When genome-wide association studies (GWAS) first started appearing 10 years ago, they were heralded as the answer to connecting human genetic variation to human disease. These kinds of studies—which sift population-level genetic data—have revealed thousands of genetic variations associated with diseases, from age-related macular degeneration to obesity to diabetes.
However, thus far GWAS have largely come up short when it comes to finding new therapies. Few significant drug targets have come to light based on GWAS data (though some studies suggest that these studies could help drug makers find new uses for existing molecules).
Part of the problem may be that, until now, the right tools haven’t been available to exploit GWAS data. But a few recent studies—including two out of Dana-Farber/Boston Children’s Cancer and Blood Disorders Center—have used GWAS data to identify therapeutically promising targets, and then manipulated those targets using the growing arsenal of gene editing methods.
Does this mean that GWAS’ day has finally come?
We have the data—now what?
“What we’re learning with GWAS is how a system may be perturbed to get a very specific, clinically relevant disease phenotype,” says Daniel Bauer, MD, PhD, a hematologist-oncologist with Dana-Farber/Boston Children’s. “I would say that’s what you want to do with a therapy, but in the other direction–to understand how to decrease susceptibility to disease rather than increasing it.”
Maybe, Bauer says, we need to look at GWAS data from a different perspective.
For starters, he notes, it was assumed that GWAS would, by and large, turn up variations in genes’ coding regions. That hasn’t been the case. “GWAS really have predominantly identified non-coding sequences. Initially that was viewed as an intractable challenge, because no one knew what to do with them.”
But those non-coding sequences have largely turned out to be in key regulatory regions of the genome, such as enhancers, which help turn genes on and off in a cell type- or tissue-specific context. “Now we’re not so confused about what these variations might be doing, and we can link them to the genes they control.
“At least with hematopoietic disorders,” he adds, “you can then imagine editing an enhancer using genome editing tools.”
Making the right cuts in the right place
Bauer knows first hand how this strategy could work. He and his mentor Stuart Orkin, MD (chairman of pediatric oncology at Dana-Farber Cancer Institute and associate chief of hematology/oncology at Boston Children’s Hospital), scanned GWAS data for naturally occurring variations in an enhancer regulating a gene called BCL11A—a genetic switch that controls hemoglobin production in red blood cells (RBCs). They then used CRISPR gene editing to make targeted cuts to the enhancer in blood stem cells in vitro.
As the pair reported in Nature, the RBCs that grew out of the edited stem cells contained substantially higher-than-normal amounts of fetal hemoglobin—a form of the oxygen-ferrying complex unaffected by the sickle cell disease (SCD) mutation—opening the door to targeting this enhancer genetically to treat SCD.
“We’ve now targeted the modifier of the modifier of a disease-causing gene,” explains Orkin. “It’s a very different approach to treating disease.”
Theirs is not the only study to try this kind of GWAS-guided genetic approach. Bauer points to a recent study in the New England Journal of Medicine where researchers at Beth Israel Deaconess Medical Center and the Massachusetts Institute of Technology used a combination of GWAS and CRISPR to home in on obesity-related gene variants and in mice turn calorie-storing white fat into calorie-burning brown fat.
And in a Dana-Farber/Boston Children’s lab within sight of Bauer’s, hematologist/oncologist Vijay Sankaran, MD, PhD, recently used GWAS data to identify genetic targets for pushing blood stem cells to produce more RBCs than normal.
Sankaran’s work centered on a gene called SH2B3; people with rare but naturally occurring variations in the gene have unusually high RBC counts. Using both CRISPR and RNA interference (RNAi, a technique for dialing down gene translation very precisely) Sankaran and his collaborators found they could mimic the effects of those variations in the laboratory, shutting off SH2B3 in blood stem cells and ramping up RBC production.
As they noted in Cell Stem Cell, their data represent a significant step towards manufacturing RBCs for transfusion or as drug delivery vehicles, potentially at a cost five-fold lower than current in vitro RBC production methods.
It’s all about the tools
Sankaran agrees that the rise of new experimental tools could really make the effort spent on GWAS analyses worthwhile.
“The first GWAS study came out in 2005, but we didn’t have the tools to really manipulate cells genetically until fairly recently,” he says. “Over the last few years, though, RNAi and genome editing tools like CRISPR have become available. “Now we have the ability to take advantage of what we learn from natural human variation to manipulate cells based on genetics and think about the next generation of therapies.”
Sankaran believes that the real promise of GWAS, genetics and gene editing will come about through the evolution of regenerative medicine and cell therapy. But where GWAS and other forms of genomic data have really shone so far, he notes, has been in helping researchers understand what biology really matters in disease processes.
“In some ways, genome editing and gene therapy are the low-hanging fruit,” he says. “We’re not going to be able to scale genome editing to hundreds of thousands of patients. But more and more, we’re learning that the genes that are mutated in disease or that explain some variation like in fetal hemoglobin could actually shine a light on other pathways to target, and other ways that we can target them.
“It’s by better understanding the biology,” he continues, “that we’ll find the druggable targets that we can address at scale.”