Disparities of birth cohort effects on pancreatic cancer incidence between the United States and urban China.

BACKGROUND: Pancreatic cancer has been associated with lifestyle factors, but few comparative studies were conducted among countries of different culture and lifestyle habits. This study compared the trends of pancreatic cancer incidence and birth cohort effects in the United States and urban China and explored the potential discrepancies of risk patterns. MATERIALS AND METHODS: Age-standardized incidence rates (ASIRs) were calculated using data from national or regional cancer registries of the United States and two large cities of China (Shanghai, Hong Kong). The temporal trends of incidence were assessed by joinpoint regression. The effects of birth cohort and calendar period were identified through age-period-cohort modeling. RESULTS: The ASIR in the United States from 1976 to 2015 was 8.26/100 000, which was higher than that in Hong Kong (4.29/100 000) and Shanghai of China (6.63/100 000). Shanghai had lower incidence (4.41/100 000) in 1976-1980 but increased annually by 1.38% in males and 1.67% in females, with a sharper upward trend than the United States and Hong Kong. Males had higher risks than females, with a male-to-female ratio of 1.34, 1.44, and 1.37 in the United States, Hong Kong, and Shanghai, respectively. A significant and prominent increase in incidence rate was observed among successive generations in China particularly for Shanghai, but such a pattern was not apparent in the United States. CONCLUSIONS: The differences in pancreatic cancer incidence by sex may be multi-factorial involving known risk factors like tobacco smoking and alcohol consumption. The significant birth cohort effects among recent and early generations in the Shanghai population were in line with a society in socioeconomic transition and adoption of Western lifestyle mainly including consumption of calorie-rich foods and physical inactivity. Differences in these risk patterns will have implications on health care efforts and policies for cancer control.

Long-term Survival Outcomes of 'Low Risk' Ductal Carcinoma in situ from a Territory-wide Cancer Registry.

AIMS: The LORIS trial is an ongoing phase III clinical trial on low risk ductal carcinoma in situ (DCIS). DCIS patients aged ≥46 years with screen-detected low/intermediate nuclear grade were considered low risk and were randomised into surveillance or standard surgery. Here we review the 10-year territory-wide breast cancer registry database and evaluate the clinical outcomes of low versus high risk DCIS patients. MATERIALS AND METHODS: This was a retrospective study of a prospectively maintained territory-wide breast cancer registry in Hong Kong. RESULTS: Between 1997 and 2006, 1391 DCIS patients were identified from the Hong Kong cancer registry breast cancer database. The mean age at diagnosis was 49.2 years (range 30-70). In total, 372 patients were classified as 'low risk', whereas the remaining 777 patients were classified as 'high risk'. After a median follow-up of 11.6 years, the 10-year overall breast cancer-specific survival of the entire DCIS cohort was 1136/1149 (98.9%). Overall breast cancer-specific survival of low risk DCIS was 99.5%, whereas that in high risk DCIS was 98.6% (Log-rank test, P = 0.208). Forty-six (12.4%) patients in the LORIS low risk group did not receive surgery, whereas 93 (12%) patients in the LORIS high risk group did not receive surgery. The 10-year breast cancer-specific survival in the non-operated low risk DCIS group was 97.8%; that in the non-operated high risk DCIS group was 96.7% (P = 1). CONCLUSION: Long-term survival of DCIS was excellent, especially in low risk DCIS, regardless of surgical treatment.

Clinical outcomes of patients with ductal carcinoma in situ in Hong Kong: 10-year territory-wide cancer registry study.

BACKGROUND: Incidence of ductal carcinoma in situ (DCIS) has increased in recent decades because of breast cancer screening. This study comprised a long-term survival analysis of DCIS using 10-year territory-wide data from the Hong Kong Cancer Registry. METHODS: This study included all patients diagnosed with DCIS in Hong Kong from 1997 to 2006. Exclusion criteria were age <30 years or ≥70 years, lobular carcinoma in situ, Paget's disease, and co-existing invasive carcinoma. Patients were stratified into those diagnosed from 1997 to 2001 and those diagnosed from 2002 to 2006. The 5- and 10-year breast cancer-specific survival rates were evaluated; standardised mortality ratios were calculated. RESULTS: Among the 1391 patients in this study, 449 were diagnosed from 1997 to 2001, and 942 were diagnosed from 2002 to 2006. The mean age at diagnosis was 49.2±9.2 years. Overall, 51.2% of patients underwent mastectomy and 29.5% received adjuvant radiotherapy. The median follow-up interval was 11.6 years; overall breast cancer-specific mortality rates were 0.3% and 0.9% after 5 and 10 years of follow-up, respectively. In total, 109 patients (7.8%) developed invasive breast cancer after a considerable delay. Invasive breast cancer rates were comparable between patients diagnosed from 1997 to 2001 (n=37, 8.2%) and those diagnosed from 2002 to 2006 (n=72, 7.6%). CONCLUSION: Despite excellent long-term survival among patients with DCIS, these patients were more likely to die of breast cancer, compared with the general population of women in Hong Kong.

Cancer survival in Hong Kong SAR, China, 1996-2001.

The Hong Kong cancer registry was established in 1963, and cancer registration is done by passive and active methods. The registry contributed data on 45 cancer sites or types registered during 1996-2001 for this survival study. Follow-up has been carried out by passive methods with median follow-up ranging from 4-60 months. The proportion of cases with histologically verified cancer diagnosis ranged from 38-100%; death certificates only (DCOs) ranged from 0-11%; 83-99% of total registered cases were included for survival analysis. The 5-year age-standardized relative survival exceeded 100% for lip and non-melanoma skin followed by thyroid (94%) and testicular (92%) cancers. The corresponding survival for common cancers were breast (90%), colon (61%), liver and Lung (22%), nasopharynx (70%), rectum (59%) and stomach (39%). The 5-year relative survival by age group showed a decreasing trend with increasing agegroups for most cancers. A decreasing survival with increasing clinical extent of disease was noted.

Incidence rate trends of histological subtypes of nasopharyngeal carcinoma in Hong Kong.

The overall decline in incidence rate of nasopharyngeal carcinoma in Hong Kong during 1988-2002 was limited primarily to a decrease in keratinising carcinoma, which could be explained by the decline in cigarette smoking. Genetic and Epstein-Barr virus interactions may explain the relatively stable incidence rate of non-keratinising carcinoma.

An ecological study of trends in cancer incidence and dietary changes in Hong Kong.

Cancer incidence rates from the Hong Kong Cancer Registry show significant increases in lung and colon cancers and decreases in nasopharyngeal cancer in both sexes from 1973 to 1992. Moreover, cervical cancer and male esophageal cancer have declined significantly, and changes in the trends of cancer of the following sites were of borderline significance: decreasing male laryngeal and female esophageal cancers and increasing prostate and female breast cancers. These changes have occurred along with dietary shifts in the population, from a diet predominantly of rice and small portions of meat, vegetables, and fish to one with larger portions of all foods but rice and eggs. The latter data were gathered from six government household surveys from 1963-64 to 1994-95. By combining the two data sets, correlation coefficients were calculated for per capita consumption patterns of eight foods (rice, pork, beef, poultry, saltwater fish, freshwater fish, fresh vegetables, and eggs) and cancer incidence data of the same year or 10 years later. Higher meat intakes were significantly and positively correlated with cancers of the colon, rectum, prostate, and female breast. The correlations also suggested that current diets were more influential than diets a decade before for cancers of the lung, esophagus, rectum, and prostate. Cancers of the nasopharynx and colon were significantly correlated with current and past diets. These results support the hypothesis that intakes of meat and its associated fat are risk factors for colon, rectal, prostate, and female breast cancers.