Randomized Placebo-Controlled Phase 2 Study of Extended-Release Ketamine Tablets (R-107) for Treatment-Resistant Depression - the BEDROC Study

Over the past two decades, research has established that ketamine can produce rapid-onset antidepressant effects in patients with treatment-resistant depression (TRD), but most trials have used intravenous racemic ketamine or intranasal esketamine. Oral dosing has been much less studied despite pharmacokinetic features of oral ketamine that increase exposure to metabolites such as norketamine and hydronorketamines, which may contribute to antidepressant activity and to a more favourable tolerability profile (for example, less dissociation and smaller blood pressure effects). An extended-release oral formulation (R-107) was developed to exploit prolonged first-pass metabolism and delayed peak concentrations, with absolute bioavailability reported as under 20% and sustained metabolite exposure over 24 hours. Glue and colleagues designed a multicentre Phase II proof-of-concept trial to evaluate the efficacy, dose–response and safety of R-107 in adults with TRD. The study used an open-label enrichment phase with once-daily 120 mg R-107 for 5 days to identify acute responders, followed by randomised, double-blind relapse-prevention treatment with multiple fixed R-107 doses or placebo administered twice weekly for 12 weeks. The authors also included a dose-finding element because the effective oral dose range was uncertain and because enrichment designs can reduce placebo response and study failure rates in antidepressant trials.

This was a multicentre Phase II trial conducted between May 2019 and August 2021. Adult men and women with TRD and baseline Montgomery–Åsberg Depression Rating Scale (MADRS) scores ≥20 entered an open-label enrichment phase during which they received 120 mg R-107 once daily for 5 days and were assessed on day 8. Responders at day 8 (defined in the extracted text as MADRS reduction ≥50% and in some places remission defined as MADRS ≤10) proceeded to randomisation; nonresponders exited the study. Responders (n = 168) were randomised on a 1:1:1:1:1 basis to double-blind R-107 doses of 30 mg, 60 mg, 120 mg or 180 mg, or to placebo, given twice weekly for 12 weeks (double-blind phase). The primary endpoint was the least squares mean change in MADRS score for each active treatment compared with placebo at week 13 (day 92), analysed using a fixed-sequence step-down closed testing procedure starting with the highest dose (180 mg). Key secondary measures included remission and response rates, Clinical Global Impression–Severity (CGI-S), Patient Global Impression–Improvement (PGI-I), time to relapse (Kaplan–Meier and restricted mean survival time), and safety assessments (adverse events, Clinician-Administered Dissociative States Scale (CADSS), bladder pain/interstitial cystitis symptom score (BPIC-SS), vitals, laboratory tests, ECGs, Brief Psychiatric Rating Scale positive symptoms (BPRS+) and Montreal Cognitive Assessment (MoCA)). The extracted text reports screening and enrolment counts (329 screened, 231 entered enrichment, 168 randomised) and indicates that most double-blind dosing occurred at home with high reported adherence (96.4% of participants with ≥80% compliance). The extraction does not clearly state whether analyses followed an intention-to-treat population or provide full statistical modelling details beyond the fixed-sequence testing approach.

From 329 individuals screened, 231 entered the open-label enrichment phase and by day 8 there was a mean MADRS reduction of 18.5 points (95% CI 17.37 to 19.69). In the enrichment phase, 132/231 (57.1%) met remission criteria (MADRS ≤10) and 168/231 (72.7%) met response criteria (≥50% reduction) and were randomised into the double-blind phase. By day 92, 100 participants had discontinued; 94 discontinued for lack of efficacy (MADRS ≥22), with counts by arm reported as placebo 26, 30 mg 22, 60 mg 19, 120 mg 16 and 180 mg 11. Completion rates rose with dose, from 29.7% in the placebo arm to 56.2% in the 180 mg arm. The primary outcome found a statistically significant benefit for the highest dose: the least squares mean difference in MADRS change from baseline at day 92 for the 180 mg group versus placebo was −6.1 (95% CI 1.0 to 11.16; P = 0.019). Numerically greater but non‑significant reductions versus placebo were observed for 120 mg (4.5; 95% CI −0.60 to 9.69; P = 0.083), 60 mg (0.7; 95% CI −4.32 to 5.70; P = 0.785) and 30 mg (1.9; 95% CI −3.08 to 6.92; P = 0.450). Secondary efficacy outcomes showed dose-related trends: remission and response rates at week 13 were numerically higher for active arms versus placebo, with a statistically significant higher response rate for the 120 mg group (48% versus 24.3%; P = 0.046) but no consistent significant differences on the CGI‑S. Patient-reported improvement (PGI‑I) showed mixed odds ratios across doses. Relapse during double-blind treatment exhibited a clear dose response. Relapse rates ranged from about 70.6% for placebo to 42.9% for the 180 mg group (reported elsewhere in the text as 70.3% and 43.7%). Median time to relapse increased with dose (placebo 45 days; 30 mg 28 days; 60 mg 56 days; 120 mg 64 days; 180 mg >85 days). The restricted mean survival time difference for 180 mg versus placebo was 19.0 days (95% CI 4.9 to 33.1), a statistically significant effect. Tolerability was generally favourable. During the open-label phase the most common adverse events were dizziness, headache, dissociation, feeling abnormal, fatigue and nausea; 26 participants (11.6%) reported dissociation but mean CADSS scores were <3. After 5 days of open‑label dosing mean systolic and diastolic blood pressure changes were −1.2 mmHg and −0.1 mmHg, respectively. In the double‑blind phase most adverse events were mild (56.7%) or moderate (18.2%), and mean CADSS scores were <1 at all visits. Sedation of mild severity occurred in five participants (four in 30 mg, one in 120 mg). Ten severe adverse events occurred in eight participants (including severe headache, severe depression, a completed suicide in a 65‑year‑old male in the 180 mg group, noncardiac chest pain, nausea, intervertebral disc protrusion and nephrolithiasis/ureterolithiasis). Five participants experienced serious adverse events: three in the 180 mg group (wound dehiscence, suicidal ideation, completed suicide), one in the 60 mg group (noncardiac chest pain) and one in the placebo group (urinary calculus). The investigators judged none of the SAEs to be treatment related; the completed suicide was attributed by the site investigator to the underlying illness. No notable changes were reported in laboratory tests, urinalyses, vital signs, body weight, ECGs, BPRS+ or MoCA scores. Bladder symptom scores (BPIC‑SS) remained low (<3/38) with no differences between placebo and 180 mg.

Glue and colleagues interpret the findings as evidence that, in an enriched sample of patients with TRD who responded acutely to R-107, twice‑weekly 180 mg extended‑release ketamine tablets produced a statistically significant and clinically meaningful reduction in depressive symptoms at week 13 versus placebo, with a between‑group MADRS difference of 6.1 points that exceeds commonly cited minimum clinically important difference thresholds. The investigators note clear dose‑related effects on relapse rates and median time to relapse and report overall good tolerability: low rates and intensity of dissociation, minimal sedation and no clinically important blood pressure elevations. The authors place their findings in the context of prior ketamine research, arguing that the extended‑release oral formulation’s pharmacokinetics—sustained metabolite exposure due to first‑pass metabolism—may underlie the favourable tolerability profile relative to parenteral or intranasal formulations. They also highlight pragmatic advantages observed in the study, such as predominantly at‑home dosing and brief clinic visits, which could support scalability and lower delivery costs. Several limitations are acknowledged. The enrichment design, which removed nonresponders before randomisation, reduces the risk of study failure but likely overestimates population‑level effectiveness and introduces expectation effects owing to the open‑label enrichment phase. The relatively short duration of the open‑label lead‑in (5 days) may have contributed to higher relapse rates after randomisation compared with studies that include longer open‑label exposure. Small group sizes for each dose may have limited statistical power to detect differences on secondary outcomes. The authors also note constraints on directly comparing these results with non‑oral ketamine studies and indicate uncertainty about differential effects in participants who were versus were not taking concurrent antidepressants, recommending larger future studies to address these questions. In summary, the investigators conclude that R‑107 extended‑release ketamine tablets were effective, safe and well tolerated in this enriched TRD population, and they suggest potential advantages of the oral extended‑release formulation including reduced dissociation, lower apparent abuse risk related to tablet hardness, fewer cardiovascular and sedative effects, and greater convenience for community administration. They stress the need for further unenriched trials and larger studies to confirm generalisability and longer‑term safety and efficacy.