Development of a Clinical Prognostic Stage Group System for Nonmetastatic Prostate Cancer

Get Permission

As reported in JAMA Oncology, Dess et al have developed a novel clinical prognostic stage group system for nonmetastatic prostate cancer that “meets criteria set forth by the American Joint Committee on Cancer [AJCC] Precision Medicine Core committee… [and outperforms] the existing [AJCC] system and commonly used risk-stratification systems.”

Study Details

The cohort study used data from 19,684 men with nonmetastatic prostate cancer from 55 sites in the United States, Canada, and Europe (the STAR-CAP cohort) who had cT1-4N0-1M0 prostate adenocarcinoma treated from January 1992 to December 2013, with follow-up through December 2017. The STAR-CAP cohort was randomly divided into training and validation data sets. A Surveillance, Epidemiology, and End Results (SEER) cohort was used as a second validation set.

Patients received curative-intent radical prostatectomy or radiotherapy with or without androgen-deprivation therapy. A points-based score staging system was developed using prostate cancer–specific mortality results from the International Staging Collaboration for Cancer of the Prostate.

Key Findings

Among the 19,684 patients included in STAR-CAP cohort, 12,421 were treated with radical prostatectomy and 7,263 with radiotherapy. Median follow-up was 71.8 months (interquartile range [IQR] = 34.3–124.3 months), with 4,078 (20.7%) being followed for ≥ 10 years.

In the training cohort, a predictive scoring system comprising age, T category, N category, Gleason grade, pretreatment serum prostate-specific antigen level, and percentage of positive core biopsy results among performed biopsies was identified. Each risk covariable was assigned a point value of 0 to 8, with total possible points ranging from 0 to 27. A total of nine score groups were defined, with scores of 0 (ie, new stage IA), 1–2 (IB), 3–4 (IC), 5–6 (IIA), 7–8 (IIB), 9–10 (IIC), 11–12 (IIIA), 13–16 (IIIB), and ≥ 17 (IIIC), corresponding to new stages IA–IIIC.

In the STAR-CAP validation set, predicted 10-year prostate cancer–specific mortality for the nine score groups ranged from 0.3% with score = 0 (stage IA) to 40.0% with score ≥17 (stage IIIC). The 10-year C index (0.796, 95% confidence interval [CI] = 0.760–0.828) was greater than that of the AJCC 8th edition (0.757, 95% CI = 0.719–0.792), with the score staging system outperforming the AJCC staging system across age, race, and treatment subgroups.

The score staging system outperformed current National Comprehensive Cancer Network® (NCCN®) three- and four-tier classification systems, with 10-year C index values of 0.729 for NCCN three-tier and 0.746 for four-tier systems vs 0.794 for the score system. Similarly, the scoring system outperformed Cancer of the Prostate Risk Assessment (CAPRA) risk grouping, with 10-year C index values of 0.782 vs 0.760, and a modified CAPRA system including T3b and T4 and clinical node–positive categories, with C index values of 0.796 vs 0.774.

In the SEER validation cohort (n = 125,575; median follow-up = 35 months, IQR = 16–53 months), the score staging system outperformed the AJCC system, with 5-year C index values of 0.838 vs 0.817.

The investigators concluded, “Using a large, diverse international cohort treated with standard curative treatment options, a proposed AJCC-compliant clinical prognostic stage group system for prostate cancer has been developed. This system may allow consistency of reporting and interpretation of results and clinical trial design.”

Daniel E. Spratt, MD, and Robert T. Dess, MD, both of the Department of Radiation Oncology, University of Michigan School of Medicine, Ann Arbor, are the corresponding authors for the JAMA Oncology article.  

Disclosure: The study was supported by the Prostate Cancer Foundation, Prostate Cancer National Institutes of Health Specialized Programs of Research Excellence, and others. For full disclosures of the study authors, visit

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.