Towards rational graft-versus-host disease prophylaxis

Calcineurin inhibition for graft-versus-host disease (GvHD) prophylaxis was introduced into clinical practice of bone marrow transplantation about 35 years ago. Through the early work of Storb, Deeg and Ringden it became clear that calcineurin inhibition with cyclosporine A as a single agent was more effective than methotrexate alone and that the two drugs acted synergistically (see Table 1 for overview). Since then, multiple strategies have been tested in randomized trials: modified doses of cyclosporine A and another calcineurin inhibitor (tacrolimus), T-cell depletion (antithymocyte globulin, CD8 antibodies), cytokine blockade (CD1 antibodies, CD25 antibodies), biologics (interleukin-11, palifermin) and reducing inflammatory triggers (gastrointestinal decontamination) to mention only a few. The results of these trials cannot be compared directly because of differences in their designs. However, these trials have a common feature: they have failed to demonstrate a clear and clinically significant improvement in overall survival. The reasons for this were increased incidence of relapse, treatment-related mortality or both.

Table 1.Efficacy of drugs used alone and in combination for the prophylaxis of acute graft-versus-host disease.

Severe GvHD is associated with high morbidity and mortality, consequently any significant reduction in the severity and incidence of GvHD should translate into measurable benefits in survival and quality of life. Obviously many of the studies done were not sufficiently powered to detect small changes in overall survival. An alternative explanation for studies showing decreased GvHD severity and incidence without survival benefit is expectation bias, as most studies were not appropriately blinded. There is no doubt that a drug combination diminishing the incidence of acute GvHD by more than half, as occurred in the study by Pidala et al.1 published in this issue of the journal, would also be expected to result in at least a decrease in transplant-related mortality and should also produce detectable changes in quality of life outcomes.

The classical dilemma of hematopoietic stem cell transplantation, that GvHD can only be completely prevented at the expense of increased relapse, has been challenged by experimental work on regulatory T cells (T-reg). T-reg can be given by adoptive transfer of ex vivo-expanded cells or can be expanded in vivo if favorable conditions for the cells can be generated in the peri-transplant period. Because in vitro expansion of T-reg for adoptive transfer is extremely resource-intensive, in vivo expansion strategies are attractive. T-reg depend on interleukin-2 for their survival. Calcineurin inhibitors, by nature of their mechanism of action, diminish T-reg expansion and fuction in most settings. In addition, other pro-inflammatory cytokines that are abundant after conditioning prevent T-reg from expanding or skew them towards other T-helper lineages. In contrast to calcineurin inhibitors, sirolimus has been shown to allow for preferential T-reg expansion in vitro and in vivo in multiple studies.2-4

The rationale of the study presented by Pidala et al.1 in this issue of Haematologica is to minimize the use of calcineurin inhibition while still preventing GvHD by preferential proliferation and proper functioning of T-reg. For this purpose the authors conducted a single center, randomized trial comparing their current standard of care, tacrolimus with plasma levels at 5-15 ng/L with methotrexate to sirolimus and tacrolimus at plasma levels of 3-7 ng/L. Tacrolimus was given at a lower dose in the experimental arm of the study in order to minimize the negative effects of calcineurin blockade on T-reg. The observed significant reduction of the incidence of acute GvHD (43% versus 89%) correlated with slightly increased Treg/total CD4 ratios on days 30 and 90 after transplantation. The incidence of chronic GvHD was 24% (95% CI 7-47%) for patients treated with sirolimus/tacrolimus and 64% (95% CI 41-79%) for those treated with methotrexate/ tacrolimus, but survival and quality of life did not differ between the patients in the two treatment groups. In accordance with concepts based on experimental models, Pidala et al. present exploratory data showing a relative (but not absolute) expansion of T-reg that might be sufficient to explain the reduced incidence of GvHD. Whether this is really the mechanism that is instrumental in patients undergoing hematopoietic stem cell transplantation remains to be clarified. Other limitations of the study are the lack of relevant clinical outcome data other than GvHD (information on overall survival, transplant-related mortality and relapse incidence) as well as a remarkably high incidence of acute GvHD in the standard-treatment arm.

In summary, the results of this single center study are in concordance with the current concept of the biology of acute GvHD and the role of T-regs and also show how current treatment standards can be modified to serve new concepts. Whether this translates into meaningful changes in clinical practice will be answered when the results of the CTN 0402 trial comparing tacrolimus (5-10 ng/L)/methotrexate with sirolimus/tacrolimus (5-10 ng/L) are communicated in a few years time. The authors are to be commended for their continued effort to study this fascinating concept. In Europe in 2010 over 12,000 patients received an allogeneic HSCT. It is a pity that only a negligible fraction of them have been entered into randomized clinical trials.5

Footnotes

• Jakob R. Passweg is Head of the Hematology Division of Basel University Hospital (Basel, Switzerland) and President of the Swiss Cancer League. His main interests are stem cell transplantation and bone marrow failure. Christoph M. Bucher is a transplant hematologist in the Hematology Division of Basel University Hospital (Basel, Switzerland). His main interest is the prevention and treatment of acute graft-versus-host disease.
• Financial and other disclosures provided by the author using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are available with the full text of this paper at www.haematologica.org.