Male pattern baldness and prostate cancer risk in a population-based case–control study

Purpose: Male pattern baldness (MPB) and prostate cancer (PCa) share commonality as prevalent, heritable and androgen-related conditions. Studies exploring the relationship between the two conditions have been inconclusive. Using a population-based, case–control study of PCa, we explore the relationship between early-onset MPB and PCa risk. Methods: Cases were men aged 35–74 diagnosed with PCa between 2002 and 2005 in King County, Washington. Controls were frequency matched by age and identified by random digit dialing. Hair pattern at age 30 and at 1 year prior to diagnosis (cases) or reference date (controls) was determined using showcards. PCa risk associated with balding was assessed with logistic regression. Results: Data from 999 cases of PCa and 942 controls were analyzed. Hair loss at age 30 was more common in controls (25.2%) than cases (19.8%, p = 0.005), and those with hair loss at age 30 had a 29% relative risk reduction for PCa (OR 0.71, 95% CI 0.56–0.91). No risk reduction was seen for men only reporting hair loss at referent age (OR 0.90, 95% CI 0.73–1.12). In men aged >60 at referent date, the risk reduction was greater for men with hair loss at age 30 from both the top of head and forehead (OR 0.55, 95% CI 0.33–0.93). Conclusion: Early-onset MPB was associated with a reduced relative risk of PCa in this population-based study. Further research into a possible mechanistic link between these prevalent and androgen-related conditions is warranted.     

History of gonorrhea and prostate cancer in a population-based case–control study in Mexico

We evaluated the association between a history of sexually transmitted diseases (STDs) and the risk for prostate cancer (PC) among Mexican males.MethodsPC incident cases (n = 402) that were identified at six public hospitals in Mexico City were matched by age (±5 years) with 805 population controls with no history of PC. By face-to-face interview, we obtained information about sexual history, previous STDs, sociodemographic characteristics, and familial history of PC. An unconditional logistic regression model was used to estimate the risk for PC.ResultsA total of 16.6% of men reported having had at least one previous STD, and the most frequently reported STD was gonorrhea (10.5%). After adjusting by PC familial history, the history of STD was associated with a two-fold greater risk of PC: odds ratio (OR) = 2.67; 95% confidence interval (95% CI = 1.91–3.73). When each STD was evaluated separately, only gonorrhea was associated with a significant increase in PC risk (OR = 3.04; 95% CI = 1.99–4.64). These associations were similar when we stratified by low-risk PC (Gleason <7) and high-risk PC (Gleason ≥7).ConclusionThese results confirm that STDs, and particularly gonorrhea, may play an etiological role in PC among Mexican males, which is consistent with a previous report from a multiethnic cohort.     

Family history of cancer and the risk of bladder cancer: A case–control study from Italy

BackgroundA family history of bladder cancer has been associated with the risk of bladder cancer, but quantification of the excess risk in different populations is still a relevant issue. Further, the role of a family history of other cancers on the risk of bladder cancer remains unclear.MethodsWe analyzed data from an Italian case–control study, including 690 bladder cancer cases and 665 hospital controls. Odds ratios (ORs) were estimated through unconditional logistic regression models, adjusted for sex, age, study center, year of interview and further for education, smoking and sibling’s number.ResultsThe OR for family history of bladder cancer was 2.13 (95% confidence intervals (95%CIs) 1.02–4.49) from the model with partial adjustment, and 1.99 (95%CI 0.91–4.32) after additional adjustment for smoking and siblings’ number, based on 23 cases (3.3%) and 11 controls (1.7%) with a family history of bladder cancer. The fully adjusted OR was 3.77 when the relative was diagnosed at age below 65 years. Smokers with a family history of bladder cancer had a four-fold increased risk compared to non-smokers without a family history. Bladder cancer risk was significantly increased among subjects with a family history of hemolymphopoietic cancers (OR = 2.97, 95%CI 1.35–6.55). Family history of cancer at other sites showed no significant association with bladder cancer risk.ConclusionThis study confirms an approximately two-fold increased risk of bladder cancer for family history of bladder cancer, and indicates a possible familial clustering of bladder cancer with cancers of the hemolymphopoietic system.     

Carcinogen DNA adducts and the risk of colon cancer: case–control study

Colorectal cancer represents 8.5% of all tumours at the King Faisal Specialist Hospital & Research Centre. Environmental and dietary carcinogens such as polycyclic aromatic hydrocarbons (PAHs) and heterocyclic amines (HCAs) have long been suspected to play a prominent role in colon cancer aetiology. We designed a case–control study to test the hypothesis of whether or not the presence of DNA adducts can play a role in the aetiology of colon cancer. DNA adducts were measured in 24 cancerous and 20 non-cancerous tissue samples of newly diagnosed colon cancer patients by ³²P-post-labelling technique. Normal tissue from 19 hospital patients served as controls. The mean levels of adducts per 10¹⁰ nucleotides in cancerous and non-cancerous tissue were 151.75±217.27 and 114.81±186.10, respectively; however, only adducts in cancerous tissue were significantly higher than controls (32.78±57.51 per 10¹⁰ nucleotides) with p-values of 0.017. No BPDE-DNA adducts were found. No relationship was found between urinary cotinine as a marker of tobacco smoke and 1-hydroxypyrene as an indicator of an individual's internal dose of PAHs and DNA adducts. In a logistic regression model, only adducts in cancerous tissue were associated with the subsequent risk of colon cancer, with an odds ratio of 3.587 (95% confidence interval 0.833–15.448) after adjustment for age and the duration of living in the current region, but of a borderline significance (p=0.086). Although it is difficult to arrive at a definite conclusion from a small dataset, our preliminary results suggest the potential role of DNA adducts in the colon carcinogenesis process. Additional studies with larger sample sizes are needed to confirm our preliminary finding. It is also important to identify the structural characterization of these unknown DNA adducts in order to have a better understanding of whether or not environmental carcinogens play a role in the aetiology of colon cancer.     

Prolonged breastfeeding reduces risk of breast cancer in Sri Lankan women: A case–control study

Goal: To assess the association between duration of breastfeeding and the risk of breast cancer in Sri Lankan women. Methods: We conducted a case–control study in women aged 30–64 years in selected health care facilities in the Western province. A total of 100 recent cases of breast cancer (histologically confirmed) and 203 controls (age and parity matched) were included. Detailed information regarding breastfeeding, menstruation, reproductive factors, passive smoking and other confounders was collected using a structured questionnaire. Adjusted odds ratios and 95% confidence intervals were calculated using multiple logistic regressions. Principle results: Multivariate analysis found that those women who breastfed for ≥24 months during lifetime had significantly lower risk of breast cancer than those who breastfed for less than 24 months (OR = 0.40; 95%CI = 0.22, 0.73). Compared to 0–11 months of lifetime breastfeeding, there was a 66.3% reduction in breast cancer risk in women who breastfed for 12–23 months, 87.4% reduction in 24–35 months and 94% reduction in 36–47 months categories. The mean duration of breastfeeding per child for ≥12 months was also associated with reduced risk of breast cancer (OR = 0.52; 95%CI = 0.28, 0.94). The significant factors associated with increased risk of breast cancer were: post-menopausal women (OR = 1.74; 95%CI = 1.01, 3.01); having an abortion in the past (OR = 3.42; 95%CI = 1.75, 6.66) and exposure to passive smoking (OR = 2.96, 95%CI = 1.53, 5.75). Major conclusions: Prolonged breastfeeding significantly reduces the risk of breast cancer and this protective effect was supported by a dose–response relationship. Risk due to passive smoking should be emphasized in anti-smoking programmes.     

ERCC1 haplotypes modify bladder cancer risk: A case–control study

Bladder cancer risk is highly influenced by environmental and/or predisposing genetic factors. In the last decades growing evidence of the major role played by DNA repair systems in the developing of bladder cancer has been provided. To better investigate the involvement of DNA repair genes previously reported to be significantly associated with bladder cancer risk, we examined in a case–control study (456 cases and 376 hospital controls) 36 single nucleotide polymorphisms (SNPs) in 10 DNA repair genes, through a better gene coverage and a deep investigation of the haplotype role. A single SNP analysis showed a significantly increased risk given by XRCC1-rs915927 G allele (OR = 1.55, CI 95% 1.02–2.37 for dominant model) and a protective effect of the rare alleles of 3 ERCC1 SNPs: rs967591 (OR = 0.66, CI 95% 0.46–0.95), rs735482 (OR = 0.62, CI 95% 0.42–0.90) and rs2336219 (OR = 0.63, CI 95% 0.43–0.93). Haplotype analysis revealed that cases had a statistically significant excess of XRCC3-TAGT and ERCC1-GAT haplotypes, whereas ERCC1-AAC, MGMT-TA, XRCC1-TGCC and ERCC2-TGAA haplotypes were significantly underrepresented. Together with other published data on large case–control studies, our findings provide epidemiological evidence supporting a link between DNA repair gene variants and bladder cancer development, and suggest that the effects of high-order interactions should be taken into account as modulating factors affecting bladder cancer risk. A detailed characterization of DNA repair genetic variation is warranted and might ultimately help to identify multiple susceptibility variants that could be responsible for joint effects on the risk.     

Associations between oral hygiene habits,diet, tobacco and alcohol and risk of oral cancer: A case–control study from India

ObjectiveThis study examines the association between the incidence of oral cancer in India and oral hygiene habits, diet, chewing and smoking tobacco, and drinking alcohol. We also assessed the effects of oral hygiene habits with oral cancer risk among chewers versus never chewers.MethodsA hospital-based case–control study was conducted in Pune, India, based on face-to-face interviews, anthropometry, and intra-oral examinations conducted for 187 oral cancer cases and 240 controls.ResultsPoor oral hygiene score was associated with a significant risk of oral cancer (adjusted OR = 6.98; 95%CI 3.72–13.05). When stratified by tobacco-chewing habit, the poor oral hygiene score was a significant risk factor only among ever tobacco chewers (adjusted OR = 14.74; 95%CI 6.49–33.46) compared with never chewers (adjusted OR = 0.71; 95%CI 0.14–3.63). Dental check-ups only at the time of pain by ever-chewers with poor oral hygiene was associated with an elevated risk (adjusted OR = 4.22; 95%CI 2.44–7.29), while consumption of green, yellow, and cruciferous vegetables and citrus fruits was protective. A linear dose–response association was observed between oral cancer and chewing tobacco in terms of age at initiation, duration, and frequency of chewing per day (P < 0.001). Smoking more than 10 bidis/cigarettes per day (adjusted OR = 2.74; 95%CI 1.28–5.89) and for a duration >25 years (adjusted OR = 2.31; 95%CI 1.14–4.71) elevated the risk of oral cancer.ConclusionGood oral hygiene habits – as characterized by healthy gums, brushing more than once daily, use of toothpaste, annual dental check-ups, and a minimal number of missing teeth – can reduce the risk of oral cancer significantly. In addition to refraining from chewing/smoking tobacco, a diet adequate in fruits and vegetables may protect against the disease.     

Parkinson’s disease and risk of prostate cancer: A Danish population-based case-control study, 1995–2010

IntroductionProstate cancer growth and progression may be linked to neurogenesis and to medical anti- Parkinson treatment, but results are inconclusive. Therefore, we examined the association between Parkinson’s disease and risk of prostate cancer in a population based case-control study.MethodsWe identified 45,429 patients diagnosed with incident prostate cancer during 1997–2010 from the National Cancer Registry. Five age-matched population controls (n = 227,145) were selected for each case. Odds ratios (ORs) adjusted for age and comorbidity for prostate cancer associated with Parkinson’s disease were computed using conditional logistic regression. Analyses were stratified by duration of Parkinson’s disease and stage of prostate cancer (localized and advanced).ResultsIn total, 245 patients (0,5%) and 1656 controls (0,7%) had Parkinson’s disease. Overall, patients with Parkinson’s disease had a 27% lower risk of prostate cancer compared with patients without Parkinson’s disease (adjusted OR (ORa) 0.73; 95% confidence interval (CI), 0.63–0.83).Risk of prostate cancer decreased with increasing duration of Parkinson’s disease.The odds ratios were slightly lower for advanced prostate cancer (ORa, 0.68; 95% CI, 0.52–0.88) than for localized prostate cancer (ORa 0.76; 95% CI, 0.61–0.93).ConclusionParkinson’s disease was associated with a risk reduction overall (27%), which decreased with increasing duration of Parkinson’s disease.     

HTERT MNS16A polymorphism in breast cancer: a case–control study

This case–control study aims to investigate the role of HTERT MNS16A polymorphism as a potential risk factors and/or a prognostic marker for breast cancer. 113 consecutive incident cases of histologically confirmed ductal breast cancer and 124 healthy controls were recruited. HTERT MNS16A polymorphism was genotyped (L: long allele, S: short allele); multivariate logistic regression was performed. No significant association was noted either at the overall analysis (OR?=?1.57, 95?% CI 0.84–2.93 for heterozygous LS carriers; OR?=?1.02, 95?% CI 0.54–1.95 for homozygous SS carriers) or at the subanalyses in premenopausal and postmenopausal women. With respect to survival analysis, HTERT MNS16A polymorphism was not associated with either disease-free survival or overall survival. HTERT MNS16A polymorphism does not seem to be a risk factor for breast cancer in the Caucasian Greek population. Further, larger studies from other countries and subjects seem to be needed as this novel polymorphism is being examined in depth.     

Mitochondrial genome variations in advanced stage breast cancer: A case–control study

Entire mitochondrial DNA (mtDNA) sequencing was carried out in 101 primary breast cancer patients and 90 controls of south Indian origin. We identified 69 novel mutations in breast cancer patients and 637 reported polymorphisms in patients and/or controls. PolyPhen-2 analysis predicted 5 out of 14 novel missense mutations as ‘probably damaging variants’. Haplogrouping analysis identified a significant association between haplogroup M5 and breast cancer risk. Microsatellite instability and tumor specific large scale mtDNA deletions were not observed in tumor tissues from the patients. In conclusion, mtDNA mutations and haplogroups may constitute an inheritable risk factor for pathogenesis of breast cancer.     

Increased risk of stroke in oral contraceptive users carried replicated genetic variants: a population-based case–control study in China

Combined oral contraceptives (COC) use is a unique risk factor for stroke in women, and may modify the associations between genetic polymorphisms and stroke. To investigate whether the genetic variants identified in a recent genome-wide association study (GWAS) could be replicated in Chinese women, as well as, whether related risk was different in COC users, 451 stroke cases and 831 age- and region-matched controls were recruited from our cohort. Genotyping of 3 SNPs (rs700651, rs10958409, and rs1333040) was performed by the polymerase chain reaction assay with TaqMan probes. The history of contraceptive use and relevant information were obtained from a face-to-face interview. Odds ratios (OR) with 95 % confidence interval (CI) were estimated under conditional logistic regression model after adjustment for cardiovascular covariates. Our study replicated the associations of rs10958409 and rs1333040, with the risk of stroke, especially hemorrhagic subtype, but failed to confirm association of rs700651. COC use was associated with a 1.56-fold (OR 1.56, 95 % CI 1.21-2.01) increased risk of stroke. COC users with rs10958409 GA/AA or rs1333040 CT/TT genotypes had an increased risk of overall stroke by 1.59-fold (OR 2.59, 95 % CI 1.59-4.19) and 3.24-fold (OR 4.24, 95 % CI 1.71-10.49), respectively, compared with the non-users with wild-type genotypes. Moreover, the risk of hemorrhagic stroke increased by 4.81- and 15.06-fold when risk allele carriers of rs10958409 or rs1333040 who took COC. Our results confirmed the associations of two GWAS SNPs, also suggested combination effects of these genetic variants and COC use on stroke risk.     

2q35-rs13387042 variant and the risk of breast cancer: a case–control study

Molecular Biology Reports - Breast Cancer is the most frequent neoplasm diagnosed among women worldwide. Genetic background and lifestyle/environment play a significant role in the disease...     

Dietary habits during the 2 months following the Chernobyl accident and differentiated thyroid cancer risk in a population-based case–control study

BackgroundThe Chernobyl nuclear power plant accident occurred in Ukraine on April 26th 1986. In France, the radioactive fallout and thyroid radiation doses were much lower than in highly contaminated areas. However, a number of risk projections have suggested that a small excess in differentiated thyroid cancer (DTC) might occur in eastern France due to this low-level fallout. In order to investigate this potential impact, a case–control study on DTC risk factors was started in 2005, focusing on cases who were less than 15 years old at the time of the Chernobyl accident. Here, we aim to evaluate the relationship between some specific reports of potentially contaminated food between April and June 1986 – in particular fresh dairy products and leafy vegetables – and DTC risk.MethodsAfter excluding subjects who were not born before the Chernobyl accident, the study included 747 cases of DTC matched with 815 controls. Odds ratios were calculated using conditional logistic regression models and were reported for all participants, for women only, for papillary cancer only, and excluding microcarcinomas.ResultsThe DTC risk was slightly higher for participants who had consumed locally produced leafy vegetables. However, this association was not stronger in the more contaminated areas than in the others. Conversely, the reported consumption of fresh dairy products was not statistically associated with DTC risk.ConclusionBecause the increase in DTC risk associated with a higher consumption of locally produced vegetables was not more important in the most contaminated areas, our study lacked power to provide evidence for a strong association between consumption of potentially contaminated food and DTC risk.     

Colorectal cancer risk and dyslipidemia: A case–cohort study nested in an Italian multicentre cohort

BackgroundDyslipidemia is an established risk factor for many diseases, but its effect on colorectal cancer risk is less clear. We investigated the association of colorectal cancer risk with plasma triglycerides, total, HDL, and LDL cholesterol in four Italian EPIC centers.MethodsWe conducted a case–cohort study on participants recruited to four Italian EPIC centers (Turin, Varese, Naples, and Ragusa; 34,148 subjects). A random subcohort of 850 subjects was obtained and 286 colorectal cancer cases were diagnosed. Triglycerides, total and HDL cholesterol were determined in plasma samples obtained at baseline and stored at −196 °C; LDL cholesterol was calculated. Hazard ratios (HR) with 95% confidence intervals (CI), adjusted for potential confounders, were estimated by Cox regression models using the Prentice method.ResultsThe highest tertiles of total (HR 1.66, 95%CI 1.12–2.45) and LDL cholesterol (HR 1.87, 95%CI 1.27–2.76) were associated with increased colorectal cancer risk compared to lowest tertiles. Risks were greater for men than women, and for postmenopausal than premenopausal women. Highest tertiles of total and LDL cholesterol were also significantly associated with increased risks of colon cancer, distal colon cancer, and rectal cancer, but not proximal colon cancer.ConclusionsOur findings suggest that high levels of total and LDL cholesterol increase colorectal cancer risk, particularly in men and postmenopausal women. However additional studies are needed to clarify the role of plasma lipids in these cancers, particularly in view of the conflicting findings of previous studies.     

Polymorphisms in TP53 and MDM2 contribute to higher risk of colorectal cancer in Chinese population: a hospital-based, case–control study

The murine double minute 2 protein (MDM2) and TP53 interact in regulating cell cycle, DNA repair and apoptosis process, which is crucial in carcinogenesis. Since functional variations in these two genes were shown to change gene expression and function, we hypothesized that potentially functional polymorphisms in these genes may contribute to colorectal cancer (CRC) susceptibility. A hospital-based case-control study consisting of 444 patients and 569 controls was conducted to explore the associations between TP53 Arg72Pro and MDM2 T309G and CRC risk in Chinese. The combined effect of TP53 Arg72Pro and MDM2 T309G was significant in a gene dose-response increasing CRC risk (trend test: P?=?0.02). Individuals carrying 3 or more potential risk alleles had 1.78 times risk (95?% CI: 1.13-2.80) to develop CRC compared with individuals without potential risk allele. This increased cancer risk was more pronounced in smokers who carried 3-4 potential risk alleles (OR?=?2.75, 95?% CI: 1.14-6.60) and in young subjects (OR?=?2.05, 95?% CI: 1.08-3.88). The gene-gene interaction between TP53 Arg72Pro and MDM2 T309G may interact in carcinogenesis of CRC in Chinese, especially in smokers, and this kind of interaction is associated with onset age of CRC.     

Influenza vaccination,pneumococcal vaccination and risk of acute myocardial infarction: matched case–control study

Background

Previous studies have shown an association between acute myocardial infarction and preceding respiratory infection. Contradictory evidence exists on the influence of influenza vaccination and pneumococcal vaccination in preventing cardiovascular disease. We aimed to investigate the possible association of influenza vaccination and pneumococcal vaccination with acute myocardial infarction.

Methods

We used a matched case–control design with data from the United Kingdom General Practice Research Database. Cases were patients who were at least 40 years of age at diagnosis of first acute myocardial infarction recorded from Nov.1, 2001, to May 31, 2007, and were matched for sex, general practice, age and calendar time (i.e., month corresponding to index date of acute myocardial infarction), with up to four controls each. Data were analyzed using conditional logistic regression, adjusted for vaccination target groups, cardiovascular risk factors, treatment medications and attendances at a general practice.

Results

We included 78 706 patients, of whom 16 012 were cases and 62 694 were matched controls. Influenza vaccination had been received in the previous year by 8472 cases (52.9%) and 32 081 controls (51.2%) and was associated with a 19% reduction in the rate of acute myocardial infarction (adjusted odds ratio [OR] 0.81, 95% confidence interval [CI] 0.77–0.85). Early seasonal influenza vaccination was associated with a lower rate of acute myocardial infarction (adjusted OR 0.79, 95% CI 0.75–0.83) than vaccination after mid-November (adjusted OR 0.88, 95% CI 0.79–0.97). Pneumococcal vaccination was not associated with a reduction in the rate of acute myocardial infarction (adjusted OR 0.96, 95% CI 0.91–1.02).

Interpretation

Influenza vaccination but not pneumococcal vaccination is associated with a reduced rate of first acute myocardial infarction. This association and the potential benefit of early seasonal vaccination need to be considered in future experimental studies.Winter peaks in incidence of acute myocardial infarction have been linked to climate,¹ metabolic factors² and infection.³ Because known risk factors do not fully account for cases of acute myocardial infarction, current interest is focused on the putative link with respiratory infection. Significant increases in acute myocardial infarction occur during peak winter incidence of pneumonia, influenza and influenza-like syndrome,⁴ particularly during years dominated by epidemic rather than nonepidemic influenza A. This association supports the notion that the increase is caused by influenza rather than cold weather.⁵Acute myocardial infarction may increase susceptibility to respiratory illness,⁶ but the association between acute myocardial infarction and respiratory infection occurring within four weeks prior to the acute myocardial infarction^7,8 supports infection as a cause of acute myocardial infarction. Although the exact mechanism is unknown, the favoured hypothesis is that infection triggers plaque rupture.⁹ Although several observational studies, as well as two randomized controlled trials,^10,11 suggest a positive effect of influenza vaccine in preventing acute myocardial infarction, they provide insufficient and conflicting evidence.¹²The aim of this study was to investigate a possible association between influenza or pneumococcal vaccination and acute myocardial infarction.     

Single-nucleotide polymorphisms in one-carbon metabolism genes,Mediterranean diet and breast cancer risk: a case–control study in the Greek-Cypriot female population

Single-nucleotide polymorphisms (SNPs) within genes of the one-carbon metabolism pathway have been shown to interact with dietary folate intake to modify breast cancer (BC) risk. Our group has previously demonstrated that the Mediterranean dietary pattern, rich in beneficial one-carbon metabolism micronutrients, protects against BC in Greek-Cypriot women. We aimed to investigate whether SNPs in the MTHFR (rs1801133 and rs1801131) and MTR (rs1805087) genes modify the effect of the Mediterranean dietary pattern on BC risk. Dietary intake data were obtained using a 32-item food-frequency questionnaire. A dietary pattern specific to the Greek-Cypriot population, which closely resembles the Mediterranean diet, was derived using principal component analysis (PCA) and used as our dietary variable. Genotyping was performed on subjects from the MASTOS study, a case–control study of BC in Cyprus, using TaqMan assays. Adjusted odds ratios (ORs) were estimated using logistic regression analyses. High adherence to the PCA-derived Mediterranean dietary pattern further reduced BC risk with increasing number of variant MTHFR 677T alleles (OR_Q4vs.Q1 for 677TT = 0.37, 95 % CI 0.20–0.69, for 677 CT = 0.60, 95 % CI 0.42–0.86). Additionally, high adherence to the Mediterranean dietary pattern decreased BC risk in subjects with at least one MTR 2756A allele (OR_Q4vs.Q1 for 2756AA = 0.59, 95 % CI 0.43–0.81, for 2756AG = 0.59, 95 % CI 0.39–0.91) and in subjects with the MTHFR 1298CC genotype (OR_Q4vs.Q1 0.44, 95 % CI 0.30–0.65). Overall P-interaction values, however, were not statistically significant. Our study suggests that these MTHFR and MTR SNPs may act as effect modifiers, highlighting their biological significance in the association between Mediterranean diet, the one-carbon metabolism pathway and BC.

Electronic supplementary material

The online version of this article (doi:10.1007/s12263-015-0453-7) contains supplementary material, which is available to authorized users.