Is G-CSF Still Needed Post-Transplant to Promote Engraftment in the Present Era? a Multi-Disciplinary Project to Evaluate Patient Safety Vs Cost Savings

Background:

The use of granulocyte colony stimulating factor (G-CSF) to promote engraftment after hematopoietic cell transplantation (HCT) remains controversial. Randomized controlled trials that showed a shorter duration of neutropenia after G-CSF in autologous (auto) HCT recipients were performed in an era when present supportive care resources were not available. The use of G-CSF after allogeneic (allo) HCT is not established by randomized trials and there is a concern that it may be associated with an increased risk of graft-versus-host disease. G-CSF is a costly drug and excluding its routine use may translate into significant cost savings for a transplant program. All inpatients transplanted routinely receive G-CSF 480 mcg/day starting day +5. We conducted a pilot study to evaluate if G-CSF post-HCT could be safely omitted after autologous and allogeneic HCT.

Methods:

2013 data was used as benchmarks for neutrophil engraftment and hospital length of stay (LOS), calculated from day 0. Three separate pilots were conducted for auto HCT, myeloablative (MAC) allo and reduced-intensity (RIC) allo HCT recipients. Eligibility criteria included sufficient cell dose for the product to be infused (PBSC ≥5.0 x10^6 CD34+ cells/kg for autos, ≥2.0 x10^6/kg for allos or BM ≥2.0x10^8 TNC/kg). G-CSF was not administered prophylactically, but could be given in clinical scenarios such as prolonged febrile neutropenia or engraftment not occurring by day +21.  Accrual target was 20 patients for each group with a planned interim analysis after ~10 patients. For each pilot patient, a control was randomly selected from 2013; controls were matched by age, graft source, diagnosis and cell dose. Implementation of the pilot was the effort of a multi-disciplinary team including physicians, mid-level providers, nurses, clinical pharmacists, data and quality assurance personnel.

Findings:

For the auto and allo MAC groups, interim analysis revealed that omission of G-CSF led to longer LOS and longer time to neutrophil engraftment (see Table).  The interim analysis of the allo RIC group appears to be comparable for the length of stay and neutrophil engraftment. 

 

Discussion:

Any cost savings of not using G-CSF are likely to be offset by the longer duration of post-transplant hospitalization and possible increased risks due to longer periods of neutropenia.  Based on our findings, the pilot has been discontinued for auto and allo MAC transplants, where we will continue to use G-CSF starting on Day +5 to promote engraftment.  This pilot will accrue the planned 20 patients for allogeneic RIC patients before a final analysis is performed.

Table