Genomics in 2025: How $500 whole genome sequencing could democratize genomic data

[Adobe Stock]

How is biobank data being practically applied in drug development? How might this evolve in 2025? 

Neil Ward

Neil Ward: Biobanks provide large-scale datasets that are invaluable for understanding the genetic factors that influence health and disease among populations. Biobank data can pinpoint new genetic variants associated with a certain disease, helping to uncover new drug targets.  For example, biobank data validated the PCSK9 gene as a target for cholesterol-lowering drugs.

Advances in whole genome sequencing technologies, now more affordable and scalable, will drive the evolution of biobanks in 2025. To date, many biobanks have relied on traditional sequencing methods to capture participants’ DNA, but these fail to capture larger and more complex variants associated with conditions including rare diseases and cancer. Long-read sequencing, such as HiFi, offers a highly accurate and comprehensive view of the genome, and can now be achieved for just $500. These developments will enable biobanks to generate high quality sequencing data to fuel research on more challenging diseases.

Another historic challenge has been biobank size, with most not including enough of the population to enable meaningful insights at scale. The UK Biobank is considered a leader, with data from 500,000 participants – but this still only represents 0.7% of the population. Increasing participation is key to unlocking deeper biological insights. One biobank leading the way on both sequencing technology and size is the Estonian Biobank, which holds data from a fifth (20%) of Estonia’s population, reflecting age, sex, and geographical distribution. The Estonian team is carving a path other nations will hope to follow in 2025 and beyond.

Please share some concrete examples of pharmacogenomic applications improving patient outcomes. How might these evolve in 2025?

Ward: The past year has seen great strides in personalized cancer vaccines, which leverage pharmacogenomics (PGx) to tailor treatments to individuals’ tumor profiles. Current immunotherapies, specifically checkpoint inhibitors, are only effective in 20-40% of patients. But new research is showing the dual action of personalized vaccines with these drugs could increase the effectiveness of treatments. The approach involves identifying unique mutations in a patient’s cancer cells, known as neoantigens, and designs vaccines that prompt the immune system to target these specific markers, so the treatment is tailored to a patient’s tumor biology.

Most PGx tests being deployed today are focused on a small number of single nucleotide polymorphisms (SNPs) in a single gene. This is because many of the most important genes involved in PGx are hard to assay, so focused projects are simply more feasible. The difficulty of understanding PGx genes is also why most health systems aren’t preventative – they wait until someone is ill and treat patients with drugs that will alleviate symptoms as quickly as possible, rather than proactively sequencing healthy individuals and intervening early.

Looking ahead, long-read sequencing definitely holds potential to uncover more relationships between genetic features and drug response. Over the next year, the significant leaps in the field will come from ambitious, larger PGx research projects that seek to compare health records, drug response, and genetic data at the population level. The Estonian Biobank is again a great example, having invested in long-read sequencing for 10,000 participants with a goal of identifying and understanding challenging PGx genes. It’s an exciting project that could lead to Estonia being one of the first countries to implement precision medicine at scale. More broadly, in 2025 PGx progress will be accelerated by new guidance aimed at further facilitating pharmacogenomics research. For example, the UAE’s  Pharmacogenomics (PGx) Guideline published in June 2024, NHS England’s Pharmacy Genomics Workforce, Education and Training Strategic Framework published in January 2024, or the FDA’s Pharmacogenomic Data Submission guidance from 2023.

Can you cite specific advances in rare disease diagnosis through genomic sequencing. What trends did we see in 2024 here?

Ward: 2024 saw transformative leaps in rare disease research, specifically conditions linked to repeat expansions. Disorders such as amyotrophic lateral sclerosis (ALS), Friedreich’s ataxia, and Huntington’s disease have long remained elusive to researchers due to the inability of existing technologies to accurately profile the “dark regions” of the genome where repeat sequences reside. But recent breakthroughs in next generation long-read sequencing such as HiFi can accurately size the repeats, understand the repeat sequence and assess the epigenetic status of these challenging regions. These insights hugely increase our ability to correctly diagnose rare conditions.

2024 also emphasized the vital role of research consortia in solving rare diseases. The Undiagnosed Hackathon held in the Netherlands gathered 120 experts from 28 countries, including doctors, geneticists, and bioinformaticians. In a 48-hour sprint, the teams diagnosed ten conditions in just two days, a significant breakthrough for families who had been waiting years for answers. Key to these discoveries was giving researchers access to the most advanced genomic technologies able to capture the challenging, yet medically relevant variants associated with rare diseases. Initiatives like the hackathon not only drive investment in advanced technologies by hospitals and research institutions but also engage and educate the rare disease community, fostering further innovation.

The role of genomic data in making clinical trials more precise/efficient? 

Ward: Genomics holds potential to greatly improve the recruitment, characterization, and stratification of clinical trial participants. For example, sponsors and CROs can better define patient groups and increase the likelihood of participants benefiting from specific therapies by selecting those with clinically-relevant structural variants, insertions, deletions, and tandem repeats. In theory, advanced sequencing technologies, like long-read sequencing, could even uncover disease states by analysing epigenetic modifications. Since epigenetic changes are reversible, these insights allow trials to more precisely assess how interventions like drugs or lifestyle changes impact disease states over time.

Despite these advantages, scaling genomic insights across clinical development has proven challenging. In clinical trials, sponsors only really want to measure things that are already known about – limiting the number of unknowns. Using sequencing technology could uncover lots of genetic variants of unknown clinical significance, introducing uncertainties into trials. Another major hurdle is the lack of universally accepted standards for collecting, storing, and analyzing genomic data, making integration into diverse trial settings complex. This is compounded by evolving discussions around data privacy and security.