Video
Steven Peskin, MD, MBA, FACP: Diagnostic tools in cancer—that’s a broad range. We, most recently, have thought about and have focused on molecular diagnostics, companion diagnostics, and the explosion in genetic testing and proteomic testing. We see some very real value.
On the other hand, there are a range of tests that have a questionable impact on the course of treatment. In fact, some of these tests can, metaphorically speaking, take one’s eye off the ball. We see that the explosion in the number of genetic molecular tests can be a cause for some overuse and inappropriate use, and again, can take us down pathways that are not in the patient’s best interest. I’ve had oncologists who know far more about this area than me express that to me as well.
There are other diagnostics. There’s the use of advanced imaging. We know that some advanced imaging is very vital. On the other hand, in certain stable breast cancer patients, for example, there’s no evidence to support repeat PET/CT (positron emission tomography and computed tomography) scans. That’s another area where we’re cognizant of working toward appropriate use.
How might we support clinicians around clinical pathways? We do not have proscribed clinical pathways that we have endorsed, or [that we are] working with a third party organization on, as some health plans that have developed pathways in cancer care do. We do provide our clinical partners with our medical policy. That reflects a lot of time, effort, energy, and thoughtfulness by clinicians with input from community thought leaders. We do provide analytics to our clinical partners. Increasingly, we’re taking what we have in the area of medical economics, analytics, and health information, to work with or harmonize that with our clinical organizations. But, we do not have explicit or proscribed clinical pathways. We do provide feedback to our clinical partners. In certain areas of cancer (chronic lymphocytic leukemia [CLL] is not one of them), we do have episode of care models that we have endorsed.