Early Biomarkers and Coverage Decisions in Immuno-Oncology

Michael Kolodziej, MD: The benefits of a biomarker in coverage policy are that the biomarker, especially if it’s a companion diagnostic test, will enrich the population for responders or identify a population of nonresponders who should not be treated with the therapy. Ultimately, it allows us to fulfill our goal of getting the right treatment to the right patient at the right time. So, it’s fair to say that payers love biomarkers. I love biomarkers! But, the challenge in immunotherapy has been that we really haven’t gotten a good biomarker. We could talk more about exactly what exists right now, but right now, we are a little bit hamstrung in our ability to use any specific marker to identify in a definitive fashion. And that’s what the payer wants to know; who’s going to benefit and who’s not. When we have such a marker, it’s virtually certain that we will require it before utilizing the therapy. And I think most oncologists understand the rationale behind the perspective of the payer, that biomarkers, when they really do help identify responders and nonresponders, are really in the best interests of everybody in the care spectrum.

Both response and nonresponse are really important outcomes identified by biomarkers. And the ability to use a biomarker to identify a population that’s not likely to respond is critical, and you can think of why a payer would think of that under three major categories. So, number one, of course, is the financial cost of a therapy that’s not likely to work: futile therapy. Nobody wants futile therapy. Number two is the cost of toxicity, and certainly no one wants to give a patient a toxic agent that’s not likely to benefit. But, the third area, which almost invariably gets forgotten, is the opportunity cost. If you go on a therapy, and it prevents you from going on a clinical trial or an alternative therapy that might have a greater likelihood of working, then there’s opportunity cost for the patient. And I think for all three reasons the payers really do like biomarkers that put you in a direction when you are confident of what the outcome will be.

Until just a short while ago, we really didn’t even think in terms of the distinction between a complimentary and companion diagnostic. We lived in a world that was largely dictated by companion diagnostics. The attractiveness of companion diagnostic is that the value of the biomarker is linked to the FDA approval of the drug. They are co-developed, and so there’s a high degree of confidence that the results of a specific test will help you make some decisions about therapy. The whole complimentary idea has created a tremendous amount of confusion, because it’s kind of like saying, “We sort of think this is going to help you make a decision.” And from a policy perspective, that’s not very helpful for a health plan. Health plans live in largely binary worlds. And so, they want to be able to say, “You know, if you have the marker it’s a really good thing for you to be treated this way,” not, “It’s kind of a good idea and you should at least consider it.” The complimentary thing has just confused the universe, as far as I’m concerned.

For, many things we do in oncology, we don’t have biomarkers. And the traditional approach is to have coverage policy reflect FDA approval or NCCN compendia guideline approval. I think that in a universe where there’s either not a biomarker, or we have candidate biomarkers, that has been and will continue to be, I believe, how payers approach a coverage policy for cancer therapies. Part of that is law, part of that is the regulations that are mandated by states. So, then I guess part of it is it’s just been that way forever, right? It’s CMS that has taken that tack, and commercial payers have followed it. I think we are now in immunotherapy, for example, in a space where we are largely governed by FDA approval and NCCN guidance. But, I am forever hopeful that we will get a biomarker.