Efficacy
The PARMA trial provided excellent data on the role of DHAP in treatment of relapsed NHL, with or without high dose chemotherapy followed by ABMT. At the end of the first two cycles of DHAP, 109 patients (41% in CR, and 59% in PR) were randomised to either BEAC followed by ABMT, or continuing with up to 4 more cycles of DHAP.r
The following points are worth noting:
- the group was by and large a favourable group, with the aim at that time to test the toxicities of ABMT
- source of stem cells was unstimulated (no G-CSF) bone marrow
- involved field radiation therapy was built into the protocol for both arms of the study, according to predefined criteria (bulky disease = 5cm or extranodal T3 or T4 lesions as defined by EORTC)
- the study was conducted well before the rituximab era.r
Comparing the ABMT arm vs chemotherapy:
- response rate (CR+PR) 84% vs 44% at completion of all treatment
- EFS 46% vs 12% (p=0.001)
- OS at 5 years 32% vs 13% (p=0.038)2


© New England Journal of Medicine 1995
Time to first relapse is correlated with response after the first two cycles of DHAP, as well as 8-year survival rates in the PARMA trial.r Age adjusted IPI at relapse is correlated with OS in the DHAP group but it has no effect in the BEAC group. In addition, ABMT was not superior in patients with an age adjusted IPI of 0.r
In the HOVON-44 trial comparing DHAP to R-DHAP, 14 out of 239 patients enrolled, were excluded from analysis. A total of 225 patients were analysed on ITT basis – 112 in DHAP arm, and 113 in R-DHAP arm. The majority of the patients had DLBCL and had received CHOP-like first line treatment. Only 4 patients in each arm had prior exposure to rituximab. The two groups were well balanced in prognostic factors and demographics.r
Response rate following chemotherapy favoured the R-DHAP arm: 75% vs 54%. Patients showing response were eligible to proceed to ASCT. Inadequate blood PBSC harvest was observed in 1 patient in the DHAP arm, and 3 in the R-DHAP arm. Six patients in DHAP and 7 in the R-DHAP arm respectively progressed prior to ASCT and went off protocol. Radiation therapy after ASCT was given to 3 patients in the DHAP arm and 9 patients in the R-DHAP arm. The overall response rates were: DHAP – CR 35%, PR 15%; R-DHAP – CR 46% and PR 27%.r
Median follow up was 31 months. At 24 months, FFS was 24% in DHAP arm, and 50% in the R-DHAP arm (p<0.001). PFS at 24 months were: 31% vs 52% (p<0.002). OS at 24 months were: 52% vs 59% (p=0.15).r

© Blood 2008
Thus, in this essentially rituximab-naïve group of patients, addition of rituximab to DHAP-like salvage regimen prior to ASCT led to significant improvement in EFS and PFS, without significant increase in toxicities or delay in haemopoietic recovery following ASCT (data not shown). However, the benefit for OS cannot be demonstrated with confidence.r This may be due to successful third line salvage treatment. Note (1) there were only 3 doses in total of rituximab in the R-DHAP arm, and (2) patients relapsing or progressing after rituximab-containing front line therapy are not adequately examined in this study (see CORAL study, below).
R-DHAP vs R-ICE (ifosfamide, carboplatin, etoposide) was examined in the CORAL study, a phase III multicentre randomised trial in relapsed or primary refractory DLBCL.rr Of a total of 396 patients (median age 55 years), 62% had received prior rituximab. Overall efficacy measurements were similar between R-DHAP and R-ICE, with different toxicity profiles. Overall, 3-year EFS was inferior in patients with prior rituximab treatment (21% vs 47%). Of interest, patients with initial CR >12 months after diagnosis had similar EFS whether they had received rituximab (D below). This suggests patients with early relapses after rituximab-containing front-line therapy constitute a prognostically adverse group (C below).r Data for rituximab maintenance was not sufficiently mature for analysis at time of publication.

© Journal of Clinical Oncology 2010
A GELA phase 2 study examining R-DHAP, alternating with R-CHOP 21, followed by ASCT in mantle cell lymphoma patients younger than 66 years was reported in 2013. Patients had at least stage-3 disease. Sixty patients were enrolled, with a median age of 57 years. This study examines the role of adding high dose cytarabine and 4 doses of rituximab to this particular chemotherapy backbone in mantle cell lymphoma, which has been investigated previously by GELA.r The following table shows the demographics:

© Blood 2013
The following is the treatment schedule

© Blood 2013
Peripheral blood stem cells were harvested after the third cycle of R-DHAP. ASCT conditioning included TBI of 10 Gy over 3 days in twice daily fractions, high dose cytarabine and high dose melphalan. Primary end point was EFS. Of 53 patients still on study at the end of induction, 49 proceeded to ASCT. BEAM was used in 7 patients because of unavailability of TBI, or induction toxicity. At end of induction, 57% of patients were in CR and 30% in PR. There was a notable proportion who converted from PR after R-CHOP to CR after R-DHAP. For patients who received the entire planned treatment, including ASCT, CR was 96% and PR 4%. On ITT analysis for the entire cohort, CR was 78%, and ORR 82%.r
Median follow-up from diagnosis was 67 months. Median EFS was 83.9 months with 5-year EFS estimate of 64%. Relapse or progression occurred in 16 patients. Median PFS was 84 months, and median OS was not reached. OS rate at 5 years was 75%.r