High Dose Chemotherapy and Autologous Stem Cell Transplantation in the Primary Therapy of Advanced Follicular Lymphoma in Adults: A Systematic Review and Meta-Analysis of Randomized Clinical Trials.

The impact of high dose therapy and autologous stem cell transplantation (ASCT) in the upfront management of adults with advanced follicular lymphoma (FL) on overall survival remains uncertain. A number of randomized clinical trials (RCTs) compared ASCT to conventional dose chemotherapy regimens. We performed a systematic review to address this question.

To compare the overall survival (OS), event free survival (EFS), treatment related mortality (TRM), incidence of secondary myelodysplasia/acute myeloid leukemia (t-MDS/AML) and secondary solid tumors within the ASCT strategy and chemotherapy.

We performed a search of MEDLINE, EMBASE, CENTRAL, American Society of Hematology (ASH), American Society of Clinical Oncology (ASCO), BIOSIS, PAPERSFIRST, PROCEEDINGS, clinical trials registries (National Cancer Institute, clinicaltrials.gov) and bibliographies of relevant studies for randomized controlled trials comparing myelo-ablative chemotherapy with ASCT to any chemotherapy in adults with untreated advanced (Ann Arbor stage III-IV) FL. Two reviewers recorded study characteristics, methodology and outcomes. Disagreement was resolved by discussion.

Seven trials were identified and included in the systematic review but only four with available data reporting 941 patients were included in the meta-analysis. Three trials excluded: one was never reported; the second was stopped early due to the increased toxicities in the transplant arm; and the third showed no significant difference in EFS or OS at an interim analysis with insufficient data for the meta-analysis. None of these studies has been published. In one of the 7 trials patients in both arms received rituximab during the induction treatment. High dose therapy and ASCT did not result in improved OS compared to standard therapy, Hazard Ratio (HR) 1.15 (0.85, 1.55). The HR for EFS was 0.61 (0.34, 1.11) with substantial heterogeneity, I² =91%, Chi² p value <0.00001. Only the duration of the follow up could explain this heterogeneity with statistically significant subgroup interaction, P <0.00001. The incidence of t-MDS/AML ranged from 1% to 7% in the ASCT arms of the four studies, with a difference in favour of standard chemotherapy, Risk Difference (RD) 4% (95%CI: -1%, 8%), and significant heterogeneity (Chi² p value of 0.008, I²=75%) due to difference in duration of follow up (p=0.005 for the subgroup differences).There was no increase in mortality directly attributable to treatment (RD 0%, 95%CI: -2%, 1%) nor in secondary solid tumors (RD 1%, 95%CI: -2%, 4%).

High dose therapy and ASCT as part of initial does not improve overall survival despite a trend towards better disease control, likely because of the effectiveness of second and subsequent lines of treatment for FL. T-MDS/AML occurs more commonly following ASCT and represents a significant risk to patients. With one exception, these studies were conducted before the introduction of monoclonal B cell antibodies. Future trials of ASCT in the context of current chemo-immunotherapy approaches to FL would further inform this issue.

No relevant conflicts of interest to declare.