ICE (Ifosfamide, carboplatin, and etoposide) Was the Best Salvage Regimen in Patients with Relapsed or Refractory Malignant Lymphoma

Background and Purpose

Several trails, which compared efficacy of salvage chemotherapies for relapsed or refractory lymphoma, have been conducted previously. In CORAL study, it documented that R-ICE was not inferior to R-DHAP in patients with relapsed or refractory DLBCL. In LY12 study, it documented that GDP was not inferior to DHAP in patients with relapsed or refractory NHL. However, there has been unknown the priority among the salvage regimens for relapsed or refractory lymphoma.

Patients and Methods

We retrospectively analyzed patients with lymphoma in first relapsed or who were refractory to initial therapies received ICE, DHAP or GDP. Salvage regimens were as followed; ICE: ifosfamide 1700 mg/m² on day1 and 2, 1600 mg/m² on day3, carboplatin AUC=5, maximum dose 800mg, on day 1, etoposide 100 mg/m² day 1 through to 3, mesna 20% dose of ifosfamide day 1 through to 3; DHAP: dexamethasone 40 mg/body day 1 thorough to 4, cisplatin 100 mg/m² over 24 hours on day 1, cytarabine 2000 mg/m² per 12 hours on day 2; GDP: gemcitabine 1000 mg/m² on day 1 and 8, dexamethasone 40mg/body day 1 through to 4 and cisplatin 25 mg/m² day 1 through to 3. Rituximab 375 mg/m² was added on the day 0 if CD20 was positive. The each doctor who was in charge of each patient determined the choice of regimen. Patients who were younger than 66 years old have received HDT followed by auto-SCT if they responded to the salvage chemotherapy and succeeded in PBSCH. Doses of chemotherapy were reduced in cases that age was older than 80 or when side effects were occurred by the treatment.

Results

Overall, 113 patients had disease progression and received salvage chemotherapies in our institute hospital from September 2003 to July 2014. Patients characteristics: median age was 62 (> 65: 63% for ICE vs. 78% for DHAP vs. 27% for GDP, p=0.001); men: 63%; histology: DLBCL 76% (11% were transformed from FL), PTCL-nos 6%, ALCL 1.8%, AITL 3%, NK/T 1.8%, HL 12%; IPI risk factors 0-2: 44%, 3: 26%, 4-5: 28% in NHL and IPS risk factors 0-2: 64% 3: 18%, 4-7: 18% in HL; refractory to initial therapies: 39%; early relapse (< 12 months) after initial therapies: 56%; prior rituximab exposure: 100% for CD20 positive; the proportion of received salvage regimen: ICE 65%, DHAP 21% and GDP 13%, respectively. Baseline characteristics were almost same except the rate of age < 66 as mentioned above and the rate of CNS involvement before salvage chemotherapy (ICE 5% vs. DHAP 30% vs. GDP 0%, p=0.002). ORR/CRR was 62%/51% for ICE vs. 43%/39% for DHAP vs. 57%/29% for GDP, respectively (p=0.23). ORR/CR for patients who had early relapse or refractory disease was 56%/15% for ICE vs. 33%/7% for DHAP vs. 43%/29% for GDP, respectively (p=0.05). The auto-SCT rate was 13% for ICE vs. 8% for DHAP vs. 7% for GDP, respectively (p=0.135). With median 28 months follow-up, the median time from first progression disease to second progression or last follow-up (2^nd PFS) was 219 days for ICE vs. 93 days for DHAP vs. 171 days for GDP, respectively (p=0.12). The median 2^nd PFS was 560.9 days and 245.1 days in ICE and non-ICE arms, respectively (p=0.01). There was also significant difference in the median 2^nd PFS between ICE and DHAP arms (p=0.04), while there was no difference between ICE and GDP arms (p=0.365). ICE showed significant improvement in the 2^nd PFS even in the patiens who had early relapsed or refractory diseases: 25.2% vs. 10.9% with non-ICE arm (p=0.04). The 2-year overall survival (OS) was no difference between ICE and non-ICE arms (68.1% and 54.8%, respectively, p=0.137). Although CNS involvement before therapy was observed significantly in DHAP arm, there was no difference in the median 2^nd PFS (321 days vs. 1053 days, p=0.369) and 2-year OS (43.6% vs. 64.2%, p=0.486) in CNS and non-CNS arms, respectively. There was no difference in grade 3/4 hematological side effects including neutropenia (p=0.12), erythrocyte transfusion (p=0.72) and platelet transfusion (p=0.09) between each arm. On the other hand, patients received DHAP experienced grade 3/4 renal dysfunction more than other arms (p=0.02). In multivariate analysis, the 2^nd PFS was significantly related to secondary CR (p=3.57×10^-10) and age less than 66 (p=0.0002).

Conclusion

ICE showed significant improvement in the 2^nd PFS but no improvement in OS. Although rate of CNS involvement before salvage therapy was significantly higher in DHAP arm, the 2^nd PFS and the OS were not affected. DHAP increased incidence of grade 3/4 renal dysfunction compared with GDP and ICE.