Tobacco smoking and the risk of subsequent primary cancer among cancer survivors: a retrospective cohort study.

BACKGROUND: Smoking is a well-known risk factor for cancer; however, there is little evidence as to whether the smoking status of cancer survivors has any risk for subsequent primary cancer (SPC) incidence, regardless of the first cancer sites. PATIENTS AND METHODS: In total, 29,795 eligible patients with a first cancer between 1985 and 2004 were examined for SPC until the end of 2006, using a record linkage between hospital-based and population-based cancer registries. The association between smoking at the time of the first cancer diagnosis and three SPC groups (i.e. specific SPC, smoking-related SPCs, and all SPCs) was calculated by Poisson regression. RESULTS: Ever smokers had 59% and 102% higher risk for all SPCs and smoking-related SPCs, respectively, than never smokers. Cancer survivors who had recently stopped smoking had 18% and 26% less risk, respectively, for these SPCs than those who smoked at the diagnosis. We also found that, compared with those who had never smoked, cancer survivors who had ever smoked had a significantly elevated risk of oral/pharyngeal, esophageal, stomach, lung, and hematological SPCs, regardless of the first cancer sites. CONCLUSIONS: These findings indicate that smoking increases not only the first cancer but also a second or SPC. Moreover, the results from recent quitters versus current smokers suggest that smoking cessation may decrease the risk for SPC, especially for smoking-related SPCs in cancer survivors. Preventive measures are necessary to reduce not only SPC incidence but also tobacco use.

Risk factors for intrahepatic cholangiocarcinoma: a possible role of hepatitis B virus.

There are several established risk factors for intrahepatic cholangiocarcinoma (ICC), namely primary sclerosing cholangitis, fibropolycystic liver disease, parasitic infection, intrahepatic biliary stones and chemical carcinogen exposure. However, the majority of patients with ICC do not have any of these risk factors. Therefore, identification of other risk factors is warranted for the prevention and early detection of ICC. We evaluated the risk factors for ICC in a large-scale cohort study in the province of Osaka, Japan. This retrospective cohort study included 154,814 apparently healthy individual blood donors, aged 40-64 years at the time of blood donation in the period 1991-1993. The average observation period was 7.6 years, resulting in 1.25 million person-years of observation. Incident ICC cases were identified by linking the blood-donor database to the records in the population-based cancer registry for the province. There were 11 incident ICC cases during follow-up, with an incidence rate of 0.88 per 100,000 person-years. Compared with subjects aged 40-49 years, the subjects aged 50-54 years and 55-59 years had a significantly higher risk for ICC (hazard ratio [HR] = 5.90; 95%CI:1.08-32.31 and 11.07; 95%CI:1.98-61.79, respectively). Compared with those with ALT level of 19 Karmen Units (KU) or less, subjects with ALT level of 40 KU or higher had a significantly higher risk for ICC (HR: 8.30; 95%CI:1.47-46.83). Compared with those who tested negative for both HBsAg and anti-HCV, those who tested HBsAg-positive had a significantly higher risk for ICC (HR: 8.56; 95%CI: 1.33-55.20). Our results suggest that HBV infection and liver inflammation are independently associated with ICC development. These findings need to be verified by further large cohort studies.

Relationship between the duration of the preoperative smoke-free period and the incidence of postoperative pulmonary complications after pulmonary surgery.

STUDY OBJECTIVE: To examine the relationship between the duration of the preoperative smoke-free period and the development of postoperative pulmonary complications (PPCs) in patients who underwent pulmonary surgery, and the optimal timing of quitting smoking. DESIGN: Retrospective cohort study. SETTING: Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan. PATIENTS: Two hundred eighty-eight consecutive patients who underwent pulmonary surgery between January 1997 and December 1998. MEASUREMENTS AND RESULTS: We collected information on the preoperative characteristics, intraoperative conditions, and occurrence of PPCs by reviewing the medical records. Study subjects were classified into four groups based on their smoking status. A current smoker was defined as one who smoked within 2 weeks prior to the operation. Recent smokers and ex-smokers were defined as those whose duration of abstinence from smoking was 2 to 4 weeks and > 4 weeks prior to the operation, respectively. A never-smoker was defined as one who had never smoked. The incidence of PPCs among the current smokers and recent smokers was 43.6% and 53.8%, respectively, and each was higher than that in the never-smokers (23.9%; p < 0.05). The moving average of the incidence of PPCs gradually decreased in patients whose smoke-free period was 5 to 8 weeks or longer. After controlling for sex, age, results of pulmonary function tests, and duration of surgery, the odds ratios for PPCs developing in current smokers, recent smokers, and ex-smokers in comparison with never-smokers were 2.09 (95% confidence interval [CI], 0.83 to 5.25), 2.44 (95% CI, 0.67 to 8.89), and 1.03 (95% CI, 0.47 to 2.26), respectively. CONCLUSIONS: These findings indicate that preoperative smoking abstinence of at least 4 weeks is necessary for patients who undergo pulmonary surgery, to reduce the incidence of PPCs.

Time trends of the prevalences of hepatitis B e antigen positives among hepatitis B virus carriers in Japan.

In order to examine time trends of the prevalences of HBeAg positives among HBV carriers in Japan, we analysed data on HBeAg of HBsAg positive voluntary blood donors (23,560 males, and 8659 females) at the Osaka Red Cross Blood Centre between January 1977 and March 1984. Age-specific prevalences of HBeAg positives decreased year by year for both sexes, especially for those in their teens and twenties. The prevalences of HBeAg positives decreased with age, but at any given age it was lower for the later than for the earlier birth cohorts. Although reasons for the secular declines are unknown, the findings suggest that the prevalence of HBeAg positives among HBV carriers will continue to decrease in Japan. This, together with the immunization programme implemented this year, may lead to a future reduction in the risk of HBV related liver diseases in Japan.

Helicobacter pylori, atrophic fundal gastritis and risk for gastric adenocarcinoma.

Fundal atrophic gastritis and Helicobacter pylori have been implicated as possible etiologic factors in gastric cancer. This case-control study was performed to determine which risk factor is more closely related to gastric cancer. The endoscopic Congo red test was performed to evaluate the extent of fundal atrophic gastritis in 43 patients with gastric cancer and 86 cancer-free control subjects, who were individually matched by age, sex, and date of endoscopy (within 3 months). The prevalance of H. pylori infection and severe fundal gastritis were significantly higher in patients with differentiated adenocarcinoma, but not with undifferentiated adenocarcinoma, than in control subjects. The odds ratios for differentiated and undifferentiated adenocarcinomas were 6.85 (95% confidence interval, 1.94-11.82) and 1.50 (95% CI, 0.84-3.11), respectively. However, the odds ratio of H. pylori infection was greater than that of severe fundal gastritis. Moreover, multivariate analysis provided similar results. H. pylori infection is an independent indicator of a higher risk of the differentiated adenocarcinomas of the stomach than is severe fundal gastritis.

Evaluation of tissue harmonic imaging for the diagnosis of focal liver lesions.

This was a prospective study to evaluate tissue harmonic imaging (THI) for the diagnosis of focal liver lesions. A total of 15 reviewers read 100 randomly arranged liver images, a fundamental grey-scale image (FGI) and a THI (transmitted: 2 MHz, received: 4 MHz) of each of 50 patients (29 with liver cirrhosis, 42 with focal lesions) taken from the same section. The mean value of overall accuracy for detecting lesions (presence or absence) was significantly higher with THI (82.3%) than with FGI (79.6%) (t = 1. 96, p< 0.05). When only the 29 cirrhosis patients were analyzed, the difference was more significant (t = 2.48, p < 0.02). The correct count rate of the number of focal lesions was higher with THI (78. 0%) than with FGI (67.0%) (t = 3.61, p< 0.005) in 23 cirrhosis patients with focal lesions. The correct diagnosis of HCC was achieved at a higher rate with THI (42.5%) than with FGI (36.8%). THI was statistically effective for detecting focal lesions, particularly in cirrhotic livers.

[Role of population-based cancer registry in cancer epidemiology--epidemiological studies in the Cancer Registration Scheme in Osaka, Japan].

A population-based cancer registration scheme started in three areas in Japan in the 1950s solely for studying cancer incidence in their respective areas. Soon thereafter, several prefectural governments started their own schemes as part of their cancer control programs, effectively expanding the aims of cancer registration: to clarify cancer facts, to elevate the medical care for cancer patients, and to plan and evaluate cancer control programs. The Osaka Cancer Registry (OCR) started in 1962, and has been using epidemiological methods as a tool in constructing its registration scheme, as well as analyzing and utilizing registry data. This report deals with the results obtained in the OCR, classifying these results into four activities of epidemiology. 1. Clarifying cancer facts (descriptive epidemiology): The OCR has been observing incidence, medical care for cancer patients, distribution of cases by clinical stage, and the 5-year relative survival rate, and has estimated the prevalence rate, cured-case rate, and future incidence into the 21st century. Population-based data on histology and multiple cancers collected at the OCR have also contributed to the new approaches in cancer epidemiology. 2. Research on risk factors (analytical epidemiology): The OCR developed a computerized record-linkage system in 1970. This not only made registry work more effective and reliable, but many cohort studies were able to be conducted with relatively little effort and highly reliable results. The cancer case file in the OCR has been linked with the newly prepared data file of the study group, and cancer incidence among the study group has been observed. Finally, cancer risks of possible causal factors in that group have been estimated quantitatively. 3. Evaluation of control programs: Secondary prevention programs (early detection) have been conducted in Japan as major cancer control programs, because effective risk factors were not previously defined. OCR data have been used for estimating sensitivity and specificity of screening tests for various cancers, as well as for evaluating the effect of clinical work on improving survival and on decreasing cancer deaths. 4. Planning future cancer control programs: The OCR has reported on the probable rapid increase of cancer incidence in the 21st century, especially of elderly cancer cases, and cancer cases with poor survival. To control these difficult problems, new cancer programs should be urgently designed and implemented. The authors have recommended that programs be prepared by cancer site, and have already presented a detailed program for lung cancer control.

[Index for evaluating completeness of registration in population-based cancer registries and estimation of registration rate at the Osaka Cancer Registry between 1966 and 1992 using this index].

The quality of population-based cancer registries has been measured by the indices of the proportion of total incident cases (DCO%) registered by death certificate only (DCO), and the ratio of incidence to mortality (I/D ratio). Recently it has been recommended that DCO% should be used as an index for the reliability of diagnosing cancers and that the proportion of cases first notified via death certificate (DCN, DCN%) be used as an index for the completeness of registration. Parkin introduced a method to estimate the registration rate, the estimated proportion of the "true incidence" that are registered in population-based registries. We recommend a modified method for estimating the registration rate for cancer registries where DCN% is relatively high, as it is in Japan, as Parkin's method may overestimate the registration rate. The method is as follows: the registration rate = (1-DCN% x 1/D ratio)/(1-DCN%). The registration rates at the Osaka Cancer Registry between 1966 and 1992 were estimated using our method. During this period, the yearly registration rate was 74.6-78.4% for males and 69.1-73.3% for females. When the cancer cases were looked at according to site, the yearly registration rate was 74.2-81.6% for stomach cancer, 81.2-89.3% for lung cancer, and 71.3-76.9% for uterine cancer. These results show that the registration rate is high for cancers that have an unfavorable prognosis and low for cancers that have a favorable prognosis. We recommend that all cancer registries in Japan calculate the completeness of registration by utilizing DCN defined as the sum of DCO plus cases not reported as cancer but with supportive clinical information of such obtained through survey of the registry for DCN.

[Risk factors of hepatitis C virus infection].

In order to clarify the risk factors of hepatitis C virus (HCV) infection, past medical histories, family histories of liver disease, and lifestyle factors (smoking and drinking habits) were compared between anti-HCV positive cases and anti-HCV negative controls. Studies were 433 anti-HCV positive cases and 298 negative controls, all of whom were out-patients with chronic hepatitis or liver cirrhosis. The findings were contrasted with the risk factors of hepatitis B virus (HBV) chronic infection. Multiple logistic regression analysis revealed the following findings. Prevalence of anti-HCV among patients with chronic hepatitis or liver cirrhosis increased with age, while prevalence of HBsAg decreased with age. Histories of blood transfusion were closely associated with the risk of HCV infection. Adjusted relative risk for patients with a history of blood transfusion was estimated to be 2.97 (95% confidence intervals: 1.88-4.71) in comparison with the risk for patients without it. Any family history of liver disease was not related with the risk of HCV infection, while mother's history of liver disease was significantly associated with the risk of chronic HBV infection (adjusted relative risk was 2.96 with 95% confidence intervals: 1.32-6.62).

[Time-trends in cancer incidence and mortality in Japan].

The cancer incidence and mortality in Japan are described herein. The total number of deaths from all malignant neoplasms in 1998 was 284,000, corresponding to 30.3% of the total number of deaths among Japanese. Lung was the leading site of cancer deaths (17.9%), followed by the stomach (17.9%), large bowel (12.1%), liver (11.8%), pancreas (6.2%), gallbladder and extrahepatic biliary tract (5.2%), lymphatic tissue (3.7%), esophagus (3.4%), breast (3.1%), prostate (2.4%), leukemia (2.3%) and uterus (1.8%). The stomach was the leading site of cancer deaths until 1997, but was replaced by the lungs in 1998. The age-standardized mortality rates (1975-1998) have increased gradually for males, but decreased slightly for females. The rates have decreased remarkably for the stomach and uterus, while increasing for the lungs, large bowel, female breast, gallbladder and extrahepatic biliary tract, pancreas, and others. The total incidence for all cancers in Japan was estimated to be 454,000 in 1995. The stomach was the leading cancer site (22.2%), followed by the large bowel (17.7%), lung (11.6%), liver (7.8%), breast (6.6%), pancreas (3.7%), gallbladder and extrahepatic biliary tract (3.4%), lymphatic tissue (3.1%) and uterus (2.9%). The age-standardized incidence rates for all sites (1975-1995) have increased gradually for males, while remaining constant for females after a slight increase in the late 1970s. The incidence rates have decreased for the stomach and uterus, but increased for the large bowel, female breast, lung, liver, gallbladder and extrahepatic biliary tract, pancreas, prostate and others. The increase in the incidence rate was prominent for the large bowel, female breast and prostate.

[Epidemiology of primary liver cancer in Japan].

In Japan, more than 90% of primary liver cancers consist of hepatocellular carcinoma (HCC), 80% of which is caused by chronic hepatitis C virus (HCV) infection, and the remaining 15% of which is caused by chronic hepatitis B virus (HBV) infection. The proportion of older patients among patients with HCC has been increasing in recent years because of the aging of the HCV-prevalent birth cohort born between 1925 and 1935. The cumulative risk of developing HCC among HCV carriers was estimated as 30% for males and 6% for females. Older age, being male, having a low platelet count, higher histological stage, genotype 1b, co-infection with HBV, heavy drinking and smoking increase the risk of developing HCC among patients with chronic HCV infection. Recent reports on the efficacy of interferon therapy on the incidence of HCC in Japanese patients with chronic hepatitis C demonstrate the importance of providing a screening system for chronic HCV infection and establishing a medical referral system so that patients undergo the appropriate therapy for the Japanese HCV carriers.

[Trends of lung cancer incidence and its prognosis in Osaka, Japan].

Lung cancer incidence figures in Japan were estimated as 30,000 for males and 11,000 for females for 1989 and represented the second and the fifth leading site of cancer, respectively, according to the Research Group for Population-based Cancer Registration in Japan. It is also estimated that lung cancer will steadily increase in the future. In relative frequencies of the major histological types, squamous cell carcinoma showed a decreasing trend, while small as well as large cell carcinomas showed an increasing trend in Osaka. In the distribution of clinical staging, the proportion of localized cases accounted for only 20% of the lung cancers in 1990. Five-year relative survival rates for lung cancer were 11.3% for all patients and 43.0% for localized cases in Osaka in 1984-1986. The rates were reported as 13.5% and 36.6% for total patients and localized cases, respectively, among caucasians from the SEER Program, a NCI project including 10 population-based cancer registries in the US. No noticeable difference was observed between Japan and the US. To control lung cancer in Japan it is considered vitally important and urgent to develop (1) new examination methods for diagnosing lung cancer in the earlier stage, (2) an efficient and effective cessation campaign for cigarettes smoking and (3) non-smoking based education.

[Descriptive and analytical epidemiology of second primaries in Osaka, Japan].

With the improvement in survival of cancer patients, the incidence of second primaries has been increasing. Data from the Osaka Cancer Registry showed that the incidence of metachronous second primaries was associated with gender (male), age and calendar year at diagnosis of the first cancer. The 10-year cumulative risk was estimated at around 10% for those who developed their first cancer in their sixties in 1978-83. The observed number (O) of second primaries (including synchronous) was compared with the expected number (E). The O/E ratios among those who developed their first cancer at ages 0-14 and 15-29 years old were much higher than the ratios among all age groups. Patients who had developed cancer of the colon, larynx, lung, bladder, or breast (1978-86) showed a significantly higher than expected risk of developing second primaries during the 1-4 years after diagnosis of the first cancer. Based on the hospital cancer registry data from Osaka Medical Center for Cancer and Cardiovascular Diseases, associations between adjuvant chemo-immunotherapy and the risk of second primaries were examined among 1,925 gastric cancer patients who underwent curative gastrectomy. The sex-, age-, and stage-adjusted hazard rate ratio of second primaries was 1.04 for patients who underwent chemotherapy and 0.70 for patients who underwent chemo-immunotherapy, when compared with the risk for patients who did not receive adjuvant chemo-immunotherapy. Some chemotherapeutic agents appeared to increase the risk of second primaries. Second primaries among 2,824 breast cancer patients were examined and their associations with adjuvant chemo-immuno-radiotherapy were analyzed. The O/E ratio for cancers of all sites was 1.28, significantly higher than 1.0. Cancer of the stomach, colon, lung and ovary were frequently observed as a second primary among them. Among 117 patients who developed second primaries, 4 developed cancer of the corpus uteri. This corresponded to 1.89 times the expected, however, only one of the 4 patients underwent tamoxifen treatment. The O/E ratio for non-Hodgkin's lymphoma was 3.40, significantly higher than 1.0. These results suggest associations between the risk for non-Hodgkin's lymphoma and chemotherapy.

Descriptive epidemiology of male breast cancer in Osaka, Japan.

Male breast cancer is rare. The total number of incidence in Osaka for the period of 1966-95 was 182. Male-to-female ratio for breast cancer incidence was 1:164 in Osaka during this period. Mean age of the male breast cancer incidence was 63.3. The numbers of incidence and the crude incidence rates for male breast cancer have increased during the last 3 decades, while the age-standardized rates have remained constant. The age-specific incidence rates for males showed a gradual increase with age, while those for females showed a steep increase beginning at twenty years of age and a peak around 45-49 or 50-54 years old. The age-standardized incidence rates of male breast cancer were lower in Japan than in European countries and North America, as were those of female breast cancer. Distributions of the histological type and the extent of disease were not significantly different between males and females. Relative 5-year survival for the male breast cancer was, however, lower than that for the female, especially in the "regional" stage and "distant" stage. Further studies on the sex-difference in survival will be mandatory based on high-quality hospital cancer registries' data, which provide detailed information on the clinical stage and treatment.

Prospective study of "early" gastric cancer.

In order to elucidate the natural history of early gastric cancer, we followed up non-concurrently certain patients who had been diagnosed endoscopically as having "early' gastric cancer and who had histological evidence of cancer by biopsy, but on whom surgical resection had been delayed or had not been conducted. At the Center for Adult Diseases, Osaka, 56 cases were eligible for this study. Out of these, 27 cases progressed to advanced cancer and 16 remained at the early stage during the follow-up period (6-88 months, mean: 29 months). The remaining 13 cases had had neither surgical resection nor examinations. The survivorship function for the duration of "early" gastric cancer was estimated by the life-table method of Kaplan and Meier. The median of the duration was estimated as 37 months. The 5-year survival rate of the 34 cases on whom surgical resection had not been conducted was estimated as 64.5%. These results suggest that early diagnosis and early treatment may lead to a reduction of mortality from stomach cancer.

Case-control study of hepatocellular carcinoma among Koreans living in Osaka, Japan.

Mortality rates from liver cancer among Koreans living in Osaka are 2-3 times higher than those among Japanese. Our previous study revealed that chronic hepatitis B virus (HBV) infection and excessive alcohol drinking are two major risk factors for hepatocellular carcinoma (HCC) among Koreans in Osaka, although more than 70% of the HCC cases were negative for hepatitis B surface antigen (HBsAg). Using a recently developed immunoassay for detecting serum hepatitis C virus antibody (HCV-Ab), the role of HCV infection was evaluated in a case-control study. The case group consisted of 90 Korean patients who were admitted to Kyowa Hospital in Osaka, and were newly diagnosed as HCC during the period from January 1989 to December 1992. The control group consisted of 249 Korean patients admitted to Kyowa Hospital during the same period and matched in age groups to the HCC cases. Seventy-four and 16.7% of cases were positive for HCV-Ab and age groups to the HCC cases. Seventy-four and 16.7% of cases were positive for HCV-Ab and HBsAg, respectively. Besides, 41.1% of cases were heavy drinkers. Multiple logistic regression analysis revealed that the adjusted relative risk was 92.4 for HCV-Ab positive and 58.2 for HBsAg positive, as compared with both HCV-Ab and HBsAg negative. Elevated risk was also demonstrated for males with a history of heavy drinking. There was no significant association between the risk of HCC and a history of blood transfusion or cigarette smoking. It was concluded that chronic HCV infection plays a major role in the etiology of HCC among Koreans living in Osaka, in addition to HBV and heavy drinking.

Effects of mass screening for neuroblastoma on incidence, mortality, and survival rates in Osaka, Japan.

OBJECTIVES: To evaluate the effects of mass screening for neuroblastoma, time trends of incidence, mortality, and survival of neuroblastoma in Osaka Prefecture were analyzed. METHODS: Data for this analysis was obtained from the population-based Osaka Cancer Registry. Time trends of incidence and mortality rates were analyzed by calendar year and by birth cohort. Survival was compared between before and after the introduction of systematic screening. RESULTS: From 1970-94, 457 cases of neuroblastoma and 182 deaths from neuroblastoma were observed in Osaka. The annual age-standardized incidence rate per million children increased from 7.5 in 1970-84 to 20.5 in 1985-94, while the mortality rates did not differ between these two periods. Analysis by birth cohort showed that the incidence rate at 0 year of age per 100,000 live births increased from 2.30 in 1970-79 (unscreened) to 19.80 in 1988-89 (screening by high-performance liquid chromatography, HPLC). The incidence rate in children 1 and 2-4 years of age also increased according to the introduction of HPLC. The mortality rate in children 1-4 years of age per 100,000 live births slightly decreased from 3.87 in 1970-79 to 3.30 in 1988-89, which was presumed to be derived from the improvement in survival due to the progress in treatment. CONCLUSIONS: It is strongly suggested that mass screening for neuroblastoma causes harm because of overdiagnosis, and it has little effect on decreasing the incidence and the mortality of neuroblastoma at 1-4 years of age.