13 November 2007

VERSION: 1 2 3 4 5 6 7 8 9

BUP-3025/Mod–Severe Knee Pain in Opioid Naïve

Norspan® Patches

A Multi-center, Randomized, Double-blind, Placebo-controlled Study with an Open-label Run-In to Assess the Efficacy, Tolerability, and Safety of BTDS 10 or BTDS 20 Compared to Placebo in Opioid-naïve Subjects with Moderate to Severe Chronic Pain due to Osteoarthritis of the Knee

Sponsor: Purdue Pharma, LP (PPLP)

Study Drug: Norspan^® Patches = Buprenorphine Transdermal Delivery System (BTDS)

Dose: 5, 10, 20

Protocol: BUP3025

Phase: III

CRO: PRA International

Site #: 114A

Screening #

CRA: Mariah Millard, MS MillardMariah@PRAIntl.com 904-335-0104

Medical Monitor: Deborah Steiner, MD (Purdue) & Sophia Fourie, MD (PRA I.)

Study Ends Recruitment:

Central Lab: Quest Laboratories

Central EKG: Biomedical Systems

IRB: Quorum

520 Subjects / 80 Sites → 7 Subjects per Site R 1:1 BTDS: Placebo

RM = Rescue Medication/Supplemental Analgesic Meds Not Allowed x 30 H Prior to Each V

Clinical Supply Services: Acculogix

Investigator Meeting: Schaumburg Illinois at:

The Renaissance Schaumburg Hotel & Convention Center 11/1-11/3/07

Call Medical Monitor for the 1^st 2 Screen Fails to discuss

13 November 2007

VERSION: 1 2 3 4 5 6 7 8 9

BUP-3025/Mod–Severe Knee Pain in Opioid Naïve

IC

Age: ≥ 40

Clinical Diagnosis OA Knee x 1 year or longer with Crepitus on active ROM (40-49)

Confirmed by Grade II-IV Radiographic Evidence within the past 2 years

Not adequately controlled with Non-Opioid Analgesic Meds

Good Candidate for around-the-clock opioid therapy

Screening “Average Pain Over the Past 14 Days” Score ≥ 5 for two consecutive days

Adjunct Rx unchanged during study

EC

Adjunct Rx changed during study; QTcB > 480 msec or + FH Long QT Sx at Screening V

Radicular Sx’s; FM; RSD; PHN; Gout, Pseudogout; Active Lyme Dz; RA; Neuropathic Pain

Surgical Procedure directed at pain relief < 6 months; SD; Unstable Psych meds x < 1 month

Unstable CV Dx; LFT > 3x; Creatinine > 2.5; Cancer x 2 yrs; MAO Inhibitor

Prior exposure to BTDS (Buprenorphine Transdermal System); WC Litigation

Muscle Relaxants if used for pain & if not stable x 1 month

PRN Opioids ≥ 5 mgm OxyIR x 2 Weeks

Daily Opioids ≥ 5 mgm OxyIR x 3 Months

Responsive to BTDS is defined by meeting both of the following two criteria:

1. Subject reports a ≥ 2 point reduction in “Average Pain Over the Last 24 H” score when compared with their score from the Screening V that occurs on the 3 consecutive days immediately prior to Randomization.

2. The “Average Pain Over the Last 24 H” score remains ≤ 4 on the 3 consecutive days immediately prior to Randomization.

22 August 2007

VERSION: 1 2 3 4 5 6 7 8 9

BUP-3025/Mod–Severe Knee Pain in Opioid Naïve

V1 D -10 to -1 Screening Period (6-10 Days), Lab, ICF, EKG x 2 > 10 min. apart;

. PS x 14 Days ≥ 5; Diary

Vp Call Subject when meets I/E criteria & D/C All Analgesic Meds & Begin recording Average Pain over the last 24 H scores in the diary at 8PM every evening. Starting 2 days after all pain meds are stopped, Call Subjects daily to assess their pain scores. Subjects who report 2 consecutive days of average pain over the last 24 H scores ≥ 5 (excluding the day when the subject stopped all pain meds) will be instructed to Stop recording pain scores in the diary & RTC for V2. (If can not RTC the next day, may resume incoming non-opioid analgesic meds until V2).

V2 Open-Label Run-In (up to 27 days); BTDS 5 x 3 D; PS x 24 H ≥ 5 (Recorded at 8PM); . . Schedule D3 Phone Contact & V2a within 13-15 days; No supplemental Analgesics!

Vp D3 Run-In Day 3 Phone Contact by PI to assess the Safety & Tolerability of BTDS 5; If tolerated, . . . increase to BTDS 10 ; Remind to record the Average Pain Over the last 24 H at 8PM nightly

V2a Received BTDS 10 x 10-12 D; If Tolerate & Respond (PS x 24 H ≤ 4 & ≥ 2 point reduction from Screening x 3 Consecutive D prior to R) to BTDS 10 enter Double-blind Phase & R BTDS vs Placebo. If can not tolerate BTDS 10 → Flunk Out! If needed may increase BTDS 20 for an additional 10 Days. If Tolerate & Respond to BTDS 20 may enter Double-blind Phase & R BTDS 20 vs. Placebo.

V3 Lab; Randomization 1:1 BTDS vs. Placebo; (84 Days/12 Wks); For the first 6 days Oxy IR 5 mgm Bid prn. Then, Tylenol or . . Advil (for Tylenol intolerant) PRN. Call & advise to Hold PRN meds x 30 H prior to q V. Assess pain during the prior 24 H while . off supplemental prn analgesic use, EKG x 2 > 10 min. apart; RTC 1 Wk

V4 W 1 Call 2 days before & remind No Supplemental Analgesics x 30 H prior to V; +/- 1 D Window

V5 W 2 Call 2 days before & remind No Supplemental Analgesics x 30 H prior to V; +/- 2 D Window

V6 W 4 Call 2 days before & remind No Supplemental Analgesics x 30 H prior to V, EKG x 2 > 10 min. apart; +/- 2 D Window

V7 W 8 Call 2 days before & remind No Supplemental Analgesics x 30 H prior to V: +/- 2 D Window

V8 W 12 Lab; EKG x 2 > 30 min. Apart; Call 2 days before & remind No Supplemental . … Analgesics x 30 H prior to V; +/- 2 D Window

22 August 2007

VERSION: 1 2 3 4 5 6 7 8 9

Norspan^® Patches

BUP-3025/Mod–Severe Knee Pain in Opioid Naïve

Opioid Narcotics Equianalgesic Conversion Table

MS PO: IV 3:1 (MSIR, MS Contin, Avinza, Kadian)

MS: Hydrocodone 1:1 (Lortab, Vicodin)

MS: Oxycodone 1.5:1 (Percocet & Oxy IR)

MS: Hydromorphone PO 4:1 (Dilaudid)

MS: Hydromorphone IV 7:1

MS: Codeine 1:7 (Tylenol # 3)

MS: Demerol 1:10

MS: Oxymorphone (Opana) 3:1

MS: Tramadol (Ultram) 1:10

MS: Propoxyphene (Darvocet) 1:7

MS: Levorphanol (Levodromoran) 1:0.125

MS: Duragesic (Fentanyl) 60 MS = 25 mcg Patch

Methadone 7.5 = about 60 MS (variable depending on dose)

The situation with Equianalgesic potency equivalents with Buprenorphine BTDS (Norspan® Patches) is less clear, and may vary according to the length of previous opioid administration, but the equipotency (dose required to obtain the same degree of analgesia) compared to oral morphine (MS) or oral Oxycodone/OxyContin has been quoted as being about:

1. Buprenorphine BTDS 5 = MS 15 = Oxycodone 10 (OxyContin 10 OD)

2. Buprenorphine BTDS 10 = MS 30 = Oxycodone 20 (OxyContin 10 BID)

3. Buprenorphine BTDS 20 = MS 45 = Oxycodone 40 (OxyContin 20 BID)

22 August 2007

VERSION: 1 2 3 4 5 6 7 8 9

BUP-3025/Mod–Severe Knee Pain in Opioid Naïve

Norspan® Patches

Osteoarthritis Severity Grades for the Knee

This Grading System for Radiologic Evidence of OA (Osteoarthritis) of the Knee in the Standing Position is to be used to qualify study subjects by documenting they have the requisite severity of OA. They require at least Grade II or more severe OA to participate.

Grade 0 Normal

Grade I Doubtful Narrowing of Joint Space and . . . Possible Osteophytic Lipping

Grade II Definite Osteophytes and

. Possible Narrowing of Joint Space

Grade III Moderate Multiple Osteophytes,

Definite narrowing of joint space, and

Some Sclerosis and Possible deformity of bone ends

Grade IV Large Osteophytes,

Marked Narrowing of Joint Space,

Severe Sclerosis and

Definite Deformity of Bone Ends

22 August 2007

VERSION: 1 2 3 4 5 6 7 8 9

BUP-3025/Mod–Severe Knee Pain in Opioid Naïve

The Science

Norspan^® Patches/Buprenorphine BTDS is a synthetic opioid analgesic with partial mu-opioid agonist and kappa-opioid antagonist properties. Buprenorphine has been widely used SL and parenteral. Buprenorphine came out as an injectable in USA 1982. The SL is approved for treating addiction. It is a Schedule III drug under the Controlled Substances Act.

Following removal of the patch, concentrations decrease in half by 12 H. Norspan^® Patches are approved already in Europe at the BTDS 20, 30 & 40 strengths.

Common AE’s (Adverse Events) seen with Norspan^® Patches

1. Patch Site Reactions:

a. Erythema

b. Pruritus

c. Rash

2. Mu-opioid agonist effects:

a. N/V

b. Dizzy

c. Dry Mouth

d. Somnolence

e. Constipation

3. QTc Prolongation similar to Avelox (Moxifloxacin HCL)

22 August 2007

VERSION: 1 2 3 4 5 6 7 8 9

BUP-3025/Mod–Severe Knee Pain in Opioid Naïve

Norspan® Patches

Questions

1. Indemnification: The sponsor likes to be held harmless for acts of negligence that the PI commits that are solely the error of the treating physician. For example, if a PI treats a subject's UTI during a study and the patient suffers an injury from this, it is squarely the PI’s fault and it would not be proper for the sponsor to sustain an injury by being held liable for it. However, we are aware of a BP study where the PI followed the protocol and had the subject wash off their BP medication and suffer a stroke from the lack of the proper BP treatment. The sponsor forced the PI to use his medical malpractice insurance to satisfy the claim for medical malpractice. Clearly, this was an injustice as the PI was following the protocol precisely and even though it is unusual for a stroke to occur during wash out, it certainly does not immunize the subject from having a stroke. When a stroke occurs during wash out it is important that the sponsor take responsibility for the liability. This should be true of any injury ex officio of changing a subject’s treatment while following the protocol. If this drug causes respiratory depression, are you going to be fully liable?
2. Study Fully Populated Causing Premature Early Termination: As sites advertise for patients to become research volunteers, the people that are good enough to step forward to join, wash off of their existing treatment, come to our site and devote a significant amount of their time with blood draws, signing paperwork and allowing the experimental protocol to progress, should be guaranteed that the sponsor will not prematurely shut down the study because they have obtained the needed number of subjects. That is, it should be the sponsor’s responsibility to do the number crunching early enough not to allow subjects to sign consent and undergo early termination just because the study is now considered full enough. This is not ethical or in the patients best interest. All subjects who sign consent must be allowed to continue thru the protocol if they are continuing to fulfill criteria regardless of how many subjects are in the trial.

3. Reimbursement for Screen Failures: Sites are taking finincial risk in the name of enrolling only those subjects who will successfully complete the trial. Sponsors need to recognize that this significantly slows enrollment. When we see a subject who may qualify but then again may not, we would be hesitant to enroll in that study. Since the exclusivity clock starts clicking as soon as the drug is patented it is imperative that the studies are done as quickly as possible. Sites that are rewarded for each piece of work they legitimately perform will have to enroll better. Sites with too high screen failure rates can be turned off to ensure the proper motivation to enroll high quality subjects is present.
4. Initial Fasting Labs and Informed Consent: Many trials have visit one with fasting labs and signing informed consent on the same day. This is less desirable than having a visit one where the subject presents non-fasting for the purpose of signing the informed consent form and having the study discussed with them. The potential problem is with a diabetic. If a diabetic patient answers an ad and presents fasting for visit one, a research related procedure was done prior to signing the ICF (Informed Consent Form). They could get a low sugar reaction while driving to the site and suffer an MVA. They may also feel pressured to sign the ICF quickly as they know when the do so they will finally get something to eat. This can be most safely avoided by allowing subjects to sign the ICF on a pre-visit one or by writing protocols that don't have fasting labs as a visit one procedure. Although I understand V1 labs are non-fasting for this trial.
5. Recruitment: Too often recruitment is an oversight thought to be the sites problem dealt with at a time other than the beginning of the trial. It becomes the sponsor’s problem after many sites under enroll. Wise sponsors address recruitment up front at the investigator’s meeting and include recruitment agencies. One budget neutral solution is for the sponsor to push to make available an IRB approved ad for those using the central IRB at the very beginning of the trial. In addition use of centralized recruitment agencies like Pi and Acurian can very useful and cost effective. All our databases could be larger. A small minority of people step forward to become research subject volunteers. Planning for recruitment issues at the beginning is much better than catch up at the end.