37INIs “hyper-progression” a relevant clinical item for patient with solid tumours candidate to check-point inhibitor treatment?

Background: Immunotherapy with immune checkpoint inhibitors (ICIs) is modifying the standard of care for several solid tumours. Nevertheless, the results in terms of efficacy are still limited to a minority of patients even if the percentage of response differed among tumours type. The evaluation of response with ICI has been improved by the definition of new immune-related Response Criteria (irRC) that identified pseudo-progression followed by delayed tumour shrinkage in up to 10% of patients treated with anti CTLA-4 or anti PD1/PDL-1 agents. Afterward, hyper-progression (HP) after ICI administration has been recently reported but a mechanicistic explanation of the relationship with the ICI administration has not yet clearly identified. For this presentation, the obvious difficulty is to distinguish between resistance and consequent progression due to the tumour biology and a truly progression primed by ICI treatment. Methods: A total of 132 patients with solid tumors treated from 2014 to 2017 with antiPD1 (n = 15), anti PDL1 (n = 47), antiPD1 plus antiCTLA4 (n = 41) or plus antiGITR (n = 14), or IDO inhibitor (n = 6) or antiOX40 (n = 2), and bifunctional anti PDL1/antiTGFbeta monoclonal antibody (n = 7) in phase I/II trial at Istituto Nazionale Tumori were retrospectively evaluated for their tumor growth rate (TGR) before treatment and upon treatment. We defined the HP as: 1. progression at first restaging on ICI; 2. increase in tumor size >50%; 3. >2-fold increase in TGR, as reported by Champiat et coll. Results: Of 132 patients analyzed, 34 patients stopped the treatment before the first tumor evaluation (21 for clinical progression and 13 for drug-related adverse events), 42 patients exhibited disease progression at the first RECIST evaluation. Based on the criteria for the definition of HP, we excluded 14 pts without a previous imaging, 4 pts without target lesions and only 3 out of 24 pts (ovarian, gastric and squamous lung cancers) can be defined as HP (2.3%). Conclusions: In our experience, the occurrence of HP is limited to a small amount of patients treated with ICI in early phase clinical trials. In our analysis most of the patients have been treated with combo of ICIs, which could be a possible explanation for the lower incidence of HP as compared to what is reported. Nevertheless, the most important bias of underestimation for the evaluation of truly HP can be due to the very stringent criteria used for their definition that exclude patients with progression of non-target lesions and patients with early clinical progressions. As matter of fact, early clinical progression related to tumor biology and not to treatment is a major issue and the criteria reported by Champiat are not adequate to assess HP in patients receiving ICIs as first line of treatment, where we should probably concentrate our effort to investigate HP. Despite the fact that prospective trials are necessary to obtain more accurate data about the true incidence of HP and its etiology, we are evaluating a wider number of cases from disease treated in clinical trials and/or with standard of care to assess if incidence is related to histology. Legal entity responsible for the study: Fondazione IRCCS Istituto Nazionale Tumori Milano Funding: Has not received any funding Disclosure: F. de Braud: Advisor for BMS, Ignyta, MSD, Novartis, Pfizer, Amgen, Roche, Merck Serono, Servier. All other authors have declared no conflicts of interest.