Safety and Efficacy of High-Dose Cyclophosphamide, Etoposide, and Ranimustine (CEM) Regimen Followed by Autologous Peripheral Blood Stem Cell Transplantation for Patients with High Risk De Novo Diffuse Large B-Cell Lymphoma or Diffuse Large B-Cell Lymphoma Associated with Follicular Lymphoma.

We evaluated the safety and efficacy of the CEM regimen containing ranimustine (MCNU) with autologous peripheral blood stem cell transplantation (PBSCT) in adult patients with relapsed or high-risk de novo DLBCL or DLBCL associated with follicular lymphoma (FL/DLBCL) in our institution.

We retrospectively analyzed 55 consecutive patients who underwent autologous PBSCT following the CEM regimen between March 1999 and June 2011 for the treatment of relapsed or high-risk DLBCL and FL/DLBCL. High-risk DLBCL was defined as partial or no response to initial treatment or high-intermediate/high risk disease according to international prognostic index (IPI) at initial diagnosis. Age-adjusted IPI was used for patients under 60. The CEM regimen consisted of CY (60 mg/kg on days -7 and -6), etoposide (500 mg/m² on days -6 to -4) and MCNU (250 mg/m²on day -3 and -2), followed by PBSCT.

Median age at transplant was 51 years (range, 23–66), and median time from diagnosis to transplant was 5 months (range, 3–241). Of them, 7 patients received radiation therapy before transplant, and 11 had no history of rituximab use before transplant. The 36 patients in CR1 at transplant were with high or high-intermediate risk according to international prognostic index at diagnosis. Other 15 patients in CR2, and 4 patients who were not in CR1/2 at transplant were included. CY dose was reduced in five patients by physicians choice. The median number of infused CD34-positive cells was 3.98 × 10⁶/kg (range, 1.36–26.87). The median post-transplant days of neutrophil recovery and platelet were 11 days (range, 9–20) and 12 days (range, 7–53), respectively. Common grade 3/4 non-hematological treatment-related toxicities within the first 100 days after transplant were nausea (24%), vomit (4%), stomatitis (15%), and diarrhea (9%). Infections included febrile neutropenia (85%), sepsis (15%) which contained catheter-related bacteremia (n=4) and pneumonia (n=2). Other less common severe toxicities were acute renal impairment (n=3), and pleural effusion (n=1). No sever cardiac, neurologic toxicity, or veno-occlusive disease of the liver was observed in any of the patients. Late-onset adverse effects during follow-up period includes chronic renal impairment (n= 5), therapy-related myelodysplastic syndrome (refractory anemia) (n=1, 120 months), and prostate cancer (n=1, 83 months). Serious late cardiac or pulmonary impairment was not diagnosed in this cohort. Median follow-up duration of 42 patients surviving at the time of the analysis was 52 months (range, 1–159). Relapse or disease progression after PBSCT was documented in 21 cases (range, 0–81 months after PBSCT). No therapy related mortality (TRM) associated with PBSCT was observed, and all the 13 deaths were due to disease relapse/progression. The 5-year estimated overall survival (OS) and progression-free survival (PFS) were 70.6% (95% CI, 54.0–82.1)% and 57.0% (95% CI, 39.5–71.2)% for all patients, respectively. In a univariate analysis for factors affecting OS and PFS, no significant factors associated with unfavorable OS and PFS was found including age at transplant, gender, history of rituximab use, histology, disease status at transplant, and previous BM involvement.

We conclude that the CEM regimen using MCNU would be a tolerable, effective conditioning regimen of autologous PBSCT for high-risk or relapsed de novo DLBCL or FL/DLBCL. Although no TRM was observed, relapse was the major cause of treatment failure. Late-onset complications in long-term survivors can occur, and should be carefully monitored.

No relevant conflicts of interest to declare.