g-NENs are a type of digestive system tumor with different clinical symptoms and biological characteristics[27]. Patients with different prognoses must be identified according to their clinical and pathological conditions to provide individualized treatment and improve the efficacy of g-NEN treatments. However, few studies have evaluated the prognostic factors for patients with g-NENs[8, 9]. Recently, the effects of preoperative body composition parameters (such as skeletal muscle mass) on postoperative short-term and long-term outcomes has attracted the attention of scholars in the East and the West. Sarcopenia is characterized by a progressive decrease in systemic muscle mass, muscle strength, or muscle physiological function associated with aging[28]. Sarcopenia has been shown to be closely related to the prognosis of patients with various malignant tumors[10-16]. However, the effect of sarcopenia on the prognosis of patients with g-NENs undergoing radical gastrectomy has not been reported. Therefore, this study combined the clinicopathological data from 138 patients treated at two institutions to explore the effects of sarcopenia on the short-term and long-term postoperative outcomes of patients with g-NENs.
Based on the definition of sarcopenia provided by the European Working Group on Sarcopenia (EWGSOP)[29] and the Asian Working Group for Sarcopenia (AWGS)[30], sarcopenia is characterized by a low skeletal muscle mass, low muscle strength and poor low physical performance. However, in the current study, low skeletal muscle mass was used as the main definition of sarcopenia. A meta-analysis exploring the relationship between sarcopenia and the risk of postoperative complications of gastrointestinal tumors included 29 studies related to sarcopenia, of which 26 used low skeletal muscle mass as the definition of sarcopenia[31]. In both studies from Eastern[10, 11] and Western[21, 28, 32] countries, researchers tend to use a low skeletal muscle mass as the definition for sarcopenia. Data on the patient’s muscle mass are obtained by analyzing the abdominal CT scan[10]. An abdominal CT scan is also a routine follow-up test performed in patients with g-NENs after radical gastrectomy[33]. The use of a low skeletal muscle mass as the definition for sarcopenia may help clinicians to make treatment decisions more conveniently and quickly.
Currently, the value of the cutoff point of sarcopenia remains controversial. The most commonly used definitions were provided by Prado et al.[21] and Martin et al.[32]. In the past, our center used X-tile software to analyze the 3-year OS rates of 924 patients with gastric adenocarcinoma after R0 resection and defined sarcopenia as a SMI<32.5 cm2/m2 for males and a SMI<28.6 cm2/m2 for females[10]. However, when previous definitions were applied, only the definitions reported by Martin et al. obtained a prevalence of sarcopenia similar to the values reported in previous studies (Supplementary Table 3). Therefore, we included the cutoff point defined by Martin et al. in the analysis. The Kaplan-Meier analysis and Cox regression analysis indicated that the cutoff points defined by Martin et al. were unable to serve as prognostic factors for patients with g-NENs in our study (Tables 3-4, Supplemental Table 2, and Supplementary Fig. 3). Therefore, this study used X-tile software to analyze the 3-year OS rates of 138 patients with g-NENs from the two institutions and defined a SMI<44.3 cm2/m2 for males and a SMI<32.4 cm2/m2 for females as sarcopenia, and the incidence of sarcopenia in our study was 42.8% (59/138). A significance difference in survival was not observed among the female group (Supplementary Fig. 2), perhaps because the proportion of female patients in this study was relatively small (33/138 cases, 23.9%). However, in the previous studies of sarcopenia, different values for the cutoff point of sarcopenia are usually used in male and female groups[14, 15, 32, 34], mainly because substantial differences in the strength and quality of skeletal muscle exist between males and females. In the present study, we compared the average SMI in male and female patients with g-NENs and observed a significant difference in the average value of the SMI between males and females (45.2 cm2/m2 in male, 37.5 cm2/m2 in female, p<0.05). Therefore, we used different diagnostic criteria for men and women in this study to better evaluate the effect of sarcopenia on the prognosis of patients with g-NENs.
The effect of sarcopenia on short-term postoperative outcomes in patients with malignant tumors remains controversial. Previous studies have confirmed that sarcopenia is associated with the postoperative short-term prognosis in patients with multiple malignant tumors[11, 13, 15, 35]. In a Chinese study, an analysis of 937 patients with gastric cancer after radical gastrectomy showed that sarcopenia was related to severe postoperative complications[11]. An American study identified an association between sarcopenia and the short-term outcomes in patients with pancreatic cancer after pancreatectomy[35]. However, some studies have reported the opposite results[34, 36]. As shown in the study by Tegels[34], the incidence of sarcopenia is higher in patients with gastric cancer, but it is not associated with a poor postoperative prognosis. According to Ouchi[36], sarcopenia does not increase the incidence of total and severe postoperative complications in patients with colorectal cancer[36]. In the present study, significant differences in the incidences of total postoperative complications, surgical complications and systemic complications were not observed between the patients with g-NENs presenting with and without sarcopenia. After stratification according to the physical location of the complications, significant correlations were not observed between sarcopenia and specific types of complications in patients with g-NENs.
In recent years, studies have confirmed that sarcopenia is closely related to the long-term prognoses of patients with multiple malignant tumors[10, 12, 14, 16]. Studies by Voron have identified sarcopenia as an independent prognostic factor for long-term outcomes in patients with hepatocellular carcinoma after hepatectomy[12]. As shown in the study by Tan, sarcopenia is associated with a poor prognosis for patients with pancreatic cancer[16]. Similar to previous studies, preoperative sarcopenia was an independent risk factor for the long-term prognosis of patients with g-NENs in the present study. We also examined the interactions between sarcopenia and the gastrectomy status and tumor aggressiveness. No significant differences in surgical methods, the extent of laparoscopic gastrectomy and pathological stages were observed between the sarcopenia group and the nonsarcopenia group (Table 1). The multivariate analysis identified the pN stage and sarcopenia as independent prognostic factors for 3-year OS and RFS rates in patients with g-NENs, while surgical methods, the extent of laparoscopic gastrectomy and pT stage were not associated with survival (Table 3). The HR value of sarcopenia changed little between the univariate and multivariate analyses in our study (Table 3). Thus, the prognostic effect of preoperative sarcopenia is less affected by the gastrectomy status and tumor aggressiveness in patients with g-NENs. However, g-NENs are divided into three different pathological types, namely, gNET, gNEC, and gMANEC. The degree of tumor differentiation, grade, and cellular components of the three pathological types are not the same[4], and the treatment strategy and prognosis are also significantly different in patients with different pathological types[37]. In the present study, a further stratified analysis showed relations between sarcopenia and the 3-year OS and RFS rates in patients with gMANEC. Potential explanations for this result are provided below. First, for the subgroup of the gNET population, gNET is a highly differentiated neuroendocrine tumor, with mainly low or moderate malignancy, and presents as stage G1 and G2[3]. The lower tumor invasiveness and the lower effect on skeletal muscle mass may explain why sarcopenia is not useful as a prognostic factor for patients with gNET. This result also may caused by the relatively small number of gNET patients, further study may be required. Second, compared with gNEC and gMANEC, gNEC is a poorly differentiated neuroendocrine carcinoma, which is generally highly malignant and manifests as stage G3. gMANEC is defined as a malignant tumor with morphological components of glandular epithelial cells and neuroendocrine cells, both of which account for at least more than 30% of the total cells[4]. The clinical characteristics of gMANEC generally depend on the proportion of neuroendocrine carcinoma components[38, 39]. Fernandes et al. postulated that the prognosis of patients with gMANEC might be related to whether certain tumor components are more invasive[40]. Furthermore, previous studies have confirmed that sarcopenia is associated with the long-term prognosis of patients with gastric adenocarcinoma[10, 11]. Therefore, we propose that the mechanism may be modulated by the presence of more adenocarcinoma components in gMANEC, and thus, sarcopenia is only related to the long-term prognosis of patients with gMANEC, but not the patients with gNET and gNEC, in the present study. The underlying molecular mechanism must be further elucidated. This result may be caused by the sample sizes of individual subgroups. Further study with bigger sample sizes of different pathological types needed to be conducted.
This study had some limitations. First, because most patients with gNET received endoscopic treatment, the number of patients with gNET included in this study was limited, which may cause bias. Second, this study employed a retrospective case-control design and was conducted in an Asian population; therefore, the results must be confirmed by prospective studies and data from Western countries. Third, the proportion of female patients in this study is relatively small (33/138 cases, 23.9%), and thus the prognostic effect of sarcopenia on female patients with g-NENs must be further analyzed in a study with a larger population. We plan to conduct related studies in the future. Fourth, this study did not analyze the effects of postoperative adjuvant chemotherapy and postoperative sarcopenia caused by the gastrectomy status and tumor aggressiveness on long-term outcomes, which may also bias the results. Fifth, Due to the relatively few cases of stratified analysis of pathological subtypes, we did not identify sarcopenia scores related to tumour type gNET, gNEC and gMANEC, and the prognostic value of sarcopenia for g-NENs may be biased. In the future, a larger sample size is needed to determine the best cut-off point of sarcopenia with different pathological types, and to verify the prognostic effect of sarcopenia on different pathological types of g-NENs. Nevertheless, to our knowledge, this study is the first to explore the effects of sarcopenia on the short-term and long-term outcomes in patients with g-NENs by using data from two independent large-volume institutions, thus providing a reference for future clinical trials.