Extensive evidence supports the use of DHAP (dexamethasone, cytarabine, cisplatin) in patients with relapsed or refractory (RR), intermediate and high-grade non-Hodgkin lymphoma (NHL). DHAP is a treatment regimen in the European Society of Medical Oncology diffuse large B-cell lymphoma (DLBCL) Clinical Practice Guidelinesr and the Australian Lymphoma Alliance Position Statement.r Salvage therapy with DHAP can be considered in appropriately selected patients fit for high-dose chemotherapy and autologous stem cell transplantation (ASCT). The evidence also supports the addition of Rituximab (R) to DHAP in patients with CD20+ B cell NHL.
DHAP was published as salvage therapy in 1988 following a single-arm, single-institution study of 90 patients with RR NHL treated between 1984 and 1986. DHAP was administered every 3-4 weeks, for 6-10 cycles (4 cycles beyond maximum tumour effect).r
In the PARMA trial,r DHAP (up to 6 cycles) has been compared with ASCT in chemo-sensitive patients with progressive relapsed NHL. Of 215 patients initially treated, median age 43 years, 109 patients completely or partially responded to 2 initial cycles of DHAP and were randomised to either ASCT with BEAC (carmustine, etoposide, cytarabine, cyclophosphamide) conditioning (n=55, “transplantation”) or up to 4 more cycles of DHAP chemotherapy (n=54, “conventional therapy”). 106 patients failed to respond to initial DHAP or were otherwise excluded from the study.r
The Collaborative Trial in Relapsed Aggressive Lymphoma (CORAL) study was a multicentre international phase III randomised controlled trial (RCT), comparing the efficacy of R-DHAP and RICE (rituximab, ifosfamide, carboplatin, etoposide) in RR CD20+ DLBCL. From 2003 - 2007, 396 patients aged 18 - 65 years in 1st relapse or with refractory lymphoma following 1st line chemotherapy were randomised 1:1 to 3 cycles of R-ICE or 3 cycles of R-DHAP. Responding patients, evaluated after initial chemotherapy, proceeded with BEAM (carmustine, etoposide, cytarabine, melphalan) ASCT. Stem cell mobilisation occurred following the second or third cycle. The ASCT realisation rate was 55%. In both regimens, rituximab (375 mg/m2) was administered before chemotherapy, and in the first course, additional rituximab was given on day −1.r
The NCIC Clinical Trials Group LY.12 study was a multicentre international phase III non-inferiority RCT to assess whether GDP (gemcitabine, dexamethasone, cisplatin) is non-inferior to DHAP in patients 18 years and older (28% >60 years, range 18.7 – 74.3 years), with RR aggressive NHL. From 2003 - 2011, 619 patients were randomly assigned 1:1 to either GDP or DHAP therapy. Patients with B cell NHL (554/619 patients) received rituximab on day 1 of each cycle of chemotherapy. 10.4% of patients were non-B-cell phenotype. The study was assessed by response rate (non-inferiority margin of 10%) and transplantation rate after 2 cycles of therapy. Responding patients proceeded to ASCT. Patients who had not achieved a complete or partial response after two treatment cycles were permitted to receive a third cycle of protocol therapy. ASCT realisation rate was 48.9% (no difference between treatment arms).r
Efficacy
In the 1998 study by Velasquez et al., there was an overall response rate (ORR) of 55% and a complete response (CR) rate of 31%.r
In the PARMA trial, after a median follow-up of 63 months, DHAP followed by ASCT was associated with an improved 5-year event-free survival (EFS) (46% vs 12%, p = 0.001) and overall survival (OS) (53% vs 32%, p=0.038). The trial was stopped early due to demonstrated superiority of ASCT. This study established the superiority of ASCT with BEAC conditioning over ongoing chemotherapy and provided information on outcomes of patients treated with up to 6 cycles of DHAP alone.r
Figure 1. Kaplan-Meier Curves for EFS in the PARMA trial.r
©The New England Journal of Medicine 1995
The CORAL study demonstrated similar outcomes between the 2 treatment arms, with the trial failing to show superior efficacy of one treatment over the other. The ORR for patients receiving RICE was 63.5% (95% CI, 56.8 – 70.7%) vs R-DHAP 62.8% (95% CI, 55.6 – 69.7%). With regard to the 191 patients treated with R-DHAP, the CR/unconfirmed CR (CRu) rate was 40% and partial response (PR) rate 24%. For the entire group, with a median follow-up of 27 months, the 3-year EFS and OS was 31% (95% CI, 26 - 36%) and 49% (95% CI, 43 - 55%), respectively. There was no difference between the RICE and R-DHAP (EFS, 26% v 35%, p = 0.6 and OS, 47% v 51%, p = 0.4, respectively).r
In the NCIC Clinical Trials Group LY.12 study, there was no difference in ORR between treatment arms following 2 cycles – 45.1% (GDP) vs 44.1% (DHAP). In the DHAP arm, CR/CR(u) rate was 14.3% and PR rate 29.8%. At a median follow-up of 53 months, no differences were detected in EFS or OS between GDP and DHAP. Survival estimates were provided for patients completing ASCT: estimated 4-year EFS for DHAP was 48% (95% CI, 39-57%) vs 43% (95% CI, 34-51%) GDP, and 4-year OS for the DHAP arm was 63% (95% CI 54-71%) and 62% (95% CI, 53-69%) in the GDP arm. GDP was found to be non-inferior to DHAP. r
Figure 2. PFS and OS from the LY.12 studyr
©Journal of Clinical Oncology 2014
Toxicity
In the Velasquez et al. study, seven (7.7%) early deaths occurred, with myelosuppression, tumour lysis syndrome and renal toxicity noted.r
The CORAL study identified that grade 3 - 4 haematological toxicities occurred more often in the R-DHAP arm compared with RICE. 57% of patients treated with R-DHAP needed platelet transfusion vs 35% in the RICE arm. Rates of infection with neutropenia were similar (16%). Grade 3 - 4 non-haematological toxicity was more common in DHAP, including grade 3 - 4 renal toxicity in 6% of patients (versus 1% RICE arm).r
In the LY.12 studyr, for those receiving DHAP, 6 patients (2%) died of treatment-related complications. Grade 3 or 4 adverse events were observed significantly less frequently during the first two cycles of chemotherapy amongst patients receiving GDP than DHAP (47% v 61%, P<0.001), including fewer episodes of febrile neutropenia (9% v 23%, P<0.001) and fewer patients requiring hospitalisation for adverse events or other illness (18% v 30%, p<0.001). In those receiving DHAP, 11% of patients had an improved, clinically meaningful, quality of life score (FACT-Total score) while 41% had a worse clinically meaningful change, compared to the GDP arm, which was 18% and 33%, respectively. r
Table 1. Grade 3 - 4 adverse events from the LY.12 studyr
©Journal of Clinical Oncology 2014