Clinical diagnosis, treatment, and prognosis look vastly different than they did a decade ago. HINTS survey results indicate that an increasing portion of Americans have undergone some sort of genetic testing — 40% in 2022 compared to 19% in 2019. These tests are used to assess an individual’s risk of developing a disease, diagnose various health conditions, or determine an individual’s chances of passing on a genetic condition.
The next ten years promise to be even more transformative. Across the board, omics technologies are becoming more accurate, more comprehensive, and more affordable. Many countries, the United States included, have adopted newborn genetic screenings. The NIH recommends that germline testing be offered to all people with male breast cancer, ovarian cancer, pancreatic cancer, and metastatic prostate cancer. The upcoming generation will have more knowledge about their genes than any generation in history.
In this article, we discuss breakout omics technologies, some of which are already changing the landscape of clinical practice. Included are several groundbreaking studies in disease research, diagnosis, treatment, and prognosis.
Long-read sequencing technologies generate significantly longer contiguous segments than were previously possible. The PacBio Revio, for example, can produce DNA or RNA reads up to 20,000 bases in length. This is a major step forward compared to short-read sequencing, which averages contiguous reads of between 50 and 300 base pairs.
With this expanded resolution, disease researchers can better assemble highly repetitive genetic regions, including GC-rich regions. This has led to many novel discoveries in disease pathogenesis. For example, a 2019 project used long-read sequencing in two siblings experiencing progressive myoclonic epilepsy. Whole exome sequencing had already been employed, and had failed to identify a causal genetic variant.
PacBio long-read low-coverage whole-genome sequencing was used to identify over 17,000 structural variants. These were then narrowed down to one 12.4 kb deletion involving the CLN6 gene. This deletion was found in a highly repetitive region, and would not have been easy to resolve with short-read sequencing.
Long-read sequencing has also been applied to oncology, where it has uncovered countless insights into cancer genes and functions. A 2024 study used PromethION long-read sequencing technology to generate a dataset of 184 patient tumors. This project showed long-read sequencing’s potential in improving our understanding of complex structural variants, viral integrations, and extrachromosomal circular DNA. Several genetic mutations were identified as probable disease factors.
We’ve long known that protein expression is a powerful measure of disease functions. Companies like Olink Proteomics offer exploratory protein technologies, as well as panels pre-designed to investigate specific disease states.
An Olink PEA study on >50,000 members of the UK Biobank was used to identify a circulating protein signature that predicts the most common form of dementia 10 years before symptoms appear. A different group of researchers used UK Biobank data to identify a blood-based protein signature predictive of Alzheimer’s disease, years before the onset of symptoms.
What makes these two studies incredibly valuable is that they have given researchers novel drug targets that may actually prevent these devastating neurodegenerative diseases from ever happening. The data coming out of the UK Biobank have been so promising that they have committed to sampling 600,000 more subjects using Olink’s HT platform.
In some instances, robust proteomic analysis holds promise for diagnostic challenges. For example, clinical tests for C-Reactive Protein (CRP) and IL-6 production are currently used to diagnose viral and bacterial infections. Because CRP and IL-6 production are upregulated in both types of infection, people with viral infections are sometimes misdiagnosed and prescribed ineffective antibiotics.
A study on 139 patients using broad biomarker profiling was able to identify host-proteins with a better chance of successfully categorizing viral or bacterial infection diagnosis than CRP and IL-6. Researchers were then able to develop a three-protein signature with a better rate of successful diagnosis than any individual protein.
Proteomics has also changed diagnostic best practices in oncology. A novel diagnostic test for 18 early-stage solid tumors was developed by researchers using circulating proteins. Scientists were able to develop a specific 160 protein-based signature for men and a separate 160 protein-based signature for women that can detect stage I cancer and identify the location of the tumor. The best odds of beating any cancer are to detect it as early as possible.
Other research has shown promise in protein profiling for risk stratification of emergency department admissions. With biomarkers of inflammation, stress, and infection, testing successfully predicted risk of death and ICU admission. Early insights like these are invaluable for determining the necessary treatment urgency and intensity.
Additional single-cell technologies are broadening the viability conditions for running a single-cell experiment. Cryopreserved live cells, frozen tissue, barcoded cells, small sample size, and even budget or instrument availability are not the barriers they once were. These technological developments have made it possible to apply single-cell sequencing across a wide range of scientific disciplines.
Oncological research has benefited greatly from single-cell sequencing technology, particularly due to cancer’s heterogeneous nature. Single-cell technologies are able to generate data on many different cell types in a sample, including rare and low-expression transcript information. Bulk RNA sequencing, by comparison, can only provide an average output of the transcriptomic content of the cell population.
Multiple research projects out of the University of California, Irvine used single-cell technology to better understand human epithelial cells in breast cancer. Prior to this study, research had proposed that several subcategories of breast epithelial cells exist.
However, this profiling effort was able to identify three epithelial cell subtypes. These populations also appeared to have additional subclusters, hypothesized to be the epithelial cells at different cell states over time. Different subtypes of epithelial cells correspond to subtypes of breast cancer, and are part of functionally distinct lineages. Understanding subtypes of cancers at their earliest stages is an important step forward in personalized treatment and early detection. Single-cell analyses like these further those efforts in a way that was previously impossible.
Spatial biology imaging capabilities have changed what’s possible in terms of understanding disease pathogenesis, spread, and function. In HIV for example, infection causes significant changes in tissues all over the body. Because the virus is difficult to detect in early post-transmission stages, little is known about this phase of infection. However, spatial imaging has led to a new level of understanding in the viral microenvironment.
A recent HIV study segmented and classified cells based on expression, location in tissue, and co-localization with virus signal. Spatial analysis found that the virus was most enriched in DCs in mucosa and macrophages in the submucosa, as opposed to finding this enrichment in the CD4+T cells as expected.
Lymphoid follicles were also identified as containing the majority of the virus, which is important to understanding the spread of HIV and the ways it can bypass the body’s defense systems. HIV-positive tissue areas contained clusters of specific HIV target cells, indicating that these clusters may promote rapid viral transfer immediately after transmission, rather than transfer occurring later in the infection via lymph nodes. These insights are critical to improving early detection, treatment, and prevention of HIV and other diseases where spatial biology can broaden our understanding.
In another example, spatial biology capabilities were tested in companion diagnostics for oncological targeted therapeutics. Historically, microscopy has been the standard for companion diagnostics. This approach is used to detect the presence or absence of a single marker protein, which is then used to select a treatment with the best probability of success.
However, the more we learn about the complexity of disease states, the less one biomarker is sufficient to predict treatment efficacy. In this immuno-oncology study, spatial phenotyping was shown to have the highest predictive value for immunotherapy success when compared to next-generation sequencing, RNA expression, and standard PLD1 IHC testing.
The true power of these developing technologies can be seen when they’re combined. Multiomic studies promise to improve the depth and confidence of understanding around disease conditions.
Proteogenomics, one of these hybrid disciplines, merges the insights available from proteomic, transcriptomic, and genomic sequencing technologies. This approach can validate the information discovered in DNA and RNA sequencing. It can also be used to better explain the relationship between the genome and the proteome, or between proteins and other proteins.
Once again, large datasets have improved what’s possible in clinical research. In a proteomic study on ovarian cancer, researchers conducted proteomic characterization of 174 tumors analyzed by The Cancer Genome Atlas. By integrating proteomic analysis with the existing genomic data, they were able to identify how the genome drives changes in the proteome in ovarian cancer. They identified factors associated with short overall survival, a potential means of patient stratification, and factors associated with treatment outcomes.
Often, researchers also combine single-cell and spatial biology tools. In many diseases, it’s important to understand the types of cells present, and their spatial organization. In a study on head and neck squamous cell carcinoma, researchers compiled single-cell composition data on nine patients. Then, they conducted deeper spatial analysis of the tumor-immune microenvironment. They were able to identify distinctions in microenvironment landscape that may impact clinical outcomes.
Multiomic approaches have revolutionized diagnostic, prognostic, and predictive medicine like never before. New technologies like long-read sequencing, improvements to single-cell technologies, and the advent of spatial biology have delivered fantastic new discoveries. Partnering these technologies with large GWAS and proteomic datasets has enabled researchers to find novel biomarkers that identify disease earlier, inform the best treatments, and predict outcomes for a variety of disease states. These evolving disciplines and technologies are essential pieces in shaping the future of patient care.