CENTER FOR DRUG EVALUATION AND
`
`RESEARCH
`
`'
`
`APPLICA TION NUMBER:
`21-976
`
`MEDICAL REVIEW
`
`

`

`NDA 21—976 PrezisIaTM (darunavir)
`Deputy Office Director Memo
`
`I
`
`Page 1 of 7
`
`Deputy Office Director Memo
`
`Applicant: Tibotec, Inc.
`
`NDA #: 21—976
`
`Drug: darunavir, tablets, 300 mg
`
`Other names used during development: TMC114
`
`Trade Name: PREZISTATM
`
`Indication: PREZISTA, co-administered with 100 mg ritonavir (PREZISTA/rtv), and
`with other antiretroviral agents, is indicated for the treatment of human
`immunodeficiency virus (HIV) infection in antiretroviral treatment—
`experienced adult patients, such as those with HIV—1 strains resistant to
`more than one protease inhibitor.
`
`Dose: darunavir/ritonavir 600/100 mg twice daily
`
`Date of Submission: December 23, 2005
`
`Action Date: June 23, 2006
`
`Recommended Regulatory Action:
`
`Approval for NDA 21—976
`
`
`The review team has reviewed the issues in detail in their respective disciplines with
`regards to the safety and efficacy of PREZISTA (darunavir), co—administered With 100
`mg of ritonavir for the treatment of HIV infection in antiretroviral experienced adult
`patients. For a detailed discussion of NDA 21-976, the reader is referred to the individual
`discipline specific reviews.
`In addition Dr. Marcus’s Team Leader’s Memo and
`Dr. Murray’s Deputy Division Director’s Memo review key issues in the NDA
`submission.
`
`The chemistry for PREZISTA is discussed in Dr. Kambhampati CMC review and is
`found to be acceptable. The recommendation regarding CMC is for approval.
`
`The recommendation from Dr. Farrelly and Dr. Verma with regards to the pharm/tox
`studies is for approval from a pharm/tox standpoint. The approval of PREZISTA will
`
`

`

`NDA 21-976 PrezislaTM (darunavir)
`Deputy Office Director Memo
`
`Page 2 of 7
`
`include a phase 4 commitment to complete the ongoing carcinogenicity studies in mice
`and rats. -
`
`The clinical pharmacology of PREZISTA is discussed in Dr. Arya’s review. Darunavir is
`primarily metabolized by oxidative metabolism, and based upon findings from in vitro
`studies this is primarily by CYP3A. Darunavir is co—administered with 100 mg of
`ritonavir as a metabolic inhibitor. When darunavir is administered with ritonavir,
`darunavir was found to be 82% bioavailable. The product label provides a table with a
`listing of drug interactions based upon results of clinical studies or expected drug—drug
`interactions and information on dose and schedule for the listed medications, as
`appropriate. A list of contraindicated medications is also included in the PREZISTA
`label. As part of the phase 4 commitments, the Applicant will conduct drug interaction
`studies evaluating darunavir/11v b.i.d with each of the following medications: rifabutin,
`buprenorphine/naloxone, and carbamazepine.
`
`The microbiologic assessment of darunavir is discussed in Dr. Naeger’s microbiologist’s
`review. Darunavir is an HIV-1 protease inhibitor that inhibits the cleavage of the HIV
`encoded Gag—Pol polyproteins in infected cells. The mutations associated with. decreased
`susceptibility of HIV to darunavir have been characterized and are summarized in the
`product label.
`
`The results of the clinical trials evaluating the safety and efficacy of darunavir are
`discussed in detail in Dr. Gibbs’s Medical Officer’s Review, Dr. .Hammerstrom’s
`Statistical Review, and also in the reviews prepared by Dr. Marcus and Dr. Murray. The
`reader is referred to their reviews for a detailed discussion of safety and efficacy. In dose—
`finding studies 202 and 213, HIV positive patients who were at least 3—class experienced
`were randomized to either one of four doses of darunavir/11v plus optimized background
`regimen or control Which was ritonavir-boosted protease inhibitor(s) plus optimized
`background regimen. The mean viral load at the time of enrollment in the studies was
`approximately 4.5 loglo viral load copies/mL, the mean CD4+ cell count was
`approximately 300 cells/uL, and the mean number of years of duration of HIV infection
`was approximately 11 to 13 years. The primary endpoint in these studies is the
`proportion of subjects with a greater than one log drop in viral load at Week 24.
`studies, all darunavir treatment arms exhibited a statistically significant greater
`proportion of subjects achieving the primary endpoint compared with the control arm
`with a trend towards greater response in the patients receiving higher doses of darunavir
`up to the highest dose studied of 600 mg po BID. Additional analyses that evaluated
`response using the more stringent endpoints of proportion of subjects with viral load
`below 400 copies/mL and proportion of subjects with viral load below the limit of
`quantification of 50 copies/mL also corroborated these findings (please see Dr.
`Hammerstrom’s review for the tabulation. of these results). Evaluation of changes in
`CD4+ cell count demonstrated higher increases in patients receiving darunavir with a
`mean increase of 92 cell/uL compared to 17 cells/uL in control subjects. These data
`provide clear evidence of the effectiveness of darunavir in reducing HIV viral load and
`increasing CD4+ cell counts.
`'
`
`In both
`
`

`

`NDA 21—976 PrezistaTM (darunavir)
`Deputy Office Director Memo
`
`Page 3 of 7
`
`A total of 810 patients received the proposed dose of darunavir/rtv 600/100 mg for any
`length of time. Of these patients, 458 initiated therapy at the proposed dose and received
`it for a mean of 35.2 weeks. Darunavir was studied in a heavily treatment—experienced
`patientpopulation on multiple concomitant medications and with limited treatment
`options; hence patients with limited treatment may have chosen to remain on darunavir
`therapy even if they experienced adverse events.
`
`Darunavir appears to be associated with skin rash. The molecule contains a sulfa moiety
`and this may, in some cases, play a role in the observed rash events.
`In the clinical
`development program rash (regardless of causality and of any grade) was reported in 7%
`of subjects with treatment discontinuations in 0.3% of subjects. Most rashes were self—
`limited maculopapular skin eruptions of mild to moderate severity. However, there were
`also infrequent reports of more severe skin reactions including cases of erythema
`multiforme and Stevens-Johnson Syndrome. Some cases of rash have been accompanied
`by fever and elevations in transaminases. The PREZISTA label provides a WARNING
`about skin rash and also includes a section warning about sulfa allergy given that
`darunavir contains a sulfonamide moiety.
`
`The most common adverse events reported with darunavir/rtv included gastrointestinal
`adverse events. Darunavir/rtv also exhibits adverse events similar to other protease
`inhibitors including effects on serum lipids. In dose—finding studies, treatment emergent
`moderate elevations in amylase and lipase were reported approximately twice as
`frequently in patients receiving darunavir/rtv as compared to control subjects. There
`were also three cases of pancreatitis reported in the clinical development program, one of
`which was fatal. However, these patients were all receiving didanosine in combination
`with tenofovir, with didanosine being a drug associated with development of pancreatitis.
`
`Elevations in liver enzymes appeared similar in patients receiving darunavir and
`comparator with about one third experiencing any grade elevation in liver enzymes.
`Across the phase 2 studies there were eight patients receiving darunavir who experienced
`serious adverse events of liver dysfunction or liver enzyme elevations. Dr. John Senior
`was consulted for evaluation of the potential hepatic effects of darunavir. As noted in Dr.
`Marcus’s review, Dr. Senior concluded that “no clear evidence of darunavir-induced liver
`
`toxicity is apparent in the data accrued so far. A couple of cases suggested possible
`contribution by darunavir to liver injury in patients with pre-existing liver problems, but
`definite attribution of causality is difficult in these patients with prolonged complex
`illnesses and exposure to many drugs. There is no indication at this time for special
`labeling of darunavir as causing clear cut liver injury.”
`
`The occurrence of death in the darunavir clinical trials was carefully evaluated by the
`clinical and statistical reviewers. In the original application it was noted that the deaths
`in the phase 2 studies were all in the pooled darunavir arms. (Randomization in studies
`202 and 213 was 4:1 darunavir to control as there were four darunavir doses being
`studied.) In study 202 there were 11 deaths across the four darunavir arms (range of
`number of deaths per darunavir arm 2—4) and 0 deaths in the control arm. In the second
`study there were 6 deaths across the four darunavir arms.(range of number of deaths per
`
`

`

`NDA 21-976 PrezisIaTM (darunavir)
`Deputy Office Director Memo
`
`Page 4 of 7
`
`darunavir arm 0—3) and 0 deaths in control arm. There was no apparent trend in dose
`response for mortality across the range of doses studied. The issue of deaths in the
`studies was evaluated extensively by the review team. There was no clear pattern to the
`cause of deaths; they appear to be causes that occur in patients with advanced HIV/AIDS.
`There are several other factors that may be contributing to the observed differences in the
`number of deaths that were observed in Studies 20 and 213 as follows. Patients were
`
`randomized four to one to darunavir versus control arm in accordance with the study
`design. Patients were more likely to remain on darunavir longer; in his review Dr.
`Hammerstrom performs an analysis which adjusts for person years of exposure and the
`difference in death rates for each of the arms is not statistically significantly different
`from the control arm. The number of patient years in the control arms are lower because
`the study design allows patients who are not achieving satisfactory viral suppression to
`receive alternative therapy rather than remaining on a failing regimen — a proviso
`necessary for safety reasons. Many of these patients who were on control and failed to
`respond were allowed to roll over into another study in which they could receive
`darunavir. Another potential contributing factor is that four times as many patients
`randomized to control (20/ 144) as compared to any darunavir arm (17/530) dropped out
`of the study prior to receiving study drug, perhaps reflecting that in this open—label study
`patients randomized to control elected to drop out in order to enroll in other clinical trials.
`The observed mortality rates in the darunavir study arms are similar to the mortality rates
`(per 100 patient years) observed in the clinical trials for tipranavir and enfuvirtide, two
`recently approved agents targeting a similarly advanced population of HIV positive
`patients. What is out of the range of past experiences is the absence of deaths in the
`control arms in studies 202 and 213.
`
`To further evaluate the occurrence of deaths additional follow-up data was requested on
`patients enrolled in the studies up to the current time, including information on patients
`who were randomized but did not receive study drug and for subjects who discontinued
`for virologic failure but did not roll over to treatment with darunavir. These data were
`analyzed in an intent—to—treat analysis and also in an “as-treated” analysis for all patients.
`In the As—Treated Analysis, for all patients, the mortality rates were 4.0% (6/149) for
`control and 3.8% (3 8/992) for darunavir. For additional analyses, including the
`sensitivity analyses, the reader is referred to Dr. Hammerstom’s Review.
`
`Based upon all of the evaluations it appears that there are factors Other than drug effect
`that are likely contributing to the apparent differences in observed mortality rates
`reported in the original NDA for studies 202 and 213.
`
`The reader is also referred to Dr. Murray’s Review that summarizes some of the recent
`discussions at the Antiviral Drugs Advisory Committee meeting. As Dr. Murray notes,
`we are not seeing mortality differences in contemporary trials of antiretroviral agents at
`24—weeks in patients with advanced HIV/AIDS, likely because of efforts to have
`adequate safeguards in place and to ethically design clinical trials.
`
`As additional data become available from the ongoing studies of darunavir, it will be
`important to continue to carefully review the findings with regards to mortality.
`
`

`

`NDA 21—976 PrezisIaTM (darunavir)
`Deputy Office Director Memo
`
`Page 5 0f 7
`
`DMETS and DDMAC have consulted on the proprietary name and do not object to the
`use of the proprietary name PREZISTA. Comments from DMETS and DDMAC have
`also been incorporated into the product labeling. The DSI inspections have been
`performed and did not identify any significant observations that would compromise the
`integrity of the data. Pediatric studies required under PREA have been deferred as noted
`in the approval letter.
`
`Postmarketing Study Commitments
`
`.
`
`0
`
`Products approved under the accelerated approval regulations, 21 CFR 314.510,
`require further adequate and well-COntrolled studies to. verify and describe clinical
`benefit. The following are the postmarketing study commitments related to approval
`under Subpart H.
`
`1. By December 31, 2007, please submit final study reports and datasets for the 96—week
`data on the ongoing Phase 2b dose—finding studies TMCl 14-C202, TMC114-C213,
`TMC 114—C208 and TMC114-C215.
`
`2. By December 31, 2007, please submit final the study reports for the 48 week data on
`the ongoing Phase 3 studies TMC114—C211 and TMC114—C214
`
`0 Deferred pediatric studies required under section 2 of the Pediatric Research Equity
`Act (PREA)
`
`3. Deferred pediatric study under PREA for the treatment of HIV—1 infection in pediatric
`patients ages 6 to 17 years. Please assess the pharmacokinetics, safety, tolerability and
`antiviral activity in two alternative doses of a suitable pediatric formulation in
`combination with ritonavir, in treatment—experienced pediatric children and adolescents
`between 6 and 17 years of age.
`'
`
`Protocol Submission:
`Final Report Submission:
`
`' Completed
`by June 2008'
`
`4. Deferred pediatric study under PREA for the treatment of HIV—1 infection in pediatric
`patients ages less than 6 years. Please evaluate dose requirements and safety in pediatric
`patients <6 years of age with HIV-1 infection after preliminary review of data from the 6
`to 17 year old children in trial TMCl l4—C212 with the Division of Antiviral Products
`(DAVP).
`
`Protocol Submission:
`Final Report Submission:
`
`by December 2008
`by June 2011
`
`

`

`NDA 21—976 PrezislaTM {darzmavir)
`Deputy Office Director Memo
`
`Page 6 0f 7
`
`0 Other postmarketing study commitments
`
`Drug-Drug Interaction Trials
`
`5. Conduct an in vivodrug—drug interaction study between Darunavir/rtv b.i.d. and
`rifabutin.
`
`6. Conduct an in vivo drug-drug interaction study between Damnavir/rtv b.i.d. and
`buprenorphine/naloxone.
`
`7. Conduct an in vivo drug—drug interaction study between Darunavir/rtv b.i.d. and
`carbamazepine.
`
`Pharmacology/Toxicology
`
`8. Complete ongoing carcinogenicity study in mice and submit final report.
`
`9. Complete ongoing carcinogenicity study in rats and submit final report.
`
`Pharmacokinetics
`
`10. Please conduct a cocktail study to determine the effects of steady state
`Darunavir/rtv 600/100 mg b.i.d. on the metabolism of CYP450 probe substrates for
`the following enzymes: CYP2C9, CYP2C19, and CYP2D6.
`
`Special Populations
`
`11. Evaluate the pharmacokinetics of Darunavir/rtv in subjects with varying degrees of
`hepatic impairment in order to determine dosing recommendations.
`
`12. Conduct a study of darunavir in treatment—experienced female patients to
`elucidate any potential gender differences in efficacy and safety.
`
`Although not post marketing study commitments we have also requested the following
`information to be submitted:
`
`Drug—Drug Interaction Trials
`
`1. The following represent clinical drug-drug interaction studies that have been planned
`by Tibotec, Inc. to be conducted with darunavir. The Division acknowledges the
`following planned studies:
`’
`
`

`

`NDA 21-976 PrezisIaTM (darunavir)
`Deputy Office Director Memo
`
`_
`
`Page 7 of 7
`
`0 TMCl 14—C127: drug—drug interaction study between Darunavir/I’tv b.i.d. and
`methadone.
`
`.W,
`_
`
`Clinical
`
`2. In addition to the required periodic adverse drug experiencereports
`[21 CFR 314.80(c)(2)], please submit a separate periodic adverse drug experience report
`for rash.
`
`Microbiology
`
`3. Determine response rates based upon presence of specific cleavage site mutations at
`baseline and submit this analysis with the PREZISTA traditional approval application.
`
`4. Determine the'protease cleavage site mutations that occur most frequently (>10%) in
`virologic failure isolates and submit this analysis with the PREZISTA traditional
`approval application.
`
`5. Determine if the most frequently occurring protease cleavage site mutations
`contributed to decreases in darunavir susceptibility through site—directed mutagenesis and
`submit this analysis with the PREZISTA traditional approval application.
`
`.
`Summary
`I concur with the assessment of the review team that PREZISTA (darunavir) tablets, co—
`administered with 100 mg ritonavir (PREZISTA/rtv), and with other antiretroviral agents,
`is indicated for the treatment of human immunodeficiency virus (HIV) infection in
`antiretroviral treatment-experienced adult patients, such as those with HIV-1 strains
`resistant to more than one protease inhibitor, under the subpart H accelerated approval
`regulations for serious or life—threatening illnesses (21CFR §3l4.510). The clinical
`studies show a clear effect on viral load and CD4 cell count and the safety profile based
`upon the available data is acceptable. The product labeling adequately describes the
`available information on PREZISTA. Approval under Subpart H is appropriate for
`PREZISTA given that it may provide meaningful treatment benefit over existing
`antiretroviral treatment options based upon its activity against clinical isolates resistant to
`multiple protease inhibitors in the treatment of patients with HIV/AIDS. As part of
`approval under Subpart H, the applicant will study the drug further to verify and describe
`its clinical benefit as described in the postmarketing study commitments under Subpart H
`. listed above.
`
`

`

`This is a representation of an electronic record that was signed electronically and
`this page is the manifestation of the electronic signature.
`
`»
`Edward Cox'
`6/23/2006 03:45:31 PM
`MEDICAL OFFICER
`
`

`

`Team Leader’s Memorandum.
`
`NDA:
`
`21—976
`
`Drug and Indication:
`
`darunavir, co—administered with 100 mg ritonavir
`(darunavir/rtv), and with other antiretroviral agents, is
`indicated for the treatment of human immunodeficiency
`virus (HIV) infection in antiretroviral treatment—
`experienced adult patients, such as those with HIV—1 strains
`resistant to more than one protease inhibitor
`
`Proposed Dose:
`
`darunavir/ritonavir 600/100 mg twice daily
`
`Dosage Form:
`
`-
`
`300 mg film-coated tablet
`
`Letter Date:
`
`December 22, 2005
`
`Stamp Date:
`
`December 23, 2005
`
`Date of Memorandum:
`
`June 23, 2006
`
`Background
`Darunavir (DRV, tradename Prezista) is a novel HIV protease inhibitor under
`development by the applicant (Tibotec, Inc.) for the treatment of HIV infection. This
`New Drug Application (NDA) was submitted in accordance with regulations and
`guidance for submission of drugs for accelerated approval; demonstration of efficacy of
`this drug is based on surrogate endpoint analyses of plasma HIV RNA and CD4+ cell
`counts in antiretroviral heavily treatment-experienced HIV—infected subjects after 24
`weeks of treatment.
`
`The clinical development package submitted to support the efficacy of DRV consists
`primarily of data from 4 controlled and 2 uncontrolled clinical trials. Two of these studies
`were 4—week randomized, controlled open-label proof—of4concept studies that were
`conducted in treatment—experienced HIV-infected subjects currently failing their
`antiretroviral regimen. One study evaluated DRV administered alone, and the other study
`evaluated DRV co-administered with a low dose (100 mg) of another protease inhibitor,
`ritonavir (rtv). In this combination, DRV is the active antiretroviral and rtv serves as a
`pharmacologic enhancer by inhibiting the metabolism of DRV via the CYP3A system,
`thereby increasing DRV concentrations. Based on the efficacy results of these proof-of-
`concept studies, the applicant chose to further develop DRV in combination with low
`dose rtv.
`
`Two randomized, controlled, partially—blinded dose—finding Phase 2b trials of 3-class
`experienced HIV-infected subjects were then initiated to evaluate four different dosing
`regimens of DRV/rtv as compared to a control regimen containing one or two rtv—boosted
`protease inhibitors (PIS). All regimens were co—administered with an optimized
`
`

`

`background of nucleoside reverse transcriptase inhibitors (NRTIS) with or without
`enfuvirtide (ENF). Statistically significantly higher rates of HIV viral load reduction were
`observed with all dosing regimens of DRV/rtv versus the control arm. Although not
`statistically significant, efficacy increased numerically with increasing dose. As a result,
`DRV/rtv 600/100 mg twice daily was chosen as the proposed dose.
`
`Based on the robust efficacy demonstrated in these dose—finding studies, it was
`determined that further controlled clinical studies in highly treatment—experienced
`patients would not be required nor would they be ethical; however, additional safety data
`was required at the proposed dose. As a result, the applicant enrolled additional patients
`in an ongoing open—label rollover safety study. An additional 327 “de novo” patients were
`enrolled and started DRV/rtv 600/ 100 mg twice daily, resulting in a safety database of
`458 patients who initiated DRV/rtv at the proposed dose.
`I
`
`Dose Selection
`
`Two proof—of—concept dose—finding studies, studies TMC114—C201 (C201) and TMC114—
`C207 (C207), were first conducted in treatment-experienced patients failing their current
`antiretroviral (ARV) regimen. Patients substituted their current protease inhibitor (P1)
`with unboosted DRV in C20] and DRV/rtv in C207 for 14 days. Each study contained a
`control arm in which subjects continued their current regimen. Demographics, including
`baseline antiretroviral (ARV) resistance, viral load, CD4+ cell count, sex, age and CDC
`class, were generally similar between the two studies. About half of patients in each study
`had no susceptible PIS at baseline. In study C201, doses ranging fromDRV 400 mg twice
`daily to 800 mg three times daily resulted in reductions of viral loads from 0.28 to 0.79
`log] 0 copies/mL, respectively. In study C207, a dose of DRV/rtv 600/ 100 twice daily
`produced the largest observed reduction in viral load on either study of 1.38 loglO
`copies/mL.
`
`To further explore optimal dose selection, the applicant then initiated two large Phase 2b
`dose—finding studies of highly treatment—experienced patients with limited to no treatment
`options, TMCl 14—C202 (C202) and TMC114—C213 (C213). Each study was a
`randomized, active—controlled partially-blinded (to DRV dose only) study comparing the
`following 4 DRV/rtv treatment groups to a cOntrol group: DRV/rtv 400/100 mg once
`daily, 800/100 mg once daily, 400/100 mg twice daily, and 600/100 mg twice daily. A
`dynamic central randomization was applied using biased coin techniques, so that all
`allocations were done randomly. Three stratification factors were applied for
`randomization: screening plasma viral load, use of ENF in the 0BR and the number of
`primary Pl mutations.
`
`Subjects randomized to DRV/rtv treatment groups substituted their Pl(s) with DRV/Itv
`and continued their same background NRTIs and ENF for 2 weeks (functional
`- monotherapy phase). After the 2-week period, subjects continued their randomized
`DRV/rtv dose and changed the screening background regimen to an optimized
`background regimen (0BR) consisting of NRTIs with or without ENF. Subjects
`randomized to the control group changed their therapy at baseline to an investigator-
`
`

`

`selected PI(s) regimen plus an 0BR (NRTIs with or without ENF); these regimens were
`selected by the investigator prior to randomization.
`
`The proposed dose of DRV/rtv 600/100 mg twice daily was selected and confirmed by a.
`. Week 16 analysis of the first 150 patients enrolled into each study, and a second analysis
`when 200 patients in each study had reached Week 16. The dose—finding portion of each
`study ended when the primary Week 24 analysis was performed. Immediately prior to the
`cut—off date of the primary analysis (February 1, 2005), the proposed dose was
`communicated to investigators and all subjects randomized to DRV/rtv were instructed to
`switch to 600/100 mg twice daily.
`
`Because of their prior ARV experience, not all subjects had a PI to which their virus was
`susceptible. Therefore, subjects were considered early failures if they did not achieve a
`0.5 log decrease in HIV RNA by Week 12. Early failures were allowed to roll over to an
`open-label DRV study. Of note, a similar early failure definition was used in both the
`“TORO” studies that formed the basis of approval of the ENF NDA and the “RESIST”
`studies that formed the basis of approval of the tipranavir (TPV) NDA. These studies also
`enrolled advanced populations for whom the open-label control for some subjects was
`expected in advance to be potentially ineffective.
`
`Inclusion Criteria and Patient Demographics for Studies C202 and C213
`
`C202 and C213 Were conducted in different countries, although the two studies were
`otherwise identical in design. C202 enrolled patients at investigative sites in the United
`States and Argentina. C213 enrolled patients at sites in Austria, Belgium, France,
`Germany, Hungary, Italy, Portugal, Spain, Switzerland, United Kingdom, Canada, Brazil
`and Australia.
`
`Eligible HIV—l infected subjects were at least 3-class experienced, were on a stable PI-
`containing regimen at screening for at least 8 weeks, and had a plasma HIV—1 RNA >
`1000 copies/mL. Three—class experience was defined as prior treatment with 2 or more
`NRTIs for at least 3 months in total and l or more NNRTIs as part of a failing regimen.
`In addition, all subjects had received at least 1 PI for at least 3 months in the past and had
`at least 1 primary PI mutation at screening (according to the IAS—USA list of March
`2003). Prior use of ENF was allowed.
`'
`
`Demographics of patients enrolled in each study are summarizedin the two following
`tables obtained from the Summary of Clinical Efficacy. Of note, just 70/596 (12%)
`enrolled subjects were female. Similar percentages of female patients were enrolled in the
`TORO and RESIST studies. In addition, just 44/596 (7.4%) patients were co—infected
`with hepatitis B or C. Low enrollment was likely due to the fact that co-infected patients
`were excluded from participation in C202 due to concerns about hepatotoxicity early in
`the development program.
`
`

`

`Table 17: Demographic Data in DICTII4—C‘202 and D—IC114—C213
`
`mic-mm
`
`.
`
`meme-m
`
`'
`
`1
`
`1
`Q
`
`‘
`
`I“.
`? '
`«
`
`.
`
`:‘
`
`:'
`
`,
`
`f
`
`.
`
`"
`
`’
`
`318
`
`34 (10.?)
`25-6 (80.5)
`12 {3.8}
`3 (O. 9::
`
`Gender, 11 (0’0)
`
`N F
`
`emale
`Male
`
`N M
`
`N
`
`ean (SD)
`
`Black
`Caucasians'white
`Hispanic
`-Oriental.-’Asian
`
`N =munber of subjects Wlill data: 11 = “subjects in class
`
`Source: Summary of Clinical Efficacy, p. 52, NDA 21976
`
`Table 18: Baseline Disease Cimmcteristics'Ill TNIC 114 C202 and T3vICll4— C213
`
`TRICl'l-l—CZOZ
`
`T31C114—C‘213
`
`Logm \1':31 load (copiess'mL)
`
`N M
`
`ean {SD}:
`
`Other
`
` Mean (SD?
`
`CD4+ cell count (106.51.)
`
`N M
`
`edian {Range}
`
`N M
`
`ean (SD)
`
`N = number of subjects with data; 11 = Isubjects in 6km
`Source: Summary of Clinical Efficacy, p. 53 NDA 21976
`
`

`

`,
`Comparator P15 and 0BR
`Determination of the number subjects with susceptible drugs at baseline was based on
`phenotypic data obtained from the Antivirograme assay. Overall, 71% of the subjects in
`C202 and 63% of subjects in C213 were infected with virus resistant to all available Pls.
`The proportion of subjects susceptible to at least one NNRTI was 49% in C202 and 46%
`in C213; however, this is likely related to the fact that 84% of the subjects in C213 were
`not using an NNRTI in their screening regimen and for C202 the use of an NNRTI during
`the screening period was disallowed. The use of NNRTIs in the OBR was not allowed in
`both trials.
`
`The choice of PI regimen in the control group was similar in both trials. Overall, almost
`all Pl regimens (98%) were co—administered with low—dose ritonavir as a pharmacologic
`enhancer, of which 75% were single boosted and 23% double boosted. No subject used
`an unboosted Pl—containing regimen. The most frequently used P1 was lopinavir (38%)
`followed by saquinavir (35%), amprenavir (33%) and atazanavir (17%). Two subjects
`used indinavir.
`
`The proportion of subjects susceptible to at least ] NRTI was 94% for C202 and 97% for
`C213. More. subjects in C213 received at least one NRTI to which they were susceptible
`(75%) as compared to C202 (65%). For both trials, the most susceptible NRTIs were
`stavudine (d4T, 84%), didanosine (ddl, 80%), and zalcitabine (ddC, 75%); however, in
`the OBR, ddI was used by only 37%, d4T by 16%, and ddC'by 0.7% of subjects. In
`contrast, the least susceptible NRTIs were tenofovir (TDF) and lamivudine (3TC), yet
`they were the most commonly used NRTls in the 0BR; TDF was used by 86% of
`subjects and 3TC by 59%.
`
`Overall, 47% of subjects used ENF in the OBR. ENF was more frequently included as
`part of the OBR in C202 than in C213 (55% versus 45%, respectively). A total of 11%‘of
`subjects had used ENF before and 36% of subjects used ENF for the first time. The
`percentage of subjects that used ENF for the first time during the trials was similar for
`both trials.
`
`The optimization of and compliance with the background regimen for the control subjects
`is supported by the observation of the applicant that 38 of the 59 subjects (64%) in the
`control group that discontinued due to virologic failure initially showed a drop in viral
`load of at least 0.5 logio. In addition, monitoring of the serum levels of Pls demonstrated
`that the investigator—selected PI could be detected on average in 90% of control subjects.
`
`Efficacy Analyses
`
`The following section highlights the major findings of the statistical review of this NDA
`by Dr. Thomas Hammerstrom.
`
`The specified primary endpoint of both studies was the proportions of subjects with a
`greater than 1 log drop in viral load at Week 24. In addition to analyses of this primary
`endpoint, analyses were performed on the more stringent endpoints of proportions of
`
`

`

`subjects with viral load below 400 copies/mL, and proportions of subjects with viral load
`below the limit of quantification of 50 copies/mL. Change in CD4+ cell count was also
`evaluated.
`
`In summary, all four doses of DRV/rtv were statistically significantly superior to the
`control arm. These findings remained significant after performance of multiple sensitivity
`analyses. The following tables summarize results of analyses of the primary endpoint,
`proportion of subjects with greater than 1 log drop in viral load at Week 24. In addition,
`analysis of the most stringent endpoint of reduction in viral load, the proportion of
`subjects with viral load <50 copies/mL, is also reported.
`
`This FDA review defines response rates by the “Time to Loss of Virologic Response
`(TLOVR)”. The TLOVR analysis is an intent-to-treat analysis that examines endpoints
`using the following definitions of treatment failure for patients who have achieved HIV
`RNA levels below the limit of quantification:
`
`For all subjects with confirmed HIV RNA levels below an assay limit, the time to failure
`is the earliest time when a specific event had occurred. These events are
`0 Death
`
`0
`é
`
`Permanent discontinuation of the study drug or loss to follow—up
`Introduction of a new antiretroviral drug (unless a background drug is changed
`for reasons of toxicity or intolerance that are clearly attributable to that drug)
`. Confirmed HIV RNA levels above or equal to an assay limit
`
`PERCENTAGE OF SUBJECTS WITH 21 LOG DROP AT WK 24, TRIAL C202
`
`Treatment
`Arm
`
`Rates of Suppression
`Mean 95%_Limits
`Diff Lower Upper DRV/r
`Control
`
`P—value
`
`400 QD
`800 QD
`400 BID
`
`600 BID
`
`34.6% 19.8% 49.4% 31/65=48% 8/61=13%
`38.4% 23.6% 53.3% 33/64=52% 8/61=13%
`47.2% 32.4% 62.0% 38/63=60% 8/61=13%
`
`50.5% 36.2% 64.9% 42/66=64% 8/61=13%
`
`<.0001
`<.0001
`<.0001
`
`<.0001
`
`Source: Statistical review of NDA 21976, by Dr. ThomasHammerstrom
`
`PERCENTAGE OF SUBJECTS WITH 21 LOG DROP AT WK 24, TRIAL C213
`
`Treatment
`Arm
`
`400 QD
`800 OD
`400 BID
`600 BID
`
`Mean
`Diff
`
`Rates of Suppression
`95%_Limits
`Lower Upper DRV/r
`Control
`
`P-value
`
`41.7% 25.9% 57.5% 45/64=70% 18/63=29% <.0001
`429% 27.1% 58.6% 45/63=71% 18/63=29% <.0001
`42.9% 27.1% 58.6% 45/63=71% 18/63=29% <.0001