This article was edited to correct that Genomics Ltd. did not conduct the linked study on CAD.
NEW YORK – Pharmaceutical giant Novo Nordisk sees an unconventional business opportunity in disease prevention and has teamed up with Genomics Ltd. to focus on preventing obesity.
The collaborators aim to identify people at higher risk for developing obesity, using a combination of polygenic risk scores (PRSs) and clinical factors, and to develop tailored solutions to help those individuals avoid growing obese.
"Our mission, essentially, is [to] stop people from becoming patients in the first place, which is kind of unusual for the pharmaceutical company," said Nadeem Sarwar, corporate VP at Novo Nordisk and cofounder of the company's new Transformational Prevention Unit (TPU).
Founded in 2023, the TPU is a relatively new group within Novo Nordisk that researches scientifically valid and commercially viable ways to prevent obesity and its consequences.
Genomics and Novo Nordisk are currently working to develop and validate an obesity PRS, using primarily data acquired by Genomics from sources that include the All of Us database and the summary statistics from numerous published studies. Once this tool is ready, the companies plan to begin screening people for genetic obesity risk.
Although he said that it remains too early to provide specific examples, Sarwar explained that Novo Nordisk hopes that this effort will help it develop commercial products to both predict obesity and to limit its occurrence.
"The focus of this partnership is to identify people who don't have disease today but are at high risk of developing disease in the future," Sarwar said.
Polygenic risk factors for obesity have been the focus of a growing body of research recently. One study published last year, for example, published PRS for obesity and nine other chronic disorders. Another study from 2023 determined that the risk of cardiac complications among obese people varied depending on the relative contributions of genetics and lifestyle.
UK-based Genomics has been developing PRS for a variety of indications, including obesity, and working to bring them into the clinical setting. Last year, the company published data suggesting that the clinical implementation of a PRS for cardiovascular risk is well accepted by both patients and practitioners.
"At the moment, we're working out what are the right things to focus on in predicting obesity and making sure we can get the most powerful tools we can, and [then] validating those," said Sir Peter Donnelly, CEO and cofounder of Genomics.
As this phase of the partnership nears completion, Donnelly said that the two companies will begin working out how to proceed, such as whether any licensing agreement would be exclusive or not.
"It'll be a question of whether we think that's the right thing, both [in terms of] general good and also from the company's point of view," Donnelly said.
In addition to genetic data, the collaborators plan to incorporate other clinical data into their final risk estimations. Multiple scientific studies have shown that combining clinical factors such as social determinants of health, smoking history, and other pieces of relevant nongenetic data to PRS can improve PRS-based risk estimations.
Other studies, however, have suggested that the relative contributions of genetic and nongenetic factors show greater nuance and may vary by condition. One study conducted last year, for example, found that PRS for coronary artery disease (CAD) –– often a comorbidity of obesity –– are sometimes similar to risk estimations made using rare CAD-related monogenic mutations.
"It's unknown how informative genetic data alone are in predicting population-based or individual obesity risk in people who don't have obesity today," Sarwar said, "so we are really excited about working with Genomics to unpack that a little bit."
Philip Jansen, a clinical geneticist and assistant professor of human genetics at the University of Amsterdam Medical Center, largely echoed Sarwar's statement, saying that it is currently unclear how PRS and environmental factors should be combined for optimal risk assessments. Jansen added that other knowledge gaps exist with respect to PRS, which could complicate their path toward clinical deployment.
"[Genome-wide association studies] that PRS are based on have a focus on European ancestry and lack of insights into non-Western ancestries," Jansen said via email.
Jansen also noted that PRS are largely based on genetic variants related to body mass index, whereas obesity is a heterogeneous phenotype with multiple subtypes, such as metabolically healthy versus unhealthy obesity.
"More research is needed into the genetics of different obesity subtypes," he said.
The issue of limited non-European representation in GWAS is a known obstacle in PRS research and in human genetic research more broadly. Donnelly said that while more diverse data will improve the predictive power of PRS, Genomics has enough data from diverse ancestry groups to validate the company's methods.
"We're quite good at squeezing a lot out of whatever data we have," he said.
Sarwar said that Novo Nordisk's TPU is currently working with other groups to gather such diverse GWAS data from around the globe. Partners in this effort include the Danish Blood Donor Study, with whom Novo Nordisk will develop prediction models and precision screening tools for obesity and its consequences, and Mayo Clinic spinout Phenomix Sciences, which recently licensed its saliva-based obesity subtyping test to precision health firm InformedDNA.
Jansen cautioned that commercial PRS tests –– direct-to-consumer tests in particular –– still exist in something of a gray space with respect to their real-life utility and typically don't provide clear advice on what to do when an unfavorable PRS is found.
"Sometimes consumers are advised to seek help from a healthcare professional if they have questions," he said, "but doctors, even clinical geneticists like myself, often do not know what to advise. Moreover, it is unclear what advice should be given to those with a high polygenic risk score. Healthy diet and regular exercise can be advised without having to know your polygenic score."
Sarwar argued that a more quantitative assessment of one's obesity risk, such as that offered via PRS, might make a difference in nudging people toward healthier preventive behaviors.
"Incorporating PRS in preventative screening is a unique opportunity to quantify people's genetic risk of developing obesity and related cardiometabolic comorbidities before they manifest and engage individuals early on in their health journey to take preventative action," he said.
Additionally, Sarwar mentioned that combining PRS and other nongenetic variables will allow Novo Nordisk and Genomics to develop individually tailored integrated risk tests.
Although the goal is to identify people who are on a trajectory toward obesity at an early enough state to meaningfully intervene and to design relevant commercial products, Sarwar said that the first step is to determine whether a PRS can identify such people in the first place.
"Once we've developed, tested, and validated a PRS, the next step will be to think about where [to] pilot and administer those tests," Sarwar said.