MP21-19: Evaluating preoperative serum laboratory values and risk of for renal cell carcinoma recurrence in high-risk patients from three independent cohorts

Emily L Davidson, Jay D Raman, Viraj A Master, Datta Patil, Glenn O Allen, Leo D Dreyfuss, David F Jarrard, Kyle A Richards, Tudor Borza, Tracy M Downs, E Jason Abel

Introduction: Multiple studies suggest that routinely collected preoperative serum labs have prognostic value to identify which patients are at increased risk for RCC recurrence. However, few studies have compared prognostic ability between lab values or among independent populations. The purpose of this study was to compare prognostic value of serum labs as biomarkers for recurrence in three independent cohorts of high risk non-metastatic RCC patients.

Methods: Clinical and pathologic data from non-metastatic =pT3a RCC patients who were treated surgically at 3 independent centers from 2000-2016 were analyzed. Cox proportional hazards analysis was used to evaluate associations of recurrence with preoperative serum lab values including: C-reactive protein (CRP), platelet count (PC), mean platelet volume (MPV), neutrophil: lymphocyte ratio (NLR), albumin, and hemoglobin (Hb).

Results: Of 746 patients with non-metastatic =pT3a RCC treated surgically at 3 institutions, the median 5-year recurrence free survival was 67%. Among individual cohorts, no differences in baseline age, gender or smoking history was identified but race and BMI were different among 3 populations (p<0.001, 0.005). 

Multivariate analysis was used to evaluate associations of race and BMI with baseline lab values considering differences in pT stage, grade, diameter, sarcomatoid features, thrombus, and systemic symptoms. Race was independently associated with Hb (p=0.002), and PC (p=0.003). BMI was independently associated with hemoglobin (p=0.02), albumin (p=0.02), and PC (p=0.02), and CRP (p=0.03).

Per cohort analysis: NLR, CRP, MPV and albumin were prognostic in 1/3 cohorts. Hb was prognostic in 2/3 populations and PC was prognostic in all 3 cohorts. 

In combined cohort analysis: CRP, albumin, MPV, Hb, and PC were significant predictors of recurrence in univariate analysis. After multivariable analysis, independent predictors of recurrence included PC (HR 1.5, 95%CI 1.1-2.1, p=0.010), Hb (HR 1.5, 95%CI 1.0-2.2, p=0.034), tumor diameter (HR 1.1, 95%CI 1.1-1.2, p<0.001), and thrombus (HR 1.4 95%CI 1.2-1.5, p<0.001).

Conclusions: Preoperative platelet count and hemoglobin are prognostic factors for postoperative recurrence in high risk RCC patients. Baseline differences in labs values are associated race and BMI, which create variability among independent cohorts and may affect interpretation of prognostic value within a specific individual population. Source of

Funding: none