Eero Pukkala

Environmental and Heritable Factors in the Causation of Cancer — Analyses of Cohorts of Twins from Sweden, Denmark, and Finland

BACKGROUND The contribution of hereditary factors to the causation of sporadic cancer is unclear. Studies of twins make it possible to estimate the overall contribution of inherited genes to the development of malignant diseases. METHODS We combined data on 44,788 pairs of twins listed in the Swedish, Danish, and Finnish twin registries in order to assess the risks of cancer at 28 anatomical sites for the twins of persons with cancer. Statistical modeling was used to estimate the relative importance of heritable and environmental factors in causing cancer at 11 of those sites. RESULTS At least one cancer occurred in 10,803 persons among 9512 pairs of twins. An increased risk was found among the twins of affected persons for stomach, colorectal, lung, breast, and prostate cancer. Statistically significant effects of heritable factors were observed for prostate cancer (42 percent; 95 percent confidence interval, 29 to 50 percent), colorectal cancer (35 percent; 95 percent confidence interval, 10 to 48 percent), and breast cancer (27 percent; 95 percent confidence interval, 4 to 41 percent). CONCLUSIONS Inherited genetic factors make a minor contribution to susceptibility to most types of neoplasms. This finding indicates that the environment has the principal role in causing sporadic cancer. The relatively large effect of heritability in cancer at a few sites suggests major gaps in our knowledge of the genetics of cancer.

Cancer risk in mutation carriers of DNA‐mismatch‐repair genes

Excessive incidence of various cancers is a challenging feature of the hereditary‐non‐polyposis‐colorectal‐cancer (HNPCC) syndrome. This study estimated the cancer incidences in HNPCC compared with the general population. Individuals in a cohort of 1763 members of 50 genetically diagnosed families were categorized according to their genetic status as mutation carriers, non‐carriers, or individuals at 50 or 25% risk of being a carrier. Incidences of cancers in these groups were compared with those in the Finnish population overall. In 360 mutation carriers, standardized incidence ratios (SIR) were significantly increased for colorectal [68; 95% confidence intervals (CI), 56 to 81], endometrial (62; 95% CI, 44 to 86), ovarian (13; 95% CI, 5.3 to 25), gastric (6.9; 95% CI, 3.6 to 12), biliary tract (9.1; 95% CI, 1.1 to 33), uro‐epithelial (7.6; 95% CI, 2.5 to 18) and kidney (4.7; 95% CI, 1 to 14) cancers and for central‐nervous‐system tumours (4.5; 95% CI, 1.2 to 12). The SIR increased with increasing likelihood of being a mutation carrier. The cumulative cancer incidences were 82, 60, 13 and 12% for colorectal, endometrial, gastric and ovarian cancers respectively. For other tumours associated with increased risk, corresponding incidences were below 4%. Interestingly, the incidence of endometrial cancer (60%) exceeded that for colorectal cancer in women (54%). The tumour spectrum associated with germline mutations of DNA‐mismatch‐repair genes involves 8 or more organ sites, suggesting a need to develop methods to screen for extra‐colonic cancer also. Int. J. Cancer 81:214–218, 1999.

Frequency of specific cancer types in dermatomyositis and polymyositis: a population-based study.

BACKGROUND Dermatomyositis and polymyositis are associated with cancer, but previous nationwide studies have not had sufficient cases to test the association between myositis and specific cancer types. Our aim was to investigate the risk of specific cancer types in individuals with dermatomyositis and polymyositis. METHODS We did a pooled analysis of published national data from Sweden, Denmark, and Finland. All patients with dermatomyositis and polymyositis (> or =15 years old) were identified by discharge diagnosis from the Swedish National Board of Health (1964-83), Danish Hospital Discharge Registry (1977-89), and Finnish National Board of Health (1969-85). Personal details were matched to national cancer registries, to identify all cases of cancer up to 1987 in Sweden, 1995 in Denmark, and 1997 in Finland, and to national death registries for the same periods. We calculated standardised incidence ratios (SIR) for individual cancer sites for dermatomyositis and polymyositis separately, using national cancer rates by country, sex, age, and date. FINDINGS We identified 618 cases of dermatomyositis, of whom 198 had cancer. 115 of the 198 developed cancer after diagnosis of dermatomyositis. This disease was strongly associated with malignant disease (SIR 3.0, 95% CI 2.5-3.6), particularly ovarian (10.5, 6.1-18.1), lung (5.9, 3.7-9.2), pancreatic (3.8, 1.6-9.0), stomach (3.5, 1.7-7.3), and colorectal (2.5, 1.4-4.4) cancers, and non-Hodgkin lymphoma (3.6, 1.2-11.1). 137 of the 914 cases of polymyositis had cancer, which developed after diagnosis of polymyositis in 95. Polymyositis was associated with a raised risk of non-Hodgkin lymphoma (3.7, 1.7-8.2), and lung (2.8, 1.8-4.4) and bladder cancers (2.4, 1.3-4.7). In both dermatomyositis and polymyositis, risk of malignant disease was highest at time of myositis diagnosis. INTERPRETATION Our results provide evidence that dermatomyositis is strongly associated with a wide range of cancers. The overall risk of malignant disease is also modestly increased among patients with polymyositis, with an excess for some cancers.

Data quality and quality control of a population-based cancer registry. Experience in Finland.

Cancer registries should pay great attention to the quality of their data, both in terms of completeness (all cancer patients in the population are registered) and accuracy (data on individual cancer patients must be correct). In addition to technical measures in the data processing, different types of checks and comparisons should be routine practice. Active research policy and ambitious, research-oriented staff with competence in medicine, biostatistics and computer science are essential in terms of maintaining good data quality.

Hopelessness and Risk of Mortality and Incidence of Myocardial Infarction and Cancer

We examined the relationship among low, moderate, and high levels of hopelessness, all-cause and cause-specific mortality, and incidence of myocardial infarction (MI) and cancer in a population-based sample of middle-aged men. Participants were 2428 men, ages 42 to 60, from the Kuopio Ischemic Heart Disease study, an ongoing longitudinal study of unestablished psychosocial risk factors for ischemic heart disease and other outcomes. In 6 years of follow-up, 174 deaths (87 cardiovascular and 87 noncardiovascular, including 40 cancer deaths and 29 deaths due to violence or injury), 73 incident cancer cases, and 95 incident MI had occurred. Men were rated low, moderate, or high in hopelessness if they scored in the lower, middle or upper one-third of scores on a 2-item hopelessness scale. Age-adjusted Cox proportional hazards models identified a dose-response relationship such that moderately and highly hopeless men were at significantly increased risk of all-cause and cause-specific mortality relative to men with low hopelessness scores. Indeed, highly hopeless men were at more than three-fold increased risk of death from violence or injury compared with the reference group. These relationships were maintained after adjusting for biological, socioeconomic, or behavioral risk factors, perceived health, depression, prevalent disease, or social support. High hopelessness also predicted incident MI, and moderate hopelessness was associated with incident cancer. Our findings indicate that hopelessness is a strong predictor of adverse health outcomes, independent of depression and traditional risk factors. Additional research is needed to examine phenomena that lead to hopelessness.

Inherited susceptibility to uterine leiomyomas and renal cell cancer

Herein we report the clinical, histopathological, and molecular features of a cancer syndrome with predisposition to uterine leiomyomas and papillary renal cell carcinoma. The studied kindred included 11 family members with uterine leiomyomas and two with uterine leiomyosarcoma. Seven individuals had a history of cutaneous nodules, two of which were confirmed to be cutaneous leiomyomatosis. The four kidney cancer cases occurred in young (33- to 48-year-old) females and displayed a unique natural history. All these kidney cancers displayed a distinct papillary histology and presented as unilateral solitary lesions that had metastasized at the time of diagnosis. Genetic-marker analysis mapped the predisposition gene to chromosome 1q. Losses of the normal chromosome 1q were observed in tumors that had occurred in the kindred, including a uterine leiomyoma. Moreover, the observed histological features were used as a tool to diagnose a second kindred displaying the phenotype. We have shown that predisposition to uterine leiomyomas and papillary renal cell cancer can be inherited dominantly through the hereditary leiomyomatosis and renal cell cancer (HLRCC) gene. The HLRCC gene maps to chromosome 1q and is likely to be a tumor suppressor. Clinical, histopathological, and molecular tools are now available for accurate detection and diagnosis of this cancer syndrome.

Occupation and cancer ― follow-up of 15 million people in five Nordic countries

We present up to 45 years of cancer incidence data by occupational category for the Nordic populations. The study covers the 15 million people aged 30–64 years in the 1960, 1970, 1980/1981 and/or 1990 censuses in Denmark, Finland, Iceland, Norway and Sweden, and the 2.8 million incident cancer cases diagnosed in these people in a follow-up until about 2005. The study was undertaken as a cohort study with linkage of individual records based on the personal identity codes used in all the Nordic countries. In the censuses, information on occupation for each person was provided through free text in self-administered questionnaires. The data were centrally coded and computerised in the statistical offices. For the present study, the original occupational codes were reclassified into 53 occupational categories and one group of economically inactive persons. All Nordic countries have a nation-wide registration of incident cancer cases during the entire study period. For the present study the incident cancer cases were classified into 49 primary diagnostic categories. Some categories have been further divided according to sub-site or morphological type. The observed number of cancer cases in each group of persons defined by country, sex, age, period and occupation was compared with the expected number calculated from the stratum specific person years and the incidence rates for the national population. The result was presented as a standardised incidence ratio, SIR, defined as the observed number of cases divided by the expected number. For all cancers combined (excluding non-melanoma skin cancer), the study showed a wide variation among men from an SIR of 0.79 (95% confidence interval 0.66–0.95) in domestic assistants to 1.48 (1.43–1.54) in waiters. The occupations with the highest SIRs also included workers producing beverage and tobacco, seamen and chimney sweeps. Among women, the SIRs varied from 0.58 (0.37–0.87) in seafarers to 1.27 (1.19–1.35) in tobacco workers. Low SIRs were found for farmers, gardeners and teachers. Our study was able to repeat most of the confirmed associations between occupations and cancers. It is known that almost all mesotheliomas are associated with asbestos exposure. Accordingly, plumbers, seamen and mechanics were the occupations with the highest risk in the present study. Mesothelioma was the cancer type showing the largest relative differences between the occupations. Outdoor workers such as fishermen, gardeners and farmers had the highest risk of lip cancer, while the lowest risk was found among indoor workers such as physicians and artistic workers. Studies of nasal cancer have shown increased risks associated with exposure to wood dust, both for those in furniture making and for those exposed exclusively to soft wood like the majority of Nordic woodworkers. We observed an SIR of 1.84 (1.66–2.04) in male and 1.88 (0.90–3.46) in female woodworkers. For nasal adenocarcinoma, the SIR in males was as high as 5.50 (4.60–6.56). Male waiters and tobacco workers had the highest risk of lung cancer, probably attributable to active and passive smoking. Miners and quarry workers also had a high risk, which might be related to their exposure to silica dust and radon daughters. Among women, tobacco workers and engine operators had a more than fourfold risk as compared with the lung cancer risk among farmers, gardeners and teachers. The occupational risk patterns were quite similar in all main histological subtypes of lung cancer. Bladder cancer is considered as one of the cancer types most likely to be related to occupational carcinogens. Waiters had the highest risk of bladder cancer in men and tobacco workers in women, and the low-risk categories were the same ones as for lung cancer. All this can be accounted for by smoking. The second-highest SIRs were among chimney sweeps and hairdressers. Chimney sweeps are exposed to carcinogens such as polycyclic aromatic hydrocarbons from the chimney soot, and hairdressers’ work environment is also rich in chemical agents. Exposure to the known hepatocarcinogens, the Hepatitis B virus and aflatoxin, is rare in the Nordic countries, and a large proportion of primary liver cancers can therefore be attributed to alcohol consumption. The highest risks of liver cancer were seen in occupational categories with easy access to alcohol at the work place or with cultural traditions of high alcohol consumption, such as waiters, cooks, beverage workers, journalists and seamen. The risk of colon cancer has been related to sedentary work. The findings in the present study did not strongly indicate any protective role of physical activity. Colon cancer was one of the cancer types showing the smallest relative variation in incidence between occupational categories. The occupational variation in the risk of female breast cancer (the most common cancer type in the present series, 373 361 cases) was larger, and there was a tendency of physically demanding occupations to show SIRs below unity. Women in occupations which require a high level of education have, on average, a higher age at first child-birth and elevated breast cancer incidence. Women in occupational categories with the highest average number of children had markedly lower incidence. In male breast cancer (2 336 cases), which is not affected by the dominating reproductive factors, there was a suggestion of an increase in risk in occupations characterised by shift work. Night-shift work was recently classified as probably carcinogenic, with human evidence based on breast cancer research. The most common cancer among men in the present cohort was prostate cancer (339 973 cases). Despite the huge number of cases, we were unable to demonstrate any occupation-related risks. The observed small occupational variation could be easily explained by varying PSA test frequency. The Nordic countries are known for equity and free and equal access to health care for all citizens. The present study shows that the risk of cancer, even under these circumstances, is highly dependent on the persons position in the society. Direct occupational hazards seem to explain only a small percentage of the observed variation – but still a large number of cases – while indirect factors such as life style changes related to longer education and decreasing physical activity become more important. This publication is the first one from the extensive Nordic Occupational Cancer (NOCCA) project. Subsequent studies will focus on associations between specific work-related factors and cancer diseases with the aim to identify exposure-response patterns. In addition to the cancer data demonstrated in the present publication, the NOCCA project produced Nordic Job Exposure Matrix (described in separate articles in this issue of Acta Oncologica) that transforms information about occupational title histories to quantitative estimates of specific exposures. The third essential component is methodological development related to analysis and interpretation of results based on averaged information of exposures and co-factors in the occupational categories.

Second Cancers Among Long-Term Survivors of Hodgkin’s Disease Diagnosed in Childhood and Adolescence

PURPOSE To quantify the risk of second cancers among long-term survivors of Hodgkins disease (HD) diagnosed before 21 years of age and to explore sex-, age-, and site-related differences. PATIENTS AND METHODS We analyzed data from 5,925 pediatric HD patients, including 2,646 10-year and 755 20-year survivors, who were reported to 16 population-based cancer registries in North America and Europe between 1935 and 1994. RESULTS A total of 157 solid tumors (observed/expected ratio [O/E] = 7.0; 95% confidence interval [CI], 5.9 to 8.2.) and 26 acute leukemias (O/E = 27.4; 95% CI, 17.9 to 40. 2) were reported. Risk of solid tumors remained significantly increased among 20-year survivors (O/E = 6.6, observed [O] = 40, cumulative risk = 6.5%) and persisted for 25 years (O/E = 4.6, O = 15, cumulative risk = 11.7%). Temporal trends for cancers of thyroid, female breast, bone/connective tissue, stomach, and esophagus were consistent with the late effects of radiotherapy. Greater than 50-fold increased risks were observed for tumors of the thyroid and respiratory tract (one lung and one pleura) among children treated before age 10. At older ages (10 to 16 years), the largest number of second cancers occurred in the digestive tract (O/E = 19.3) and breast (O/E = 22.9). Risk of solid tumors increased with decreasing age at HD on a relative but not absolute scale. CONCLUSION Children and adolescents treated for HD experience significantly increased risks of second cancers at various sites for 2 to 3 decades. Although our results reflect the late effects of past therapeutic modalities, they underscore the importance of lifelong follow-up of pediatric HD patients given early, more aggressive treatments.

Early-Onset Renal Cell Carcinoma as a Novel Extraparaganglial Component of SDHB-Associated Heritable Paraganglioma

Hereditary paraganglioma syndrome has recently been shown to be caused by germline heterozygous mutations in three (SDHB, SDHC, and SDHD) of the four genes that encode mitochondrial succinate dehydrogenase. Extraparaganglial component neoplasias have never been previously documented. In a population-based registry of symptomatic presentations of phaeochromocytoma/paraganglioma comprising 352 registrants, among whom 16 unrelated registrants were SDHB mutation positive, one family with germline SDHB mutation c.847-50delTCTC had two members with renal cell carcinoma (RCC), of solid histology, at ages 24 and 26 years. Both also had paraganglioma. A registry of early-onset RCCs revealed a family comprising a son with clear-cell RCC and his mother with a cardiac tumor, both with the germline SDHB R27X mutation. The cardiac tumor proved to be a paraganglioma. All RCCs showed loss of the remaining wild-type allele. Our observations suggest that germline SDHB mutations can predispose to early-onset kidney cancers in addition to paragangliomas and carry implications for medical surveillance.

Cancer risk after metal on metal and polyethylene on metal total hip arthroplasty.

The incidence of cancer after metal on metal total hip arthroplasty (McKee-Farrar) and polyethylene on metal total hip arthroplasty (Brunswik, Lubinus) was compared with that of the general population in Finland. The mean followup time for the patients who had metal on metal total hip arthroplasty was 15.7 (9092 person years) and for the patients who had polyethylene on metal total hip arthroplasty it was 12.5 years (19,846 person years). One hundred thirteen malignant cancers were observed in patients who had metal on metal total hip arthroplasty and 212 were observed in patients who had polyethylene on metal total hip arthroplasty. The standardized incidence ratio for all cancers of the metal on metal arthroplasty group was 0.95 (95% confidence limits 0.79-1.13) and that of the polyethylene on metal arthroplasty group was 0.76 (95% confidence limits 0.68-0.86). The risk of total cancer in the patients who had metal on metal total hip arthroplasty was 1.23-fold compared with that of the patients who had polyethylene on metal total hip arthroplasty. Both groups had significantly less lung cancer than the general population: the leukemia incidence in the patients who had metal on metal total hip arthroplasty was slightly increased (observed to experienced 7/3.03, standardized incidence ratio 0.61; 95% confidence limits 0.17-1.56). The leukemia rate of the patients who had metal on metal total hip arthroplasty was 3.77-fold compared with that of the patients who had polyethylene on metal total hip arthroplasty, but this difference was not statistically significant. No sarcomas were observed at the site of the prosthesis. The incidence of the other forms of cancers did not differ significantly from those in the general population. The observed variation in the incidence of different cancers among patients who had total hip arthroplasty compared with the general population suggests that factors other than total hip arthroplasty play a major role in the origin of cancer.