Unavailability of Carmustine and its dramatically increased cost by nearly 10 fold has led to its replacement by bendamustine (Be) and an increase use of Bendamustine-based (BeEAM) conditioning regimen before autologous stem cell transplantation for lymphoma. The aim of this study was to evaluate the safety of the BeEAM regimen in the real life use in several French centers. Median age of the 386 patients (233 male, 153 female) was 55 years (17-72). The majority of patients had aggressive B-cell (40%), follicular (17%), Hodgkin's (17%) and mantle cell lymphoma (15%). Two hundred and forty six patients (64%) were transplanted after 2nd line chemotherapy and the median number of prior lines of chemotherapy of 2 (1-9). Two hundred and sixty patients (67%) had received prior platinum chemotherapy and only 4% had a history of prior chronic renal failure (Table 1). The median creatinine level at conditioning initiation was 71(36-206) µmol/l.

Be was administered at a median dose of 191 mg/m2/day (50-250) on day -7 and -6. Thirty two percent received 100-160 mg/m²/d and 58% of patients received more than 160 mg/m²/d. Median 24h-hydration volume was 3 l (1-6) which began within median time of 18 h (1-72) prior to conditioning initiation. The median duration time of Be perfusion was 60 min (30-143).

Grade 1-4 acute renal failure (ARF) was reported in 107 cases (28%) (G ≥2; 34%) and appeared after a median time of 2 days (1-18) after conditioning start. Melphalan dosage was reduced, due to renal failure, in 10% of patients (Table 2). 84% of patients normalized their creatinine level within a median time of 10 days (1-77). The most frequent reported Grade 1-4 toxicities were mucositis (84%), gastroenteritis (53%), skin toxicity (34%), colitis (29%), liver toxicity (19%), pneumonitis (5%) and cardiac rhythm disorders (4%). Opportunistic infection was documented in 9.5% of patients, HHV6 reactivation in 6% of patients. Ten percent of patients needed intensive care management and toxic death was estimated at 5% (Table 3). Patients received a median of 3 (0-34) packed red blood cells and 4 (0-56) platelets units. The median time to neutrophils > 0.5 G/L and platelets > 20 G/L were 10 (0-50) and 12 (0-210) days. The median time of hospital stay was 23 (12-85) days. In comparison with the group of patient without renal failure, the group of patients with ARF had older age (58 vs 54), higher rate of pre-transplant chronic renal failure (10% vs 1%), higher rate of platinum treatment (77% vs 64%), higher day1 creatinine level (81 vs 69) and received higher median bendamustine (199 vs 182). Colitis (p=0.039), pneumonitis (p=0.008), cardia arrhythmia (p=<0.0001), intensive care admission (p=<0.0001), need for blood transfusion (p=<0.0001), hospital stay duration (p=<0.0001), and death (p=<0.0001), were more frequent in patients with post conditioning renal failure. In multivariate analysis, creatinine level at day1, bendamustine dose>160mg/m2 and age were independent prognostic factors for ARF. A three-point clinical predictor score for acute renal failure was identified and included creatinine level>65µmol/l, bendamustine dose >160mg/m² and age>57 years. ARF stratified by score, was 4% with score 0, 17% with score 1, 30% with score 2 and 45% with score 3.

In conclusion, BeEAM induced 5% non-relapse mortality with high rate of renal toxicity. A simple, three-point scoring system can stratify patients by levels of risk for ARF. Rapid identification of higher risk patients may allow a reduction of the bendamustine dose to improve clinical outcomes. Prospective comparative studies are needed to confirm toxicity extents of this conditioning as compared with other type of high dose therapy.


Soussain:Pharmacyclics: Research Funding; Celgene: Research Funding; Roche: Research Funding. Malak:Novartis: Membership on an entity's Board of Directors or advisory committees.

Author notes


Asterisk with author names denotes non-ASH members.