A bioassay ultimately bridged conflict between the Office of New Drug Quality Assessment and the Botanical Review Team over the adequacy of product characterization in Salix Pharmaceuticals Ltd.’s Fulyzaq (crofelemer) NDA, review documents show – but only after the dispute was adjudicated by CDER Director Janet Woodcock.

“Late phase development of botanical new drugs remains a rarity and review experience with botanical NDAs is limited,” lead botanical reviewer Shaw Chen observed in an Aug. 8, 2012 review. “We are now setting precedents with the first few botanical NDAs.”

The final agreement on crofelemer identity testing is a “holistic approach to controlling this extremely complicated drug substance,” chemistry reviewer Nina Ni concluded.

“This is a botanical product whose constitution by its very nature is more complex and heterogeneous than the purified new molecular entity classically approved by the FDA,” Division of Gastroenterology and Inborn Errors Products Deputy Directory Andrew Mulberg noted in a Dec. 14, 2012 summary review. Crofelemer is derived from the latex of a South American tree; it is only the second botanical NDA reviewed by FDA since the agency’s 2004 guidance on botanical drug products. Manufacturing specifications were also the major concern in the review of the first NDA, MediGene Inc.’s Veregen (kunecatechins), a topical extract of green tea approved in 2006 for genital warts Also see "No FDA Compromise On Veregen Specifications In First Botanical NDA Review" - Pink Sheet, 15 February, 2007..

The main point of contention centered on a question of identity: how to ensure that the marketed product is the same drug substance tested in clinical trials. “There is notable discordance of opinion between the CMC and Botanical reviewers,” Mulberg said. “The primary ONDQA concern is the lack of a reproducible and robust test for the identity of the drug substance.” The botanical reviewers, however, argued that “botanical new drugs can rarely have CMC specifications as precise as those of pure chemical drugs.”

Identity is “especially difficult to determine for botanical drugs with unknown number and identities of active ingredients,” Mulberg said. “Even in the best case, the chemical composition of a botanical preparation is not completely defined, nor can all active ingredients be necessarily identified,” he observed. “Strength and potency of these vaguely defined products are not easy to determine, adding to the difficulties in CMC controls and clinical pharmacology studies.”

The first ONDQA review of Fulyzaq recommended against approval, countering the botanical review team’s support. While chemistry reviewer Nina Ni noted that the sponsor had conducted a range of spectroscopic analyses, physical tests and chemical tests, she found those values to be at best “useful surrogate markers for the identity of the drug substance.”

“Although all the physical and chemical identity tests are placed in the specification, they, as a whole, are still deemed far from fulfilling the statutory requirement for the identity of a drug,” Ni concluded in her July 13, 2012 review.

The Botanical Review Team’s Chen, on the other hand, argued that the guidance allows for botanical drug products to remain as complex mixtures. Therefore, “both purification and identification of the active ingredients in botanicals are optional and not required.”

Chen attributed the dispute between the two review disciplines to ONDQA’s “strict reading of regulations” from a “pure small molecule drug perspective.” Regulatory interpretation “needs to be extended from a narrow, chemical mindset to accommodate the complex nature of [a] botanical drug substance,” Chen declared.

“From BRT’s experience, attempts to require high degree molecular characterization will usually be futile,” Chen stated. “It would be technically impossible, financially prohibitive or clinically unnecessary.”

Chen acknowledged that “standard CMC characterization … remains the most important method to ensure quality and therapeutic consistency” but described an emerging broad paradigm for control of botanical drug substances that ranges from “pre-CMC” steps focused on growing and harvesting medicinal plants with good agricultural and collection practices to “post-CMC” approaches looking at clinically relevant bioassays and clinical experience with various batches.

[Sidebar#06130501023]

Unmet Medical Need

FDA strove to resolve the stand-off between the botanical and CMC review teams quickly because Fulyzaq, the first treatment specifically for secretory diarrhea associated with antiretroviral therapy for HIV infection, addressed an unmet medical need. Control of diarrhea “not only improves the quality of patients’ lives, but also may improve the compliance to antiretroviral therapy,” clinical reviewer Wen-Yi Gao explained in an Aug. 31, 2012 review. Gao noted that 40% of patients on highly active antiretroviral therapy regimens have moderate to severe diarrhea, and 28% of patients discontinue HAART due to gastrointestinal intolerance.

Fulyzaq is the first drug to make “treatment of persistent and refractory HIV diarrhea” possible, Gao said. Not only were no drugs indicated for HIV diarrhea, but available anti-diarrheal medications like loperamide induce tolerance with repeated use. “This is the first demonstration of a 20-week persistent anti-HIV diarrheal effect in this field,” Gao said.

The adaptive clinical trial included a five-month uncontrolled (placebo-free) phase after a four-week placebo-controlled phase; concomitant antidiarrheals were allowed only in the placebo-free portion. While the extension phase results are not definitive, placebo crossover patients and patients continuing on crofelemer “appeared to be responsive,” Gao said (see related story,Also see "Fulyzaq Review Shows FDA Adapting Smoothly To Adaptive Trial Design" - Pink Sheet, 22 May, 2013.). An open-label 48-week Phase III trial, Study CFHD3092, is ongoing; the safety and tolerability trial has completed enrollment of 250 subjects.

Regulatory Briefing

“Due to the extensive ONDQA concerns related to assurance of quality and the perceived medical need for this product, a CDER Center Director Briefing was held on Aug. 6, 2012,” ONDQA Division Of New Drug Quality Assessment II Director Terrance Ocheltree reported in a Dec. 31, 2012 memo.

Regulatory briefings are sometimes called “internal advisory committees” – sessions for senior CDER staff to hash out difficult questions arising in the drug review process Also see "Regulatory Briefings: FDA's "Internal Advisory Committees" " - Pink Sheet, 1 November, 2008.. Thorny manufacturing issues presented by new technologies or product categories are not infrequent subjects for regulatory briefings. Recently, CMC issues concerning Genentech Inc.’s antibody-drug conjugate Kadcyla (ado-trastuzumab emtansine) were brought to CDER Director Woodcock ([A#06130401018]).

The Fulyzaq regulatory briefing addressed three questions, according to Ocheltree:

“Is the benefit for public health of crofelemer greater than the potential risk that may originate due to unknown and inconsistent composition of the drug product?”
“Is there any concern if the commercial product may not be the same as that used in the clinical trials?”
“Is it sufficient to control identity and other quality attributes solely through the control of plant source and manufacturing process?”

The discussion revolved around whether the “urgent” medical need merited moving the application forward and the concern that if the drug were approved, what if the sponsor were asked to submit additional data post-approval and could not, Mulberg reported.

“Dr. Woodcock does not believe that all of the CMC characterization is needed for approvability” of crofelemer, Mulberg stated. Woodcock agreed that the sponsor “should be asked to submit a potency assay, i.e. a bioassay that is clinically relevant that would support the approvability of this compound by demonstrating a relevant relationship to the putative pathobiology of HIV-associated diarrhea and impact on the pathophysiological process of chloride secretion.”

The bioassay, when combined with updated drug substance specifications proposed by Salix, assuaged the ONDQA review team’s concerns. “This information was deemed sufficient to establish the identity of the crofelemer drug substance and drug product, if a validated, clinically relevant bioassay were implemented as part of release specifications,” Office of Drug Evaluation III Director Julie Beitz emphasized in the final Dec. 31, 2012 approval memo.

Too Late For Typical Acceptance Criteria

After the regulatory briefing, FDA moved quickly to ask Salix to “develop a bioassay that would allow for potency testing and for release and stability testing on future batches,” review team leader Anil Rajpal reported in a Dec. 28, 2012 review (see timeline,[A#06130501006]). By Aug. 10, Salix had submitted a bioassay development proposal for review by the Office of Biotechnology Products’ Division of Therapeutic Proteins, which was concerned that bioassay development and validation “will require a time-frame significantly longer than the one proposed by the applicant,” Rajpal noted.

The OBP review team initially had concerns about the inspection status and GMP compliance status of the release and stability testing site. A new inspection often “requires several months” to arrange, Rajpal pointed out, but the OBP reviewers suggested that “for unmet medical need, perhaps the inspection could be speedily arranged.” The final validation report for the bioassay was received on Dec. 14, 2012. “The applicant proposed an acceptance criterion for the bioassay … based on pooled results obtained from the bioassay validation on drug substance and drug product,” Rajpal reported.

OBP deemed the proposal “acceptable,” but only “because the applicant was asked to establish this assay very late in drug development,” the team leader said. OBP called the bioassay plan “not optimal” because “typically acceptance criteria are established based on the results of the material used in the clinical trial. In this case, the typical approach is not feasible because the applicant has little to no clinical trial material still available.” OBP signed off on the bioassay on Dec. 21, 2012; the review division adopted OBP’s recommendation for post-marketing commitments aimed at “improvement to the assay and acceptance criteria.”

A GMP inspection of the facility performing the bioassays was the final piece of the approval picture, and that did not come until the very end of the review – literally: the Office of Compliance signed off on the inspection on Dec. 31, the day of approval.

Senior review officials appeared to have been confident that the facility would pass inspection. Mulberg, on Dec. 14, provisionally recommended approval “at the current time based on the current state of bioassay development and the criticality of this assay for making a final decision on action for the NDA, an approval decision is being made as there is establishment of the validated bioassays which are reproducible with crofelemer.”

A “Holistic Approach” To CMC

The agreement to implement a bioassay that addressed the concerns of both ONDQA and the Botanicals Review Team produced a “holistic approach to controlling this extremely complicated drug substance,” chemistry reviewer Ni declared in her final Dec. 31, 2012 review. “Although no single test can give an unequivocal identity of crofelemer, it is believed that when all the tests are combined, the composite results are expected to provide a reasonable insight into the nature of the identity of crofelemer.”

The “holistic approach” includes “strict controls on the raw material,” “robust manufacturing process controls” and “adequate specifications” with “validated chemical and physical tests” as well as a “validated clinically relevant bioassay.”

“This is a new paradigm that bears further deliberation as this application bears truth to the complexities of the scenario,” Mulberg concluded.

The “holistic approach” could prove fruitful as FDA faces greater challenges in botanical products regulation. “Compared with most botanical preparations in the INDs we have received, crofelemer is a relatively simple botanical derived from a single part of a single plant,” botanicals team leader Chen observed.

A Question Of Identity: Drug & Botanical Perspectives On Manufacturing Clash In Fulyzaq Review

More from Manufacturing

More from Compliance