Cerhan, James R.

Abstract Background Mendelian randomization (MR) studies are susceptible to metadata errors (e.g. incorrect specification of the effect allele column) and other analytical issues that can introduce substantial bias into analyses. We developed a quality control (QC) pipeline for the Fatty Acids in Cancer Mendelian Randomization Collaboration (FAMRC) that can be used to identify and correct for such errors. Methods We collated summary association statistics from fatty acid and cancer genome-wide association studies (GWAS) and subjected the collated data to a comprehensive QC pipeline. We identified metadata errors through comparison of study-specific statistics to external reference data sets (the National Human Genome Research Institute-European Bioinformatics Institute GWAS catalogue and 1000 genome super populations) and other analytical issues through comparison of reported to expected genetic effect sizes. Comparisons were based on three sets of genetic variants: (i) GWAS hits for fatty acids, (ii) GWAS hits for cancer and (iii) a 1000 genomes reference set. Results We collated summary data from 6 fatty acid and 54 cancer GWAS. Metadata errors and analytical issues with the potential to introduce substantial bias were identified in seven studies (11.6%). After resolving metadata errors and analytical issues, we created a data set of 219 842 genetic associations with 90 cancer types, generated in analyses of 566 665 cancer cases and 1 622 374 controls. Conclusions In this large MR collaboration, 11.6% of included studies were affected by a substantial metadata error or analytical issue. By increasing the integrity of collated summary data prior to their analysis, our protocol can be used to increase the reliability of downstream MR analyses. Our pipeline is available to other researchers via the CheckSumStats package (https://github.com/MRCIEU/CheckSumStats).

ImportanceAnthracyclines increase the risk for congestive heart failure (CHF); however, long-term cumulative incidence and risk factors for CHF after anthracycline therapy are not well defined in population-based studies.ObjectiveTo compare the long-term cumulative incidence of CHF in patients with breast cancer or lymphoma treated with anthracycline therapy compared with healthy controls from the same community.Design, Setting, and ParticipantsThis retrospective population-based case-control study included data from the Rochester Epidemiology Project. Participants included residents of Olmsted County, Minnesota, diagnosed with breast cancer or lymphoma from January 1985 through December 2010 matched for age, sex, and comorbidities with healthy controls, with a final ratio of 1 case to 1.5 controls. Statistical analysis was performed between July 2017 and February 2022.ExposuresCancer treatment and CHF risk factors.Main Outcomes and MeasuresThe main outcome was new-onset CHF, as defined by the modified Framingham criteria. Cox proportional hazards regression was used to estimate hazard ratios (HRs) to compare the risk of CHF in participants with cancer vs controls, adjusted for age, sex, diabetes, hypertension, hyperlipidemia, coronary artery disease, obesity, and smoking history.ResultsA total of 2196 individuals were included, with 812 patients with cancer and 1384 participants without cancer. The mean (SD) age was 52.62 (14.56) years and 1704 participants (78%) were female. Median (IQR) follow-up was 8.6 (5.2-13.4) years in the case group vs 12.5 (8.7-17.5) years in the control group. Overall, patients with cancer had higher risk of CHF compared with the control cohort even after adjusting for age, sex, diabetes, hypertension, coronary artery disease, hyperlipidemia, obesity, and smoking status (HR, 2.86 [95% CI, 1.90-4.32]; P < .001). After adjusting for the same variables, CHF risk was greater for patients with cancer receiving anthracycline (HR, 3.25 [95% CI, 2.11-5.00]; P < .001) and was attenuated and lost statistical significance for patients with cancer not receiving anthracyclines (HR, 1.78 [95% CI, 0.83-3.81]; P = .14). Higher cumulative incidence for patients treated with anthracyclines vs comparator cohort was observed at 1 year (1.81% vs 0.09%), 5 years (2.91% vs 0.79%), 10 years (5.36% vs 1.74%), 15 years (7.42% vs 3.18%), and 20 years (10.75% vs 4.98%) (P < .001). There were no significant differences in risk of CHF for patients receiving anthracycline at a dose of less than 180 mg/m2 compared with those at a dose of 180 to 250 mg/m2 (HR, 0.54 [95% CI, 0.19-1.51]) or at a dose of more than 250 mg/m2 (HR, 1.23 [95% CI, 0.52-2.91]). At diagnosis, age was an independent risk factor associated with CHF (HR per 10 years, 2.77 [95% CI, 1.99-3.86]; P < .001).Conclusions and RelevanceIn this retrospective population-based case-control study, anthracyclines were associated with an increased risk of CHF early during follow-up, and the increased risk persisted over time. The cumulative incidence of CHF in patients with breast cancer or lymphoma treated with anthracyclines at 15 years was more than 2-fold that of the control group.

Abstract Context Metformin is the first-line drug for treating diabetes but has a high failure rate. Objective To identify demographic and clinical factors available in the electronic health record (EHR) that predict metformin failure. Methods A cohort of patients with at least 1 abnormal diabetes screening test that initiated metformin was identified at 3 sites (Arizona, Mississippi, and Minnesota). We identified 22 047 metformin initiators (48% female, mean age of 57 ± 14 years) including 2141 African Americans, 440 Asians, 962 Other/Multiracial, 1539 Hispanics, and 16 764 non-Hispanic White people. We defined metformin failure as either the lack of a target glycated hemoglobin (HbA1c) (<7%) within 18 months of index or the start of dual therapy. We used tree-based extreme gradient boosting (XGBoost) models to assess overall risk prediction performance and relative contribution of individual factors when using EHR data for risk of metformin failure. Results In this large diverse population, we observed a high rate of metformin failure (43%). The XGBoost model that included baseline HbA1c, age, sex, and race/ethnicity corresponded to high discrimination performance (C-index of 0.731; 95% CI 0.722, 0.740) for risk of metformin failure. Baseline HbA1c corresponded to the largest feature performance with higher levels associated with metformin failure. The addition of other clinical factors improved model performance (0.745; 95% CI 0.737, 0.754, P < .0001). Conclusion Baseline HbA1c was the strongest predictor of metformin failure and additional factors substantially improved performance suggesting that routinely available clinical data could be used to identify patients at high risk of metformin failure who might benefit from closer monitoring and earlier treatment intensification.

Abstract The prognostic relevance of diagnosis to treatment interval (DTI) in patients with newly diagnosed mantle cell lymphoma (MCL) is unknown. Hence, we sought to evaluate the impact of DTI on outcomes in MCL using 3 large datasets (1) the University of Iowa/Mayo Clinic Specialized Program of Research Excellence Molecular Epidemiology Resource, (2) patients enrolled in the ALL Age Asthma Cohort/CALGB 50403, and (3) a multisitecohort of patients with MCL. Patients were a priori divided into 2 groups, 0 to 14 days (short DTI) and 15 to 60 days (long DTI). The patients in whom observation was deemed appropriate were excluded. One thousand ninety-seven patients newly diagnosed with MCL and available DTI were included in the study. The majority (73%) had long DTI (n=797). Patients with short DTI had worse eastern cooperative oncology group performance status (ECOG PS ≥2), higher lactate dehydrogenase, bone marrow involvement, more frequent B symptoms, higher MCL International Prognostic Index (MIPI ≥6.2), and were less likely to receive intensive induction therapy than long DTI group. The median progression-free survival (2.5 years vs 4.8 years, p<0.0001) and overall survival (7.8 years vs. 11.8 years, p<0.0001) were significantly inferior in the short DTI group than the long DTI cohort and remained significant for progression-free survival and overall survival in multivariable analysis. We show that the DTI is an important prognostic factor in patients newly diagnosed with MCL and is strongly associated with adverse clinical factors and poor outcomes. DTI should be reported in all the patients newly diagnosed with MCL who are enrolling in clinical trials and steps must be taken to ensure selection bias is avoided.