By Diane Allingham-Hawkins, PhD, FCCMG, FACMG, Senior Director, Genetic Test Evaluation Program
Hayes has commented several times on this page regarding the potential negative implications of direct-to-consumer (DTC) genetic testing. You can read our most recent blog on this topic here.
A recent commentary written by researchers at Boston Children’s Hospital and Harvard Medical School and published online in the journal Clinical Pharmacology & Therapeutics highlighted some real-life examples of how DTC results can be misinterpreted and potentially harmful. You can access the full article here.
The authors described the case of Dr. J., a half-Ashkenazi Jewish, half-Irish man with a family history of Crohn’s disease. The DTC test performed on Dr. J.’s sample included a pharmacogenetic test (i.e. a test designed to predict response to, or an adverse reaction to, a drug based on a person’s genes) to predict response to thiopurine drugs, agents used to treat a variety of conditions including hematologic cancers, inflammatory bowel disease (including Crohn’s disease), and rheumatoid arthritis, as well as being used as immunosuppressants in organ transplantation. Since thiopurine drugs are metabolized in part by thiopurine methyltransferase (TPMT), common genetic variants in the TPMT gene, have been investigated for their impact on drug metabolism and adverse events. At least 20 different variants (often called polymorphisms) have been described that decrease TPMT enzyme activity and can increase the likelihood of myelosuppression during treatment. Dr. J.’s reported results suggested that he had 2 different variant copies of the TPMT gene that would significantly reduce his TPMT activity and thereby put him at higher risk of an adverse reaction to treatment with thiopurine drugs. The problem was that the DTC testing lab did not adequately investigate Dr. J.’s TPMT genes to verify their interpretation of his results. Family studies and retesting revealed that Dr. J. does not carry 2 different variant TPMT copies but rather 1 variant copy and 1 normal copy. The difference is that his TPMT enzyme activity would be only moderately reduced, rather than the significant reduction predicted by the DTC test. At the very least, this error would lead to use of an ineffective starting dose of thiopurine drugs, should Dr. J. ever be prescribed such drugs. At worst, treatment of a serious condition such as a hematologic cancer might be significantly delayed due to this information.
In the same article, the authors point out that Dr. J.’s wife, Mrs. J., received a falsely reassuring result for her risk of familial breast cancer, a condition that exists in her family, from the same DTC testing company. The test classified Mrs. J. as having “typical risk” of breast cancer, with a 4-star confidence rating. Further assessment of the results revealed that Mrs. J. had only been tested for 3 variants in BRCA1 and BRCA2 that are common in Ashkenazi Jewish individuals but found rarely in other populations. Since Mrs. J. is not Ashkenazi Jewish, this test did nothing to address her personal risk of familial breast cancer.
Fortunately, Dr. and Mrs. J. had knowledge and access to professionals who could clarify the results they received. Sadly, most individuals receiving DTC genetic test results have neither, and, consequently, there is a very real risk that these results could be misinterpreted and could lead to poor healthcare decisions. For this reason, Hayes continues to encourage and support greater FDA oversight of all genetic testing, including both DTC genetic tests and clinically available tests marketed as laboratory-developed tests (LDTs).