A retrospective study of R‐DHAP/Ox for early progressing follicular lymphoma

Data from the LymphoCare study (Casulo et al, 2015) demonstrated that patients with follicular lymphoma (FL) progressing within 2 years of diagnosis (early POD), after first-line chemo-immunotherapy with R-CHOP (rituximab, cyclophosphamide, vincristine, adriamycin, prednisone) have a median overall survival (OS) of 5 years from the time of progression. These patients represent around 20% of all treated FL and, for them, optimal management remains uncertain. Second-line chemo-immunotherapy and consolidation with autologous stem cell transplantation (ASCT) has demonstrated significant activity on previously treated FL (Sabloff et al, 2007; Pettengell et al, 2013). While selection of second-line therapy is frequently based on institutional standards, the combination of R-DHAP/Ox (rituximab, dexamethasone, cytarabine, cisplatin/oxaliplatin) (Lignon et al, 2010) is an attractive option because it demonstrated superior outcomes to ifosfamide-based therapy in previously treated diffuse large B cell lymphoma (DLCBL) with germinal centre subtype (Thieblemont et al, 2011). We hypothesized that R-DHAP/Ox would be active in patients with high-risk FL and thus performed a retrospective study. The study was performed in accordance with Institutional Review Board-approved protocols at three Italian institutions (Torino University, San Bortolo Vicenza, Torino Città della Salute). All subjects gave informed consent for the use of the clinical data. Patients with FL were considered eligible if they had received at least one line of prior chemo-immunotherapy, progressed within 2 years from first-line and had received R-DHAP/Ox following disease progression. Grade 3B FL or histological transformations prior to receiving R-DHAP/Ox were excluded. Investigators recorded clinical data from medical records, including number and type of therapies before and after R-DHAP/Ox, response to therapies, main adverse events (infections requiring admission, inadequate stem cell collection, secondary malignancies, transformation to DLBCL), and survival outcomes. Response was based on standard International Working Group Criteria (Cheson et al, 2014). Overall survival was calculated from the time of POD after first-line chemo-immunotherapy. Time to next therapy (TTNT) was calculated from the end of one line of therapy to start of the subsequent one. Survival probability was estimated with the Kaplan–Meier method. Median follow-up time was estimated by reverse Kaplan–Meier method. Analyses were performed using sas version 9.4 (SAS Institute, Cary, NC). A total of 46 patients met the eligibility criteria; their characteristics at diagnosis, first-line therapy and treatments before R-DHAP/Ox are summarized in Table 1. Eleven patients (24%) received rituximab maintenance after first-line therapy. Response to first-line chemo-immunotherapy was achieved in 100% [48% complete response (CR)]; five patients (11%) reached only a stable disease (SD). The median TTNT from first-line was 14·3 months [95% confidence interval (CI) 10·6–19·4]. Response to R-DHAP/Ox +/- ASCT is reported in Table 2. Thirty-four patients (74%) underwent consolidation with ASCT, 12 patients did not because of disease progression in five cases (42%), stem cell collection problems in three cases (25%) and infectious complications during R-DHAP/Ox precluded further high dose treatment in two cases (16·5%). The remaining two patients consolidated R-DHAP/Ox with radio-immunotherapy. ORR: 86·9% (CR 54%) SD: 6·6% PD: 4·4% NA: 2·1% ORR: 91·3% (CR 79·4%) SD: 0 PD: 2·9% NA: 5·8% Febrile neutropenia was reported in eight cases (17%), and chemotherapy was delayed in five cases (11%). In three cases, an 80% dose reduction of the therapy from starting dose was reported, due to age or comorbidities. One patient needed a dose reduction for renal toxicity. Stem cell collection failure occurred in 3/41 cases (7·3%). Four patients (8·6%) experienced transformation to DLBCL; one acute leukaemia and one myelodysplastic syndrome were also reported. Acute toxicities were consistent with previously reported series of ASCT, while rates of secondary malignancies were comparable to other series with less intensive treatments (McLaughlin et al, 2005; Xu et al, 2013; Andrade-Campos et al, 2016). At the time of the analysis, 30 patients (65·2%) were still alive. Causes of death were progression of lymphoma in nine subjects (56·3%), secondary myeloid disorders in two patients (12·5%) and other toxicity in three patients (18·7%: 1 related to ACST and 1 to allogeneic transplant, 1 renal toxicity plus unspecified neurological syndrome). In 1 case (6·2%) the death was unrelated to lymphoma or toxicity, and in another case (6·2%) the cause was not reported. After a median follow-up of 66 months, median OS, OS at 5 years and TTNT after R-DHAP/OX +/- ASCT for the population were as reported in Table 2. Notably, the TTNT after R-DHAP/OX was nearly three times longer than after first-line therapy. Our data suggest that treatment of patients with early POD FL by R-DHAP/Ox and ASCT is active and reasonable in selected patients. Despite bias arising from inclusion of patients that did not receive R-CHOP as front-line therapy and selection of patients eligible to receive intensive therapy, compared to the LymphoCare population, our series had similar characteristics with the sole exception being that the average age in our cohort was slightly less than the LymphoCare patients (Casulo et al, 2015). The OS in this series appears superior to the LymphoCare study (OS at 5 years 74·3% vs. 50%), which is particularly evident for the group undergoing ASCT (OS at 5 years 83·8%). In the LymphoCare study, only 7·3% of patients experiencing early POD were treated with ASCT, making a direct comparison impossible (Casulo et al, 2015). To our knowledge, only one other series reported on outcomes of early POD FL patients treated with a specific therapy. In an unplanned subgroup analysis of a phase II trial of the PI3K inhibitor idelalisib in FL, the ORR was 56·8% (CR 13·5%), median duration of response was 11·8 months, and OS at 5 years from relapse after first-line chemotherapy 79% (Gopal et al, 2017). Although these data are encouraging, the ideal approach to early POD FL is still far from standardized. Our data confirm that patients with FL and early POD have a poor prognosis, but also suggest that R-DHAP/Ox +/- ASCT may be effective therapy in selected patients. Prospective studies of R-DHAP/Ox and ASCT are warranted in high-risk FL. PM and PG designed the study and wrote the paper. FC, CV, UV and MB approved the study design and enrolled patients. AC, MCT, ID and MN enrolled patients and collected clinical data. ZC and PG assembled the data and performed statistical analysis. PM, PG, CV and JPL interpreted the data. MB, FC and JPL provided critical organizational support. All authors approved the final version of the manuscript for submission.