Before he stepped in front of an audience of 1,600 people representing the top global leaders in health care, biotech and venture capital, William Lane, MD, PhD, collected his thoughts before he shared his big idea to transform blood typing. Lane knew that he needed to convince his audience at the 2018 World Medical Innovation Forum (WMIF) that the field of blood typing needed to be revolutionized; many people assume that medical professionals already know everything they need to know about blood type to keep patients safe. But there are more than 300 antigens on red blood cells, each of which can contribute to a person’s unique blood type and to which an individual can become sensitized. Determining multiple antigens is costly and arduous and rarely done today. But Lane has an idea for how to identify these antigens as part of standard practice before a patient receives a blood transfusion. Since he first began working on this new tool known as bloodTyper, Lane has been convincing scientific and clinical experts about the importance of this resource that has the potential to change the field and save patients’ lives. He hoped that taking the stage at WMIF would help him reach an even wider audience and persuade his colleagues to help advance this tool to clinical implementation.
Why Blood Type Matters
Before a transfusion, it is standard practice to determine a person’s ABO blood group by testing for the presence or absence of A and B antigens on the surface of red blood cells to avoid a severe acute hemolytic reaction, renal failure and shock can occur, and death is possible. In addition, it’s common to test for D antigen status, known as Rh, which is crucial in prenatal medicine, as transfusing Rh+ blood into a pregnant patient who is Rh- is the leading cause of hemolytic disease among newborns. This means that patients are tested for a total of four antigens on red blood cells. However, there are more than 300 antigens on red blood cells, for which none are currently typed before routine blood transfusions.
Although between five and 16 deaths reported to the Federal Drug Administration are attributed to mismatches in donor and recipient blood types annually, Lane and other blood type experts believe complications are under reported or not appreciated when they occur after transfusion. Lane estimates that the risk of complications after blood transfusion is 3 to 30 percent – or about 3.4 million cases worldwide.
“While complications can be life threatening, with current technology, it isn’t currently cost effective to do blood typing on a range of antigens for all patients who receive transfusions,” says Lane.
Next-Generation Sequencing: What’s Next for Blood Typing
Recently developed techniques – such as genotyping, or sampling single “letters” of DNA – make it possible for medical professionals to routinely type more antigens, but still only classify approximately 30 antigens. Lane saw a clear opportunity for improvement here. “There was a consensus among pathologists that new methods for blood typing were needed, but nobody knew how to begin to classify more than 300 antigens,” says Lane. “I had an idea that I wanted to use next generation sequencing with automated analysis software to do next-gen blood grouping. It was a pretty radical approach.”
Even among pathologists – specialists who are concerned with the diagnosis of disease – Lane’s radical approach was met with some resistance. “It was hard at first to convince people that this idea was worth sinking a lot of time and money into, and that they should trust me to develop it, as I was just a resident at the time.”
Lane found an opportunity and support to create his algorithm when he connected with the leaders of the MedSeq Project, including principal investigator Robert Green, MD, MPH, from the Division of Genetics, and co-investigator Heidi Rehm, PhD, from the Division of Pathology. The MedSeq Project is the first randomized trial of whole genome sequencing in healthy adults. By leveraging an extra blood sample from each of the 110 participants, Lane obtained the data he needed to develop bloodTyper – a computer software that rapidly and accurately returns results on an individual’s blood antigens. bloodTyper will allow for comprehensive and cost-effective determination of subtle differences in blood types with more than 99 percent accuracy. Lane believes that bloodTyper is a fundamental tool for the future of pathology and transfusion medicine.
Curating the database needed to build bloodTyper and developing the algorithm to classify antigens was the exact sort of challenge Lane had always hoped to work on. “I have always felt passionate about impacting patient care, but at the same time, I enjoyed wading into the technical aspects of how things work. I loved that I was able to combine my enjoyment of technology and medicine when developing bloodTyper.”
Ready for the Clinic
bloodTyper has been validated using the 110 genomes from the MedSeq Project, and Lane is now looking to implement his software on a larger scale. In an international collaboration with the INTERVAL study set up by the Universities of Cambridge and Oxford in the UK, Lane is leveraging blood samples from 50,000 participants to test bloodTyper’s performance on large-scale data sets. At the WMIF, professionals with cloud computing platforms were interested in collaborating with Lane to make bloodTyper available for clients on their cloud platforms.
This may only be the beginning of Lane’s global collaborations.
“With the National Healthcare System in the UK, the framework is already in place for us to implement bloodTyper to screen donors today,” said Lane. “We are also looking forward to typing the one million whole genomes from the recently begun “NIH All of Us” research program. Here at BWH, we are thinking about beginning to implement bloodTyper to screen complicated reference cases. bloodTyper will be ready for clinical use when the data is ready for us.”