Excess Late Gastrointestinal Toxicity Associated with Leam Conditioning Compared to BEAM for Autologous Stem Cell Transplantation in Lymphoma, a Single Institution Experience

Introduction

A standard conditioning regimen for patients undergoing autologous stem cell transplant (ASCT) for relapsed/refractory lymphoma in Europe has consisted of BCNU, etoposide, cytarabine and melphalan (BEAM).

BCNU has not been readily available for several years. Thus, in our institution, since August 2008 it has been substituted by oral lomustine (LEAM). There is very limited information on the relative toxicities of the two regimes, and to address this we undertook a retrospective analysis of two comparable groups of transplant patients, to assess the relative toxicity profiles.

Methods and Patients

Two cohorts of 50 patients undergoing ASCT for relapsed/refractory lymphoma were compared, one conditioned with BEAM (BCNU 300mg/m², etoposide 1600mg/m² (in divided doses), cytarabine 800mg/m² (in divided doses) and melphalan140mg/m² and the other with LEAM (lomustine 200mg/m² instead of BCNU).

Toxicities were assessed for each cohort, using the nursing and medical notes, according to CTCAEv3.0 criteria: nausea, vomiting, stomatitis and diarrhoea, alanine liver transaminase (ALT) for hepatic toxicity and creatinine for renal toxicity. Surrogate markers for mucosal toxicity were also assessed, such as use of anti-diarrhoeals, opiates and syringe-driver anti-emetics.

Results

The unpaired t-test was used for continuous variables and Fisher's exact test for categorical values.

8 LEAM + 5 BEAM patients were not evaluable 2 weeks post discharge.

à9 LEAM + 8 BEAM patients were not evaluable at 3 months.

º47 LEAM and 46 BEAM patients assessable at 3 months.

There was minimal non-relapse mortality in both cohorts, with 2 early deaths with LEAM (at D+6 and D+8 respectively) and 1 early death with BEAM (day +7) all secondary to neutropenic sepsis.

As expected, significant mucosal toxicity was seen with both regimens, with 51% of patients having at least grade 3 diarrhoea, and approximately 30% having grade 3 stomatitis. There was no statistically significant difference in any of the toxicity parameters measured between the 2 regimens during the inpatient admission. The finding of a statistically significant excess of GI toxicity with LEAM at 3 months (mainly residual abdominal pain and diarrhoea) was not expected, and was not seen when assessed at an earlier time point (2 weeks post discharge).

Both regimens were associated with minimal hepatic and renal toxicity.

Conclusions

There appears to be an excess of late gastrointestinal toxicity associated with the LEAM regimen. However, this toxicity was not associated with a difference in time to engraftment and there was no detectable difference in non-relapse mortality. The cause of this GI toxicity signal is unclear, and requires confirmation and further investigation. There also appears to be no difference in clinical outcomes, however longer follow-up of patients conditioned with LEAM is required to confirm this finding in a more robust manner.