From observational studies to personalized treatment in spondylarthritis: focus on IL-17 inhibition

KEYWORDS:

• Spondyloarthritis
• treatment
• anti-IL-17
• anti-TNF

Spondyloarthritis (SpA) is a chronic inflammatory disease involving predominantly the spine and sacro-iliac joints, but with the possibility of peripheral involvement (joints, enthesitis, dactylitis). Extra articular manifestations (EAM) may be associated in a number of cases (uveitis, psoriasis, inflammatory bowel disease: IBD). Therefore, SpA is a multifaceted disease with several phenotypes: axial radiographic SpA (representing ankylosing spondylitis) or non-radiographic axial SpA, peripheral arthritic, peripheral enthesitic SpA [1,2].

Immunopathological knowledge of the disease has led to the development of new targeted therapies, and after anti-TNF [3], targeting of the IL-23/IL-17 axis [4] and then JAK inhibitors [5] have been proposed or are under development in this condition. In this context of expansion of therapeutic means, it is important to have guidance or choice of indication for these different treatments.

1. Value of observational studies and cohorts

This kind of study allows adding information about safety of a new treatment in post-marketing survey. But another benefit is the possibility, in case of longitudinal follow-up, to catch, in a real-life setting, factors associated or even predictive of a good response to treatment or favorable outcome of the disease. Such findings, along with the results of head-to-head studies will pave the way of a predictive or personalized medicine. This is particularly important regarding the cost and the diversity of the therapeutic arsenal available for SpA.

In a recent issue of Expert Opinion on Biological Therapy [6], authors evaluated effectiveness, retention rate, and safety of secukinumab, an anti-IL17A monoclonal antibody, over 1 year in 169 patients with SpA (ankylosing spondylitis and psoriatic arthritis, PsA). This real-life observational study brings confirmation of secukinumab’s efficacy in the several phenotypic forms of SpA with different evaluation tools. Clinical efficacy of secukinumab, assessed by Ankylosing Spondylitis Disease Activity Score with C Reactive Protein (ASDAS-CRP), correlates with biologic inflammation at baseline and absence of smoking. This efficacy seems independent from the line of treatment (no difference if secukinumab was the first line of biologic or second or more), from methotrexate co-medication and from Body Mass Index (BMI). However, efficacy was reduced in female gender and in case of metabolic syndrome. Safety data from this study are in compliance with previously reported results from controlled studies, without any unexpected new adverse event. These results may be compared to data from the literature. Due to the recent availability of anti-IL17 agents, such long-term real-life data are scarce.

A French retrospective, longitudinal, observational, and multicentre study [7] was conducted on 556 patients (338 (61%) axial SpA, 183 (33%) Psoriatic Arthritis), 324 (58%) females, mean age 47 ± 12 years, 221 (40%) smokers, 160 (29%) radiographic sacroiliitis, 238 (43%) Magnetic resonance Imaging (MRI) sacroiliitis, 206 (37%) high CRP, 239 (43%) HLA B27 positive, mean BASDAI 48.9 ± 27. Secukinumab was associated with methotrexate (MTX) in 140 pts (25%) and was the first-line biologic Disease Modifying Anti Rheumatic Drug (bDMARD) in 56 patients (10%). The median drug survival of Secukinumab was 76 weeks. First-line administration was associated with longer survival compared to second-line administration and above: 111 [83–138] versus 69 weeks [57–81] (p = 0.01). Maintenance was not significantly different by gender, disease, associated MTX, elevated CRP, axial SpA vs PsA, and smoking status. Among axial SpA patients, the absence of radiographic or MRI sacroiliitis and normal CRP did not significantly affect the survival of the biodrug.

Another study investigated the retention rates of secukinumab at 6 and 12 months and the proportions of patients in remission after 6 and 12 months of treatment in routine care in Europe [8]. These results were evaluated overall and stratified according to prior use of targeted bDMARD/synthetic anti-rheumatic drugs (ts)DMARDs. A total of 941 axial SpA patients were included. Overall, the retention rate of secukinumab at 6/12 months was 82% to 73% and higher in b/tsDMARD naïve patients compared to non-naïve patients. The same was found for remission rates.

These two studies are in line with the pivotal studies in ankylosing spondylitis (MEASURE 1 and MEASURE 2 studies) finding a better response rate of use of secukinumab as first line bDMARD [4]. A post hoc analysis of pooled data from MEASURE 1 and MEASURE 2 studies on secukinumab in ankylosing spondylitis found that therapeutic efficacy was not related to baseline CRP [9].

More data are available for anti-TNFα agents in SpA [10]. Overall, for both radiographic and non-radiographic axial SpA, the same predictors of good response were identified: male sex, young age, short disease duration, high initial Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), presence of HLA-B27, high initial CRP, and sacroiliac inflammation on MRI. Some factors were found to be associated with poor response to anti-TNF therapy in axial SpA: smoking [11], increased BMI, and obesity [12]. Of interest, these two factors had the same impact in ankylosing spondylitis (axial disease) and psoriatic arthritis (peripheral disease).

Observational studies may help to find out some factors associated with the therapeutic response, or confirm results from pivotal studies. But they may also highlight some differences in the responders’ profile from one biodrug to another pending its target or mode of action.

2. Expert opinion

The results of the work presented in Expert Opinion on Biological Therapy [6] are a contribution to the development of therapeutic strategies in a disease such as spondyloarthritis, with a large clinical and paraclinical polymorphism and a broadening of the range of possible therapeutic options. In this context, the idea is to provide the right treatment for the right patient (and at the right time). At present, international or national recommendations [13] outline strategies, but do not necessarily provide the keys to choice in the absence of evidence-based data.

These observational studies allow the individualization of factors associated with therapeutic maintenance, resulting from the efficacy/tolerance balance. They may catch safety signals not reported during RCTs. Some of these, such as IBD, demyelinating diseases, latent TB, could play an important role in personalized treatment. The work presented here [6] on secukinumab shows small differences with other similar studies [7,8], especially concerning the impact of the treatment line. Differences in the numbers and heterogeneity of patients included in these different studies may be a beginning of explanation of this discrepancy. There are also differences in comparison with anti-TNFα data, especially concerning BMI, CRP, or inflammatory elements in MRI. Presence of objective signs of inflammation (MRI or CRP) at inclusion may avoid overdiagnosis and subsequent enrollment of ‘false’ non-radiographic axial SpA. These differences in the profile of good responder patients according to the targeted therapy illustrate the different involvement of these cytokines in the expression and progression of the disease. This allows us to sketch the beginning of an algorithm for choosing between these classes of biomedical drugs.

These first data concerning secukinumab will have to be confirmed by other studies on the same product, but also with other biological treatments targeting interleukin 17, in particular ixekizumab (another anti-IL-17A monoclonal antibody) and bimekizumab (anti-IL-17A and F) [4]. It will also be necessary to check whether these associated/predictive factors for good response are also observed in the different phenotypic forms of the disease: axial/peripheral, radiographic/non-radiographic, and taking into account the possible presence of extra-articular manifestations. For example, it was unexpectedly observed that therapeutic targeting of the IL-23/IL-17 axis in SpA had different effects depending on the disease phenotype: efficacy of anti-IL23 therapies on peripheral rheumatologic and gut inflammation, but not on axial disease, and efficacy of anti-IL17 therapy on axial and peripheral rheumatologic disease, but not on inflammatory bowel disease [14].

Finally, clinical data from observational studies must be compared and complemented by ongoing head-to-head studies and other theranostic approaches to search for biomarkers predictive of therapeutic response using genomic and metabolomic methods, with the limitations and difficulties of these techniques [15].

Real-life observational studies are therefore useful for the detection of factors associated with the tolerance and efficacy of treatments, laying the basis for a tailored prescription, particularly in SpA, a highly polymorphic disease with a diversity of costly therapeutic options.

Declaration of interest

D Wendling has received speaking fees from AbbVie, BMS, MSD, Pfizer, Roche Chugai, Amgen, Nordic Pharma, UCB, Novartis, Janssen, Celgene, Hospira, Lilly, Sandoz, Grunenthal, Mylan; and has worked on advisory boards for Abbvie, Pfizer, Grunenthal, Novartis; and has received hospitality from Abbvie, Novartis, Roche Chugai, MSD, UCB, and Fresenius Kabi. P Goupille has received speaking fees from AbbVie, BMS, MSD, Pfizer, Chugai, Biogaran, MEDAC, UCB, Novartis, Janssen, Sanofi, Lilly; and has worked on advisory boars for Abbvie, BMS, MSD, Pfizer, Biogaran, UCB, Novartis, Janssen, Celgene, Sanofi, Lilly. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial relationships or otherwise to disclose.

Additional information

Funding

This paper is not funded.