Abstract:

Objective To explore the predictive value of nutritional risk screening-2002 (NRS-2002) on the efficacy and prognosis of chemoradiotherapy (CRT) in patients with lymph node metastasis (LNM) after esophagectomy of thoracic esophageal squamous cell carcinoma (ESCC). Methods From January 2013 to December 2016, 140 patients with LNM after esophagectomy of thoracic ESCC receiving curative radiotherapy (RT) only or concurrent CRT in the Department of Radiation Oncology, Affiliated Taixing People's Hospital of Bengbu Medical College were retrospectively analyzed. According to the median hemoglobin before treatment, the patients were divided into two groups: hemoglobin<127 g/L group (n=65) and hemoglobin≥127 g/L group (n=75). According to NRS-2002 tool standard, the nutritional risk was scored and grouped by the best cutoff value, patients were divided into NRS-2002<3 group (n=65) and NRS-2002≥3 group (n=75). The logistic regression was used to analyze the factors influencing the short-term efficacy. Kaplan-Meier method was used to calculate survival rate and survival comparison was performed by log-rank test. The univariate and multivariate analyses were carried out by Cox proportional hazards model. Finally, receiver operating characteristic (ROC) curve was plotted and the area under the curve (AUC) was calculated to confirm the prediction accuracy of NRS-2002 in efficacy and survival prognosis. Results The treatment effective rate of the whole group was 65.0% (91/140). The effective rates of the patients in hemoglobin<127 g/L group and hemoglobin≥127 g/L group were 52.3% (34/65) and 76.0% (57/75) respectively, and there was a statistically significant difference (χ²=8.592, P=0.003). The effective rates of the patients in NRS-2002<3 group and NRS-2002≥3 group were 81.5% (53/65) and 50.7% (38/75) respectively, and there was a statistically significant difference (χ ²=14.588, P<0.001). The univariate logistic regression analysis showed that the short-term efficacy of CRT was highly associated with treatment method (HR=1.322, 95%CI: 1.121-3.632, P=0.012), hemoglobin (HR=0.346, 95%CI: 0.169-0.711, P=0.004) and NRS-2002 score (HR=3.374, 95%CI: 1.185-9.611, P<0.001). Multivariate logistic regression analysis found that treatment method (HR=1.125, 95%CI: 1.098-2.927, P=0.042) and NRS-2002 score (HR=3.868, 95%CI: 1.723-8.685, P<0.001) were independent factors for short-term effect of CRT. The median overall survival (OS) of all patients was 16.0 months (95%CI: 14.3-17.7), the 1-, 2- and 3-year survival rates were 69.3%, 28.6% and 21.4%, respectively. In the hemoglobin<127 g/L group, the 1-, 2-, and 3-year survival rates were 62.1%, 21.5% and 10.8%, and in the hemoglobin≥127 g/L group, the 1-, 2-, and 3-year survival rates were 74.7%, 34.7% and 30.7%, with a statistically significant difference (χ ²=7.370, P=0.007). In the NRS-2002<3 group, the 1-, 2-, and 3-year survival rates were 89.2%, 50.8% and 43.1%, and in the NRS-2002≥3 group, the 1-, 2-, and 3-year survival rates were 52.0%, 9.3% and 2.7%, with a statistically significant difference (χ ²=48.778, P<0.001). Cox univariate factor analysis found that treatment method (HR=1.375, 95%CI: 1.108-1.968, P=0.019), hemoglobin (HR=0.607, 95%CI: 0.417-0.885, P=0.009) and NRS-2002 score (HR=0.471, 95%CI: 0.349-0.636, P<0.001) were closely related to OS of patients with LNM after esophagectomy of thoracic ESCC. However, Cox multivariate analysis showed that only NRS-2002 score (HR=0.494, 95%CI: 0.365-0.669, P<0.001) was an independent prognostic factor for OS. ROC curve showed that NRS-2002 score predicted AUC of short-term efficacy and OS were 0.669 (95%CI: 0.576-0.762, P=0.001) and 0.798 (95%CI: 0.719-0.878, P<0.001), respectively. Conclusion Thoracic ESCC patients with LNM receiving CRT after esophagectomy with NRS-2002 score<3 before treatment have a better short-term efficacy and long-term survival.

Key words: Esophageal neoplasms, Lymphatic metastasis, Prognosis, Chemoradiotherapy, Nutritional risk screening