Prostate Risk Index (PRIX) as a new method of risk classification for clinically localized prostate cancer

Abstract

Purpose: The aim of this study is (1) to develop a new method of risk classification for clinically localized prostate cancer; (2) to examine it in terms of compatibility with existing data such as nomograms; and (3) to compare it with existing risk-grouping methods.

Material and methods: The new grading system introduced here consists of three factors. The first is a prostate-specific antigen (PSA) of 4.1-10.0 ng/ml (score 0), 10.1-20.0 ng/ml (score 1), and >20.0 ng/ml (score 2). The second is a Gleason score (GS) of 6 (score 0), 7 (score 1), and 8-10 (score 2). The third is T classifications (UICC 2002) of T1c-T2a (score 0), T2b-T2c (score 1), and T3a (score 2). The sum of the three scores was named Prostate Risk Index (PRIX). Then, the compatibility of PRIX with the Partin Table, Kattan Nomogram, and Roach's formula was examined. At the same time, PRIX was compared with D'Amico, the National Comprehensive Cancer Network (NCCN), and Seattle classifications.

Results: PRIX 0 corresponded to 1-2% of pathologic lymph node involvement (pLN+) according to the Partin Table; PRIX 1 to 3-4%; PRIX 2 to 7-10%; PRIX 3 to 14-18%; PRIX 4 to 24-29%; PRIX 5 to 32-37%; and PRIX 6 to 42%. PRIX well separated the risks with relatively narrow ranges of probability, while D'Amico, NCCN, and Seattle classifications generally gave wide ranges especially for high-risk groups, both in the Partin Table and Kattan Nomogram. Roach's formula sometimes overestimated the risk compared to the Partin Table.

Conclusion: PRIX fully corresponded to the Partin Table in terms of pLN+, and corresponded to the other nomograms better than any existing risk-grouping method. PRIX may thus function as a prognostic factor or contribute to patient selection in clinically localized prostate cancer.