Clinical Trial Design Features of Myelofibrosis Trials during the Last Decade: Comprehensive Review of Clinicaltrials.Gov Data 2010-2019

Introduction: FDA approval of JAK2-inhibitors for myelofibrosis (MF) has changed the landscape for treatment introducing possible increased complexity of clinical trial designs in MF. Many patients with MF will have insufficient response or intolerance to ruxolitinib, opening an opportunity for testing of novel agents alone or in combination. Novel trial designs exist for early stage (phase 1 and 2) studies. The Myeloproliferative Neoplasms Research Consortium (MPN-RC) is an NCI-funded academic group of laboratory and clinical scientists working to develop and evaluate treatments that will improve the survival of patients with MPNs focused on early stage trials in MF. This review was undertaken to understand the landscape of MF clinical trials and improve our approach to trial design.

Methods: Studies with “myelofibrosis” as a condition were abstracted from clinicaltrials.gov using an application programming interface from 2010-2019 and reviewed for phase, trial design features, and endpoints. Trial design review captured such features as presence of dose escalation and approach (rule-based versus model-based/assisted) in the phase 1 setting. Observational and behavioral based studies were excluded. A total of 165 treatment interventional studies including MF patients (either MF as the primary cohort, or MF along with other MPN, myelodysplastic syndrome, or hematologic malignancies related cohorts) were reviewed. Primary and secondary endpoints were reviewed for quality of life, symptoms, or other patient-reported outcomes (PRO).

Results: A median of 16.5 (range 13-20) trials were initiated per year; 92 (56%) phase 2, 54 (33%) phase 1; 17 (10%) phase 3 and 2 (1%) phase 4, with 77 (47%) industry sponsored (Figure 1). Among phase 1 trials, 41/54 (76%) included dose escalation, the majority employing a standard 3+3 design (or other rule-based design) or not specifically detailing the dose escalation schema. Less than 5 dose escalation trials were explicitly described as model-based or model-assisted. Almost half of the phase 1 trials (23/54; 43%) included combination treatment, with 17/23 (74%) using ruxolitinib or fedratinib. Among phase 2 trials, 82 (89%) were non-randomized and 10 (11%) randomized trials; 35/92 (38%) involved combination treatment and 57 (62%) single agent. Median actual/target enrollment for the phase 3 trials (n=16) was 192 patients (range 49-2233 patients); only 2 phase 3 trials were > 500 patients.

Median number of total outcomes was 6.0 (range 1-28) across all studies with a median of 5.5 (range 1-28) for 2010-2014 versus 7.0 (range 1-22) for 2015-2019 (p=0.09). Overall 63/165 (38%) studies included at least one PRO endpoint; 15 (28%) phase 1, 33 (36%) phase 2 and 14 (82%) phase 3. Inclusion of PRO endpoints for years 2010-2014 was similar to studies started in years 2015-2019 [31/82 (38%) and 32/83 (39%)]. Industry sponsored studies included PRO endpoints more frequently [46/77 (60%) versus 17/88 (19%); p<0.001] than non-industry sponsored trials but this is primarily due to a higher proportion of phase 3/4 trials (n=17) being conducted by industry versus non-industry (n=2) (Figure 2).

Conclusions: For phase 1 trials in MF, few dose-escalation trials included model-based or model-assisted methods. These designs are encouraged in order to maximize therapeutic benefits in MF, a rare cancer. Clinical trial designs in myelofibrosis have shifted in recent years as evidenced by an increase in number of endpoints, likely to accommodate new challenges in the post JAK2-inhibitors approval era. Use of patient-reported outcomes including quality of life and symptom assessment was high in the phase 3 setting, and continued use should be encouraged as endpoint assessments in MF clinical trial designs.