The NEUTRODIET trial: Is a protective diet necessary after SCT?

Infections are a major cause of morbidity and mortality during neutropenia following stem cell transplant (SCT).¹ It was hypothesized over 50 years ago that reducing pathogens in the gastrointestinal (GI) tract by excluding specific foods that may act as a vector for bacteria could be beneficial in neutropenic patients.²

The first diet with the aim of providing a completely germ‐free food intake was developed in the 1960s,² and today, a low-microbial restrictive diet is a standard of care adopted in around 90% of transplant centers to prevent infections.¹

Despite the widespread use of this strategy as a standard of care, its efficacy had never been tested prospectively; therefore, evidence-based results and standardization are lacking.¹ Observational data have highlighted a paradoxical link between use of a restrictive neutropenic diet and higher rates of infection. Furthermore, a period of low oral feeding after allogeneic SCT correlates directly with a higher incidence and severity of acute graft-versus-host disease (GvHD).¹

The phase III NEUTRODIET trial was designed to assess the role of a non-restrictive diet (NRD) versus a protective diet (PD) in terms of infection rate, feeding outcomes, and acute GvHD incidence in autologous and allogeneic hematopoietic SCT (HSCT) recipients.¹ Here, we summarize key results and clinical impacts from this trial, as presented by Stella at the 64th American Society of Hematology (ASH) Annual Meeting and Exposition.¹

Study design¹

The NEUTRODIET trial was a multicenter, randomized, non-inferiority, phase III trial consisting of adult patients undergoing HSCT or high-dose induction chemotherapy. A total of 247 patients were enrolled, of which 222 patients were randomized 1:1 to PD or NRD from initiation of chemotherapy for the entire duration of neutropenia. The PD consisted of foods cooked at >80ᵒC and excluded all raw foods, while the NRD diet was compliant with hospital hygiene standards and excluded raw fish and meat. There was a planned follow-up of 100 days for allogeneic recipients and 30 days for other patients.

The primary endpoint was to:

• demonstrate the absence of significant differences in infection (Grade ≥3) and death rates during the period of neutropenia between the two arms (PD vs NRD).

Secondary endpoints included assessment of:

• GI infections and fever of unknown origin;
• overall survival;
• the cumulative incidence of acute GvHD; and
• nutritional status.

Further details on the study design and patient characteristics can be found in this previous article published on the GvHD Hub.

Results³

No significant difference was found between the two arms in the cumulative incidence of infections, incidence of fever of unknown origin, sepsis, documented GI infection, GI symptoms, diarrhea, and a microbiological isolation, bodyweight variations, incidence of nausea and mucositis, hospitalization length, parenteral nutrition use, duration of parenteral nutrition, incidence of acute Grade ≥2 GvHD in allogeneic HSCT recipients, any grade acute GvHD, or intestinal acute GvHD (Table 1).

Table 1. Primary and secondary endpoint outcomes in the NEUTRODIET trial*

CI, confidence interval; GI, gastrointestinal; GvHD, graft-versus-host disease; HSCT, hematopoietic stem cell transplant RR, risk ratio. *Data from Stella, et al.3
Outcome, % (unless stated otherwise)	Protective diet (n = 112)	Non-restrictive diet (n = 112)	RR (95% CI)	p value
Cumulative incidence of Grade >2 infections	34	39	0.86 (0.6–1.2)	0.5
Fever of unknown origin	43	39	1.3 (0.9–1.7)	0.2
Sepsis	11	14	0.7 (0.4–1.5)	0.5
Documented GI infection	7	3	2.7 (0.8–9.1)	0.2
GI symptoms, diarrhea, and a microbiological isolation	34	30	0.9 (0.6–1.3)	0.7
Mean bodyweight variations, kg	−3.6	−3.2	—	0.33
Incidence of nausea	16	15	1.1 (0.6–1.9)	>0.99
Incidence of mucositis	60	62	1.05 (0.8–1.3)	0.8
Mean hospitalization length, days	21	22	—	0.47
Parenteral nutrition use	23	26	0.9 (0.4–1.4)	0.8
Mean duration of parenteral nutrition, days	6.9	6.7	—	0.8
Incidence of Grade ≥2 acute GvHD in allogeneic HSCT recipients	17	25	0.7 (0.2–2)	0.7
Any grade GvHD	30	33	0.9 (0.4–2.1)	>0.99
Intestinal GvHD	13	8	1.6 (0.3–7.4)	0.6

Analysis of daily diaries completed by patients revealed that the NRD was associated with higher satisfaction, with 16% of patients receiving PD vs 35% receiving NRD reporting that “diet prescription did not negatively impact my alimentation” (risk ratio, 0.5; 95% confidence interval, 0.3–0.8; p = 0.006).

In both arms, the bacteria isolated most frequently from blood cultures belonged to the Enterobacteriaceae family, and the most frequently isolated bacteria from stool samples was Clostridium difficile. Only one death was reported at 30 days; the patient was in the NRD arm, but the death was unrelated to diet or infection.

Conclusion

No significant difference in infection rates, feeding outcomes, and acute GvHD incidence was found between patients receiving a PD versus a NRD. In addition, patient satisfaction was improved with a NRD versus a PD. These results, combined with data published in settings other than post-transplantation, suggest that adherence to a restrictive diet is an unnecessary burden for patients in this setting, and that patient quality of life can be improved without detrimental impact by adopting a NRD. These findings indicate that a shift is required in the current dietary standard of care adopted in the majority of transplant centers.