This current study showed that psoas muscle loss during treatment affects survival negatively in non-metastatic gastric cancer patients.
Despite the emerging new strategies, historically locally advanced gastric cancer treatment includes surgery +/- CRT or perioperative chemotherapy plus surgery. During all these treatments, oral intake can be deteriorated due to the disease itself, surgical morbidity and toxicities. In the cornerstone Intergroup 0116 trial, %33 Grade 3 Gastrointestinal toxicity was observed during adjuvant CRT , which brings a malnutrition risk and weight loss, especially in this patient group.
Malnutrition is one of the most important prognostic factors in cancer patients. Some studies showed an adverse relation between malnutrition and surviva [11,12]. This syndrome is not only related to poor oncological outcomes but also associated with deterioration of the immune system, delayed wound healing, higher infection rate and longer hospital stay [13–16]. All these negative factors may also diminish the patient's compliance with the treatment. In the recent GLIM consensus, reduced muscle loss was accepted as one of the strongest phenotypic criteria of malnutrition. Chronic inflammation and reduced food intake lead cancer patients to altered metabolism and body composition that manifests with a decrease in any marker of muscle mass like fat-free mass, muscle mass index or body cell mass .
There are several techniques to measure lean body mass and detect muscle loss. Magnetic resonance imaging (MRI) and CT are the best methods to quantify the skeletal muscle mass highly correlated with the cadaveric measurements [17,18]. Despite the high accuracy and reproducibility, these techniques are not easy to perform for each patient and have a high cost of instrumentation. Bioelectric impedance is also used for this purpose. This noninvasive technique measures the body composition indirectly using electric signals . This method is faster and easier than whole-body MRI and CT but this also needs extra effort and cost. The psoas muscle is one of the most important muscle groups for the perpendicular system and can be evaluated on CT images for staging and follow-up period and no need for an extra process that can bring an easier evaluation of muscle mass status instead of the whole body muscle mass evaluation. So calculating the PMA for detecting reduced muscle mass on CT images in cancer patients seems to be reasonable. In recent trials, measuring PMA on CT images was found as a non-invasive tool to predict sarcopenia [20,21].
Some studies showed that perioperative nutritional support for gastrointestinal malignancies reduces the number and severity of postoperative complications even if they do not have any sign of malnutrition [22,23]. These studies also underlie the important role of nutritional support on morbidity and mortality, especially in gastrointestinal cancers.
In this study, we hypothesised that psoas muscle loss as a sign of sarcopenia and malnutrition during the treatment is a negative prognostic factor on survival.
Cheng-Le Zhuang et al. retrospectively reviewed the gastric cancer patients who undergone curative surgery, and they found low skeletal muscle index, calculated with PMA and patient height, were related with postoperative severe complications as an independent risk factor and they also found sarcopenia as an independent risk for overall and disease-free survival especially in stage 2 and 3 patients, as well . They found 3-year overall survival 53.8% vs 73.6% (p < 0.001) and 3-year disease-free survival as 54.7% vs 73.5% (p < 0.001) in favour of patients without sarcopenia. A similar study was held in bladder cancer patients and sarcopenia was also found as related with a longer hospital stay, higher rate of perioperative complications and worse overall survival . Park et al. also found preoperative low psoas muscle area as a negative risk factor on overall survival in surgically treated oesophagal cancer patients . They found 3-year overall survival 64.9% in the high PMA group vs 37.1% in the low PMA group (p = 0.002). The results were similar in patients with upper urinary tract urothelial carcinoma among preoperative cases  and rectal cancer patients before neoadjuvant CRT . A systematic review of 13 studies and meta-analysis also denotes sarcopenia is significantly related with all-cause mortality in hepatocellular cancer patients .
These trials in different types of cancer patients showed that muscle loss is related with poorer outcomes, but all these trials were based on only a single measurement of the psoas muscle area before surgery or CRT. In this trial, we aimed to assess the change of the psoas muscle area before and after the whole cancer treatment modalities including surgery, chemotherapy and radiation therapy and to investigate the effects of the change on survival as well. All these treatments have serious side effects and may let the patients get deteriorated so we tried to evaluate the impact of side effects and morbidities on sarcopenia and we found a significant relationship between muscle loss and overall survival (42% vs 84%, p = 0.05)
Two studies from the United States and Korea examined the change of psoas muscle volume (PMV) and area and its effects on patients treated with chemotherapy and radical cystectomy for muscle-invasive bladder cancer and surgically treated oesophagal cancer patients, respectively. Zargar et al. from the US measured all psoas volume and calculated the change during the neoadjuvant chemotherapy undergoing radical cystectomy in bladder cancer. In this study, median %5 PMV and higher loss are associated with decreased but not statistically significant complete and partial pathological complete response rates and overall survival . Park et al. also focused on the prognostic effect of the psoas muscle area change in oesophagal cancer patients after one year who underwent surgery and they found that psoas muscle loss more than %10 as a significant risk factor on overall survival . In the low ∆ PMA group, they have found 3-year overall survival 58.2% and 18.9% in the high ∆ PMA group (p = 0.049). Three-year disease-free survival rates were found 47.3% and 18.8% in favour of the low ∆ group. We have found three year overall and progression-free survival rates as 84% vs 42% (p = 0.05) and %80 vs 38% (p = 0.07) in favour of the low ∆ PMA group. In our small cohort, we also examined the effects of age and stage on overall and disease-free survival. There was a trend for early-stage and younger ages but we could not find a statistical difference. The small number of cohorts should be the possible explanation of this issue.
Despite the uncertainties on measuring methods, accessibility and cut-off values of reduced muscle mass; there is strong evidence to use it as a single phenotypic criterion in the diagnosis of malnutrition . In the current study, we have tried to evaluate the dynamic changes, not a sectional evaluation because of the difference of our patient population then other chronic diseases and older adults. These gastric cancer patients generally have more acute/subacute reversible changes due to the treatments and treatment-related toxicities so dynamic measurements should be better than one sectional measurement to estimate survival or morbidity.