FDA’s approval of Corcept Therapeutics Inc.’s Korlym for Cushing’s disease without a Risk Evaluation and Mitigation Strategy shows how the agency put practical considerations and precedent above potential political implications – but not without taking the decision all the way to Center for Drug Evaluation and Research Director Janet Woodcock for vetting.

The active ingredient in Korlym, mifepristone, also is approved as an abortifacient as Danco Laboratories’ Mifeprex, and as such is subject to a REMS. FDA review documents reveal that early on, the agency recommended a REMS for Korlym as well – but it reversed that stance based on evaluation of REMS precedents and the needs of the indicated patient population.

FDA’s decision-making process looked at the need for a REMS from two perspectives, review documents show: first, was a risk management strategy necessary to ensure safe use by the population of Cushing’s syndrome patients specified in the Korlym indication; and, second, was a REMS needed to maintain the integrity of the existing mifepristone REMS for Mifeprex for early pregnancy termination?

The agency review team agreed upon the answer to the first question, focused on the type of risk/benefit calculation that underlies many drug reviews, early on. The frontline review teams from both the Office of Surveillance and Epidemiology’s Division of Risk Management and the Office of New Drugs’ Division of Metabolism and Endocrinology Drug Products “agreed that a REMS with [Elements To Assure Safe Use] for Korlym would not improve the benefit/risk balance for the intended use (Cushing’s) population, and would impose an unnecessary barrier to their access to the drug,” clinical reviewer Marina Zemskova reported in a Jan. 13, 2012 review.

The more politically fraught second question, on maintaining the integrity of the Mifeprex REMS, was the one of most interest to FDA officials above the division level. The Korlym REMS question was discussed by the REMS Oversight Committee on Sept. 29, 2011, and then in the center director briefing Nov. 3, 2011. “The REMS Oversight Committee and the Center Director provided additional guidance and affirmed that although a REMS is required for Mifeprex, a REMS for Korlym is not necessary,” given that Korlym’s safety and drug utilization could be “monitored” via post-marketing requirements, DRISK analyst Suzanne Robottom stated in a Jan. 27, 2012, review.

The CDER director briefing brought together high-level staff from the drug review, drug safety, regulatory policy and executive operations branches of the center to address “whether it is necessary to require a REMS with ETASU for approval of Korlym … in order to maintain the integrity of the current Mifeprex REMS program,” minutes of the Nov. 3, 2011, briefing state. “By approving Korlym without an ETASU, we discussed the [possibility that] Korlym might be used off-label to terminate a pregnancy or misused or diverted in order to terminate a pregnancy without proper healthcare provider supervision,” the minutes state.

CDER All-Stars Ratify Decision To Forgo Korlym REMS

On Nov. 3, 2011, CDER Director Janet Woodcock was briefed on the question of whether a REMS with elements to assure safe use was necessary for Korlym in order to “maintain the integrity” of the Mifeprex REMS. In addition to Woodcock, attendees at the briefing included:

Office of Regulatory Policy Director Jane Axelrad
Office of the CDER Director Associate Director for Drug Safety Operations Teresa Toigo
Office of New Drugs Director John Jenkins
OND Associate Director for Safety, Policy and Research Mwango Kashoki
OND Office Drug Evaluation II Director Curtis Rosebraugh
Office of Pharmacovigilance and Epidemiology Director Gerald Dal Pan
OPE Division of Risk Management Director Claudia Karwoski
DRISK Team Leader Cynthia LaCivita, and DRISK Risk Management Analyst Suzanne Robottom
OND Division of Metabolism and Endocrinology Products Deputy Director For Safety Amy Egan
DMEP clinical reviewer for Korlym Marina Zemskova
Office of the Chief Counsel attorney Carla Cartwright
Office of Regulatory Policy Senior Program Manager Kristin Everett
Office of Executive Programs Policy Analyst Mary Gross and Regulatory Project Manager Lee Lemley

The Importance Of Treatment Setting

The severity and relative rarity of Cushing’s syndrome was a keystone of Woodcock’s decision to support the DMEP and DRISK review teams in not requiring a REMS for Korlym. The minutes note that the endogenous syndrome is “a serious multisystem disorder” that is “associated with a decreased quality of life and increased mortality.” Korlym received orphan designation for Cushing’s on July 5, 2007, for treatment of endogenous Cushing’s syndrome. And while Cushing’s itself is rare – FDA estimates a prevalence in the U.S. of about 20,000 patients – the limitation of the Korlym indication only to patients with hyperglycemia induced by diabetes or glucose intolerance due to Cushing’s syndrome mean that only about 5,000 patients would be eligible for drug treatment.

“Treatment requires highly specialized care in tertiary care centers,” the meeting minutes emphasize. “The number of prescribers who would prescribe Korlym is also limited; there are only 35 endocrinologists in the U.S. who are recommended by the Cushing’s Support and Research Foundation for treatment and care of Cushing’s syndrome patients.”

FDA is typically very sensitive to the sophistication of the prescribers in a given setting; oncology provides the clearest example, where the agency has generally relied on the expertise and judgment of oncologists to comprehend and manage the sometimes fine line between risk and benefit with cancer treatment without the edifice of formal REMS programs. Office of Hematology and Oncology Products Director Richard Pazdur has described oncology’s “de facto restricted distribution system” in explaining why cancer drugs have different risk management standards and incur fewer REMS than other therapeutic classes Also see "Oncology Has Different Risk Management Standards, FDA’s Pazdur Acknowledges" - Pink Sheet, 13 June, 2011..

At roughly the same time that FDA was considering the need for a Korlym REMS, similar themes arose in the agency’s debate over the need for a REMS regarding teratogenicity associated with Genentech Inc.’s pioneering Hedgehog inhibitor for basal cell carcinoma, Erivedge (vismodegib). Division of Oncology Products 2 Director Patricia Keegan’s summary review explains that a REMS would not be required, “despite the seriousness of the teratogenic risk,” for reasons echoing some of the reasons behind the Korlym decision: the lack of effective alternative therapies, the small number of patients potentially at risk, and a “standard of medical care” that provides “adequate safeguards through familiarity with the risks, risk communication, and patient monitoring” (see related item,[A#06120801038]).

The REMS That Wasn’t, And The Central Pharmacy That Is

In the background of FDA’s deliberations on the need for a REMS with ETASU for Korlym was an actual REMS proposal, submitted by Corcept with the original April 15, 2011, NDA submission per FDA’s instruction. The agency told Corcept that a “REMS with restricted distribution will be necessary for this NDA/indication” during the Sept. 14, 2010, pre-NDA meeting, according to Zemskova.

An Oct. 17, 2011 meeting request indicates that the company was proceeding to lay the groundwork for the REMS. “Due to the time frame for setting up and implementing REMS elements, Corcept has initiated activities with vendors and begun preparing draft materials needed for the program,” Corcept said in its request. Presciently, the company continued, “we would like to obtain agency input and agreement on the program to be sure we are progressing in the right direction before spending more time and resources to complete these activities.”

While the description of Corcept’s REMS plan is considerably redacted, the company’s correspondence with FDA outlines the plan. The proposed Korlym REMS would have mandated registration by prescribers, patients, and specialty and hospital pharmacies, with specific communication material for each group. The drug would “only be dispensed to patients with documentation of safe use conditions,” Corcept said, and “female patients of child bearing potential must have had a negative pregnancy test prior to receiving Korlym.”

A key element of the REMS proposal – preventing Korlym retail distribution by making the drug available only through selected specialty pharmacies – was resurrected as a voluntary limited distribution system. “We do agree with your proposal to distribute Korlym solely to ‘a limited number of specialty pharmacies,’” FDA told Corcept during the Dec. 13, 2011 meeting where the decision not to require a REMS was communicated. “We anticipate that Korlym would be shipped from these specialty pharmacies directly to patients or hospitals. This should prove more convenient for patients and will facilitate monitoring the use of Korlym to ensure that it is being used in the indicated population.” Ultimately, Korlym was approved with one central pharmacy.

REMS or no, “the need for some monitoring” was a factor in FDA discussions. Commercial drug use databases are insufficient for tracking the use of both Korlym and Mifeprex, neither of which are available through retail pharmacy channels. “Using a single specialty pharmacy does allow the sponsor the ability to monitor use more closely through its business contract with the specialty pharmacy,” DRISK’s Robottom noted.

“A post-marketing requirement (PMR) study to obtain Korlym use data (age, gender, dose, duration of treatment) ‘to better characterize the incidence rates of adverse events with Korlym’ is prudent,” she stated. “Monitoring drug use data for both Mifeprex and Korlym, in conjunction with reports of serious adverse events resulting from pregnancy terminations outside of the Mifeprex REMS, will be important factors in future regulatory action to address any compromise to the Mifeprex REMS.”

If the PMR data “indicate that this approach compromises the integrity of the Mifeprex REMS and results in serious adverse events, or additional serious safety signals arise, further regulatory action must be considered,” Robottom emphasized.

Looking For Consistency In The Use Of REMS

While FDA’s REMS authority is broad, the agency does not see it as a blunt instrument. “The agency evaluates an active ingredient based on the risk-benefit profile for the intended population,” Robottom clarified.

Significant new uses for available drugs pose a particular challenge for that principle.

Concerns that one indication could pose significant risks, but that restricting access to a drug could pose a different set of risks to patients in another setting, also were raised during the reviews of Vivus Inc.’s obesity treatment Qsymia (topiramate/phentermine) and Gilead Sciences Inc.’s Truvada (emtricitabine/tenofovir) for HIV pre-exposure prophylaxis. For Qsymia, the teratogenicity risk of topiramate posed a different risk/benefit calculation in the obesity context than in the drug’s previously approvals in epilepsy and migraine indications, where FDA was loath to limit access. Truvada posed a public health conundrum: how to expand use for PrEP while preventing development of resistance in HIV-positive patients without restricting access for either group.

Ensuring access was a key aspect of FDA’s risk strategy in both approvals. Qsymia was approved in July with a REMS regarding risk of birth defects that includes restricted distribution. While FDA accepted Vivus’ proposal for dispensing only through select mail-order pharmacies, the agency requested that the company develop a plan for dispensing through a broader range of pharmacies, including at retail Also see "REMS For Qsymia Limits Distribution, But Vivus To Explore Expanding Access" - Pink Sheet, 23 July, 2012.. Truvada cleared FDA for PrEP, also in July, with a REMS with ETASU, but the agency declined to require a restricted distribution system linking access to documented evidence of a negative HIV test ([A#00120723003])

“It is important to ensure that the intended treatment population can receive Korlym in a timely, dependable manner in the least burdensome way,” Robottom emphasized in her review of the Cushing’s drug. “Any restrictions will impede access with little to no benefit to Cushing’s syndrome patients.” This consideration outweighs the chance that the lack of restrictions on Korlym could compromise the safe use conditions set up by the Mifeprex REMS, she said.

FDA has not required a REMS for any product based solely on the fact that the active ingredient is covered by a REMS for another indication.

Applying a REMS to Korlym because mifepristone has a REMS as Mifeprex would break new ground in safety regulation, Robottom indicated. “To date, the agency has not required a REMS for a product based only on the fact that the active ingredient already has a REMS for one population,” she stated.

“Denosumab was originally approved under two tradenames for different indications,” Robottom pointed out. As Prolia, Amgen Inc.’s biologic was approved for treatment of post-menopausal osteoporosis with a REMS regarding “the risks of serious infections, dermatologic adverse reactions, and suppression of bone turnover, including osteonecrosis of the jaw,’” she reported. But when denosumab was approved as Xgeva for prevention of skeletal-related events in patients with bone metastases from solid tumors, “a REMS was not required given the resulting differences in the risk-benefit profile when considering the patient populations (post-menopausal women vs. cancer patients with bone metastases) and prescribing populations (internists vs. oncologists).”

Looking For Precedents For Controlling Misuse

Placing Korlym in context was not entirely straightforward. “There are no other glucocorticoid receptor antagonists approved in the U.S. for comparison,” the risk management review noted (see related story, ([A#06120801014]). Some anti-corticolic drugs are used in Cushing’s treatment, like ketoconazole and mitotane, but “because these drugs have a different mechanism of action, they are not associated with the same potential risks as mifepristone,” and none have a REMS.

In terms of the signal risk of mifepristone – fetal loss – Robottom looked at how FDA handled other drugs with similar risk, chiefly methotrexate and misoprostol, “At present, this risk is addressed through labeling for these drugs,” Robottom noted. “There are no REMS approved that address only fetal loss without also the accompanying risk of birth defect.”

The potential for off-label use of misoprostol and methotrexate to terminate pregnancy is also not addressed by a REMS. While the extent of off-label abortifacient use of either drug is “unknown,” Robottom reports that “with each drug the risk of termination of pregnancy is managed through labeling (Contraindication, Boxed Warning).”

FDA has used a variety of strategies to handle the broader problem of drug misuse. Robottom highlighted the agency’s handling of Egrifta and Xyrem to illustrate the two main pathways: voluntary and required restricted distribution.

Growth hormones are at risk for misuse and abuse for off-label uses like weight loss and body building, but “none of the growth hormone products have a REMS,” Robottom observed. When reviewing Theratechnologies Inc.’ synthetic growth-hormone releasing factor analog Egrifta (tesamorelin) for HIV lipodystrophy, FDA concerns about unapproved use initially spurred a request for a REMS, but – in a solution very similar to the final Korlym risk management plan – the agency instead settled on a voluntary restricted distribution system coupled with sponsor monitoring of prescribing activity for evidence of inappropriate use. FDA backed away from the REMS largely as a matter of fairness and consistency with the lack of such programs for other growth hormone products Also see "Egrifta Approval: FDA Balanced Modest Efficacy, Off-Label Use Potential With Unmet Medical Need" - Pink Sheet, 1 February, 2011..

Jazz Pharmaceuticals PLC’s Xyrem (sodium oxybate) was approved for cataplexy in narcolepsy patients in 2002, before the enactment of REMS authority (it would be one of the first tranche of products with “deemed REMS” under the regs). Nonetheless, its approval was contingent on a RiskMAP under which the sponsor agreed to use a central pharmacy to dispense the drug, which would fill prescriptions only after prescriber and patient had received and read educational material, and to maintain prescriber and patient registries, Robottom summarized. The RiskMAP was “deemed” a REMS under the FDA Amendments Act, requiring Jazz to pursue an actual REMS.

Xyrem also posed another instance of one drug/two names when Jazz pursued a fibromyalgia claim for sodium oxybate in a different dose strength under the name Rekinla. The potential approval of Rekinla held the prospect of separate REMS, with separate enrollment forms and pharmacy dispensing programs, to prevent medication errors and abuse (sodium oxybate is known as a “date rape drug”).

FDA’s briefing material for the Rekinla advisory committee highlighted the agency’s anxiety about the potential for confusion Also see "Two REMS, Two Names, One Drug: Is It Safe? FDA Questions Jazz's Proposal For Fibromyalgia Indication" - Pink Sheet, 18 August, 2010.. The advisory committee recommended against approval 20-2, and FDA followed with a broad “complete response” letter requesting additional clinical and risk management work in October 2010 Also see "Narrower Indication, New Studies Are Likely Riffs For Jazz Fibromyalgia Drug" - Pink Sheet, 18 October, 2010..

Mifeprex In The Real World

FDA’s risk management thinking also took into account real-world experience with Mifeprex, using it to shape predictions of the likelihood of off-label use of mifepristone. A Mifeprex drug use review requested by DRISK analyzed off-label use of mifepristone from 2000 to 2011. Off-label Mifeprex use for Cushing’s syndrome “is uncommon,” Office of Surveillance and Epidemiology Division of Epidemiology II Drug Use Analyst Patty Greene concluded Sept. 19, 2011.

“Overall drug exposure and subsequent access is anticipated to be small given the small size of the intended use population and lack of a signal for substantially broader use,” Robottom stated.

The different regimens, dosing, and distribution of Mifeprex and Korlym also argued against widespread off-label use of the Cushing’s syndrome product to terminate pregnancies.

One Ingredient, Two Paths To Patients

Differences in dosing and availability reduce the likelihood of off-label use of one brand of mifepristone for the other.

Mifeprex

Korlym

Available Forms

200 mg tablets in blister packages of 3 pills per carton

300 mg tablets, packaged in bottles of 28 or 280 tabs

Dosing Regimen

One-time use: 600 mg followed by misoprostol 400 µg (label); 200 mg followed by 800 µg misoprostol (standard practice)

300 mg to 1,200 mg daily for chronic use

Distribution restrictions

REMS requiring prescriber certification; distribution only though medical facilities

Voluntary restricted distribution program (SPARK) limiting distribution to central specialty pharmacy

“Third party payors/reimbursement may play a substantial role in influencing prescribing behavior,” Robottom predicted. With an average wholesale price of approximately $6,700 to $26,800 for a monthly supply of Korlym (daily dose can range from 300 mg to 1,200 mg), managed care tools like prior authorization can be expected to control for off-label use.

Robottom acknowledged that “patient privacy is a significant issue with Mifeprex,” but said that privacy concerns do not meet the criteria to trigger a REMS because “the purpose of a REMS is to ensure that the benefits of the drug outweigh its risks.” The Mifeprex REMS responds to concerns about safety and proper use.

REMS Are Not The Only Risk Management Tools

FDA used risk management techniques at a less intensive level than the REMS to address the potential dangers of Korlym. The “serious risks associated with the use of this product … can be adequately addressed through physician and patient labeling,” FDA told Corcept during the Dec. 13, 2011 meeting where the decision not to require a REMS was communicated.

“In final analysis it was agreed that the main risks associated with Korlym use in patients with Cushing’s syndrome (adrenal insufficiency, hypokalemia, endometrial thickening and vaginal bleeding, unintended fetal loss, and risks due to drug-drug interactions) could be communicated effectively to physicians via information contained in the package insert, while a non-REMS Medication Guide could achieve the same function for individual patients,” Cross-Disciplinary Team Leader Dragos Roman said in a Feb. 13, 2012 review.

Pregnancy information could be provided through the non-REMS Medication Guide, without the proposed REMS requirement for patients to register and acknowledge receipt of the risk information. The agreed-upon Medication Guide would address “the need to have a negative pregnancy test prior to Korlym treatment initiation and following any discontinuations longer than seven days, about the need to use a medically effective form of contraception during Korlym treatment to reduce the albeit low risk of pregnancy and of unintended Korlym-induced pregnancy termination,” Roman said.

DMEP Director Mary Parks commented that “the probability that a Cushing’s patient will become pregnant while on Korlym is very low” but, “regardless, the label will include a boxed warning and a contraindication for its use in pregnant women.” Parks called a contraindication “the most stringent safety warning in an FDA approved labeling” in her summary review. “It means that the risk from use of Korlym clearly outweighs any possible therapeutic benefit in the pregnant patient,” she said.

And, of course, patients have a choice. “Patients who do not want to face this extremely small but nevertheless real risk have the option to forego medical treatment with Korlym,” Roman stated.

FDA’s Decision To Back Off Korlym REMS Shows Context Is King

More from Drug Safety

More from Pink Sheet