Genomic medicine year in review: 2024

Introduction

Starting in 2019, AJHG has published an annual feature^1,2,3,4,5 identifying ten key advances in applying genomic information to clinical care that were reported in the previous 12 months of published literature. The Genomic Medicine Working Group of the National Advisory Council for Human Genome Research of the National Human Genome Research Institute (NHGRI) has authored these reviews based on its broader effort to identify notable accomplishments in genomic medicine on a monthly basis and posts them on a searchable website (see web resources). From this larger set of published accomplishments, the Working Group has continued to select ten papers annually to be highlighted as the most significant.

This year the Working Group has selected its ten most significant advances among the 21 recognized works published during the year ending August 31, 2024. This is fewer than the total numbers of papers recognized in 2023 (34), 2022 (38), 2021 (30), 2020 (45), and 2019 (47). Criteria for selecting publications and highlights were similar to 2019–2023 (Box 1), though somewhat more emphasis may have been placed on accomplishments that were truly novel or first of their type. Population-based screening, polygenic risk prediction, and associations with actionable variants continued to be well represented among topics covered. Highlighted papers are listed and described in order of their publication below.

Box 1. Criteria for inclusion of papers in genomic medicine rear in review, 2024.

• •Involve use of patients’ genomic variant information in clinical decision making
• •Demonstrate impact of direct clinical implementation on health outcomes or behaviors
• •Demonstrate the potential for clinical implementation
• •Are likely to be generalizable beyond original setting
• •Are likely to have implications for healthcare systems or practice guidelines
• •Are important considerations for diversity and health equity
• •Are sufficiently large and rigorous to overcome sampling error and other bias
• •Are broadly representative of the field beyond NHGRI-sponsored or US-funded programs

Testing and managing iron overload after genetic screening-identified hemochromatosis

Savatt, J.M. et al. (2023). Testing and management of iron overload after genetic screening-identified hemochromatosis. JAMA Netw. Open 6(10):e2338995.

Savatt and colleagues used the Geisinger MyCode biobank to assess whether screening for HFE p.Cys282Tyr homozygosity is associated with recognition and management of asymptomatic iron overload in a rural, integrated health care system. The authors identified 201 participants with a p.Cys282Tyr homozygous result via exome sequencing, of whom 57 (28%) were also identified clinically as having hereditary hemochromatosis, and 86,300 participants negative for p.Cys282Tyr homozygosity. Thus, 144 participants first learned of their homozygosity status through genomic screening. Following result disclosure, 99 of 144 (69%) participants identified only through MyCode had recommended follow-up testing. Of the 144 participants, 52 (36%) met laboratory or liver biopsy criteria for iron overload, 36 (69%) of whom began phlebotomy or chelation therapy. This study demonstrates the actionability of genomic screening results, confirms that iron overload is underdiagnosed among those with HFE p.Cys282Tyr homozygosity, and documents the ability of genomic screening to prompt relevant health behaviors.

Actionable genotypes and their association with lifespan in Iceland

Jensson, B.O. et al. (2023). Actionable genotypes and their association with life span in Iceland. N Engl J Med. 389(19):1741–1752.

The prevalence of actionable genetic variants (using ACMG v3.0, 73 genes) and their associations with life span and cause of death were investigated among 57,933 Icelanders. Manual curation of 4,405 genetic variants identified 235 actionable genotypes across 53 genes; 2,306 participants (4%) carried at least one actionable genotype. Carriers had a slightly shorter median life span than noncarriers (86 vs. 87 years) while 10% of carriers died before reaching age 69 compared to 10% of non-carriers who died before age 73. Notably, carriers of actionable genotypes in cancer-related genes had a significantly shorter median life span—by 3 years—compared to noncarriers. Actionable variants in two cardiovascular genes, LDLR and MYBPC3, were also associated with shorter life spans. This study highlights the potential impact of actionable genetic variants on life expectancy and potential importance of population-level genomic screening; however, the reduced genetic variability of the Icelandic population limits its generalizability.

Impact of digitally enhanced genetic results disclosure in diverse families

Suckiel, S.A. et al. (2023). The NYCKidSeq randomized controlled trial: Impact of GUÍA digitally enhanced genetic results disclosure in diverse families. Am J Hum Genet. 110(12):2029–2041.

To address the need for more efficient genetic counseling as genetic testing increases, the authors developed a web-based tool, GUÍA (Genomic Understanding, Information, and Awareness), designed to capture key elements of a return of genetic test results session in Spanish or English at an accessible reading level supported by illustrations. Among 551 diverse families in whom genetic testing was ordered for children, 270 were randomized to GUÍA-directed return and 281 to standard of care. The primary endpoints were participants’ perceived understanding of and confidence in explaining genetic test results assessed by surveys at baseline, post-disclosure, and at six months. Use of the tool improved both early and late perceived understanding, with no effect on confidence. The effectiveness of GUÍA in improving understanding was most significant in Hispanic/Latino(a) individuals. This randomized trial supports the further development of digital approaches to facilitate the delivery of genetic test results in diverse populations.

Chronic disease polygenic risk scores for clinical implementation in diverse US populations

Lennon, N.J. et al. (2024). Selection, optimization, and validation of ten chronic disease polygenic risk scores for clinical implementation in diverse US populations. Nat Med. 30, 480–497.

Lennon et al. described the selection and optimization of 10 polygenic risk scores (PRSs) for use in the Electronic Medical Records and Genomics (eMERGE) Network. Ten scores were selected from among 23 nominated conditions based on PRS performance characteristics, medical actionability, potential clinical utility, and evidence from diverse ancestry populations. Three conditions were cardiovascular, three metabolic, two cancer, one respiratory, and one renal. Genetic ancestry was used to calibrate PRS mean and variance utilizing genetically diverse data from 13,475 participants of the All of Us Research Program. Results for the first 2,500 eMERGE participants (of whom only 1/3 self-identified as White) showed that over 20% had a high PRS for at least one of the selected conditions; these distributions appeared independent of genetic ancestry. This paper informs the approach for developing PRS-based testing that accurately reflects disease risk in diverse populations and supports further studies of integrating PRS in clinical settings.

Skeletal muscle ryanodine receptor 1 variants and malignant hyperthermia

Yu, K.D. et al. (2024). Evaluation of malignant hyperthermia features in patients with pathogenic or likely pathogenic RYR1 variants disclosed through a population genomic screening program. Anesthesiology. 140(1), 52–61.

To identify the incidence of malignant hyperthermia (MH) and potentially fatal hypermetabolic response to triggering anesthesia agents, records of unselected carriers of pathogenic or likely pathogenic genetic variants of the most common MH susceptibility gene—skeletal muscle ryanodine receptor 1 (RYR1)—were reviewed as disclosed through the MyCode Community Health Initiative. No participant had prior documented genetic testing for MH susceptibility. Among 152 participants with an actionable RYR1 variant, 68 (45%) had anesthesia records documenting exposure to a triggering agent. None, however, had documented clinical features of MH (muscle rigidity, myoglobinuria, hyperkalemia, elevated creatine kinase, severe myalgia, dark urine), and none received dantrolene treatment. Of 120 participants with possibly MH-related findings, 112 (93%) were considered unlikely to be MH events, and 8 (7%) had insufficient records to determine etiology. These results show that MH events in response to triggering anesthesia agents are infrequent in an unselected population with actionable RYR1 variants.

Treating inherited retinal disease with gene editing

Pierce, E.A. et al. (2024). Gene editing for CEP290-associated retinal degeneration. N Engl J Med. 390(21):1972–1984.

Pierce et al. investigated using CRISPR-Cas9 genome editing to treat inherited, severe early-onset vision loss in patients with homozygosity or compound heterozygosity for an IVS26 pathogenic variant in CEP290 (centrosomal protein, 290 kDa). CEP290 is associated with Leber congenital amaurosis 10 (LCA10), the most common inherited blindness disorder in children. A phase 1–2, open-label trial was conducted that included 14 participants aged 3 and older, each receiving a single subretinal injection of the gene-editing complex (EDIT-101) at low, intermediate, or high dose in their worse eye. The study’s primary focus was on safety, with visual acuity, retinal sensitivity, mobility, and quality of life measured as key secondary outcomes. No serious adverse events or dose-limiting toxic effects were observed in the study. Participants experienced notable improvements in vision and quality of life, indicating the promising therapeutic potential of in vivo gene editing for treating inherited degenerative retinal disease.

Validation of a clinical breast cancer risk assessment tool for all ancestries

Mabey, B. et al. (2024). Validation of a clinical breast cancer risk assessment tool combining a polygenic score for all ancestries with traditional risk factors. Genet Med. 26(7):101128.

Mabey et al. used longitudinal insurance claim data to validate a combined risk score that integrates a polygenic risk score with the widely used Tyrer-Cuzick model to assess breast cancer risk. The authors linked genetic test and tokenized linked claim data from 130,000 women ages 18–84 referred for hereditary cancer genetic testing and negative for breast cancer germline genomic variants to identify incident invasive breast cancer diagnoses. By following these women over about 148,000 person years (median follow up of 12.1 months) they demonstrated that the PRS model performs better than Tyrer-Cusick alone and the combined model had a roughly 2-fold great discriminatory accuracy than either individual model. Not only does this approach provide strong validity data for the combined risk score, but it also demonstrates the effective use of real-world data across different healthcare systems and payers that could be replicated across a wide range of predictive models.

Broader access to clinical genome sequencing benefits diverse individuals with rare diseases

Thorpe, E. et al. (2024). The impact of clinical genome sequencing in a global population with suspected rare genetic disease. Am J Hum Genet. 111(7):1271–1281.

Thorpe and colleagues investigated the success of clinical genome sequencing (cGS) in a diverse cohort of individuals from both high-income countries (HICs) and low- and middle-income countries (LMICs), considering the impact of such testing on health disparities. Providing cGS through the philanthropic iHope program to 1,004 international, diverse individuals meeting criteria for a rare and undiagnosed disease, the authors evaluated diagnostic rates and changes in management. Individuals from LMICs were 1.7 times more likely than those from HICs to have a positive test (57% vs. 34%), and their likelihood of a positive test changing care was not significantly different from HIC individuals. Those from HICs were more likely to have had a prior genetic test and were more likely to have had parental data to consider. The authors suggest that greater access to cGS is indicated overall and that deployment of cGS to LMICs could reduce resource-dependent health disparities.

Benefits for children with suspected cancer from routine whole-genome sequencing

Hodder, A. et al. (2024). Benefits for children with suspected cancer from routine whole-genome sequencing. Nat Med. 30:1905–1912.

This study aimed to determine whether offering whole-genome sequencing (WGS) to children in England suspected to have cancer improves patient management when compared to current standard of care. Genetic variants detected by WGS in 281 children across two English cancer centers, together with clinical and diagnostic information, were collected and compared to standard-of-care genetic panel results. Variants uniquely identified by WGS provided clinical benefit in 80 instances from 69 cases (24%), aiding diagnoses, providing additional therapeutic options, or changing clinical management. Additional disease-relevant variants were detected in 108 instances from 83 cases (29%). Notably, WGS identified novel disease-defining variants in 6 cases. WGS not only reproduced the findings of panels typically employed, but also identified previously unknown genomic features of childhood tumors. The study concluded that WGS improved patient management when implemented in routine clinical practice and may be useful for children with neoplastic disorders.

Clinical signatures of genetic epilepsies precede diagnosis in electronic medical records

Galer P.D. et al. (2024). Clinical signatures of genetic epilepsies precede diagnosis in electronic medical records of 32,000 individuals. Genet. Med 26(11):101211.

The authors aimed to identify early clinical features suggestive of genetic diagnoses in individuals with epilepsy through large-scale analyses of electronic medical records. A Natural Language Processing pipeline was applied to extract clinical features from the clinical notes of 32,112 individuals with childhood epilepsy, including 1,925 with genetic epilepsies. Age-dependent associations between known epilepsy genes and clinical features were then applied to train machine learning models to classify patients as having or not having a genetic diagnosis. They identified nearly 48,000 age-dependent associations of clinical features with genetic etiologies 3.6 years (median) before molecular diagnosis. For example, neurodevelopmental differences between ages 6 to 9 months increased the likelihood of a later molecular diagnosis 5-fold. They also produced a machine learning model that reliably predicted a later diagnosis of SCN1A-related disorders. Such models are promising to apply for earlier diagnosis and implementation of targeted therapeutic strategies in genetic epilepsies.

Conclusion

Progress in genomic medicine continues to grow, expanding on prior research findings and moving into new areas. This growing body of knowledge and the confirmatory research it engenders may reduce the likelihood that a subsequent paper would be considered a notable advance, regardless of its importance. This may be reflected in the smaller number of papers cited by the Genomic Medicine Working Group as notable accomplishments in 2024 than in prior years. Because recognition as a significant advance over existing literature and practice was not one of the Working Group’s stated criteria, it will be added for the 2025 year in review.

Highlighted papers continued themes from earlier years, particularly in genetic diagnosis of rare syndromes (Thorpe et al.), population-based screening for monogenic syndromes (Savatt et al. and Yu et al.), and safety and efficacy of gene editing (Pierce et al.). Polygenic risk prediction (Lennon et al. and Mabey et al.) and associations of actionable genotypes (Hodder et al. and Jensson et al.) with cancer treatment and lifespan were also highlighted. Clinical informatics research demonstrated the value of a digital assistant in results disclosure (Suckiel et al.) and diagnosis of genetic epilepsies from electronic medical records (Galer et al.). Notably, no papers in pharmacogenomics or neonatal genome sequencing were highlighted this year, though incremental progress in these fields continues. Genomic medicine resources continued to comprise a large share of the papers recognized as notable accomplishments, again reflecting the growing body of evidence of the feasibility and practical value of genomic medicine implementation in clinical care.

Acknowledgments

The authors gratefully acknowledge the assistance of Jacob Baroch in preparation of this manuscript.

Web resources

• NHGRI, Accomplishments in Genomic Medicine, https://www.genome.gov/health/Genomics-and-Medicine/accomplishments