Interim analysis of a placebo controlled study of dronabinol for adults with sickle cell disease and chronic pain Free

Introduction:

Chronic pain is the greatest contributor to poor quality of life in people living with sickle cell disease (SCD). Many individuals with SCD report using cannabis to treat this pain. We hypothesized that dronabinol, an FDA approved capsule containing synthetic tetrahydrocannabinol (THC) could be effective to treat pain and improve quality of life in people with SCD and chronic pain. We designed a placebo-controlled study with 8 weeks of individualized dosed dronabinol and included a planned interim safety analysis to ensure that dronabinol did not increase rates of severe adverse events (SAEs) or worsen cognition.

Methods:

Subjects enrolled were adults (age>18) with SCD of any genotype and chronic pain, on stable doses of pain and disease modifying medications for at least 3 months, at steady state, with urine toxicology negative for THC in the past 30 days, and without a history of psychosis or substance use disorder.

Subjects were randomized 1:1 to dronabinol or placebo. Doses started at 5mg twice daily (bid) and were titrated daily based on subject preference from minimum of 2.5 mg daily to maximum of 10 mg bid and then continued for 8 weeks. All laboratory and pain and quality of life (QOL) measures (PROMIS: pain impact, neuropathic and nociceptive pain quality, cognition, ASCQ-Me: stiffness, sleep, emotional, and social impact) were taken at baseline and at 4 and 8 weeks on study. Mass spectrometry for minor cannabinoids was measured to assess for abstinence from other cannabinoids during the study period.

Results:

30 subjects have been randomized (15 dronabinol/15 placebo), median age 34.3 SD 6.8 years old, 79% were female, 76% HbSS/HbSβ0, 24% HbSC/Hbβ+, 79% on hydroxyurea, 38% on crizanlizumab, and 41% on voxelotor (when it was available). There were not different between dronabinol and placebo. Six subjects withdrew early but completed evaluations for safety (dronabinol: 3 due to side effects, 1 due to pregnancy, 1 lost to follow up, placebo: 1 due to side effects). There was no significant difference in rates of SAEs between groups (dronabinol: pain events (3 admissions, 6 infusion center, and 3 emergency department (ED) visits), 3 admissions for other reasons unrelated to drug; placebo: pain events (3 admissions, 3 infusion center, and 1 ED visit), 3 admissions for other reasons.) Rates of experiencing at least 1 adverse event (AE) were higher in the dronabinol group vs placebo (10 vs 4, p=0.03). All AEs were grade 1 events and known side effects of dronabinol including fatigue, dizziness, disorientation, and headache. Cognition scores (higher is better) at visit 3 was not different between dronabinol vs placebo (50% (25% - 75%) (63.9 (46.9 - 63.9) vs 53.6 (46.8 - 58.2) p=0.20). Mass spectrometry confirmed no subjects used cannabinoids other than dronabinol during the study.

At 8 weeks, measures of pain (lower is better) were all numerically improved in the dronabinol group but have not yet met statistical significance: pain impact (58.6 (56.1 - 61.6) vs 62.1 (58.3-66.6) p=0.18, nociceptive pain (48.1(42.4-57.6) vs 49.6 (41.7-53) p=0.19, neuropathic pain 44.7 (41.6-51.5) vs 48.1 (42.4-57.6) p=0.26). Measures of QOL (higher is better) were all numerically improved and improvement in social impact was statistically significant: stiffness (49.3 (48-52.2) vs 43.6 (36.6-52.9), p=0.27, sleep (49.7 (41.4-56.3) vs 43 (40.0-49.7) p=0.29, emotional impact (54.2 (52.3-56.9) vs 49.5 (42.6 - 54.7) p=0.055, social impact (53.9 (49.8-56.1) vs 46.8 (44.9-48.4) p=0.0061).

Conclusions:

In an interim analysis dronabinol was not associated with SAEs or worsened cognition, the only AEs were grade 1 and previously known. Personalized dose titration allowed those who were intolerant to side effects to withdraw early. Dronabinol improved social impact of disease. Other measures of QOL and pain were numerically better but have not yet reached statistical significance. Given the encouraging safety interim analysis the study will continue enrollment until it has reached full power.