Safety and Feasibility of Outpatient Autologous Hematopoietic Stem Cell Transplantation in Pediatric Patients with Central Nervous System Tumors

Introduction

High-dose intensive chemotherapy with autologous hematopoietic stem cell transplantation (autoHSCT) is a well-established treatment for pediatric central nervous (CNS) tumors. Historically, autoHSCT was done primarily in the inpatient setting given risk of toxicities and infection risk in the setting of prolonged cytopenias. Prior data in adult literature has shown excellent outcomes for multiple myeloma patients who underwent outpatient autoHSCT (Abid et. al. Bone Marrow Transplant 2017). To date, there is no published pediatric data comparing outcomes and complications of pediatric CNS tumor patients who have undergone inpatient and outpatient autoHSCT.

Materials and Methods

We performed a retrospective cohort study analyzing a total of 37 pediatric CNS tumor patients, with a total of 49 outpatient and 41 inpatient transplant episodes who underwent autoHSCT at a single institution. Our primary endpoints included treatment-related mortality (TRM), overall survival (OS), time to neutrophil and platelet engraftment, types and sites of infection, duration of antibiotic therapy and length of hospitalization and intensive care (ICU) stays.

Results

The most common primary diagnosis was medulloblastoma (51.4%). Median age at time of diagnosis was 2 years (range 0.6-21 years). Median age at time of autoHSCT was 3 years (range 1-23). Median length of hospital stay for transplant episodes in the outpatient setting was 0 days and 17 days for transplant episodes in the inpatient setting. Patients who had stem cell infusions performed in the outpatient setting had a median time to hospitalization of day +6, with febrile neutropenia being the most common reason for admission. In total, 7 outpatient transplant episodes did not require any inpatient hospital care. A total of 40 outpatient transplant episodes (81.6%) did not require admission to the intensive care unit (ICU) compared to 32 (78%) inpatient transplant episodes. Median time spent in the intensive care unit was 0 days (range 0-4 days) for the outpatient cohort compared to 0 days (0-37 days) for the inpatient cohort. Median time to neutrophil engraftment was the same in both cohorts at 11 days (p=0.02). Time to platelet engraftment for the outpatient cohort was 13 days (range 8-82 days) compared to 16 days (range 8-106 days) in the inpatient cohort (p=0.0008). Bacteremia was the most common infection isolated in the inpatient cohort (40%) compared to bacterial stool infections (46.2%) of the outpatient cohort. Average number of days of antibiotics for outpatient cohort was 7.2 days compared to 18.6 days in the inpatient cohort. Median time to infection identified in the outpatient group was day +6 (range days +2 to +28) compared to day +7 (range days -5 to +120) in the inpatient group. At a median follow-up of 1.67 years, overall survival for all patients was 66.1%. At a median follow-up of 619 post-transplant days, transplant-related mortality (TRM) was 0% for both cohorts. There were a total of 12 deaths that occurred, 100% of cause of death was secondary to progressive disease or non-transplant related treatment adverse effects.

Conclusion

AutoHSCT for properly selected pediatric CNS tumor patients performed in the outpatient setting is safe and effective.