Safety Overview of Postmarketing and Clinical Experience of Sodium Oxybate (Xyrem): Abuse, Misuse, Dependence, and Diversion

Abstract

Study Objectives:

This study reviewed the cumulative postmarketing and clinical safety experience with sodium oxybate (Xyrem®), a treatment approved for cataplexy and excessive daytime sleepiness in narcolepsy. Study objectives were to investigate the occurrence of abuse/misuse of sodium oxybate since first market introduction in 2002, classify cases using DSM-IV criteria for substance abuse and dependence, and describe specific characteristics of these cases.

Methods:

We retrospectively reviewed postmarketing spontaneous adverse event (AE) reports from 15 countries for all cases containing reporting terminology related to abuse/misuse to determine its occurrence. All death cases independent of causality were reviewed to identify associated risk factors.

Results:

Approximately 26,000 patients worldwide received sodium oxybate from first market introduction in 2002 through March 2008. Of those 26,000 patients, 0.2% reported ≥ 1 of the events studied. These included 10 cases (0.039%) meeting DSM-IV abuse criteria, 4 cases (0.016%) meeting DSM-IV dependence criteria, 8 cases (0.031%, including 3 of the previous 4) with withdrawal symptoms reported after discontinuation of sodium oxybate, 2 confirmed cases (0.008%) of sodium oxybate–facilitated sexual assault, 8 cases (0.031%) of overdose with suicidal intent, 21 deaths (0.08%) in patients receiving sodium oxybate treatment with 1 death known to be related to sodium oxybate, and 3 cases (0.01%) of traffic accidents involving drivers taking sodium oxybate. During this period, approximately 600,000 bottles of sodium oxybate were distributed, and 5 incidents (0.0009%) of diversion were reported.

Conclusion:

Cumulative postmarketing and clinical experience indicates a very low risk of abuse/misuse of sodium oxybate.

Citation:

Wang YG; Swick TJ; Carter LP; Thorpy MJ; Benowitz NL. Safety overview of postmarketing and clinical experience of sodium oxybate (xyrem): abuse, misuse, dependence, and diversion. J Clin Sleep Med 2009;5(4):365-371.

Sodium oxybate (Xyrem) is the sodium salt of gamma-hydroxybutyric acid (GHB), an endogenous short-chain fatty acid with properties consistent with those of a neurotransmitter and/or neuromodulator.¹ GHB is found throughout the brain as well as in tissues outside the central nervous system (CNS). GHB is a rapidly acting CNS depressant. In the 1990s, GHB gained a reputation as a club drug because of its use at raves and dance parties, and as a date rape drug because of reports of intoxication, obtundation, and amnesia.^2,3 The concentration or doses of illicitly produced GHB are often unknown and can be contaminated with GHB prodrugs such as 1,4-butanediol (1,4-BD) or gamma-butyrolactone (GBL).⁴ Illicit GHB, particularly in combination with alcohol, can result in greater than intended effects for both the voluntary and involuntary (assault victim) user. The FDA has issued warnings stating that GBL and 1,4-BD, analogs of GHB, which are metabolized to GHB when ingested, also have the potential for abuse and are a danger to public health.⁵ Illicit GHB has been reported to decrease level of consciousness to varying degrees and has been associated with overdose,^6,7 likely because of its relatively steep dose-effect relationship.⁸

GHB and the pharmaceutical product sodium oxybate have been evaluated for abuse liability in non-human species and humans, respectively, using well-established laboratory methodologies. Most non-human studies have shown that GHB is not self-administered to the same extent as other drugs of abuse,⁹ although Weerts et al. showed GHB self-administration under conditions of infrequent and extended access.¹⁰ Similarly, human abuse liability studies have shown that the abuse liability of sodium oxybate is comparable to that of other scheduled and unscheduled depressant drugs such as triazolam and alcohol.^11,12 However, national surveys of drug use, illicit drug confiscation, and emergency room and poison control center calls in the U.S. indicate that the prevalence of illicit use of GHB is very low, has declined since 2000,⁴ and is much lower than that of the major drugs of abuse and club drugs including 3,4,-methylenedioxymethamphetamine (MDMA, street name: ecstasy).¹³ Both epidemiological and laboratory studies have reported that frequent, around-the-clock patterns of GHB administration can produce physical dependence, as evidenced by a withdrawal syndrome.¹⁴ Physical dependence (characterized by the emergence of withdrawal signs and/or symptoms upon cessation of use) was reported with daily use of illicit GHB at supratherapeutic doses of 18 to 250 g/day^15–17 (therapeutic doses of sodium oxybate for narcolepsy range from 4.5 to 9 g/night). In patients who ingested these high doses of illicit GHB around the clock, withdrawal signs and symptoms included psychosis, agitation, tachycardia, hypertension, delirium with auditory and visual hallucinations, diaphoresis, nausea, and vomiting. The onset of these reactions occurred within 1 to 6 hours after the last dose of GHB and lasted 5 to 15 days. One death of a hospitalized patient occurred as the withdrawal symptoms were resolving.¹⁶

In a randomized double-blind placebo-controlled trial designed to evaluate the effects of abrupt discontinuation of sodium oxybate on patients with narcolepsy with cataplexy after a mean nighttime exposure of 21 months, 17% of the patients (5 of 26 patients) withdrawn from active drug reported symptoms of anxiety (2), dizziness (1), insomnia (1), and somnolence (1). These symptoms were interpreted as possibly representing the return of the narcoleptic symptoms or mild withdrawal. This further suggests that when taken under approved and appropriate prescribing conditions, sodium oxybate has little in the way of significant discontinuation effects.¹⁸

Early studies with GHB and later studies with pharmaceutical grade sodium oxybate demonstrated increased duration and intensity of slow wave activity and slow wave sleep in non-human species^19,20 and in patients with narcolepsy,²¹ fibromyalgia,²² Parkinson disease,²³ and obstructive sleep apnea.^24,25 The pivotal clinical trials supporting approval of sodium oxybate for cataplexy and excessive daytime sleepiness (EDS) in patients with narcolepsy demonstrated its clinical efficacy in decreasing the number of cataplectic attacks, and objective and subjective measures of EDS.¹⁷ These studies also demonstrated that sodium oxybate reduced nocturnal awakenings and had an acceptable safety profile, with no long-term safety concerns when used as recommended.^17,26

Sodium oxybate has been approved by the FDA for the treatment of cataplexy (July 2002) and EDS (November 2005) in patients with narcolepsy. In 2005, sodium oxybate was also approved for the treatment of cataplexy in patients with narcolepsy by Health Canada's Therapeutic Products Directorate, and for the treatment of cataplexy in adult patients with narcolepsy by the European Medicines Agency for the European Union (EU) and the Swiss Agency, Swissmedic, for Therapeutic Products. In 2006, sodium oxybate received an expanded indication for narcolepsy with cataplexy in the EU. Sodium oxybate is currently marketed as Xyrem (sodium oxybate) oral solution in 15 countries.

Narcolepsy is a condition that has been documented worldwide, with a prevalence ranging from 0.002% in Israel²⁷ to 0.18% in Japan.²⁸ Prevalence in the US is approximately 0.05%.²⁹ The pentad of clinical symptoms includes: EDS, cataplexy, hypnagogic hallucinations, sleep paralysis, and fragmented sleep. EDS and cataplexy are the most common daytime symptoms of narcolepsy. Historically, treatment of narcolepsy involved the use of one medication to address EDS and a different medication to control cataplexy. This paradigm changed with the availability of sodium oxybate for the treatment of both EDS and cataplexy. Sodium oxybate is the only product approved for the treatment of cataplexy with narcolepsy and is a first-line therapy for patients with narcolepsy.³⁰

In the US, sodium oxybate is classified as a Schedule III controlled substance for medicinal use under the Controlled Substances Act, with illicit use subject to Schedule I penalties. In Canada and the EU, it is classified as a Schedule III and a Schedule IV controlled substance, respectively.

As part of the current US risk management program, sodium oxybate is available through a unique prescribing and dispensing program, using a centralized pharmacy for all prescriptions. The Xyrem Success Program (the risk management program) provides educational materials to the prescriber and the patient explaining the risks and proper use of the product. Once the central pharmacy has documented that the patient has read and/or understood the materials, the pharmacy ships the drug directly to the patient. The Success Program also recommends patient follow-up with the prescribing physician every 3 months. Physicians are expected to report all serious adverse events to the manufacturer directly or through the Success Program. Each patient and each prescribing physician is registered with the central pharmacy in a secure database. To minimize abuse and diversion, the central pharmacy always verifies the legitimacy of any requests for early refills with the prescribing physician.

Since sodium oxybate is a sodium salt of GHB, and illicit GHB is a street drug with a history of abuse and criminal misuse, there has been a perception among some physicians and patients that all of the risks associated with illicit GHB also apply to sodium oxybate. Pharmaceutical sodium oxybate can differ from illicit GHB in accessibility, purity, and dosing.¹³ Therefore, it is important to determine whether the adverse events and abuse/misuse associated with sodium oxybate also differ from those reported with the use of illicit GHB. To determine whether the dangers and concerns attributed to illicit GHB use reflect risks associated with marketed sodium oxybate, we systemically reviewed worldwide postmarketing data on the overall safety and on the occurrence of the abuse/misuse of this product, including abuse, dependence, sexual assault, suicide-related overdose, automobile accidents, death, and diversion.

METHODOLOGY

The postmarketing safety of sodium oxybate is monitored through the spontaneous adverse event reporting system, which collects reports on all marketed pharmaceutical products in the US. In addition, a Post-Marketing Evaluation Program (PMEP) for Xyrem was implemented from 2002 to November 2005 to solicit safety data on an additional 1,000 patients after market introduction. In addition to reporting adverse events through the MedWatch system, under the PMEP, physicians were asked to see their patients at least every 3 months and to write prescriptions that covered no more than 3 months of therapy. Physicians were asked to follow up with patients, to provide adverse event information for the first 6 months of therapy, and to reevaluate patients after 3 and 6 months. Physicians used a company-provided questionnaire to query patients specifically about any instances of vomiting, incontinence, sleepwalking, confusion, and convulsions, and generally, about any other adverse events. The PMEP also captured reports of patients who had not experienced any adverse events while receiving sodium oxybate treatment.

In addition to adverse event reporting, physicians were asked to report any potentially inappropriate use of the product, including ingestion of excessive quantities by the patient, accidental or deliberate use by someone else, and any uncertain dosing resulting from difficulty in dose preparation of the liquid solution. Physicians were asked to record and report to the company all early or overly frequent requests for prescription refills; requests to replace spilled, lost, or stolen product; and patient requests for inappropriate dose increases.

Cases were collected through spontaneous adverse event reporting after market introduction. In addition to spontaneous reports from consumers, healthcare providers, literature case reports, and other sources, the central pharmacy is required to report to the manufacture all adverse events received during contact with consumers or physicians. As part of the central pharmacy's follow-up, a nurse from the pharmacy contacts each patient after the first shipment and at 3 other times during the first 2 months of treatment. Because the pharmacy is required to initiate and maintain initial and longitudinal contact with both patients and physicians, patients have more frequent opportunities to report events than is the case with other products.

We searched all the adverse event cases that were reported to Jazz Pharmaceuticals since the US launch of Xyrem in 2002 through March 2008, which included cases from the US, the EU, Switzerland, and Canada. We reviewed cases with the terms from the Medical Dictionary for Regulatory Activities (MedDRA) that contained the following complete or partial words for adverse events:

• “abuse”

• “misuse”

• “dependence”

• “withdraw”

• “toleran”

• “overdose” (based on reporter's verbatim: cases for attempted or successful suicide)

• “suicid”

• “accident”

• “sleepwalking”

• “somnambulism”

• “breathing-related sleep disorder”

Additionally, we searched for cases with an outcome of “death” or “fatal.” We reviewed all risk management reports from the central pharmacy to identify incidents of diversion. We further analyzed cases reported as possible “abuse” or “dependence” for satisfying the DSM-IV criteria for substance abuse and/or dependence. All reported abuse or dependence cases were rated on a scale of 1 to 4 by a physician, with 1 being least likely to be a case of abuse or dependence, and 4 being most likely to be a case of abuse or dependence. All reports were reviewed by the authors.

Rating Scale

1. Reported abuse or dependence case with no supported evidence or the description was not consistent with an abuse or dependence case. The case did not satisfy DSM-IV criteria for abuse or dependence.

2. Reported abuse or dependence case with some, but not significant, information suggesting the case might be an abuse or dependence case. The case description did not satisfy DSM-IV criteria for abuse or dependence.

3. Reported abuse or dependence case with some information suggesting the case might be an abuse or dependence case. The case description likely satisfied DSM-IV criteria for abuse or dependence.

4. Reported abuse or dependence case with information suggesting the case was very likely to be an abuse or dependence case. The case description satisfied DSM-IV criteria for abuse or dependence.

RESULTS

Overall Safety

A total of 928 PMEP reports were received by July 2006, representing 692 unique patients. Of those reports, 64% (593/928) reported no adverse event. The 10 most commonly reported adverse events from solicited and spontaneous reports were nausea, insomnia, headache, dizziness, vomiting, somnolence, initial insomnia, feeling abnormal, confusional state, and tremor/anxiety. After data from these 692 patients were collected, the PMEP requirement was terminated with the agreement of the FDA (November 2005) because the data were consistent with the adverse events reported in the clinical trials and because of the relatively low rate of physician response. The PMEP results as of September 2004 were summarized by Wedin et al. in 2006.³¹

From product introduction in the US, the EU, and Canada through March 2008, an estimated 26,000 patients have received treatment with commercial sodium oxybate at different doses. During that period, 3,781 adverse event reports were reported to the manufacturer worldwide, representing 973 unique adverse events (excluding reports of ineffectiveness). Given the limitations of postmarketing adverse event reporting, the assessment of causality is usually limited, making causation and/or association in many cases speculative at best. Table 1 lists the 20 most frequent adverse events reported during the period under study. As is typical with postmarketing spontaneous reporting, the incidence of postmarketing adverse events has been lower than that from the clinical trials. Overall, there were no new significant safety findings from the postmarketing adverse event profile compared to what was reported in controlled clinical trials as described in the product prescribing information.

Table 1.

Twenty Most Frequently Reported Adverse Events in Postmarketing Use Through March 2008

Events (MedDRA Preferred Term)	Number of Reports (Incidence)
Nausea	578 (2.2%)
Insomnia	365 (1.4%)
Headache	362 (1.4%)
Dizziness	339 (1.3%)
Vomiting	264 (1.0%)
Somnolence	234 (0.9%)
Initial insomnia	207 (0.8%)
Feeling abnormal	195 (0.8%)
Weight decreased	169 (0.7%)
Confusional state	166 (0.6%)
Tremor	164 (0.6%)
Anxiety	162 (0.6%)
Depression	158 (0.6%)
Fatigue	149 (0.6%)
Diarrhea	111 (0.4%)
Dyspnea	110 (0.4%)
Hyperhidrosis	100 (0.4%)
Blood pressure increased	99 (0.4%)
Paresthesias	99 (0.4%)
Sleep walking	94 (0.4%)

Motor Vehicle Accidents

Sodium oxybate is a CNS depressant, with dizziness as one of the most frequently reported postmarketing adverse events. Sleepwalking/somnambulism, which is among the 20 most frequently reported events, might be associated with the drug's reported effect in increasing slow wave sleep.²¹ Episodes of somnambulism can range from simply sitting up in bed, to walking and attempting to leave one's home. Cases of more complex behavior such as driving, with or without confusional states, have been posited to be associated with slow wave sleep (NREM parasomnias). Because sedation, dizziness, and somnambulism raise concern about traffic accidents in patients taking sodium oxybate, we reviewed all traffic accident and/or somnambulism cases involving driving during the day and/or night.

Through March 2008, 3 cases (0.01% of the 26,000 patients who received sodium oxybate treatment) involving motor vehicle accidents (MVAs) were reported. In one case, the MVA occurred during the day, and it could not be determined whether the patient was receiving sodium oxybate therapy at the time of the accident. A second case concerned a patient who reported dizziness, cold sweats, vomiting, chills, and the inability to think clearly after the first night of sodium oxybate therapy and had an MVA during the day, 36 hours after the last dose of sodium oxybate; the half-life of the drug is 60–90 minutes. The third case concerned a patient who experienced an episode of somnambulism with driving, resulting in an MVA. The patient had experienced sleepwalking and sleep eating on 5 previous occasions during stressful times, and reported being stressed at the time of the event. The patient had ingested Adderall (amphetamine and dextroamphetamine) the previous morning and alcohol 1.5 hours before taking sodium oxybate on the night of the incident. Sodium oxybate therapy was continued, and somnambulism did not reoccur. This is the only case reported of somnambulism with driving.

Abuse

Through March 2008, 14 cases of abuse of sodium oxybate were reported worldwide after market introduction. These cases were reported as possible abuse because they had one or more of the following characteristics (reason for reporting in Table 2):

1. Patient intentionally took more doses and/or more frequently than prescribed.

2. Patient continued to take sodium oxybate after being instructed to stop by the physician.

3. Patient inappropriately used the drug during daytime.

4. Patient attempted to obtain a dose higher than prescribed.

5. Patient engaged in inappropriate activities, such as driving too soon after dosing.

6. Patient exhibited inappropriate behavior.

7. Patient had intentional non-suicidal overdose.

8. Patient used this drug inappropriately with other drugs.

9. Patient provided the drug to others.

Table 2.

Reported Abuse Cases Through March 2008

Prescribed Use/Dose/Frequency	Reported Actual Dose/frequency	Reason for Reportinga	Other Concomitant Psychoactive Drugs	Reported Other Substance Abuse History	Adequate description of Abuse/ Abuse rating per DSM-IV (1-4)b	Resolution	Comment
Narcolepsy/6 g/nightly	Unknown doses during the day	C, E	Fluoxetine	Y	Yes/4	Drug discontinued, event resolved
Social anxiety/4 g/twice nightly	Unknown doses throughout the day	D, I	Fluoxetine	Y	Yes/4	Physician was notified
Narcolepsy/4.5 g/qid or 9 g/bid	Varies, 9 g/ qid was reported	A, C, E, G, H	Zolpidem, alprazolam, escitalopram oxalate, acetaminophen and hydrocodone, trazodone, oxcarbazepine	Y	Yes/4	Drug discontinued
Depression/NIc	Unknown doses anytime during the day	C		Unknown	Yes/4	Drug discontinued and restarted without problem of abuse
Cataplexy/3 g/twice nightly	More than prescribed and during the day	A, B, C	Modafinil, dextroamphetamine, venlafaxine hydrochloride	N	Yes/4	Drug discontinued	Likely abuse case per DSM-IV
Sleep problem/NI	More than prescribed	A	Zolpidem, duloxetine, cyclobenzaprine, tramadol, primidone, buprenorphine/naloxone, Benzedrine, varenicline,	Y	Yes/4	Drug discontinued	Patient was abusing multiple drugs.
Narcolepsy with cataplexy/4.5 g/twice nightly	Unknown dose anytime during the day	C, E	Methylphenidate, alprazolam	N	Yes/4	Physician's office was notified	Likely abuse case per DSM-IV
Unknown/NI	3 times prescribed dose	A	Unknown	Unknown	Yes/4	Drug discontinued	Likely abuse case per DSM-IV
Narcolepsy and cataplexy/9 g/unknown frequency	NI	B	Topiramate, oxcarbazepine	Y	No/3	Drug discontinued, admitted for detoxification
Narcolepsy/NI	More frequent than prescribed	A, F		Unknown	No/3	Drug discontinued	Reported taking doses more frequently than prescribed
NI/NI	More than prescribed	A	Unknown	Unknown	No/2	Drug discontinued	Physician reported patient was taking more medication than was prescribed. No other information.
Insomnia/4.5 g/ nightly	6 g/nightly	A	Sertraline hydrochloride	N	No/1	NA	No evidence of recurrent maladaptive pattern of Xyrem use
Narcolepsy/2.5 g/twice nightly	13.5 g throughout the day	C	Fluoxetine, alprazolam	N	Yes/1	Physician was notified	This case is a dependence case; therefore, the case is not counted as abuse per DSM-IV.
Alpha intrusion associated with fibromyalgia/NI	NI	F	Methadone, tizanidine, clonazepam	Possible	No/1	Drug discontinued	Xyrem was discontinued in Feb 2007. The patient was exhibiting inappropriate behavior in July 2007. Patient was on other medications that may have caused the symptoms.

^a

reasons for reports coded as follows:

1. Patient intentionally took more doses and/or more frequently than prescribed.
2. Patient continued to take sodium oxybate after being instructed to stop by the physician.
3. Patient inappropriately used the drug during daytime.
4. Patient attempted to obtain a dose higher than prescribed.
5. Patient engaged in inappropriate activities, such as driving too soon after dosing.
6. Patient exhibited inappropriate behavior.
7. Patient had intentional non-suicidal overdose.
8. Patient used this drug inappropriately with other drugs.
9. Patient provided the drug to others.

^b

rating scale:

1. Reported abuse or dependence case with no supported evidence or the description was not consistent with an abuse or dependence case. The case did not satisfy DSM-IV criteria for abuse or dependence.
2. Reported abuse or dependence case with some but not significant information suggesting the case might be an abuse or dependence case. The case description did not satisfy DSM-IV criteria for abuse or dependence.
3. Reported abuse or dependence case with some information suggesting the case might be an abuse or dependence case. The case description likely satisfied DSM-IV criteria for abuse or dependence.
4. Reported abuse or dependence case with information suggesting the case was very likely to be an abuse or dependence case. The case description satisfied DSM-IV criteria for abuse or dependence.

^cNI – no information

Of these 14 cases, 10 (0.039%) of the patients who took the commercial product during the time evaluated satisfied DSM-IV criteria for substance abuse and 1 case satisfied DSM-IV criteria for substance dependence, which is discussed separately in the section on dependence. The remaining 3 cases contained no definitive evidence of abuse and were considered suspicious of abuse. All 14 cases were either reported by or to the physicians who prescribed sodium oxybate at the time of reporting. During clinical trials with sodium oxybate in individuals with narcolepsy, 781 patients received the drug at different dosages in controlled clinical trials. No cases of abuse of sodium oxybate were reported in these clinical trials.

Dependence

Physical dependence refers to adaptations resulting in withdrawal signs or symptoms when the drug is discontinued.³² Dependence per DSM-IV criteria (i.e., addiction) refers to the loss of control over the desire to acquire and use the drug regardless of adverse consequences associated with obtaining and using it. Physical dependence and dependence per DSM-IV criteria are distinct phenomena, and the terminology should not be used interchangeably.

In worldwide postmarketing adverse event reporting, there have been 20 cases of dependence on, withdrawal from, or tolerance to sodium oxybate from 2002 through March 2008. In reviewing these 20 cases, we found 4 cases (0.016% of patients who took the commercial product during the time evaluated) that satisfied DSM-IV criteria for substance dependence. Three of these cases also satisfied DSM-IV criteria for physical dependence. One patient reported taking 20 g sodium oxybate at bedtime followed by 20 g throughout the day; one patient reported taking sodium oxybate 8 to 10 times during a 24-hour period at doses from 1.5 g to 4.5 g; and one patient reported taking sodium oxybate 13.5 g (unknown whether this was the total daily dose or one dose) throughout the day. A fourth case involved a patient who had already developed dependence on Alcover (a formulation of sodium oxybate available in Europe) before starting Xyrem. When Alcover was no longer available in that country, the physician prescribed Xyrem.

We examined all the remaining cases of reported abuse and dependence that did not satisfy DSM-IV criteria for abuse or dependence for evidence of withdrawal signs or symptoms from sodium oxybate. Five of these remaining cases included adverse events that were reported after sodium oxybate was discontinued, which add to a total of 8 (0.031%) physical dependence cases. The doses in these 5 cases were reported as 4.5 g nightly (2 cases), 9 g nightly, and unknown (2 cases; Table 3). Withdrawal signs and symptoms included headache, dizziness, auditory hallucinations, questionable psychotic behavior, thought disturbance, agitation, anxiety, tremors, rebound fatigue and sleepiness, confusion, and paranoia (see Table 3).

Table 3.

Cases of Dependence/Withdrawal/Tolerance Reported Through March 2008

Prescribed Use and Dose – Frequency at the Time of Dependence/ Titration Exceeds PIa Guidelines	Other Concomitant Psychoactive Drugs	Other Sub-stance Abuse History	Withdrawal Symptoms	Resolution	Dependence Rating per DSM-IV (1-4)b/ Physical Dependence (withdrawal after discontinuation) (1-4)b	Comments
DSM-IV DEPENDENCE CASES
Narcolepsy 13.5 g throughout the day/Y	Fluoxetine, alprazolam	N	sweating, weakness, dizziness, upset stomach	Physician was notified	4/4
Narcolepsy with cataplexy 40 g per day/NIc	Unknown	Y	peripheral tingling, goose bumps, sweating, inability to sleep	Drug discontinued	4/4
Cataplexy and Alcoverd dependence 4.5 g bid plus 3g tid/NAe	Alcover	Y	weakness, shaking, nausea, disorientation	Drug treatment on-going	4/4	Patient had pre-existing dependence to Alcover. Xyrem was prescribed for Alcover dependence.
Insomnia 4.5 g 8-10 times a day/Y	Hydrocodone	Y	No withdrawal symptoms	Drug discontinued	3/1
PHYSICAL DEPENDENCE CASES
Insomnia 6 g nightly/N	Sertraline hydrochloride	N	Visual hallucination, agitation, tremors, tachycardia, nausea, dilated pupils, “picking”, increased muscle tone, hyperreflexia, dehydration	Drug discontinued	1/3	Patient was taking 6 g nightly instead of prescribed 4.5 g for several nights prior to discontinuation. The patient was discontinuing other medications at the same time.
Narcolepsy with cataplexy 4.5 g nightly/N	Gabapentin, topiramate	N	Headache after 9 days discontinuation	Drug continued	1/3
Narcolepsy 9 g nightly/Y	Methylphenidate	N	dizziness, auditory hallucinations, chest pain, paranoia, and thought disturbance	Drug discontinued	1/3	Symptoms started four days after the discontinuation of sodium oxybate
Idiopathic hypersomnia Unknown/NI	Fluoxetine, methylphenidate	Y	Confusion, paranoia	Drug discontinued	1/3
Sleeping disorder Unknown/NI	Venlafaxine	Unknown	headache, feeling agitated anxious, tremulous, rebound, and sleepiness	Unknown	1/4
OSAf 2 g, 2.5 g, 3.5 g, 3.5 g nightly/NA	Unknown	N	No withdrawal symptoms	Drug continued	1/1
Narcolepsy 3 g three times a night/Y	Unknown	N	anxiety, tightening of abdomen, the urge to defecate	Drug continued	1/1	Symptoms occurred every morning after taking Xyrem
Fibromyalgia 6.5 g, then 5 g nightly/Y	Unknown	N	No withdrawal symptoms	Unknown	1/1
Fibromyalgia 5.5 g twice nightly/Y	Unknown	N	crying, emotional, tired	Unknown	1/2	Physician felt the events were signs and symptoms of the patient's anxiety and depression
Insomnia 4.5 g twice nightly/NI	Unknown	N	No withdrawal symptoms	Drug discontinued	1/1	Patient reported tolerance because it took longer for the patient to fall asleep than before
Narcolepsy with cataplexy and EDS 12 g nightly/Y	Unknown	Unknown	No withdrawal symptoms	Drug continued	1/1
EDS with narcolepsy 15 g unknown frequency/Y	Amphetamine	Unknown	No withdrawal symptoms	Unknown	1/1	Term used in searching index can not be collaborated with evidence
Narcolepsy with cataplexy Twice nightly/NI	Venlafaxine, escitalopram oxalate, trazodone	N	No withdrawal reported	Drug discontinued	1/1	Term used in searching index can not be collaborated with evidence
Unknown Unknown/NI	Fluoxetine, methylphenidate	Unknown	No withdrawal reported	Unknown	2/1	Term used in searching index can not be collaborated with evidence
Narcolepsy 2.25 g twice nightly?/NI	Oxycodone, clonazepam, bupropion,	N	No withdrawal reported	Unknown	1/1	Term used in searching index can not be collaborated with evidence
Unknown Unknown/NI	Unknown	N	No withdrawal reported	Unknown	1/1	Term used in searching index can not be collaborated with evidence

^aPI - prescribing information

^b

Rating scale:

• Reported abuse or dependence case with no supported evidence or the description was not consistent with an abuse or dependence case. The case did not satisfy DSM-IV criteria for abuse or dependence.
• Reported abuse or dependence case with some but not significant information suggesting the case might be an abuse or dependence case. The case description did not satisfy DSM-IV criteria for abuse or dependence.
• Reported abuse or dependence case with some information suggesting the case might be an abuse or dependence case. The case description likely satisfied DSM-IV criteria for abuse or dependence.
• Reported abuse or dependence case with information suggesting the case was very likely to be an abuse or dependence case. The case description satisfied DSM-IV criteria for abuse or dependence.

^cNI – no information

^dAlcover - a formulation of sodium oxybate available in Europe

^eNA – not applicable

^fOSA - obstructive sleep apnea

Drug-Facilitated Sexual Assault

To determine the incidence of sexual assault in which sodium oxybate was involved, we examined all cases from clinical studies and from postmarketing surveillance through March 2008. We found 4 cases, which are described below.

Three of the 4 cases were postmarketing cases involving sexual assault, and one case was from a clinical trial. The involvement of sodium oxybate was confirmed in 2 of the 3 postmarketing cases (0.008%).

• In 2003, a female patient reported being sexually assaulted in her home by another member of the household. The patient stated that the assailant was aware of the effect of the drug and knew that she would be “incapacitated.”

• In 2005, a report was received that a male stole sodium oxybate that had been prescribed to his family member. He transported the sodium oxybate to a social event with an acquaintance. The acquaintance alleged that they used the drug together, after which he sexually assaulted her.

The third postmarketing case, which occurred in 2007, consisted of a physician who reported that a local law authority had notified him that one of his patients had been sexually assaulted. Although the physician reported prescribing sodium oxybate in February 2005, the central pharmacy had no record that the drug had ever been shipped to this patient.

The fourth case occurred during a clinical trial in an individual with narcolepsy in 2002. A participant in the trial reported being sexually assaulted while enrolled in the study and chose to continue in the study.

Overdose Associated with Suicide or Suicide Attempt

We found 1 (0.0039%) completed suicide case with polydrug use including sodium oxybate and 7 (0.027%) reported suicide attempts involving sodium oxybate overdose with or without the involvement of other drugs or substances. These cases are presented in Table 4.

Table 4.

Cases of Suicide or Attempted Suicide with Overdosing Reported Through March 2008

Prescribed Use	Prescribed Dose	Overdose Amount	Other Products Involved in Overdosing	Risk Factors	Outcome	Comment
Insomnia	4.5 g nightly	unknown	Possible temazepam, zolpidem	Previous suicide attempt a year ago via poly-drug ingestions	Death caused by polydrug intoxication associated with drowning	The postmortem GHB concentration in this case (81mcg/L) must be interpreted cautiously. Endogenous postmortem GHB concentrations rise significantly after death. The GHB concentration cited in this case is comparable to GHB concentrations reported postmortem in patients where the cause of death could exclude GHB exposure.
Severe insomnia	4.5 g nightly	20 g	None	History of suicide attempt	Recovered	The prescribing physician did not consider the event was related to sodium oxybate.
Unknown	Unknown	3-5 doses (dosage unknown)	None	Unknown	Recovered
Narcolepsy	4.5 g twice nightly	Unknown	Unknown	History of suicide attempt	Recovered	Suicide attempt with unspecified medications overdose was suspected by admitting physician
Fibromyalgia	7.5 g nightly	Unknown	Clonazepam, zolpidem tartrate	History of depression	Recovered
Unknown	Unknown	< 270 g	None	Unknown	Recovered
Narcolepsy	5-6 g nightly	Unknown	None	Unknown	Recovered
Mild obstructive sleep apnea and insomnia	3.75 g twice nightly	3.75 g followed by another 3.75 g in 5 min	None	The patient experienced hallucinations where she was convinced to kill herself.	Recovered

Fatalities

In the US, in the event of a patient death for any reason, the central pharmacy is likely to receive a notification from the patient's family or prescribing physicians to stop drug shipments. As a result, reports of deaths are likely to be more numerous than is typical for a product with a similar patient exposure (approximately 26,000 patients) but without a central distribution system. During the period under this study for postmarketing safety, 30 deaths were reported worldwide. In 2 of these cases, sodium oxybate had been discontinued prior to death; in 7 cases, it is unknown, unclear, or unlikely that the patients were taking sodium oxybate at the time of death; and in 6 of the remaining 21 cases, the deaths were considered not related to sodium oxybate by the prescribing physicians. Causes of death for the remaining 15 cases were as follows: 7 with unknown causes, 3 from drug overdose (1 involving sodium oxybate; 2 not involving sodium oxybate), and 1 each from lung cancer, suicide, drowning, renal failure, and cardiac arrest (Table 5).

Table 5.

Cases with Death as an Outcome Reported Through March 2008

Prescribed Use	Prescribed Dose	Cause of Death	Other Relevant information	Reporting Physician's Causality Assessment to the Drug
NAa	2.25 g twice nightly	Unknown	Urine was positive for amphetamines, benzodiazepines, and TCAb	Possibly related
NA	4.5 g nightly	Unknown	Unknown if the patient had discontinued the drug prior to death	NA
Narcolepsy	5 g nightly	Metastatic lung cancer	Longstanding Hxc of smoking	NA
Insomnia	4.5 g nightly	Suicide with polydrug intoxication in association with drowning	Hx of a previous suicide attempt one year ago	NA
Trouble sleeping	NA	Unknown	Hx of bipolar disorder and overdose attempts	NA
Fibromyalgia	3 g twice nightly	Heart disease		Not related
Insomnia	2 g twice nightly	Unknown	Hx of severe insomnia, mild OSA, and obesity. Was taking alprazolam concomitantly. Urine was positive for benzodiazepines, acetaminophen, diphenhydramine, and quetiapine. Blood was positive for GHB, alprazolam, nordazepam, and quetiapine.	NA
Fibromyalgia – Insomnia	3 g twice nightly	Polydrug accidental overdose excluding Xyrem	Serum was positive for tramadol, fluoxetine, lidocaine, and pseudoephedrine	NA
Insomnia – Pain	3 g once nightly	OxyContin overdose		NA
Narcolepsy	NA	Heart attack	Hx of severe sleep apnea, not compliant with CPAP	Not related
Narcolepsy with cataplexy	7 g twice nightly	Fell asleep and drowned in tub	Reported to have traces of Xyrem found in her system	NA
Insomnia	6 g nightly	Renal failure	Hx of diabetes and renal problem	NA
NA	NA	Unknown	Central pharmacy received returned drug shipment with a note of pt had passed away	NA
Sleep disturbance	2 g nightly	Overdose on Klonopin and opioids	Physician believed the patient had run out of Xyrem at the time of death. Hx of being hospitalized every other month of depression	Not related
NA	NA	Unknown	Unknown if the patient was taking Xyrem at the time of death	NA
NA	NA	Cardiac arrest and chronic obstructive pulmonary disease exacerbation		NA
Fibromyalgia	NA	Suicide	Death of a family member and familiar illness prior to death	Not related
Diffuse Lewy body dementia	1st dose: 5 g 2nd dose: 4 g	Natural causes		Not related
Insomnia	4.5 g twice nightly	Unknown	Hx of seizure, unknown if the patient took the drug prior to death	NA
Sleep cycle problem	NA	Post-op GI tract hemorrhage	GI bleeding post hip replacement, unknown if the patient was taking the drug during this period	Not related
Fibromyalgia	3 g twice nightly	GHB intoxication	Toxicology was positive for GHB, amitriptyline, morphine, tramadol, but the reporter noted the interpretation of postmortem GHB concentration should be made with caution.	NA
EDS and fibromyalgia	2.25 g twice nightly	Unknown	Died the next day following elbow surgery, unknown if the patient was taking the drug during this period	NA
Insomnia	1st dose: 4.5 g 2nd dose: 5 g	Unknown, possible heart disease		Not related
Fibromyalgia	2.25 g twice nightly	Myocardial infarction	Drug was discontinued several weeks prior to the death	Not related
Narcolepsy with cataplexy	1.25 g twice nightly	Unknown	Hx of disturbed metabolism, developed auditory hallucination after the drug, all tests were within normal limits, did not take the drug the day before the death, however, toxicology reported presence of amphetamine and sodium oxybate	NA
NA	NA	Unknown	Hospitalized prior to death for viral syndrome and dehydration	NA
Narcolepsy	NA	Unknown	Track marks were found on arm. Physician unsure if the patient had been taking the drug prior to death	NA
NA	9 g twice nightly	Unknown	Unclear if the patient was taking the drug at the time of death	NA
NA	NA	Unknown	Hx of drug abuse and illicit GHB use	NA
NA	NA	Possible liver failure		Not related

^aNA - Not available

^bTCA - tricyclic antidepressant

^bHx - history

In 4 cases, postmortem toxicology showed positive serum GHB in varied concentrations. Toxicology reports of GHB are difficult to interpret in such cases. For example, one of the toxicology reports included a caution about the interpretation of postmortem GHB. It has been reported that endogenous GHB blood levels rise significantly following death.³³ Therefore blood levels of GHB following death do not necessarily indicate consumption of exogenous GHB or sodium oxybate.

Diversion

Under the current risk management program, sodium oxybate is shipped directly to each patient via express courier to avoid diversion and to minimize the risk of misuse. The central pharmacy reports all discrepancies, including lost, stolen, or diverted shipments and obtains the prescribing physician's approval for any early refill requests.

Of the approximately 26,000 patients who have received the commercial drug worldwide and the approximately 600,000 bottles of sodium oxybate distributed during the same time period, there have been 5 confirmed incidents (0.0009%) of drug diversion, where the drug was taken from patients and was used or was intended to be used by others. Incidents of diversion included 1 patient who was taking a sibling's drug because the patient couldn't afford the product; 1 case reported by a father who found drug that did not belong to his child in the child's room but did not know if the child had ingested any; 1 case reported by a mother who stated that her child had overdosed on the drug but did not know where her child had obtained the drug; 1 patient who used a friend's drug; and 1 person who had stolen a family member's drug and allegedly used the drug with an acquaintance (discussed earlier under drug-facilitated sexual assault).

We further identified 6 incidents of possible diversion where the drug was reported to have been stolen with no information about subsequent use, 9 incidents of coincidental theft where the drug was part of a general theft including other items, and 22 incidents where drug was lost or missing in various situations and theft was not suspected.

DISCUSSION

Over the past 5 years, the overall postmarketing safety profile of sodium oxybate relative to issues of misuse has been very good and similar to that observed in controlled clinical trials in individuals with narcolepsy. Approximately 26,000 patients worldwide received commercial sodium oxybate from first market introduction in 2002 through March 2008. Of those 26,000 patients, 0.2% reported ≥ 1 of the events in this study. These included 3 cases describing traffic accidents, 10 cases that satisfied DSM-IV criteria for substance abuse, 4 cases that satisfied DSM-IV criteria for substance dependence, 5 cases that did not satisfy DSM-IV criteria for substance dependence, but satisfied criteria for physical dependence, 2 confirmed cases of sexual assault involving sodium oxybate, 8 cases of overdose with attempted suicide, and 21 deaths from various causes. Of the approximately 600,000 bottles distributed during this period, there were 5 incidents of diversion. Cumulative commercial experience indicates that the incidence of misuse and the risk of abuse, dependence, diversion, and drug-facilitated sexual assault with sodium oxybate appear to be very low. Concern that sodium oxybate might present the same risk profile as illicit GHB is not supported by the available data.

These results should be viewed with the caveat that postmarketing safety data are often underreported because they typically rely on spontaneous reporting. Underreporting occurs for various reasons: the event was tolerable, no time to report, fear of reprisal associated with misuse of a product, and lack of familiarity with the reporting process and requirements. In this case, however, given the frequency and longitudinal nature of the central pharmacy's contacts with patients and physician, the early refill follow-up, and the requirement for verification prior to drug shipment, it is likely that adverse events, including misuse and death, have been more completely captured with sodium oxybate than with other products with similar patient exposure. The persistent decrease in the incidence of illicit GHB abuse and intoxication reported by Drug Abuse Warning Network (DAWN) and the American Association of Poison Control Centers^4,13 over the past several years indirectly suggests that the incidence of sodium oxybate misuse is low. This adverse event database set includes reports received spontaneously from patients, physicians, family members, regulatory agencies, the central pharmacy, the literature, and international product licensing partners.

Since the completion of the analysis presented in the results section of this paper, reports of 4 sodium oxybate-related fatalities were published. One case was previously identified during our study. The remaining 3 cases could not be definitively linked to a previously identified case. Akins et al. reported 1 fatality in a patient with a medical history of narcolepsy, sleep apnea, and sarcoidosis, who was prescribed other CNS-depressant drugs in addition to sodium oxybate.³⁴ The cause of death was reported as acute intoxication due to the combined effects of GHB, tramadol, and carisoprodol. Zvosec et al. reported 3 fatalities involving sodium oxybate.³⁵ The 3 cases included one overdose case in a patient with a history of multiple drug overdoses. In the other 2 cases, the causes of death were not established; however, postmortem toxicology detected other CNS-depressant drugs in both patients' systems. Both patients were obese, and one of them had mild obstructive sleep apnea. No cardiac profile was provided for either patient.

Sleep apnea has been reported with a high incidence in some cohorts of narcoleptic patients.¹⁷ While this paper has focused primarily on issues on misuse of sodium oxybate, a study by George et al.³⁶ and a case report from Seeck-Hirschner³⁷ noted a worsening of sleep disordered breathing after the initiation of sodium oxybate. Caution should be observed when sodium oxybate is prescribed to patients with suspected compromised respiratory function. If there is a concern, the patient should be assessed for sleep disordered breathing by a physician before prescribing sodium oxybate. Ways to assess potential sleep disordered breathing include home sleep testing, which includes pulse oximetry, or full in-laboratory nocturnal polysomnography. Sodium oxybate is contraindicated in patients being treated with sedative hypnotic agents.

Risks in association with the use of sodium oxybate, which is a CNS depressant, should be monitored carefully. Some studies with GHB have demonstrated significant CNS depression without respiratory depression,³⁸ while others have reported respiratory depression with GHB abuse, often in the presence of cointoxicants, such as alcohol.^39,40 Clinical trials of sodium oxybate in narcolepsy have also reported respiratory depression.^17,21

Sodium oxybate has been used safely and successfully in treating narcolepsy patients with EDS and cataplexy for several years. As with any controlled substance, it is very important for clinicians to monitor the safe use of sodium oxybate in patients regularly. The Xyrem Success Program recommends patient follow-up every 3 months. Clinicians should also assess a patient's potential for drug abuse and suicide before issuing the prescription. Good clinical practice dictates that patients be followed to ensure continuing safe use of the drug and monitor for signs of misuse, abuse, and dependence. All healthcare practitioners need to evaluate early refill requests, to assess potential tolerance to the drug, and to work closely with the pharmacy to prevent potential abuse or diversion. When used according to prescribing and monitoring guidelines, sodium oxybate has been shown to be a safe and effective treatment for cataplexy and EDS in narcolepsy and is associated with a very low incidence of misuse, abuse, and dependence.

DISCLOSURE STATEMENT

This manuscript was prepared by employees of Jazz Pharmaceuticals and outsde medical experts. None of the medical experts were paid for their work on this paper. Dr. Wang is an employee of Jazz Pharmaceuticals and owns stock and stock options in the company. Dr. Carter is a former employee of Jazz Pharmaceuticals and owns stock in the company. Dr. Swick has received research support from Jazz Pharmaceuticals/Orphan Medical, Takeda, Merck, Cephalon, Sanofi-Aventis, Pfizer, Somaxon, GlaxoSmithKline, and Epix Pharmaceuticals; has consulted for Jazz Pharmaceuticals and Boehringer Ingelheim; and is on the Speakers' Bureaus of GlaxoSmithKline, Sepracor, Jazz Pharmaceuticals, Sanofi-Aventis, and Boehringer-Ingelheim. Dr. Thorpy is on the Speakers' bureaus for Cephalon Inc., Jazz Pharmaceuticals, Sanofi-Aventis, GlaxoSmithKline and Takeda Pharmaceuticals. Dr. Benowitz has conducted clinical research for which medication was provided as a gift by Jazz Pharmaceuticals.