Now that the interleukin-17 inhibitor secukinumab and tumor necrosis factor inhibitors are competing options for treatment of patients with ankylosing spondylitis, the companies that make those drugs must feel pressure to find some sort of advantage for their agents.
How else to explain the remarkable pair of similar post hoc analyses presented in June at the European Congress of Rheumatology in London? One of the analyses was funded by Novartis – the company that markets secukinumab (Cosentyx) – and included several Novartis employees as coauthors. The second study, presented immediately afterward in the main session at the meeting devoted to ankylosing spondylitis (AS) treatments, had backing from AbbVie, which markets adalimumab (Humira), the largest-selling tumor necrosis factor inhibitor worldwide, and had several AbbVie employees as coauthors.
Both analyses used a “matching adjusted indirect comparison,” a fairly new way to compare the performance of interventions studied in two totally independent trials by propensity matching patients from each of the two trials. It’s purportedly a way to make a legitimate comparison in the absence of head-to-head data.
Making the two reports even more surreal was their use of essentially the same data.
The first report came from Walter P. Maksymowych, MD, an AS clinician and researcher from the University of Alberta, who with his coauthors used data collected on secukinumab in the MEASURE 1 pivotal trial and on adalimumab in the ATLAS pivotal trial. He spent much of his presentation describing the methods behind the indirect comparison, and I don’t think I can be blamed for calling the results of this Novartis-sponsored analysis predictable: overall better performance by secukinumab, compared “indirectly” with adalimumab for clinical responses and patient quality of life.
Mitchel L. Zoler/Frontline Medical News
Dr. Walter P. Maksymowych and Dr. Keith A. Betts
The second report, the one sponsored by AbbVie, came from Keith A. Betts, PhD, a biostatistician who works for the Analysis Group, an international consulting firm. He also used the ATLAS database as the source for adalimumab outcomes, and differed marginally from Dr. Maksymowych by taking data on secukinumab patients from both the MEASURE 1 and MEASURE 2 pivotal trials. Although Dr. Betts also used the matching adjusted indirect comparison approach and broadened his data source modestly, his results showed a distinctly different outcome: similar efficacy for the two drugs. Dr. Betts also included a cost efficacy analysis, and in this part adalimumab showed superior performance after he factored in the cost per responding AS patient.
During the combined discussion period following the two talks, both presenters defended the legitimacy of their approaches, although Dr. Maksymowych conceded that these indirect comparisons are “hypothesis generating rather than producing a definitive answer.” But a couple of active European AS researchers rose to comment from the floor and discredit the whole process.
“These two presentations show why I am not a proponent of indirect comparisons. The statistical models squeeze the data until they confess,” said Robert Landewé, MD, an AS specialist at the University of Amsterdam. “This is now a commercial rather than a scientific clash between two important drugs. I challenge these companies to perform a head-to-head trial. Indirect comparisons are not good,” he concluded, to a round of audience applause.
“There are so many methodological issues,” said Désirée van der Heijde, MD, another Dutch AS clinician and researcher who rose to critique both studies. “The only thing you can rely on is head-to-head trials.”
I later spoke with Dr. Maksymowych, and he expressed some pessimism about the prospects for a fully-powered, head-to-head trial of an interleukin-17 inhibitor and tumor necrosis factor inhibitor because it would need to enroll so many patients. “Randomized studies of active comparators need to be huge because it’s hard to show improvements when the response rates are high,” he said. Plus, he added, it isn’t entirely about a drug’s efficacy against AS spinal symptoms anyway.
“We also have to think about the impact of treatment on other aspects of this disease, such as psoriasis and colitis, as well as radiographic disease progression,” he said. These aspects of the activity of both classes of drugs have not received much study in AS patients until now.
In other words, the battle between treatment options for AS has just begun, and seems likely to be fought on many fronts.
On Twitter @mitchelzoler