PurposeNeoadjuvant chemoradiotherapy (nCRT) is a standard treatment option for patients with stage III oesophageal cancer. Approximately 30% of oesophageal cancer patients will have a pathological complete response (pCR) after nCRT. However, available clinical methods cannot accurately predict pCR for patients. We aimed to find more indicators that could be used to predict the pathological response to nCRT.MethodA total of 84 patients with stage III oesophageal squamous cell cancer were enrolled in this study. Ten patients failed to have surgery as a result of progressive disease (PD). Among the patients who underwent surgery, 32 patients had a pathologic complete response (pCR), whereas 42 patients showed no or partial response (npCR) after nCRT. Routine blood test results and lymphocyte subset assessments before and after nCRT were retrospectively analysed. Univariate and multivariate analyses were used to identify independent predictors of the clinical curative effect of nCRT. Eventually, nomograms were established for predicting the PD and pCR rates.ResultsThe numbers of lymphocytes, B lymphocytes, T lymphocytes, Th lymphocytes, Ts lymphocytes, and NK cells and the percentages of B lymphocytes and NK cells were decreased significantly after nCRT (P < 0.0001), whereas the percentages of T lymphocytes and Ts lymphocytes increased (P < 0.0001). Univariate analysis showed that age, the length of the lesion, the level of haemoglobin before nCRT, and the amount of change in haemoglobin were related to PD, and the percentage of NK cells after nCRT was related to pCR. Multivariate logistic analysis demonstrated that the length of the lesion, the neutrophil-to-lymphocyte ratio (NLR) before nCRT, and the amount of change in haemoglobin were independent predictors of PD, whereas the percentage of NK cells after nCRT was an independent predictor of pCR.ConclusionLymphocyte subsets changed dramatically during nCRT, and these changes together with baseline and posttreatment lymphocyte subsets have predictive value in determining the response to nCRT for oesophageal cancer.