Person:

Kraft, Phillip

An Essay by A. Cecile Janssens and Peter Kraft discusses the limitations inherent in research involving collection of self-reported data by self-selected participants, and makes proposals for upfront communication of such limitations to study participants.

Early menopause (EM) affects up to 10% of the female population, reducing reproductive lifespan considerably. Currently, it constitutes the leading cause of infertility in the western world, affecting mainly those women who postpone their first pregnancy beyond the age of 30 years. The genetic aetiology of EM is largely unknown in the majority of cases. We have undertaken a meta-analysis of genome-wide association studies (GWASs) in 3493 EM cases and 13 598 controls from 10 independent studies. No novel genetic variants were discovered, but the 17 variants previously associated with normal age at natural menopause as a quantitative trait (QT) were also associated with EM and primary ovarian insufficiency (POI). Thus, EM has a genetic aetiology which overlaps variation in normal age at menopause and is at least partly explained by the additive effects of the same polygenic variants. The combined effect of the common variants captured by the single nucleotide polymorphism arrays was estimated to account for ∼30% of the variance in EM. The association between the combined 17 variants and the risk of EM was greater than the best validated non-genetic risk factor, smoking.

Diffuse large B-cell lymphoma (DLBCL) is the most common lymphoma subtype and is clinically aggressive. To identify genetic susceptibility loci for DLBCL, we conducted a meta-analysis of three new genome-wide association studies (GWAS) and one prior scan, totaling 3,857 cases and 7,666 controls of European ancestry, with additional genotyping of nine promising SNPs in 1,359 cases and 4,557 controls. In our multi-stage analysis, five independent SNPs in four loci achieved genome-wide significance marked by rs116446171 at 6p25.3 (EXOC2; (P=2.33x10^{-21})), rs2523607 at 6p21.33 (HLA-B; (2.40x10^{-10})), rs79480871 at 2p23.3 (NCOA1; (P=4.23x10^{-8})), and two independent SNPs, rs13255292 and rs4733601, at 8q24.21 (PVT1; (P=9.98x10^{-13}) and (P=3.63x10^{-11}), respectively). These data provide substantial new evidence for genetic susceptibility to this B-cell malignancy, and point towards pathways involved in immune recognition and immune function in the pathogenesis of DLBCL.

Purpose We developed an absolute risk model to identify individuals in the general population at elevated risk of pancreatic cancer. Patients and Methods Using data on 3,349 cases and 3,654 controls from the PanScan Consortium, we developed a relative risk model for men and women of European ancestry based on non-genetic and genetic risk factors for pancreatic cancer. We estimated absolute risks based on these relative risks and population incidence rates. Results: Our risk model included current smoking (multivariable adjusted odds ratio (OR) and 95% confidence interval: 2.20 [1.84–2.62]), heavy alcohol use (>3 drinks/day) (OR: 1.45 [1.19–1.76]), obesity (body mass index >30 kg/m2) (OR: 1.26 [1.09–1.45]), diabetes >3 years (nested case-control OR: 1.57 [1.13–2.18], case-control OR: 1.80 [1.40–2.32]), family history of pancreatic cancer (OR: 1.60 [1.20–2.12]), non-O ABO genotype (AO vs. OO genotype) (OR: 1.23 [1.10–1.37]) to (BB vs. OO genotype) (OR 1.58 [0.97–2.59]), rs3790844(chr1q32.1) (OR: 1.29 [1.19–1.40]), rs401681(5p15.33) (OR: 1.18 [1.10–1.26]) and rs9543325(13q22.1) (OR: 1.27 [1.18–1.36]). The areas under the ROC curve for risk models including only non-genetic factors, only genetic factors, and both non-genetic and genetic factors were 58%, 57% and 61%, respectively. We estimate that fewer than 3/1,000 U.S. non-Hispanic whites have more than a 5% predicted lifetime absolute risk. Conclusion: Although absolute risk modeling using established risk factors may help to identify a group of individuals at higher than average risk of pancreatic cancer, the immediate clinical utility of our model is limited. However, a risk model can increase awareness of the various risk factors for pancreatic cancer, including modifiable behaviors.

Genome-wide association studies (GWAS) have identified common variants that predispose individuals to a higher body mass index (BMI), an independent risk factor for endometrial cancer. Composite genotype risk scores (GRS) based on the joint effect of published BMI risk loci were used to explore whether endometrial cancer shares a genetic background with obesity. Genotype and risk factor data were available on 3,376 endometrial cancer case and 3,867 control participants of European ancestry from the Epidemiology of Endometrial Cancer Consortium GWAS. A BMI GRS was calculated by summing the number of BMI risk alleles at 97 independent loci. For exploratory analyses, additional GRSs were based on subsets of risk loci within putative etiologic BMI pathways. The BMI GRS was statistically significantly associated with endometrial cancer risk (P = 0.002). For every 10 BMI risk alleles a woman had a 13% increased endometrial cancer risk (95% CI: 4%, 22%). However, after adjusting for BMI, the BMI GRS was no longer associated with risk (per 10 BMI risk alleles OR = 0.99, 95% CI: 0.91, 1.07; P = 0.78). Heterogeneity by BMI did not reach statistical significance (P = 0.06), and no effect modification was noted by age, GWAS Stage, study design or between studies (P≥0.58). In exploratory analyses, the GRS defined by variants at loci containing monogenic obesity syndrome genes was associated with reduced endometrial cancer risk independent of BMI (per BMI risk allele OR = 0.92, 95% CI: 0.88, 0.96; P = 2.1 x 10−5). Possessing a large number of BMI risk alleles does not increase endometrial cancer risk above that conferred by excess body weight among women of European descent. Thus, the GRS based on all current established BMI loci does not provide added value independent of BMI. Future studies are required to validate the unexpected observed relation between monogenic obesity syndrome genetic variants and endometrial cancer risk.

Epidemiological studies have reported inconsistent associations between telomere length (TL) and risk for various cancers. These inconsistencies are likely attributable, in part, to biases that arise due to post-diagnostic and post-treatment TL measurement. To avoid such biases, we used a Mendelian randomization approach and estimated associations between nine TL-associated SNPs and risk for five common cancer types (breast, lung, colorectal, ovarian and prostate cancer, including subtypes) using data on 51 725 cases and 62 035 controls. We then used an inverse-variance weighted average of the SNP-specific associations to estimate the association between a genetic score representing long TL and cancer risk. The long TL genetic score was significantly associated with increased risk of lung adenocarcinoma (P = 6.3 × 10−15), even after exclusion of a SNP residing in a known lung cancer susceptibility region (TERT-CLPTM1L) P = 6.6 × 10−6). Under Mendelian randomization assumptions, the association estimate [odds ratio (OR) = 2.78] is interpreted as the OR for lung adenocarcinoma corresponding to a 1000 bp increase in TL. The weighted TL SNP score was not associated with other cancer types or subtypes. Our finding that genetic determinants of long TL increase lung adenocarcinoma risk avoids issues with reverse causality and residual confounding that arise in observational studies of TL and disease risk. Under Mendelian randomization assumptions, our finding suggests that longer TL increases lung adenocarcinoma risk. However, caution regarding this causal interpretation is warranted in light of the potential issue of pleiotropy, and a more general interpretation is that SNPs influencing telomere biology are also implicated in lung adenocarcinoma risk.

Objective: Many confirmed genetic loci for obesity are expressed in regions of the brain that regulate energy intake and reward-seeking behavior. Whether these loci contribute to the development of specific eating behaviors has not been investigated. We examined the relationship between a genetic susceptibility to obesity and cognitive restraint, uncontrolled and emotional eating. Design and Methods Eating behavior and body mass index (BMI) were determined by questionnaires for 1471 men and 2381 women from two U.S cohorts. Genotypes were extracted from genome-wide scans and a genetic-risk score (GRS) derived from 32 obesity-loci was calculated. Results: The GRS was positively associated with emotional and uncontrolled eating(P<0.002). In exploratory analysis, BMI-increasing variants of MTCH2, TNNI3K and ZC3H4 were positively associated with emotional eating and those of TNNI3K and ZC3H4 were positively associated with uncontrolled eating. The BMI-increasing variant of FTO was positively and those of LRP1B and TFAP2B were inversely associated with cognitive restraint. These associations for single SNPs were independent of BMI but were not significant after multiple-testing correction. Conclusions: An overall genetic susceptibility to obesity may also extend to eating behaviors. The link between specific loci and obesity may be mediated by eating behavior but larger studies are warranted to confirm these results.

Standard statistical approaches for prioritization of variants for functional testing in fine-mapping studies either use marginal association statistics or estimate posterior probabilities for variants to be causal under simplifying assumptions. Here, we present a probabilistic framework that integrates association strength with functional genomic annotation data to improve accuracy in selecting plausible causal variants for functional validation. A key feature of our approach is that it empirically estimates the contribution of each functional annotation to the trait of interest directly from summary association statistics while allowing for multiple causal variants at any risk locus. We devise efficient algorithms that estimate the parameters of our model across all risk loci to further increase performance. Using simulations starting from the 1000 Genomes data, we find that our framework consistently outperforms the current state-of-the-art fine-mapping methods, reducing the number of variants that need to be selected to capture 90% of the causal variants from an average of 13.3 to 10.4 SNPs per locus (as compared to the next-best performing strategy). Furthermore, we introduce a cost-to-benefit optimization framework for determining the number of variants to be followed up in functional assays and assess its performance using real and simulation data. We validate our findings using a large scale meta-analysis of four blood lipids traits and find that the relative probability for causality is increased for variants in exons and transcription start sites and decreased in repressed genomic regions at the risk loci of these traits. Using these highly predictive, trait-specific functional annotations, we estimate causality probabilities across all traits and variants, reducing the size of the 90% confidence set from an average of 17.5 to 13.5 variants per locus in this data.

Primary open angle glaucoma (POAG) is a leading cause of blindness world-wide. To identify new susceptibility loci, we meta-analyzed GWAS results from 8 independent studies from the United States (3,853 cases and 33,480 controls) and investigated the most significant SNPs in two Australian studies (1,252 cases and 2,592 controls), 3 European studies (875 cases and 4,107 controls) and a Singaporean Chinese study (1,037 cases and 2,543 controls). A meta-analysis of top SNPs identified three novel loci: rs35934224[T] within TXNRD2 (odds ratio (OR) = 0.78, P = 4.05×10−11 encoding a mitochondrial protein required for redox homeostasis; rs7137828[T] within ATXN2 (OR = 1.17, P = 8.73×10−10), and rs2745572[A] upstream of FOXC1 (OR = 1.17, P = 1.76×10−10). Using RT-PCR and immunohistochemistry, we show TXNRD2 and ATXN2 expression in retinal ganglion cells and the optic nerve head. These results identify new pathways underlying POAG susceptibility and suggest novel targets for preventative therapies.

Introduction: Breast cancer in situ (BCIS) diagnoses, a precursor lesion for invasive breast cancer, comprise about 20 % of all breast cancers (BC) in countries with screening programs. Family history of BC is considered one of the strongest risk factors for BCIS. Methods: To evaluate the association of BC susceptibility loci with BCIS risk, we genotyped 39 single nucleotide polymorphisms (SNPs), associated with risk of invasive BC, in 1317 BCIS cases, 10,645 invasive BC cases, and 14,006 healthy controls in the National Cancer Institute’s Breast and Prostate Cancer Cohort Consortium (BPC3). Using unconditional logistic regression models adjusted for age and study, we estimated the association of SNPs with BCIS using two different comparison groups: healthy controls and invasive BC subjects to investigate whether BCIS and BC share a common genetic profile. Results: We found that five SNPs (CDKN2BAS-rs1011970, FGFR2-rs3750817, FGFR2-rs2981582, TNRC9-rs3803662, 5p12-rs10941679) were significantly associated with BCIS risk (P value adjusted for multiple comparisons <0.0016). Comparing invasive BC and BCIS, the largest difference was for CDKN2BAS-rs1011970, which showed a positive association with BCIS (OR = 1.24, 95 % CI: 1.11–1.38, P = 1.27 x 10−4) and no association with invasive BC (OR = 1.03, 95 % CI: 0.99–1.07, P = 0.06), with a P value for case-case comparison of 0.006. Subgroup analyses investigating associations with ductal carcinoma in situ (DCIS) found similar associations, albeit less significant (OR = 1.25, 95 % CI: 1.09–1.42, P = 1.07 x 10−3). Additional risk analyses showed significant associations with invasive disease at the 0.05 level for 28 of the alleles and the OR estimates were consistent with those reported by other studies. Conclusions: Our study adds to the knowledge that several of the known BC susceptibility loci are risk factors for both BCIS and invasive BC, with the possible exception of rs1011970, a putatively functional SNP situated in the CDKN2BAS gene that may be a specific BCIS susceptibility locus. Electronic supplementary material The online version of this article (doi:10.1186/s13058-015-0596-x) contains supplementary material, which is available to authorized users.