2. UroToday Home
  3. Conference Highlights
  4. AUA 2019
  5. AUA 2019 Prostate Cancer

AUA 2019

AUA 2019: External Validation of the PREDICT Prostate Tool for Prognostication in nmCRPC: A Study from Prostate Cancer Database Sweden

Chicago, IL (UroToday.com) The PREDICT Prostate is a novel prognostic model that provides individualized long-term prostate cancer-specific and overall survival estimates using baseline clinical and pathological data. The model, derived from UK data and validated in an East Asian cohort, estimates potential treatment benefit.1 In that model, the same authors used 10,000 men from the United Kingdom National Cancer Registration and Analysis Service (NCRAS) and 2500 men from Singapore as a validation cohort.

They identified a multivariable model estimating individualized 10- and 15-year survival outcomes using age, prostate-specific antigen (PSA), histological grade, biopsy core involvement, stage, and primary treatment, which were each independent prognostic factors for PCSM, and age and comorbidity, which were prognostic for NPCM. The model demonstrated good discrimination, with a C-index of 0.84 (95% CI: 0.82-0.86) and 0.84 (95% CI: 0.80-0.87) for 15-year PCSM in the UK and Singapore validation cohorts, respectively, comparing favourably to international risk-stratification criteria.

In this abstract, the authors validate their data using 69,206 men from Swedish Prostate Cancer database. Data on age, PSA, clinical T stage, grade group, biopsy involvement, primary treatment and comorbidity (variables used in the original study) were retrieved from the nation-wide population-based Prostate Cancer data Base Sweden (PCBase). Men with non-metastatic prostate cancer and PSA < 100 ng/ml diagnosed between 2000 and 2010 were included. 

This cohort had 13 years of median follow-up. Of the total population, 20,384 men underwent conservative management and 32,842 men received radical treatment. In terms of efficacy, the overall discrimination of PREDICT model was excellent: C-indices of 0.85 for PCSM and 0.79 for ACM were noted. Calibration of the model was excellent with 25,925 deaths predicted and 25,849 deaths observed in PCBaSe. 

Within the population of men undergoing conservative treatment and those undergoing radical treatment, c-indices for 15-year PCSM were 0.81 and 0.78 respectively. Among men on well-defined active surveillance, c-indices were further improved at 0.88 for PCSM and 0.75 for ACM. Calibration also remained good within treatment groups with differences between observed and predicted numbers of overall deaths 1.4%, 2.4% and 3.1% among men who received active surveillance, external beam radiotherapy, and prostatectomy respectively.

Therefore, in an even larger, well maintained national database, the PREDICT prostate model is highly accurate in predicting PCSM and NPCM. Unlike other available models it incorporates competing risk for death and provides individualized prognostic estimates of mortality, with and without upfront radical treatment, for up to 15 years from diagnosis. This is important information that can be used to help guide physicians and patients in their decision making process.

Presented by: Vincent Gnanapragasam, FRCSUrol, University of Cambridge

Co-authors: David Thurtle, Cambridge, United Kingdom, Ola Bratt, Gothenburg, Sweden, Pär Stattin, Uppsala, Sweden, Paul Pharoah

Written by: Thenappan Chandrasekar, MD (Clinical Instructor, Thomas Jefferson University) (twitter: @tchandra_uromd, @JEFFUrology) at American Urological Association's 2019 Annual Meeting (AUA 2019), May 3 – 6, 2019 in Chicago, Illinois

Reference:

  1. Thurtle DR, Greenberg DC, Lee LS, Huang HH, Pharoah PD, Gnanapragasam VJ. Individual prognosis at diagnosis in nonmetastatic prostate cancer: Development and external validation of the PREDICT Prostate multivariable model. PLoS Med. 2019 Mar 12;16(3):e1002758. doi: 10.1371/journal.pmed.1002758. eCollection 2019 Mar.