The prognosis of young DLBCL patients at high risk treated with standard R-CHOP is still rather poor. The role of intensified HDC+ASCT in first line treatment is still a matter of debate in the Rituximab era. The FIL planned a prospective randomized phase III trial with a 2×2 factorial design aimed at investigating the possible benefit of intensification with R-HDC+ASCT after R-dose-dense chemotherapy delivered at two different level of dose (R-CHOP14/R-MegaCHOP14). Patients and methods. The primary end-point was to increase 2-year Progression Free Survival (PFS) from 50% of the standard dose-dense arm (R-dose-dense chemotherapy) to 65% in the experimental arm (R-dose-dense chemotherapy followed by R-HDC +ASCT). Secondary end-point was the comparison between two different schemes of dose dense chemotherapy, R-CHOP14 and R-MegaCHOP14. Inclusion criteria were: age 18–65; untreated DLBCL; age-adjusted IPI (aa-IPI) score 2 or 3. Patients were stratified according to aa-IPI and randomized at diagnosis to receive: R-CHOP14 × 8 cycles; R-MegaCHOP14 (1200 mg/sqm Cyclophosphamide, 70 mg/sqm Doxorubicin, standard dose Vincristine and Prednisone) × 6; R-CHOP14 × 4 + R-HDC (Rituximab + High Dose Cytarabine + Mitoxantrone + Dexamethasone) followed by BEAM and ASCT; R-MegaCHOP14 × 4 + R-HDC + BEAM and ASCT. G-CSF support was mandatory. Central nervous system prophylaxis was done according to the Italian Society of Hematology guidelines.
From June 2005 to September 2010, 412 patients were enrolled. Histology was centrally reviewed in 90% of cases. Thirteen patients were excluded because of different histological subtypes in 10 and active hepatitis HCV and HBV in 3. 399 patients were eligible and randomized: 199 to R-HDC+ASCT and 200 to R-dose-dense chemotherapy without ASCT; according to the type of chemotherapy 203 were randomized to RCHOP14 and 196 to R-MegaCHOP14. All patients were evaluable for analysis. Clinical characteristics were: median age 49 (range 18;65); stage II/III/IV 6/29/65%; LDH higher than normal value 89%; ECOG PS >1 43%; aa-IPI score 2/3 74/26%; all characteristics were well balanced between patients treated with or without ASCT. In the R-HDC+ASCT group, 151 patients (76%) completed the treatment and 177 (88%) in the R-dose-dense chemotherapy arms. Complete Remission (CR) was seen in 296 (74%) patients; CR was 76% in R-HDC+ASCT vs. 72% in the R-dose-dense chemotherapy arms. Overall 26 patients (7%) had a partial remission and 64 (16%) did not respond. Treatment-related deaths occurred in 13 (3%) patients: 8 (4%) in the R-HDC+ASCT arms vs. 5 (2.5%) in R-dose-dense arms. Grade III/IV extrahematological toxicities were reported in 85 patients (43%) in the R-HDC+ASCT vs. 38 (19%) in R-dose-dense arms. With a median follow-up of 36 months, 3-year PFS and 3-year Overall Survival (OS) rates for the whole series were: 64% (95% CI:59–69) and 79% (95% CI:74–83) respectively. Patients randomized to R-HDC+ASCT had a 3-year PFS of 70% (95% CI:63–76) compared to 59% of those treated with R-dose-dense only (95% CI:51–66); p = .010 (Figure 1) with a HR of 0.64 (95% CI: 0.46–0.91, p = .012). No difference in 3-year PFS was observed between R-CHOP14 and R-MegaCHOP14. According to aa-IPI, 3-year PFS in R-HDC+ASCT vs R-dose-dense arms were: aa-IPI score 2 74% (95%CI:65–80) vs. 63% (95%CI:54–71) p = .047; aa-IPI score 3 59% (95% CI:45–71) vs. 46% (95% CI:32–59), p= .121. At the time of this analysis, 3-year OS is 81% (95% CI:74–86) in R-HDC+ASCT vs. 78% (95% CI:70–83) in R-dose-dense chemotherapy patients, p= .556. In a Cox-model including the four arms and assuming R-CHOP14 as reference, the risk of relapse was significantly reduced in both ASCT arms with a major effect in the R-CHOP14+R-HDC+ASCT arm (HR=0.58, 95% CI=0.36–0.93, p= .025) and a slight minor HR reduction in the R-MegaCHOP14+R-HDC+ASCT arm (HR=0.68, 95% CI=0.42–1.09 p= .109). Moreover PFS in both arms (ASCT vs no ASCT) was further compared within pre-planned subgroups analysis according to the type of dose-dense chemotherapy, age, gender, aa-IPI and bone marrow involvement: the benefit of R-HDC+ASCT in PFS was maintained across all subgroups with no statistically significant interaction.
R-dose-dense chemotherapy followed by R-HDC and BEAM with ASCT significantly reduced the risk of progression compared to standard dose-dense chemotherapy (R-CHOP14 or R-MegaCHOP14) in young patients with high-risk DLBCL.
Vitolo:Roche: Membership on an entity's Board of Directors or advisory committees. Zaja:Roche: Membership on an entity's Board of Directors or advisory committees.
Asterisk with author names denotes non-ASH members.