MP47-10: Differences in predictive ability of CRP and De Ritis Ratio for Outcomes in Renal Cell Carcinoma.

John Perry*, La Jolla, CA, Viraj Master, Atlanta, GA, Hajime Tanaka, Tokyo, Japan, Arman Walia, Ava Saidian, Rekha Narasimhan, Mimi Nguyen, Madison Chakoumakos, Margaret Meagher, Juan Javier-Desloges, La Jolla, CA, Dattatraya Patil, Atlanta, GA, Kazutaka Saito, Yasuhisa Fujii, Tokyo, Japan, Ithaar Derweesh, La Jolla, CA

Introduction: C-reactive protein (CRP) and De Ritis ratio (aspartate transaminase/alanine transaminase or AAR) have been described as predictors of outcomes in renal cell carcinoma (RCC). We sought to investigate predictive ability of CRP and AAR in different ethnic groups.

Methods: Retrospective analysis of the International Marker Consortium for Renal Cancer (INMARC) of patients who underwent surgery for RCC. Patients were grouped according to ethnicity and African American (AA), White, Asian, and Hispanic groups with preoperative CRP and AAR were selected for comparison. Primary outcome was overall survival (OS). Secondary outcome was cancer specific survival (CSS). Cox-regression multivariable analysis (MVA) was performed for predictors of all-cause (ACM) and cancer specific mortality (CSM). Kaplan-Meyer Analysis (KMA) was utilized to evaluate OS and CSS within different ethnic groups based on CRP and AAR value.

Results: A total of 4810 patients were analyzed (627 AA, 2344 White, 462 Hispanic, 1094 Asian; median follow up 33 months). MVA revealed elevated CRP (>5 mg/L) to be predictive of ACM in AA (HR 2.8, p<0.001) and White (HR 2.9, p<0.001) patients. Elevated AAR (>1.26) was predictive for ACM in Asian (HR 2.5, p=0.004) patients. CRP (p=0.390) and AAR (p=0.201) were not predictive of ACM for Hispanic patients. MVA showed similar results for CSM with elevated CRP noted to be a risk factor for worsened CSM in AA (HR 7.01, p<0.001) and White (HR 3.4, p<0.001) patients while elevated AAR was a risk factor for CSM in Asian (HR 2.4, p=0.041) patients. Neither CRP (p=0.569) nor AAR (p=0.837) were predictive of CSM for Hispanic patients.  KMA comparing non-elevated vs. elevated CRP values noted significantly greater 5-year OS for non-elevated CRP in AA (82.9% vs. 48.4%, p<0.001), White (81.4% vs. 48.2%, p<0.001), and Asian (94.3% vs. 69.4%, p<0.001), but not Hispanic (p=0.07), as well as significantly greater 5-year CSS in AA (94.6% vs. 62.6%, p<0.001), White (90.3% vs. 54.6%, p<0.001) and Asian (97.1% vs. 69.4%, p<0.001), but not Hispanic (p=0.925). KMA comparing non-elevated vs. elevated AAR revealed higher 5-year OS (96.4 vs. 88.7%, p<0.001) and 5-year CSS (97.5% vs. 94.4%, p=0.026) in Asian patients but demonstrated no significant differences in other groups.

Conclusions: CRP has broad utilities in prognostic abilities for non-Asian sub-cohorts, while AAR has predictive abilities in only Asian patients. Our findings suggest differing efficacy in ability of these markers to predict prognosis in various ethnic groups, highlighting need for further research in ethnic differences in RCC.

Source of Funding: Stephen K Weissman Kidney Cancer Research Fund