Hyperprogressive Disease (HPD) in Solid Tumours Receiving Immune Checkpoint Inhibitors in a Real-World Setting

Introduction: Hyperprogressive disease (HPD) is a state of accelerated tumor growth from cancer immunotherapy, associated with poor outcome. The reported incidence is 6% to 29% among studies using varying definitions of HPD, with no predictive biomarkers. Tumor infiltrating lymphocytes (TILs) are prognostic and predictive for immunotherapy benefit in various tumor types, but have only been tested for correlation with HPD in one study. Objectives: The objective of the study was to determine the prevalence of HPD in solid tumor patients treated with immune checkpoint inhibitor therapy in a real-world setting, and to assess clinicopathological features as potential biomarkers for HPD. Methods: We conducted a retrospective analysis of solid tumor patients treated with immune checkpoint inhibitors at a single institution. Imaging pre-immunotherapy and postimmunotherapy were assessed for HPD, and correlated against clinicopathological factors, including TILs and programmed death-ligand 1 (PD-L1) status through archival tumor assessment. HPD was defined per Matos et al as response evaluation criteria in solid tumors (RECIST) progressive disease, minimum increase in measurable lesions of 10 mm, plus increase of ≥40% in sum of target lesions compared with baseline and/or increase of ≥20% in sum of target lesions compared with baseline plus new lesions in at least 2 different organs. Results: HPD occurred in 11 of 87 patients (13%), and associated with inferior overall survival (median 5.5 months vs 18.3 months, P = .002). However, on multivariate analysis, only liver metastases (hazard ratio [HR] 4.66, 95% confidence interval [CI] 2.27-9.56, P < .001) and PD-L1 status (HR 0.53, 95% CI 0.30-0.95, P = .03) were significantly associated with survival. Presence of liver metastases correlated with occurence of HPD (P = .01). Age, sex, and monotherapy versus combination immunotherapy were not predictive for HPD. PD-L1 status and TILs were not associated with HPD. Conclusions: We found 13% HPD among solid tumor patients treated with immunotherapy, consistent with the range reported in prior series. Assessment for HPD is feasible outside of a clinical trials setting, using modified criteria that require comparison of 2 imaging studies. Liver metastases were associated with risk of HPD, while TILs and PD-L1 status were not predictive for HPD.