Clinical OMICS

JAN-FEB 2019

Healthcare magazine for research scientists, labs, pathologists, hospitals, cancer centers, physicians and biopharma companies providing news articles, expert interviews and videos about molecular diagnostics in precision medicine

Issue link:

Contents of this Issue


Page 47 of 51

46 Clinical OMICs January/February 2019 vajal-Carmona said in a press release. Another reason, he added, is that there are not enough data and models for the scientific community to develop precision medicine studies with minority populations included. Carvajal-Carmona said that while gastric cancer is a lead- ing cancer killer of Latinos, the largest gastric cancer genetic sequencing study to date, which analyzed 300 tumors, included only three tumor samples from patients of Latino ancestry. In 2017, Carvajal-Carmona and his colleagues con- ducted a study in which they established that PALB2—a gene involved in DNA repair—as a gastric cancer gene. For 20 years previously, only one gene, CDH1, had been iden- tified as being associated with gastric cancer. But, beyond basic research, Carvajal-Carmona said the new grant will be focused on conducting clinical trials in minority popula- tions. "We want to generate a body of data that can be used in clinical trials and move from cancer disparity to cancer equity," Carvjal-Carmona told Clinical OMICs. Likewise, Esteban Burchard, M.D., is hoping a new birth-cohort study of Puerto Ricans will have an impact on the rates of asthma in that population. "Asthma has the big- (continued from previous page) Refining Genetic Risk Scores By some accounts, the field of clinical genomics is poised to expe- rience a breakthrough thanks to the innovation of polygenic risk scores, also called genetic risk scores (GRS). These scores, which are determined by genome-wide association studies (GWAS), are based on variation in multiple genetic loci and their associated weights. Researchers believe these numbers serve as the best prediction for the disease that can be made when taking into ac- count variation in multiple genetic variants. Some researchers, however, believe that the lack of samples from non-white populations results in GWAS that paint biased pictures and cannot always be generalized to other populations. "Most of the existing genome-wide association studies haven't really reflected our planet's diversity. They are basically massively enriched with European samples," said Joseph Lachance, Ph.D., an assistant professor of biological sciences at the Georgia Institute of Technology. In a November 2018 study published in Genome Biology, Lachance and his colleagues used GWAS to see how often they would get genetic predictions of disease risk and how much they would vary by ancestry or by location. "What we found is that when you take published GWAS results, it turns out that your pre- dictions of African risk are too high or too low. We're doing a poor job of predicting risk for individuals that have genomes that are largely African. One major byproduct of this is that you can trans- late results from the U.K. to Denmark, but it doesn't work if trying to translate findings from the U.K. to Senegal." On the other hand, after studying the genetics of hypertension in populations of the African diaspora, Richard Cooper, M.D., now believes the empirical support for the common sense argument of diversifying studies is limited—especially when it comes to predicting risk using polygenic risk scores. "New discoveries will occur in new populations, but the overall value of omics needs to be better defined before one can actually assess the potential importance of widespread application of risk assessment and pre- dictive testing," said Cooper, who is professor and chair of public health sciences at Stritch School of Medicine at Loyola University Chicago. He also wrote: "As genomics has started to address di- rect clinical problems (risk predication, etc.), it has become clear that it has limited (not zero) utility, outside (potentially) cancer." According to Cooper, the push for broader studies of molecular genetics resulted from the availability of technology for cheap, fast genotyping and sequencing. This served a very useful pur- pose in clarifying basic issues, but was not helpful for complex traits—like blood pressure and obesity—related to heritable risks. Cooper said that the value of GRS for conditions like hyperten- sion and coronary heart disease has been grossly overstated. "The GRS methods—now seeking commercial application—yield triv- ial amounts of new information and their value added has never been rigorously assessed." Still, researchers developing GRS in other diseases remain con- vinced that diversifying databases and clinical research studies is the right thing to do, socially and scientifically. "Most of the world does not look like Europe," Lachance said. "You don't want to have only a subset of the world's population benefitting from (omics research)," Lachance said. "If you are trying to match genetics with health, it's important that you are not looking at a subset of the world's population…You need to reflect the diversity of the world as best as you can." —Camille Mojica Rey n Research from November by Joseph Lachance, Ph.D., and colleagues at the Georgia Institute of Technology found that GWAS does a poor job of predicting disease risk among the African population due to a lack of relevant sample data.

Articles in this issue

Links on this page

Archives of this issue

view archives of Clinical OMICS - JAN-FEB 2019