Thank you, Dave, for the question. So I think we recognize that it is important to invest in clinical studies. But when we are investing in clinical studies, we are also really trying to understand what the question that we are answering is and whether that is going to be incremental to what physicians want. So with respect to therapy selection, of course, I think agree with you that we are finding -- and we have shown, I think, pretty conclusively that we are finding more mutations. Does that lead to better outcomes, clinical outcomes for the patients. In our discussions, at least with the physicians, that is not a question or a concern that they have. So we could run those studies, but I don't think it would change essentially the adoption curve that we are seeing in this very established area of therapy selection. On the other hand, though, as we go into more new areas like response monitoring and MRD, I believe that larger studies and more investment is certainly needed because those are the areas that physicians are maybe a little less comfortable, right? If they see essentially an EGFR mutation or a KRAS mutation that we detect, I think the physicians, I think, intrinsically understand and know that, that is going to lead to a better outcome for the patient. But it is not, I think, as clearly demonstrated in the response and MRD areas, not just by us, but broadly by all of the different diagnostics companies that especially outside of the adjuvant setting, if you detect progression early, if you are doing surveillance with MRD, is that leading to better clinical outcomes. And in those areas, I think it is important to continue to invest, but we are always going to be very thoughtful about what studies that we are running. So we will continue our investments. We will increase our investments as we go into -- especially into MRD. But I would also think that there are a lot of investments that happen in this field that might not really move the needle. So we really want to work on studies that actually make -- that answer the questions that oncologists have and that is, I think, really important and different in different areas. For therapy selection, I don't think that is OS, but in response monitoring and MRD, it may very well be.