Reviews/Rheumatoid Arthritis Targeted Therapy/Tofacitinib cardiovascular and malignancy safety signal reflects a JAK class effect
JAK Inhibitors · RA-JAK-02 · v2
Borderline: within 4 points of the Doubtful/Refuted boundary, so a re-grade could land either side.
This corpus shows the tofacitinib safety signal is real and comes from a randomised trial, but the claim stretches it into two much bigger statements the evidence does not carry. A raised cancer signal does recur across several Janus kinase inhibitor datasets, but the heart-and-stroke half of the claim is actually contradicted by three of the class-level studies, which found no higher cardiovascular risk. No study in this corpus tests the idea that these drugs should be reserved for patients who have already failed biologic treatment, so that policy conclusion has nothing behind it. The one trial that looked at who is most at risk pointed to older age and smoking, not to how many prior treatments a patient had. The panel landed on Doubtful because the claim overreaches on both its cardiovascular limb and its restriction-to-biologic-failure limb.
The tofacitinib signal is established at trial grade: in ORAL Surveillance the hazard ratios were 1.33 (95% CI, 0.91 to 1.94) for MACE and 1.48 (95% CI, 1.04 to 2.09) for cancers versus a TNF inhibitor [1], with the malignancy excess emerging from month 18 onwards [2]. A malignancy signal recurs across the class: a nationwide emulation found malignancy HR 1.61 (1.08-2.41) [6], and the WHO database flagged neoplasms across JAK inhibitors [7]. The cardiovascular limb does not replicate: [6] reported 3P-MACE HR 0.92 (95% CI 0.59-1.42), [7] found no significant increase in major cardiovascular events, and [8] reported MACE HR 0.71, all non-significant. Pharmacovigilance [4] is the only source asserting a cardiovascular class effect, from disproportionate reporting. No paper evaluates restriction to biologic-failure patients [3]. The contested point: The corpus splits on the cardiovascular limb: spontaneous-report data [4] read a class-wide cardiovascular signal, while the cohort and emulation studies [6][7][8] find no significant excess. The malignancy limb is comparatively consistent [6][7], but even the index cardiovascular signal in [1] crosses unity.
All 10 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 Rheumatoid Arthritis Targeted Therapy, 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 is compound and its policy limb is untested: no corpus paper evaluates restriction to biologic-failure patients or its outcomes, so that clause is unsupported by the evidence presented.
The cardiovascular class effect is contradicted, not established: [7] found no significant major-CV increase, [6] 3P-MACE HR 0.92 (NS) and [8] MACE HR 0.71 (NS); only pharmacovigilance [4] asserts a CV class effect, and disproportionality cannot establish causation.
Where it remains uncertain. Three class-level datasets show no cardiovascular excess ([6] 3P-MACE HR 0.92; [7] no significant increase; [8] MACE HR 0.71), and no paper tests restriction to biologic-failure patients, which [3] contradicts by locating risk in age >=65 or smoking.
Aleatory
The ORAL Surveillance MACE HR 1.33 has a 95% CI of 0.91 to 1.94 [1], so sampling variation leaves the tofacitinib cardiovascular effect statistically compatible with no increase.
Epistemic
No corpus paper measures outcomes under restriction to biologic-failure patients, so the policy limb is simply unmeasured [3].
Inferential
[4] infers a cardiovascular class effect from reporting ratios (e.g. VTE adj.ROR=2.11), but disproportionality cannot establish incidence or causation.
The load-bearing papers from a corpus of 10, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Establishes the tofacitinib-specific index signal; the CV HR crosses 1.0, so even the source trial's MACE finding is not statistically definitive.
“The hazard ratios were 1.33 (95% CI, 0.91 to 1.94) for MACE and 1.48 (95% CI, 1.04 to 2.09) for cancers; the noninferiority of tofacitinib was not shown.”
Panel auditInternal validity: high · scope of inference: RA patients >=50 with >=1 additional cardiovascular risk factor on tofacitinib versus TNFi; speaks to tofacitinib, not the JAK class.Key limitations: single JAK inhibitor only · open-label · risk-enriched population · no total N in abstract
Directly undercuts the restriction rationale: risk is stratified by age/smoking, not biologic-failure status.
“In patients 'aged <65 years and never smokers' ('low-risk'), there was no detectable risk increase with tofacitinib versus TNFi (HRs approximately 1.0) up to 6 years of follow-up.”
Panel auditInternal validity: medium · scope of inference: Which RA subpopulations drive the tofacitinib excess risk versus TNFi; speaks to the basis for any restriction, still tofacitinib-only.Key limitations: post-hoc subgroups · single JAK inhibitor · wide HR range · no N in abstract
The strongest corpus support for the class-effect limb, but from spontaneous-report disproportionality, which cannot establish incidence or causation.
“These adverse events were reported following all JAK inhibitors that were studied, suggesting a class effect.”
Panel auditInternal validity: low · scope of inference: Relative reporting frequency of cardiovascular/VTE events across JAK inhibitors versus bDMARDs; hypothesis-generating for class effect only.Key limitations: spontaneous reporting bias · no denominators · notoriety bias after ORAL Surveillance · no causal inference
Confirms elevated malignancy incidence with tofacitinib versus TNFi but examines no other JAK inhibitor, so the class-effect element of the claim is not addressed.
Panel auditInternal validity: high · scope of inference: Malignancy risk in risk-enriched RA patients on tofacitinib versus TNFi; not a cross-JAK comparison.Key limitations: single JAK inhibitor only · open-label · post-hoc time-window analysis · risk-enriched population
4 of 10 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.
The claim overreaches on both limbs — the cardiovascular class effect is contested and the biologic-failure restriction is unsupported and contradicted by the corpus's own subpopulation analysis.
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.
No test of biologic-failure restriction. The claim's policy conclusion depends on outcomes under a line-of-therapy restriction that no corpus paper measures [3]. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A prospective or emulated comparison of JAK inhibitor safety in biologic-naive versus biologic-failure patients, matched on cardiovascular risk.
Open question: Does reserving JAK inhibitors for biologic-failure patients change cardiovascular or malignancy outcomes? Why it is still open: No corpus paper evaluates a biologic-failure restriction or its outcomes [3].
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 18 Sep 2026 · protocol v0.19