly associated with BCR, but do not add important discrimination to standard clinicopathologic variables for BCR prediction nor provide benefit across a range of clinically-relevant decision thresholds. Volume-related measurement is not warranted in routine pathologic evaluation and reporting.
The Journal of urology. 2018 Oct 06 [Epub ahead of print]
Yujiro Ito, Kazuma Udo, Emily A Vertosick, Daniel D Sjoberg, Andrew J Vickers, Hikmat A Al-Ahmadie, Ying-Bei Chen, Anuradha Gopalan, S Joseph Sirintrapun, Satish K Tickoo, Peter T Scardino, James A Eastham, Victor E Reuter, Samson W Fine
Department of Surgery (Urology Service)., Department of Epidemiology and Biostatistics., Department of Pathology, Memorial Sloan Kettering Cancer Center, NY, NY., Department of Pathology, Memorial Sloan Kettering Cancer Center, NY, NY. Electronic address: .