One of the core components of our MatchSourceSM technology platform is a powerful, predictive search algorithm we call HapLogicSM. We sat down with Katie Howe, Ph.D., senior manager, Immunogenetic Operations and Research, and Caleb Kennedy, Ph.D., senior manager, Bioinformatics Research, to learn more about HapLogic and the important role search algorithms play in allogeneic drug development.
Q: What’s the purpose of an algorithm?
Katie: Algorithm sounds like a fancy word, but what we’re really talking about is a predictive search engine—like Google or Amazon. We have 19 million potential donors in the Be The Match Registry®. Our algorithm—HapLogic—was designed to rank order those donors according to population-specific frequency data. It’s a predictive way of saying to physicians, “here is the most likely donor to match your patient, and here’s how likely it is that the match will be successful.” From an initial focus on matching donors for transplant, we are now translating our expertise to support cell therapy clinical development and manufacturing by third parties.
Caleb: It’s about helping our customers arrive at the right treatment for their patient. Taking something that’s very complex and making it easier to digest so that the user can make good decisions.
Q: Since you launched HapLogic in 2006, how has the technology advanced and improved?
Caleb: It’s advanced in two areas: the speed and accuracy of the algorithm itself, and the amount of population-specific frequency data we’ve collected. When it comes to predictive algorithms, your predictions are only as good as your data. Today, we have the largest and most diverse donor registry in the world.
Katie: Beyond size, one of the things that sets our registry apart is its diversity. We have a huge range of characteristics and backgrounds—and this makes us the ideal partner for companies in the biotherapies space.
Caleb: For example, some companies are looking for rare or specific HLA alleles with particular clinical benefits—and those alleles are represented and searchable in our registry.
Q: How has the role of HapLogic changed?
Katie: We designed HapLogic to quickly go through millions of donors and come up with the best options—and that same concept and application applies, whether that’s an inventory of our donors, or an inventory of cellular products.
Caleb: We are able to adapt MatchSource to use HapLogic to search a variety of inventories—from our registry of unrelated donors to manufactured cell products that will be used for immediate infusion.
Q: What’s on the horizon?
Katie: Right now, HapLogic only applies to donors who are listed through the Be The Match Registry—but there are millions of other potential donors out there in smaller registries in other countries. In almost 50% of our cases, either the patient or the donor is international. Unifying the match process across donors worldwide is an important focus for us. We’re working to create a new prediction algorithm that would unify the match process and provide match predictions for donors worldwide.