Background
The association between smoking and nasopharyngeal carcinoma (
NPC
) is still uncertain. The aim of this study was to validate smoking effect on
NPC
and explore if smoking can induce
NPC
by persistently reactivating
EBV
in long‐term based on a prospective cohort design.
Methods
A
NPC
screening cohort with 10 181 eligible residents in Sihui city, southern China was conducted from 2008 to 2015. The smoking habit was investigated through the trained interviewers and
EBV
antibodies (
VCA
‐IgA,
EBNA
1‐IgA) as screening markers were tested periodically. New
NPC
cases were identified through local cancer registry. Cox's regression model was used to estimate the adjusted hazard ratios (
aHR
s) of smoking on
NPC
incidence. In the non‐
NPC
participants, the associations between smoking and
EBV
seropositivity in different periods were assessed by logistic regression and generalized estimating equations (
GEE
).
Results
With a median of 7.54 years, 71
NPC
s were diagnosed ≥1 year after recruitment. Compared with never smokers, the
aHR
s of developing
NPC
among ever smokers were 3.00 (95%
CI
: 1.46‐6.16). Stratified by sex, the
HR
s of ever smoking were 2.59 (95%
CI
: 1.07‐6.23) for male and 3.75 (95%
CI
: 1.25‐11.20) for female, respectively. Among the non‐
NPC
individuals, ever smoking was not only associated with
EBV
seropositivity at baseline, but also in the 3‐5 years of follow up, with adjusted odds ratios (
aOR
s) of 1.68 (95%
CI
: 1.29‐2.18) for
VCA
‐IgA and 1.92 (95%
CI
: 1.42‐2.59) for
EBNA
1‐IgA. Among the smokers who were tested
EBV
antibodies at least twice, the similar results were obtained using GEE.
Conclusion
Smoking could significantly increase the long‐term risk of
NPC
in southern China, partly by persistently reactivating
EBV
.