Why a JAK inhibitor was stopped was linked to how the next drug fared

Design

Multicentre observational cohort (ANSWER)

Population

500 treatment courses started after JAK inhibitor discontinuation in RA

Primary outcome

Drug retention by reason for prior discontinuation

Effect

After inefficacy, abatacept vs JAKi discontinuation for ineffectiveness HR 2.83 (95% CI 1.20-6.70)

This analysis used 500 treatment courses from the multicentre ANSWER cohort in which a new biologic or targeted synthetic DMARD was started after a JAK inhibitor was stopped. Next treatments were grouped as TNF inhibitors, IL-6 receptor antibodies, abatacept (CTLA4-Ig) or a second JAK inhibitor, and retention was analysed by the reason the first JAK inhibitor was stopped.

When the first JAK inhibitor failed for lack of effect, abatacept was more likely than a second JAK inhibitor to be stopped for ineffectiveness (HR 2.83, 95% CI 1.20-6.70). When the first was stopped for a toxic adverse event, TNF inhibitors and IL-6 receptor antibodies were less often stopped for toxicity than a second JAK inhibitor. Disease activity trajectories over 12 months did not differ between groups.

This is observational, and choice of next drug reflects clinicians' judgement about each patient. Group sizes for some comparisons were small.

The practical point is to record why a JAK inhibitor was stopped, and let that reason shape the next choice: cycling to another JAK inhibitor looks reasonable after inefficacy, less so after toxicity.

• Record the specific reason a JAK inhibitor is stopped — inefficacy or toxicity.
• After inefficacy, a second JAK inhibitor, TNF or IL-6 receptor inhibitor all looked reasonable.
• After inefficacy, abatacept was stopped for lack of effect about three times as often as a second JAK inhibitor.
• After a toxic adverse event, consider TNF or IL-6 receptor inhibitors over JAK cycling.
• Disease activity at 12 months did not differ between next treatments.

Why it matters

Sequencing after JAK inhibitors is a growing question with little trial evidence, and the reason for stopping is often not recorded.

Don't overread it

This is observational retention data, not a randomised comparison of next treatments.

The statistics, in plain English

A hazard ratio of 2.83 is a large difference, but the interval (1.20-6.70) is wide, reflecting small numbers. Retention is a practical but indirect measure of effectiveness.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.