Fractional Polynomials in a New Metastatic Renal Carcinoma Continuous Prognostic Index Involving Histology, Laboratory, and Clinical Predictors

Abstract

We analyzed the effect of histology and various clinical and laboratory predictors in a new continuous prognostic index for metastatic renal-cell carcinoma based on fractional polynomials. We evaluated 322 metastatic renal-cell carcinoma patients treated with subcutaneous recombinant cytokine-based home therapies in consecutive trials. We evaluated papillary histology, along with age, disease localizations, C-reactive protein, and neutrophil count in a new prognostic index, which was based on the multivariable fractional polynomial (MFP) algorithm. The MFP model allowed us to divide patients into three risk groups, using seven selected significant prognostic factors: histology, neutrophils, C-reactive protein, bone metastases, liver metastases, lymph node metastases, and age. Using the multivariable fractional polynomial model, median overall survival for high-, intermediate-, and low-risk patients was 10 months (n=80), 23 months (n=162), and 41 months (n=80) (scheme A; p <or= 0.01), or 7.9 months (n=16), 20 months (n=226), and 41 months (n=80) (scheme B; p <or= 0.001), respectively. Histology, clinical, and laboratory predictors contribute to a MFP algorithm-based prognostic index, allowing for a highly effective long-term risk discrimination in metastatic renal-cell carcinoma patients receiving recombinant immunotherapy. The flexibility in creating patient groups by using a continuous prognostic index allows to meet clinical needs in terms of improved prediction and optimized treatment selection.