Accepted Article
Epidemiological Risk Factors for Adrenocortical Carcinoma: A Hospital-based Case- Control Study
Mouhammed Amir Habra 1, Mohamad Anas Sukkari 1, Ansam Hasan1, Youssef Albousen ”, Mohamed A. Elsheshtawi 1, Camilo Jimenez 1, Matthew Campbell 2, Jose A. Karam 3, Paul H. Graham4, Rikita I. Hatia5, Alexandria T. Phan 6, Jeena Varghese 1, Manal M. Hassan
From the Departments of Endocrine Neoplasia and Hormonal Disorders ”, Genitourinary Medical Oncology 2, Urology 3, Surgical Oncology 4, and Epidemiology 5. The University of Texas MD Anderson Cancer Center, Houston, Texas, United States of America. Department of Hematology-Oncology, The University of Texas Health Science Center at Tyler, Texas, United States of America 6.
Short Title: Risk factors for Adrenocortical Carcinoma
Corresponding Author: Mouhammed Amir Habra, MD, Department of Endocrine Neoplasia and Hormonal Disorders, Unit 1461, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, Texas 77030, USA, Email: mahabra@mdanderson.org;
Phone: (713) 792-2841; Fax: (713) 794-4065
Conflicts of Interest
The authors declare no conflicts of interest with this manuscript.
Other financial disclosures:
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/ijc.32534
Accepted Article
· Mouhammed Amir Habra:
o Exelixis: Research funding (to institution) ☐
o HRA Pharma: Research Support (to institution), Honorarium/advisory board ☐
· Jose Karam:
o Consulting/Advisory Board/Honorarium: Pfizer, Roche/Genentech, Novartis, EMD Serono ☐
o Research funding ( to institution): Roche/Genentech, Merck, Mirati ☐
Funding Sources
This study is supported in part by the NIH/NCI under award number P30CA016672 and The Beverlin Fund for Adrenal Cancer Research.
Key Words: smoking, alcohol use, adrenocortical carcinoma, family history
Abbreviation: ACC, adrenocortical carcinoma; AOR, Adjusted odds ratio; CI, confidence interval, BMI: body mass index.
What’s New: Findings from a retrospective study of a large cohort demonstrated an increased risk of adrenocortical carcinoma (ACC) in male smokers and in male and female individuals with a family history of malignancy. A history of alcohol use was associated with reduced risk of
ACC in men. Understanding these risk factors and their underlying mechanisms may help prevent ACC in susceptible individuals and eventually identify new therapeutic options for ACC. Article Category: Cancer Epidemiology
Accepted Article
Accepted Article
ABSTRACT
Adrenocortical carcinoma (ACC) is a rare malignancy whose risk factors are unclear. We explored the association of ACC risk with exposure to selected environmental factors, with a focus on cigarette smoking. We conducted a hospital-based case-control study at The University of Texas MD Anderson Cancer Center. Cases (N=432) patients with ACC treated at MD Anderson, and controls (N=1204) were healthy and genetically unrelated spouses of patients at MD Anderson who had cancers not associated with smoking. Information on the subjects’ demographic features and selected risk factors was collected using a structured, validated questionnaire and medical records review. Unconditional logistic regression was used to calculate adjusted odds ratios (AORs) via the maximum-likelihood method. Cases had a younger mean (± standard deviation) age than did controls (47.0± 0.7 and 60.0 ±0.3 years, respectively), and the majority of cases were female (60.6%) and non-Hispanic white (82.4%). We found a markedly increased risk of ACC among male cigarette smokers, with an AOR=1.8 (95% Confidence Interval [CI] =1.2-2.9), but not among female smokers (AOR=1.1, 95% CI=0.7-1.6). Family history of cancer was associated with increased risk of ACC (AOR=2.8, 95% CI 1.9-4.3) and in both men and women, whereas alcohol consumption was associated with reduced risk in men (AOR=0.2, 95% CI=0.1-0.3) but not women (AOR=0.7, 95% CI=0.5-1.1). Understanding these risk factors and their underlying mechanisms may help prevent ACC in susceptible individuals and eventually identify new therapeutic options for ACC.
Accepted Article
INTRODUCTION
Adrenocortical carcinoma (ACC) is an “orphan” malignancy with an annual incidence rate of 0.7-2 cases per million individuals.1,2 Despite its rarity, ACC poses a clinical challenge. It is a complex malignancy with a heterogeneous presentation and poor prognosis as patients have a high risk of recurrence and low cure rates. ACC is slightly more common in women than in men, and its peak diagnosis is in the fourth and fifth decades of life 3-5. The overwhelming majority of ACC cases are sporadic, and less than 10% of ACC cases in adults can be seen as part of hereditary syndromes such as Li-Fraumeni syndrome, Lynch syndrome, and multiple endocrine neoplasia type 1.º Because of the rarity of ACC, little is known about other potential risk factors.
Limited retrospective data published 2 decades ago suggested an association between ACC risk and smoking in men, but the association between ACC and alcohol consumption was not established. 7,8 A subset of ACC cases carried the genomic signature seen in smokers, but the mechanisms of this association remain unclear. 9 To clarify this question, we investigated the association between multiple environmental factors with ACC risk in a large single-institution database, with the primary aim of characterizing the association between smoking and ACC risk. We hypothesized that a potential association between ACC risk and environmental factors such smoking and alcohol use. Our findings will guide future exploration of the mechanisms behind this potential association, which may help prevent ACC in susceptible individuals and eventually result in new treatments in selected ACC patients.
Accepted Article
PATIENTS AND METHODS
Study Design and Population
In this hospital-based case-control study, ACC patients (cases) and control subjects were recruited at The University of Texas MD Anderson Cancer Center. The study was approved by the MD Anderson Institutional Review Board. De-identified data can be made available upon reasonable request while following current regulations and ethical standards to protect patients’ privacy.
Cases were identified retrospectively through a detailed review of the medical records of patients with pathologically confirmed ACC referred to MD Anderson from 1998 to 2016 for treatment at the institution’s clinics. At their initial visit to MD Anderson, all patients gave written informed consent to participate in epidemiological studies by responding to a self- administered questionnaire requesting information about demographic characteristics (age, sex, residence, marital status, and education), past history of risk factors for cancer (e.g., personal smoking history, alcohol use, medical history [body mass index (BMI), type 2 diabetes mellitus, personal history of malignancies]), and family history of cancer. Completed questionnaires were electronically documented in the patients’ medical records. Data were extracted from the epidemiological questionnaire results or documentation of risk factors in patients’ medical records.
Control subjects were spouses of other MD Anderson patients with cancers other than gastrointestinal, lung, or head and neck cancer (smoking-related cancers) who agreed to
Accepted Article
participate in this survey. All controls reported that they had no prior history of cancer at the time of enrollment and completed the self-administered questionnaire given to the ACC patients.
Cigarette smokers were defined as subjects who had smoked ≥100 cigarettes during their lifetimes. Smokers were asked to provide the average number of cigarettes they smoked per day, the age at which they began smoking, and the duration of smoking. Former smokers were questioned about the age at which they stopped smoking. Number of pack-years smoking were calculated by multiplying the number of years of smoking by the number of packs of cigarettes smoked per day (1 pack-year = 1 pack of cigarettes/day for 1 year). Heavy smokers were defined as those with more than 20 pack-years of smoking. Also, alcohol drinkers were defined as subjects who had consumed at least four alcoholic drinks of beer, wine, or hard liquor each month for 6 months during their lifetimes. Finally, information about cancer history in first- and second-degree relatives was collected from both cases and controls.
Statistical Methods
The Access software program (Microsoft, Redmond, WA) was used for data entry and management, and Stata software (version 13; StataCorp, College Station, TX) was used for statistical analysis. Chi-square tests were used to determine the significance of differences in the distributions of categorical variables between the cases and control subjects. P values <0.05 were considered significant. Univariate single-factor, unconditional logistic regression analyses were performed to assess the marginal effects of cigarette smoking on ACC risk using maximum-
Accepted Article
likelihood estimation. Also, multivariate unconditional logistic regression analyses were performed taking into account variables that could confound the risk of smoking, using those with P < 0.1 in the single-factor analyses. Each factor examined (age, sex, race/ethnicity, education level, history of type 2 diabetes mellitus, marital status, place of residence, alcohol use, and family history of cancers) was analyzed using the likelihood-ratio test. The final model was chosen on the basis of biological plausibility and lowest -2 Log-Likelihood function.
RESULTS
We analyzed a total of 1636 subjects (432 cases and 1,204 controls). The majority of cases were female (60.6%) and non-Hispanic white (82.4%). The cases were younger than the control subjects, with a mean difference (± standard deviation) in age of 12.2 ± 0.7 years (47.0±0.7 years and 60.0 ± 0.3 years in cases and controls, respectively). The cases had no significant differences in education level or state of referral from the controls. The majority of the cases and controls were Texas residents. In the multivariate logistic regression analysis, we adjusted for all significant demographic variables.
Forty ACC patients (9%) had a history of other malignancies prior to ACC diagnosis. They included two patients with three prior malignancies (testicular carcinoma, prostate carcinoma, and nonmelanoma skin cancer in one patient and Hodgkin lymphoma, differentiated thyroid carcinoma, and breast cancer in the other) and one patient with two prior malignancies (nonmelanoma skin cancer and prostate carcinoma). All of the remaining 37 patients had one
Accepted Article
prior malignancy (eight breast carcinomas, seven nonmelanoma skin cancers, six prostate carcinomas, three endometrial carcinomas, three melanomas, two differentiated thyroid carcinomas, two renal cell carcinomas, two cervical carcinomas, one lung carcinoma, one bladder carcinoma, one sarcoma, and one urothelial carcinoma).
Fourteen patients (3%) had development of other malignancies after ACC diagnosis. This included one patient with two malignancies (sarcoma and breast carcinoma) who had Li- Fraumeni syndrome. The remaining 13 patients had one subsequent malignancy (two with differentiated thyroid carcinoma, two with nonmelanoma skin cancer, and one each with vulvar squamous cell carcinoma, appendiceal adenocarcinoma, breast cancer, prostate cancer, endometrial cancer, thymic carcinoma, renal cell carcinoma, ovarian carcinoma, and colon cancer as part of Lynch syndrome).
Table 1 shows that cigarette smoking was a significant risk factor for ACC, with a 50% greater risk in smokers than in nonsmokers. When we analyzed the data according to sex, this relationship was statistically significant in men but not women. Smokers among the cases and control subjects started smoking at similar ages (mean [± standard error] age, 18.0 ± 0.5 years in cases and 18.0 ± 0.2 years in controls; p=0.1). In addition, the overall duration of smoking and number of cigarettes smoked per day were similar in the cases and controls. In comparing nonsmokers and smokers, the only significant trend in risk of ACC according to pack-years of smoking was in male subjects. The median number of pack-years was slightly higher for the cases (23.3 [range, 0.1-140.0]) than for the controls (20.0 [range, 0.1-145.0]) after controlling for
Accepted Article
education level, age, and race (p-value: 0.06), and the estimated odds ratios (95% confidence intervals) in men were 1.4 (0.8-2.7) for fewer than 20 pack years of smoking and 1.3 (0.7-2.4) for at least 20 pack-years of smoking as compared with those in nonsmokers after adjustment for age, race, alcohol use, and family history of cancer. Restricting the analysis to non-Hispanic white subjects and age matched in men did not meaningfully change the observed significant association between cigarette smoking and ACC risk.
In contrast with smoking, alcohol use was associated with a reduced risk of ACC in men (AOR=0.2, 95% CI=0.1-0.3) while this association in women was not significant (AOR=0.7, 95% CI=0.5-1.1). A family history of cancer was a significant risk factor for ACC in all patients overall as well as the male and female patients separately (Table 1). Four patients (1%) reported family history of malignancies that could be associated with Li-Fraumeni syndrome (brain tumors and leukemia) and none reported family history of sarcoma.
BMI was not significantly associated with risk of ACC, as the estimated odds ratio (95% confidence interval) for BMI for this association was 0.98 (0.94-1.03). Inclusion and exclusion of BMI in the final logistic regression model did not result in meaningful differences in the findings.
DISCUSSION
The results of this large study suggest that an increased risk of ACC is associated with cigarette smoking in men. Also, we observed a greater risk of ACC in men and women with family
Accepted Article
histories of cancer than in those without them. In contrast, we observed an 80% reduction in ACC risk among male drinkers.
Our findings of increased risk of ACC in male smokers are in line with reports from 2 decades ago that raised suspicion about a link between tobacco use and ACC. In a case-control study, Hsing et al. sent a questionnaire about smoking, oral contraceptives use, and alcohol use to the next of kin of about 20,000 dead adults. They included 176 individuals who died of ACC and 352 control subjects who died of causes unrelated to smoking, oral contraceptive use, or drinking. The ACC risk was increased in men who were heavy smokers but not in women, and the researchers found no association of ACC risk with alcohol use or body habitus 8. In contrast to this earlier study, we found reduced risk of ACC in association with alcohol use. In contrast with that study, we found a reduced risk of ACC in association with alcohol use in men. Furthermore, Hsing and colleagues found an increased risk of ACC in association with oral contraceptive use, especially in women younger than 25 years. We had limited data about the use of estrogen products in our series. This was a limitation of our study, and estrogen use should be documented in future prospective studies of environmental exposures in ACC.
Cancer develops in only a minority of smokers, and a genetic predisposition likely increases the risk of malignant tumors in these individuals.1º Other environmental exposures, such as alcohol use, are associated with increased risk of cancers of the oral cavity, pharynx, larynx, esophagus, liver, colon/rectum, pancreas, stomach, and breast. The risk of kidney cancer and non-Hodgkin lymphoma is lower in association with alcohol use. 11-16
Accepted Article
Aside from genetic predisposition in a minority of adult ACC patients, no risk factors for ACC are firmly established. ACC can be associated with a few hereditary cancer syndromes, such as Lynch syndrome, Li-Fraumeni syndrome, multiple endocrine neoplasia type 1, familial adenomatous polyposis, and Beckwith-Wiedemann syndrome. However, studies have revealed that these syndromes cannot be viewed as the only drivers of malignant adrenocortical tumorigenesis, raising the likelihood of interaction of ACC with environmental factors.17
Although knowledge of them is increasing, considerable gaps in understanding of the basic mechanisms of tobacco related malignancies remain. The fact that a minority of smokers get lung cancer suggests that genetic factors are involved in smoking-induced carcinogenesis. Many studies of human smoking-related cancers, such as lung, bladder, and head and neck cancers, have suggested that factors including genetic variations in carcinogen metabolism, cell- cycle regulation, and DNA repair determine susceptibility to tobacco carcinogenesis.10,18
Under oxidative stress, various defense mechanisms may evolve to counteract it. Plasma antioxidants such as bilirubin, ascorbic acid, uric acid, and a- and y-tocopherols are considered the first line of defense against exogenous oxidative stress because they scavenge reactive oxygen species prior to their interaction with cellular components. The second line of defense against exogenous and endogenous oxidative stress is provided by several antioxidant enzymes as well as the phase II enzymes glutathione S-transferases.19,20 All of these enzymes work to reduce the reactive oxygen species to less reactive components and to excrete the end products. However, if reactive oxygen species escape these defense lines and DNA damage occurs, the
Accepted Article
body’s defense against oxidative stressors shifts to the DNA repair system. Thus, we hypothesize that deficient antioxidant and DNA repair mechanisms increase an individual’s susceptibility to oxidative stress-induced DNA damage and cancer, including ACC. Our findings in the present case-control study add to the importance of understanding the underlying mechanisms that link tobacco use with ACC development. In the meantime, minimizing or eliminating smoking seems to be a prudent approach to prevention of ACC.
The relationship of alcohol consumption with cancer incidence, morbidity, and mortality is complex and likely influenced by multiple genetic and environmental factors. Alcohol use is associated with increased risk of multiple malignancies (breast, colorectal, esophageal, hepatocellular, and head and neck cancer) but not with that of others (neuroendocrine tumors),2” and it has been associated with reduced risk of multiple myeloma and ACC in men based on our data.22,23 A clear mechanism explaining the difference in the association of alcohol consumption with ACC risk between men and women is lacking. This difference may be related to genetic, metabolic, and hormonal factors causing lower activity of gastric alcohol dehydrogenase and to slower hepatic alcohol metabolism in women.24,25 We did not have sufficient data to establish a dose-response relationship between alcohol use and ACC risk or explain the different effects of alcohol use on ACC risk according to sex.
In our study, BMI was not significantly associated with risk of ACC, and BMI at the time of cancer diagnosis is well accepted to potentially result in misclassification bias, as the disease may affect patients’ weight, which usually differs from the weight of healthy individuals.
Accepted Article
Further research is needed to validate our findings in different populations as well as identify the underlying mechanisms and genetic predisposition factors that contribute to ACC development in men and women separately.
The strengths of this study include the large number of ACC cases included despite the rarity of this disease. In addition, we used a detailed, well-validated questionnaire to collect information on the cases and controls. Furthermore, we carefully adjusted for potentially confounding factors, such as race, alcohol use, and family history of cancer. Our study reflects the known limitations of a retrospective design as well as referral bias, as the cases were seen in a major referral center. Other limitations include a lack of information on potential changes in environmental exposures over time, especially smoking and alcohol use. Given the retrospective nature of our study, we speculate that a family history of malignancies is more prone to recollection bias as well as uncertainty about the exact cancer diagnosis in family members than are other parts of the medical history. Two patients had hereditary cancer syndromes associated with ACC one with Li-Fraumeni syndrome and the other with Lynch syndrome and four patients reported a family history of malignancies suspicious for Li-Fraumeni syndrome.
In conclusion, our findings support an association between ACC risk and tobacco use, especially in men, and an association between alcohol use and reduced ACC risk in men only. Also, a family history of cancer was a significant risk factor for ACC in all patients. Exploring
Accepted Article
the underlying mechanisms of these associations is needed to further our understanding of ACC development and potentially identify new therapeutic options for this rare malignancy.
Acknowledgements
We thank Ms. Sunita Patterson and Mr. Donald R. Norwood from MD Anderson Research Medical Library Scientific Publication Services for their editorial support.
References
1. Kebebew E, Reiff E, Duh QY, Clark OH, McMillan A. Extent of disease at presentation and outcome for adrenocortical carcinoma: have we made progress? World J Surg. 2006;30(5):872- 878.
2. Kerkhofs TM, Verhoeven RH, Van der Zwan JM, et al. Adrenocortical carcinoma: a population- based study on incidence and survival in the Netherlands since 1993. Eur J Cancer. 2013;49(11):2579-2586.
3. Else T, Kim AC, Sabolch A, et al. Adrenocortical carcinoma. Endocr Rev. 2014;35(2):282-326.
4. Fassnacht M, Kroiss M, Allolio B. Update in adrenocortical carcinoma. J Clin Endocrinol Metab. 2013;98(12):4551-4564.
5. Ayala-Ramirez M, Jasim S, Feng L, et al. Adrenocortical carcinoma: clinical outcomes and prognosis of 330 patients at a tertiary care center. Eur J Endocrinol. 2013;169(6):891-899.
6. Else T, Rodriguez-Galindo C. 5th International ACC Symposium: Hereditary Predisposition to Childhood ACC and the Associated Molecular Phenotype. Horm Cancer-Us. 2016;7(1):36-39.
7. Chow WH, Hsing AW, Mclaughlin JK, Fraumeni JF, Jr. Smoking and adrenal cancer mortality among United States veterans. Cancer Epidemiol Biomarkers Prev. 1996;5(2):79-80.
8. Hsing AW, Nam JM, Co Chien HT, Mclaughlin JK, Fraumeni JF, Jr. Risk factors for adrenal cancer: an exploratory study. Int J Cancer. 1996;65(4):432-436.
9. Zheng S, Cherniack AD, Dewal N, et al. Comprehensive Pan-Genomic Characterization of Adrenocortical Carcinoma. Cancer Cell. 2016;29(5):723-736.
Accepted Article
10. Wu X, Zhao H, Suk R, Christiani DC. Genetic susceptibility to tobacco-related cancer. Oncogene. 2004;23(38):6500-6523.
11. Connor J. Alcohol consumption as a cause of cancer. Addiction. 2017;112(2):222-228.
12. Gupta S, Wang F, Holly EA, Bracci PM. Risk of pancreatic cancer by alcohol dose, duration, and pattern of consumption, including binge drinking: a population-based study. Cancer Causes Control. 2010;21(7):1047-1059.
13. Ma K, Baloch Z, He TT, Xia X. Alcohol Consumption and Gastric Cancer Risk: A Meta- Analysis. Med Sci Monit. 2017;23:238-246.
14. Mahabir S, Leitzmann MF, Virtanen MJ, et al. Prospective study of alcohol drinking and renal cell cancer risk in a cohort of Finnish male smokers. Cancer Epidem Biomar. 2005;14(1):170- 175.
15. Lee JE, Hunter DJ, Spiegelman D, et al. Alcohol intake and renal cell cancer in a pooled analysis of 12 prospective studies. J Natl Cancer Inst. 2007;99(10):801-810.
16. Psaltopoulou T, Sergentanis TN, Ntanasis-Stathopoulos I, Tzanninis IG, Tsilimigras DI, Dimopoulos MA. Alcohol consumption and risk of hematological malignancies: A meta-analysis of prospective studies. International Journal of Cancer. 2018;143(3):486-+.
17. Assie G, Jouinot A, Bertherat J. The ‘omics’ of adrenocortical tumours for personalized medicine. Nat Rev Endocrinol. 2014;10(4):215-228.
18. Lee HW, Wang HT, Weng MW, et al. Cigarette side-stream smoke lung and bladder carcinogenesis: inducing mutagenic acrolein-DNA adducts, inhibiting DNA repair and enhancing anchorage-independent-growth cell transformation. Oncotarget. 2015;6(32):33226-33236.
19. Singhal SS, Singh SP, Singhal P, Horne D, Singhal J, Awasthi S. Antioxidant role of glutathione S-transferases: 4-Hydroxynonenal, a key molecule in stress-mediated signaling. Toxicol Appl Pharmacol. 2015;289(3):361-370.
20. Deger Y, Yur F, Ertekin A, Mert N, Dede S, Mert H. Protective effect of alpha-tocopherol on oxidative stress in experimental pulmonary fibrosis in rats. Cell Biochem Funct. 2007;25(6):633- 637.
21. Hassan MM, Phan A, Li D, Dagohoy CG, Leary C, Yao JC. Risk factors associated with neuroendocrine tumors: A US-based case-control study. International Journal of Cancer. 2008;123(4):867-873.
22. Kroll ME, Murphy F, Pirie K, et al. Alcohol drinking, tobacco smoking and subtypes of haematological malignancy in the UK Million Women Study. Br J Cancer. 2012;107(5):879-887.
23. Santo L, Liao LM, Andreotti G, Purdue MP, Hofmann JN. Alcohol consumption and risk of multiple myeloma in the NIH-AARP Diet and Health Study. Int J Cancer. 2019;144(1):43-48.
24. Baraona E, Abittan CS, Dohmen K, et al. Gender differences in pharmacokinetics of alcohol. Alcohol Clin Exp Res. 2001;25(4):502-507.
25. Mancinelli R, Binetti R, Ceccanti M. Woman, alcohol and environment: Emerging risks for health. Neurosci Biobehav Rev. 2007;31(2):246-253.
Accepted Article
| Variable | Cases N=432 No. (%) | Controls N=1204 No. (%) | All Subjects OR (95% CI) | Men OR (95% CI) | Women OR (95% CI) |
|---|---|---|---|---|---|
| Sex | |||||
| Male | 170 (39.4) | 709 (58.9) | 1 (reference) | ||
| Female | 262 (60.6) | 495 (41.1) | 0.6 (0.5-0.9) | ||
| Race | |||||
| Non-white | 76 (17.6) | 107 (8.9) | 1 (reference) | 1 (reference) | 1 (reference) |
| White | 356 (82.4) | 1097 (91.1) | 0.7 (0.4-1.1) | 0.7 (0.3-1.2) | 0.7 (0.4-1.3) |
| Ever smoker | |||||
| No | 231 (54.4) | 621 (52.3) | 1 (reference) | 1 (reference) | 1 (reference) |
| Yes | 194 (45.6) | 567 (47.7) | 1.5 (1.1-2.1) | 1.8 (1.2-2.9) | 1.1 (0.7-1.6) |
| Ever alcohol use | |||||
| No | 203 (65.1) | 522 (43.9) | 1 (reference) | 11 (reference) | 1 (reference) |
| Yes | 109 (34.9) | 666 (56.1) | 0.4 (0.3-0.6) | 0.2 (0.1-0.3) | 0.7 (0.5-1.1) |
| Type II diabetes | |||||
| No | 393 (91) | 1104 (91.7) | 1 (reference) | 1 (reference) | 1 (reference) |
| Yes | 39 (9) | 100 (8.3) ) | 0.9 (.5-1.6) | 0.7 (0.3-1.6) | 1.1 (0.5-2.3) |
| Family history of cancer | |||||
| No | 36 (12.8) | 367 (30.5) | 1 (reference) | 1 (reference) | 1 (reference) |
| Yes | 245 (87.2) | 837 (69.5) | 2.8 (1.9-4.3) | 3 (1.6-5.6) | 2.8 (1.6-4.9) |
OR, odds ratio; CI, confidence interval. Bold signifies statistically significant differences between patients and controls (p<0.05).
* Missing from patients: smoking (N= 7), alcohol (N=120), and family history (N=151) and from controls smoking (N= 16) and alcohol (N=16).