Associations of Lifestyle and Genetic Risks with Obesity and Related Chronic Diseases in the UK Biobank: A Prospective Cohort Study.

BACKGROUND: Interplay between lifestyle risk scores (LRSs) and genetic risk scores (GRSs) on obesity and related chronic diseases are underinvestigated and necessary for understanding obesity causes and developing prevention strategies. OBJECTIVES: This study aimed to investigate independent and joint associations and interactions of LRS and GRS with obesity prevalence and risks of diabetes, cardiovascular disease (CVD), and obesity-related cancer. METHODS: In this cohort study of 444,957 UK Biobank participants [age: 56.5 ± 8.1 y; BMI (in kg/m2): 27.4 ± 4.7], LRS included physical activity, dietary score, sedentary behavior, sleep duration, and smoking (range: 0-20, each factor had 5 levels). GRS was calculated based on 941 genetic variants related to BMI. Both scores were categorized into quintiles. Obesity (n = 106,301) was defined as baseline BMI ≥30. Incident diabetes (n = 16,311), CVD (n = 18,076), and obesity-related cancer (n = 17,325) were ascertained through linkage to registries over a median of 12-y follow-up. RESULTS: The LRS and GRS were independently positively associated with all outcomes. Additive interactions of LRS and GRS were observed for all outcomes (P < 0.021). Comparing the top with bottom LRS quintile, prevalence differences (95% CIs) for obesity were 17.8% (15.9%, 19.7%) in the top GRS quintile and 10.7% (8.3%, 13.1%) in the bottom GRS quintile; for diabetes, CVD, and obesity-related cancer, incidence rate differences associated with per SD increase in LRS were greater in the top than that in the bottom GRS quintile. Participants from top quintiles of both LRS and GRS had 6.16-fold, 3.81-fold, 1.56-fold, and 1.44-fold higher odds/risks of obesity, diabetes, CVD, and obesity-related cancer, respectively, than those from bottom quintiles of both scores. CONCLUSIONS: Higher LRS was associated with higher obesity prevalence and risks of related chronic diseases regardless of GRS, highlighting the broad benefits of healthy lifestyles. Additive gene-lifestyle interactions emphasize the public health importance of lifestyle interventions among people with high genetic risks.

Circulating levels of biomarkers and risk of ductal carcinoma in situ of the breast in the UK Biobank study.

Observational studies have shown associations between circulating levels of various biomarkers (eg, total cholesterol [TC], low-density lipoprotein cholesterol [LDL], insulin-like growth factor-1 [IGF-1], C-reactive protein [CRP] and glycated hemoglobin-1c [HbA1c]) and the risk of invasive breast cancer (IBC). Ductal carcinoma in situ of the breast (DCIS) is a nonobligate precursor of IBC and shares several risk factors with it. However, the relationship between these biomarkers and DCIS risk remains unexplored. We studied the association between circulating levels of TC, LDL-C, high-density lipoprotein cholesterol (HDL-C), Lipoprotein (a) (Lp-(a)), IGF-1, CRP and HbA1c, with the risk of DCIS in 156801women aged 40 to 69 years and breast cancer-free at enrolment when blood samples and information on demographic and health-related factors were collected. Incident cases of DCIS were ascertained during the follow-up via linkage to the UK cancer registries Multivariable-adjusted Cox proportional hazards models were used to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for the associations of interest. In all, 969 DCIS incident cases were diagnosed during 11.4 years of follow-up. Total cholesterol was inversely associated with the risk of DCIS (HRquintile(Q)5vsQ1 = 0.47, 95% CI: 0.27-0.82, Ptrend = .008). Conversely, LDL-C was positively associated with DCIS risk (HRQ3vsQ1 = 1.43, 95% CI: 1.01-2.04, HRQ4vsQ1 = 1.60, 95% CI: 1.04-2.47, HRQ5vsQ1 = 2.29, 95% CI: 1.36-3.88, Ptrend = .004). In postmenopausal women, CRP had a weak positive association with DCIS risk, while HbA1c showed a nonlinear association with the risk. These results, in conjunction with those from previous studies on IBC, provide support for the association of several biomarkers with the risk of an early stage of breast cancer.

Metabolic Syndrome and the Risk of Ductal Carcinoma In Situ of the Breast in the UK Biobank.

BACKGROUND: Metabolic syndrome (MetS), defined by the presence of three of more metabolic dysregulations such as hyperlipidemia, hyperinsulinemia, central obesity, and hypertension, has been associated with increased risk of cardiovascular disease, diabetes, and various cancers, including invasive breast cancer (IBC). Whether MetS is a risk factor for ductal carcinoma in situ of the breast (DCIS), a nonobligate precursor of IBC, remains unknown. METHODS: A total of 198,748 women ages 40 to 69 years, DCIS- and IBC-free at enrolment in UK Biobank, were included in the current study. Multivariable-adjusted Cox proportional hazards models were used to estimate the association between MetS and DCIS. RESULTS: A total of 1,251 DCIS cases were ascertained during an average follow-up of 11.4 years. There was no association between MetS and the risk of DCIS overall, or by menopausal status. Analysis of individual components of MetS showed an association between central obesity (waist circumference ≥88 cm) and increased DCIS risk in postmenopausal women. CONCLUSIONS: In this prospective study, we found no association between MetS and DCIS risk. IMPACT: The study findings do not support an association between MetS and this breast cancer precursor.

Menopausal hormone therapy and change in physical activity in the Women's Health Initiative hormone therapy clinical trials.

OBJECTIVE: The menopausal transition results in a progressive decrease in circulating estrogen levels. Experimental evidence in rodents has indicated that estrogen depletion leads to a reduction of energy expenditure and physical activity. It is unclear whether treatment with estrogen therapy increases physical activity level in postmenopausal women. METHODS: A total of 27,327 postmenopausal women aged 50-79 years enrolled in the Women's Health Initiative randomized double-blind trials of menopausal hormone therapy. Self-reported leisure-time physical activity at baseline, and years 1, 3, and 6 was quantified as metabolic equivalents (MET)-h/wk. In each trial, comparison between intervention and placebo groups of changes in physical activity levels from baseline to follow-up assessment was examined using linear regression models. RESULTS: In the CEE-alone trial, the increase in MET-h/wk was greater in the placebo group compared with the intervention group at years 3 ( P = 0.002) and 6 ( P < 0.001). Similar results were observed when analyses were restricted to women who maintained an adherence rate ≥80% during the trial or who were physically active at baseline. In the CEE + MPA trial, the primary analyses did not show significant differences between groups, but the increase of MET-h/wk was greater in the placebo group compared with the intervention group at year 3 ( P = 0.004) among women with an adherence rate ≥80%. CONCLUSIONS: The results from this clinical trial do not support the hypothesis that estrogen treatment increases physical activity among postmenopausal women.

Association Between the Healthy Lifestyle Index and Risk of Multimorbidity in the Women's Health Initiative.

BACKGROUND: Multimorbidity, defined as the presence of 2 or more chronic health conditions, is increasingly common among older adults. The combination of lifestyle characteristics such as diet quality, smoking status, alcohol intake, physical activity (PA), sleep duration, and body fat as assessed by body mass index (BMI) or waist circumference, and risk of multimorbidity are not well understood. OBJECTIVES: We investigated the association between the healthy lifestyle index (HLI), generated by combining indicators of diet quality, smoking, alcohol, PA, sleep amount, and BMI, and risk of multimorbidity, a composite outcome that included cardiovascular disease (CVD), diabetes, cancer, and fracture. METHODS: We studied 62 037 postmenopausal women aged 50-79 years at enrollment in the Women's Health Initiative, with no reported history of CVD, diabetes, cancer, or fracture at baseline. Lifestyle characteristics measured at baseline were categorized and a score (0-4) was assigned to each category. The combined HLI (0-24) was grouped into quintiles, with higher quintiles indicating a healthier lifestyle. Multivariable adjusted estimates of hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the risk of developing multimorbidity were obtained using Cox proportional hazard models. RESULTS: Over an average follow-up period of 16.3 years, 5 656 women developed multimorbidity. There was an inverse association between the HLI levels and risk of multimorbidity (compared to the HLI_1st quintile: HR_2nd quintile = 0.81 95% CI 0.74-0.83, HR_3rd quintile = 0.77 95% CI 0.71-0.83, HR_4th quintile = 0.70 95% CI 0.64-0.76, and HR_5th quintile = 0.60 95% CI 0.54-0.66; p trend < .001). Similar associations were observed after stratification by age or BMI categories. CONCLUSIONS: Among postmenopausal women, higher levels of the HLI were associated with a reduced risk of developing multimorbidity.

Healthy Lifestyle Index and Risk of Cardiovascular Disease Among Postmenopausal Women With Normal Body Mass Index.

Background A lifestyle comprising a healthy diet, light alcohol consumption, no smoking, and moderate or intense physical activity has been associated with reduced risk of cardiovascular disease (CVD). We examined the association of a healthy lifestyle index (HLI), derived from scores for each of these components plus waist circumference, with the risk of incident CVD and CVD subtypes in postmenopausal women with normal body mass index (18.5-<25.0 kg/m2). Methods and Results We studied 40 118 participants in the Women's Health Initiative, aged 50 to 79 years at enrollment, with a normal body mass index and no history of CVD. The HLI score was categorized into quintiles. We estimated multivariable adjusted hazard ratios (HR) and 95% CIs for the association of HLI with risk of CVD and CVD subtypes using Cox regression models. A total of 3821 cases of incident CVD were ascertained during a median follow-up of 20.1 years. Compared with the lowest quintile (unhealthiest lifestyle), higher HLI quintiles showed inverse associations with the risk of CVD (HRquintile-2=0.74 [95% CI, 0.67-0.81]; HRquintile-3=0.66 [95% CI, 0.60-0.72]; HRquintile-4=0.57 [95% CI, 0.51-0.63]; and HRquintile-5=0.48 [95% CI, 0.43-0.54], P-trend=<0.001). HLI was also inversely associated with risks of stroke, coronary heart disease, myocardial infarction, angina, and coronary revascularization. Subgroup analyses, stratified by age (≤63 years vs >63 years), body mass index (

Healthy lifestyle index and the risk of ductal carcinoma in situ of the breast in the Women's Health Initiative.

A relatively high healthy lifestyle index (HLI) score, representing a healthy diet, participation in moderate to vigorous physical exercise, no smoking, low to no alcohol intake and a normal body mass index, has been associated with a reduced risk of invasive breast cancer. However, no study has shown an association between the HLI and the risk of ductal carcinoma in situ of the breast (DCIS), which is considered to be a nonobligate precursor of invasive breast cancer. We evaluated this association in a prospective cohort of 132 230 postmenopausal women, aged 50 to 79 years, recruited between 1993 and 1998 across the United States and enrolled in the Women's Health Initiative study. The HLI score was created and categorized into quartiles. During an average follow-up of 15.4 years, 2035 DCIS cases were ascertained. Multivariable-adjusted Cox regression was used to estimate hazard ratios (HR) and 95% confidence intervals (CI) for the association of HLI with the risk of DCIS. Women in the highest HLI quartile had a lower DCIS risk than those in the lowest quartile (HR4thQT  = 0.80, 95% CI, 0.70-0.92) and this association was stronger in women with a family history of breast cancer (HR4thQT  = 0.70, 95% CI, 0.52-0.93), and for ER+/PR+ DCIS (HR4thQT  = 0.66, 95% CI, 0.52-0.83). These findings suggest that there is an inverse association between HLI and risk of DCIS, and suggest that the adoption of a healthy lifestyle might lower the risk of DCIS.

Association of Prediagnostic Serum Levels of Vitamin D with Risk of Ductal Carcinoma In Situ of the Breast in the UK Biobank Cohort Study.

BACKGROUND: Higher circulating levels of vitamin D [25(OH)D] have been associated with reduced risk of developing invasive breast cancer. However, their association with risk of ductal carcinoma in situ (DCIS) of the breast has received little attention. METHODS: We examined the association of serum vitamin D with risk of DCIS in 231,203 women, aged 40 to 69 years at enrollment in the UK Biobank cohort study. Multivariable Cox proportional hazards models were used to estimate HRs and 95% confidence intervals for the association of vitamin D with DCIS risk. RESULTS: There was no association between serum 25(OH)D levels and risk of DCIS overall, or by menopausal status. The association was not modified by body mass index category, family history of breast cancer, or current use of menopausal hormone therapy. CONCLUSIONS: In this large prospective cohort study, we did not observe an association between circulating serum levels of vitamin D and risk of DCIS. IMPACT: While previous studies have suggested that serum vitamin D has an inverse association with risk of invasive breast cancer, the present results do not provide evidence for an association with DCIS, a non-obligate precursor of invasive disease.

Healthy lifestyle index and risk of pancreatic cancer in the Women's Health Initiative.

PURPOSE: Lifestyle factors such as smoking, alcohol, body weight, physical activity, and diet quality have been associated with the risk of pancreatic cancer. However, studies of their combined association in women are limited. METHODS: Data on smoking habits, alcohol intake, diet composition, recreational physical activity, body weight, and waist circumference, obtained at recruitment for 136,945 postmenopausal women (aged 50-79 years) participating in the Women's Health Initiative study, were categorized separately, with higher scores for each variable assigned to the categories representing healthier behaviors. The combined healthy lifestyle index (HLI) score, created by summing the scores for each risk factor, was grouped into quartiles. We used multivariable-adjusted Cox regression to estimate hazard ratios (HR) and 95% confidence intervals (CI) for pancreatic cancer risk in association with the HLI. RESULTS: Over an average follow-up period of approximately 16.0 years, 1,119 incident cases of pancreatic cancer were ascertained. Compared to women in the lowest HLI quartile, those in the upper quartiles (qt) had a reduced risk of pancreatic cancer (multivariable-adjusted HRqt3rd 0.83, 95% CI 0.74-0.99; and HRqt4th 0.74, 95% CI 0.62-0.88, respectively, p trend = 0.001). Use of waist circumference instead of BMI in the HLI score yielded similar results. Among women who were either non-diabetic or non-smokers, high HLI was also associated with reduced risk (HRqt4th 0.78, 95% CI 0.65-0.85 and HRqt4th 0.80, 95% CI 0.66-0.97, respectively). Stratification by BMI categories (18.5- < 25.0, 25.0- < 30.0 and > 30.0 kg/m2) showed similar results in all groups. CONCLUSIONS: Our findings suggest that in postmenopausal women, a healthy lifestyle is associated with reduced risk of pancreatic cancer.

Physical activity and risk of benign proliferative epithelial disorders of the breast, in the Women's Health Initiative.

BACKGROUND: Recreational physical activity (PA) has been shown to be inversely associated with breast cancer risk. However, the association of recreational PA with benign proliferative epithelial disorders (BPED) of the breast, conditions associated with increased risk of breast cancer, has not been adequately studied. METHODS: We used data from an ancillary study of benign breast disease conducted among the 68 132 postmenopausal women (aged 50-79 at recruitment) participating in the Women's Health Initiative randomized clinical trials. All clinical trial participants underwent annual or biennial mammogram screening. During the follow-up, for women who reported breast biopsies but were cancer free, the associated histological sections were obtained and subjected to standardized central pathology review. Self-reported recreational PA at baseline (n = 61 684) and at 3 years of the follow-up (n = 55 923) were quantified as metabolic equivalents [MET]-h/week. There were 1624 confirmed BPED cases during an average follow-up time of 7.7 years. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). RESULTS: Higher average PA over 4 years was associated with lower risk of non-atypical BPED (P-trend = 0.02). There was a 6% lower risk of non-atypical BPED for every 5 MET-h/week increase between baseline and year 3 (HR = 0.94, 95% CI 0.89-0.99). Compared with women who remained inactive (PAbaseline and PAyear3 <9 MET-h/week), those who became active (PAbaseline<9 MET-h/week to PAyear3 ≥9 MET-h/weekee), remained active (PAbaseline and PAyear3 ≥9 MET-h/week), or decreased activity (PAbaseline ≥9 MET-h/week to PAyear3 <9 MET-h/week) had lower BPED risk. CONCLUSIONS: Recreational physical activity after menopause was associated with lower BPED risk among postmenopausal women.

Association of a Healthy Lifestyle Index with Risk of Breast Cancer among Women with Normal Body Mass Index in the UK Biobank.

BACKGROUND: A high healthy lifestyle index (HLI), a composite score based on good diet quality, low alcohol consumption, no smoking, moderate to high physical activity, and waist circumference <80 cm, has been consistently associated with a reduced risk of breast cancer. Recently, high levels of body fat were found to be associated with an elevated risk of breast cancer in postmenopausal women with a normal body mass index (BMI; 18.5-<25 kg/m2). Whether the HLI is associated with breast cancer risk in women with normal BMI is unknown. METHODS: We studied 102,572 women aged 40 to 69 years with a normal BMI at enrollment into the UK Biobank cohort study. The HLI was created by assigning to each component higher scores for healthier behaviors and then summing the scores. The HLI was categorized by tertiles and age- and multivariable-adjusted HRs for the association of the HLI with breast cancer risk by menopausal status were estimated using Cox proportional hazards models. RESULTS: In postmenopausal women, compared with a low HLI, higher scores were associated with a reduced risk of breast cancer [HRHLI-3rd tertile = 0.76; 95% confidence interval (CI), 0.64-0.91]. Findings were similar for premenopausal women, although they did not reach statistical significance, except when smoking status was excluded from the HLI score (HLIwithout smoking: HR3rd tertile = 0.71; 95% CI, 0.56-0.90). CONCLUSIONS: In normal BMI postmenopausal women, a high HLI score was associated with a reduced risk of breast cancer. IMPACT: Following a healthy lifestyle may reduce the risk of breast cancer among normal weight postmenopausal women.

A Randomized Trial of Calcium Plus Vitamin D Supplementation and Risk of Ductal Carcinoma In Situ of the Breast.

Background: The effect of calcium plus vitamin D (CaD) supplementation on risk of ductal carcinoma in situ (DCIS) of the breast, a nonobligate precursor of invasive ductal carcinoma, is not well understood. In this secondary analysis, we examined this association in the Women's Health Initiative CaD trial over approximately 20 years of follow-up. Methods: A total of 36 282 cancer-free postmenopausal women (50-79 years) were randomly assigned to daily (d) calcium (1000 mg) plus vitamin D (400 IU) supplementation or to a placebo. Personal supplementation with vitamin D (≤600 IU/d, subsequently raised to 1000 IU/d) and calcium (≤1000 mg/d) was allowed. The intervention phase (median = 7.1 years), was followed by a postintervention phase (additional 13.8 years), which included 86.0% of the surviving women. A total of 595 incident DCIS cases were ascertained. Hazard ratios (HRs) plus 95% confidence intervals (CIs) were calculated. Results: The intervention group had a lower risk of DCIS throughout follow-up (HR = 0.82, 95% CI = 0.70 to 0.96) and during the postintervention phase (HR = 0.76, 95% CI = 0.61 to 0.94). The group that used CaD personal supplements in combination with the trial intervention had a lower risk of DCIS compared with the trial placebo group that did not use personal supplementation (HR = 0.72, 95% CI = 0.56 to 0.91). Conclusions: CaD supplementation in postmenopausal women was associated with reduced risk of DCIS, raising the possibility that consistent use of these supplements might provide long-term benefits for the prevention of DCIS.

Low-Fat Dietary Modification and Risk of Ductal Carcinoma In Situ of the Breast in the Women's Health Initiative Dietary Modification Trial.

BACKGROUND: Results of observational studies of the association between dietary fat and risk of invasive breast cancer have been inconsistent. In the Women's Health Initiative dietary modification (DM) randomized trial designed to lower fat intake, the intervention was not associated with a statistically significant reduction of overall breast cancer risk. However, the DM association with risk of ductal carcinoma in situ (DCIS) of the breast, a putative breast cancer precursor, has not been reported. METHODS: A total of 48,835 postmenopausal women, ages 50-79 years at enrollment, with no breast cancer history and ≥32% of total energy intake from fat, were randomly assigned either to a dietary intervention (n = 19,541) designed to reduce total fat intake to 20% of energy and to increase vegetable, fruit, and grain consumption, or to a comparison group (n = 29,294). Cox proportional hazards models were used to estimate HRs and 95% confidence intervals for the association between the intervention and DCIS risk. RESULTS: During 18.7 years (median) cumulative follow-up, including intervention (∼8.7 years) and post-intervention phases (∼13.0 years), 817 DCIS cases were ascertained. No evidence of an association between the DM intervention and DCIS risk was observed overall, or by trial phase (intervention and post-intervention). Similarly, no associations were found in subgroups defined by potential risk factors for DCIS. CONCLUSIONS: DM aiming to reduce fat intake was not associated with altered risk of DCIS. IMPACT: These results do not provide evidence of an association between dietary fat reduction and the risk of DCIS among postmenopausal women.

Longitudinal multi-gene panel assessment of circulating tumor DNA revealed tumor burden and molecular characteristics along treatment course of non-small cell lung cancer.

BACKGROUND: Most studies associating circulating tumor DNA (ctDNA) with outcome in lung cancer treatment were either cross-sectional or, if longitudinal, only analyzed a limited number of genes. This study evaluated the potential of utilizing ctDNA profiled by a panel of common cancer genes to monitor tumor burden and to reveal molecular characteristics of tumor along treatment course. METHODS: Twenty Chinese non-small cell lung cancer (NSCLC) patients with serial plasma samples collected (I) before starting on either first- or second-line treatment, (II) at stable disease on treatment, and (III) upon disease progression, were analyzed for mutations in ctDNA using the PGDx 64-gene panel. Paired statistics compared mutation profiles between any two of the three time points. RESULTS: Proportions with detectable ctDNA decreased from 65% at baseline to 35% at stable disease and rose to 80% at progression (P=0.012, between stable disease and progression); median ctDNA levels (mutated fragments per mL) were 7.8, 0, and 24.7 at the three time points, respectively (P=0.013 between baseline and progression; P=0.007 between stable disease and progression). Although plasma epidermal growth factor receptor (EGFR) mutations were commonly detected, 15% of patients had mutations other than EGFR detected during progression, such as various types of TP53 mutations. CONCLUSIONS: ctDNA profiling in serial blood samples reflected tumor burden over time, and a multi-gene panel was more sensitive in indicating lung cancer progression on treatment than a single gene approach. The detection of additional oncogenic mutations or their disappearance suggested evolution of tumor heterogeneity along treatment course.

Sex hormones, SHBG and risk of colon and rectal cancer among men and women in the UK Biobank.

BACKGROUND: Experimental studies have suggested a role for sex hormones in the etiology of colorectal cancer (CRC) but epidemiological data are inconclusive. METHODS: We examined the associations of testosterone, estradiol, and sex hormone binding globulin (SHBG), with risk of CRC (n = 3,247) in 206,508 men and 219,106 women enrolled in the UK Biobank. Participants were followed for a median of 7.1 years. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95 % confidence intervals (CIs) for CRC risk. RESULTS: In men, in multivariable adjusted models testosterone and SHBG were not associated with CRC. Among men in the highest tertile of physical activity, SHBG was inversely associated with risk (HRq5vs. q1 0.75, 0.56-0.99,). In women, testosterone and SHBG were not associated with CRC risk. There were no differences in the associations between testosterone, SHBG and CRC risk in the analyses stratified by menopausal status. We did not observe an association between estradiol and CRC risk; however, given the limited number of individuals with detectable values of estradiol (13.2 % of the total sample) we are unable to draw a firm conclusions regarding this association. CONCLUSION: The results of this study did not provide support for associations of sex hormones and SHBG with CRC risk.

Association of Sex Hormones with Risk of Cancers of the Pancreas, Kidney, and Brain in the UK Biobank Cohort Study.

BACKGROUND: There is some evidence to suggest that endogenous levels of sex hormones might influence the etiology of cancers of the pancreas, kidney, and brain, but epidemiologic data are lacking. METHODS: We evaluated the association of circulating levels of total and free testosterone, and of sex hormone-binding globulin (SHBG), with the risk of cancers of the pancreas, kidney, and brain, and of total and free estradiol with the risk of kidney cancer, in the UK Biobank cohort study (n = 425,793; 225 pancreatic cancers, 749 kidney cancers, 467 brain cancers). Multivariable Cox proportional hazards models were used to estimate HRs and 95% confidence intervals for the associations. RESULTS: Testosterone and SHBG levels were not associated with risk of pancreatic cancer. Most of the associations for the other two anatomic sites were null. There were inverse associations between total testosterone and brain cancer in men and between SHBG and risk of kidney cancer in the total sample and in women. Estradiol was not associated with the risk of kidney cancer. CONCLUSIONS: The results of this study provide little support for associations between sex hormones/SHBG and risk of cancers of the pancreas, kidney, and brain. Larger studies are warranted. IMPACT: Although these results provide little support for roles for sex hormones and SHBG in the etiology of cancers of the pancreas, kidney, and brain, there is a need for studies with larger numbers of cases.

Diabetes, Glycated Hemoglobin, and Risk of Cancer in the UK Biobank Study.

BACKGROUND: Evidence suggest that diabetes and glycated hemoglobin (HbA1c) levels are associated with cancer risk. However, previous studies have been limited variably by failure to adjust for cancer-specific risk factors (e.g., body mass index), inattention to diabetes duration and use of antidiabetic medications, and failure to stratify by obesity. METHODS: We examined the association between diabetes, HbA1c, and cancer risk in the UK Biobank, using data from 476,517 participants (54% women), followed for an average period of 7.1 years. Diabetes was defined on the basis of baseline self-reported diagnosis of diabetes and/or use of diabetes medication, while HbA1c measured at baseline was categorized as low (<31 mmol/mol), normal (31-<39 mmol/mol), increased risk (39-<48 mmol/mol), and high risk for diabetes (≥48 mmol/mol). Multivariable Cox proportional hazards models were used to estimate the association of diabetes and cancer at different anatomical sites, with adjustment for cancer-specific risk factors. RESULTS: Diabetes was associated with increased risk of cancers of the stomach, liver, bladder, endometrium, and lung among smokers, and with decreased risk of prostate cancer. Compared with the normal HbA1c category, the increased risk category was positively associated with risk of cancers of the colon, liver, bladder, and lung among smokers, and the high-risk category was associated with increased risk of cancers of the esophagus, liver, pancreas, and bladder, and with decreased risk of prostate cancer. CONCLUSIONS: These results suggest that both diabetes and/or elevated HbA1c are associated with risk of cancer at several anatomic sites. IMPACT: The associations of diabetes and HbA1c levels with cancer suggest their importance in cancer prevention.

Risk factors for ductal carcinoma in situ of the breast in the UK Biobank cohort study.

Ductal carcinoma in situ of the breast (DCIS) is considered to be a non-obligate precursor of invasive breast cancer (IBC). This suggests that risk factors for DCIS should be a subset of those for IBC. To this end, we investigated whether demographic, lifestyle, and reproductive factors that have been linked to IBC risk are also associated with DCIS risk. This study was conducted in 263,788 women aged 40-69 years at enrolment into the UK Biobank population-based cohort. Information on demographic, reproductive and health factors was collected at baseline using computerized questionnaires, while incident DCIS was ascertained through linkage to UK cancer registries. Age-adjusted and multivariable hazard ratios were estimated using Cox proportional hazards models in the total sample and by menopausal status. During an average of 7 years of follow-up, 1,016 women developed DCIS. Multivariable analysis indicated that age, physical activity, height, family history of breast cancer, menopausal status, parity, and years between menarche and first live birth had associations with DCIS risk. Among post-menopausal women not using hormone replacement therapy, body mass index ≥30 kg/m2 was associated with increased DCIS risk. This study, the largest to date including both pre-menopausal and post-menopausal women, confirms previous findings indicating correspondence between risk factors for DCIS and IBC and highlights the potential contribution to DCIS risk of anthropometric measures not previously reported to be associated with the disease, such as height and BMI amongst post-menopausal women.

Smoking and alcohol drinking effect on radiotherapy associated risk of second primary cancer and mortality among breast cancer patients.

BACKGROUND: Smoking and alcohol consumption are potential risk factors for breast cancer (BC) and may modify the risk of radiotherapy-associated second primary cancer (SPC) occurrence and total mortality. We explored the joint effect of smoking, or alcohol drinking, and radiotherapy on the risk of SPC and overall mortality among BC survivals. METHODS: We conducted a cancer registry-based study of 10,676 BC cases (stage 0-III) with data on smoking and alcohol consumption at time of diagnosis and clinical and therapeutics characteristics. Multivariable Cox proportional hazard models were used to estimate Hazard Ratios [HRs] and 95% confidence interval [CI] of total and site-specific SPC and mortality adjusting for demographic and cancer related characteristics. RESULTS: The SPC risk associated with radiotherapy was higher among ever-smokers than never-smokers (p for interaction = 0.04). Compared to never-smokers/unirradiated, the adjusted HR for ever-smokers/irradiated was 1.79 (95%CI, 1.43-2.23), and for never-smokers/irradiated was 1.31 (95%CI, 1.06-1.63). Analysis by cancer site showed that for ever-smokers/irradiated the risk for hematological, gastrointestinal, gynecological urological and lung/pulmonary cancer was significantly increased by two to five-fold. Mortality was significantly higher for ever-smokers/irradiated (HR = 1.25; 95%CI, 1.06-1.47), but was lower for never-smokers/irradiated (HR = 0.85; 95%CI, 0.73-0.99). Alcohol consumption did not alter the association between radiotherapy and SPC risk, but was associated with lower mortality risk. CONCLUSION: Patients who received radiotherapy and smoked before or at time of BC diagnosis have an increased risk for specific SPCs; drinking alcohol did not alter the effect of radiotherapy. Smoking significantly increased mortality risk reducing the protective effect of radiotherapy treatment.