Utilizing pre-therapy clinical schema and initial CT changes to predict progression-free survival in patients with metastatic renal cell carcinoma on VEGF-targeted therapy: a preliminary analysis.

OBJECTIVE: Because of varying treatment effectiveness with vascular endothelial growth factor (VEGF)-targeted therapy in patients with metastatic renal cell carcinoma (RCC), the association of prognostic pre-therapy clinical schema, initial post-therapy computed tomography (CT) findings, and combination thereof in predicting progression-free survival (PFS) was investigated. A predictive biomarker that combines clinical risk factors and CT imaging features associated with initial response to therapy would be useful in stratifying patients into risk groups to guide therapy, in designing and interpreting results of clinical trials, in planning risk-directed therapy, and in patient counseling. Early identification of poor responders using an imaging biomarker may reduce drug-related toxicity and cost and allow for a therapeutic intervention before disease burden significantly advances.

MATERIALS AND METHODS: For this institutional review board-approved HIPAA-compliant retrospective study, baseline data for 82 patients with metastatic RCC treated with sunitinib or sorafenib was obtained for risk stratification by Memorial Sloan Kettering Cancer Center (MSKCC) criteria and criteria by Heng et al. (J Clin Oncol 2009;27:5794-9), (described here as "VEGF prognostic factors criteria"). The initial post-therapy CT was evaluated by Response Assessment Criteria in Solid Tumors (RECIST), Choi criteria, and Morphology, Attenuation, Size, and Structure (MASS) criteria. Kaplan-Meier estimates of PFS (the reference standard) for each patient group and overall accuracy of each method and combined criteria were calculated.

RESULTS: The MSKCC model, VEGF prognostic factors criteria, RECIST, MASS criteria, MSKCC + MASS criteria, and VEGF prognostic factors + MASS criteria each demonstrated significant differences in PFS among patient groups (P < 0.005 for each, Log-rank test). Stratification of patient groups by Choi criteria was not statistically significant with respect to PFS (P = 0.101). MSKCC + MASS criteria yielded the highest overall accuracy for identifying PFS ≥ 1 year (77%) and for identifying PFS < 1 year (77%).

CONCLUSIONS: A combination of pre-therapy clinical risk factors and CT imaging response by MASS criteria more effectively predicted PFS in patients with metastatic RCC on VEGF-targeted therapy than any single method.