Predictive utility of a genetic risk score of common variants associated with type 2 diabetes in a black South African population

Assessing the prediction of type 2 diabetes risk using polygenic and clinical risk scores in South Asian study populations

Background

Genome-wide polygenic risk scores (PRS) have shown high specificity and sensitivity in predicting type 2 diabetes (T2D) risk in Europeans. However, the PRS-driven information and its clinical significance in non-Europeans are underrepresented. We examined the predictive efficacy and transferability of PRS models using variant information derived from genome-wide studies of Asian Indians (AIs) (PRSAI) and Europeans (PRSEU) using 13,974 AI individuals.

Methods

Weighted PRS models were constructed and analyzed on 4602 individuals from the Asian Indian Diabetes Heart Study/Sikh Diabetes Study (AIDHS/SDS) as discovery/training and test/validation datasets. The results were further replicated in 9372 South Asian individuals from UK Biobank (UKBB). We also assessed the performance of each PRS model by combining data of the clinical risk score (CRS).

Results

Both genetic models (PRSAI and PRSEU) successfully predicted the T2D risk. However, the PRSAI revealed 13.2% odds ratio (OR) 1.80 [95% confidence interval (CI) 1.63-1.97; p = 1.6 × 10-152] and 12.2% OR 1.38 (95% CI 1.30-1.46; p = 7.1 × 10-237) superior performance in AIDHS/SDS and UKBB validation sets, respectively. Comparing individuals of extreme PRS (ninth decile) with the average PRS (fifth decile), PRSAI showed about two-fold OR 20.73 (95% CI 10.27-41.83; p = 2.7 × 10-17) and 1.4-fold OR 3.19 (95% CI 2.51-4.06; p = 4.8 × 10-21) higher predictability to identify subgroups with higher genetic risk than the PRSEU. Combining PRS and CRS improved the area under the curve from 0.74 to 0.79 in PRSAI and 0.72 to 0.75 in PRSEU.

Conclusion

Our data suggest the need for extending genetic and clinical studies in varied ethnic groups to exploit the full clinical potential of PRS as a risk prediction tool in diverse study populations.

Predicting type 2 diabetes risk before and after solid organ transplantation using polygenic scores in a Danish cohort

Type 2 diabetes mellitus (T2DM) can be multifactorial where both genetics and environmental factors play a role. We aimed to investigate the use of polygenic risk scores (PRS) in the prediction of pre-transplant T2DM and post-transplant diabetes mellitus (PTDM) among solid organ transplant (SOT) patients. Using non-genetic risk scores alone; and the combination with PRS, separate logistic regression models were built and compared using receiver operator curves. Patients were assessed pre-transplant and in three post-transplant periods: 0-45, 46-365 and >365 days. A higher PRS was significantly associated with increased odds of pre-transplant T2DM. However, no improvement was observed for pre-transplant T2DM prediction when comparing PRS combined with non-genetic risk scores to using non-genetic risk scores alone. This was also true for predictions of PTDM in all three post-transplant periods. This study demonstrated that polygenic risk was only associated with the risk of T2DM among SOT recipients prior to transplant and not for PTDM. Combining PRS with a clinical model of non-genetic risk scores did not significantly improve the predictive ability, indicating its limited clinical utility in identifying patients at high risk for T2DM before transplantation, suggesting that non-genetic or different genetic factors may contribute to PTDM.

Genetics of diabetes

Diabetes mellitus is a complicated disease characterized by a complex interplay of genetic, epigenetic, and environmental variables. It is one of the world's fastest-growing diseases, with 783 million adults expected to be affected by 2045. Devastating macrovascular consequences (cerebrovascular disease, cardiovascular disease, and peripheral vascular disease) and microvascular complications (like retinopathy, nephropathy, and neuropathy) increase mortality, blindness, kidney failure, and overall quality of life in individuals with diabetes. Clinical risk factors and glycemic management alone cannot predict the development of vascular problems; multiple genetic investigations have revealed a clear hereditary component to both diabetes and its related complications. In the twenty-first century, technological advancements (genome-wide association studies, next-generation sequencing, and exome-sequencing) have led to the identification of genetic variants associated with diabetes, however, these variants can only explain a small proportion of the total heritability of the condition. In this review, we address some of the likely explanations for this "missing heritability", for diabetes such as the significance of uncommon variants, gene-environment interactions, and epigenetics. Current discoveries clinical value, management of diabetes, and future research directions are also discussed.

Genetics of Type 2 Diabetes: Implications from Large-Scale Studies

PURPOSE OF REVIEW

Type 2 diabetes (T2D) is a multifactorial, heritable syndrome characterized by dysregulated glucose homeostasis that results from impaired insulin secretion and insulin resistance. Genetic association studies have successfully identified hundreds of T2D risk loci implicating many genes in disease pathogenesis. In this review, we provide an overview of the recent T2D genetic studies from the past 3 years with particular focus on the effects of sample size and ancestral diversity on genetic discovery as well as discuss recent work on the use and limitations of genetic risk scores (GRS) for T2D risk prediction.

RECENT FINDINGS

Recent large-scale, multi-ancestry genetic studies of T2D have identified over 500 novel risk loci. The genetic variants (i.e., single nucleotide polymorphisms (SNPs)) marking these novel loci in general have smaller effect sizes than previously discovered loci. Inclusion of samples from diverse ancestral backgrounds shows a few ancestry specific loci marked by common variants, but overall, the majority of loci discovered are common across ancestries. Inclusion of common variant GRS, even with hundreds of loci, does not substantially increase T2D risk prediction over standard clinical risk factors such as age and family history. Common variant association studies of T2D have now identified over 700 T2D risk loci, half of which have been discovered in the past 3 years. These recent studies demonstrate that inclusion of ancestrally diverse samples can enhance locus discovery and improve accuracy of GRS for T2D risk prediction. GRS based on common variants, however, only minimally enhances risk prediction over standard clinical risk factors.

Polygenic Prediction of Type 2 Diabetes in Africa

OBJECTIVE

Polygenic prediction of type 2 diabetes (T2D) in continental Africans is adversely affected by the limited number of genome-wide association studies (GWAS) of T2D from Africa and the poor transferability of European-derived polygenic risk scores (PRSs) in diverse ethnicities. We set out to evaluate if African American, European, or multiethnic-derived PRSs would improve polygenic prediction in continental Africans.

RESEARCH DESIGN AND METHODS

Using the PRSice software, ethnic-specific PRSs were computed with weights from the T2D GWAS multiancestry meta-analysis of 228,499 case and 1,178,783 control subjects. The South African Zulu study (n = 1,602 case and 981 control subjects) was used as the target data set. Validation and assessment of the best predictive PRS association with age at diagnosis were conducted in the Africa America Diabetes Mellitus (AADM) study (n = 2,148 case and 2,161 control subjects).

RESULTS

The discriminatory ability of the African American and multiethnic PRSs was similar. However, the African American-derived PRS was more transferable in all the countries represented in the AADM cohort and predictive of T2D in the country combined analysis compared with the European and multiethnic-derived scores. Notably, participants in the 10th decile of this PRS had a 3.63-fold greater risk (odds ratio 3.63; 95% CI 2.19-4.03; P = 2.79 × 10-17) per risk allele of developing diabetes and were diagnosed 2.6 years earlier than those in the first decile.

CONCLUSIONS

African American-derived PRS enhances polygenic prediction of T2D in continental Africans. Improved representation of non-European populations (including Africans) in GWAS promises to provide better tools for precision medicine interventions in T2D.

Genetic risk score to improve prediction and treatment in gestational diabetes mellitus

Diabetes mellitus is a chronic disease caused by the interaction of genetics and the environment that can lead to chronic damage to many organ systems. Genome-wide association studies have identified accumulating single-nucleotide polymorphisms related to type 2 diabetes mellitus and gestational diabetes mellitus. Genetic risk score (GRS) has been utilized to evaluate the incidence risk to improve prediction and optimize treatments. This article reviews the research progress in the use of the GRS in diabetes mellitus in recent years and discusses future prospects.

Circulating Nucleic Acid-Based Biomarkers of Type 2 Diabetes

Type 2 diabetes (T2D) is a deficiency in how the body regulates glucose. Uncontrolled T2D will result in chronic high blood sugar levels, eventually resulting in T2D complications. These complications, such as kidney, eye, and nerve damage, are even harder to treat. Identifying individuals at high risk of developing T2D and its complications is essential for early prevention and treatment. Numerous studies have been done to identify biomarkers for T2D diagnosis and prognosis. This review focuses on recent T2D biomarker studies based on circulating nucleic acids using different omics technologies: genomics, transcriptomics, and epigenomics. Omics studies have profiled biomarker candidates from blood, urine, and other non-invasive samples. Despite methodological differences, several candidate biomarkers were reported for the risk and diagnosis of T2D, the prognosis of T2D complications, and pharmacodynamics of T2D treatments. Future studies should be done to validate the findings in larger samples and blood-based biomarkers in non-invasive samples to support the realization of precision medicine for T2D.

Genetic risk scores for cardiometabolic traits in sub-Saharan African populations

BACKGROUND

There is growing support for the use of genetic risk scores (GRS) in routine clinical settings. Due to the limited diversity of current genomic discovery samples, there are concerns that the predictive power of GRS will be limited in non-European ancestry populations. GRS for cardiometabolic traits were evaluated in sub-Saharan Africans in comparison with African Americans and European Americans.

METHODS

We evaluated the predictive utility of GRS for 12 cardiometabolic traits in sub-Saharan Africans (AF; n = 5200), African Americans (AA; n = 9139) and European Americans (EUR; n = 9594). GRS were constructed as weighted sums of the number of risk alleles. Predictive utility was assessed using the additional phenotypic variance explained and the increase in discriminatory ability over traditional risk factors [age, sex and body mass index (BMI)], with adjustment for ancestry-derived principal components.

RESULTS

Across all traits, GRS showed up to a 5-fold and 20-fold greater predictive utility in EUR relative to AA and AF, respectively. Predictive utility was most consistent for lipid traits, with percentage increase in explained variation attributable to GRS ranging from 10.6% to 127.1% among EUR, 26.6% to 65.8% among AA and 2.4% to 37.5% among AF. These differences were recapitulated in the discriminatory power, whereby the predictive utility of GRS was 4-fold greater in EUR relative to AA and up to 44-fold greater in EUR relative to AF. Obesity and blood pressure traits showed a similar pattern of greater predictive utility among EUR.

CONCLUSIONS

This work demonstrates the poorer performance of GRS in AF and highlights the need to improve representation of multiple ethnic populations in genomic studies to ensure equitable clinical translation of GRS.

A Genetic Risk Score Improves the Prediction of Type 2 Diabetes Mellitus in Mexican Youths but Has Lower Predictive Utility Compared With Non-Genetic Factors

Background

Type 2 diabetes (T2D) is a multifactorial disease caused by a complex interplay between environmental risk factors and genetic predisposition. To date, a total of 10 single nucleotide polymorphism (SNPs) have been associated with pediatric-onset T2D in Mexicans, with a small individual effect size. A genetic risk score (GRS) that combines these SNPs could serve as a predictor of the risk for pediatric-onset T2D.

Objective

To assess the clinical utility of a GRS that combines 10 SNPs to improve risk prediction of pediatric-onset T2D in Mexicans.

Methods

This case-control study included 97 individuals with pediatric-onset T2D and 84 controls below 18 years old without T2D. Information regarding family history of T2D, demographics, perinatal risk factors, anthropometric measurements, biochemical variables, lifestyle, and fitness scores were then obtained. Moreover, 10 single nucleotide polymorphisms (SNPs) previously associated with pediatric-onset T2D in Mexicans were genotyped. The GRS was calculated by summing the 10 risk alleles. Pediatric-onset T2D risk variance was assessed using multivariable logistic regression models and the area under the receiver operating characteristic curve (AUC).

Results

The body mass index Z-score (Z-BMI) [odds ratio (OR) = 1.7; p = 0.009] and maternal history of T2D (OR = 7.1; p < 0.001) were found to be independently associated with pediatric-onset T2D. No association with other clinical risk factors was observed. The GRS also showed a significant association with pediatric-onset T2D (OR = 1.3 per risk allele; p = 0.006). The GRS, clinical risk factors, and GRS plus clinical risk factors had an AUC of 0.66 (95% CI 0.56-0.75), 0.72 (95% CI 0.62-0.81), and 0.78 (95% CI 0.70-0.87), respectively (p < 0.01).

Conclusion

The GRS based on 10 SNPs was associated with pediatric-onset T2D in Mexicans and improved its prediction with modest significance. However, clinical factors, such the Z-BMI and family history of T2D, continue to have the highest predictive utility in this population.

Non-communicable diseases pandemic and precision medicine: Is Africa ready?

Non-communicable diseases (NCDs) kill more than 41 million people every year, accounting for 71% of all deaths globally. The prevalence of NCDs is estimated to be higher than that of infectious diseases in Africa by 2030. Precision medicine may help with early identification of cases, resulting in timely prevention and improvement in the efficacy of treatments. However, Africa has been lagging behind in genetic research, a key component of the precision medicine initiative. A number of genomic research initiatives which could lead to translational genomics are emerging on the African continent which includes the Non-communicable Diseases Genetic Heritage Study (NCDGHS) and the Men of African Descent and Carcinoma of the Prostate (MADCaP) Network. These offer a promise that precision medicine can be applied in African countries. This review evaluates the advances of genetic studies for cancer, hypertension, type 2 diabetes and body mass index (BMI) in Africa.

Genetic Risk Score Increased Discriminant Efficiency of Predictive Models for Type 2 Diabetes Mellitus Using Machine Learning: Cohort Study

Background: Previous studies have constructed prediction models for type 2 diabetes mellitus (T2DM), but machine learning was rarely used and few focused on genetic prediction. This study aimed to establish an effective T2DM prediction tool and to further explore the potential of genetic risk scores (GRS) via various classifiers among rural adults.

Methods: In this prospective study, the GRS for a total of 5,712 participants from the Henan Rural Cohort Study was calculated. Cox proportional hazards (CPH) regression was used to analyze the associations between GRS and T2DM. CPH, artificial neural network (ANN), random forest (RF), and gradient boosting machine (GBM) were used to establish prediction models, respectively. The area under the receiver operating characteristic curve (AUC) and net reclassification index (NRI) were used to assess the discrimination ability of the models. The decision curve was plotted to determine the clinical-utility for prediction models.

Results: Compared with the individuals in the lowest quintile of the GRS, the HR (95% CI) was 2.06 (1.40 to 3.03) for those with the highest quintile of GRS (P_trend < 0.05). Based on conventional predictors, the AUCs of the prediction model were 0.815, 0.816, 0.843, and 0.851 via CPH, ANN, RF, and GBM, respectively. Changes with the integration of GRS for CPH, ANN, RF, and GBM were 0.001, 0.002, 0.018, and 0.033, respectively. The reclassifications were significantly improved for all classifiers when adding GRS (NRI: 41.2% for CPH; 41.0% for ANN; 46.4% for ANN; 45.1% for GBM). Decision curve analysis indicated the clinical benefits of model combined GRS.

Conclusion: The prediction model combined with GRS may provide incremental predictions of performance beyond conventional factors for T2DM, which demonstrated the potential clinical use of genetic markers to screen vulnerable populations.

Clinical Trial Registration: The Henan Rural Cohort Study is registered in the Chinese Clinical Trial Register (Registration number: ChiCTR-OOC-15006699). http://www.chictr.org.cn/showproj.aspx?proj=11375.

Systematic Review of Polygenic Risk Scores for Type 1 and Type 2 Diabetes

Recent studies have led to considerable advances in the identification of genetic variants associated with type 1 and type 2 diabetes. An approach for converting genetic data into a predictive measure of disease susceptibility is to add the risk effects of loci into a polygenic risk score. In order to summarize the recent findings, we conducted a systematic review of studies comparing the accuracy of polygenic risk scores developed during the last two decades. We selected 15 risk scores from three databases (Scopus, Web of Science and PubMed) enrolled in this systematic review. We identified three polygenic risk scores that discriminate between type 1 diabetes patients and healthy people, one that discriminate between type 1 and type 2 diabetes, two that discriminate between type 1 and monogenic diabetes and nine polygenic risk scores that discriminate between type 2 diabetes patients and healthy people. Prediction accuracy of polygenic risk scores was assessed by comparing the area under the curve. The actual benefits, potential obstacles and possible solutions for the implementation of polygenic risk scores in clinical practice were also discussed. Develop strategies to establish the clinical validity of polygenic risk scores by creating a framework for the interpretation of findings and their translation into actual evidence, are the way to demonstrate their utility in medical practice.

Genetic Basis of Obesity and Type 2 Diabetes in Africans: Impact on Precision Medicine

PURPOSE OF REVIEW

Recent advances in genomics provide opportunities for novel understanding of the biology of human traits with the goal of improving human health. Here, we review recent obesity and type 2 diabetes (T2D)-related genomic studies in African populations and discuss the implications of limited genomics studies on health disparity and precision medicine.

RECENT FINDINGS

Genome-wide association studies in Africans have yielded genetic discovery that would otherwise not be possible; these include identification of novel loci associated with obesity (SEMA-4D, PRKCA, WARS2), metabolic syndrome (CA-10, CTNNA3), and T2D (AGMO, ZRANB3). ZRANB3 was recently demonstrated to influence beta cell mass and insulin response. Despite these promising results, genomic studies in African populations are still limited and thus genomics tools and approaches such as polygenic risk scores and precision medicine are likely to have limited utility in Africans with the unacceptable possibility of exacerbating prevailing health disparities. African populations provide unique opportunities for increasing our understanding of the genetic basis of cardiometabolic disorders. We highlight the need for more coordinated and sustained efforts to increase the representation of Africans in genomic studies both as participants and scientists.

Comparing distributions of polygenic risk scores of type 2 diabetes and coronary heart disease within different populations

Polygenic risk scores are gaining more and more attention for estimating genetic risks for liabilities, especially for noncommunicable diseases. They are now calculated using thousands of DNA markers. In this paper, we compare the score distributions of two previously published very large risk score models within different populations. We show that the risk score model together with its risk stratification thresholds, built upon the data of one population, cannot be applied to another population without taking into account the target population’s structure. We also show that if an individual is classified to the wrong population, his/her disease risk can be systematically incorrectly estimated.