Thrombocytopenia Is a Biomarker for Response in Patients Treated with Anti PD-1/Pdl-1 Therapy

Background: Immunotherapy directed against the PD-1/PDL-1 (programmed cell death protein [ligand] 1) pathway has shown activity across a range of malignancies. Immune-related adverse events have been shown to be associated with the efficacy of PD-1/PDL-1 inhibitors in patients with melanoma and non-small cell lung cancer. Thrombocytopenia has been reported in up to 33% of patients receiving anti PD-1/PDL-1 therapy. Our hypothesis is that immune-mediated thrombocytopenia may be a biomarker for response to anti PD-1/PDL-1 therapy.

Methods: We collected data on 250 patients aged 18 years and older who were treated with anti PD-1/PDL-1 therapy at the University of Oklahoma Health Sciences Center between January 2014 and January 2016. Initial platelet count at baseline as well as nadir platelet count during treatment were examined. Patient's clinicopathologic characteristics were analyzed using simple descriptive statistics [median (range) for continuous covariates and n (%) for categorical variables]. Kaplan Meier Analysis was performed to assess how thrombocytopenia was associated with overall survival (OS) and progression free survival (PFS). Patients were grouped according to their nadir platelet count: normal platelets (≥150k), grade 1 (75k-150k), and combined grades 2, 3, 4 (<75k). Cox proportional hazard regression models were used to assess the association of thrombocytopenia with OS and PFS adjusted for various covariates. The objective of the study is to assess whether the degree of thrombocytopenia correlates with the PFS and OS in patients receiving anti PD-1/PDL-1 therapy.

Results: A total of 250 patients were reviewed in this study. After excluding 27 patients with baseline platelet count <100k and 14 patients with missing platelet count data, 209 patients were included in the analysis. Fifty-five percent were females. Median age at diagnosis was 62 (23-91) years. The most common cancer diagnoses were lung cancer (21.1%), melanoma (17%), and ovarian cancer (9%). Nivolumab was the most commonly used agent (40.9%), followed by Pembrolizumab (38.4%). Median number of treatments received was 5. Median platelet count at baseline was 227 (102-535). Median nadir platelet count was 180 (4-552). Thrombocytopenia of any grade was observed in 70 (33%) patients .The number of patients with grade 1, 2, and 3/4 were 56 (26.8%), 1 (0.5%) and 13 (6.2%), respectively. Median follow-up time was 226 days. Univariate analysis showed that patients with grade I thrombocytopenia had higher PFS and OS when compared to patients who did not develop thrombocytopenia [HR 0.48 (95% CI, 0.29-0.81), P = 0.0053 and HR 0.49 (95% CI, 0.27-0.9), P = 0.0209, respectively]. When adjusted for age, cancer type, performance status, and line of treatment, grade 1 thrombocytopenia remained positively associated with survival outcome [HR 0.50 (95% CI, 0.29-0.87), P = 0.0146 for PFS and HR 0.40 (95% CI, 0.21-0.77), P = 0.0061 for OS]. Only 14 patients had grade 2-4 thrombocytopenia, and there was no statistically significant difference in survival compared to patients with no thrombocytopenia [adjusted HR 0.52 (95% CI, 0.2-1.38), P = 0.189 for OS].

Conclusion: Our study shows that in patients treated with anti PD-1/PDL-1 therapy, grade 1 thrombocytopenia appears to be positively correlated with survival outcomes. This survival advantage was not seen with higher grades of thrombocytopenia. PD-L1 was recently shown to be expressed on human platelets; thus, thrombocytopenia is likely immune-mediated. Thrombocytopenia may be a biomarker for efficacy of immune checkpoint therapy. The predictive significance of thrombocytopenia in patients receiving PD-1/PDL-1 therapy warrants further investigations.