Reviews/TNF Alpha Inhibitors/TNF-alpha inhibitors are effective in ankylosing spondylitis and axial spondyloarthritis
· TN-SPA-01 · v1
Borderline: within 4 points of the Established/Likely boundary, so a re-grade could land either side.
Across this corpus, every study that reports how patients did on TNF inhibitors points the same way: symptoms and disease-activity scores improve, and many patients stay on the drug. The single direct head-to-head-against-placebo signal comes from one adalimumab arm, where more patients hit the response target than on placebo [10], and a very large treatment study found most patients responding [15]. Big real-world registries across several countries agree that substantial fractions of patients reach remission and keep taking the drug [4][11][13]. The main caution is that most of this evidence watches patients on treatment rather than comparing them against an untreated group, so the striking consistency partly reflects how the studies were built. The panel landed at the lower edge of its top tier for exactly this reason: the direction is clear and independently repeated, but the controlled efficacy evidence inside this corpus is thin.
The corpus is uniformly favourable in direction. The clearest controlled signal is the adalimumab reference arm in [10], where ASAS40 reached 36% versus 19% on placebo at week 16, and a 1,250-patient adalimumab study reported ASAS20 of 70.5% [15]. An adalimumab and biosimilar RCT reached ASDAS inactive disease of 0.63 in its favourable-response cluster at week 24 [3], and a cohort showed all disease-activity measures falling significantly by six months (P < .001) [5]. Large independent registries corroborate this: DANBIO reported TNFi retention up to 72% and remission up to 32% [4], twelve European registries pooled a 71% 12-month retention rate [11], and a Swiss cohort reported ASAS40 of 44/34/29% across weight groups [13]. The contested point: The division is whether the corpus demonstrates efficacy or merely consistency on treatment: only [10]'s non-powered reference arm supplies a placebo comparison, while most included papers ([1][4][7][9][12][13][16][17][18][19]) are observational registries or dose-reduction and predictor studies that assume TNFi work. A second fault line is depth of effect: outcomes are surrogate disease-activity indices, and [6] found TNFi did not significantly retard radiographic spinal progression (mSASSS 1.7 vs 1.1, p=0.79).
All 22 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from TNF Alpha Inhibitors, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
Only [10]'s adalimumab arm provides a placebo comparison, and it was a non-powered reference arm rather than the trials' primary test; the pivotal placebo-controlled TNFi efficacy evidence sits in the study-type-excluded guidelines, so the corpus leans heavily on uncontrolled data.
A large share of included papers ([1][4][7][9][12][13][16][17][18][19]) are observational registries or dose-reduction/predictor studies that presuppose TNFi effectiveness rather than test it against a comparator, so consistency partly reflects design rather than confirmed efficacy.
Where it remains uncertain. The sole placebo comparison is one non-powered arm [10]; most papers are observational and outcomes are surrogate indices, with no structural-progression benefit shown [6].
Aleatory
Response varies by subgroup and cannot be eliminated: [3] splits into favourable and less-favourable clusters, and sex, obesity, HLA-B27 and comorbidity modify response [12][13][17][18].
Epistemic
The pivotal placebo-controlled TNFi trials sit in the study-type-excluded guidelines [20][21], leaving only one non-powered placebo arm [10]; more controlled data would reduce this gap.
Inferential
Effectiveness is inferred from on-treatment surrogate indices (ASAS, BASDAI, ASDAS) and open-label or observational designs [4][11][13][15], where regression-to-mean and confounding are not controlled.
The load-bearing papers from a corpus of 22, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Phase III adalimumab RCT in active AS: majority favourable-response cluster reaches 63% ASDAS-inactive disease at week 24
“C1 achieved significantly higher rates of ASDAS-Inactive Disease (ASDAS-ID) than C2 at week 2 (0.21 [95% CI 0.16, 0.26] vs. 0.01 [95% CI 0.00, 0.02]; Risk Difference [RD]: -0.21 [95% CI -0.26, -0.15], p < 0.001) and week 24 (0.63 [95% CI 0.57, 0.69] vs. 0.14 [95% CI 0.09, 0.18]; RD: -0.49 [95% CI -0.57, -0.42], p < 0.001).”
Panel auditInternal validity: medium · scope of inference: Heterogeneity and early divergence of adalimumab response patterns in active AS.Key limitations: no placebo contrast · post hoc clustering · needs external validation
Nationwide registry: substantial first-TNFi 12-month retention (63-72%) in axSpA
“The 12-month retention rates stratified by baseline CRP (low/intermediate/high) were 51%/53%/55% for IL-17i and 63%/66%/72% for TNFi, with similar results when stratified by sex and radiographic status.”
Panel auditInternal validity: medium · scope of inference: Whether baseline CRP predicts TNFi (and IL-17i) retention and remission in routine axSpA care.Key limitations: no untreated comparator · confounding by indication · moderator-focused
Prospective cohort: significant reduction in all disease-activity measures within 6 months of first TNFi
“Disease activity parameters (Bath Ankylosing Spondylitis Disease Activity Index, Ankylosing Spondylitis Disease Activity Score-based C-reactive protein, and Bath Ankylosing Spondylitis Disease Functional Index) and C-reactive protein were significantly decreased at the sixth month of TNFi therapy (all P < .001).”
Panel auditInternal validity: low · scope of inference: Change in disease activity and VEGF over 6 months of first TNFi in axSpA.Key limitations: no control arm · small sample · single-arm pre-post
The abstract reports no difference in response between disease-duration subgroups and gives no TNFi-versus-control contrast from which overall efficacy can be read.
Panel auditInternal validity: medium · scope of inference: Whether TNFi works better in very early than established axSpA in routine care.Key limitations: no untreated/placebo arm · subgroup comparison only · observational
4 of 22 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Consistent, independent primary evidence with a placebo-controlled signal supports effectiveness, but limited controlled data and reliance on surrogate outcomes keep it at the lower bound of Established.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Powered placebo-controlled TNFi efficacy within corpus. Only one non-powered reference arm [10] provides a placebo contrast; the directional claim rests on uncontrolled data. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: Include or retrieve adequately powered placebo-controlled randomised trials of TNFi in AS and axSpA with ASAS40 as the primary endpoint.
Open question: Does TNFi effectiveness in AS and axSpA hold against an adequately powered placebo comparator within this evidence base, not only a single reference arm? Why it is still open: The sole placebo contrast is the non-powered adalimumab arm in [10]; pivotal controlled trials are in the excluded guidelines.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 26 Oct 2026 · protocol v0.19