NEW YORK (GenomeWeb) – In an analysis published in Nature Medicine today, the Broad Institute's David Scott and Feng Zhang said that prescreening patients with whole-genome sequencing and then integrating that information with empirical methods for guiding RNA selection could help researchers design CRISPR-based therapeutics that are both efficacious and safe.
"In contrast to typical drug development strategies aimed at targets that are highly conserved among individual patients, treatment at the genomic level must contend with substantial inter-individual natural genetic variation," Scott and Zhang wrote. "This genetic variation confounds the target sites of certain Cas endonucleases more than others…. If this variation disrupts the therapy target site, it can affect the efficacy of a CRISPR-based therapeutic; if it generates off-target candidate sites, it can affect the safety of a CRISPR-based therapeutic."
The researchers analyzed the recently released Exome Aggregation Consortium (ExAC) dataset with variants from 60,706 individuals and 1000 Genomes Project dataset to determine the impact of population genetic variation on therapeutic genome editing with Streptococcus pyogenes Cas9 (SpCas9), the SpCas9 variants VQR and VRER, Staphylococcus aureus Cas9 (SaCas9), and Acidaminococcus sp. Cpf1 (AsCpf1).
They found that extensive variation was likely to affect the efficacy of each enzyme, and could result in "unique, patient-specific off-target candidates." In fact, about 93 percent to 95 percent of possible genome editing targets contained variants in the ExAC data set that are likely to alter the efficiency of target cleavage, they added, "supporting the notion that using multiple enzymes with distinct [protospacer adjacent motif] requirements should enhance both safety and efficacy by increasing the number of available targets for therapeutically relevant genomic loci."
Further, they noted, the existence of multiple haplotypes in any population compounds the number of off-target candidates at a given locus. So, they used the current 1000 Genomes dataset — which provides comprehensive coverage of alleles occurring at a frequency of up to 0.1 percent in the population — to identify targets with minimal off-target candidates in high-frequency haplotypes in the human population. "Use of the enhanced-specificity enzymes eSpCas9 and Cas9-HF1 will further reduce the likelihood of cleavage at off-target candidate sites, but it will still remain important to avoid target regions that are repetitive or have off-target candidates in high-frequency haplotypes," Scott and Zhang wrote.
They also pointed to the demographic information, including sex and ancestry, for each individual in the 1000 Genomes Project dataset — information that can also be used to explore how population demographics can affect off-target candidate variation for a given individual. They found that "pre-therapeutic whole-genome sequencing of individual patients will be needed to select a single approved guide RNA-enzyme combination for treatment that is a perfect match to the patient's genome and free of patient-specific off-target candidates. The selection of specific targets to pursue for therapeutic development will also depend on the type of gene edit desired."
The authors concluded that this type of analysis will be especially important as researchers and companies begin to design clinical trials for CRISPR-based therapeutics. If they do not take genetic variation into consideration, Scott and Zhang said, they risk confounding the trials.