Liposomal Formulation of Vincristine Allows for Doubling the Dose Compared to Conventional Vincristine: Results of the First Futility Analysis of the OPTIMAL>60 Study of the German High-Grade Non-Hodgkin Lymphoma Study Group (DSHNHL)

Background: Cumulative neurotoxicity is a major issue and dose limiting factor of vincristine in the treatment of diffuse large B-cell-lymphoma (DLBCL) and other malignancies, especially in elderly patients. In the RICOVER-60 trial of the German High Grade Non-Hodgkin Lymphoma Study Group (DSHNHL) [Pfreundschuh et al., Lancet Oncol. 2008] only 59% of patients received full-dose vincristine at cycle 6, 22% of patients received no vincristine at cycle 6 and 36% at cycle 8 due to the neurotoxicity of the drug. Thus, vincristine-induced polyneuropathy is a long-term toxicity that heavily affects the quality of life of successfully treated elderly patients with lymphoma, because of slow and often incomplete recovery. Moreover, retrospective analyses suggest that dose reductions of vincristine lead to a considerable loss of efficacy [Fisher et al., Am J Med. 1977; De VV et al., Cancer Res. 1987]. Since preclinical and clinical studies suggest a high activity and excellent tolerability of liposomale vincristine, a randomized comparison of the two formulations is warranted.

Methods: In the OPTIMAL>60 study elderly patients with DLBCL (61- to 80 years-old, ECOG 0-4, IPI 1 [age>60] with bulk, or IPI 2-5) are randomized to receive eight 2-week applications or an optimized schedule of 12 application of rituximab in combination with 6 cycles of CHOP-14. In a 2x2 factorial design, patients receive a second randomization into conventional vincristine (1.4 mg/m², capped at 2 mg) or liposomal vincristine (2mg/m², uncapped). One secondary endpoint of the OPTIMAL>60 trial of the DSHNHL is the determination of the median dose of either conventional or liposomal vincristine, that can be given to patients applying identical dose reduction rules. Neurotoxicity was determined as polyneuropathy according to the Common Terminology Criteria for Adverse Events (CTCAE v. 4.03).

Results: In this planned futility analysis of the OPTIMAL>60 trial, we investigated the cumulative dose of liposomal (VLIP, Marqibo®) and conventional vincristine (VCR) and the neurotoxic effect of the first 105 patients. Fifty-three patients received VCR and 52 patients VLIP. In comparing the median cumulative dose received for each group. The median cumulative dose of VLIP group (10.09mg/m²) was nearly double the median cumulative dose of VCR group (5.42mg/m²) (p<0.001). Since identical dose reduction rules were applied, we expected equal neurotoxic effects with VLIP and VCR, which was indeed the case. Before initiation of therapy 100% of patients receiving VCR showed PNP grade 0 while 15.2% of patients in the VLIP-group showed PNP grade 1 (84.8% grade 0) which was significantly worse (p=0.018). After cycles 1 and 2 the median grade of polyneuropathy was 0 in both groups, which increased to a median toxicity of grade 1 after cycles 3, 4, 5 and 6. The maximum grades of polyneuropathy observed in both groups were not significantly different (p=0.865). In the VLIP group no liposomal vincristine was administered in 18 patients (34.62%) , in the VCR arm group no vincristine in 14 patients (26.42%) in at least one cycle of therapy (p=0.361). Due to polyneuropathy no VLIP was given in 16 patients (30.77%), no VCR in 12 patients (22.64%) for the minimum of one cycle of therapy (p=0.346).

Conclusions: With nearly twice the dose of VLIP achieved in elderly patients treated in the OPTIMAL>60 study, the preset goals were met and the study will continue as planned. A total recruitment of 864 patients is necessary to demonstrate whether the double dose of VLIP will translate into a significantly improved outcome in 3-year PFS (HR 0.68 or 9% with an alpha=5% and a power of 80%). Acknowledgment: This study was supported by Spectrum, Amgen and Roche.