Non-Restrictive Diet Does Not Increase Infections in Patients with Neutropenia after Stem Cell Transplantation: Final Analysis of the Neutrodiet Multicenter, Randomized Trial

Background – Infections are a leading cause of morbidity and mortality during the period of neutropenia following high-dose chemotherapy, especially after hematopoietic stem cells transplantation (HSCT).

Among measures applied to prevent infections, the use of low-microbial Protective Diet (PD) is a standard of care adopted in 80-90% of Bone Marrow Transplantation (BMT) centers.

The extensive use of PD happens despite its efficacy has never been tested prospectively and standardization among BMT centers is still lacking. Our aim was to assess the role of a Non-Restrictive Diet (NRD) versus PD in terms of infection rate, feeding outcomes and aGVHD incidence in autologous and allogeneic-HSCT recipients.

Study design – We conducted a multicenter, randomized, non-inferiority, phase III trial enrolling all consecutive adult patients (pts) undergoing HSCT or high-dose induction chemotherapy. Eligible patients were randomly assigned in a 1:1 ratio to receive PD or NRD from the start of chemotherapy for the entire duration of neutropenia. Planned follow-up was 100 days for allogeneic recipients and 30 days for other pts.

Results – Overall, 247 pts were enrolled. Between July 2016 and March 2022, a total of 224 pts underwent randomization and were analyzed: 112 were assigned to the PD arm and 112 to the NRD arm (Figure 1). Two pts withdrew consent, 11 were improperly enrolled for protocol violation, 10 were lost at follow-up and were not included in this analysis.

The clinical and demographic characteristics were similar in the two groups. The arms were well balanced in terms of gender, age, disease, use of antimicrobial prophylaxis and duration of neutropenia. Seventy-six percent of the analyzed pts received autologous-HSCT and 21% received allogeneic-HSCT (Table 1).

No difference was found in cumulative incidence of infections: G>2 infections (according to CTCAE 4.0) occurred in 38 (34%) and 44 patients (39%), assigned to PD and NRD respectively (RR = 0.86; 95% Confidence Interval [CI] = 0.6-1.2, p=0.5).

The incidence of fever of unknown origin (including febrile neutropenia) and sepsis was similar in the two arms, respectively 43% in PD vs 39% in NRD (RR=1.3; 95% CI = 0.9-1.7, p=0.2) and 11% in PD vs 14% in NRD (RR=0.7; 95% CI = 0.4-1.5, p=0.5). Bacteria most frequently isolated from blood cultures belonged to Enterobacteriaceae family in both arms.

Only one death was reported at 30 days; death occurred due to cytokine release syndrome after ruxolitinib suspension in a pt, assigned to NRD arm, undergoing allogeneic-HSCT for myeloproliferative neoplasm.

The incidence of documented GI infection, defined as the presence of GI symptoms, diarrhea, and a microbiological isolation, were comparable between arms (7% in PD vs 3% in NRD, RR=2.7; 95% CI = 0.8-9.1, p=0.2 and 30% in NRD vs 34% in PD; RR=0.9; 95% CI = 0.6-1.3, p=0.7). In both arms Clostridium difficile was the most frequently isolated bacteria at stool test.

No differences were found between arms in terms of body weight variations (mean -3.6 kg in PD vs -3.2 kg in NRD, p=0.33), incidence of nausea and mucositis (16% in PD vs 15% in NRD; RR 1.1, 95% CI = 0.6-1.9, p>0.99 and 62% in PD vs 60% in NRD; RR 1.05, 95% CI = 0.8-1.3, p=0.8), hospitalization length (mean 21 days in PD vs 22 days in NRD, p= 0.47), parenteral nutrition (PN) use (23% vs 26%; RR=0.9; 95% CI = 0.4-1.4, p=0.8) and PN duration (mean 6.9 days vs 6.7 days, p=0.8).

Focusing on allogeneic-HSCT recipients, the incidence of aGVHD grade ≥2 was similar between the arms (17% in PD vs 25% in NRD; RR=0.7; 95% CI=0.2-2, p=0.7).

No difference was found in terms of any grade aGVHD and intestinal aGVHD, respectively 30% in PD vs 33% in NRD (RR=0.9; 95% CI = 0.4-2.1, p>0.99) and 13% in PD vs 8% in NRD (RR=1.6; 95% CI = 0.3-7.4, p=0.6).

From an analysis of daily diaries filled by pts Non-Restrictive Diet was associated with higher satisfaction, with 16% of pts receiving PD vs 35% in NRD reporting that “diet prescription did not negatively impact my alimentation” (RR=0.5; 95%CI = 0.3-0.8; p=0.006).

Conclusions – To the extent of our knowledge this is the first randomized trial investigating the role of diet after HSCT. No significant difference in terms of infection rates, feeding outcomes and aGVHD incidence were found between Protective Diet and Non-Restrictive Diet. These results, together with data published in settings other than post-transplantation, demonstrate that the use of a restrictive diet is an unnecessary burden for patients’ quality of life