Causes of death analysis and the prognostic model construction in neuroendocrine carcinoma of the cervix: A SEER‐based study

Abstract

AbstractPurposeNeuroendocrine carcinoma of the cervix (NECC) is rare but results in poor prognosis. The causes of death (CODs) in NECC patients are rarely reported. Our study aimed to explore the distributions of death causes of NECC patients compared with squamous cell carcinoma (SCC) and adenocarcinoma (ADC) and to develop a validated survival prediction model.MethodsPatients diagnosed with NECC, SCC, or ADC were identified from the Surveillance, Epidemiology, and End Results Program database from 1975 to 2019. We analyzed the standardized mortality ratio (SMR) to determine each cause of death for each survival time category. The Kaplan–Meier method was used for survival analysis. Univariate and multivariate Cox regression analyses were used to establish a nomogram model.ResultsA total of 358 NECC patients were included in this study, and 270 (75.4%) died during the follow‐up period. Patients with NECC had 5.55 times (95% CI, 4.53–6.79, p < 0.0001) higher risk of death compared with patients with SCC and 10.38 times (95% CI, 8.28–13.01, p < 0.0001) higher compared with ADC. Cervical cancer is the main cause of death in NECC. As the diagnosis time increased, the risk of death from all causes and cervix cancer gradually decreased. While after at least 10 years of follow‐up time, the highest and most dramatical SMR values were observed for metastasis (SMR, 138.81; 95% CI, 37.82–355.40; p < 0.05) and other cancers as the reason for death has an over 7‐fold higher SMR (SMR: 7.07; 95% CI: 2.60–15.40, p < 0.05) more than 5 years after the cancer diagnosis. Race, FIGO stage, and surgery were independent risk factors for the overall survival (OS) of NECC patients. For the predictive nomogram, the C‐index was 0.711 (95% CI: 0.697–0.725) and was corrected to 0.709 (95% CI: 0.680, 0.737) by bootstrap 1000 resampling validation.ConclusionCompared with SCC and ADC, NECC patients have an elevated risk of mortality due to cervical cancer and metastasis. We successfully constructed a prognostic nomogram for patients with NECC. Based on refractoriness and high mortality of NECC, targeted treatment strategies and follow‐up plans should be further developed according to the risk of death and distribution characteristics of CODs.