Celltrion Biosimilar Package Is Strong, But How Many Indications Will It Get?

Like Zarxio before it, Inflectra is coming before an advisory committee having received a glowing review from FDA staff; the key question seems to be how many of Remicade's indications will it be able to extrapolate.

The FDA Arthritis Advisory Committee's Feb. 9 meeting on Celltrion Inc.'s Inflectra (CT-P13), a proposed biosimilar to Janssen Biotech Inc.'s Remicade (infliximab), will feature a greater emphasis on clinical data extrapolation across indications than did the only other panel review of a 351(k) application.

FDA will ask its external advisors to discuss whether there is sufficient scientific justification to extrapolate data from comparative clinical studies in rheumatoid arthritis (RA) and ankylosing spondylitis (AS) to support a determination of biosimilarity of CT-P13 for all of the other indications for which Remicade is licensed in the US, including inflammatory bowel disease claims, according to draft questions.

In the lone voting question on licensure, FDA asks the panel to explain whether any "no" votes apply to some or all of the indications on the Remicade label and, if so, why (see sidebar).

Such focused questions and language specific to data extrapolation across indications were not asked in the inaugural review of a 351(k) application – the Oncologic Drugs Advisory Committee's January 2015 meeting on Sandoz Inc.'s Zarxio (filgrastim-sndz), a biosimilar to Amgen Inc.'s Neupogen (filgrastim) (Also see "Sandoz’s Biosimilar Filgrastim ‘Highly Similar’ To Neupogen, FDA Staff Say" - Pink Sheet, 5 Jan, 2015.).

Yet, the heightened focus on data extrapolation is not unexpected given the wide array of indications for which Remicade is labeled, questions about the tumor necrosis factor-inhibitor's mechanism of action in Crohn's disease and ulcerative colitis, and the decision by Canadian regulators not to grant to the inflammatory bowel disease claims for Celltrion's biosimilar in that country.

There is, however, a striking similarity between the first two biosimilar candidates to go before FDA's external advisors. As with Zarxio, agency staff have given CT-P13 a glowing review heading into the all-important advisory committee meeting.

FDA's briefing document released Feb. 5 concludes that Celltrion's analytical, preclinical, clinical pharmacology, comparative efficacy and immunogenicity data demonstrate that CT-P13 is "highly similar" to US-licensed Remicade with no clinically meaningful differences in the studied indications.

In addition, Celltrion provided "an extensive data package" to address the scientific considerations for extrapolation of data to support biosimilarity to other conditions on the Remicade label, FDA said.

"In aggregate, the evidence indicates that the extrapolation of biosimilarity to the indications for which Celltrion is seeking licensure … may be scientifically justified," the agency concluded.

Extrapolation Justification

FDA's briefing document reviews a number of issues considered in the scientific justification for extrapolation of data to support biosimilarity for Remicade's indications beyond RA and AS.

These include the lack of differences in observed PK parameters for US-licensed Remicade in Crohn's disease patients as compared to patients with other conditions of use, and the similarities in PK between CT-P13 and Remicade.

In addition, there is an expectation for similar immunogenicity between CT-P13 and Remicade in other indications, and the primary mechanism of action in psoriatic arthritis and plaque psoriasis is the same as that for RA and AS.

Foreign Development And Approval

Celltrion will finally get to make its case to the Arthritis Advisory Committee after an 11-month delay (Also see "Celltrion's Infliximab Biosimilar Heads To FDA Panel After 11-Month Delay" - Pink Sheet, 15 Jan, 2016.).

FDA originally scheduled the meeting for March 2015 but postponed it due to information requests pending with the sponsor (Also see "Biosimilar Bummer: FDA Postpones Celltrion Infliximab Advisory Committee" - Pink Sheet, 25 Feb, 2015.).

The product's June 8 user fee goal date passed without word of any regulatory action (Also see "Celltrion's Biosimilar User Fee Deadline Passes With Conspicuous Silence" - Pink Sheet, 9 Jun, 2015.).

It appears that FDA simply let the user fee goal date pass; there is no mention in the agency's briefing document about having issued a complete response letter for CT-P13.

Pfizer Inc. holds US marketing rights to CT-P13 through its acquisition of Hospira Inc. in September.

The Celltrion product, which is also known as Remsima in some ex-US markets, has been approved for the same indications as Remicade in the EU, South Korea, Japan and India.

In 2014, Health Canada approved CT-P13 for all indications except ulcerative colitis and Crohn’s disease after concluding that extrapolation of data from RA and AS to inflammatory bowel disease could not be recommended due to residual uncertainty about the role and impact of small differences in antibody-dependent cellular cytotoxicity (ADCC) that might have relevance in IBD, FDA's briefing document notes (Also see "As They Look To The U.S., Celltrion And Hospira Win Approval For Remicade Biosimilars In Canada" - Pink Sheet, 21 Jan, 2014.).

"The development of CT-P13 was conducted entirely outside of the US and was directed towards meeting the product approval requirements of non-US regulatory agencies," the briefing document states.

Celltrion's first interaction with FDA on the product occurred though a Biosimilar Biological Product Development Type 3 meeting in July 2013 followed by a Type 4 meeting in April 2014.

Celltrion's 351(k) package included:

• Extensive analytical data;
• A single-dose pharmacokinetic study providing a three-way comparison of CT-P13, US-licensed Remicade and EU-approved Remicade;
• A comparative clinical study intended to demonstrate similarity in efficacy and safety between CT-P13 and EU-approved Remicade in patients with severely active RA;
• A PK and efficacy study in patients with moderate to severe AS;
• An assessment of safety and immunogenicity in patients undergoing single transition from EU-approved Remicade to CT-P13 during open-label extensions of the RA and AS studies; and
• Scientific justification for extrapolation of data across Remicade's indications.

Activity 'Within The Quality Range'

In keeping with FDA's recommended stepwise approach to demonstrating biosimilarity, the agency's briefing document goes through Celltrion's evidence starting with analytical data and ending with the clinical efficacy, safety and immunogenicity findings.

For any 351(k) submission, the analytical data are of primary concern to FDA because they form the basis for a finding of biosimilarity.

Following the Zarxio panel review, FDA stated a desire to have its experts pay more attention to the analytical characterization rather than focusing heavily on clinical data (Also see "Biosimilar Advisory Committees Will Focus Less On Clinical Data" - Pink Sheet, 25 Jan, 2016.).

The advisory committee roster for the CT-P13 meeting includes experts in protein engineering (see related story, (Also see "Biosimilar Committee Roster Shows More Focus On Analytics For Inflectra" - Pink Sheet, 5 Feb, 2016.)).

FDA's briefing document concludes that various comparisons of US- and EU-licensed versions of Remicade and CT-P13 "met the pre-specified criteria for analytical similarity, including statistical criteria for the critical potency bioassay (TNF-α neutralization) and TNF-α binding strength. Thus, a pair-wise analytical comparison of CT-P13 to US-licensed Remicade is consistent with the conclusion that CT-P13 is highly similar to the reference product."

Celltrion also established an adequate analytical bridge between EU-approved Remicade, US-licensed Remicade and CT-P13 to justify the relevance of the comparative data generated using EU-approved Remicade to support a demonstration of the biosimilarity to US-licensed Remicade, FDA said.

The analytical review includes a discussion of ADCC, which the agency described as "one of several plausible mechanisms of action" in Crohn's disease and ulcerative colitis.

The binding affinity of CT-P13 to the natural killer cells expressed as FcyRIIIa and b was shifted lower compared to US-licensed Remicade. "This was associated with subtle shifts in glycosylation at Asn297 on the heavy chain of the two antibody products detected in the analysis using a HPAEC-PAD chromatography method," FDA said.

When features across the broad class of TNF-inhibitors are examined, there is some evidence to suggest that low or no levels of ADCC are associated with a lack of efficacy in inflammatory bowel disease, FDA said.

"ADCC activity may vary with the strength of the FcyR:Fc bridging, which in turn may be dependent on the glycan composition of the antibody," FDA said.

An orthogonal ADCC assay using enriched natural killer cells suggested a small downward shift of approximately 20% in ADCC activity with CT-P13 compared with EU- and US-licensed Remicade at each concentration tested, FDA said.

Nevertheless, "the data support a demonstration that CT-P13 is highly similar to US-licensed Remicade because of the largely overlapping data (90% of proposed biosimilar lots were within the quality range defined by three standard deviations around the mean set by Celltrion’s data on the reference product) and the nature of the measured activity (i.e., an Fc function of uncertain importance)."

In sum, FDA concluded that the ADCC activity of CT-P13 "is within the quality range set by Celltrion's data on the reference product."

In addition, "the data submitted by Celltrion support the conclusion that CT-P13 and US-licensed Remicade have the same mechanisms of action for specified indications, to the extent that the mechanisms of action are known or can reasonably be determined."

Clinical Efficacy Data

Turning to the clinical data, Study 3.1 was a 54-week randomized, double-blind, parallel group trial in 606 RA patients. The primary endpoint was the proportion of patients who remained in the study and achieved an American College of Rheumatology 20% (ACR20) response at Week 30. Approximately 60.9% of patients on CT-P13 met this endpoint, compared to 58.9% of patients on EU-approved Remicade.

The estimated absolute difference between treatments was 2.0% (90% confidence interval [CI]: -4.6%, +8.7%), FDA said, adding: "The 90% CI successfully ruled out the similarity margin of ±12% that the agency has determined reasonable."

Study 1.1 was a 54-week randomized, double-blind, parallel group trial in 250 patients with AS; it was designed to compare PK profiles, with safety and efficacy comparisons as secondary objectives.

"Among the subset of randomized patients remaining in the study at Week 30, 70.5% of patients randomized to CT-P13 and 72.4% of patients randomized to EU-approved Remicade achieved an Assessment of SpondyloArthritis International Society 20% (ASAS20) response, for an estimated odds ratio comparing treatments of 0.91 (95% CI: 0.51, 1.62)," FDA said.

FDA noted concern about the high rate of early treatment discontinuation in the RA and AS studies, with 25% and 16% of patients failing to complete the 54-week, double-blind periods in each respective study. Since the high drop-out rates led to substantial missing data in important analyses, FDA conducted tipping point analyses to explore the sensitivity of results.

However, FDA seems to have been satisfied by the results of these analyses.

For the RA study, "confidence intervals for the differences between CT-P13 and EU-approved Remicade failed to rule out concerning losses in efficacy only under the assumption that patients who dropped out on CT-P13 had much worse outcomes than dropouts on EU-approved Remicade," FDA said.

"Given the similar proportions of patients and distributions of reasons for early withdrawal on the two treatment arms, in addition to the similar baseline characteristics between dropouts on the two arms, an assumption of such large differences between the outcomes in dropouts on the two treatments seems implausible. Therefore, these tipping point sensitivity analyses largely support the findings of the key efficacy analyses in Study 3.1."

FDA said no new safety signals were identified in the CT-P13 clinical program compared to the known adverse event profile of US-licensed Remicade. Although there was an observed numerical imbalance in active tuberculosis and pneumonia with CT-P13, these do not indicate a clinically meaningful difference, the agency said.

Immunogenicity

On the question of immunogenicity, FDA said there were numerical imbalances in the incidence and titer of anti-drug antibodies (ADA) in Study 1.4, the single-dose PK study of CT-P13, US-licensed Remicade and EU-approved Remicade. However, this imbalance was not associated with a difference in PK, and the low incidence of immunogenicity with US-approved Remicade in the same study was not consistent with the published data, FDA said.

The agency also noted that an interim analysis of immunogenicity from Study 3.4, an ongoing randomized, double-blind trial in patients with active Crohn's disease, indicated a comparable incidence of ADA formation between CT-P13 and US-licensed Remicade.

"In light of these additional contextual pieces, the agency does not believe that the results of Study 1.4 are likely to represent real or clinically meaningful differences between US-licensed Remicade and CT-P13. Therefore, there are sufficient data supporting similar immunogenicity between CT-P13, EU-approved Remicade and US-licensed Remicade and that immunogenicity adds to the totality of the evidence to support a demonstration of no clinically meaningful differences between CT-P13 and US-licensed Remicade."