Relative lymphocyte count is a prognostic parameter in cancer patients with catumaxomab immunotherapy

Background: Catumaxomab (anti-EpCAM x anti-CD3) treatment in peritoneal carcinomatosis (PC) of EpCAM-positive cancers was effective in phase I and II studies. Recently, it was approved in the EU for treatment of peritoneal carcinomatosis and malignant ascites. Aim of this hypothesis-generating study was to identify predictive or prognostic biomarkers with relevance for overall survival. Methods: 34 patients with PC in phase I/II studies with catumaxomab treatment were assessed for age, Karnofsky Index (KI), relative (RLC) and absolute lymphocyte count, relative and absolute granulocyte count, T-cell subsets, NK cells, and monocytes before catumaxomab therapy. Disease control (responder) was defined by stable disease, partial response or complete response (RECIST v1.0) >3 months or survival >6 months. Correlation analysis, Kaplan-Meier curves, ROC calculation, and multivariate regression were used for statistical analysis. Results: Mean RC values significantly differed between the non-responder (14.0%) and the responder group (23.9%; p = 0.001). RLC was correlated with overall survival (p = 0.03). RLC of >12% defined by ROC calculation was associated with prolonged survival (p = 0.035; hazard ratio of 2.775 for patients with RLC <12%). Patients with RLC >12% showed a mean survival of 15.6 versus 5.6 months in patients with RLC <= 12% (p = 0.001). Multivariate analysis found the individual RLC before therapy (p = 0.039) and the KI performance status (p = 0.002) to be independent prognostic parameters. Increasing KI by 1% resulted in a risk decrease of 10.1%. Increasing RLC by 1% resulted in a risk decrease of 4.6%. Age and the extent of PC did not significantly influence survival. Conclusions: RLC and KI were identified as potential prognostic parameters for superior disease control and overall survival after catumaxomab treatment. RLC may be used as a biomarker to indicate a suitable immune status for catumaxomab therapy. The predictive impact has to be confirmed in further studies. (C) 2014 Elsevier Ltd. All rights reserved.