5-year prospective cluster randomised controlled study of a new nasopharyngeal carcinoma screening programme

• Published in: The Lancet (British edition) 2015-10, Vol.386, p.S4-S4
• Main Authors: Ji, Mingfang, MD, Sheng, Wei, MSc, Cheng, Weimin, MD, Zhang, Jun, Prof, Cao, Su Mei, MD, Yu, Xia, MD, Huang, Songlin, MD, He, Shuming, Prof, Huang, Hanwei, MD, Wu, Biaohua, MD, Wei, Kuangrong, MD, Liu, Xiaodong, MD, Lian, Shifeng, MD, Chen, Honglin, PhD, Zheng, Yingjie, Prof, Huang, Shoujie, MSc, Ge, Shengxiang, PhD, Ng, Park S P, PhD, Liu, Qing, Prof, Hong, Minghuang, MD, Zeng, Yi Xin, Prof, Yuan, Yong, Prof, Xia, Ningshao, Prof, Ng, Mun-Hon, Prof
• Format: Article
• Language: English
• Subjects: Internal Medicine; Towns
• ISSN: 0140-6736; 1474-547X
• DOI: 10.1016/S0140-6736(15)00582-6

Abstract Background Nasopharyngeal carcinoma is common in China, where the incidence is as high as 2·44 per 100 000 people per year, and usually presents at advanced stage when the prognosis is poor. The human Epstein-Barr tumour virus is thought to be an important cause of this cancer. Screening programmes using the Epstein-Barr virus-specific IgA immunoglobulin antibody subtype as the principle screening test can result in early diagnosis, but the long-term effects have not been ascertained. Methods In a mass screening programme between August, 2009, and December, 2014, three townships of Zhongshan city in southern China were randomly assigned: one as the screening town and two as control towns. 16 695 residents of the screening town aged 30–59 years were enrolled, and each was randomly matched for age, sex, and date of enrolment with two residents of the control towns, resulting in 33 390 control participants in total. Participants of the screening group were tested for serum anti-Epstein-Barr virus IgA concentrations using two ELISAs. Those with moderate antibody concentrations or higher (≥0·65 probability of nasopharyngeal carcinoma units [logit P]) were invited to be retested annually in the following 3 years, and those with high antibody concentrations (≥0·98 logit P) were referred to otorhinolaryngologists for diagnostic work-up. Findings 43 nasopharyngeal carcinoma cases, with two deaths, were reported in the screening group, compared with 52 cases, with eight deaths, in the control group. The sensitivity of the screening triage was 95%, specificity was 94%, positive predictive value was 3·4%, and negative predictive value was 99·9%. Cumulative incidence was 0·26% per person for the screening group and 0·013% for those yielding a negative result, compared with 0·16% for the control group. 35 (82%) cases identified by the screening triage were localised disease, compared with 9 (19%) control cases (p