Yesterday, a member here made a very fair comment (here) after one of my recent posts in which I raised my concern about giving treatments without evidence to support it. The issue was whether it is really appropriate to be so protective and insular in my thinking in a setting in which patients with lung cancer may have exhausted all of the proven treatment options. Would I really defend offering no further treatment to someone in whom the only proven path is to a very disappointing and to some people an unaccaptable place?
In my posts and comments I often urge restraint about using untested drugs and other approaches because they don’t have evidence to support them, but I should qualify that. It really all depends on the risk vs. potential benefit. It’s absolutely true that in lung cancer we are often left with situations in which we run out of evidence-based appealing options and the choice is between a known bad outcome and the great unknown. We don’t have proven fourth-line treatments for advanced NSCLC, but some patients are still feeling well and strongly inclined toward more treatment, even with the understanding that there are no therapies with any proven benefit. Would I just take the easy way out and reflexively wrap myself in the flag of ironclad data or enrollment on a clinical trial? When we have oncology conferences and there is a question of how we would manage a very challenging case, the answer of “enroll on a clinical trial” is the cop-out answer that keeps you from having to make a real decision. It’s often a great answer, and I am very, very supportive of pursuing treatment AND moving the field forward in a clinical trial, but in real life that’s not always an option.
So getting back to the question, every day we face situations that don’t have a proven best answer, and pretty much every day I treat patients outside of the bounds of evidence-based medicine. I do think there is an important distinction that needs to be made based on the risks and benefits of the situation, and in a situation where all good options and you’ve got nothing to lose, the risks of an unproven treatment are quite low, even if the anticipated benefits are not necessarily high or are just completely unknown. On the other hand, in the situation I was describing for the SWOG 0023 trial, or similar situations like treatment options after potentially curative surgery, patients could already be cured without more treatment, so the stakes of harmful therapy could be much greater.
The potential risk vs. benefit decision doesn’t just simplify to extreme caution for earlier stage, potentially curative disease and a free-for-all with advanced disease. There could be situations where “pushing the envelope” is appropriate in earlier stage, curable cancer. A well-informed patient who has a very careful discussion of the potential advantages and disadvantages of something like post-operative tarceva in a never-smoker or someone with an EGFR mutation might reasonably consider the potential upside to be so great that it’s worth taking as an adjuvant therapy despite the lack of data. Even the use of consolidation taxotere after concurrent chemo and radiation, as was pioneered with the SWOG 9504 trial (discussed in more detail here), really hasn’t had a trial prove the benefit, and has a risk of dying from the treatment, but we consider the attractiveness of the one option against all of the alternatives and make the best decisions we can.
Patients with advanced disease, NSCLC or SCLC, are treated to improve survival and maximize quality of life, but not generally for cure. I would be much more inclined to pursue a treatment like standard doublet chemo, potentially with avastin, before doing something with little or no evidence to back it. I would not be inclined to treat a patient with a high risk of bleeding from a squamous NSCLC with avastin, since the risk of a fatal bleed in the first or second cycle could markedly shorten the survival of that person compared to a non-avastin option. That said, I and other experts have sometimes advocated first-line tarceva-based therapy for never-smokers or a patient with the EGFR mutation, or someone with bronchioloalveolar carcinoma, not based on irrefutable evidence, but based on an interpretation of the limited evidence to support it.
So I would definitely not say there are any absolutes here. Some people are more risk-averse while others are much more “all or nothing” than other people. I am not the most conservative oncologist, but I try to use the evidence in making recommendations, and I’m not a real cowboy. My style resonates with some patients, some might want someone who will try anything more readily, and some might want someone who sticks purely to the evidence. I would never want to haughtily suggest that cancer treatment should be a cookbook approach, purely by the numbers.
One more monkey wrench is the reality of the costs of therapy. In single-payer systems like Canada, the UK, and many other parts of the world, there is a fairly rigid calculus of allocating resources, where certain treatments that are not effective or not effective enough, or just too expensive, are not covered. In the US, the system is much more erratic, with some patients unable to receive therapies proven to be beneficial (not just cancer, but any treatment), while we have patients with really no chance of recovering who remain on a ventilator for a month before they ultimately die or support is withdrawn. In between are a lot of other marginally or non-effective therapies that can be quite expensive, and payment for them may come out of the same pot that now cannot provide adequate care for the less well-covered. This same pressure leads to increased copayments for Tarceva and other medications, which some patients are forced to forego because they cannot cover the increased costs, even though these medications can provide meaningful survival benefits.
There are no easy answers to these questions. I am very relieved to not be in a position of deciding who gets what, but there is an element of a zero sum game here. It does concern me that in the US, we are denying some people basic care while providing harmful, expensive, ineffective care to others. I don’t believe that there should be an expectation that insurance companies or Medicare pay for any therapy that hasn’t been proven to be deadly, no matter the cost, just because it exists. The limited resources that exist in the health care system I work in is just one of the many factors that I think should be considered, at least at some level. On the other hand, this isn’t an issue is a patient is saying they will pay for it themselves.
This is a much more contentious topic than just describing the established standard treatments or the promising new option being tested. Feel free to question, challenge, agree with, or object to what I’m saying. We all come into this with our own perspective, and I can’t pretend to really know what it is like to be a patient or a family member directly affected by cancer. My take comes as a physician who wants to do the best for my patient, as a friend to my patients, and also as a citizen. I’d love to hear the views of people with a different perspective, whether you agree or disagree.