Predicting Oncologic Outcomes in Renal Cell Carcinoma After Surgery

Area of Science:

• Urology
• Oncology
• Nephrology

Background:

• Accurate prediction of oncologic outcomes is crucial for patient management, clinical trial development, and biomarker research.
• Renal cell carcinoma (RCC) encompasses several subtypes, each with distinct prognostic characteristics.

Purpose of the Study:

• To develop and validate prognostic models for progression-free survival (PFS) and cancer-specific survival (CSS).
• Models were specifically created for clear cell (ccRCC), papillary (papRCC), and chromophobe (chrRCC) subtypes of renal cell carcinoma.

Main Methods:

• Retrospective analysis of the Mayo Clinic Nephrectomy registry (1980-2010) including 3633 patients with nonmetastatic RCC.
• Multivariable Cox proportional hazards regression was employed to build parsimonious models using clinicopathologic features.
• Models were evaluated using c-indexes and converted into risk scores for predicting PFS and CSS, accounting for competing risks.

Main Results:

• Prognostic models were generated for each RCC histologic subtype, with c-indexes for PFS ranging from 0.77 to 0.83 and for CSS from 0.83 to 0.86.
• Risk scores were developed for each subtype and outcome, enabling prediction of survival rates.
• A multivariable model for chrRCC CSS was not assessed due to a limited number of events (22 deaths).

Conclusions:

• The study successfully generated specific prognostic models for ccRCC, papRCC, and chrRCC using established clinicopathologic features.
• These updated models are expected to enhance patient prognosis assessment, inform biomarker study design, and guide clinical trial enrollment.
• The identified features accurately predict progression and death from RCC post-surgery, aiding in patient counseling and treatment planning.