Predictive value of tumor regression rate after induction chemotherapy for survival of patients with locally advanced nasopharyngeal carcinoma
宋扬,王斌,王阁,肖何,耿明英,陈川
DOI: https://doi.org/10.3760/cma.j.cn371439-20201116-00031
2021-01-01
Abstract:Objective:To investigate the predictive value of tumor regression rate after induction chemotherapy for survival of patients with locally advanced nasopharyngeal carcinoma.Methods:A total of 161 patients with stage Ⅲ-ⅣA nasopharyngeal carcinoma newly diagnosed at the Daping Hospital of Army Medical University from January 2009 to December 2012 were selected as the research subjects. The relationships between tumor size changes before and after induction chemotherapy and survival time were analyzed. Kaplan-Meier method was used to draw the survival curve accompanied with log-rank test. Cox regression analysis was used to analyze the risk factors affecting the prognosis of patients with nasopharyngeal carcinoma.Results:There were statistically significant differences in the tumor regression rate of primary lesions between N 1and N 2-3( Z=2.177, P=0.029), T 1-2and T 3-4( Z=-4.501, P<0.001)patients after induction chemotherapy. In N 1stage patients, the 5-year overall survival (OS) rates of patients with primary lesions achieving objective response ( n=18) and those without objective response ( n=19) after induction chemotherapy were 88.89% and 57.45%, and patients with cervical lymph node metastatic lesions achieving objective response ( n=19) and those without objective response ( n=18) were 86.72% and 49.10% respectively, with statistically significant differences ( χ2=6.023, P=0.014; χ2=7.441, P=0.006). In N 2-3stage patients, the 5-year OS rates of patients with primary lesions achieving objective response ( n=81) and those without objective response ( n=43) after induction chemotherapy were 77.56% and 50.70%, and patients with cervical lymph node metastatic lesions achieving objective response ( n=85) and those without objective response ( n=39) were 75.11% and 52.04% respectively, with significant differences ( χ2=8.037, P=0.005; χ2=7.268, P=0.007). Univariate Cox regression analysis showed that in patients with stage N 1, the tumor regression rate of primary lesions ( HR=0.048, 95% CI: 0.004-0.644, P=0.022), the efficacy of primary lesions ( HR=0.174, 95% CI: 0.037-0.830, P=0.028), the efficacy of cervical lymph node metastatic lesions ( HR=0.154, 95% CI: 0.033-0.725, P=0.017) after induction chemotherapy were significantly associated with OS; in N 2-3stage patients, the tumor regression rate of primary lesions ( HR=0.178, 95% CI: 0.056-0.564, P=0.003), the tumor regression rate of cervical lymph node metastatic lesions ( HR=0.081, 95% CI: 0.020-0.324, P<0.001), the efficacy of primary lesions ( HR=0.422, 95% CI: 0.228-0.781, P=0.006), the efficacy of cervical lymph node metastatic lesions ( HR=0.439, 95% CI: 0.238-0.813, P=0.009) after induction chemotherapy were significantly associated with OS. In multivariate Cox regression including N stage and tumor regression rate, N stage and efficacy, the interaction items were not statistically significant (all P>0.05). In T 1-2stage patients, the 5-year OS rates of patients with primary lesions achieving objective response ( n=45) and those without objective response ( n=13) after induction chemotherapy were 77.55% and 84.62%, and patients with cervical lymph node metastatic lesions achieving objective response ( n=43) and those without objective response ( n=15) were 78.89% and 80.00% respectively, with no significant differences ( χ2=0.239, P=0.625; χ2=0.005, P=0.943); in T 3-4stage patients, the 5-year OS rates of patients with primary lesions achieving objective response ( n=54) and those without objective response ( n=49) after induction chemotherapy were 78.90% and 45.00%, and patients with cervical lymph node metastatic lesions achieving objective response ( n=61) and those without objective response ( n=42) were 75.10% and 42.89% respectively, with significant differences ( χ2=13.615, P<0.001; χ2=12.752, P<0.001). Univariate Cox regression analysis showed that in patients with stage T 1-2, the tumor regression rate, the efficacy of primary lesions and cervical lymph node metastatic lesions after induction chemotherapy were not related to OS (all P>0.05); in T 3-4stage patients, the tumor regression rate of primary lesions ( HR=0.121, 95% CI: 0.033-0.444, P=0.001), the tumor regression rate of cervical lymph node metastatic lesions ( HR=0.126, 95% CI: 0.036-0.442, P=0.001), the efficacy of primary lesions ( HR=0.297, 95% CI: 0.150-0.588, P<0.001), the efficacy of cervical lymph node metastatic lesions ( HR=0.329, 95% CI: 0.173-0.625, P=0.001) after induction chemotherapy were significantly associated with OS. Multivariate Cox regression analysis showed that the interaction test of T stage and the efficacy of primary lesion trended to be statistically significant ( P=0.062). Conclusion:In patients with stage Ⅲ-ⅣA nasopharyngeal carcinoma, the responsiveness to induction chemotherapy in stage T 3-4patients has important value in predicting survival prognosis.