Abstract 1486P
Background
Sarcopenia is a multifactorial syndrome defined by progressive and generalized loss of skeletal muscle mass, reduction of strength and physical performance which has been increasingly correlating with impaired cancer patient outcomes. High accuracy and reproducible results make CT one of the gold standard for body composition measurement. We aimed at assessing the impact of CT-based body composition parameters on the survival outcomes of resected GEA.
Methods
cT2-T4 and/or N-positive GEA pts undergoing curative-intent resection at our Institution between 2008-2018 were eligible. Presurgical clinicopathologic, biochemical, and antropometric data were retrospectively retrieved, while body composition parameters and their changes over time were derived by CT scan using GE Healthcare AW VolumeShare 7 software. Univariate and multivariate analyses for DFS and OS were performed.
Results
A total of 107 pts were included in the analysis. Median age was 66.1 years (range 20-85), 61 (57%) were males. 45 pts (42.1%) had stage II, while 62 pts (57.9%) had stage III. 98 (91%) pts had a non-cardia GEA. 85 (79%) pts received adjuvant treatment, consisting of fluoropyrimidine-based doublet in 85% of cases. Mean preoperative BMI was 23.9 kg/m2. CT scans were performed presurgically and from 4 to 15 months after resection. In the whole population, the 3-year DFS and OS were 48% and 49%, respectively. Out of 27 tested covariates, baseline IntraMuscular Adipose tissue Content (IMAC), together with ECOG PS and disease stage, was significantly associated with DFS (HR 1,97; p=0.03) and OS (HR 1,77; p=0.04) at the multivariate analysis. Specifically, pts with high baseline IMAC (≥ -0.42) had a shorter 3-year DFS (35% vs 62%, p<0,018) and 5-year OS (32% vs 55%, p=0,037).
Conclusions
We showed for the first time that presurgical IMAC, which reflects the quality of skeletal muscle, was an independent predictor of survival in radically resected GEA. This easy-to-calculate and largely available CT-derived parameter, may identify high-risk patients in need for a prompt and tailored nutritional intervention aimed at improving their outcome.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Massimiliano Salati.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.