The challenges faced by oncologists in using biomarker-based tests were underlined in a recent article by Dr. Marcus Neubauer in OncLive.
In recent posts in this blog I have focused on the challenges of interpreting genomic data, but Dr. Neubauer emphasizes two additional and highly relevant points.
The first is that just selecting the right tests to order for each patient can be a challenge. "A proliferation of diagnostic companies offer molecular tests, but only limited guidance on which tests are clinically appropriate and, also important, which tests insurance companies will cover," states Dr. Neubauer. As reported in this blog, validation data for new tests can be limited and difficult to access, making the physician's job ever harder.
Dr. Neubauer also points out that community oncologists have a particularly challenging job, as they must "practice as generalists, treating patients with all types of cancer." The National Cancer Institute estimates that 85% of cancer patients are treated at community hospitals. If researchers at Stanford Medical School are struggling to interpret genomic data in a timely and effective manner, I can only imagine the scope of the challenge for an over-burdened community oncologist.
The ever-present interpretation challenge may have become even more pronounced as well, based on the results of a study just published in Science Translational Medicine. This study compared mutations discovered in the tumor cells of patients with samples of their normal tissue, and found that an average of 1/3 of the variants in the tumor samples were false positives, as they also occurred in the normal tissue.
Since most genetic pathology tests only measure tumor samples, this basic understanding is being missed in most cases. It may be time to re-evaluate basic testing practices.