OBJECTIVE To explore the difference between familial and sporadic nasopharyngeal carcinoma patients on risk factors and family history and provide evidence on genetic counseling and screening strategy for relatives of nasopharyngeal carcinoma patients in Guangdong province. METHODS The Cantonese nasopharyngeal carcinoma patients diagnosed in Cancer Center, Sun Yat-sen University from October, 2005 to October, 2007 were recruited as subjects. 1877 patients were collected, including 181 familial nasopharyngeal carcinoma patients and 1696 sporadic nasopharyngeal carcinoma patients. The demographic characteristics, clinical characteristics, risk factors and family history between two groups were compared. Moreover, the distribution of nasopharyngeal carcinoma patients in first-degree relatives and the time interval between proband and the affected first-degree relatives in familial nasopharyngeal carcinoma patients was analyzed. RESULTS All 9.64% of 1877 nasopharyngeal carcinoma patients had affected relatives in first-degree relatives, among them, 58.49% (124/212) were siblings and 41.51% (88/212) were parents. The mean time interval between siblings and proband were (7.40 +/- 5.41) years while the mean time interval between parents and proband were (15.55 +/- 10.61) years when nasopharyngeal carcinoma occurred, and the difference was statistically significant (t = -5.78, P < 0.01). More than 80% patients of the two group were at advanced stage when they were diagnosed. There were no difference (P values were all > 0.05) both in adulthood and childhood in salted fish (OR = 1.01; 95% CI: 0.59 - 1.75 vs OR = 1.31; 95% CI: 0.92 - 1.86), preserved vegetables (OR = 0.93; 95% CI: 0.58 - 1.49 vs OR = 1.12; 95% CI: 0.80 - 1.57), fermented pastes (OR = 0.37; 95% CI: 0.14 - 1.01 vs OR = 1.61; 95% CI: 0.99 - 2.48), fresh fruits (OR = 0.87; 95% CI: 0.60 - 1.26 vs OR = 0.65; 95% CI: 0.20 - 2.12) and cured meat (OR = 1.26; 95% CI: 0.87 - 1.83 vs OR = 1.28; 95% CI: 0.71 - 2.30) diet. No significant difference (P > 0.05) was obtained on smoking (OR = 0.99; 95% CI: 0.68 - 1.45) and incidence of other cancers in first-degree relatives (OR = 0.85; 95% CI: 0.56 - 1.28) in the two groups. CONCLUSION Familial nasopharyngeal carcinoma was 9.64% in the observed subjects. In the familial nasopharyngeal carcinoma, the time interval at diagnosis was shorter between proband and siblings as compared with parents. Most of the patients were at advanced stage. So, we recommend the first-degree relatives of nasopharyngeal carcinoma patients, especially siblings, should be screened regularly according to the specific conditions.