Whether or not you are aware, the field of laboratory medicine has already entered into the era of pharmacogenomics in personalized medicine. In yesterday’s AACC University morning short course, Advanced Applications of Molecular Diagnostics and Pharmacogenomics in Targeted Therapies, Gregory Tsongalis, PhD, and J. Greg Howe, PhD highlighted the ways in which next-generation sequencing technologies are already guiding targeted cancer therapy.
Using a case-based approach to guide the discussion, Tsongalis demonstrated how next-generation sequencing is being used at Dartmouth-Hitchcock Medical Center to analyze biopsied tumor tissue for mutations in a panel of 50 genes. This strategic use of next-generation sequencing allows the clinical team to target the tumor with the combination of chemotherapeutic agents most likely to be effective. Next-generation sequencing also allows the clinical team to look at mutations that, while not currently actionable, may qualify a patient for one of many clinical trials for new cancer therapeutics.
Howe similarly presented data from his practice at Yale on the use of next-generation gene panels for the analysis of prognostic markers of Acute Myeloid Leukemia (AML). The diagnosis and prognosis of AML increasingly includes analysis of several genes. However, unlike the genetic coverage of the available solid tumor panels, the available next-generation panels for AML did not provide the necessary coverage for actionable genes. Because of this shortfall in coverage, Howe shared how Yale developed a next-generation sequencing panel to provide adequate coverage of the prognostic markers in AML.
This reliance on the patient’s genotype has been costly for laboratories using a traditional single-sequencing model, but the switch to next-generation sequencing has changed that. When asked to quantify this savings, Tsongalis shared as an example: the cost of analyzing only the two most common mutations in EGFR and seven most common mutations in KRAS was approximately three times the cost of the 50-gene panel used at Dartmouth-Hitchcock Medical Center. Howe agreed that the magnitude of cost reduction was similar at Yale.
As next-generation sequencing moves into more diagnostic algorithms, there will be hurdles to overcome, Tsongalis noted. One of these hurdles is the reliance on the availability of tumor tissue suitable for analysis. For example, following minimally invasive procedures, occasionally there is insufficient tissue after other diagnostic studies have been performed to conduct next-generation testing.
One possible solution to this problem is the use of a liquid biopsy, Tsongalis said. A liquid biopsy, otherwise known as a peripheral blood sample, could be used in the near future to detect circulating cell-free tumor DNA. Many tumors have increased cell turnover or necrosis, and the cell-free DNA released during this process may be of sufficient concentration in peripheral circulation to allow for the early detection and monitoring of cancer with a procedure no more invasive than routine phlebotomy.
These new pushes into pharmacogenomics are in stark contrast to the much promoted—and now maligned—pharmacogenomic dosing of warfarin. Warfarin has a wide range of effective doses that are due at least in part to the genotype of CYP2C9 and VKORC1. While seemingly a logical place to start pharmacogenomic testing, this strategy did not fare well for several reasons, including firmly established clinical dosing schemes and problems with genotyping turnaround time.
While the large gene panels are revolutionizing cancer diagnosis, prognosis, and therapy, it is likely that targeted somatic genotyping, such as that used for assessing risk of severe neutropenia associated with Irinotecan, will continue. However, as next-generation technology is more widely implemented and its cost falls, it is possible that these single-gene tests will also become obsolete in favor of whole genome sequencing.