CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`
`
`
`APPLICATION NUMBER:
`022272Orig1s014
`
`
`OFFICE DIRECTOR MEMO
`
`
`
`
`
`
`

`

`
`
`DEPARTMENT OF HEALTH & HUMAN SERVICES
`
`
`
`
`
`Public Health Service
`Food and Drug Administration
`Rockville, MD 20857
`
`
`
`
`
`To:
`
`
`
`From:
`
`
`
`
`
`Subject:
`
`Janet Woodcock, MD
`Director, Center for Drug Evaluation and Research (CDER)
`Food and Drug Administration (FDA)
`
`Douglas C. Throckmorton, MD
`Deputy Director for Regulatory Programs
`CDER, FDA
`
`Abuse-Deterrent Properties of Purdue’s Reformulated OxyContin (oxycodone
`hydrochloride) Extended-Release Tablets
`
`
`
`This memorandum summarizes the complex and technical multidisciplinary review of scientific data
`regarding Purdue’s reformulated OxyContin (oxycodone hydrochloride) extended release tablets1
`(OCR) (NDA 22-272) and its potential abuse deterrent properties for the purposes of regulatory
`decision-making. This matter has been the subject of extensive consideration by Agency experts over
`the course of many months. Given the nature of this matter, an integrated assessment of the scientific
`data and regulatory issues is particularly important.
`
`Accordingly, in my capacity as Deputy Director for Regulatory Programs for the Center for Drug
`Evaluation and Research (CDER or the Center) -- and given my extensive involvement in scientific
`and policy decisions on issues related to drugs with abuse potential within CDER -- you have asked me
`to provide recommendations on the following two issues:
`
`
`1. Whether the labeling for Purdue Pharma LP’s (Purdue’s) reformulated OxyContin
`(oxycodone hydrochloride) extended release tablets (OCR) should be revised to include
`language describing abuse-deterrent properties of the new formulation along with relevant
`caveats (the “labeling issue”); and
`
`
`
`2. Whether Purdue’s original formulation of OxyContin (oxycodone hydrochloride) extended
`release tablets (OC) should be determined to be withdrawn for reasons of safety or
`effectiveness (the “relisting issue”).
`
`
`These two issues should be considered in light of the respective standards in the Federal Food, Drug,
`and Cosmetic Act (the Act) and implementing regulations. FDA has long considered abuse and
`dependence in different contexts of regulatory decision-making (e.g., product labeling, drug approvals,
`adverse events).
`
`
`
`1 The approved labeling refers to “OxyContin (oxycodone hydrochloride controlled-release) Tablets.”
`
`
`
`1
`
`Reference ID: 3294145
`
`

`

`Fundamental to consideration of both issues here is an overall assessment of potential abuse-deterrent
`properties of OCR relative to OC. Accordingly, this memo focuses on my assessment of the data
`regarding those properties. The materials used in preparing this assessment are listed at the end of the
`document, and consist of extensive scientific reviews of data generated by multiple components of
`CDER, including the Controlled Substances Staff (CSS), the Division of Anesthesia, Analgesia, and
`Addiction Products (DAAAP), and the Office of Surveillance and Epidemiology (OSE).
`
`The memorandum concludes by explaining my recommendations on both the labeling issue and the
`relisting issue. Those recommendations are:
`
`
`1. That the labeling for OCR should be revised to include language describing the abuse-
`deterrent properties of
`the new formulation along with relevant caveats; and
`
`2. That OC should be determined to be withdrawn for reasons of safety or effectiveness.
`
`
`
`
`
`I.
`
` Background
`
`A. Opioids Generally
`
`
`Prescription opioid analgesics are an important component of modern pain management. Abuse and
`misuse of these products, however, have created a serious and growing public health problem. FDA
`has worked to address this problem while ensuring that patients in pain have appropriate access to
`opioid analgesics. FDA has approved labeling and a risk evaluation and mitigation strategy (REMS) to
`address the problem of abuse and misuse.
`
`Another important step towards the goal of creating safer opioid analgesics has been the development
`of opioids that are formulated to deter abuse and misuse. FDA considers the development of these
`products a high public health priority. Because opioid analgesics must be able to deliver the opioid to
`patients for the management of pain, the extent to which an abuse-deterrent product is able to reduce
`misuse and abuse will not be absolute. Therefore, the extent of abuse deterrence can only be
`understood when studied relative to a comparator.
`
`
`B. Purdue’s OC and OCR
`
`FDA originally approved Purdue’s new drug application for OxyContin extended release tablets (OC)
`(NDA 20-553) on December 12, 1995. The labeling stated that the product should only be taken
`orally, and warned that taking crushed, chewed, or broken tablets could lead to the rapid release and
`absorption of a potentially toxic dose of oxycodone. The product was not formulated with properties
`to deter abuse, and approved labeling did not include language on abuse-deterrent properties.
`
`Purdue subsequently submitted and received approval of another new drug application for
`reformulated OxyContin extended release tablets (OCR) (NDA 22-272) on April 5, 2010, with the goal
`of making it more difficult to misuse and abuse by changing the physical and chemical properties of
`the formulation. The NDA included studies assessing these product attributes. The approved labeling
`did not include language on abuse-deterrent properties. In addition, as a part of the approval of OCR
`in 2010, FDA imposed post-marketing requirements to assess the impact of these changes on abuse
`and misuse patterns in the real world.
`
`Shortly after approval of OCR, Purdue notified FDA by letter dated August 10, 2010, that it had ceased
`shipment of OC, and FDA subsequently moved OC to the “Discontinued Drug Product List” section of
`2
`
`
`
`Reference ID: 3294145
`
`

`

`the “Approved Drug Products With Therapeutic Equivalence Evaluations” (commonly known as the
`Orange Book). Purdue also asked FDA by letter dated March 19, 2013, to withdraw approval of OC
`for reasons of safety. There are several pending abbreviated new drug applications (ANDAs) that cite
`OC as the reference listed drug and propose to duplicate OC. In addition, several citizen petitions have
`been submitted requesting that FDA determine whether OC (10 mg, 15 mg, 20 mg, 30 mg, 40 mg, 60
`mg, 80 mg, and 160 mg strengths) was voluntarily withdrawn from sale for reasons other than safety or
`effectiveness.
`
`Purdue submitted data regarding the abuse-deterrent properties of OCR in a citizen petition dated
`August 28, 2012.2 On September 14, 2012, Purdue submitted Supplement 014 to NDA 22-272 (S-14)
`requesting prior FDA approval of labeling describing the abuse-deterrent properties of OCR. The CP
`and S-014 included data from in vitro, pharmacokinetic (PK), clinical abuse potential and
`epidemiologic studies.
`
`
`II. Findings from Multidisciplinary Review
`
`A. Summary of Multidisciplinary Review
`
`
`FDA evaluated data on the following four categories of abuse-deterrent properties of OCR:3
`o Laboratory-based in vitro manipulation and extraction studies
`o Pharmacokinetics studies
`o Clinical abuse potential studies
`o Investigations analyzing postmarketing data on abuse
`
`
`These data have been the subject of extensive evaluation by multiple components of CDER over the course
`of many months. Key studies discussed in this memo are summarized in the attached chart. Because the
`extent of abuse deterrence for OCR can best be understood when studied relative to a comparator much of
`the data compare OC and OCR. The data from all investigations relevant to the potentially abuse-
`deterrent properties of OCR need to be evaluated together, considering the totality of the evidence, to
`assess whether and the degree to which OCR can be expected to deter abuse relative to OC.4
`
`
`1. Laboratory-based in vitro manipulation and extraction studies5
`
`
`The goal of these studies was to assess the effects of the new formulation on a variety of in vitro
`measures related to the manipulation of the formulation for the purposes of abuse and misuse. In other
`words, the goal is to evaluate the ease with which the potentially abuse-deterrent properties of the
`formulation can be defeated or compromised. Because the original formulation’s extended-release
`properties were easy to defeat with simple methods such as chewing, one goal was to have the product
`retain a degree of the extended-release properties after dissolution or extraction of crushed tablets
`
`2 The petition asked FDA, among other things, not to approve any generic versions of oxycodone hydrochloride extended-
`release tablets that are not as abuse-deterrent as OCR. FDA issued a non-substantive denial of the petition, which was
`subject to FDCA 505(q), on January 26, 2013. (Docket FDA-2012-P-0939).
`3 See Draft Guidance for Industry: Abuse-Deterrent Opioids — Evaluation and Labeling (January 2013) (hereinafter Draft
`Guidance) for background on categories of studies that may be relevant for evaluating abuse deterrence.
`4 The evaluation of an abuse-deterrent formulation takes into consideration the most common routes of abuse which in this
`case are the oral, nasal, and intravenous routes. As reflected in the approved OCR labeling, the use of opioid analgesic
`products carries the risk of addiction even under appropriate medical use, so it is appropriate to evaluate data on abuse
`deterrent properties regardless of the population.
`5 See e.g., DAAAP and CSS reviews for more information.
`
`
`
`Reference ID: 3294145
`
`3
`
`

`

`(physical manipulation followed by chemical manipulation or dissolution). Polyethylene oxide CEO)
`is the main excipient that imparts these properties. The reformulated OxyContin (OCR) consists of
`oxycodone hydrochloride in a
`(mo matrix of
`‘5’“)
`(m4)
`
`Important measures of the impact of the new formulation on in vitro properties relevant to abuse are
`summarized in table one, appended to this review. Several points need to be made about the testing
`conducted on OCR:
`
`testing strategy was driven by the type of abuse deterrent
`l. The physical chemical
`technology being used. If the product had employed a different strategy (e.g., inclusion of
`an aversive substance to decrease abuse) other testing methods generally would need to be
`applied.
`2. The testing compared CC to OCR to assess relative improvement in the ability to deter
`abuse or misuse. This type of assessment was necessary because the science of abuse
`deterrence is relatively new and the ability of physical and chemical properties needed to
`reduce abuse should be assessed on a case by case basis.
`3. The use of repeated measures and, where possible, multiple approaches to physical and
`chemical manipulation increased the confidence in the validity of the measures.
`
`Overall, the in vitro studies demonstrate that manipulation of OCR tablets is more difficult compared
`with the manipulation of OC tablets. Compared with the OC formulation, the extended-release
`mechanism of OCR tablets requires a higher amount of effort, time, experience and tools to defeat
`making it more difficult to create a fine powder for insufflation. This is important because particle size
`may influence the rate of opioid release from the manipulated product. For extraction for intravenous
`abuse, OCR is particularly challenging as it turns into a viscous gel that is resistant to injection. This
`feature may make abuse via insufflation more difficult also, but whether this is so cannot be measured
`using mechanical testing methods only and should be considered in the context of the other categories
`of testing below. Oxycodone in both the OC and OCR formulations is not appropriate for vaporization
`(e.g., for smoking) as the oxycodone degrades at temperatures close to where vaporization occurs.
`
`To summarize the results, the in vitro data suggests that OCR represents an improvement over OC in
`that it increases the ability of OCR to resist crushing, breaking, and dissolution. The in vitro data also
`demonstrate that OCR has physicochemical properties expected to make abuse by injection difficult.
`The in vitro data provide support, together with other categories of data below, that OCR has
`physicochemical properties that are expected to reduce abuse via the intranasal route.
`
`2. Pharmacokinetic Studies6
`
`The goal of these studies was to assess the effects and better understand the in vivo properties of the
`new formulation on the release of oxycodone from the formulation, both when intact and following
`manipulation. The comparison of the intact products was to assure that the OCR formulation is equally
`bioavailable (when swallowed whole) to the OC formulation. The comparison of the manipulated
`products was to assess the impact of the new formulation on the abusability of the new formulation
`
`6 See e.g.. DAAAP and CSS reviews for more information.
`
`Reference ID: 3294145
`
`

`

`when taken via known oral and non-oral routes of abuse (i.e., injection, insufflation). For both
`comparisons, standard pharmacokinetic measures were assessed (including Tmax, Cmax, and AUC).
`
`As summarized in the tables appended to this memo:
`1. When serum concentrations of whole OC and OCR were compared, the two formulations
`are equally bioavailable.
`2. When serum concentrations of OC and OCR were compared, both formulations are
`susceptible to being defeated when chewed vigorously
`), with
`shorter periods of chewing defeating the OC formulation to a larger degree than the OCR
`formulation. The result is a longer Tmax for the OCR formulation when chewed routinely
`compared with OC, which may predict a lower abuse potential.7 However, given the ability
`of chewing to defeat the extended-release features of both formulations, the impact of OCR
`on oral abuse has yet to be adequately demonstrated.
`3. Serum concentrations following crushing of OCR with mortar and pestle show that OCR
`retains its extended release properties after crushing. In vitro dissolution data following
`crushing of OC and OCR with mortar and pestle show that crushing by manual means (such
`as mortar and pestle) defeats the extended-release properties of OC but not OCR tablets.
`
`
`To summarize the results from this category, the PK and other data demonstrate that (while OCR is as
`bioavailable as OC), OCR is more resistant than OC to some forms of manipulation but not others.
`Notably, however, vigorous chewing is still able to disrupt the controlled-release mechanism of the
`OCR, such that an impact of the formulation change on oral routes of abuse cannot yet be adequately
`assessed.
`
`
`3. Clinical Abuse Potential Studies8
`
`
`The goal of these studies was to compare the attractiveness (“likability”) of manipulated OxyContin
`formulations (OC, OCR) by exposing individuals experienced in the abuse of opioids to these products
`and assessing their responses.9 The focus of these studies was on relevant routes of abuse (particularly
`insufflation).
`
`As summarized in table three appended to this memo:
`
`
`1. Finely-crushed and coarsely-crushed OCR had lower liking scores than OC when
`insufflated. The results were consistent whether evaluated using median liking scores or
`using a responder analysis.
`2. While the full amount of finely- and coarsely-crushed OC could be insufflated by 25 of 27
`subjects, only 17 of the 27 subjects tested given OCR were able to insufflate the full
`amount. OCR and OC caused similar degrees of intranasal irritation.
`
`To summarize the results from this category, the clinical abuse potential studies reinforce data from the
`other categories of data discussed in this document. The data from the clinical studies, along with
`
`7 The Tmax (time to maximum concentration) for the OCR and OC formulations when taken without chewing were 4.5
`hours and 2.5 hours, respectively. A longer Tmax has been suggested to be associated with a reduced risk of abuse, but no
`quantitative link has been established to date.
`8 See e.g., DAAAP and CSS reviews for more information.
`9 Clinical abuse potential studies are generally conducted in a drug-experienced abuse population as a common enrichment
`strategy. These subjects generally are more capable of distinguishing active drug from placebo reproducibly which
`improves the power of the study to distinguish differences between treatments.
`
`
`
`Reference ID: 3294145
`
`5
`
`(b) (4)
`
`

`

`support from the in vitro and other data, indicate that OCR has physicochemical properties that are
`expected to reduce abuse via insufflation. That is, the reformulation resulted in reduced attractiveness
`for insufflation of the manipulated OCR to individuals experienced in the abuse of opioids when
`compared with manipulated OC.
`
`
`4. Postmarketing Studies10
`The sponsor submitted data from 11 postmarketing investigations, three of which met the
`characteristics (as set forth in the Draft Guidance) of formal investigations.11 The data from the 11
`investigations were reviewed by OSE and DAAAP and the results are summarized in the OSE and
`DAAAP reviews and in table four appended to this memo. There were three goals of the investigations
`into the effects of OxyContin reformulation in the real world setting:
`
`1. Compare the rates of abuse, diversion, and outcomes of abuse (e.g., overdose,
`hospitalization, death) for OCR and OC
`
`2. Compare the rates of adverse events and therapeutic errors for OCR and OC
`
`3. Compare the rates of accidental exposures for OCR and OC
`
`
`
`Formal Epidemiologic Studies
`
`
`These studies submitted by the sponsor assessed the impact of the new OxyContin formulation in
`several important areas, including overall and route-specific abuse of OCR, OC, and other extended-
`release oxycodone products. Another study, the Client Treatment Study (CTS) Investigation was
`submitted to the docket by another company and was also reviewed.
`
`
`
`NAVIPPRO™ Study
`
`
`The NAVIPPRO™ study looked at data from standardized self-administered questionnaires completed
`by clients entering substance abuse treatment centers to examine the prevalence of OxyContin abuse in
`the 30 days prior to admission. The results suggested that the replacement of OC with OCR in the
`marketplace has led to reduced abuse, including reduction of abuse of both OxyContin and other forms
`of extended-release oxycodone products. These conclusions are limited by the persistence of reported
`use and abuse of OC long after Purdue had withdrawn OC from marketing.12 The results of the
`NAVIPPRO™ study also differ from those reported from the National Survey on Drug Use in
`Households (NSDUH), which showed no change in non-medical use of OxyContin in the 30 days prior
`to the interview following replacement of OC with OCR in the marketplace.
`
`
`
`10 See e.g., DAAAP and OSE reviews for more information.
`11 The National Addiction Vigilance Intervention and Prevention Program (NAVIPPRO™) Investigation, the Researched
`Abuse Diversion and Addiction-Related Surveillance (RADARS®) System Poison Center Program (SPCP) Investigation,
`and the National Poison Data System (NPDS) Investigation.
`12 Current data do not explain the persistence of use of OC. Explanations that have been offered include the persistence of
`the OC formulation in the supply chain and in patients’ homes, as well as mis-identification by individuals of what form of
`OxyContin they took.
`
`
`
`Reference ID: 3294145
`
`6
`
`

`

`The data from NAVIPPRO™ also suggest an effect of OCR in deterring route-specific abuse. Prior to
`the introduction of OCR, the rate of non-oral abuse of OxyContin was 70% among abusers. Since the
`introduction of OCR, the corresponding rate of non-oral abuse of OCR was 40% among abusers.
`
`
`Client Treatment Study (CTS) Investigation
`
`
`This study collected information about the routes of abuse of OxyContin in similar ways to
`NAVIPPRO, collecting data from standardized self-administered questionnaires completed by clients
`entering substance abuse treatment centers to examine the prevalence of OxyContin abuse. The data
`from NAVIPPRO™ are in conflict with the data from the CTS. Instead of decreases in the prevalence
`of overall OxyContin abuse, the CTS reported increases in the prevalence of overall OxyContin abuse
`in the previous 30 days (via both oral and non-oral routes) from 2.6 to 2.9%.
`
`With regard to route-specific abuse, here again the data from NAVIPPRO do not agree with data from
`the CTS investigation. In the CTS data, the prevalence of OxyContin abuse via non-oral routes
`increased from 44 to 48% following OCR introduction. While there are differences in how the data are
`collected in these two studies and differences in how they were analyzed,13 the reason for these
`different results from similar data sources is not known.
`
`
`RADARS® System Poison Control Program (SPCP) Investigation
`
`
`The RADARS data assessed OxyContin abuse by measuring the numbers of poison control calls
`related to intentional abuse of OxyContin. After the introduction of OCR, there was a 32% decline in
`call numbers14. The decline was precipitous after the market introduction of OCR, which was
`inconsistent with the gradual decline in the numbers of OxyContin prescriptions.
`
`No information about route-specific abuse was obtained.
`
`
`National Poison Control Data System (NPDS) Investigation
`
`
`Like RADARS, the NPDS investigation assessed OxyContin abuse by measuring the numbers of
`poison control calls related to intentional abuse of OxyContin. After the introduction of OCR, there
`was a 30% decline in call numbers and a 19% decline when the calls were adjusted for the number of
`prescriptions dispensed for OxyContin. The decline was precipitous after the market introduction of
`OCR, which was inconsistent with the gradual decline in the numbers of OxyContin prescriptions.
`
`While some of the formal epidemiologic studies suggest a decline in OCR abuse via non-oral routes,
`other studies do not support such a finding. Longer-term follow-up of the ongoing studies, as well as
`additional data from other formal studies would be useful.
`
`
`Additional Post-marketing Investigations
`
`
`The sponsor conducted additional studies to evaluate the impact of OCR on a variety of outcomes that
`are plausibly linked to changes in patterns of abuse and misuse.15 They include information from a
`variety of relevant sources on the following:
`
`13 These differences are discussed in greater detail in the OSE reviews.
`14 Expressed as numbers of calls per unique recipient of an OxyContin prescription.
`15 See reviews from OSE and DAAAP for details.
`
`
`
`Reference ID: 3294145
`
`7
`
`

`

`• Diversion
`• Doctor shopping
`• Street price of the new OxyContin formulation
`• Social use patterns reflected in internet discussions
`• Prescribing of OxyContin by physicians identified as ‘high risk’ based on past
`prescribing patterns.
`• OxyContin abuse by abusers followed longitudinally in a cohort of abusers in Kentucky
`• Adverse events from sponsor’s International Drug Safety Database (ARGUS).
`• Statement of law enforcement personnel involved in enforcement related to prescription
`drug abuse made to Purdue’s Law Enforcement Liaison and Education trainers from
`2010-2012.
`
`Data on Clinical Consequences
`
`
`While some of the informal studies are designed to assess death related to prescription drug abuse,
`mortality data are not yet available. To date, the available data from the informal studies, together with
`the formal epidemiologic studies, are not sufficient to adequately assess the clinical consequences of
`abuse (e.g., overdose, death, or hospitalizations). Longer-term follow-up of the ongoing studies, as
`well as additional data from other formal studies and other postmarketing investigations would be
`useful.
`
`Based on aggregate data from these postmarketing investigations, I believe that while the
`investigations have some limitations, these data support the findings from other data sources that OCR
`can be expected to reduce abuse via the intravenous and intranasal routes, and possibly reduce overall
`abuse.16 Particularly encouraging data from these sources include the following:
`• Declines in prescribing behaviors that have been linked to misuse and abuse: doctor
`shopping, prescriptions paid for with cash, prescriptions for the highest doses of
`OxyContin, and prescriptions by physicians identified as ‘high risk’ based on past
`prescribing patterns.
`• Declines in reported rates of insufflation and injection of OCR, and decreases in the
`average number of days of abuse of OCR via injection or insufflation, compared with
`OC in a cohort of abusers followed longitudinally in Kentucky.
`
`
`
`
`
`B. Conclusions from Scientific Multidisciplinary Review
`
`
`Given the complex, technical, and multidisciplinary nature of this matter, I am including an integrated
`summary assessment of the scientific data below:
`
`
`• Both OC and OCR are equally bioavailable when swallowed whole.
`• The in vitro data, together with the other data show that, relative to OC, there is an increase in
`the ability of OCR to resist cutting, crushing, chewing, breaking, and dissolution using a variety
`of tools and solvents.17
`
`16 DAAAP and OSE express consistent views on the more formal postmarketing investigations. Although DAAAP and
`OSE use different language to characterize the other postmarketing data and have differing views of the robustness of that
`data, there is general agreement with my conclusions regarding the regulatory implications of the postmarketing data.
`17 OxyContin is also misused for therapeutic purposes. For example, a patient or caregiver may crush the product to
`administer it if the patient lacks the ability to swallow an intact tablet. As noted in the boxed warning of the labeling,
`disruption of the tablet and controlled-release mechanism for abuse or misuse “can lead to rapid release and absorption of a
`8
`
`
`
`
`
`Reference ID: 3294145
`
`

`

`
`
`
`
`
`
`
`
`• The in vitro testing of the physical properties of OCR was extensive18 and rigorously
`conducted, and the effects shown19 provide a strong basis for predicting an effect of OCR on
`route-specific abuse. The studies were appropriate for a formulation that is focused on physical
`changes to deter abuse and misuse. The testing of the physical properties was robust in that it
`was also sufficient to reveal how the current formulation could be defeated (e.g., vigorous
`chewing or robust mechanical grinding).
`
`• The in vitro data, together with the pharmacokinetic data, show that while OCR is more
`difficult to crush than OC, vigorous chewing is sufficient to defeat the extended-release
`features of OCR to a similar degree as that seen with OC.20 Accordingly, the impact of the
`formulation change on the oral route of abuse cannot yet be adequately assessed.
`
`•
`
`Intranasal and intravenous opioid abuse are associated with serious adverse events including
`overdose and death and the OCR can be expected to have a positive public health impact.21
`Intravenous opioid abuse is associated with HIV and hepatitis B and C infection risk as well as
`organ damage.22 Intranasal opioid abuse is associated with nasal, palatal, and pharyngeal
`necrosis.23
`
`• With regard to intravenous abuse, the in vitro data demonstrate that OCR has physicochemical
`properties expected to make abuse by injection difficult. When subjected to an aqueous
`environment, OCR gradually forms a viscous hydrogel (i.e., a gelatinous mass) that resists
`passage through a needle. The in vitro testing was sufficient to demonstrate that OCR prevents
`oxycodone from being drawn into a syringe to any meaningful extent. Although there are no
`supporting pharmacokinetic or clinical data available (and it is not ethically feasible to obtain
`such data), the in vitro data, coupled with the post-marketing reports, supports the conclusion
`that OCR has properties that are predicted to reduce abuse by the intravenous route compared
`with OC.
`
`• With regard to intra-nasal abuse, the clinical studies showed OCR resulted in lower liking
`scores than OC. The in vitro data also showed that OCR requires a higher amount of effort,
`time, experience and tools to crush, which could make more difficult the creation of a fine
`powder for intranasal use. There is also supportive pharmacokinetic and postmarketing data.
`
`
`potentially fatal dose of oxycodone.” The increased ability of OCR to resist crushing and breaking may have an impact on
`deterring such use.
`18 For example, the fractionation studies used the full range of particle sizes likely to be achieved for misuse and abuse
`purposes, and the extraction studies included a broad array of “household”, “industrial,” and pH buffered solvents. The
`syringeability testing included multiple needle sizes and extraction times and temperatures.
`19 For example, the maximum amount of oxycodone expelled via injection of crushed OCR (through a 27 gauge needle)
`was
` as compared with a range of
` oxycodone expulsion for OC.
`20 The pharmacokinetic studies, together with in vitro data, suggest that shorter periods of chewing defeated OC to a larger
`degree than the OCR formulation, and it is reasonable to conclude that such an effect could have an impact on misuse.
`21 See Refs. 5-9 (in “Materials Used for This Review”).
`22 See Refs. 5-7. According to Section 9.2 of the approved labeling for both OC and OCR, injection of the OxyContin
`tablet excipients “can result in death, local tissue necrosis, infection, pulmonary granulomas, and increased risk of
`endocarditis and valvular heart injury.”
`23 See Refs. 8-9.
`
`
`
`Reference ID: 3294145
`
`9
`
`(b)
`(4)
`
`(b) (4)
`
`

`

`Based on these data, OCR is predicted to reduce abuse via the intranasal route compared with
`OC.
`
`
`• Abuse of OC and OCR is still possible by the intravenous and intranasal routes, as well as by
`the oral route.
`
`• Oxycodone in both the OC and OCR formulations is not appropriate for vaporization (e.g., for
`smoking) as the oxycodone degrades at temperatures close to where vaporization occurs. The
`difficulty of inhalation is a reflection of properties of the drug substance, not the reformulation.
`
`• The postmarketing data support the conclusions reached using the in vitro, PK, and clinical
`data, but do not yet demonstrate, a reduction in OCR abuse following replacement of OC with
`OCR in the marketplace. Additional data, including epidemiological data, when available, will
`provide further information on the impact of OCR on the abuse liability of the drug.
`
`III. Implications of the Above Assessment for Regulatory Decision-making
`
`
`
`
`
`
`
`
`
`
`The data relevant to the potentially abuse-deterrent properties of OCR need to be evaluated together,
`considering the totality of the evidence, to assess the implications for regulatory decisions. My
`recommendations regarding the labeling and relisting issues are discussed below and take into account
`my assessment of the multidisciplinary review together with applicable standards for evaluating the
`respective issues. A summary of the salient data considered as a part of this review is included in the
`tables appended to this document.
`
`
`
`21 CFR 314.127(b)(2), (3), and (4). An NDA supplement is considered an “application” that must
`meet the same evidentiary standards for approval as the NDA itself. See 21 CFR 314.3 and 314.71.
`
`
`FDA has long considered abuse and misuse information a necessary element in evaluating the safety
`of a drug. For example, if a drug has potential for abuse, agency regulations require NDA and IND
`sponsors to provide the agency with “a description and analysis of studies or information related to
`abuse of the drug” and any studies related to overdosage. 21 CFR 312.23(a)(10)(i) and
`314.50(d)(5)(vii). FDA regulations (in effect since 1985) that require sponsors to review, investigate,
`and submit to FDA adverse drug experience reports define “adverse drug experience” broadly to
`
`
`
`Reference ID: 3294145
`
`10
`
`A.
`
`Labeling Issue
`
`
`FDA approves an NDA only