Lower risk of hospitalisation for depression following hospitalisation for COVID-19 versus for another reason

INTRODUCTION

Although hospitalisation for COVID-19 is associated with a higher post-discharge risk of mood disorders, including major depressive disorder (MDD) and bipolar disorder (BD), this risk has not been compared to that following hospitalisation for a reason other than COVID-19.

METHODS

Using data from France's National Health Data System (SNDS) database, we compared patients hospitalised for mood disorders in the 12 months following COVID-19/another reason hospitalisation.

RESULTS

96,313 adult individuals were hospitalised for COVID-19, and 2,979,775 were hospitalised for another reason. In the 12 months post-discharge, 110,976 (3.83 %) patients were hospitalised for mood disorders. In unadjusted analyses, patients initially hospitalised for COVID-19 (versus another reason) were more likely to be subsequently hospitalised for a mood disorder (4.27 % versus 3.82 % versus, respectively, p < 0.0001). These patients were also more likely to have a history of mood disorders, especially depressive disorders (6.45 % versus 5.77 %, respectively, p < 0.0001). Women, older age, lower social deprivation, a history of mood disorders, longer initial hospitalisation (COVID-19 or other), and a higher level of clinical care during initial hospitalisation were all significantly associated with the risk of subsequent hospitalisation for MDD and BD. In contrast, after adjusting for all these factors, persons initially hospitalised for COVID-19 were less likely to be subsequently hospitalised for MDD (OR = 0.902 [0.870-0.935]; p < 0.0001). No difference between both groups was observed for BD.

LIMITATIONS

Other reasons were not separately studied.

CONCLUSIONS

After adjusting for confounding factors, initial hospitalisation for COVID-19 versus for another reason was associated with a lower risk of hospitalisation for a mood disorder.

Adverse effects of immunotherapies for multiple sclerosis: a network meta-analysis

BACKGROUND

Multiple sclerosis (MS) is a chronic disease of the central nervous system that affects mainly young adults (two to three times more frequently in women than in men) and causes significant disability after onset. Although it is accepted that immunotherapies for people with MS decrease disease activity, uncertainty regarding their relative safety remains.

OBJECTIVES

To compare adverse effects of immunotherapies for people with MS or clinically isolated syndrome (CIS), and to rank these treatments according to their relative risks of adverse effects through network meta-analyses (NMAs).

SEARCH METHODS

We searched CENTRAL, PubMed, Embase, two other databases and trials registers up to March 2022, together with reference checking and citation searching to identify additional studies.

SELECTION CRITERIA

We included participants 18 years of age or older with a diagnosis of MS or CIS, according to any accepted diagnostic criteria, who were included in randomized controlled trials (RCTs) that examined one or more of the agents used in MS or CIS, and compared them versus placebo or another active agent. We excluded RCTs in which a drug regimen was compared with a different regimen of the same drug without another active agent or placebo as a control arm.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methods for data extraction and pairwise meta-analyses. For NMAs, we used the netmeta suite of commands in R to fit random-effects NMAs assuming a common between-study variance. We used the CINeMA platform to GRADE the certainty of the body of evidence in NMAs. We considered a relative risk (RR) of 1.5 as a non-inferiority safety threshold compared to placebo. We assessed the certainty of evidence for primary outcomes within the NMA according to GRADE, as very low, low, moderate or high.

MAIN RESULTS

This NMA included 123 trials with 57,682 participants. Serious adverse events (SAEs) Reporting of SAEs was available from 84 studies including 5696 (11%) events in 51,833 (89.9%) participants out of 57,682 participants in all studies. Based on the absolute frequency of SAEs, our non-inferiority threshold (up to a 50% increased risk) meant that no more than 1 in 18 additional people would have a SAE compared to placebo. Low-certainty evidence suggested that three drugs may decrease SAEs compared to placebo (relative risk [RR], 95% confidence interval [CI]): interferon beta-1a (Avonex) (0.78, 0.66 to 0.94); dimethyl fumarate (0.79, 0.67 to 0.93), and glatiramer acetate (0.84, 0.72 to 0.98). Several drugs met our non-inferiority criterion versus placebo: moderate-certainty evidence for teriflunomide (1.08, 0.88 to 1.31); low-certainty evidence for ocrelizumab (0.85, 0.67 to 1.07), ozanimod (0.88, 0.59 to 1.33), interferon beta-1b (0.94, 0.78 to 1.12), interferon beta-1a (Rebif) (0.96, 0.80 to 1.15), natalizumab (0.97, 0.79 to 1.19), fingolimod (1.05, 0.92 to 1.20) and laquinimod (1.06, 0.83 to 1.34); very low-certainty evidence for daclizumab (0.83, 0.68 to 1.02). Non-inferiority with placebo was not met due to imprecision for the other drugs: low-certainty evidence for cladribine (1.10, 0.79 to 1.52), siponimod (1.20, 0.95 to 1.51), ofatumumab (1.26, 0.88 to 1.79) and rituximab (1.01, 0.67 to 1.52); very low-certainty evidence for immunoglobulins (1.05, 0.33 to 3.32), diroximel fumarate (1.05, 0.23 to 4.69), peg-interferon beta-1a (1.07, 0.66 to 1.74), alemtuzumab (1.16, 0.85 to 1.60), interferons (1.62, 0.21 to 12.72) and azathioprine (3.62, 0.76 to 17.19). Withdrawals due to adverse events Reporting of withdrawals due to AEs was available from 105 studies (85.4%) including 3537 (6.39%) events in 55,320 (95.9%) patients out of 57,682 patients in all studies. Based on the absolute frequency of withdrawals, our non-inferiority threshold (up to a 50% increased risk) meant that no more than 1 in 31 additional people would withdraw compared to placebo. No drug reduced withdrawals due to adverse events when compared with placebo. There was very low-certainty evidence (meaning that estimates are not reliable) that two drugs met our non-inferiority criterion versus placebo, assuming an upper 95% CI RR limit of 1.5: diroximel fumarate (0.38, 0.11 to 1.27) and alemtuzumab (0.63, 0.33 to 1.19). Non-inferiority with placebo was not met due to imprecision for the following drugs: low-certainty evidence for ofatumumab (1.50, 0.87 to 2.59); very low-certainty evidence for methotrexate (0.94, 0.02 to 46.70), corticosteroids (1.05, 0.16 to 7.14), ozanimod (1.06, 0.58 to 1.93), natalizumab (1.20, 0.77 to 1.85), ocrelizumab (1.32, 0.81 to 2.14), dimethyl fumarate (1.34, 0.96 to 1.86), siponimod (1.63, 0.96 to 2.79), rituximab (1.63, 0.53 to 5.00), cladribine (1.80, 0.89 to 3.62), mitoxantrone (2.11, 0.50 to 8.87), interferons (3.47, 0.95 to 12.72), and cyclophosphamide (3.86, 0.45 to 33.50). Eleven drugs may have increased withdrawals due to adverse events compared with placebo: low-certainty evidence for teriflunomide (1.37, 1.01 to 1.85), glatiramer acetate (1.76, 1.36 to 2.26), fingolimod (1.79, 1.40 to 2.28), interferon beta-1a (Rebif) (2.15, 1.58 to 2.93), daclizumab (2.19, 1.31 to 3.65) and interferon beta-1b (2.59, 1.87 to 3.77); very low-certainty evidence for laquinimod (1.42, 1.01 to 2.00), interferon beta-1a (Avonex) (1.54, 1.13 to 2.10), immunoglobulins (1.87, 1.01 to 3.45), peg-interferon beta-1a (3.46, 1.44 to 8.33) and azathioprine (6.95, 2.57 to 18.78); however, very low-certainty evidence is unreliable. Sensitivity analyses including only studies with low attrition bias, drug dose above the group median, or only patients with relapsing remitting MS or CIS, and subgroup analyses by prior disease-modifying treatments did not change these figures. Rankings No drug yielded consistent P scores in the upper quartile of the probability of being better than others for primary and secondary outcomes.

AUTHORS' CONCLUSIONS

We found mostly low and very low-certainty evidence that drugs used to treat MS may not increase SAEs, but may increase withdrawals compared with placebo. The results suggest that there is no important difference in the occurrence of SAEs between first- and second-line drugs and between oral, injectable, or infused drugs, compared with placebo. Our review, along with other work in the literature, confirms poor-quality reporting of adverse events from RCTs of interventions. At the least, future studies should follow the CONSORT recommendations about reporting harm-related issues. To address adverse effects, future systematic reviews should also include non-randomized studies.

Can machine learning provide preoperative predictions of biological hemostasis after extracorporeal circulation for cardiac surgery?

OBJECTIVES

The goal of this study was to improve decision making regarding the transfusion of patients at the end of extracorporeal circulation for cardiac surgery through machine learning predictions of the evolution of platelets counts, prothrombin ratio, and fibrinogen assay.

METHODS

Prospective data with information about patient preoperative biology and surgery characteristics were collected at Institut Mutualiste Montsouris Hospital (Paris, France) for 10 months (n = 598). For each outcome of interest, instead of arbitrarily choosing 1 machine learning algorithm, we trained and tested a variety of algorithms together with the super learning algorithm, a state-of-the-art ensemble method that aggregates all the predictions and selects the best performing algorithm (total, 137 algorithms). We considered the top-performing algorithms and compared them to more standard and interpretable multivariable linear regression models. All algorithms were evaluated through their root mean squared error, a measure of the average difference between true and predicted values.

RESULTS

The root mean squared error of the top algorithms for predicting the difference between pre- and postoperative platelet counts, prothrombin ratio, and fibrinogen assay were 38.27 × 10e9/L, 8.66%, and 0.44 g/L, respectively. The linear models had similar performances.

CONCLUSIONS

Our machine learning algorithms accurately predicted prothrombin ratio and fibrinogen assay and less accurately platelet counts. As such, our models could provide an aid-decision tool for anesthetists in an operating room; future clinical trials addressing this hypothesis are warranted.

The risk of hospitalization for psychotic disorders following hospitalization for COVID-19: a French nationwide longitudinal study

COVID-19, like other infectious diseases, may be a risk factor for psychotic disorders. We aimed to compare the proportions of hospitalizations for psychotic disorders in the 12 months following discharge from hospital for either COVID-19 or for another reason in the adult general population in France during the first wave of the pandemic. We conducted a retrospective longitudinal nationwide study using the national French administrative healthcare database. Psychotic disorders were first studied as a whole, and then chronic and acute disorders separately. The role of several adjustment factors, including sociodemographics, a history of psychotic disorder, the duration of the initial hospitalization, and the level of care received during that hospitalization, were also analyzed. Between 1 January 2020 and 30 June 2020, a total of 14,622 patients were hospitalized for psychotic disorders in the 12 months following discharge from hospital for either COVID-19 or another reason. Initial hospitalization for COVID-19 (vs. another reason) was associated with a lower rate of subsequent hospitalization for psychotic disorders (0.31% vs. 0.51%, odds ratio (OR) = 0.60, 95% confidence interval (CI) [0.53-0.67]). This was true for both chronic and acute disorders, even after adjusting for the various study variables. Importantly, a history of psychotic disorder was a major determinant of hospitalization for psychotic disorders (adjusted OR = 126.56, 95% CI [121.85-131.46]). Our results suggest that, in comparison to individuals initially hospitalized for another reason, individuals initially hospitalized for COVID-19 present a lower risk of hospitalization for first episodes of psychotic symptoms/disorders or for psychotic relapse in the 12 months following discharge. This finding contradicts the hypothesis that there is a higher risk of psychotic disorders after a severe COVID-19.

Mediating Role of Lifestyle Behaviors in the Association between Education and Cancer: Results from the European Prospective Investigation into Cancer and Nutrition

BACKGROUND

Many studies have shown that socioeconomic position (SEP) is associated with the incidence of malignant tumors at different sites. This study aims to estimate the association between educational level (as proxy for SEP) and cancer incidence and to understand whether the observed associations might be partially explained by lifestyle behaviors.

METHODS

The analyses were performed on data from the European Prospective Investigation into Cancer and Nutrition (EPIC) study, globally and by sex. We used Cox proportional hazards models together with mediation analysis to disentangle the total effect (TE) of educational level [measured through the Relative Index of Inequality (RII)] on cancer incidence into pure direct (PDE) and total indirect (TIE) effect, unexplained and explained by mediators, respectively. PDE and TIE were then combined to compute the proportions mediated (PM).

RESULTS

After an average of 14 years of follow-up, 52,422 malignant tumors were ascertained. Low educated participants showed higher risk of developing stomach, lung, kidney (in women), and bladder (in men) cancers, and, conversely, lower risk of melanoma and breast cancer (in post-menopausal women), when compared with more educated participants. Mediation analyses showed that portions of the TE of RII on cancer could be explained by site-specific related lifestyle behaviors for stomach, lung, and breast (in women).

CONCLUSIONS

Cancer incidence in Europe is determined at least in part by a socioeconomically stratified distribution of risk factors.

IMPACT

These observational findings support policies to reduce cancer occurrence by altering mediators, such as lifestyle behaviors, particularly focusing on underprivileged strata of the population.

Error rate control for classification rules in multiclass mixture models

In the context of finite mixture models one considers the problem of classifying as many observations as possible in the classes of interest while controlling the classification error rate in these same classes. Similar to what is done in the framework of statistical test theory, different type I and type II-like classification error rates can be defined, along with their associated optimal rules, where optimality is defined as minimizing type II error rate while controlling type I error rate at some nominal level. It is first shown that finding an optimal classification rule boils down to searching an optimal region in the observation space where to apply the classical Maximum A Posteriori (MAP) rule. Depending on the misclassification rate to be controlled, the shape of the optimal region is provided, along with a heuristic to compute the optimal classification rule in practice. In particular, a multiclass FDR-like optimal rule is defined and compared to the thresholded MAP rules that is used in most applications. It is shown on both simulated and real datasets that the FDR-like optimal rule may be significantly less conservative than the thresholded MAP rule.

Mediation models of anxiety and depression between temperament and drive for thinness and body dissatisfaction in anorexia nervosa

PURPOSE

Anorexia nervosa (AN) is a life-threatening condition in which temperament, anxiety, depression, and core AN body-related psychopathology (drive for thinness, DT, and body dissatisfaction, BD) are intertwined. This relationship has not been to date disentangled; therefore, we performed a multiple mediation analysis aiming to quantify the effect of each component.

METHODS

An innovative multiple mediation statistical method has been applied to data from 184 inpatients with AN completing: Temperament Evaluation of Memphis, Pisa, Paris, and San Diego Autoquestionnaire, Eating Disorders Inventory-2, State-Trait Anxiety Inventory, and Beck Depression Inventory.

RESULTS

All affective temperaments but the hyperthymic one were involved in the relationship with DT and BD. Only the anxious temperament had a significant unmediated direct effect on DT after the strictest correction for multiple comparisons, while the depressive temperament had a significant direct effect on DT at a less strict significance level. State anxiety was the strongest mediator of the link between affective temperament and core AN body-related psychopathology. Depression showed intermediate results while trait anxiety was not a significant mediator at all.

CONCLUSION

Affective temperaments had a relevant impact on body-related core components of AN; however, a clear direct effect could be identified only for the anxious and depressive temperaments. Also, state anxiety was the strongest mediator thus entailing interesting implications in clinical practice.

LEVEL OF EVIDENCE

V, cross-sectional study.

Associations between plasma levels of brominated flame retardants and methylation of DNA from peripheral blood: A cross-sectional study in a cohort of French women

BACKGROUND

Brominated flame retardants (BFRs) are organic compounds that are widespread in the environment. Because of their persistence, they are able to bioaccumulate with major impacts on human health. It has been hypothesized that the effect of BFRs on human health is mediated by alterations of DNA methylation.

OBJECTIVE

The aim of this study was to examine the association between methylation of DNA extracted from peripheral blood and circulating levels of BFRs measured in plasma.

METHODS

We conducted a methylation wide association study on 336 blood samples from a study within the E3N (Etude Epidémiologique auprès de femmes de l'Education Nationale) cohort, a long-term longitudinal cohort of French women. DNA methylation at more than 850 000 cytosine-guanine dinucleotide (CpG) sites was measured with the Illumina Infinium HumanMethylation - EPIC BeadChip. Circulating levels of seven BFRs (BDE-28, BDE-47, BDE-99, BDE-100, BDE-153, BDE-154 and PBB-153) were measured by gas chromatography coupled to high-resolution mass spectrometry in plasma samples. The association between DNA methylation and BFRs plasma levels was assessed through linear mixed-effects models followed by gene-set enrichment analyses (GSEA).

RESULTS

We identified 253 CpG sites whose methylation levels were significantly associated with exposure to BFRs after Bonferroni correction. For 50 of these CpGs the p-values were less than 2.2x10^-9 with the strongest association being between BDE-154 and cg23619365 (4.32x10^-13). GSEA of CpG sites associated with exposure to BFRs identified significant enrichment of genes involved in hypoxia, glycolysis and adipogenesis.

CONCLUSIONS

Exposure to BFRs appears to be related to numerous alterations in DNA methylation. These findings, if replicated in independent studies, provide insights into the biological and health effects of BFRs.

Application of two statistical approaches (Bayesian Kernel Machine Regression and Principal Component Regression) to assess breast cancer risk in association to exposure to mixtures of brominated flame retardants and per- and polyfluorinated alkylated substances in the E3N cohort

BACKGROUND

Brominated flame retardants (BFR) and per- and polyfluorinated alkylated substances (PFAS) are two groups of substances suspected to act as endocrine disruptors. Such substances could therefore be implicated in the occurrence of breast cancer, nevertheless, previous studies have led to inconstant results. Due to the large correlation between these substances, and the possibly non-linear effects they exert, evaluating their joint impact as mixtures on health remains challenging. This exploratory study aimed to generate hypotheses on the relationship between circulating levels of 7 BFR (6 polybrominated diphenyl ethers and 1 polybrominated biphenyls) and 11 PFAS and the risk of breast cancer in a case-control study nested in the E3N French prospective cohort by performing two methods: Principal Component Regression (PCR) models, and Bayesian Kernel Machine Regression (BKMR) models.

METHODS

194 post-menopausal breast cancer cases and 194 controls were included in the present study. Circulating levels of BFR and PFAS were measured by gas chromatography coupled to high-resolution mass spectrometry and liquid chromatography coupled to tandem mass spectrometry, respectively. The first statistical approach was based on Principal Component Analysis (PCA) followed by logistic regression models that included the identified principal components as main exposure variables. The second approach used BKMR models with hierarchical variable selection, this latter being suitable for highly correlated exposures. Both approaches were also run separately for Estrogen Receptor positive (ER +) and Estrogen Receptor negative (ER-) breast cancer cases.

RESULTS

PCA identified four principal components accounting for 67% of the total variance. Component 3 showed a marginal association with ER + breast cancer risk. No clear association between BFR and PFAS mixtures and breast cancer was identified using BKMR models, and the credible intervals obtained were very wide. Finally, the BKMR models suggested a negative cumulative effect of BFR and PFAS on ER- breast cancer risk, and a positive cumulative effect on ER + breast cancer risk.

CONCLUSION

Although globally no clear association was identified, both approaches suggested a differential effect of BFR and PFAS mixtures on ER + and ER- breast cancer risk. However, the results for ER- breast cancer should be interpreted carefully due to the small number of ER- cases included in the study. Further studies evaluating mixtures of substances on larger study populations are needed.

Hospitalization for COVID-19 is associated with a higher risk of subsequent hospitalization for psychiatric disorders: A French nationwide longitudinal study comparing hospitalizations for COVID-19 and for other reasons

Introduction

Although COVID-19 has been associated with psychiatric symptoms in patients, no study to date has examined the risk of hospitalization for psychiatric disorders after hospitalization for this disease.

Objective

We aimed to compare the proportions of hospitalizations for psychiatric disorders in the 12 months following either hospitalization for COVID-19 or hospitalization for another reason in the adult general population in France during the first wave of the current pandemic.

Methods

We conducted a retrospective longitudinal nationwide study based on the national French administrative healthcare database.

Results

Among the 2,894,088 adults hospitalized, 96,313 (3.32%) were admitted for COVID-19. The proportion of patients subsequently hospitalized for a psychiatric disorder was higher for COVID-19 patients (11.09 vs. 9.24%, OR = 1.20 95%CI 1.18–1.23). Multivariable analyses provided similar results for a psychiatric disorder of any type and for psychotic and anxiety disorders (respectively, aOR = 1.06 95%CI 1.04–1.09, aOR = 1.09 95%CI 1.02–1.17, and aOR = 1.11 95%CI 1.08–1.14). Initial hospitalization for COVID-19 in intensive care units and psychiatric history were associated with a greater risk of subsequent hospitalization for any psychiatric disorder than initial hospitalization for another reason.

Discussion

Compared with hospitalizations for other reasons, hospitalizations for COVID-19 during the first wave of the pandemic in France were associated with a higher risk of hospitalization for a psychiatric disorder during the 12 months following initial discharge. This finding should encourage clinicians to increase the monitoring and assessment of psychiatric symptoms after hospital discharge for COVID-19, and to propose post-hospital care, especially for those treated in intensive care.

Genetic architectures of proximal and distal colorectal cancer are partly distinct

OBJECTIVE

An understanding of the etiologic heterogeneity of colorectal cancer (CRC) is critical for improving precision prevention, including individualized screening recommendations and the discovery of novel drug targets and repurposable drug candidates for chemoprevention. Known differences in molecular characteristics and environmental risk factors among tumors arising in different locations of the colorectum suggest partly distinct mechanisms of carcinogenesis. The extent to which the contribution of inherited genetic risk factors for CRC differs by anatomical subsite of the primary tumor has not been examined.

DESIGN

To identify new anatomical subsite-specific risk loci, we performed genome-wide association study (GWAS) meta-analyses including data of 48 214 CRC cases and 64 159 controls of European ancestry. We characterised effect heterogeneity at CRC risk loci using multinomial modelling.

RESULTS

We identified 13 loci that reached genome-wide significance (p<5×10-8) and that were not reported by previous GWASs for overall CRC risk. Multiple lines of evidence support candidate genes at several of these loci. We detected substantial heterogeneity between anatomical subsites. Just over half (61) of 109 known and new risk variants showed no evidence for heterogeneity. In contrast, 22 variants showed association with distal CRC (including rectal cancer), but no evidence for association or an attenuated association with proximal CRC. For two loci, there was strong evidence for effects confined to proximal colon cancer.

CONCLUSION

Genetic architectures of proximal and distal CRC are partly distinct. Studies of risk factors and mechanisms of carcinogenesis, and precision prevention strategies should take into consideration the anatomical subsite of the tumour.

Association between menopausal hormone therapy, mammographic density and breast cancer risk: results from the E3N cohort study

BACKGROUND

Menopausal hormone therapy (MHT) is a risk factor for breast cancer (BC). Evidence suggests that its effect on BC risk could be partly mediated by mammographic density. The aim of this study was to investigate the relationship between MHT, mammographic density and BC risk using data from a prospective study.

METHODS

We used data from a case-control study nested within the French cohort E3N including 453 cases and 453 matched controls. Measures of mammographic density, history of MHT use during follow-up and information on potential confounders were available for all women. The association between MHT and mammographic density was evaluated by linear regression models. We applied mediation modelling techniques to estimate, under the hypothesis of a causal model, the proportion of the effect of MHT on BC risk mediated by percent mammographic density (PMD) for BC overall and by hormone receptor status.

RESULTS

Among MHT users, 4.2% used exclusively oestrogen alone compared with 68.3% who used exclusively oestrogens plus progestogens. Mammographic density was higher in current users (for a 60-year-old woman, mean PMD 33%; 95% CI 31 to 35%) than in past (29%; 27 to 31%) and never users (24%; 22 to 26%). No statistically significant association was observed between duration of MHT and mammographic density. In past MHT users, mammographic density was negatively associated with time since last use; values similar to those of never users were observed in women who had stopped MHT at least 8 years earlier. The odds ratio of BC for current versus never MHT users, adjusted for age, year of birth, menopausal status at baseline and BMI, was 1.67 (95% CI, 1.04 to 2.68). The proportion of effect mediated by PMD was 34% for any BC and became 48% when the correlation between BMI and PMD was accounted for. These effects were limited to hormone receptor-positive BC.

CONCLUSIONS

Our results suggest that, under a causal model, nearly half of the effect of MHT on hormone receptor-positive BC risk is mediated by mammographic density, which appears to be modified by MHT for up to 8 years after MHT termination.

Reducing socio-economic inequalities in all-cause mortality: a counterfactual mediation approach

Background

Socio-economic inequalities in mortality are well established, yet the contribution of intermediate risk factors that may underlie these relationships remains unclear. We evaluated the role of multiple modifiable intermediate risk factors underlying socio-economic-associated mortality and quantified the potential impact of reducing early all-cause mortality by hypothetically altering socio-economic risk factors.

Methods

Data were from seven cohort studies participating in the LIFEPATH Consortium (total n = 179 090). Using both socio-economic position (SEP) (based on occupation) and education, we estimated the natural direct effect on all-cause mortality and the natural indirect effect via the joint mediating role of smoking, alcohol intake, dietary patterns, physical activity, body mass index, hypertension, diabetes and coronary artery disease. Hazard ratios (HRs) were estimated, using counterfactual natural effect models under different hypothetical actions of either lower or higher SEP or education.

Results

Lower SEP and education were associated with an increase in all-cause mortality within an average follow-up time of 17.5 years. Mortality was reduced via modelled hypothetical actions of increasing SEP or education. Through higher education, the HR was 0.85 [95% confidence interval (CI) 0.84, 0.86] for women and 0.71 (95% CI 0.70, 0.74) for men, compared with lower education. In addition, 34% and 38% of the effect was jointly mediated for women and men, respectively. The benefits from altering SEP were slightly more modest.

Conclusions

These observational findings support policies to reduce mortality both through improving socio-economic circumstances and increasing education, and by altering intermediaries, such as lifestyle behaviours and morbidities.

Adiposity and Endometrial Cancer Risk in Postmenopausal Women: A Sequential Causal Mediation Analysis

BACKGROUND

Adiposity increases endometrial cancer risk, possibly through inflammation, hyperinsulinemia, and increasing estrogens. We aimed to quantify the mediating effects of adiponectin (anti-inflammatory adipocytokine); IL6, IL1-receptor antagonist, TNF receptor 1 and 2, and C-reactive protein (inflammatory status biomarkers); C-peptide (hyperinsulinemia biomarker); and free estradiol and estrone (estrogen biomarkers) in the adiposity-endometrial cancer link in postmenopausal women.

METHODS

We used data from a case-control study within the European Prospective Investigation into Cancer and Nutrition (EPIC). Eligible women did not have cancer, hysterectomy, and diabetes; did not use oral contraceptives or hormone therapy; and were postmenopausal at recruitment. Mediating pathways from adiposity to endometrial cancer were investigated by estimating natural indirect (NIE) and direct (NDE) effects using sequential mediation analysis.

RESULTS

The study included 163 cases and 306 controls. The adjusted OR for endometrial cancer for body mass index (BMI) ≥30 versus ≥18.5-<25 kg/m2 was 2.51 (95% confidence interval, 1.26-5.02). The ORsNIE were 1.95 (1.01-3.74) through all biomarkers [72% proportion mediated (PM)] decomposed as: 1.35 (1.06-1.73) through pathways originating with adiponectin (33% PM); 1.13 (0.71-1.80) through inflammation beyond (the potential influence of) adiponectin (13% PM); 1.05 (0.88-1.24) through C-peptide beyond adiponectin and inflammation (5% PM); and 1.22 (0.89-1.67) through estrogens beyond preceding biomarkers (21% PM). The ORNDE not through biomarkers was 1.29 (0.54-3.09). Waist circumference gave similar results.

CONCLUSIONS

Reduced adiponectin and increased inflammatory biomarkers, C-peptide, and estrogens mediated approximately 70% of increased odds of endometrial cancer in women with obesity versus normal weight.

IMPACT

If replicated, these results could have implications for identifying targets for intervention to reduce endometrial cancer risk in women with obesity.

The impact of lifecourse socio-economic position and individual social mobility on breast cancer risk

BACKGROUND

Women with an advantaged socioeconomic position (SEP) have a higher risk of developing breast cancer (BC). The reasons for this association do not seem to be limited to reproductive factors and remain to be understood. We aimed to investigate the impact of lifecourse SEP from childhood and social mobility on the risk of BC considering a broad set of potential mediators.

METHODS

We used a discovery-replication strategy in two European prospective cohorts, E3N (N = 83,436) and EPIC-Italy (N = 20,530). In E3N, 7877 women were diagnosed with BC during a median 24.4 years of follow-up, while in EPIC-Italy, 893 BC cases were diagnosed within 15.1 years. Hazard ratios (HR) were estimated using Cox proportional hazard models on imputed data.

RESULTS

In E3N, women with higher education had a higher risk of BC (HR [95%CI] = 1.21 [1.12, 1.30]). This association was attenuated by adjusting for reproductive factors, in particular age at first childbirth (HR[95%CI] = 1.13 [1.04, 1.22]). Health behaviours, anthropometric variables, and BC screening had a weaker effect on the association. Women who remained in a stable advantaged SEP had a higher risk of BC (HR [95%CI] = 1.24 [1.07; 1.43]) attenuated after adjustment for potential mediators (HR [95%CI] = 1.13 [0.98; 1.31]). These results were replicated in EPIC-Italy.

CONCLUSIONS

These results confirm the important role of reproductive factors in the social gradient in BC risk, which does not appear to be fully explained by the large set of potential mediators, including cancer screening, suggesting that further research is needed to identify additional mechanisms.

Causal mediation analysis in presence of multiple mediators uncausally related

Mediation analysis aims at disentangling the effects of a treatment on an outcome through alternative causal mechanisms and has become a popular practice in biomedical and social science applications. The causal framework based on counterfactuals is currently the standard approach to mediation, with important methodological advances introduced in the literature in the last decade, especially for simple mediation, that is with one mediator at the time. Among a variety of alternative approaches, Imai et al. showed theoretical results and developed an R package to deal with simple mediation as well as with multiple mediation involving multiple mediators conditionally independent given the treatment and baseline covariates. This approach does not allow to consider the often encountered situation in which an unobserved common cause induces a spurious correlation between the mediators. In this context, which we refer to as mediation with uncausally related mediators, we show that, under appropriate hypothesis, the natural direct and joint indirect effects are non-parametrically identifiable. Moreover, we adopt the quasi-Bayesian algorithm developed by Imai et al. and propose a procedure based on the simulation of counterfactual distributions to estimate not only the direct and joint indirect effects but also the indirect effects through individual mediators. We study the properties of the proposed estimators through simulations. As an illustration, we apply our method on a real data set from a large cohort to assess the effect of hormone replacement treatment on breast cancer risk through three mediators, namely dense mammographic area, nondense area and body mass index.

Systematic review with meta-analysis: comparative risk of lymphoma with anti-tumour necrosis factor agents and/or thiopurines in patients with inflammatory bowel disease

BACKGROUND

The risk of lymphoma in patients with inflammatory bowel disease (IBD) treated with anti-TNF agents remains unclear.

AIM

To assess the comparative risk of lymphoma with anti-TNF agents and/or thiopurines in IBD METHODS: We searched PubMed, EMBASE and Cochrane Library to identify studies that evaluated lymphoproliferative disorders associated with anti-TNF agents with or without thiopurines. The risk of lymphoma was assessed through four comparator groups: combination therapy (anti-TNF plus thiopurine), anti-TNF monotherapy, thiopurine monotherapy and control group. Pooled incidence rate ratios (IRR) were estimated through Poisson-normal models.

RESULTS

Four observational studies comprising 261 689 patients were included. As compared with patients unexposed to anti-TNF and thiopurines, those exposed to anti-TNF monotherapy, thiopurine monotherapy or combination therapy had pooled IRR (per 1000 patient-years) of lymphoma of 1.52 (95% CI: 1.06-2.19; P = 0.023), 2.23 (95% CI: 1.79-2.79; P < 0.001) and 3.71 (95% CI: 2.30-6.00; P ≤ 0.01), respectively. The risk of lymphoma associated with combination therapy was higher than with thiopurines or anti-TNF alone with pooled IRR of 1.70 (95% CI: 1.03-2.81; P = 0.039) and 2.49 (95% CI: 1.39-4.47; P = 0.002), respectively. The risk did not differ between anti-TNF monotherapy and thiopurine monotherapy with pooled IRR of 0.72 (95% CI: 0.48-1.07; P = 0.107). All observational studies were of high quality according to the Newcastle-Ottawa scale.

CONCLUSIONS

There is an increased risk of lymphoma in IBD patients treated with anti-TNF agents, either alone or when combined with thiopurines.

Serum levels of hsa-miR-16-5p, hsa-miR-29a-3p, hsa-miR-150-5p, hsa-miR-155-5p and hsa-miR-223-3p and subsequent risk of chronic lymphocytic leukemia in the EPIC study

Chronic lymphocytic leukemia (CLL) is an incurable disease accounting for almost one-third of leukemias in the Western world. Aberrant expression of microRNAs (miRNAs) is a well-established characteristic of CLL, and the robust nature of miRNAs makes them eminently suitable liquid biopsy biomarkers. Using a nested case-control study within the European Prospective Investigation into Cancer and Nutrition (EPIC), the predictive values of five promising human miRNAs (hsa-miR-16-5p, hsa-miR-29a-3p, hsa-miR-150-5p, hsa-miR-155-5p and hsa-miR-223-3p), identified in a pilot study, were examined in serum of 224 CLL cases (diagnosed 3 months to 18 years after enrollment) and 224 matched controls using Taqman based assays. Conditional logistic regressions were applied to adjust for potential confounders. The median time from blood collection to CLL diagnosis was 10 years (p25-p75: 7-13 years). Overall, the upregulation of hsa-miR-150-5p, hsa-miR-155-5p and hsa-miR-29a-3p was associated with subsequent risk of CLL [OR1∆Ct-unit increase (95%CI) = 1.42 (1.18-1.72), 1.64 (1.31-2.04) and 1.75 (1.31-2.34) for hsa-miR-150-5p, hsa-miR-155-5p and hsa-miR-29a-3p, respectively] and the strongest associations were observed within 10 years of diagnosis. However, the predictive performance of these miRNAs was modest (area under the curve <0.62). hsa-miR-16-5p and hsa-miR-223-3p levels were unrelated to CLL risk. The findings of this first prospective study suggest that hsa-miR-29a, hsa-miR-150-5p and hsa-miR-155-5p were upregulated in early stages of CLL but were modest predictive biomarkers of CLL risk.

Mediating effect of soluble B-cell activation immune markers on the association between anthropometric and lifestyle factors and lymphoma development

Sustained B-cell activation is an important mechanism contributing to B-cell lymphoma (BCL). We aimed to validate four previously reported B-cell activation markers predictive of BCL risk (sCD23, sCD27, sCD30, and CXCL13) and to examine their possible mediating effects on the association between anthropometric and lifestyle factors and major BCL subtypes. Pre-diagnostic serum levels were measured for 517 BCL cases and 525 controls in a nested case-control study. The odds ratios of BCL were 6.2 in the highest versus lowest quartile for sCD23, 2.6 for sCD30, 4.2 for sCD27, and 2.6 for CXCL13. Higher levels of all markers were associated with increased risk of chronic lymphocytic leukemia (CLL), follicular lymphoma (FL), and diffuse large B-cell lymphoma (DLBCL). Following mutual adjustment for the other immune markers, sCD23 remained associated with all subtypes and CXCL13 with FL and DLBCL. The associations of sCD23 with CLL and DLBCL and CXCL13 with DLBCL persisted among cases sampled > 9 years before diagnosis. sCD23 showed a good predictive ability (area under the curve = 0.80) for CLL, in particular among older, male participants. sCD23 and CXCL13 showed a mediating effect between body mass index (positive) and DLBCL risk, while CXCL13 contributed to the association between physical activity (inverse) and DLBCL. Our data suggest a role of B-cell activation in BCL development and a mediating role of the immune system for lifestyle factors.

Predicting breast cancer metastasis from whole-blood transcriptomic measurements

Objective

In this exploratory work we investigate whether blood gene expression measurements predict breast cancer metastasis. Early detection of increased metastatic risk could potentially be life-saving. Our data comes from the Norwegian Women and Cancer epidemiological cohort study. The women who contributed to these data provided a blood sample up to a year before receiving a breast cancer diagnosis. We estimate a penalized maximum likelihood logistic regression. We evaluate this in terms of calibration, concordance probability, and stability, all of which we estimate by the bootstrap.

Results

We identify a set of 108 candidate predictor genes that exhibit a fold change in average metastasized observation where there is none for the average non-metastasized observation.

Circulating Levels of Insulin-like Growth Factor 1 and Insulin-like Growth Factor Binding Protein 3 Associate With Risk of Colorectal Cancer Based on Serologic and Mendelian Randomization Analyses

BACKGROUND & AIMS

Human studies examining associations between circulating levels of insulin-like growth factor 1 (IGF1) and insulin-like growth factor binding protein 3 (IGFBP3) and colorectal cancer risk have reported inconsistent results. We conducted complementary serologic and Mendelian randomization (MR) analyses to determine whether alterations in circulating levels of IGF1 or IGFBP3 are associated with colorectal cancer development.

METHODS

Serum levels of IGF1 were measured in blood samples collected from 397,380 participants from the UK Biobank, from 2006 through 2010. Incident cancer cases and cancer cases recorded first in death certificates were identified through linkage to national cancer and death registries. Complete follow-up was available through March 31, 2016. For the MR analyses, we identified genetic variants associated with circulating levels of IGF1 and IGFBP3. The association of these genetic variants with colorectal cancer was examined with 2-sample MR methods using genome-wide association study consortia data (52,865 cases with colorectal cancer and 46,287 individuals without [controls]) RESULTS: After a median follow-up period of 7.1 years, 2665 cases of colorectal cancer were recorded. In a multivariable-adjusted model, circulating level of IGF1 associated with colorectal cancer risk (hazard ratio per 1 standard deviation increment of IGF1, 1.11; 95% confidence interval [CI] 1.05-1.17). Similar associations were found by sex, follow-up time, and tumor subsite. In the MR analyses, a 1 standard deviation increment in IGF1 level, predicted based on genetic factors, was associated with a higher risk of colorectal cancer risk (odds ratio 1.08; 95% CI 1.03-1.12; P = 3.3 × 10-4). Level of IGFBP3, predicted based on genetic factors, was associated with colorectal cancer risk (odds ratio per 1 standard deviation increment, 1.12; 95% CI 1.06-1.18; P = 4.2 × 10-5). Colorectal cancer risk was associated with only 1 variant in the IGFBP3 gene region (rs11977526), which also associated with anthropometric traits and circulating level of IGF2.

CONCLUSIONS

In an analysis of blood samples from almost 400,000 participants in the UK Biobank, we found an association between circulating level of IGF1 and colorectal cancer. Using genetic data from 52,865 cases with colorectal cancer and 46,287 controls, a higher level of IGF1, determined by genetic factors, was associated with colorectal cancer. Further studies are needed to determine how this signaling pathway might contribute to colorectal carcinogenesis.

Mitochondrial DNA Copy-Number Variation and Pancreatic Cancer Risk in the Prospective EPIC Cohort

BACKGROUND

Mitochondrial DNA (mtDNA) copy number in peripheral blood has been found to be associated with risk of developing several cancers. However, data on pancreatic ductal adenocarcinoma (PDAC) are very limited.

METHODS

To further our knowledge on this topic, we measured relative mtDNA copy number by a quantitative real-time PCR assay in peripheral leukocyte samples of 476 PDAC cases and 357 controls nested within the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.

RESULTS

We observed lower mtDNA copy number with advancing age (P = 6.54 × 10-5) and with a high body mass index (BMI) level (P = 0.004) and no association with sex, smoking behavior, and alcohol consumption. We found an association between increased mtDNA copy number and decreased risk of developing PDAC with an odds ratios (OR) of 0.35 [95% confidence interval (CI), 0.16-0.79; P = 0.01] when comparing the fifth quintile with the first using an unconditional logistic regression and an OR of 0.19 (95% CI, 0.07-0.52; P = 0.001) with a conditional analysis. Analyses stratified by BMI showed an association between high mtDNA copy number and decreased risk in the stratum of normal weight, consistent with the main analyses.

CONCLUSIONS

Our results suggest a protective effect of a higher number of mitochondria, measured in peripheral blood leukocytes, on PDAC risk.

IMPACT

Our findings highlight the importance of understanding the mitochondrial biology in pancreatic cancer.

Physical activity and risks of breast and colorectal cancer: a Mendelian randomisation analysis

Physical activity has been associated with lower risks of breast and colorectal cancer in epidemiological studies; however, it is unknown if these associations are causal or confounded. In two-sample Mendelian randomisation analyses, using summary genetic data from the UK Biobank and GWA consortia, we found that a one standard deviation increment in average acceleration was associated with lower risks of breast cancer (odds ratio [OR]: 0.51, 95% confidence interval [CI]: 0.27 to 0.98, P-value = 0.04) and colorectal cancer (OR: 0.66, 95% CI: 0.48 to 0.90, P-value = 0.01). We found similar magnitude inverse associations for estrogen positive (ER+ve) breast cancer and for colon cancer. Our results support a potentially causal relationship between higher physical activity levels and lower risks of breast cancer and colorectal cancer. Based on these data, the promotion of physical activity is probably an effective strategy in the primary prevention of these commonly diagnosed cancers.

Systematic review with meta-analysis: mortality in acute severe ulcerative colitis

BACKGROUND

Acute severe ulcerative colitis (ASUC) is a life-threatening condition. Mortality in ASUC decreased in published series but there is uncertainty as to whether this also applies to the real-life setting.

AIM

To perform a systematic review and meta-analysis of mortality in ASUC in studies from referral centres and in population-based studies, separately and combined. A second aim was to identify risk factors of mortality in ASUC.

METHODS

We searched pubmed and embase from 1998 to 2016, to identify studies that reported 3-month or 12-month mortalities of acute UC in adult patients treated in referral centres, and in population-based studies.

RESULTS

Six population-based studies with 741 743 patients and 47 referral centre-based studies with 2556 patients were included. The pooled 3-month and 12-month mortalities were respectively 0.84% and 1.01%. Advanced age was significantly associated with both 3 month and 12 month mortalities (OR = 1.15 per year, 95% CI: 1.10-1.20 and OR = 1.19 per year, 95% CI: 1.15-1.23 respectively). The pooled 3-month and 12-month mortalities were 0.78% and 0.85% in studies with median age of less than 50 and 2.81% and 4.17% in studies with median age of 50 or more, respectively. After adjustment for age, 3-month and 12-month mortalities did not differ between population-based and referral centre-based studies.

CONCLUSIONS

Mortality in acute severe ulcerative colitis is approximately 1%; it is higher in older patients. Efforts should be made to improve the care of elderly patients with severe UC.

Appraising the causal relevance of DNA methylation for risk of lung cancer

BACKGROUND

DNA methylation changes in peripheral blood have recently been identified in relation to lung cancer risk. Some of these changes have been suggested to mediate part of the effect of smoking on lung cancer. However, limitations with conventional mediation analyses mean that the causal nature of these methylation changes has yet to be fully elucidated.

METHODS

We first performed a meta-analysis of four epigenome-wide association studies (EWAS) of lung cancer (918 cases, 918 controls). Next, we conducted a two-sample Mendelian randomization analysis, using genetic instruments for methylation at CpG sites identified in the EWAS meta-analysis, and 29 863 cases and 55 586 controls from the TRICL-ILCCO lung cancer consortium, to appraise the possible causal role of methylation at these sites on lung cancer.

RESULTS

Sixteen CpG sites were identified from the EWAS meta-analysis [false discovery rate (FDR) < 0.05], for 14 of which we could identify genetic instruments. Mendelian randomization provided little evidence that DNA methylation in peripheral blood at the 14 CpG sites plays a causal role in lung cancer development (FDR > 0.05), including for cg05575921-AHRR where methylation is strongly associated with both smoke exposure and lung cancer risk.

CONCLUSIONS

The results contrast with previous observational and mediation analysis, which have made strong claims regarding the causal role of DNA methylation. Thus, previous suggestions of a mediating role of methylation at sites identified in peripheral blood, such as cg05575921-AHRR, could be unfounded. However, this study does not preclude the possibility that differential DNA methylation at other sites is causally involved in lung cancer development, especially within lung tissue.

Computational tools to detect signatures of mutational processes in DNA from tumours: A review and empirical comparison of performance

Mutational signatures refer to patterns in the occurrence of somatic mutations that might be uniquely ascribed to particular mutational process. Tumour mutation catalogues can reveal mutational signatures but are often consistent with the mutation spectra produced by a variety of mutagens. To date, after the analysis of tens of thousands of exomes and genomes from about 40 different cancer types, tens of mutational signatures characterized by a unique probability profile across the 96 trinucleotide-based mutation types have been identified, validated and catalogued. At the same time, several concurrent methods have been developed for either the quantification of the contribution of catalogued signatures in a given cancer sequence or the identification of new signatures from a sample of cancer sequences. A review of existing computational tools has been recently published to guide researchers and practitioners through their mutational signature analyses, but other tools have been introduced since its publication and, a systematic evaluation and comparison of the performance of such tools is still lacking. In order to fill this gap, we have carried out an empirical evaluation of the main packages available to date, using both real and simulated data. Among other results, our empirical study shows that the identification of signatures is more difficult for cancers characterized by multiple signatures each having a small contribution. This work suggests that detection methods based on probabilistic models, especially EMu and bayesNMF, have in general better performance than NMF-based methods.

Efficacy and safety of pharmacological treatments for patent ductus arteriosus closure: A systematic review and network meta-analysis of clinical trials and observational studies

Efficacy and safety profiles of different pharmacological interventions used to treat patent ductus arteriosus (PDA) are relatively unexplored. Integrating the findings of randomized clinical trials (RCTs) with those from observational studies may provide key evidence on this important issue. We aimed at estimating the relative likelihood of failure to close the PDA, need for surgical closure, and occurrence of adverse events among preterm and full-term infants treated with indomethacin, ibuprofen, or acetaminophen, placebo, or no treatment including both RCTs and observational studies. We searched PubMed, Embase, and the Register of Controlled Trials from inception to October 30, 2018. We first estimated proportions of subjects with failure to close the PDA, subjects in whom surgical closure was performed after pharmacological treatment, death, and subjects with selected adverse events (AEs). These estimates were obtained using frequentist random-effect meta-analysis of arm-specific proportions. We then compared active drugs with each other and with control (either placebo or no treatment) by summarizing results at the end of treatment reported in the papers, regardless of number of administration(s), dose, route and type of administration, and study design and quality. We also summarized primary outcome results separately at first, second and third cycles of treatment. These estimates were obtained using Bayesian random-effects network meta-analysis for mixed comparisons, and frequentist random-effect pairwise meta-analysis for direct comparisons. We included 64 RCTs and 24 observational studies including 14,568 subjects (5339 in RCTs and 9229 in observational studies, 8292 subjects received indomethacin, 4761 ibuprofen, 574 acetaminophen, and 941 control (including placebo or no intervention).The proportion of subjects with failure to close the PDA was 0.24 (95% Confidence Interval, CI: 0.20, 0.29) for indomethacin, 0.18 (0.14, 0.22) for ibuprofen, 0.19 (0.09, 0.30) for acetaminophen, and 0.59 (0.48, 0.69) for control. At end of treatment, compared to control, we found inverse associations between all active drugs and failure to close PDA (for indomethacin Odds Ratio, OR, was 0.17 [95% Credible Interval, CrI: 0.11-0.24], ibuprofen 0.19 [0.12-0.28], and acetaminophen 0.15 [0.09-0.26]), without differences among active drugs. We showed inverse associations between effective drugs and need for surgical closure, as compared to control (for indomethacin OR was 0.28 [0.15-0.50], ibuprofen 0.30 [0.16-0.54], and acetaminophen 0.19 [0.07-0.46]), without differences among drugs. Indomethacin was directly associated with intraventricular hemorrhage (IVH) (1.27; 1.00, 1.62) compared to ibuprofen, and to oliguria as compared to ibuprofen (3.92; 1.69, 9.82) or acetaminophen (10.8; 1.86, 93.1). In conclusion, active pharmacological treatment, with indomethacin, ibuprofen, or acetaminophen, is inversely associated with failure to close the PDA compared to non-treatment. Ibuprofen should be preferred to indomethacin to avoid occurrence of IVH or oliguria, acetaminophen should be preferred to indomethacin to avoid oliguria.

Vitamin D-Related Genes, Blood Vitamin D Levels and Colorectal Cancer Risk in Western European Populations

Higher circulating 25-hydroxyvitamin D levels (25(OH)D) have been found to be associated with lower risk for colorectal cancer (CRC) in prospective studies. Whether this association is modified by genetic variation in genes related to vitamin D metabolism and action has not been well studied in humans. We investigated 1307 functional and tagging single-nucleotide polymorphisms (SNPs; individually, and by gene/pathway) in 86 vitamin D-related genes in 1420 incident CRC cases matched to controls from the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. We also evaluated the association between these SNPs and circulating 25(OH)D in a subset of controls. We confirmed previously reported CRC risk associations between SNPs in the VDR, GC, and CYP27B1 genes. We also identified additional associations with 25(OH)D, as well as CRC risk, and several potentially novel SNPs in genes related to vitamin D transport and action (LRP2, CUBN, NCOA7, and HDAC9). However, none of these SNPs were statistically significant after Benjamini-Hochberg (BH) multiple testing correction. When assessed by a priori defined functional pathways, tumor growth factor β (TGFβ) signaling was associated with CRC risk (P ≤ 0.001), with most statistically significant genes being SMAD7 (PBH = 0.008) and SMAD3 (PBH = 0.008), and 18 SNPs in the vitamin D receptor (VDR) binding sites (P = 0.036). The 25(OH)D-gene pathway analysis suggested that genetic variants in the genes related to VDR complex formation and transcriptional activity are associated with CRC depending on 25(OH)D levels (interaction P = 0.041). Additional studies in large populations and consortia, especially with measured circulating 25(OH)D, are needed to confirm our findings.

Stem cell replication, somatic mutations and role of randomness in the development of cancer

An intense scientific debate has recently taken place relating to the "bad luck" hypothesis in cancer development, namely that intrinsic random, and therefore unavoidable, mutagenic events would have a predominant role in tumorigenesis. In this article we review the main contributions to this debate and explain the reasons why the claim that cancer is mostly explained by intrinsic random factors is unsupported by data and theoretical models. In support of this, we present an analysis showing that smoking-induced mutations are more predictive of cancer risk than the lifetime number of stem cell cellular divisions.

Epigenome-wide association study for lifetime estrogen exposure identifies an epigenetic signature associated with breast cancer risk

BACKGROUND

It is well established that estrogens and other hormonal factors influence breast cancer susceptibility. We hypothesized that a woman's total lifetime estrogen exposure accumulates changes in DNA methylation, detectable in the blood, which could be used in risk assessment for breast cancer.

METHODS

An estimated lifetime estrogen exposure (ELEE) model was defined using epidemiological data from EPIC-Italy (n = 31,864). An epigenome-wide association study (EWAS) of ELEE was performed using existing Illumina HumanMethylation450K Beadchip (HM450K) methylation data obtained from EPIC-Italy blood DNA samples (n = 216). A methylation index (MI) of ELEE based on 31 CpG sites was developed using HM450K data from EPIC-Italy and the Generations Study and evaluated for association with breast cancer risk in an independent dataset from the Generations Study (n = 440 incident breast cancer cases matched to 440 healthy controls) using targeted bisulfite sequencing. Lastly, a meta-analysis was conducted including three additional cohorts, consisting of 1187 case-control pairs.

RESULTS

We observed an estimated 5% increase in breast cancer risk per 1-year longer ELEE (OR = 1.05, 95% CI 1.04-1.07, P = 3 × 10-12) in EPIC-Italy. The EWAS identified 694 CpG sites associated with ELEE (FDR Q < 0.05). We report a DNA methylation index (MI) associated with breast cancer risk that is validated in the Generations Study targeted bisulfite sequencing data (ORQ4_vs_Q1 = 1.77, 95% CI 1.07-2.93, P = 0.027) and in the meta-analysis (ORQ4_vs_Q1 = 1.43, 95% CI 1.05-2.00, P = 0.024); however, the correlation between the MI and ELEE was not validated across study cohorts.

CONCLUSION

We have identified a blood DNA methylation signature associated with breast cancer risk in this study. Further investigation is required to confirm the interaction between estrogen exposure and DNA methylation in the blood.

Association of Selenoprotein and Selenium Pathway Genotypes with Risk of Colorectal Cancer and Interaction with Selenium Status

Selenoprotein genetic variations and suboptimal selenium (Se) levels may contribute to the risk of colorectal cancer (CRC) development. We examined the association between CRC risk and genotype for single nucleotide polymorphisms (SNPs) in selenoprotein and Se metabolic pathway genes. Illumina Goldengate assays were designed and resulted in the genotyping of 1040 variants in 154 genes from 1420 cases and 1421 controls within the European Prospective Investigation into Cancer and Nutrition (EPIC) study. Multivariable logistic regression revealed an association of 144 individual SNPs from 63 Se pathway genes with CRC risk. However, regarding the selenoprotein genes, only TXNRD1 rs11111979 retained borderline statistical significance after adjustment for correlated tests (PACT = 0.10; PACT significance threshold was P < 0.1). SNPs in Wingless/Integrated (Wnt) and Transforming growth factor (TGF) beta-signaling genes (FRZB, SMAD3, SMAD7) from pathways affected by Se intake were also associated with CRC risk after multiple testing adjustments. Interactions with Se status (using existing serum Se and Selenoprotein P data) were tested at the SNP, gene, and pathway levels. Pathway analyses using the modified Adaptive Rank Truncated Product method suggested that genes and gene x Se status interactions in antioxidant, apoptosis, and TGF-beta signaling pathways may be associated with CRC risk. This study suggests that SNPs in the Se pathway alone or in combination with suboptimal Se status may contribute to CRC development.

Discovery of common and rare genetic risk variants for colorectal cancer

To further dissect the genetic architecture of colorectal cancer (CRC), we performed whole-genome sequencing of 1,439 cases and 720 controls, imputed discovered sequence variants and Haplotype Reference Consortium panel variants into genome-wide association study data, and tested for association in 34,869 cases and 29,051 controls. Findings were followed up in an additional 23,262 cases and 38,296 controls. We discovered a strongly protective 0.3% frequency variant signal at CHD1. In a combined meta-analysis of 125,478 individuals, we identified 40 new independent signals at P < 5 × 10-8, bringing the number of known independent signals for CRC to ~100. New signals implicate lower-frequency variants, Krüppel-like factors, Hedgehog signaling, Hippo-YAP signaling, long noncoding RNAs and somatic drivers, and support a role for immune function. Heritability analyses suggest that CRC risk is highly polygenic, and larger, more comprehensive studies enabling rare variant analysis will improve understanding of biology underlying this risk and influence personalized screening strategies and drug development.

KIM-1 as a Blood-Based Marker for Early Detection of Kidney Cancer: A Prospective Nested Case-Control Study

Purpose: Renal cell carcinoma (RCC) has the potential for cure with surgery when diagnosed at an early stage. Kidney injury molecule-1 (KIM-1) has been shown to be elevated in the plasma of RCC patients. We aimed to test whether plasma KIM-1 could represent a means of detecting RCC prior to clinical diagnosis.Experimental Design: KIM-1 concentrations were measured in prediagnostic plasma from 190 RCC cases and 190 controls nested within a population-based prospective cohort study. Cases had entered the cohort up to 5 years before diagnosis, and controls were matched on cases for date of birth, date at blood donation, sex, and country. We applied conditional logistic regression and flexible parametric survival models to evaluate the association between plasma KIM-1 concentrations and RCC risk and survival.Results: The incidence rate ratio (IRR) of RCC for a doubling in KIM-1 concentration was 1.71 [95% confidence interval (CI), 1.44-2.03, P = 4.1 × 10-23], corresponding to an IRR of 63.3 (95% CI, 16.2-246.9) comparing the 80th to the 20th percentiles of the KIM-1 distribution in this sample. Compared with a risk model including known risk factors of RCC (age, sex, country, body mass index, and tobacco smoking status), a risk model additionally including KIM-1 substantially improved discrimination between cases and controls (area under the receiver-operating characteristic curve of 0.8 compared with 0.7). High plasma KIM-1 concentrations were also associated with poorer survival (P = 0.0053).Conclusions: Plasma KIM-1 concentrations could predict RCC incidence up to 5 years prior to diagnosis and were associated with poorer survival. Clin Cancer Res; 24(22); 5594-601. ©2018 AACR.

Identifying and correcting epigenetics measurements for systematic sources of variation

Background

Methylation measures quantified by microarray techniques can be affected by systematic variation due to the technical processing of samples, which may compromise the accuracy of the measurement process and contribute to bias the estimate of the association under investigation. The quantification of the contribution of the systematic source of variation is challenging in datasets characterized by hundreds of thousands of features.

In this study, we introduce a method previously developed for the analysis of metabolomics data to evaluate the performance of existing normalizing techniques to correct for unwanted variation. Illumina Infinium HumanMethylation450K was used to acquire methylation levels in over 421,000 CpG sites for 902 study participants of a case-control study on breast cancer nested within the EPIC cohort. The principal component partial R-square (PC-PR2) analysis was used to identify and quantify the variability attributable to potential systematic sources of variation. Three correcting techniques, namely ComBat, surrogate variables analysis (SVA) and a linear regression model to compute residuals were applied. The impact of each correcting method on the association between smoking status and DNA methylation levels was evaluated, and results were compared with findings from a large meta-analysis.

Results

A sizeable proportion of systematic variability due to variables expressing ‘batch’ and ‘sample position’ within ‘chip’ was identified, with values of the partial R² statistics equal to 9.5 and 11.4% of total variation, respectively. After application of ComBat or the residuals’ methods, the contribution was 1.3 and 0.2%, respectively. The SVA technique resulted in a reduced variability due to ‘batch’ (1.3%) and ‘sample position’ (0.6%), and in a diminished variability attributable to ‘chip’ within a batch (0.9%). After ComBat or the residuals’ corrections, a larger number of significant sites (k = 600 and k = 427, respectively) were associated to smoking status than the SVA correction (k = 96).

Conclusions

The three correction methods removed systematic variation in DNA methylation data, as assessed by the PC-PR2, which lent itself as a useful tool to explore variability in large dimension data. SVA produced more conservative findings than ComBat in the association between smoking and DNA methylation.

Electronic supplementary material

The online version of this article (10.1186/s13148-018-0471-6) contains supplementary material, which is available to authorized users.

Genealogy of breastfeeding

Decline and resurgence of breastfeeding (BF) characterized last century. Several factors influencing BF outcome were identified. Despite the huge literature on BF, no data on its matrilineal transmission are available. BF practice was prospectively followed in 2546 Italian mothers. Lactation and BF outcome were related, besides to known factors interfering with BF, to the occurrence of previous maternal and paternal BF. Recalls of grandmaternal and grand-grandmaternal BF behaviours allowed the construction of familiar pedigrees of BF across three generations. Having been breastfed was the strongest factor addressing successful BF establishment (odds ratio (OR) 9.33; 95% confidence interval (CI) 7.40-11.84; p < 0.0001) and BF duration (at 6 months: OR 3.79; 95% CI 3.11-4.64; p < 0.0001). The hazard ratio for breastfed vs non-breastfed mothers was 0.46 (95% CI 0.41-0.50; log-rank p < 0.0001). The rate of BF failures was fivefold higher in non-breastfed mothers, mostly occurring during lactogenesis when the let-down reflex becomes essential.

CONCLUSION

At any generation, mothers are likely to have daughters repeating their BF experience. Differently from the intergenerational effects of environmental factors responsible for the BF secular trend, this trait is transgenerationally transmitted and reversible, with temporal and clinical features of lactation failure. Accordingly, we speculate that epigenetic mechanisms might alter offspring's oxytocinergic receptor signalling.

WHAT IS KNOWN

Several cultural and socio-demographic factors are known to influence breastfeeding outcome. The generational effects of breastfeeding itself have not been investigated so far.

WHAT IS NEW

Maternal breastfeeding is the most important factor addressing daughters' breastfeeding outcome. This behavior is transmitted transgenerationally, with features suggesting epigenetic mechanisms.

Alternative methods for H1 simulations in genome-wide association studies

OBJECTIVE

Assessing the statistical power to detect susceptibility variants plays a critical role in genome-wide association (GWA) studies both from the prospective and retrospective point of view. Power is empirically estimated by simulating phenotypes under a disease model H1. For this purpose, the gold standard consists in simulating genotypes given the phenotypes (e.g. Hapgen). We introduce here an alternative approach for simulating phenotypes under H1 that does not require generating new genotypes for each simulation.

METHODS

In order to simulate phenotypes with a fixed total number of cases and under a given disease model, we suggest 3 algorithms: (1) a simple rejection algorithm; (2) a numerical Markov chain Monte-Carlo (MCMC) approach, and (3) an exact and efficient backward sampling algorithm. In our study, we validated the 3 algorithms both on a simulated dataset and by comparing them with Hapgen on a more realistic dataset. For an application, we then conducted a simulation study on a 1000 Genomes Project dataset consisting of 629 individuals (314 cases) and 8,048 SNPs from chromosome X. We arbitrarily defined an additive disease model with two susceptibility SNPs and an epistatic effect.

RESULTS

The 3 algorithms are consistent, but backward sampling is dramatically faster than the other two. Our approach also gives consistent results with Hapgen. Using our application data, we showed that our limited design requires a biological a priori to limit the investigated region. We also proved that epistatic effects can play a significant role even when simple marker statistics (e.g. trend) are used. We finally showed that the overall performance of a GWA study strongly depends on the prevalence of the disease: the larger the prevalence, the better the power.

CONCLUSIONS

Our approach is a valid alternative to Hapgen-type methods; it is not only dramatically faster but has 2 main advantages: (1) there is no need for sophisticated genotype models (e.g. haplotype frequencies, or recombination rates), and (2) the choice of the disease model is completely unconstrained (number of SNPs involved, gene-environment interactions, hybrid genetic models, etc.). Our 3 algorithms are available in an R package called 'waffect' ('double-u affect', for weighted affectations).