PDF: COBRIT
(2009)Objective
To determine whether 90 days of citicoline (CDP-choline) 2000 mg/day, started within 24 hours of injury, improves global functional and cognitive outcomes in patients with complicated mild, moderate, or severe TBI.
Study Summary
• Primary endpoint: global composite odds ratio across 9 dichotomized neurocognitive/functional measures (GOS-E, CVLT-II, Trails A/B, Stroop 1/2, COWAT, PSI, Digit Span) at 90 days
• This publication reports only design and methods; efficacy and safety results are not presented here (as of July 31, 2009, 800 patients enrolled; enrollment 77% of expected to date)
Intervention
Citicoline 1000 mg twice daily (2000 mg/day total) orally or enterally for 90 days vs matched placebo, first dose within 24 hours of TBI
Inclusion Criteria
Age 18–70 (19 in Alabama), non-penetrating TBI, enrolled within 24 hours of injury, English-speaking. GCS off paralytics: 3–12 with motor score ≤5 (no CT criteria required); OR 3–12 with motor=6 meeting specified CT criteria; OR 13–15 meeting the CT criteria. On paralytics: GCS 3TP meeting the CT criteria.
Study Design
Arms: Citicoline 2000 mg/day vs Placebo
Patients per Arm: Per-arm counts not reported in this design paper (1292 planned total; 800 enrolled as of July 31, 2009)
Outcome
• Secondary: rate of recovery at 30/90/180 days; disability, life satisfaction, and psychological well-being at 30/90/180 days; treatment differences at 90 days vs 180 days (3 months after treatment discontinued); survival; safety profile — not reported in this paper
• Efficacy results not reported in this design/methods publication
Bottom Line
This is the design and methods paper for COBRIT; efficacy and safety results are not reported here. The trial is powered at 85% (two-sided α=0.05) to detect a global odds ratio of 1.40 across 9 dichotomized neurocognitive and functional outcome measures at 90 days, with 1292 planned enrollees (accounting for ~15% anticipated loss to follow-up). As of July 31, 2009, 800 participants had been enrolled.
Major Points
- COBRIT is a phase III, randomized, double-blind, placebo-controlled, multicenter trial testing citicoline 1000 mg twice daily (2000 mg/day) vs matched placebo for 90 days in TBI patients enrolled within 24 hours of injury across 8 US clinical centers plus one data coordinating center (TBI-CT Network).
- Target enrollment: 1292 eligible patients over approximately 32 months (estimated enrollment completion August 2010). As of July 31, 2009, 800 patients had been enrolled, representing 77% of expected enrollment to date and 13% of all patients screened.
- Among screened patients at time of interim report: 6% were unable to consent and lacked a legally authorized representative, and 6% refused participation.
- Enrolled severity distribution at time of interim report: severe (GCS 3–8 with motor ≤5) 29%, moderate (GCS 9–12 with motor ≤5) 4%, complicated mild (GCS 13–15 or motor=6) 67%. Planned distribution was 38% severe, 16% moderate, 46% complicated mild.
- Primary outcome is a global test of 9 dichotomized functional/cognitive measures at 90 days analyzed by global odds ratio using stratified logistic regression. This approach avoids multiple-comparison penalties while capturing TBI's multidimensional impact.
- Citicoline (CDP-choline) is a naturally occurring intermediate in phosphatidylcholine biosynthesis with dual action: acute neuroprotection (attenuates arachidonic acid/cardiolipin/sphingomyelin release, increases glutathione, reduces reactive oxygen species) and subacute neurorecovery (increases dopamine and acetylcholine levels, enhances cerebral blood flow).
- Dose rationale: 2000 mg/day was chosen because prior U.S. trials employed 2000 mg as the maximum dosage. Doses as high as 4000 mg/day have been safely used but side effects increase at that level. Animal data used up to 500 mg/kg intraperitoneally (which would grossly exceed recommended human levels).
- Treatment window: first dose within 24 hours of injury. Rationale: animal and human data suggest a 6–48 hour window; 24-hour window chosen to maximize both neuroprotective and neurorepair opportunities.
- Power analysis: 1124 participants needed for 85% power (two-sided α=0.05) to detect a global OR of 1.40 (corresponding to an absolute improvement in GOS-E good-outcome rate of 7.7%). With ~15% anticipated loss to follow-up, target enrollment of 1292.
- Stratification at randomization by: (1) accrual center and (2) TBI severity in 3 strata: GCS 3–5; GCS 6–8 with motor ≤5; GCS ≥9 or GCS 6–8 with motor=6.
Study Design
- Study Type
- Randomized, double-blind, placebo-controlled, multicenter phase III trial
- Randomization
- Yes
- Blinding
- Double-blind
- Sample Size
- 1292
- Follow-up
- 180 days
- Centers
- 8
- Countries
- United States
Primary Outcome
Definition: Global test of 9 dichotomized functional/cognitive measures at 90 days (global odds ratio via stratified logistic regression)
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| Not reported (design paper only) | Not reported (design paper only) | - |
Limitations & Criticisms
- This publication reports only design and methods — no efficacy or safety results are presented.
- The enrolled severity distribution at interim report (67% complicated mild, 4% moderate, 29% severe) diverged substantially from the planned distribution (46% complicated mild, 16% moderate, 38% severe), potentially reducing power to detect effects in the more severely injured subgroups.
- Use of a global composite endpoint across 9 heterogeneous cognitive/functional measures may mask differential effects on individual domains — if citicoline affects only one measure, the global test will not detect efficacy.
- The assumption that treatment effect is constant across all 9 battery measures (required by the global logit model) may not hold for a drug targeting specific neurotransmitter systems (cholinergic, dopaminergic).
- Restriction to English-speaking patients limits generalizability to non-English-speaking TBI populations, who may constitute a significant fraction of urban TBI admissions.
- 24-hour enrollment window is broader than some prior trials (8–12 hours) — earlier administration might be necessary for maximal neuroprotection, and the broad window may dilute effect.
- Loss to follow-up of 15% was anticipated but not yet observed — actual loss may differ and affect power.
- No a priori plan to test pharmacogenomic interactions (e.g., APOE-4 status), though DNA banking for future secondary analyses was included.
- Prior positive trials of citicoline in TBI were small, used lower doses, and showed benefit only on secondary endpoints — the evidentiary basis for expecting efficacy at 2000 mg/day over 90 days in this broad population was modest.
Citation
Zafonte R, et al. J Neurotrauma. 2009;26(12):2207–2216. doi:10.1089/neu.2009.1015