As a clinical trialist, I look at it as my duty to make sure clinical trials are available at my cancer center for the particular diseases I treat—myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML).
I mean, let's face it—we haven't exactly changed the world in these diseases. In MDS, the median age at diagnosis is 71 years, and the only curative therapy is bone marrow transplantation. Those two facts are mutually exclusive 95% of the time. In AML, for the average patient walking into my clinic (or, more commonly, being transferred onto our leukemia floor), the chance of being cured is 30%, meaning that 70% of the time, our best treatment efforts are merely palliative.
Over the past three decades, the backbone of remission induction therapy has changed little, as has the chance of being cured. Given these fairly dismal odds, we even avoid the term “cure,” preferring “long-term, disease-free survival,” meaning alive and without MDS or AML at five years of follow-up.
Most would agree, to use the euphemistic phraseology, we have a lot of “opportunities” in these areas. In other words, we can do a heck of a lot better. And the only way to do better is to study new drugs or new drug combinations, or to choose more ideal patients. And I don't plan on turning anyone with AML or MDS away anytime soon.
So I make it my responsibility to have available a menu of trials for various indications, and look at it as a personal embarrassment when I don't have a study open for someone with particularly refractory disease. One of the most crucial aspects to conducting trials in human subjects is the informed consent process. The intent and ethics of informed consent is straightforward, and can be defined: We must ensure that we explain to a patient, in plain, understandable terms, the potential risks and benefits of a specific therapy, along with alternative therapies that can be chosen, and the people who will be involved in that person's care.
It is our duty to then assess that patient's understanding of what we communicated, give an opportunity to answer questions, and reinforce that consent to receive a therapy can be withdrawn at any time, without penalty or prejudice.
But then the rubber hits the road.
“The potential benefit of giving you chemotherapy is that we can get your leukemia under control, at least for a while, and maybe even cure you. But there are risks…”
“I don't want to hear about the risks.”
But I have to tell you about them. Don't I? Sometimes people don't want to know. They may not want to hear anything negative, for fear it will impair their ability to will the cancer away. And many times, we tell them, or think we tell them about complete remission rates, chances of cure, and risks of dying, and they don't hear anything beyond, “You have leukemia and here's how we can treat it.”
We conducted a study of older AML patients about to undergo intensive induction chemotherapy (both on and off clinical trials) in which we deliberately provided numerical estimates of chances of cure (5-10%) and chances of dying (15-25%) from therapy, in addition to reviewing a panoply of side effects. We then returned to their hospital room minutes afterwards to administer a questionnaire asking them to repeat back to us their estimates of cure and treatment-related mortality.
The results? Approximately three-quarters of patients reported their chances of being cured to be greater than 50%, while half of patients estimated their chance of dying from therapy to be less than 10%.
So, did this really represent informed consent? Sure, we deliberately provided concrete risks of receiving an extremely toxic therapy. Most patients clearly did not integrate that information in a way that allowed them to convey it back to us. Should we have used this as an opportunity to tell those who were overly optimistic, “Oh no, you're way off, your chances are actually much worse than you think”? Or leave well enough alone?
Bioethicists might argue that we should give patients time to receive information about their cancers and therapies. If necessary, we might have to arrange multiple visits, even over a number of days, to deliver this information in digestible pieces, until we (and more importantly, our patients!) feel comfortable about how well this information is understood.
I would counter that this recommendation is rarely possible outside of the showroom of medical theory. Most cancers, and AML in particular, are considered forms of medical emergencies, both by those having just received a diagnosis, and by those of us who treat them. Neither patients nor their families are particularly welcoming to the notion of delaying therapy for days, and in truth most oncologists do not arrange multiple visits to discuss diagnoses and treatment—we rely heavily on family members of patients, and on nurses, who are much more cognizant of our patients' levels of understanding than we are. Yet, there's an even greater argument against this iterative process of informed consent.
What's wrong with a little denial?
On some level, whether they express it on a questionnaire or not, our patients know that their diagnosis is serious, and that they can die from it or from therapy used to treat it. And they've known it their entire lives. I still remember when I was a kid and heard that John Wayne died while in treatment for an abdominal cancer. It's cancer. Nobody wants it. Even children know it's bad. So if one of my patients decides subconsciously to suppress all the stuff about side effects and death from therapy as she puts one foot in front of the other and reframes her life, who am I to say that's wrong. Or somehow, not informed?
I propose we change the definition of informed consent to incorporate risks, benefits, alternatives, and personnel involved in therapy, with a patient expressing understanding, to the extent that he or she wants to.
Isn't that, ultimately, the most patient-focused approach?