Characterizing genetic profiles for high triglyceride levels in U.S. patients of African ancestry.

Hypertriglyceridemia (HTG) is a common cardiovascular risk factor characterized by elevated triglyceride (TG) levels. Researchers have assessed the genetic factors that influence HTG in studies focused predominantly on individuals of European ancestry. However, relatively little is known about the contribution of genetic variation of HTG in people of African ancestry (AA), potentially constraining research and treatment opportunities. Our objective was to characterize genetic profiles among individuals of AA with mild-to-moderate HTG and severe HTG versus those with normal TGs by leveraging whole-genome sequencing data and longitudinal electronic health records available in the All of Us program. We compared the enrichment of functional variants within five canonical TG metabolism genes, an AA-specific polygenic risk score for TGs, and frequencies of 145 known potentially causal TG variants between HTG patients and normal TG among a cohort of AA patients (N = 15,373). Those with mild-to-moderate HTG (N = 342) and severe HTG (N ≤ 20) were more likely to carry APOA5 p.S19W (odds ratio = 1.94, 95% confidence interval = [1.48-2.54], P = 1.63 × 10-6 and OR = 3.65, 95% confidence interval: [1.22-10.93], P = 0.02, respectively) than those with normal TG. They were also more likely to have an elevated (top 10%) polygenic risk score, elevated carriage of potentially causal variant alleles, and carry any genetic risk factor. Alternative definitions of HTG yielded comparable results. In conclusion, individuals of AA with HTG were enriched for genetic risk factors compared to individuals with normal TGs.

Characterizing genetic profiles for high triglyceride levels in U.S. patients of African ancestry.

Hypertriglyceridemia (HTG) is a common cardiovascular risk factor characterized by elevated circulating triglyceride (TG) levels. Researchers have assessed the genetic factors that influence HTG in studies focused predominantly on individuals of European ancestry (EA). However, relatively little is known about the contribution of genetic variation to HTG in people of AA, potentially constraining research and treatment opportunities; the lipid profile for African ancestry (AA) populations differs from that of EA populations-which may be partially attributable to genetics. Our objective was to characterize genetic profiles among individuals of AA with mild-to-moderate HTG and severe HTG versus those with normal TGs by leveraging whole genome sequencing (WGS) data and longitudinal electronic health records (EHRs) available in the All of Us (AoU) program. We compared the enrichment of functional variants within five canonical TG metabolism genes, an AA-specific polygenic risk score for TGs, and frequencies of 145 known potentially causal TG variants between patients with HTG and normal TG among a cohort of AA patients (N=15,373). Those with mild-to-moderate HTG (N=342) and severe HTG (N≤20) were more likely to carry APOA5 p.S19W (OR=1.94, 95% CI [1.48-2.54], p=1.63×10 -6 and OR=3.65, 95% CI [1.22-10.93], p=0.02, respectively) than those with normal TG. They were also more likely to have an elevated (top 10%) PRS, elevated carriage of potentially causal variant alleles, and carry any genetic risk factor. Alternative definitions of HTG yielded comparable results. In conclusion, individuals of AA with HTG were enriched for genetic risk factors compared to individuals with normal TGs.

Clinical associations with a polygenic predisposition to benign lower white blood cell counts.

Polygenic variation unrelated to disease contributes to interindividual variation in baseline white blood cell (WBC) counts, but its clinical significance is uncharacterized. We investigated the clinical consequences of a genetic predisposition toward lower WBC counts among 89,559 biobank participants from tertiary care centers using a polygenic score for WBC count (PGSWBC) comprising single nucleotide polymorphisms not associated with disease. A predisposition to lower WBC counts was associated with a decreased risk of identifying pathology on a bone marrow biopsy performed for a low WBC count (odds-ratio = 0.55 per standard deviation increase in PGSWBC [95%CI, 0.30-0.94], p = 0.04), an increased risk of leukopenia (a low WBC count) when treated with a chemotherapeutic (n = 1724, hazard ratio [HR] = 0.78 [0.69-0.88], p = 4.0 × 10-5) or immunosuppressant (n = 354, HR = 0.61 [0.38-0.99], p = 0.04). A predisposition to benign lower WBC counts was associated with an increased risk of discontinuing azathioprine treatment (n = 1,466, HR = 0.62 [0.44-0.87], p = 0.006). Collectively, these findings suggest that there are genetically predisposed individuals who are susceptible to escalations or alterations in clinical care that may be harmful or of little benefit.

Large language models facilitate the generation of electronic health record phenotyping algorithms.

OBJECTIVES: Phenotyping is a core task in observational health research utilizing electronic health records (EHRs). Developing an accurate algorithm demands substantial input from domain experts, involving extensive literature review and evidence synthesis. This burdensome process limits scalability and delays knowledge discovery. We investigate the potential for leveraging large language models (LLMs) to enhance the efficiency of EHR phenotyping by generating high-quality algorithm drafts. MATERIALS AND METHODS: We prompted four LLMs-GPT-4 and GPT-3.5 of ChatGPT, Claude 2, and Bard-in October 2023, asking them to generate executable phenotyping algorithms in the form of SQL queries adhering to a common data model (CDM) for three phenotypes (ie, type 2 diabetes mellitus, dementia, and hypothyroidism). Three phenotyping experts evaluated the returned algorithms across several critical metrics. We further implemented the top-rated algorithms and compared them against clinician-validated phenotyping algorithms from the Electronic Medical Records and Genomics (eMERGE) network. RESULTS: GPT-4 and GPT-3.5 exhibited significantly higher overall expert evaluation scores in instruction following, algorithmic logic, and SQL executability, when compared to Claude 2 and Bard. Although GPT-4 and GPT-3.5 effectively identified relevant clinical concepts, they exhibited immature capability in organizing phenotyping criteria with the proper logic, leading to phenotyping algorithms that were either excessively restrictive (with low recall) or overly broad (with low positive predictive values). CONCLUSION: GPT versions 3.5 and 4 are capable of drafting phenotyping algorithms by identifying relevant clinical criteria aligned with a CDM. However, expertise in informatics and clinical experience is still required to assess and further refine generated algorithms.

Serious Bleeding in Patients With Atrial Fibrillation Using Diltiazem With Apixaban or Rivaroxaban.

Importance: Diltiazem, a commonly prescribed ventricular rate-control medication for patients with atrial fibrillation, inhibits apixaban and rivaroxaban elimination, possibly causing overanticoagulation. Objective: To compare serious bleeding risk for new users of apixaban or rivaroxaban with atrial fibrillation treated with diltiazem or metoprolol. Design, Setting, and Participants: This retrospective cohort study included Medicare beneficiaries aged 65 years or older with atrial fibrillation who initiated apixaban or rivaroxaban use and also began treatment with diltiazem or metoprolol between January 1, 2012, and November 29, 2020. Patients were followed up to 365 days through November 30, 2020. Data were analyzed from August 2023 to February 2024. Exposures: Diltiazem and metoprolol. Main Outcomes and Measures: The primary outcome was a composite of bleeding-related hospitalization and death with recent evidence of bleeding. Secondary outcomes were ischemic stroke or systemic embolism, major ischemic or hemorrhagic events (ischemic stroke, systemic embolism, intracranial or fatal extracranial bleeding, or death with recent evidence of bleeding), and death without recent evidence of bleeding. Hazard ratios (HRs) and rate differences (RDs) were adjusted for covariate differences with overlap weighting. Results: The study included 204 155 US Medicare beneficiaries, of whom 53 275 received diltiazem and 150 880 received metoprolol. Study patients (mean [SD] age, 76.9 [7.0] years; 52.7% female) had 90 927 person-years (PY) of follow-up (median, 120 [IQR, 59-281] days). Patients receiving diltiazem treatment had increased risk for the primary outcome (RD, 10.6 [95% CI, 7.0-14.2] per 1000 PY; HR, 1.21 [95% CI, 1.13-1.29]) and its components of bleeding-related hospitalization (RD, 8.2 [95% CI, 5.1-11.4] per 1000 PY; HR, 1.22 [95% CI, 1.13-1.31]) and death with recent evidence of bleeding (RD, 2.4 [95% CI, 0.6-4.2] per 1000 PY; HR, 1.19 [95% CI, 1.05-1.34]) compared with patients receiving metoprolol. Risk for the primary outcome with initial diltiazem doses exceeding 120 mg/d (RD, 15.1 [95% CI, 10.2-20.1] per 1000 PY; HR, 1.29 [95% CI, 1.19-1.39]) was greater than that for lower doses (RD, 6.7 [95% CI, 2.0-11.4] per 1000 PY; HR, 1.13 [95% CI, 1.04-1.24]). For doses exceeding 120 mg/d, the risk of major ischemic or hemorrhagic events was increased (HR, 1.14 [95% CI, 1.02-1.27]). Neither dose group had significant changes in the risk for ischemic stroke or systemic embolism or death without recent evidence of bleeding. When patients receiving high- and low-dose diltiazem treatment were directly compared, the HR for the primary outcome was 1.14 (95% CI, 1.02-1.26). Conclusions and Relevance: In Medicare patients with atrial fibrillation receiving apixaban or rivaroxaban, diltiazem was associated with greater risk of serious bleeding than metoprolol, particularly for diltiazem doses exceeding 120 mg/d.

Leveraging generative AI to prioritize drug repurposing candidates for Alzheimer's disease with real-world clinical validation.

Drug repurposing represents an attractive alternative to the costly and time-consuming process of new drug development, particularly for serious, widespread conditions with limited effective treatments, such as Alzheimer's disease (AD). Emerging generative artificial intelligence (GAI) technologies like ChatGPT offer the promise of expediting the review and summary of scientific knowledge. To examine the feasibility of using GAI for identifying drug repurposing candidates, we iteratively tasked ChatGPT with proposing the twenty most promising drugs for repurposing in AD, and tested the top ten for risk of incident AD in exposed and unexposed individuals over age 65 in two large clinical datasets: (1) Vanderbilt University Medical Center and (2) the All of Us Research Program. Among the candidates suggested by ChatGPT, metformin, simvastatin, and losartan were associated with lower AD risk in meta-analysis. These findings suggest GAI technologies can assimilate scientific insights from an extensive Internet-based search space, helping to prioritize drug repurposing candidates and facilitate the treatment of diseases.

Large Language Models Facilitate the Generation of Electronic Health Record Phenotyping Algorithms.

Objectives: Phenotyping is a core task in observational health research utilizing electronic health records (EHRs). Developing an accurate algorithm demands substantial input from domain experts, involving extensive literature review and evidence synthesis. This burdensome process limits scalability and delays knowledge discovery. We investigate the potential for leveraging large language models (LLMs) to enhance the efficiency of EHR phenotyping by generating high-quality algorithm drafts. Materials and Methods: We prompted four LLMs-GPT-4 and GPT-3.5 of ChatGPT, Claude 2, and Bard-in October 2023, asking them to generate executable phenotyping algorithms in the form of SQL queries adhering to a common data model (CDM) for three phenotypes (i.e., type 2 diabetes mellitus, dementia, and hypothyroidism). Three phenotyping experts evaluated the returned algorithms across several critical metrics. We further implemented the top-rated algorithms and compared them against clinician-validated phenotyping algorithms from the Electronic Medical Records and Genomics (eMERGE) network. Results: GPT-4 and GPT-3.5 exhibited significantly higher overall expert evaluation scores in instruction following, algorithmic logic, and SQL executability, when compared to Claude 2 and Bard. Although GPT-4 and GPT-3.5 effectively identified relevant clinical concepts, they exhibited immature capability in organizing phenotyping criteria with the proper logic, leading to phenotyping algorithms that were either excessively restrictive (with low recall) or overly broad (with low positive predictive values). Conclusion: GPT versions 3.5 and 4 are capable of drafting phenotyping algorithms by identifying relevant clinical criteria aligned with a CDM. However, expertise in informatics and clinical experience is still required to assess and further refine generated algorithms.

Association between APOL1 risk variants and the occurrence of sepsis in Black patients hospitalized with infections: a retrospective cohort study.

Background: Two risk variants in the apolipoprotein L1 gene (APOL1) have been associated with increased susceptibility to sepsis in Black patients. However, it remains unclear whether APOL1 high-risk genotypes are associated with occurrence of either sepsis or sepsis-related phenotypes in patients hospitalized with infections, independent of their association with pre-existing severe renal disease. Methods: A retrospective cohort study of 2242 Black patients hospitalized with infections. We assessed whether carriage of APOL1 high-risk genotypes was associated with the risk of sepsis and sepsis-related phenotypes in patients hospitalized with infections. The primary outcome was sepsis; secondary outcomes were short-term mortality, and organ failure related to sepsis. Results: Of 2242 Black patients hospitalized with infections, 565 developed sepsis. Patients with high-risk APOL1 genotypes had a significantly increased risk of sepsis (odds ratio [OR]=1.29 [95% CI, 1.00-1.67; p=0.047]); however, this association was not significant after adjustment for pre-existing severe renal disease (OR = 1.14 [95% CI, 0.88-1.48; p=0.33]), nor after exclusion of those patients with pre-existing severe renal disease (OR = 0.99 [95% CI, 0.70-1.39; p=0.95]). APOL1 high-risk genotypes were significantly associated with the renal dysfunction component of the Sepsis-3 criteria (OR = 1.64 [95% CI, 1.21-2.22; p=0.001]), but not with other sepsis-related organ dysfunction or short-term mortality. The association between high-risk APOL1 genotypes and sepsis-related renal dysfunction was markedly attenuated by adjusting for pre-existing severe renal disease (OR = 1.36 [95% CI, 1.00-1.86; p=0.05]) and was nullified after exclusion of patients with pre-existing severe renal disease (OR = 1.16 [95% CI, 0.74-1.81; p=0.52]). Conclusions: APOL1 high-risk genotypes were associated with an increased risk of sepsis; however, this increased risk was attributable predominantly to pre-existing severe renal disease. Funding: This study was supported by R01GM120523 (QF), R01HL163854 (QF), R35GM131770 (CMS), HL133786 (WQW), and Vanderbilt Faculty Research Scholar Fund (QF). The dataset(s) used for the analyses described were obtained from Vanderbilt University Medical Center's BioVU which is supported by institutional funding, the 1S10RR025141-01 instrumentation award, and by the CTSA grant UL1TR0004from NCATS/NIH. Additional funding provided by the NIH through grants P50GM115305 and U19HL065962. The authors wish to acknowledge the expert technical support of the VANTAGE and VANGARD core facilities, supported in part by the Vanderbilt-Ingram Cancer Center (P30 CA068485) and Vanderbilt Vision Center (P30 EY08126). The funders had no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

When the body is fighting off an infection, the processes it uses to protect itself can sometimes overreact. This results in a condition known as sepsis which can cause life-threatening damage to multiple organs. In the United States, Black patients are 60-80% more likely to develop sepsis compared to individuals who identify as White; differences remain even after accounting for socio-economic status and presence of other illnesses. Recent work has suggested that two variants of the APOL1 gene which are almost exclusively found in people with African ancestry may be a contributing factor to this disparity. These 'high-risk' genetic variants have also been shown to increase the likelihood of kidney diseases. It is therefore possible that the elevated chance of sepsis is not directly linked to these variations of APOL1, but rather is the result of patients already having reduced kidney function. To understand the relationship between APOL1 and sepsis, Jiang et al. analyzed data from patients admitted to Vanderbilt University Medical Centre in the United States between 2000 and 2020. This included 2,242 patients who identified as Black and had been hospitalized with an infection. The analyses showed that 16% of these individuals were carriers of the APOL1 high-risk variants. The high-risk patients were more likely to experience sepsis and demonstrate kidney damage. But other organs commonly damaged by sepsis were not affected more in these individuals compared to the other 84% of patients who did not have these variants. Furthermore, when individuals with pre-existing kidney diseases were removed from this high-risk group, the increased likelihood of sepsis was no longer prominent. These findings suggest that the APOL1 variants do not directly increase the risk of sepsis, and this association is primarily due to patients with these genetic variations being more susceptible to kidney diseases. There are new drugs under development targeting the APOL1 variants. While these may provide protection against kidney diseases, they are unlikely to be successful at preventing or treating sepsis once a patient has been hospitalized with an infection.

Clinical consequences of a polygenic predisposition to benign lower white blood cell counts: Consequences of benign WBC count genetics.

Polygenic variation unrelated to disease contributes to interindividual variation in baseline white blood cell (WBC) counts, but its clinical significance is undefined. We investigated the clinical consequences of a genetic predisposition toward lower WBC counts among 89,559 biobank participants from tertiary care centers using a polygenic score for WBC count (PGSWBC) comprising single nucleotide polymorphisms not associated with disease. A predisposition to lower WBC counts was associated with a decreased risk of identifying pathology on a bone marrow biopsy performed for a low WBC count (odds-ratio=0.55 per standard deviation increase in PGSWBC [95%CI, 0.30 - 0.94], p=0.04), an increased risk of leukopenia (a low WBC count) when treated with a chemotherapeutic (n=1,724, hazard ratio [HR]=0.78 [0.69 - 0.88], p=4.0×10-5) or immunosuppressant (n=354, HR=0.61 [0.38 - 0.99], p=0.04). A predisposition to benign lower WBC counts was associated with an increased risk of discontinuing azathioprine treatment (n=1,466, HR=0.62 [0.44 - 0.87], p=0.006). Collectively, these findings suggest that a WBC count polygenic score identifies individuals who are susceptible to escalations or alterations in clinical care that may be harmful or of little benefit.

Concurrent Gabapentin and Opioid Use and Risk of Mortality in Medicare Recipients with Non-Cancer Pain.

Gabapentin is prescribed for pain and is perceived as safe generally. However, gabapentin can cause respiratory depression, exacerbated by concomitant central nervous system depressants (e.g., opioids), a concern for vulnerable populations. We compared mortality rates among new users of either gabapentin or duloxetine with or without concurrent opioids in the 20% Medicare sample. We conducted a new-user design retrospective cohort study, in Medicare enrollees ages 65-89 years with noncancer chronic pain and no severe illness who filled prescriptions between 2015 and 2018 for gabapentin (n = 233,060) or duloxetine (n = 34,009). Daily opioid doses, estimated in morphine milligram equivalents (MMEs), were classified into none, low (0 < MME < 50), and high (≥ 50 MME), based on Centers for Disease Control and Prevention (CDC) recommendations. The outcomes were all-cause mortality (primary) and out-of-hospital mortality (secondary). We used inverse probability of treatment weighting to adjust for differences between gabapentin and duloxetine users. During 116,707 person-years of follow-up, 1,379 patients died. All-cause mortality rate in gabapentin users was 12.16 per 1,000 person-years vs. 9.94 per 1,000 in duloxetine users. Risks were similar for users with no concurrent opioids (adjusted hazard ratio (aHR) = 1.03, 95% confidence interval (CI): 0.80-1.31) or low-dose daily opioids (aHR = 1.06, 95% CI: 0.63-1.76). However, gabapentin users receiving concurrent high-dose daily opioids had an increased rate of all-cause mortality compared with duloxetine users on high-dose opioids (aHR = 2.03, 95% CI: 1.19-3.46). Out-of-hospital mortality yielded similar results. In this retrospective cohort study of Medicare beneficiaries, concurrent use of high-dose opioids and gabapentin was associated with a higher all-cause mortality risk than that for concurrent use of high-dose opioids and duloxetine.

Leveraging Generative AI to Prioritize Drug Repurposing Candidates: Validating Identified Candidates for Alzheimer's Disease in Real-World Clinical Datasets.

Drug repurposing represents an attractive alternative to the costly and time-consuming process of new drug development, particularly for serious, widespread conditions with limited effective treatments, such as Alzheimer's disease (AD). Emerging generative artificial intelligence (GAI) technologies like ChatGPT offer the promise of expediting the review and summary of scientific knowledge. To examine the feasibility of using GAI for identifying drug repurposing candidates, we iteratively tasked ChatGPT with proposing the twenty most promising drugs for repurposing in AD, and tested the top ten for risk of incident AD in exposed and unexposed individuals over age 65 in two large clinical datasets: 1) Vanderbilt University Medical Center and 2) the All of Us Research Program. Among the candidates suggested by ChatGPT, metformin, simvastatin, and losartan were associated with lower AD risk in meta-analysis. These findings suggest GAI technologies can assimilate scientific insights from an extensive Internet-based search space, helping to prioritize drug repurposing candidates and facilitate the treatment of diseases.

Leveraging Generative AI to Prioritize Drug Repurposing Candidates: Validating Identified Candidates for Alzheimer's Disease in Real-World Clinical Datasets.

Drug repurposing represents an attractive alternative to the costly and time-consuming process of new drug development, particularly for serious, widespread conditions with limited effective treatments, such as Alzheimer's disease (AD). Emerging generative artificial intelligence (GAI) technologies like ChatGPT offer the promise of expediting the review and summary of scientific knowledge. To examine the feasibility of using GAI for identifying drug repurposing candidates, we iteratively tasked ChatGPT with proposing the twenty most promising drugs for repurposing in AD, and tested the top ten for risk of incident AD in exposed and unexposed individuals over age 65 in two large clinical datasets: 1) Vanderbilt University Medical Center and 2) the All of Us Research Program. Among the candidates suggested by ChatGPT, metformin, simvastatin, and losartan were associated with lower AD risk in meta-analysis. These findings suggest GAI technologies can assimilate scientific insights from an extensive Internet-based search space, helping to prioritize drug repurposing candidates and facilitate the treatment of diseases.

Risk for Bleeding-Related Hospitalizations During Use of Amiodarone With Apixaban or Rivaroxaban in Patients With Atrial Fibrillation : A Retrospective Cohort Study.

BACKGROUND: Amiodarone, the most effective antiarrhythmic drug in atrial fibrillation, inhibits apixaban and rivaroxaban elimination, thus possibly increasing anticoagulant-related risk for bleeding. OBJECTIVE: For patients receiving apixaban or rivaroxaban, to compare risk for bleeding-related hospitalizations during treatment with amiodarone versus flecainide or sotalol, antiarrhythmic drugs that do not inhibit these anticoagulants' elimination. DESIGN: Retrospective cohort study. SETTING: U.S. Medicare beneficiaries aged 65 years or older. PATIENTS: Patients with atrial fibrillation began anticoagulant use between 1 January 2012 and 30 November 2018 and subsequently initiated treatment with study antiarrhythmic drugs. MEASUREMENTS: Time to event for bleeding-related hospitalizations (primary outcome) and ischemic stroke, systemic embolism, and death with or without recent (past 30 days) evidence of bleeding (secondary outcomes), adjusted with propensity score overlap weighting. RESULTS: There were 91 590 patients (mean age, 76.3 years; 52.5% female) initiating use of study anticoagulants and antiarrhythmic drugs, 54 977 with amiodarone and 36 613 with flecainide or sotalol. Risk for bleeding-related hospitalizations increased with amiodarone use (rate difference [RD], 17.5 events [95% CI, 12.0 to 23.0 events] per 1000 person-years; hazard ratio [HR], 1.44 [CI, 1.27 to 1.63]). Incidence of ischemic stroke or systemic embolism did not increase (RD, -2.1 events [CI, -4.7 to 0.4 events] per 1000 person-years; HR, 0.80 [CI, 0.62 to 1.03]). The risk for death with recent evidence of bleeding (RD, 9.1 events [CI, 5.8 to 12.3 events] per 1000 person-years; HR, 1.66 [CI, 1.35 to 2.03]) was greater than that for other deaths (RD, 5.6 events [CI, 0.5 to 10.6 events] per 1000 person-years; HR, 1.15 [CI, 1.00 to 1.31]) (HR comparison: P = 0.003). The increased incidence of bleeding-related hospitalizations for rivaroxaban (RD, 28.0 events [CI, 18.4 to 37.6 events] per 1000 person-years) was greater than that for apixaban (RD, 9.1 events [CI, 2.8 to 15.3 events] per 1000 person-years) (P = 0.001). LIMITATION: Possible residual confounding. CONCLUSION: In this retrospective cohort study, patients aged 65 years or older with atrial fibrillation treated with amiodarone during apixaban or rivaroxaban use had greater risk for bleeding-related hospitalizations than those treated with flecainide or sotalol. PRIMARY FUNDING SOURCE: National Heart, Lung, and Blood Institute.

Duloxetine, Gabapentin, and the Risk for Acute Myocardial Infarction, Stroke, and Out-of-Hospital Death in Medicare Beneficiaries With Non-Cancer Pain.

OBJECTIVE: Duloxetine is a serotonin-norepinephrine reuptake inhibitor prescribed for musculoskeletal and other forms of chronic pain. Its dual pharmacologic properties have the potential to either raise or lower cardiovascular risk: adrenergic activity may increase the risk for acute myocardial infarction (AMI) and stroke, but antiplatelet activity may decrease risk. Gabapentin is another nonopioid medication used to treat pain, which is not thought to have adrenergic/antiplatelet effects. With the current emphasis on the use of nonopioid medications to treat patients with chronic pain, assessing cardiovascular risks associated with these medications among high-risk patients is important. MATERIALS AND METHODS: We conducted a retrospective cohort study among a 20% sample of Medicare enrollees, aged 65 to 89, with chronic pain who were new users between 2015 and 2018 of either duloxetine (n = 34,009) or gabapentin (n = 233,060). We excluded individuals with cancer or other life-threatening conditions at study drug initiation. The primary outcome was a composite of AMI, stroke, and out-of-hospital mortality. We adjusted for comorbidity differences with time-dependent inverse probability of treatment weighting. RESULTS: During 115,668 person-years of follow-up, 2361 patients had the composite primary outcome; the rate among new users of duloxetine was 16.7/1000 person-years compared with new users of gabapentin (21.1/1000 person-years), adjusted hazard ratio = 0.98 (95% CI: 0.83, 1.16). Results were similar for the individual components of the composite outcome as well as in analyses stratified by demographic and clinical characteristics. DISCUSSION: In summary, cohort Medicare patients with non-cancer pain beginning treatment with duloxetine had rates of AMI, stroke, and out-of-hospital mortality comparable to those who initiated gabapentin.

APOL1 and the risk of adverse renal outcomes in patients of African ancestry with systemic lupus erythematosus.

BACKGROUND: Systemic lupus erythematosus (SLE) disproportionately affects individuals of African ancestry (AA) compared to European ancestry (EA). In the general population, high risk (HR) variants in the apolipoprotein L1 (APOL1) gene increase the risk of renal and hypertensive disorders in individuals of AA. Since SLE is characterized by an interferon signature and APOL1 expression is driven by interferon, we examined the hypothesis that APOL1 HR genotypes predominantly drive higher rates of renal and hypertensive-related comorbidities observed in SLE patients of AA versus those of EA. METHODS: We performed a retrospective cohort study in patients with SLE of EA and AA using a genetic biobank linked to de-identified electronic health records. APOL1 HR genotypes were defined as G1/G1, G2/G2, or G1/G2 and low risk (LR) genotypes as 1 or 0 copies of the G1 and G2 alleles. To identify renal and hypertensive-related disorders that differed in prevalence by ancestry, we used a phenome-wide association approach. We then used logistic regression to compare the prevalence of renal and hypertensive-related disorders in EA and AA patients, both including and excluding patients with the APOL1 HR genotype. In a sensitivity analysis, we examined the association of end stage renal disease secondary to lupus nephritis (LN-related ESRD) with ancestry and the APOL1 genotype. RESULTS: We studied 784 patients with SLE; 195 (24.9%) were of AA, of whom 27 (13.8%) had APOL1 HR genotypes. Eighteen renal and hypertensive-related phenotypes were more common in AA than EA patients (p-value ≤ 1.4E-4). All phenotypes remained significantly different after exclusion of patients with APOL1 HR genotypes, and most point odds ratios (ORs) decreased only slightly. Even among ORs with the greatest decrease, risk for AA patients without the APOL1 HR genotype remained significantly elevated compared to EA patients. In the sensitivity analysis, LN-related ESRD was more prevalent in SLE patients of AA versus EA and AA patients with the APOL1 HR genotype versus LR (p-value < .05 for both). CONCLUSION: The higher prevalence of renal and hypertensive disorders in SLE patients of AA compared to those of EA is not fully explained by the presence of APOL1 high risk variants.

Association between APOL1 risk variants and the occurrence of sepsis in Black patients hospitalized with infections: a retrospective cohort study.

Background: Two risk variants in the apolipoprotein L1 gene ( APOL1 ) have been associated with increased susceptibility to sepsis in Black patients. However, it remains unclear whether APOL1 high-risk genotypes are associated with occurrence of either sepsis or sepsis-related phenotypes in patients hospitalized with infections, independent of their association with pre-existing severe renal disease. Methods: A retrospective cohort study of 2,242 Black patients hospitalized with infections. We assessed whether carriage of APOL1 high-risk genotypes was associated with the risk of sepsis and sepsis-related phenotypes in patients hospitalized with infections. The primary outcome was sepsis; secondary outcomes were short-term mortality and organ failure related to sepsis. Results: Of 2,242 Black patients hospitalized with infections, 565 developed sepsis. Patients with high-risk APOL1 genotypes had a significantly increased risk of sepsis (odds ratio [OR]=1.29 [95% CI, 1.00-1.67; p=0.047]); however, this association was not significant after adjustment for pre-existing severe renal disease (OR=1.14 [95% CI, 0.88-1.48; p=0.33]), nor after exclusion of those patients with pre-existing severe renal disease (OR=0.99 [95% CI, 0.70-1.39; p=0.95]. APOL1 high-risk genotypes were significantly associated with the renal dysfunction component of the Sepsis-3 criteria (OR=1.64 [95% CI, 1.21-2.22; p=0.001], but not with other sepsis-related organ dysfunction or short-term mortality. The association between high-risk APOL1 genotypes and sepsis-related renal dysfunction was markedly attenuated by adjusting for pre-existing severe renal disease (OR=1.36 [95% CI, 1.00-1.86; p=0.05]) and was nullified after exclusion of patients with pre-existing severe renal disease (OR=1.16 [95% CI, 0.74-1.81; p=0.52]). Conclusion: APOL1 high-risk genotypes were associated with an increased risk of sepsis; however, this increased risk was attributable predominantly to pre-existing severe renal disease. Funding: This study was supported by R01GM120523 (Q.F.), R01HL163854 (Q.F.), R35GM131770 (C.M.S.), HL133786 (W.Q.W.), and Vanderbilt Faculty Research Scholar Fund (Q.F.). The dataset(s) used for the analyses described were obtained from Vanderbilt University Medical Center's BioVU which is supported by institutional funding, the 1S10RR025141-01 instrumentation award, and by the CTSA grant UL1TR0004from NCATS/NIH. Additional funding provided by the NIH through grants P50GM115305 and U19HL065962. The authors wish to acknowledge the expert technical support of the VANTAGE and VANGARD core facilities, supported in part by the Vanderbilt-Ingram Cancer Center (P30 CA068485) and Vanderbilt Vision Center (P30 EY08126).The funders had no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Association Between Genetically Predicted Expression of TPMT and Azathioprine Adverse Events.

Polymorphisms thiopurine-S-methyltransferase (TPMT) and nudix hydrolase 15 (NUDT15) can increase the risk of azathioprine myelotoxicity, but little is known about other genetic factors that increase risk for azathioprine-associated side effects. PrediXcan is a gene-based association method that estimates the expression of individuals' genes and examines their correlation to specified phenotypes. As proof of concept for using PrediXcan as a tool to define the association between genetic factors and azathioprine side effects, we aimed to determine whether the genetically predicted expression of TPMT or NUDT15 was associated with leukopenia or other known side effects. In a retrospective cohort of 1364 new users of azathioprine with EHR-reported White race, we used PrediXcan to impute expression in liver tissue, tested its association with pre-specified phecodes representing known side effects (e.g., skin cancer), and completed chart review to confirm cases. Among confirmed cases, patients in the lowest tertile (i.e., lowest predicted) of TPMT expression had significantly higher odds of developing leukopenia (OR=3.30, 95%CI: 1.07-10.20, p=0.04) versus those in the highest tertile; no other side effects were significant. The results suggest that this methodology could be deployed on a larger scale to uncover associations between genetic factors and drug side effects for more personalized care.

The relationship between high density lipoprotein cholesterol and sepsis: A clinical and genetic approach.

Sepsis accounts for one in three hospital deaths. Higher concentrations of high-density lipoprotein cholesterol (HDL-C) are associated with apparent protection from sepsis, suggesting a potential therapeutic role for HDL-C or drugs, such as cholesteryl ester transport protein (CETP) inhibitors that increase HDL-C. However, these beneficial clinical associations might be due to confounding; genetic approaches can address this possibility. We identified 73,406 White adults admitted to Vanderbilt University Medical Center with infection; 11,612 had HDL-C levels, and 12,377 had genotype information from which we constructed polygenic risk scores (PRS) for HDL-C and the effect of CETP on HDL-C. We tested the associations between predictors (measured HDL-C, HDL-C PRS, CETP PRS, and rs1800777) and outcomes: sepsis, septic shock, respiratory failure, and in-hospital death. In unadjusted analyses, lower measured HDL-C concentrations were significantly associated with increased risk of sepsis (p = 2.4 × 10-23 ), septic shock (p = 4.1 × 10-12 ), respiratory failure (p = 2.8 × 10-8 ), and in-hospital death (p = 1.0 × 10-8 ). After adjustment (age, sex, electronic health record length, comorbidity score, LDL-C, triglycerides, and body mass index), these associations were markedly attenuated: sepsis (p = 2.6 × 10-3 ), septic shock (p = 8.1 × 10-3 ), respiratory failure (p = 0.11), and in-hospital death (p = 4.5 × 10-3 ). HDL-C PRS, CETP PRS, and rs1800777 significantly predicted HDL-C (p < 2 × 10-16 ), but none were associated with sepsis outcomes. Concordant findings were observed in 13,254 Black patients hospitalized with infections. Lower measured HDL-C levels were significantly associated with increased risk of sepsis and related outcomes in patients with infection, but a causal relationship is unlikely because no association was found between the HDL-C PRS or the CETP PRS and the risk of adverse sepsis outcomes.

Genetic and Survey Data Improves Performance of Machine Learning Model for Long COVID.

Over 200 million SARS-CoV-2 patients have or will develop persistent symptoms (long COVID). Given this pressing research priority, the National COVID Cohort Collaborative (N3C) developed a machine learning model using only electronic health record data to identify potential patients with long COVID. We hypothesized that additional data from health surveys, mobile devices, and genotypes could improve prediction ability. In a cohort of SARS-CoV-2 infected individuals (n=17,755) in the All of Us program, we applied and expanded upon the N3C long COVID prediction model, testing machine learning infrastructures, assessing model performance, and identifying factors that contributed most to the prediction models. For the survey/mobile device information and genetic data, extreme gradient boosting and a convolutional neural network delivered the best performance for predicting long COVID, respectively. Combined survey, genetic, and mobile data increased specificity and the Area Under Curve the Receiver Operating Characteristic score versus the original N3C model.