Efficacy
DA-R-EPOCH was prospectively compared to R-CHOP for unselected DLBCL patients in a CALGB trial.1 This prospective phase III study of 491 patients showed more early discontinuations in the DA-R-EPOCH arm with increased toxicity and no difference in either progression-free survival (PFS) (hazard ratio, 0.93; 95% CI, 0.68 to 1.27; p = 0.65) or overall survival (OS) (2-year 86.5% versus 85.7%). Factors such as MYC, BCL2 and BCL6 expression or rearrangement on fluorescence in situ hybridisation (FISH) were not included prospectively and only analysed on a subset of patients, with only three identified cases of double-hit lymphoma (DHL) found. 35 patients with primary mediastinal B-cell lymphoma were included in the analysis.
Figure 1. PFS, OS and PFS by subset in the Phase III Intergroup Trial Alliance/CALGB 503031



© Journal of Clinical Oncology 2019
Several studies have looked at dose intensification utilising DA-R-EPOCH as a means of improving outcomes in patients with non-GCB tumours which have shown poorer response to therapy and inferior survival compared to GCB type DLBCL. Wilson et al. looked at 69 patients with previously untreated DLBCL who received DA-R-EPOCH, and compared outcomes between different prognostic groups according to international prognostic index (IPI) and to germinal centre (GC) vs non-GC phenotype.10 High-risk by IPI had 54% event-free survival (EFS) at 62 months compared with 92% and 87% in the high-intermediate and low-intermediate groups, respectively. EFS was 94% vs 58% for GC vs non-GC phenotype, respectively.
Outcomes in GCB type DLBCL over two trials show 100% EFS at 5 years in HIV-negative GCB DLBCL,10 and 95% EFS at 5 years in HIV-positive GCB DLBCL. A modification of the DA-EPOCH regime substituting high-dose dexamethasone for prednisolone, and administering cycles at 14-day intervals, also achieved encouraging PFS and OS rates, especially in high-risk disease.11
High-grade B-cell lymphoma (including DLBCL) with MYC and BCL-2 or BCL-6 translocations (WHO 2016)
Garcia-Suarez et al.8 studied 33 patients with poor prognosis previously untreated DLBCL who received DA-R-EPOCH. All patients had at least 2 adverse prognostic factors on the IPI. 24% had bulky disease. Overall response rate (ORR) was 97%, with complete response (CR) 81% and partial response (PR) 16%. Long-term follow-up of this study demonstrated 10-year EFS and OS rates of 47.8% and 63.6%, respectively.9 Patients with BCL6 rearrangement had an OS of 100%, with the authors postulating that the addition of etoposide may downregulate BCL6 expression.
The utility of “aggressive” chemoimmunotherapy regimens in the treatment of double- and triple-hit lymphomas (DHL/THLs) is broadly supported in view of the poor outcomes of such patients with R-CHOP alone.18 There were insufficient numbers of DHL/THLs for post hoc subset analysis in the CALGB study to assess superiority of DA-R-EPOCH over R-CHOP for unselected DLBCL,1 however, there is some data to suggest superior response rates (improved PFS) of “intensive” immunochemotherapy regimens over R-CHOP.14 These retrospective, pooled data appears to support a more intensive regimen (DA-R-EPOCH, hyper CVAD/MA, R-CODOX-M/IVAC), however, utility of one regimen over another is not suggested, and difference in results may be attributed to other issues studied e.g. patient population. Consolidation with autologous stem cell transplant (ASCT) appears not to improve survival in the de novo setting.
A phase II, multicentre prospective study by Dunleavy et al. examined DA-R-EPOCH in patients with MYC rearranged DLBCL.20 Of the 53 patients treated, 19 patients had sole rearrangement of MYC, and 24 had additional rearrangement of BCL2, BCL6 or both. There were 3 treatment-related deaths. The 48-month EFS and OS were 71% and 76.7% respectively.
Oki et al. published a case series of DHL which included retrospective analysis of 129 patients with DHL confirmed by FISH.15 Similar to the Petrich data,14 there is a suggestion that DA-R-EPOCH is a superior regimen to R-CHOP for DHL, certainly with respect to EFS, (see below) however, this is retrospective data, and relevant biases need to be therefore considered. CNS relapse occurred in 10 - 15%, and CNS prophylaxis could be considered.
Howlett et al. published a systematic review and meta-analysis of the use of R-CHOP, DA-R-EPOCH and intensive dose chemotherapy protocols such as CODOX-M/R-IVAC, R-hyper CVAD/R-M/C. They included 394 patients from 11 studies in the analysis, of which 180 received R-CHOP, 91 received DA-R-EPOCH, and 123 received dose-intensive protocols. The median PFS for the groups was 12.1, 22.2 and 18.9 months, respectively, with a significant improvement in PFS with DA-R-EPOCH compared to R-CHOP, however, there was no significant improvement in OS.21
Figure 2. (A) Progression-free and (B) overall survival curves estimated from the meta-analysis 21

© British Journal of Haematology 2015
There is interest in determining whether the use of dose escalation confers an advantage in double- or triple-expressor lymphomas. An Italian multicentre retrospective series of 122 sequential patients with double-expressor lymphoma (DEL) treated with DA-R-EPOCH demonstrated 2-year PFS 74% and OS 84%.16 Of these, 19 were double-hit cytogenetics with trend towards poorer survival in this subgroup (66%). TP53 mutation was found in 16 patients with statistically significant poorer PFS (58%) and OS (62%). A University of California study compared 122 sequential DELs receiving DA-R-EPOCH (61%) or R-CHOP (39%), specifically excluding double-hit. There was no difference between R-CHOP and DA-R-EPOCH for 3-year PFS (33% vs 57%) and OS (72% vs 72%), however it appeared that the poorer PFS in R-CHOP did not translate to reduced OS due to salvage chemotherapy and ASCT.17 Other smaller series similarly have not demonstrated clear advantage in DA-R-EPOCH over R-CHOP in DELs. Thus recommendations remain to consider DA-R-EPOCH in cases demonstrating MYC, BCL2 and/or BCL6 rearrangements only.
CNS prophylaxis
CNS disease poses a particular challenge in patients with DLBCL, including patients with CNS involvement at the time of presentation, as well as the risk of CNS relapse. DA-R-EPOCH does not demonstrate good CNS penetration so additional measures could be considered for patients with CNS involvement, or deemed to be at high risk of CNS involvement. Several strategies have been explored, including use of intrathecal methotrexate incorporated into the DA-R-EPOCH protocol, and use of high-dose methotrexate given in between cycles.22, 23, 24
In the prospective trial of DA-R-EPOCH in DLBCL with MYC rearrangements, IT methotrexate 12 mg was delivered on days 1 and 5 of cycles three to six as CNS prophylaxis for a total of 8 doses.20 There are only small studies evaluating intravenous methotrexate in combination with DA-R-EPOCH. A prospective trial of 47 patients with poor prognostic CD5+ DLBCL delivered four cycles of DA-R-EPOCH followed by two cycles of IV methotrexate at a dose of 3.5 g/m2 on a 14-day cycle, then completed a further four cycles of DA-R-EPOCH.25 Patients did not receive IT methotrexate. The 2-year CNS relapse rate was 9%. A common practice is to administer 2 doses of IV methotrexate 3.5 g/m2 two weeks apart following completion of the DA-R-EPOCH regimen, however, there are no published studies evaluating efficacy and toxicity with this approach.