`

`

`

`Cross Discipline Team Leader Review
`
`76,480 for Cushing's syndrome, the Korlym pharmacology/toxicology and clinical
`pharmacology programs had been quite extensive, and human exposure across a variety of
`indications and doses exceeded 1100 subjects/patients. The IND was in fact opened with what
`was planned to be the registration clinical trial (Study C 1 073-400), the results of which are
`submitted in the current NDA. This study was planned as a 24-week, open-label, uncontrolled
`(single-arm), Phase 3 clinical trial to be conducted in 50 patients with Cushing's syndrome and
`glucose intolerance or diabetes (29 patients in the end) and hypertension (21 patients).
`Cushing's disease is a rare disease and Corcept has received orphan indication on July 5, 2007
`for the "treatment of clinical manifestations of endogenous Cushing's syndrome". At the time
`when the IND was opened, DMEP provided answers to a series of questions submitted by the
`sponsor regarding the development program for the Cushing's syndrome indication. In
`summary:
`• The Division agreed that the toxicology studies conducted up to that time, along with
`ongoing carcinogenicity studies, would be sufficient for a Cushing's syndrome
`indication.
`• Given the rarity of the disease and the ethical issues raised by conducting a placebo
`controlled trial in a condition of such severity when there is preliminary evidence of
`efficacy from published reports, submission of a single Phase 3 clinical study using a
`single-arm, open-label design was found to be acceptable for an NDA submission.
`• The Division provided advice regarding efficacy endpoints selected to be evaluated in
`the pivotal study, and specifically indicated that, due to the fact that cortisol levels
`cannot be used as a measure of efficacy in the case of mifepristone, the primary
`efficacy endpoints should be clinical, i.e. change in blood pressure and/or glycemic
`control. The Division also advised to use of area under the time vs. concentration curve
`for glucose during an oral glucose tolerance test as a study endpoint. In the end it was
`selected as one of the two primary endpoints.
`• Following review of the dosing information accumulated in healthy volunteers and
`across various patient populations studied under mifepristone INDs, the Division
`recommended that Korlym doses should not exceed a maximum of20 mg/kg/day.
`• The Division also advised that every woman with an intact uterus undergo transvaginal
`ultrasound at baseline and completion of the study.
`
`In subsequent correspondence (September, 2, 2009) the Division added a request to obtain
`baseline and end-of-trial endometrial biopsy in order to evaluate the proliferative effect of the
`drug on the endometrium. In doing so, the Division decided that, given the severity of the
`condition being studied and the lack of an approved therapy, the potential effects of
`mifepristone on reproduction should be studied in the registration clinical trial rather than in a
`dedicated reproductive study that will slow the Korlym clinical program.
`
`In a communication dated March 3, 2010, DMEP asked the sponsor to add baseline and end(cid:173)
`of-trial ophthalmological exams to the safety evaluations of the pivotal trial. This request was
`triggered by the observation of retinal atrophy in the preclinical program in a single animal
`species (Sprague Dawley rats) that was not confirmed in a second species (i.e. not observed in
`mouse or dog). As was the case with endometrial biopsies, this request and the subsequent
`implementation in the clinical trial were made while the trial was in progress.
`
`Page 20f30
`
`Reference ID: 3086567
`
`2
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2055, Page 3
`
`

`

`

`

`Cross Discipline Team Leader Review
`
`Korlym has received orphan designation. A disease of adult age mostly (peak incidence is
`between 25 and 40 years of age) that affects women more than men (8-fold higher rate of
`pituitary tumor and 5-fold higher rate of cortisol-secreting adrenal tumors in women with
`Cushing syndrome than in men), Cushing's syndrome, ifleft untreated, has an extremely poor
`prognosis: a mean duration from presentation to death of 4.7 years in Cushing's original series
`and a mortality rate that is 5-fold higher than that of age and gender-matched sUbjects2
`•
`Medical treatment of Cushing's syndrome is secondary to surgical management that can be
`curative. However, a significant proportion of patients is not cured by surgery. For instance,
`remission rates following initial surgery for Cushing's disease due to micro adenomas of the
`pituitary are between 70-90% and smaller (50-65%) if due to macroadenomas3
`. Patients who
`fail surgery have several therapeutic options that include repeat surgery at the original site
`(pituitary or ectopic) or removal of the adrenals, radiotherapy, or medical therapy. Medical
`therapy may also be used rarely in some patients who are not candidates for surgery or
`radiotherapy (including patients with metastatic disease), or when immediate control of the
`hypercortisolemia is required prior to surgery due to the severity of the disease.
`
`Currently there are no approved medical therapies for Cushing's syndrome. Several drugs that
`reduce cortisol secretion (adrenal-directed therapy) are used offlabel in clinical practice
`(metyrapone, etomidate, ketoconazole) with the goal of reduction or normalization of cortisol
`secretion. In many respects mifepristone stands alone in the context of the above-mentioned
`steroidogenesis inhibitors, because it does not reduce cortisol synthesis. Rather, mifepristone
`reduces the biological effects of the existing endogenous cortisol by competing effectively
`with it for binding to the type II nuclear glucocorticoid receptors for which mifepristone has an
`18-fold higher affinity than cortisol.
`
`The fact that the potential efficacy of mifepristone cannot be measured by quantifying
`endogenous cortisol secretion (e.g. measuring urinary free cortisol, an important efficacy
`measure in clinical practice and an equally relevant endpoint in clinical trials) has had direct
`consequences in the way clinical trials with Korlym have been planned and conducted. In
`absence of any other qualified biomarkers of disease improvement, the applicant had to select
`clinical endpoints. Of the many clinical manifestations of Cushing's disease (glucose
`intolerance and diabetes, hypertension, obesity, myopathy, bone loss, decreased quality of life,
`gonadal dysfunction, dermatological changes, compromised immune function, psychiatric
`symptoms, and fluid and electrolyte disturbances) glucose intolerance and diabetes, on one
`hand, and hypertension, on the other hand, were selected as primary measures of efficacy for
`the Korlym phase 3 clinical trial.
`
`Thus, central to this application is whether the Korlym clinical program has provided
`substantial evidence of effectiveness in adults with Cushing syndrome who have not
`adequately responded to surgical treatment. This determination is a particularly challenging
`task for a variety of reasons. First and foremost, Korlym has been studied in a single-arm
`clinical trial with no comparator; since there are no approved medical therapies for this
`
`2 Etxabe J, Vazquez JA 1994 Morbidity and mortality in Cushing's disease: an epidemiological approach. Clin
`Endocrinol (Oxf) 40:479-484
`3 Blevins LS et al. An approach to the management of patients with residual Cushing's disease. J Neurooncol
`(2009) 94; 313-319.
`
`Page 4 0[30
`
`Reference ID: 3086567
`
`4
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2055, Page 5
`
`

`

`Cross Discipline Team Leader Review
`
`indication to compare with, and since the use of a placebo ann in a disease of such severity
`would be unethical. Such a design in which patients serve as their own control is acceptable
`because the disease is not expected to show spontaneous improvement, and favorable changes
`observed at the end of the trial relative to baseline can be presumed to be treatment-related if
`there are no major confounders. Even so, additional challenges in intelpreting the l'esults of
`the Phase 3 clinical trial are posed by the small size of the study (50 patients), the confounding
`effects of concomitant medications such as antihypertensive dmgs and glucose lowering
`medications that may interfere with the metrics of the primary efficacy variables, the expected
`biochemical and clinical variability of the patients enrolled, their relatively diverse surgical,
`radiological and medical history, and the heterogeneity in Cushing's syndrome etiology
`(although most patients having pituitary disease, a few had ectopic forms of Cushing's
`syndrome), Last but not least, methodological shortcomings appear to have been generated in
`one of the primalY analysis cohorts, the "hypeliension" cohort, which enrolled patients on the
`basis of elevated systolic and/or diastolic blood pressure, while diastolic blood pressure only
`was selected as primruy efficacy variable. This memorandum will thus focus on the
`aforementioned limitations and on the interpretability of the Phase 3 program results with
`respect to efficacy, in addition to discussing the standard safety fmdings.
`
`3. CMC/Device
`
`The CMC review (DARRTS 111212012 and 1117/2012) recommends approval oftms
`application. There are no recommendations for Phase 4 studies. The OUice of Compliance has
`issued on Janumy 12,2012 an «acceptable" recommendation for the manufacturing facilities.
`
`mifepristone, has a molecular weight of 429.60 g/mol. It is a
`made to be structurally similar to progesterone and
`as immediate-release tablets and formulated to
`
`closure
`
`The CMC review indicates that Corcept has authorization to the relevant D1v1Fs for the dmg
`substance, ingredients, container-closure system, and mallufacturing processes. The dmg
`specificatiolls were reviewed and found to be acceptable. Impurities and degradants met ICH
`
`50f30
`
`Reference ID: 3086567
`
`5
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2055, Page 6
`
`

`

`Cross Discipline Team Leader Review
`
`requirements. Based on the stability data submitted, an expiry of 24 months is granted at room
`temperature.
`
`A categorical exclusion from an Environmental Assessment was granted.
`
`4. Nonclinical Pharmacology/Toxicology
`
`The nonclinical pharmacology/toxicology reviews (DARRTS, 1120/2012 and 2/7/2011)
`recommend approval. They do not make any recommendations for additional studies.
`
`The applicant submitted in this NDA a 12-month toxicology study in dogs and a 2-year
`carcinogenicity study in the mice and rats. Both studies were conducted at the request of the
`Agency and were in addition to a wide range of preclinical studies previously submitted under
`IND 76,480. The applicant did not submit any reproductive and developmental toxicology
`studies of their own; instead they are referencing the non clinical fertility and genotoxicity data
`in the Mifeprex® label (Danco Laboratories). Reliance on Danco's data was found to be
`scientifically valid by the toxicology/pharmacology team on the basis ofthe demonstration
`that the active ingredient in Korlym is mifepristone and that the pharmacological/toxicological
`studies conducted with Korlym demonstrate pharmacodynamic effects consistent with the
`known effects of mifepristone.
`
`Relevant safety findings of the pharmacology/toxicology program (other then those related to
`the known pharmacodynamic effect of the drug) include liver toxicity4, thyroid tumors5
`, retinal
`atrophy, and QT prolongation. The liver and thyroid findings have been attributed to
`induction of enzyme activity in the liver (mainly CYP3A) that results in hyperplasia and
`eventually neoplasia, as well as increased thyroid hormone metabolism in the liver. In clinical
`studies LFT elevation has not been a problem (there were no cases that met the definition of
`Hy's law, and LFT outlier values were rare, transient, of moderate magnitude, and resolved
`without intervention or could be explained by underlying liver disease. Thyroid laboratory
`changes have been minimal (transient TSH elevations with minimal changes in free T4) and
`are all monitorable (See also Section 8 for clinical discussion). The retinal atrophy (single
`species observation) and the QTc prolongation observed in dogs will also be discussed in
`Section 8 of this memorandum.
`
`5. Clinical Pharmacology/Biopharmaceutics
`
`The clinical pharmacology review (DARRTS 1113/20012) finds the data submitted in the
`NDA acceptable. However, due to the fact that mifepristone, the active ingredient in Korlym,
`
`4 Elevated liver function tests (LFTs), hepatocellular hypertrophy (especially in rat), hepatocellular toxicity in
`mouse at doses of about 5X clinical exposure; hepatocellular adenomas (rat-specific).
`5 Thyroid follicular cell adenomas, carcinomas in female rats.
`
`Page 6 of30
`
`Reference ID: 3086567
`
`6
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2055, Page 7
`
`

`

`

`

`Cross Discipline Team Leader Review
`
`No justification for a body weight cut-off of 60 kg was found despite the use of such a dosing
`threshold by the applicant in the Phase 3 clinical trials.
`
`Food effect studies conducted with 1200 mg single dose, 1200 mg multiple doses, and 600 mg
`single dose indicate that food increases Cmax and AUC ofmifepristone (e.g. an increase of
`19% in Cmax and 29% in AUC following a single dose of600 mg ofmifepristone).
`Mifepristone is recommended to be taken with food.
`
`Results from a study conducted in patients with severe renal impairment showed no significant
`change in the pharmacokinetics ofmifepristone following administration of 1200 mg of
`mifepristone.
`
`Finally, based on the known metabolism ofmifepristone and the drug-drug interaction studies
`conducted, the clinical pharmacology reviewer makes the following recommendations:
`No dose adjustment is necessary for patients with hepatic impairment, but doses in
`excess of 600 mg should not be used in patients with moderate hepatic impairment, and
`the drug should not be used in patients with severe impairment.
`Concomitant use of Korlym with simvastatin or lovastatin should be contraindicated.
`When given concomitantly with Korlym, substrates of CYP2C8/9 should be used at the
`smallest recommended doses and closely monitored for adverse effects.
`• Other oral drugs with CYP3A mediated metabolism may need the lowest or a reduced
`dose when used with Korlym.
`Concomitant use of strong inhibitors of CYP3A is contraindicated.
`Mild to moderate inhibitors ofCYP3A do not require dose adjustment ofKorlym.
`Use of moderate inhibitors ofCYP3A4 should be avoided.
`Use ofmidazolam should be contraindicated.
`Use of CYP2B6 substrates should be avoided.
`
`6. Clinical Microbiology
`
`Not applicable.
`
`7. Clinical/Statistical .. Efficacy
`
`The main source of clinical data for the Cushing's syndrome indication is Study C 1 073-400
`(further referred to as Study 400), a 24-week, single-arm, multicenter (17 US sites) open-label
`study conducted in 50 adult patients with Cushing's syndrome. Patients who completed the
`study and benefited from Korlym were allowed participation in an extension trial, Study
`C 1 073-415 (further referred to as Study 415) following a 6-week period of observation off
`mifepristone. Study 415 is ongoing.
`
`Page 8 0[30
`
`Reference ID: 3086567
`
`8
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2055, Page 9
`
`

`

`Cross Discipline Team Leader Review
`
`Study 400 enrolled adult patients with a diagnosis of endogenous Cushing's syndrome who
`, and who had clinical evidence ofhypercortisolemia7, biochemical evidence of cortisol excess8
`
`
`required medical treatment in the opinion of the investigator. There was no specific metric
`associated with the clinical determination that a patient was a candidate for medical treatment
`or not; instead, not unlike a clinical practice scenario, the decision appears to have been based
`on the best judgment of the caring physician/investigator9
`. Not surprisingly, given the
`heterogeneity of Cushing's syndrome in general, the etiology of Cushing's syndrome among
`the patients enrolled in Study 400 was quite diverse, and included Cushing's disease (most
`patients), ectopic sources of ACTH, or excess of cortisol secretion due to adrenal sources
`(adrenal carcinoma).
`
`Mifepristone treatment was started in all patients with a dose of 300 mg once a day. This was
`followed by dose escalations of 300 mg at a time, which were done at several weeks interval in
`absence of clinical improvement, provided that the drug was well tolerated. The highest
`allowed doses were 900 mg for patients with weights < 60 kg (this dose could be reached as
`early as Week 6) and 1200 mg for those with weights> 60 kg (this was achieved as early as
`Week 10). No patients received doses in excess of 1200 mg (although the protocol allowed
`exceptions under certain circumstances) and no dose was allowed to exceed 20 mg/kg/day.
`Dose reductions were also permitted for safety or poor tolerability. If Korlym treatment
`resulted in oversuppression of cortisol activity and evidence of adrenal insufficiency (an event
`anticipated based on prior experience with this drug and its known mechanism of action)
`mifepristone was to be interrupted for 2: 7 days (given the long half-life of the drug) and
`treatment with exogenous corticosteroid was to be initiated; if Korlym treatment was re(cid:173)
`initiated it had to be done with the initial starting dose of 300 mg.
`
`Two inclusion criteria deserve special attention because they are related in a fundamental way
`to the clinical endpoints selected for the primary efficacy analysis (glucose control and
`reduction in diastolic blood pressure). As designed, patients were to be enrolled in Study 400
`in two distinct cohorts: a cohort of patients who had either glucose intolerance or type 2
`diabetes (from now on referred to, for simplicity, as the "diabetes" cohort because 83% of
`patients in this cohort had diabetes) and a cohort of patients with hypertension (from now on
`referred to as the "hypertension" cohort). Please note that there was no randomization to any
`of these two groups, and patients were assigned to each of them simply according to pre(cid:173)
`specified baseline characteristics. Of the 50 patients enrolled, 29 were assigned to the diabetes
`cohort and 21 to the hypertension cohort. Each cohort had a different pre specified primary
`
`7 Patients had to have two or more of the following signs and symptoms: Cushingoid appearance (moon facies,
`dorsal-cervical fat pad, plethora), increased body weight or central obesity, proximal muscle weakness, low bone
`mass (DXA T-score < -1.0), psychiatric symptoms (including depression or psychosis), hirsutism and/or
`violaceous striae and/or acne.
`S This consisted in above-normal urinary free cortisol (UFC) on at least two complete 24-hour urine collections
`collected within 4 months of the baseline visit with at least one done during the screening period which was 6
`weeks in length (all UFC collections were accompanied by concomitant measuring of urine creatinine to ensure
`appropriate collection of the urine specimens).
`9 The applicant specifically states that "the study enrolled subjects for whom the investigator had determined that medical
`treatment of endogenous hypercortisolemia was needed. Medical treatment could have been intended to treat the effects of
`persistent or recurrent hypercortisolemia after surgery and/or radiation for Cushing's disease, to bridge the period of time for
`radiation to become effective, when surgery was not feasible, or in subjects with non-pituitary based ACTH-dependent and
`ACTH-independent disorders" (CSR C1073-400, page 21).
`
`Page 9 of 30
`
`Reference ID: 3086567
`
`9
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2055, Page 10
`
`

`

`Cross Discipline Team Leader Review
`
`efficacy endpoint and primary efficacy analysis. As such, Study 400 could didactically be
`seen as two different small trials, one conducted in patients with glucose intolerance or
`diabetes and the other in patients with hypertension. In these two "trials", the primary efficacy
`assessments were different but the safety evaluations were identical.
`
`To be enrolled in the diabetes cohort, patients had to show evidence of glucose intolerance!O or
`diabetes!! based on the results of a baseline oral glucose tolerance test or based on standard
`diabetes diagnostic criteria. Of note, there criteria were not applied to those patients who had
`previously been diagnosed with diabetes mellitus and were on anti-hyperglycemic medications
`prior to study initiation (i.e. these patients were allowed enrollment on the basis of prior
`diagnosis of diabetes). Such patients, however, had to be on stable doses of antidiabetic
`medications prior to enrollment and no changes in antidiabetic medication(s) were allowed
`from 2 weeks prior to the initiation of the study drug through its completion12. No new
`antihyperglycemic medications were allowed once enrolled, and patients who required and
`received additional antihyperglycemic medications during the study were to be discontinued.
`The primary efficacy analysis for the diabetes cohort was a responder analysis; a responder
`was defined as a patient who had a 2: 25% reduction in the serum glucose area under the time
`vs. concentration curve at end of trial relative to baseline following an oral glucose tolerance
`(oGTT) test.
`
`The blood pressure criterion required for assignment to the hypertension group included a
`systolic blood pressure 2: 140 mmHg and/or a diastolic blood pressure> 90 mmHg; patients
`were also allowed enrollment if they had a history of hypertension caused by or aggravated by
`hypercortisolemia and were currently receiving antihypertensive medication(s); such patients
`had to be on stable doses of antihypertensive medications, and no changes were allowed for 4
`weeks prior to enrollment. No new antihypertensive medications were allowed during the
`trial; if used, patients were to be discontinued13
`. The primary efficacy analysis for the
`hypertension cohort was also a responder analysis, a responder being a patient that experienced
`a 2: 5 mm reduction in diastolic blood pressure. It should be mentioned from the start (and this
`issue will be further expanded with the actual presentation of the clinical efficacy results) that
`there is some degree of disconnect between the blood pressure inclusion criterion and the
`primary efficacy analysis for this cohort in that the above-described inclusion criterion allowed
`patients to be enrolled with normal diastolic blood pressure, and in fact the mean diastolic
`blood pressure at baseline (82.9± 11.42 mm Hg) was within the normal range.
`
`10 Evidence of impaired glucose tolerance was to consist in a plasma glucose level between 140 mg/dL and 199
`mg/dL after a 2-hour 75-gram oral glucose load.
`11 Diabetes mellitus had to be confirmed by a fasting plasma glucose 2: 126 mg on two measurements OR a 2-
`hour plasma glucose level 2: 200 mg/dL after a 75-gram oral glucose load.
`12 19/29 (65.5%) ofthe patients in the diabetes group fell in this category and used glucose lowering drugs during
`the trial, including exenatide, sulphonylureas, metformin, DPP-IV inhibitors, or various insulins.
`Thiazolidinediones were not allowed during the study and for 4 months prior to enrollment.
`13 10/21 (47.6%) patients used diuretics, l3/21 (6l.9%) used drugs that acted on the renin-angiotensin system,
`7/21 (33.3%) used beta blocking agents, and 6/21 (28.6%) used calcium channel blockers.
`
`Page 100[30
`
`Reference ID: 3086567
`
`10
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2055, Page 11
`
`

`

`Cross Discipline Team Leader Review
`
`Of note, patients assigned to the diabetes/glucose intolerance cohort were allowed to have
`concomitant hypertensionl4, but patients assigned to the hypertension group could not have
`diabetes/glucose intolerance. The management of hypertension was different in the diabetes
`group in that patients were allowed changes in antihypertensive mediations and initiation of
`new antihypertensive drugs in accordance to standard of care.
`
`The patients' characteristics at enrollment in Study 400 were generally consistent with those of
`patients with Cushing's syndrome seen routinely in clinical practice. The mean age at baseline
`was of 45.4 years (range 26 to 71 years). The mean BMI of 35.7 kg/m2 was consistent with an
`obese population (range 24.1 to 66.4); mean weight was 99.5 kg (range 61.3 to 198.7 kg), and
`the mean waist circumference was 119 cm (range 88.5, 178.4); 35/50 (70%) of all patients
`were female, consistent with the higher prevalence of Cushing's disease in women, and 42/50
`(84%) of patients were white. From a pure clinical perspective, 98% of patients had a
`Cushingoid phenotype, 60% had evidence of hirsutism and skin manifestations such as
`violaceous striae or acne, 98% had central obesity or were overweight, 54% had proximal
`muscle weakness, 53% had evidence of psychiatric symptoms (psychosis, depression), and
`26% had evidence of low bone mass (T-score < -1.0 by DEXA).
`
`With respect to the underlying diagnosis of Cushing's syndrome, 43 (86%) patients were
`described as having Cushing's disease, 3(6%) patients had cortisol excess due to adrenal
`carcinomas, and 4 (8%) patients had an ectopic source of ACTH. For the 43 patients with
`Cushing disease medical treatment was initiated because of persistence of signs/symptoms of
`hypercorticolism despite pituitary surgery (i.e. failed pituitary surgery) in 30 (70%) patients,
`disease recurrence after primary pituitary surgery in 21 (49%) patients, and not being
`candidates or treatable with surgery in 6 (14%); only one patient had a de novo Cushing's
`disease diagnosis l5; N.B. patients could have more than one reason for being eligible for
`medical treatment. Of interest, 18 (42%) patients had received radiation to the pituitaryl6. In
`retrospect, it would have been desirable to avoid enrolling such patients in an uncontrolled trial
`because of possible interference with the progression of the disease (improvement of pituitary
`function following radiation therapy can occur and continue to occur over long periods of
`time). The significance of this finding will be addressed when the primary analysis results will
`be discussed.
`
`Several patients received medical therapy in the past, but all patients enrolled were naIve to
`mifepristone therapy17. The applicant indicates that 18 patients were entirely naIve to medical
`therapy, at least 23 patients received prior medical therapy and 13 of them failed medical
`
`14 79 % patients were also hypertensive in the diabetes group. Mean systolic pressure at screening was 138
`mmHg (range 100-181 mmHg) and mean diastolic pressure was 86.4 mm Hg (range 64-116 mm Hg)
`15 Table 14.1.4.1 CSR for Study 400
`16 The duration between the initial radiotherapy and the beginning of the trial was as short as 25, 23 days, several
`months (7 and 9) but mostly over 1 year and as long as 7 years.
`17 The Division asked the applicant to present more detailed baseline information regarding previous treatments
`for Cushing's syndrome. In their response the applicant indicated that Study 400 did not enroll patients on the
`basis of having used or failed prior medical treatments, but rather on the basis of the global assessment of the
`physician whether medical treatment was indicated, and detailed prior treatment data were available only in 44/50
`patients. As such a precise chronology of prior therapeutic interventions (surgical, radiotherapeutic or medical)
`and for each patient could not be reconstructed and the data presented here is a best estimate.
`
`Page 11 of 30
`
`Reference 10: 3086567
`
`11
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2055, Page 12
`
`

`

`Cross Discipline Team Leader Review
`
`therapy with a variety of agents (mostly ketoconazole, only a few used cabergoline, octreotide,
`and metyrapone or combinations thereof). Patients discontinued prior medical therapi(es) at
`least 30 days prior to beginning of the trial with some for more than 3 months. All three
`patients with adrenal cortical carcinoma continued mitotane during the trial (as allowed in the
`protocol). A carryover effect of previous medications is unlikely because they were
`discontinued ahead of the trial and many have short half-lives (for instance, ketoconazole, by
`far the most widely used has a half- life of 2-8 hours.and cortisol elevations follow promptly
`after being discontinued).
`
`Compliance with Korlym in Study 400 was generally good. It reached 92% in the diabetes
`cohort for the primary efficacy analysis population, 71.4 % in the hypertension cohort for the
`primary efficacy analysis population and 100% in completers regardless of cohort. The
`completion rate was only 68% (34/50) for the whole trial; 69% (20/29) for the diabetes cohort
`and 67% (14/21) in the hypertensive cohort. It was in anticipation of discontinuations that the
`applicant prespecified that the primary analyses should be conducted in the mITT population
`for each cohort.
`
`Finally, due to the fact that Korlym blocks the progesterone receptor as well as the
`glucocorticoid receptor and may result in pregnancy loss, women enrolled in Study 400 had to
`have a negative pregnancy test prior to receiving Korlym; in addition, they had to agree to
`continue using a non-hormonal, medically acceptable method of contraception if they were of
`childbearing potential. Of note, fertility is significantly decreased in patients with Cushing's
`syndrome (136 reported pregnancies in the literature18
`).
`
`At the completion of Study 400, patients were allowed to be enrolled in an extension trial
`following an observation period of 6 weeks off medication. Study 415 was a multicenter,
`open-label, extension study conducted in patients with endogenous Cushing's syndrome who
`completed Study 400 and who, in the judgment of the investigator, received clinical benefit
`from Korlym in Study 400. Of the 34 patients who completed Study 400, 30 patients entered
`Study 415. Dosing principles in the extension study were identical to those in the pivotal
`study. The purpose of Study 415 was to collect additional safety data.
`
`Efficacy in the diabetes cohort
`
`The pre specified primary efficacy analysis was a responder analysis of the number of patients
`who experienced at least a 25% decrease in AUCg!ucose from baseline to end-of-trial during a
`2-hour oral glucose tolerance test. The analysis prespecified that the lower bound of the
`confidence interval for the proportion of responders should be >20%, a threshold which was
`chosen as clinically significant on the basis of being very different from the rate of
`spontaneous remission in the intended population, which is close to zero. The analysis was to
`
`18 Lindsay JR, Jonklaas J, Oldfield EH, Nieman LK 2005 Cushing's syndrome during pregnancy: personal
`experience and review ofthe literature. The Journal of clinical endocrinology and metabolism 90:3077-
`3083
`
`Page 12 of30
`
`Reference ID: 3086567
`
`12
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2055, Page 13
`
`

`

`

`

`

`

`Cross Discipline Team Leader Review
`
`reviewer re-calculated the mean change from baseline in HbAlc and anived at similar values:
`-1.11 (2-sided 95% CI = (-1.56, -0.65), which were statistically significant (p=.OOOl) HbAle
`was measured in a centrallaboratOlY.
`
`Of the 21 subjects in the diabetes group available for HbAlc analyses, 14 had above normal
`HbAlc levels at baseline, ranging behveen 6.7% and 10.4%; all of them had reductions in
`HbAlc by the end of the study (range 0.4 to 4.4%). Table 23 in the Clinical Review
`(reproduced below) presents baseline and change to end-of-trial in HbA1c for these 14
`patients; 8 of them (57 %) normalized HbA1c at the fmal visit (HbAlc < 6.4%). It is worth
`noting that only 6 of them received radiation therapy, a potential confounder of treatment.
`Such a degree of HbAlc reduction occurring over snch a relatively short period of time is not
`expected to be secondmy to radiation treatment. Besides, and importantly, all 6 patients had
`elevated ACTH levels at baseline and the levels remained elevated at Week 24,