This article is more than 5 years old. Information may no longer be current.
The ethics of treatment-related harm in stem cell transplantation
William Wood
At a recent conference, my colleagues and I had the opportunity to discuss an individual who I had considered for a stem cell transplant.
This person — aged 65 to 70 years — had lived with aplastic anemia for several years and had been on and off standard therapies and a clinical trial. He also had several other ongoing issues, including modest renal dysfunction with a creatinine between 1.5 and 2; iron overload; platelet transfusion refractoriness; and coronary and cerebrovascular disease. In addition, he appeared to be developing a secondary myelodysplastic syndrome.
An HLA match was available. My inclination was not to pursue transplant, and I admit this was borne of instinct and a conglomeration of individual risk factors that did not fit neatly into a unifying global risk model. A hematopoietic cell transplantation-specific comorbidity index (HCT-CI) would indicate moderate risk, although this index did not capture all of this individual’s issues or the relative lack of data to support transplant for this unusual set of circumstances.
Perhaps not surprisingly, then, an interesting discussion ensued. Some were in favor of transplant and others were not. The case raised an important question that we encounter frequently: How much treatment-related risk is acceptable, and what is the threshold for potential efficacy to counterbalance this risk? Should this decision be made by clinicians, patients or a combination?
Risk vs. benefit
In many ways, risk and benefit in transplant are misleading and depend on how data are presented. If median survival estimates for a transplanted population are discussed, particularly in comparison with non-transplant strategies, benefits appear modest. On the other hand, if 5-year OS curves are shown and demonstrate a population of “cured” long-term survivors, transplant becomes more enticing — especially when the nontransplant alternative is described as palliative in nature.
When this discussion comes down to an individual — and the choice is cast as a “chance” at cure (no matter how small) or “no chance” at all — the understandable potential for emotional bias on both sides, patient and physician, becomes almost overwhelming. This leads to the kinds of decisions that many of us can regretfully remember, whether we or others have made them, in which the tragic outcome that results seems all too predictable. Sometimes the unfortunate outcome is relapse rather than treatment-related mortality and we tell ourselves that we’ve fought the good fight, but this fight comes at a price, too. Transplants have significant and guaranteed toxicity, even if they are not ultimately responsible as a cause of death.
How much risk, then, is acceptable, and for how much benefit? Is a 10% treatment-related mortality risk an acceptable risk to take? How about 20%, 30% or 50%? And for how much potential benefit — a 5% chance at long-term survival? A 20% chance? Or more?
These are difficult ethical questions, as we practice in a profession in which one of our fundamental tenets is to “at first, do no harm.” In a sense, treatment-related death should never be “acceptable” at any level. Those of us who have had one of our patients succumb to treatment-related death know this instinctively.
My first experience with transplant-related mortality will be seared forever into my consciousness. I remember telling myself that that particular individual had a high-risk disease, that her consent was informed and that the outcome was unexpected — but then again, I had known I was exposing her to significant potential suffering and harm, and she died from it. I have never felt worse and less comfortable in my own skin as a physician and as a person. Although I could rationalize the case indefinitely, I had not gone through medical training to cause suffering and harm, but that is what had happened.
It seems that only unusual circumstances should overcome the threat of transplant-related harm, whether guaranteed (short-term toxicity) or possible (treatment-related disability or death). I would argue that these circumstances should meet some specified preconditions, such as 1) clear data to support the effectiveness of the indication for individuals similar in risk to the individual being considered for transplant, or 2) a clinical trial designed to generate such data, and 3) a clear understanding of these risks by the patient, with truly informed consent.
Individual risk
Although there are potential problems with all aspects of these conditions, a particularly nebulous area surrounds the issue of individual risk. The previously mentioned HCT-CI is a good start to quantifying individual risk, although the measure is cumbersome and — as an aggregate of comorbidities — does not measure patient vulnerability directly. New and improved measures are needed. Some researchers are looking at variations of a comprehensive geriatric assessment, while some of my own research investigates cardiopulmonary fitness and patient-reported outcomes. Until new and improved measures are available, though, we’re left with imperfect models and subjective judgment on a case-by-case basis.
The HCT-CI and our subjective judgments are not frequently used to include or exclude individuals from consideration of transplant. But should they? When the alternative is characterized as palliation, many of us would recoil and say that they should not.
On the other hand, although we have successfully developed reduced-intensity regimens to extend transplantation to older and less fit populations, is it possible that we have reached the limitations of this technology in its current form? Was transplantation really designed to apply to everyone?
A renewed debate
Perhaps we need a renewed debate in the field about the ethical considerations involved in determining the balance between treatment-induced harm and potential benefit. A better understanding of individual pre-transplant risk may allow us to define populations of patients, based on diseases and physiological vulnerabilities, for whom the risks of transplantation, in its current form, are indeed unacceptable — and for whom an entirely new paradigm, a “third way” between modern transplantation and non-transplant palliation, is needed.
What this new paradigm might look like is, of course, not known at this time, but we should continue to promote and support creative approaches to find it. In this regard, a recent report and editorial in the Journal of Clinical Oncology about “microtransplantation” — HLA-mismatched peripheral blood stem cell transfusions without conditioning or subsequent engraftment, putatively separating graft-versus-tumor and graft-versus-host disease, and thus reducing treatment-related toxicity — are intriguing.
It is far too early to tell whether this type of approach will ultimately be proven effective, but it is unconventional, promising and illustrates the general point.
To be sure, for the right patients and the right indications, stem cell transplantation is a life-saving therapy, and it is one of the greatest and most astonishing successes of modern medicine. We need to exercise caution and prudence to apply this great power ethically.
Reference:
Guo M. J Clin Oncol. 2012;30:4084-4090.
For more information:
William Wood, MD, is an assistant professor of medicine in the division of hematology/oncology at the University of North Carolina in Chapel Hill. He may be reached at UNC Health Care System, Division of Hematology and Oncology, 101 Manning Drive, Chapel Hill, NC 27514; email: william_wood@med.unc.edu. You also may follow him on Twitter (@WoodBD).
Disclosure: Wood reports no relevant financial disclosures.