Regulatory Decision Summary for Hydromorphone HP 10, Hydromorphone HP 20, Hydromorphone HP 50, Hydromorphone HP Forte

In this Supplemental New Drug Submission (SNDS) the safety and efficacy of hydromorphone for supervised injectable opioid agonist therapy (siOAT) in opioid-injecting individuals with severe, treatment-resistant Opioid Use Disorder (OUD) was based on 2 Phase 3 clinical trials, the NAOMI and SALOME studies. The SNDS consisted of published reports of the two studies, as well as two systematic literature reviews and corresponding clinical studies which provided further evidence of the efficacy of diacetylmorphine, the comparator for hydromorphone in SALOME. The submission was reviewed in accordance with the Health Canada Guidance Drug Submissions Relying on Third-Party Data (Literature and Market Experience).

The NAOMI trial was conducted in Vancouver and Montreal, and examined whether supervised injection of diacetylmorphine (DAM) was superior to further treatment with oral methadone, in opioid injecting patients with severe opioid dependence (DSM-IV) who had previously failed oral opioid agonist therapy (OAT). Patients in both the NAOMI and SALOME studies were individuals with advanced, treatment-resistant opioid dependence. Patients were predominantly male (mean 61-69%), had an average age of 40-44.3 years, had been using intravenous drugs for a mean of 15.4-16.5 years, and were using heroin a mean of 25-27 days per month. In the NAOMI study, patients had a mean of 11 previous attempts at drug treatment, including 3.2 courses of methadone maintenance therapy, and in the SALOME study patients had a mean of 2.8 attempts at methadone maintenance therapy in the 5 years prior to study enrolment.

The primary efficacy outcomes in NAOMI at 12 months were retention in addiction treatment or drug-free status, and reduction in illicit-drug use or other illegal activity. Although NAOMI was an open-label trial comparing methadone and DAM, it included a substudy of patients who received injectable hydromorphone, and these patients and those receiving DAM were treated in a double-blind fashion.

Overall, 111 patients received methadone, 115 received injectable DAM, and 25 received injectable hydromorphone, and patients receiving injectable drugs could receive supplementary methadone if needed. At 12 months there were statistically significant reductions in rates of illicit-drug use or other illegal activity in the diacetylmorphine group compared to the methadone group, as well as an increased rate of retention in addiction treatment.

In addition to demonstrating greater efficacy of diacetylmorphine compared to methadone, NAOMI revealed that subjects were unable to distinguish hydromorphone from diacetylmorphine. Patient outcomes were also comparable between the two drugs, and 0 of 46 urine samples from 23 patients taking hydromorphone tested positive for 6-monoacetylmorphine or morphine; the study was not sufficiently powered to formally establish non-inferiority of hydromorphone compared to DAM, however.

The SALOME trial further explored the efficacy of hydromorphone for siOAT by examining whether injectable hydromorphone was non-inferior to injectable DAM in reducing illicit heroin use in chronic injection opioid users. SALOME was a double-blind non-inferiority trial conducted in Vancouver by the NAOMI group of researchers. The primary efficacy outcomes in SALOME, assessed at 6 months, were self-reported days of street heroin use in the previous 30 days, days of use of any street-acquired opioids in the prior 30 days, and the proportion of urinalyses positive for street heroin markers at 6 months.

A total of 100 patients received injectable hydromorphone and 102 patients received injectable diacetylmorphine. Medication was self-injected intravenously or intramuscularly, under supervision, up to three times per day, and subjects were allowed to supplement their study drug with oral methadone; roughly 78% of subjects received supplementary methadone at least once over the study period. Following dose titration, patients in the hydromorphone group received a mean of 261.2 mg of drug daily, compared to a daily mean of 506.4 mg of DAM received in the DAM group.

For the difference in adjusted self-reported days of use of any street-acquired opioids in the prior month, hydromorphone was non-inferior to DAM in both per protocol (PP) and intent to treat (ITT) analyses.

Hydromorphone was also noninferior to DAM for the difference in proportion of urinalyses positive for street heroin markers at 6 months, for both PP and ITT analyses.

For the difference in adjusted self-reported days of use of street heroin in the prior month, hydromorphone was non-inferior to DAM in the Per Protocol (PP) analysis. For the ITT analysis, the prespecified non-inferiority margin (NI) (4 days) was not excluded by the lower 90% confidence interval (CI) (4.14 days). Health Canada considered that the choice of a 4 day NI margin was based on a consensus opinion of investigators from jurisdictions with access to DAM treatment. In the absence of market-authorized DAM in Canada, and in the current environment of potentially lethal street opioids, the results of this study, including for the ITT analysis of the street heroin endpoint, were considered consistent with substantial evidence of efficacy of hydromorphone for OAT.

With respect to safety, in the SALOME study there were 29 serious adverse events (SAEs) considered possibly related to injected study medication, including 5 SAEs in the hydromorphone group and 24 SAEs in the DAM group. Seizures and overdoses accounted for 25 of these 29 SAEs. All 11 seizures occurred in patients taking diacetylmorphine, and 11 overdoses were reported in DAM-treated subjects, compared to 3 overdoses in the hydromorphone-treated subjects.

The safety profile of injectable hydromorphone was judged acceptable under the proposed conditions of use, in the context of a potentially fatal condition for which alternate therapies have been unsuccessful. The risks of injectable hydromorphone, including overdose and severe respiratory depression, were considered adequately mitigated in the setting of close monitoring by appropriately trained health care professionals, and outweighed by the substantial risks of continued unsupervised injection of illicit drugs.

Limited safety data are available from ongoing, off-label use of injectable hydromorphone as siOAT in Canada, including an open-label observational cohort study and spontaneous pharmacovigilance. No new safety signals have been identified to date.

In light of the above, the benefit-harm-uncertainty assessment of Hydromorphone HP 10, Hydromorphone HP 20, Hydromorphone HP 50 and Hydromorphone HP Forte is considered favourable, for an indication for supervised injectable opioid agonist therapy in adult patients with severe opioid use disorder who are using injectable opioids and have failed previous attempts at opioid agonist therapy.