Combined Features Based on Preoperative Controlling Nutritional Status Score and Circulating Tumour Cell Status Predict Prognosis for Colorectal Cancer Patients Treated with Curative Resection

Background: The preoperative controlling nutritional status (CONUT) score and circulating tumour cell (CTC) status are associated with poor prognosis of colorectal cancer (CRC). The aim of the present study is to determine whether the combination of CONUT and CTC (CONUT-CTC) could better predict the prognosis of CRC patients treated with curative resection. Methods: Preoperative CONUT score was retrospectively calculated in 160 CRC patients who underwent curative resection at Zhongnan Hospital of Wuhan University from 2015 to 2017. Preoperative CTC counts were enumerated from 5 ml peripheral vein blood by a CTCBIOPSY (R) device. According to the preoperative CONUT and CTC status, the patients were divided into three groups: CONUT-CTC (0), CONUT-CTC (1) and CONUT-CTC (2). The relationship between CONUT score and CTC, as well as the associations of CONUT-CTC status with clinicopathological factors and survival, were evaluated. Results: Preoperatively, the number and positive rate of CTC were positively correlated with the preoperative CONUT score (P<0.01). An elevated CONUT-CTC score was significantly associated with deeper tumour invasion (P=0.025), lymphatic vessel invasion (P=0.002), venous invasion (P<0.001) and higher pTNM stage (P=0.033). Kaplan-Meier analysis and log-rank tests revealed significant decreases in recurrence-free survival (RFS) and cancer-specific survival (CSS) among CRC patients with CONUT-CTC score of 0, 1 and 2 (P<0.001). In pTNM stage-stratified analysis, high CONUT-CTC score was significantly associated with the poor (P<0.001) and CSS (P<0.001) of patients with stage III disease, but not correlated with the prognosis of patients with stage II disease (RFS: P=0.077; CSS: P<0.090). Further univariate and multivariate analyses showed that CONUT-CTC was an independent factor affecting patients' RFS [hazard ratio (HR)=2.66, 95% confidence interval (CI):1.79-3.96, P<0.001] and CSS (HR=3.75, 95%CI: 2.14-6.57, P<0.001). In time-dependent receiver operating characteristics (ROC) analyses, CONUT-CTC score had a higher area under the ROC curve (AUC) for the prediction of RFS and CSS than did preoperative CONUT score or CTC status. Conclusion: The preoperative CONUT-CTC score is associated with tumour progression and poor prognosis in patients with CRC treated with curative resection, indicating that better information on CRC prognosis could be obtained from combined preoperative host immune-nutritional status and CTC detection.