953P Sarcopenia as predictive factor of response to nivolumab in recurrent-metastatic head and neck (RM-SCCHN) cancer patients

Abstract

Sarcopenia is an emerging problem in cancer patients (pts), significantly higher in SCCHN pts. Aim of this study is to analyze the impact of sarcopenia on efficacy of Nivolumab in RM-SCCHN cancer pts after platinum failure. We retrospectively reviewed clinical and radiological data of 24 RM-SCCHN pts undergoing Nivolumab at Modena University Hospital from 2016 to 2019. Sarcopenia was assessed using abdominal computed tomography scan (CT) before Nivolumab starting and after 8-12 weeks of therapy. Skeletal Muscle Mass (SMI cm2/m2) at L3 levels, Subcutaneous Fat Index (SFI cm2/m2), Visceral Fat Index (VFI cm2/m2) were considered for the analysis; variables were correlated with survival (short vs long survivors, SS vs LS, considering mOS <6month vs ≥ 6mo) and disease control (responders, R, considered as OR + SD vs non-responders, nR). In a subgroup of pts, we described the variation of SMI between the first and the second CT and correlate it with survival and response. 20 pts were evaluable for analysis: 15 (75%) were male; median age: 61 y-o (52-77), most of them had oropharynx (7, 35%) or oral cavity cancers (5, 25%) as primary sites; 13 pts (65%) had a normal BMI before starting Nivolumab. Median OS was 3 mo (0-11): 13 pts (65%) were SS, 7 (35%) LS; R vs nR were 13 (65%) and 7 (35%), respectively. At univariate analysis, no correlation between baseline SMI, SFI, VFI and survival was observed. On the contrary, we found a correlation between mean baseline values of SMI, SFI, VFI and response, in particular for SMI 55,4 in R vs 42,82 in nR, p=0,0008; SFI 61,79 in R vs 38,59 in nR, p=0,0009; for VFI 65,33 in R vs 44,77 in nR, p=0,0431. If we consider 15 pts evaluable for both CTs, 11 pts (73%) had a reduction of SMI at the second CT: mOS was 4 mo (2-9), 7 pts (64%) were SS and 7 pts (64%) were nR. Oppositely, 4 pts (27%) had an increase of SMI: mOS was 10 mo (0-11), 3 pts (75%) were LS and 2 (50%) were R. Sarcopenia is a widespread, but underestimated problem of SCCHN; it may be simply assessed using CT, generally employed for disease evaluation. Detection of body composition and identification of sarcopenia may be a suitable tool to predict response to immunotherapy of SCCHN.