Abstract 685


In 2004 we have shown that first-line intensive therapy plus transplantation of autologous hematopoietic stem cells lead to an improved EFS and OS over standard CHOP regimen in adults with disseminated aggressive lymphomas (NEJM 2004; 350: 1287). Then the rituximab era came. We have shown that the addition of R to the intensive therapy was feasible and might further improve the results (BBMT 2010; 16: 672).

We present here the preliminary analysis of a multicenter randomized trial aiming at comparing the results of R- CHOP every 14 days to R- HDT in adults with untreated confirmed DLBCL (clinical trial.gov: NCT 00561379).


The patients were 18 to 60 y.o. with CD20 + DLBCL Ann Arbor stage 1 or 2 with bulk >/= 7 cm or 3 and 4 were randomized at diagnosis (with a stratification according to age adjusted IPI) between R-CHOP every 14 days at standard doses (8 consecutive courses if a response was observed after the first 4 courses) and R-HDT. This program consisted of 2 courses of high-dose CHOP-like regimen, 15 days apart, with rituximab (375/mg/m2) on day 1 of each course, followed by rituximab on d 22, harvest of G-CSF mobilised peripheral blood stem cells on d 28,29, then rituximab on d 36 followed by a course of high-dose methotrexate with cytarabin. For patients who achieved at least a partial response after these 3 courses, a BEAM regimen started on d 66 to 80 followed by the infusion of stem cells. In each arm, the intermediate evaluation of response was assessed by means of standard CT Scan and PET. Patients not achieving a PR or a negative PET were proposed a salvage Tx followed in case of response by an autologous transplantation. The main objective of the study was the EFS with an event defined as insufficient intermediate response, progression, relapse or death. Analysis was performed on intent to treat basis.


331 pts were included from 01/2005 to 05/2010. 305 patients were found eligible and 286 are fully evaluable at time of this abstract, 143 in each arm. The main characteristics of the pts in the 2 arms are strictly super imposable. Overall 56% had a bulk as defined, 72% had a 3 or 4 AA stage and 58% had an aa IPI 2 or 3. The treatment as scheduled in the protocol was completed in 71% in the R-CHOP arm and 60% in the R-HDT arm. The intermediate evaluation showed exactly the same rate of response with Standard CT scan (CR + CRu: 65% and 62% following 4 courses R-CHOP and first 3 courses of R-HDT respectively). PET remained significantly more often + after the first 3 courses of R-HDT (43% vs 30% following 4xR-CHOP; p=0.03). The ORR (CR + Cru) was 78% and 71% in the R-CHOP and R-HDT arms respectively (p=0.8).

With a med FU of surviving patients of 25 m, the probability of survival for the whole group of patients is 88% and 74% at 2 and 5 years. There is no significant difference according to the treatment arm in any of the aaIPI strata. The EFS with the standard CT Scan as an intermediate evaluation is 74% and 66% at 2y and 5 y with no difference according to the treatment arm in any aaIPI strata. The EFS with PET scan as an intermediate evaluation is 49% and 45% at 2y and 5 y and is statistically worse with R-HDT for aaIPI 2 and 3 pts (p=0.01).


Based on these results, one cannot recommend first-line R-HDT as performed in that trial for adult patients with DLBCL. R-CHOP 14 is as efficient, less toxic and spares resources.


Milpied:Roche: Membership on an entity's Board of Directors or advisory committees.

Author notes


Asterisk with author names denotes non-ASH members.