Identification of new prognostic factors in phase I patients treated by immunotherapy

Abstract

Background: Evaluation of patient's life expectancy is crucial to select patient who may benefit from phase I studies. The Royal Marsden Hospital score (RMH) established a prognostic tool validated in prospective studies and based on 3 objectives variables: number of metastatic sites, LDH and serum albumin. Nevertheless, it remains unclear if those factors could be extended to immunotherapy phase I trials. Method(s): A retrospective analysis of risk factors of early death for patients enrolled into immune checkpoint inhibitor phase I trials in our institution was conducted. Demographic and biological characteristics (age, gender, number of metastatic site, LDH, albumin, neutrophil count, lymphocyte count, neutrophil-to-lymphocyte ratio, hemoglobin, platelet count) were analyzed with univariate and multivariable analysis for overall survival. Result(s): 155 patients (Male: 83; Median age was 59 (22-88)) treated with an experimental immunotherapy between March 2012 and January 2016 were analyzed. The most frequent tumor types were non-small cell lung cancer (16.7%), head and neck cancer (12.2%), urothelial cancer (10.3%), renal cancer (9%), breast cancer (7.7 %), cervical cancer (6.4%). In the multivariable model, Albumin (HR: 1.7; 95%CI 1.06-2.9) and LDH (HR: 1.9; 95%CI 1.2-3.2) remained statistical significant; nevertheless the number of metastases had no impact on the patients treated with immunotherapy (HR: 0.8; 95%CI 0.5-1.3). Interestingly, a higher neutrophil-to-lymphocyte ratio (NLR) was associated with a worse prognosis (HR = 1.8; 95%CI 1.1-3). We generated a new score with albumin (<35g/dl = +1), LDH (>upper normal limit = +1) and NLR (>6 =+ 1). We performed a Harrell c-index for evaluating the suitability of this new score (c index = 0.70; 95%CI 0.64-0.77) which was slightly higher than the c-index for RMH score in our series (0.65; 95% CI 0.58-0.72). Conclusion(s): Our study points that the addition of NLR to classical prognostic variables such as albumin and LDH, may predict better the overall survival of patients that are treated with immunotherapy. The number of metastasis has no prognostic value in our series. Although this score needs an external validation, this is a valuable tool to make a better selection of patients for phase 1 immunotherapy trials.