FDA Panel Looks To Insurers To Enforce TNF Inhibitor Use Restrictions In Spondyloarthritis

Arthritis Advisory Committee recommends adding an axial spondyloarthritis claim to the label of UCB’s Cimzia but not Abbott’s Humira; concerns about potential misuse in back pain patients who do not have the inflammatory condition weigh heavily during the panel’s two-day meeting on the novel indication.

A limited indication for axial spondyloarthritis and third-party payer mechanisms to constrain drug use, such as prior authorization, could help ensure that TNF inhibitors are prescribed only for those patients with evidence of the inflammatory condition and not a much broader group of patients with back pain, an FDA advisory committee said.

During a two-day meeting focused on axial spondyloarthritis (axial SpA) as an indication for drug development in general and for two TNF inhibitors specifically, members of the Arthritis Advisory Committee were preoccupied with the potential for the biologics, which are expensive and have well-known safety risks, to be used in a broad swath of patients who do not meet imaging and clinical criteria for the inflammatory condition.

However, they suggested these concerns could be remedied, in part, by developing a narrowly tailored indication that is linked to documented evidence of the disease. Such a highly specific indication would then likely be enforced by insurers who already place TNF inhibitors on prior authorization.

The de facto role of payers as gatekeepers to ensuring appropriate use could allow the novel axial SpA indication to move forward.

The de facto role of payers as a gatekeeper to ensuring appropriate use in axial SpA could be appealing to FDA given its inability to restrict prescribing of the currently marketed agents to rheumatologists or other practitioners experienced in diagnosing the condition.

And, as much as the biopharma industry may dislike prior authorization, in the case of axial SpA the insurance mechanism could actually allow a novel indication to move forward as FDA gains more confidence that the appropriate patient population is being treated.

Though not a certainty, UCB Inc.’s Cimzia (certolizumab) could be the first TNF inhibitor approved for patients with axial SpA following the committee’s narrow endorsement of a supplemental indication July 23. In contrast, the panel’s same-day, overwhelming rejection of AbbVie Inc.’s Humira (adalimumab) for use in a subset of axial SpA patients suggests the company may have to conduct another trial if it wants to add the novel claim (Also see "FDA Panel Gives Nod To Spondyloarthritis Claim For Cimzia, But Not Humira" - Pink Sheet, 23 Jul, 2013.).

ASAS Patient Classification Criteria

Axial SpA is a broad spectrum of chronic inflammatory disease that primarily affects the axial skeleton. Ankylosing spondylitis (AS) is a well-characterized, chronic and progressive form of axial SpA. Four TNF inhibitors, including Humira, are approved for treating AS, but there are no approved treatments for the broader umbrella condition of axial SpA.

Drugs that carry an AS indication were approved on the basis of studies that used the modified New York criteria to identify patients for enrollment. Under this classification system, patients are deemed to have AS if they have radiological evidence of sacroiliitis and meet at least one of three clinical criteria.

Advisory Committee Votes

Humira

• 12-1 (1 abstention) that the data do not provide substantial evidence that adalimumab provides a clinically meaningful beneficial effect in the treatment of active non-radiographic axial spondyloarthritis in adults with objective signs of inflammation by elevated CRP or MRI, who have had inadequate response or are intolerant to nonsteroidal anti-inflammatory drugs
• 10-4 that the safety profile is adequate to support approval for the proposed indication
• 12-1 (1 abstention) recommending against approval for the proposed indication

Cimzia

• 8-5 (1 abstention) that the data provide substantial evidence that certolizumab provides a clinically meaningful beneficial effect in the treatment of active spondyloarthritis, including patients with ankylosing spondylitis
• 13-1 that the safety profile is adequate to support approval in the proposed indication
• 7-6 (1 abstention) recommending approval for the proposed indication

On July 22, the advisory committee discussed the Assessment of Spondyloarthritis International Society (ASAS) criteria for axial SpA and implications of using these criteria for drug approval. ASAS developed new classification criteria to encompass more patients with inflammatory back pain, including those that do not yet have X-ray changes, also known as nonradiographic axial SPA (nr-axSpA).

The ASAS criteria require patients to have back pain for at least three months and an age of onset of less than 45 years. Beyond this threshold, patients must satisfy one of two criteria arms. The imaging arm requires evidence of sacroiliitis on X-rays or magnetic resonance imaging plus at least one typical SpA feature, while the clinical arm requires presence of the genetic marker human leukocyte antigen (HLA)-B27 coupled with at least two typical SpA features.

Both the Cimzia and Humira pivotal trials supporting the proposed axial SpA indications used modified versions of the ASAS criteria for enrollment purposes.

In briefing documents released ahead of the meeting, FDA questioned whether axial SpA is a sufficiently well understood and defined population to support a new indication for drug development. The agency also raised concerns about basing a new indication on the ASAS criteria, suggesting this classification system may be overly broad, encompassing subgroups of conditions with various disease progression and resolution trajectories (Also see "Humira, Cimzia Arthritis Indications May Rest On Adequacy Of Classification Criteria" - Pink Sheet, 19 Jul, 2013.).

Preventing Inappropriate Treatment

During the course of their discussion on the ASAS criteria, and on the Humira and Cimzia sBLAs the following day, panelists fretted about how to prevent widespread prescribing of TNF inhibitors, particularly by providers who are not well versed in diagnosing axial SpA, to patients with mechanical, non-inflammatory back pain.

“I very much like the ASAS inclusion of these new sets of people who we’ve all seen clinically and clearly have this disease, even though the radiograph isn’t yet positive, and I think coming up with therapeutic options for them is important,” said temporary voting member David Felson, Boston University. “My main concern is … that you’re going to approve these drugs and they’re going to be indiscriminately used by patients treated by primary care physicians out there who say, ‘Oh, this patient might have this entity,’ and subject a lot of patients who are unlikely to have” axial SpA to the risks of TNF inhibitor treatment.

Felson questioned whether FDA had a mechanism to constrain prescribing by only rheumatologists “so that it isn’t indiscriminately used [by] millions of patients who have low back pain.”

“I think that’s part of what our concern is,” responded Sarah Yim, associate director of FDA’s Division of Pulmonary, Allergy and Rheumatology Products. “Our mechanisms for constraining use, there are very few of them and they’re very burdensome, the elements to assure safe use, for example. Those are very difficult to control, very difficult to enforce. So we try and sort of take it from the top, and if the indication is overly broad, perhaps we shouldn’t have such an overly broad indication because enforcement and restricting the use of a product, especially those that are already out there, is very difficult.”

Insurer scrutiny of TNF inhibitor use will increase, not decrease, if an axial SpA indication is granted, industry consultant Deodhar predicted.

During the panel’s deliberations on the Humira sBLA, Aileen Pangan, senior medical director of immunology for AbbVie, said that 80% of TNF inhibitor prescriptions in AS are written by rheumatologists, suggesting a high degree of expertise in diagnosing this condition.

Industry consultant and rheumatologist Atul Deodhar, Oregon Health & Science University, told the committee that prescribers already have to jump through many hoops created by insurers’ pre-authorization requirements to get TNF inhibitors for patients with AS and rheumatoid arthritis. “I have to show them the diagnosis, how I came to the conclusion, my chart notes,” Deodhar said of insurers. “My guess is that if this indication becomes available, then that scrutiny is going to become even more and not less.”

Advisory committee members generally agreed that axial SpA would be a suitable indication for clinical development programs but nevertheless cited a need for more data on the rates of spontaneous clinical remission and progression to AS among nr-axSpA patients.

To reduce the risk of misclassification, several panelists favored restricting the indication for axial SpA to a subset of patients with MRI evidence of inflammation. Such a restriction would essentially rely solely on the imaging arm of the ASAS classification criteria and would not encompass patients who only meet the clinical arm’s criteria.

Narrowing Cimzia’s Indication

The committee’s discussions about insurers as gatekeepers for appropriate TNF inhibitor use carried over into its deliberations on the Cimzia and Humira applications.

Although they recommended approval of Cimzia for active axial SpA by the slimmest of margins, panelists sought to add more specificity around the term “active” in the broadly written indication proposed by UCB.

UCB representatives told the committee they intended the indication to reflect the patient population studied in the 325-patient pivotal trial, which encompassed individuals with a high level of disease activity with or without radiographic evidence of AS. Enrollment criteria for the pivotal trial required patients to have a clinical diagnosis of axial SpA, meet modified ASAS classification criteria, have evidence of active disease as demonstrated through disease activity and spinal pain measures, and objective evidence of inflammation in the form of either elevated C-reactive protein and/or MRI evidence of sacroiliitis.

Some panelists who voted for approval said they were confident FDA and UCB could reach agreement on the indication’s wording, while some who opposed approval said they were concerned about the breadth of the proposed statement.

Temporary voting member Donald Miller, North Dakota State University, was among the seven panelists who favored approval and pointed to the gatekeeper role that payer restrictions can serve in assuring that prescribing occurs in accordance with the label.

By linking the indication to imaging data, “insurance companies and third-party payers would really require that evidence before they were willing to pay for these therapies,” Boston University’s Felson said.

“I’m optimistic that we can negotiate an appropriate label, and I think third-party payers will look to that labeling for guidance as to appropriate use and hopefully that will provide appropriate use in practice,” Miller said.

Boston University’s Felson abstained from the approval vote but nevertheless suggested alternative language for the indication statement because “I think we’re all uncomfortable with the wording that has been provided.”

He suggested limiting the indication to patients with imaging-proven axial spondyloarthropathy defined by a positive sacroiliac joint radiograph or MRI evidence of inflammatory lesions in the joint. “Rather than call it active … spondyloarthritis, I think I would label it as we’ve described it, which is imaging positive and define it explicitly,” he said. “You could imagine that if you did that, that insurance companies and third-party payers would really require that evidence before they were willing to pay for these therapies, and I think that that’s probably not unhealthy.”

Humira Study’s Patient Population Unclear

In contrast with Cimzia, AbbVie’s attempt to draw a narrow indication linking Humira use with objective signs of inflammation was not enough to sway the advisory committee that a new indication in nr-axSpA was supported by the evidence.

AbbVie sought a claim for use in adults with active, nr-axSpA with objective signs of inflammation by elevated CRP or MRI, who have had an inadequate response, or are intolerant, to nonsteroidal anti-inflammatory drugs. However, panelists said the targeted population of patients that lack radiological evidence of disease was not sufficiently well defined to permit approval.

The Humira pivotal trial randomized 192 patients who met the ASAS classification to either Humira or placebo. Humira was associated with a statistically significant improvement for both the overall population and the subgroup of patients with objective evidence of inflammation on baseline MRI or elevated CRP (Also see "Humira Shows Benefit In New Arthritis Indications Based On Novel Criteria" - Pink Sheet, 7 Jun, 2012.).

However, FDA raised questions about the patient population studied, noting that 20% of patients in the trial had radiographic evidence of AS, according to a centralized review of patient X-rays. Since Humira already is approved for AS, patients with this condition were supposed to have been excluded from the study. Adalimumab’s efficacy in these AS patients appeared to drive the efficacy results for the overall study population, the agency said.

While FDA highlighted similar concerns about misclassification of AS and nr-axSpA patients in the Cimzia study, they were of less consequence because that trial was intended to include a broad spectrum of patients and the treatment difference from placebo was similar across subgroups, suggesting that no single subpopulation was driving the overall efficacy results.

Another problem for AbbVie was that its proposed restrictions relative to objective signs of inflammation represented a subpopulation of the pivotal trial, with as few as 45 patients in this group. Panelists said a new study was warranted in the subgroup of patients with objective evidence of inflammation.

“I would really want to see a study that has MRI of the [sacroiliac] joints and inflammatory markers in that group with symptoms, because then I think that’s the clinically appropriate group to evaluate and potentially treat,” said temporary voting member Nancy Lane, University of California-Davis.