Computed Tomography Determined Sarcopenia is a Significant Predictor of Poor Clinical Outcome in Gastric Cancer


 Background Sarcopenia is defined as a progressive and extensive loss of muscle mass and function. This study aimed to investigate the impact of sarcopenia and body composition on survival outcomes in patients with newly diagnosed gastric carcinoma (GC).Materials and methods Skeletal muscle area was measured at the level of the third lumbar vertebra (L3) using baseline CT images in patients with GC. Sarcopenia was defined as a L3 Skeletal muscle index (SMI) of < 41.6 cm2/m2 for men and < 32 cm2/m2 for women using van der Werf’s cutoffs. The disease free survival (DFS), overall survival (OS) and clinical characteristics of patients with and without sarcopenia were compared.Results A total of 226 patients were included. The median age of patients was 62 years (range 18–85) and 154 patients (68.1%) were men. Of the patients 37.6% were metastatic. Sarcopenia was present in 75 patients (33.2%) and was at significantly higher frequencies in men, metastatic disease, lower body mass index (BMI < 30), and higher ages (≥ 65 years) (respectively; all p < 0.05). The median DFS was 27 months (95% CI, 1.5 to 52.4) in patients with sarcopenia and non-appliceable in patients without sarcopenia in patients undergoing curative surgery for gastric cancer (p=0.041). The median OS was 10 months (95%CI, 7.730 to 12.270) in patients with sarcopenia and 29 months (95%CI, 21.307 to 36.693) in patients without sarcopenia (p<0.0001). In the multivariate Cox regression model, sarcopenia (with or without) (B=1.101; p<0.0001), stage (1-3 vs 4) (B=1.087; p=0.016) and surgical resection margin (R0 vs R1-2) (B=1.029; p<0.0001) were statistically significant predictors for OS.Conclusion Sarcopenia is an independent clinical predictor for poor prognosis in patients with GC. Early diagnosis and screening of sarcopenia may have a positive effect on survival outcomes.


Introduction
Gastric cancer is the fth most common cancer and third leading cause of cancer-related deaths in the world 1 . Due to the aggresiveness of disease and the lack of effective screening methods, the majority of patients are diagnosed at advanced stage, and the 5-year overall survival rate is generally less than 20% 2 . There is a number of clinical indicators which are related to prognosis in GC. Tumor stage is the most important predictor of the long-term prognosis 3 . In addition, studies have demonstrated well-known clinical and pathological features such as the presence of lymph node involvement, lymphovascular invasion, perineural invasion, D2 plus lymphadenectomy, RO resection margin and performance status of the patient [3][4][5] . Despite that, there is still a need for the detection of potentially modi able prognostic factors. Recently, sarcopenia has been identi ed as a negative prognostic factor which has an effect on morbidity and mortality in many cancer types 6-8 .
Sarcopenia was rstly described in the 1980s as an age-related decrease in lean body mass affecting nutritional status, mobility and independence 9 . The de nition of sarcopenia has been since developed. Sarcopenia is a syndrome de ned as a progressive and generalised skeletal muscle disorder that is related to increased probability of adverse outcomes including physical deformity, falls, fractures, treatment complications and mortality by The European Working Group on Sarcopenıa in Older People 2 (EWGSOP2) and European Society of Parenteral and Enteral Nutrition (ESPEN) 10 . Primary causes of sarcopenia are nutritional factors and age related inactivity. Cancer-related diseases are the secondary cause. The risk of sarcopenia is higher in untreated cancer patients due to wasting syndrome (cachexia), which is a characteristic feature of many malignancies, and its incidence is between 16 and 75%. [11][12] .
The gold standard method recommended by international working groups for the detection of sarcopenia is computed tomography (CT) [13][14] . The Skeletal muscle index (SMI) is a semiquantitative parameter which is calculated by dividing the muscle area given by CT with the square of the patient's height. Sexspeci c SMI cut-offs are used to diagnose sarcopenia 13,15 .
There are numerous studies reporting the impact of CT-determined sarcopenia in various cancers 6,15−19 . Presence of CT-determined sarcopenia in cancer patients increases perioperative comorbidity such as a higher risk of postoperative infections, longer hospitalization periods and high hospital costs 7,15,20−23 . Sarcopenia can also cause early termination of neoadjuvant chemotherapy because of dose-limiting effect [24][25] .
Presence of sarcopenia statistically reduces both DFS and OS in patients undergoing curative radical gastric surgery 7-8 and predicts survival in patients with advanced gastric cancer 26 .
Data about effects of sarcopenia on DFS and OS reported from Europe is more limited than eastern Asia and there is not any study on this subject in Turkish population. In this study,we aimed to evaluate the effect of CT-dedected sarcopenia on prognosis in gastric cancer patients.

Patient selection
This study is approved by the local Institutional Review Board and conducted in accordance with Helsinki declaration. The medical records of patients with gastric adenocarcinoma who were admitted to the medical oncology department of Izmir Katip Celebi University Faculty of Medicine Ataturk Training and Research Hospital between 2008 and 2020 were retrospectively reviewed. Among these patients, those with CT examination at the time of diagnosis were included. The clinical and histopathological features of all patients, including sex, age, body mass index (BMI), serum albumin level, Eastern cooperative oncology group performance status (ECOG PS), survival outcomes were recorded. Patients with severe chronic in ammatory or autoimmune disease, having steroid treatment and having severe comorbidity were excluded.
De nition of CT-Based Sarcopenia

Treatment endpoints
Disease free survival (DFS) was de ned as the time from randomization (pathological diagnosis) to the rst event of either recurrent disease or death.
Overall Survival (OS) was de ned as the time from randomization (pathological diagnosis) to death from any cause or the nal follow-up visit.

Statistical analyses
Statistical analyses were made by using the Statistical Package of Social Science (SPSS) version 16.0 software (Chicago, IL). The Kaplan-Meier method was performed to estimate survival outcomes and groups were compared by the log-rank test. Cox proportional hazards models were t to determine the association between sarcopenia with survival outcomes after adjustment for patient and disease characteristics. The 95% con dence interval (CI) was used to quantify the relationship between survival time and each independent factor and all statistical tests were carried out two-sided and a P value ≤0.05 was considered statistically signi cant.

Results
Four hundred eighty GC patients were found and 226 of them who met the inclusion criteria were included. Of the patients, 154 (68.1%) were men and 72 (31.9%) were women and median age was 62 years (range 17-85). 141 patients (62.4%) presented at non-metastatic and 85 (37.6%) patients presented at metastatic stages. The clinicopathological features of the patients are listed in Table 1.

Discussion
In this study, we analyzed the CT-determined sarcopenia retrospectively in gastric cancer patients. Male gender, advanced age, lower BMI, metastatic stage and lower ECOG PS were signi cantly related to sarcopenia. CT determined sarcopenia was independently associated with a worse DFS and OS.
Frailty is a condition of increased vulnerability to poor resolution of homoeostasis after a stressor event, resulting in an increased risk of developing adverse health outcomes 28 . Although the de nition of frailty is controversial and its assessment is subjective, in various studies, sarcopenia has been de ned as an accurate and objective measure of frailty 28-29 . It is important to identify and assess sarcopenia at the time of diagnosis. The detection of an optimal sarcopenia cut-off value on CT imaging depends on the type of disease and patient factors like sex, age, and race 30 31 . The prevalence of sarcopenia is higher in patients with locally advanced disease compared to patients with early-stage disease 32,34 . In our study, while the prevalence of sarcopenia was 24% in the group with non-metastatic disease, this rate was found to be 42.2% in the group with metastatic disease. This is because tumors that have progressed from local disease to metastatic stage may also produce more cancer-related cytokines, more oral intake disturbance due to the local effect of the tumor that may induce sarcopenia 33 . Hong-Bo Zou found the sarcopenia ratio as 20.0% in non-metastatic stage gastric cancer patients 32 .
We found a higher prevalance of sarcopenia in men (40.3%) than women (18.1%). Fukuda et al. reported that the prevalance of men in the sarcopenic group was higher than in the non-sarcopenic group in elderly gastric cancer patients undergoing gastrectomy (28.7% vs. 6.06%; p = 0.009) 35 . In another study including 276 newly diagnosed cancer patients, the prevalence of sarcopenia in men was found to be statistically higher than in women (25.4% vs 5,9%.; p<0.0001) 36 . Our results were also similar to the literature data.
Another common indicator of a patient's general condition is BMI. We found that sarcopenia prevalence was lower in obese group (15.6%) compared to non-obese group (36.1%). Zhuang et al. reported that sarcopenia was negatively correlated with lower BMI in patients undergoing radical gastrectomy for GC 7 .
Similarly, O azoglu et al. reported the prevalence of sarcopenia was lower in obese and overweight patients compared to normal and underweight ones with newly diagnosed cancer 11 .
In this study the prevalence of sarcopenia was found to be lower in the group with good ECOG status (28.9% in ECOG PS 0-1) than the worse one (59.4% in ECOG PS 2-3). Similarly, Prado et al. investigated the prevalence of sarcopenia in patients with solid tumors of the respiratory and gastrointestinal tract and they found lower ratios in patients with good functional scores (ECOG PS 0-1) compared to patients with poor functional scores (ECOG PS2-3) (47% vs 53%, respectively P=0.009) 30 . A similar result was observed in the study of O azoğlu et al. While the rate of sarcopenia was 14.3% in patients with ECOG PS 0, the rate of sarcopenia was signi cantly higher at 23% in patients with ECOG PS 1-2 (p=0.026) 11 .
O azoglu et al. hypothesized that the reason for the higher prevalence of sarcopenia in the group of patients with low performance scores might be because of less physical activity and less food intake, which eventually caused loss of muscle mass 11 .
In the present study, sarcopenia prevalence was higher in patients over 60 years (47.2%) than those under 60 years (15.8%). It has been shown in previous studies that the prevalence of sarcopenia increases with aging 11,20,32,34 . The possible reasons for the accelerated loss of muscle mass in elderly people, can be the higher prevalance of chronic diseases, aging, drug usage and more sedentary lifestyle.
Many reports have suggested that sarcopenia was associated with both short and long-term survival outcomes 7,8,15,20,37,39 , but in fewer studies no signi cant relation was found between sarcopenia and short 12,34,38 or long-term outcomes in GC 12,24  In conclusion, CT determined sarcopenia is a signi cant predictor of poor clinical outcome s in gastric cancer. Each gastric cancer patient should also be evaluated in terms of sarcopenia besides clinical, radiologic and histopatologic examination.

Declarations
Con ict of interest: The authors declare that they have no con ict of interest. Caption not included with this version.

Figure 2
Caption not included with this version.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.