David Coffey, MD, is a physician scientist with the Fred Hutchinson Cancer Research Center in Seattle, Washington. When he’s not treating patients, he’s in the lab researching blood cancers such as multiple myeloma and lymphoma. His research centers on the detection of circulating DNA in peripheral blood samples to identify mutations involved in these cancers. We recently spoke with Dr. Coffey about his research, the challenges he faces, and his vision for the future.
NanoString (NS): What’s the biggest challenge you face in your research?
David Coffey (DC): One of the major challenges we face in studying multiple myeloma, a cancer of bone marrow plasma cells, is that DNA mutations that drive the disease are both a spatially and temporally heterogeneous. Despite that, our current approach for characterizing the genomic alterations relies on a single bone marrow biopsy that may not be entirely representative of the disease. To overcome this, we are attempting to identify circulating tumor DNA in patient blood samples that may originate from multiple sites of disease. This DNA is released into the circulation from dying cancer cells. Our hope is that the DNA mutations we detect in the blood may be more representative of the entire disease in a single patient rather than a single biopsy.
NS: How were you identifying mutations from patient samples for your research?
DC: It really can be like searching for a needle in a haystack. We rely on next generation sequencing methods to identify mutations in target genes. One problem is that the type of data generated can be overwhelming. The file sizes are extremely large and we often needed a dedicated biostatistician to analyze the data. There can also be inconsistencies in the software available. It works, but it’s far from ideal.
NS: What are you learning from these sequencing projects?
DC: We’re looking for mutations specific to genes that will drive cancer development or may impact clinical decisions. The challenge with multiple myeloma is that there are only twenty or so genes that mutate consistently in the disease and only a few of these mutations are typically seen in one individual. What we’ve learned is that there’s not one single abnormality that presents an Achilles Heel to target in the disease. What we’re hoping is that we can start to use NanoString’s 3D Biology™ Technology on blood samples to identify DNA and SNV that are relevant to the disease.
NS: What led you to begin working with NanoString?
DC: When trying to analyze small numbers of cells, we found that the signal from the background was often greater than the signal from the target. However, we are limited in the amount of sample from our patient. We may only have 1 mL of patient bone marrow aspirate to work with and we need to maximize the amount of information we collect from it. We needed a different technology that would help us hone in on our target cells.
NS: Your sample volumes are so small and rare; trying a new technology might seem risky to some. Why did you move forward with NanoString?
DC: It’s true that every sample is precious and in limited supply. Our existing technology forced us to choose what information we needed to have and what information we were willing to sacrifice simply because we might only have enough sample to run one assay. So we can collect genomic data but not proteomic information, or vice versa. NanoString’s 3D Biology™ Technology, and specifically the Vantage 3D ™ DNA:RNA:Protein Heme Assay*, presented a new option where we didn’t have to compromise. What we found was that we could run multiple assays in parallel on the same sample. This lets us collect data faster and more effectively cross-correlate our results. The data files are smaller and much easier to interpret so we can all work with the data.
NS: That’s a lot of data—how are you going to use what you’ve collected?
DC: Yes, we have an extraordinary amount of data and big clinical data sets. My vision is to turn these into a single, queryable database that we can correlate molecular data with important clinical outcomes. Using machine learning, we can potentially unlock a lot of insight from these data. This hinges on having a reliable source of data that is not limited to just partial genomics or just proteomics.
NS: Beyond the potential of diagnostic or predictive screening, what can these unified data sets do to enhance clinical research?
DC: In some cases, we can already sequence a patient and assess their risk level for developing a given type of cancer. But let’s go beyond that and ask, “will a certain type of treatment work well for me given my genetic profile?” That will influence the treatments we select and cut down on some of the existing clinical trial and error. Furthermore, we would like to be able to test a patient’s blood to monitor the efficacy of their treatment by looking at the RNA and proteins. For example, we can compare the RNA and protein levels of cancer indicators prior to beginning treatment and see how they change as a result of the treatment. Potentially we can get real time, individualized feedback as to how long we need to keep this patient on their chemotherapy. It may then be possible to decrease the amount of chemotherapy required and get the patient healthier sooner.
NS: What is the most important thing to you in your research?
DC: My goal is to change patients’ outcomes for the better. It’s great to see intriguing research published but I am disappointed when that research is not translated to the patients it was intended for. It’s not about building a publication record; it’s about changing peoples’ lives. That’s what we’re here for.
NanoString is proud to be a pivotal part of David’s research. Learn more about 3D Biology™ Technology and this study at ASH 2017 “Simultaneous detection of single nucleotide variants, mRNA transcripts, and protein expression in multiple myeloma bone marrow aspirates.”
*The Vantage 3D™ DNA SNV Heme and Vantage 3D™ Protein Heme panels are currently in development. Please contact us for more information.
FOR RESEARCH USE ONLY. Not for use in diagnostic procedures.