Federico Stella, MD, University of Milan, discusses study results demonstrating no significant difference in infection rates between restrictive and nonrestrictive neutropenic diets following hematopoietic stem cell transplantation. These phase 3 data were presented at the 64th American Society of Hematology Annual Meeting and Exposition.
Transcript
Can you give a summary of the study design?
We designed a multicenter randomized study trying to demonstrate that the use of a restrictive neutropenic diet is basically useless for preventing infections in our patients undergoing hematopoietic stem cell transplantation. We designed and conducted a randomized noninferiority trial that compared the 2 arms. Patients randomized in the standard arm received during the period of neutropenia—so low blood cell counts—a restrictive diet consisting of only foods cooked above 80 °C. That is, more or less, 175 °F. Patients randomized in the experimental arm received during the period of neutropenia no restrictive diet. So it’s a diet compliant to hospital hygiene standards that allows patients to eat fresh fruit, fresh vegetables, cold cuts, and sausages that were forbidden in the standard arm.
Did you have any pushback when designing this study, which challenges a long-held belief that the neutropenic diet is necessary?
It is a very long-lasting belief, as you have correctly highlighted, but several centers have the feeling that restrictive diet is basically not capable to prevent infections. It was a widespread idea, even because we have observational data that highlighted a paradoxical link between the use of neutropenic diet and higher incidence of infection. And it will be explained from the impact of diet on our microbiota, so the content of bacteria in our gut. It was challenging the status quo, it’s true. But it was a status quo that, thanks to other observational studies and studies in other settings rather than hematopoietic stem cell transplantation, was already challenged. And I think that the merit of these studies is to demonstrate something that everyone already thought, but [was] never tested prospectively. In 2016, when Professor Paolo Corradini—that is the mind under this study—read over this observational data [and] said we have to stop restricting the diet of our patients, it is useless. But, on the other side methodologically, scientifically, the right thing to do to stop the use of this diet safely is to conduct a randomized trial that definitely demonstrates that the use of it, because prospective data was lacking. So we wanted to, and he wanted to, fill this gap and to put a cross on the use of a nonrestrictive diet in the most methodologically correct way. So we designed this randomized study.
Can you explain the distinction between testing for noninferiority rather than superiority?
This distinction is crucial basically for a statistical design because, if you want to test the hypothesis of noninferiority, we need a lower number of patients to be enrolled. We wanted to test the noninferiority because we want to be sure that we are not exposing our patients to a higher risk of infection, obviously. We don’t want to demonstrate the superiority of nonrestrictive diet, because it’s quite obvious that if you understand that 2 things are equal and one is more complex than the other, you will choose the easy path. So that’s the rationale behind the design of the study.
This transcript has been lightly edited for clarity.
