0 citations0 references

Commentary on Degenhardt et al (2015): A new formulation for research

Addiction2015Vol. 110(2), pp. 238–239

Citations Over TimeTop 24% of 2015 papers

Abstract

Concerns over the abuse of codeine and morphine have presented themselves historically as market opportunities for the pharmaceutical industry to develop medicines with less risk and more nociception (and cough suppression), leading to marketing heroin. As the initial belief of heroin's lower relative addiction potential was fading, in 1929 the Committee (later College) on the Problems of Drug Dependence was chartered for ‘the replacement of all present use of addicting alkaloids by substitutes having no addiction properties’ 1. The ensuing capitalist competition yielded dozens of synthetics, each settling in different medical niches, and expanding to other controlled substances. Two decades ago, an ‘abuse-deterrent’ reformulation of temazepam in the United Kingdom 2 and Australia 3 was introduced to prevent intravenous injection of original gel capsules. Heating the waxy excipient led to gruesome complications of intra-arterial injection when cooled, resulting in partial withdrawal of the abuse deterrent formulation. Later, the promise of a new approach to abuse-deterrence devolved into disappointment: ‘the OxyContin label originally approved by FDA indicated that the controlled-release characteristics of OxyContin were believed to reduce its potential for abuse’ 4. Exuberance curtailed, claim discredited, the manufacturer(s) of OxyContin spent a decade engineering versions of extended-release oxycodone less prone to injection, resulting in an oxycodone–naloxone combination product in Europe 5 and reformulation of the original in Australia, North America and elsewhere; the latter is the subject of the National Opioid Medications Abuse Deterrence (NOMAD) study 6. In order to evaluate claims of abuse deterrence systematically, pre-authorization abuse liability studies solicit data from ‘experienced’ drug users in laboratory settings to determine how much less they like novel hard-to-abuse formulations relative to morphine or hydromorphone standards 7, 8. These elegant studies aim to isolate inherent properties of the medicine, eliminating bias through blinding and randomization. In this way, subjective impressions of drug liking coexist alongside objective pharmacological fact in drug labels. (Presumably, clear guidance on claims would incentivize manufacturers by allowing abuse deterrence claims for promotional advantage or prolonging exclusivity by differentiating from unprotected generics). However, how do we translate with fidelity the findings of pre-marketing abuse liability trials to observational epidemiological studies? Quantitative studies compare rates of outcome (e.g. abuse, overdose) for a reformulation versus traditionally unprotected opioids, or historical comparators 9-17. Observational studies are conducted with the complexity that individuals who have never encountered the reformulation are not choosing between it and another drug; some would deem lack of availability as success itself. If the street availability of reformulated OxyContin is nearly zero, can we ascribe lower relative likeability based on an absolute lower rate of endorsement? The danger is that if few people endorse a reformulation, the natural reaction may be to attribute lower abuse risk as a fundamental property of the product, instead of an interaction between social availability and physical formulation. If heroin suddenly disappeared, would the resulting decline in injection be a fundamental property of the drug or the social causes that led to its disappearance? Future cohort studies could identify drug users endorsing a spectrum of opioids, with equal opportunity to obtain each (including reformulation) to ensure uniform epidemiological exposure. The mosaic approach in NOMAD sets the benchmark for regulatory science: ‘the more any quantitative social indicator is used for social decision making, the more subject it will be to corruption pressures and the more apt it will be to distort and to corrupt the social processes it is intended to monitor’ 18. Degenhardt et al. have set about an ambitious agenda to synthesize results from various, potentially conflicting, studies. They are scientifically experienced and maintain the credibility to conduct this analysis. Public clinical trial registries brought overdue transparency to the vital endeavour of making better medicines. With the publication of this research protocol, NOMAD sets the standard for post-marketing observational studies. Confidential post-marketing statistical analysis plans, required by regulators and drafted by industry, hardly suffice to defend proprietary interests while severely diminishing credibility. Notably, Australian regulations have no path for label claims of real world abuse deterrence; however, it would be of interest to know if a claim is intended to be made elsewhere. Degenhardt et al. are commended for including needle exchange and other direct service organizations. While the outcomes of this study are abuse/addiction and overdose, bloodborne infection transmission should not be overlooked. The promise of tamper-deterring formulations lies in reducing overall drug use harms. Although seroconversion studies require large cohorts, one recent study found that prescription opioid injectors had higher viral hepatitis prevalence than heroin injectors 19. Finally, drug user representation in contextualizing study design is an ethical necessity, adhering to the ‘nothing about us without us’ principle. Drug user representation has been lacking in regulatory and industry discussions worldwide. The NOMAD study's close connection to drug users is the very earliest shift towards more civil society representation in research. More can be accomplished. None.

Citations Over TimeTop 24% of 2015 papers

Abstract

Related Papers