Published online Feb 27, 2022. doi: 10.4240/wjgs.v14.i2.143
Peer-review started: June 6, 2021
First decision: July 16, 2021
Revised: July 24, 2021
Accepted: January 6, 2022
Article in press: January 6, 2022
Published online: February 27, 2022
Processing time: 261 Days and 5.1 Hours
Gastric cancer (GC) is an important public health burden worldwide. The TNM staging system based on tumor infiltration, regional lymph node metastasis, and distant metastasis is considered as the conventional criterion for evaluating prognosis and guiding treatment after surgery. Adjuvant chemotherapy can effectively reduce the disease recurrence. Based on the results of the Adjuvant Chemotherapy Trial of TS-1 for Gastric Cancer (ACTS-GC), stage II/III disease as the standard target of adjuvant chemotherapy after surgery, with the exception of pathological stages T1N2-3 (pT1N2-3) and pT3N0. However, in these two groups, there is still a portion of high-risk patients with a poor prognosis.
Analyzing the independent risk factors for the prognosis of pT1N2-3 and pT3N0 GC patients will provide a basis for clinicians to treat and predict the prognosis of these patients in the future.
To identify the high-risk group among these patients after radical surgery by analyzing biomarkers and clinicopathological features and construct prognostic models for them.
This retrospective study analyzed the clinicopathological characteristics and long-term survival data of 459 patients with pT1N2-3/pT3N0 GC, all of whom underwent radical gastrectomy at the Harbin Medical University Cancer Hospital between January 2000 and April 2016. The chi–square test was used to analyze the differences in the clinicopathological features between the pT1N2-3 and pT3N0 groups. The Kaplan–Meier analysis and log-rank test were used to analyze overall survival (OS). The independent risk factors for patient prognosis were analyzed by univariate and multivariate analyses based on the Cox proportional hazards regression model. The cutoff values of continuous variables were analyzed by receiver operating characteristic curve. The nomogram models were constructed with R studio.
According to the postoperative pathology report, there were 89 and 370 patients in the pT1N2-3 group and pT3N0 group, respectively. There was no statistically significant difference in OS between the pT1N2-3 and pT3N0 groups (P = 0.374). Prealbumin (P = 0.040), carcino-embryonic antigen (CEA) (P = 0.021), and metastatic lymph node ratio (mLNR) (P = 0.035) were independent risk factors for prognosis in the pT1N2-3b group. Age (P = 0.039), body mass index (BMI) (P = 0.002), and gastrectomy (P < 0.001) were independent risk factors for prognosis in the pT3N0 group. The area under the curve values of the nomogram models predicting the 5-year prognosis of the pT1N2-3 group and pT3N0 group were 0.765 and 0.699, respectively.
The nomogram model based on peripheral blood biomarkers and clinicopathological features, including prealbumin, CEA, and mLNR, can be used to predict the prognosis of pT1N2-3 GC patients. Age, BMI, and gastrectomy can be used to predict the prognosis of pT3N0 GC patients.
Further multicentric studies are needed to expand the sample size and external validation of the nomogram models will be performed to determine their predictive ability.