Reviews/Psoriasis Psoriatic Arthritis/Upadacitinib ACR50 versus placebo in TNF-inhibitor inadequate responders
Psa JAK · PSA-JAK-01 · v2
The claim asks whether upadacitinib beats placebo on a 50% improvement response in psoriatic arthritis patients whose earlier TNF-inhibitor treatment failed. One phase 3 trial, SELECT-PsA 2, reports that more patients on upadacitinib than on placebo reached that response at 12 weeks [2], and follow-up analyses show the benefit holding through 56 weeks [3][5] and 2 years [6]. The direction of benefit is clear and consistent. But the trial enrolled patients who had failed any biologic drug, a wider group than the TNF-only wording of the claim, and the corpus never states an actual number, confidence interval, or p-value for the 50% response. It is also a single trial with no independent study confirming it. The panel judged the benefit real and durable but held back because of the missing numbers, the broader population, and the lack of replication.
The corpus contains only the SELECT-PsA 2 phase 3 trial and its extensions. Block [2] states that a greater proportion of patients achieved ACR50 at week 12 with upadacitinib versus placebo, and [3] and [5] report that ACR20/50/70 responses were maintained through 56 weeks. Overall superiority over placebo is anchored by ACR20 figures of 56.9% and 63.8% versus 24.1% (p<0.001) in [1], with MDA of 25.1% and 28.9% versus 2.8% (p<0.001). Durability extends to 2 years in [6]. No block reports a numeric ACR50 point estimate, confidence interval, or p-value. The contested point: The ACR50 result is stated only qualitatively as a 'greater proportion' in [2], while the fully quantified endpoint in the corpus is ACR20 in [1]. The population enrolled was biologic-DMARD-IR, broader than the claim's TNF-inhibitor-IR wording, so the recorded benefit is not a TNF-IR subgroup result.
All 6 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 Psoriasis Psoriatic Arthritis, 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.
The claim specifies inadequate response to TNF inhibitors, but every corpus paper enrolled a broader biologic-DMARD-IR population; superiority in the pooled biologic-IR group is not a TNF-IR subgroup result.
ACR50 superiority is stated only qualitatively ('greater proportion') in [2]; the corpus provides no ACR50 point estimate, confidence interval, or p-value, whereas ACR20 carries full statistics.
Where it remains uncertain. The ACR50 signal is qualitative only, with quantified superiority carried by ACR20 (56.9%/63.8% vs 24.1%, p<0.001) [1], and the enrolled population is biologic-IR rather than the claim's TNF-IR [2][3].
Aleatory
The estimate rests on a single trial's patient sample, so ordinary between-study and sampling variation in the ACR50 rate is unquantified with no second trial to bound it.
The load-bearing papers from a corpus of 6, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Anchors magnitude and statistical significance of superiority over placebo in the biologic-IR population (ACR20 as the quantified companion endpoint)
“At week 12, significantly more patients receiving upadacitinib 15 mg and 30 mg versus placebo achieved ACR20 (56.9% and 63.8% vs 24.1%; p<0.001 for both).”
Panel auditInternal validity: medium · scope of inference: Upadacitinib versus placebo for ACR20 and MDA at weeks 12 and 24 in patients with inadequate response to at least one biologic DMARD.Key limitations: ACR50 not reported · population is biologic-IR, not specifically TNF-inhibitor-IR · sample size and design detail absent
Direct statement of ACR50 superiority over placebo in the bDMARD-IR population
“A greater proportion achieved ACR50 and ACR70 at Wk 12 with UPA vs PBO, and more achieved MDA at Wk 24.”
Panel auditInternal validity: medium · scope of inference: Upadacitinib versus placebo for ACR50 at week 12 in patients with active PsA and inadequate response to biologic DMARDs.Key limitations: no number or significance test reported for ACR50 · population is biologic-IR, not specifically TNF-inhibitor-IR · sample size absent
Shows ACR50 response is durable through 56 weeks in the biologic-IR population
“Improvements in the proportion of patients achieving ACR20/50/70, PASI75/90/100, and minimal disease activity were maintained with both doses of upadacitinib through 56 weeks.”
Panel auditInternal validity: medium · scope of inference: Durability of ACR and other responses with upadacitinib through 56 weeks in a biologic-IR PsA population.Key limitations: no placebo contrast · no numbers or significance for ACR50 · population is biologic-IR, not specifically TNF-inhibitor-IR
The abstract states ACR responses were sustained through 2 years but reports neither an ACR50-specific result nor a comparison against placebo, so the claim's contrast cannot be read.
Panel auditInternal validity: medium · scope of inference: Two-year maintenance of ACR, MDA and PASI responses and safety with upadacitinib in biologic-IR PsA.Key limitations: no placebo contrast · ACR50 not separately reported · population is biologic-IR, not specifically TNF-inhibitor-IR
4 of 6 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.
Directionally well-supported by the primary RCT but discounted for single-trial dependence, a biologic-IR population broader than the claim's TNF-IR wording, and absent ACR50 effect statistics.
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.
ACR50 numeric effect in the TNF-IR subgroup. The corpus states ACR50 superiority only qualitatively [2] and never for a TNF-IR-only group, so the claim's exact endpoint and population lack a quantified estimate. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A prespecified TNF-inhibitor-IR subgroup analysis reporting ACR50 rates, difference versus placebo, confidence interval and p-value.
Open question: What is the ACR50 response rate and its difference versus placebo specifically in patients with inadequate response to TNF inhibitors? Why it is still open: Every block enrolled a broader biologic-DMARD-IR population [1][2][3] and none isolates a TNF-IR subgroup.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 18 Sep 2026 · protocol v0.19