Tapering a Working Biologic in a Patient Who Fits Every Predictor of Failing To
She has every reason to want to taper: three years of remission, a second pregnancy she's planning, and a guideline that now supports the idea. She also has nearly every documented risk factor for a taper actually failing.
Sofia M. teaches kindergarten and has spent the three years since her adalimumab-induced remission from axial spondyloarthritis training, in her words, "for the marathon my body wouldn't let me run before" — a literal half-marathon this spring, and a second child she and her partner are hoping to conceive within the year. Both goals sit behind the same question she brought to this visit: can she taper off, or at least down, a drug she's been on for four years and hasn't needed a dose adjustment for since. Her disease has been genuinely, durably quiet — BASDAI under 1.5 at every visit for the past two years, CRP normal, no morning stiffness she can recall well enough to describe.
The 2025 BSR axSpA guideline gives her a real answer to build on: people in sustained remission, generally defined as at least six months of low disease activity, should be offered therapeutic tapering as a shared decision, typically implemented by extending the interval between doses rather than stopping outright. That is genuine, guideline-level support for what she's asking. It sits next to a second, less convenient finding from the same evidence base: across the randomized trials behind that recommendation, female sex, HLA-B27 negativity, a high physician global assessment score, and elevated CRP at baseline were identified as negative predictors of successfully maintaining remission after tapering — and Sofia, HLA-B27 positive herself, still shares one of those four factors outright. One of the same trials, Yates et al., found dose reduction actually failed to show non-inferiority against standard dosing, with roughly half of tapered patients maintaining response compared to eighty percent on standard dose — a genuinely discordant result sitting inside the same literature the group is now trying to apply to her, though the guideline's own reading across the full trial set still lands on non-inferiority overall.
On exam today she has full, pain-free spinal mobility and a negative Schober's test for restriction — a genuinely reassuring baseline against which any early return of stiffness during a taper would be easy to detect rather than lost in ordinary day-to-day variation. She has already mapped her training schedule against a hypothetical taper timeline on her own, unprompted. What none of the evidence can tell the room is the thing it most needs: the predictor list describes which populations taper badly, not which individuals do, and no trial in that set was designed to say whether a woman whose disease has been silent for two years is described better by her sex or by her two years.
What the taper studies actually predicted for her
The 2025 BSR guideline is direct about this: offer therapeutic tapering to anyone in sustained remission, defined as at least six months at low disease activity, as a shared decision between patient and clinician. She's been quiet for two years on every measure we track. The guideline doesn't carve out an exception for sex, and I don't think we should be inventing one that isn't written into the recommendation she's asking us to apply.
The guideline's silence on sex doesn't mean the evidence behind it is silent, though. The same body of tapering trials that produced that recommendation also identified female sex, HLA-B27 negativity, high physician global score, and elevated baseline CRP as negative predictors of successfully staying in remission after a taper. She's HLA-B27 positive, which cuts the other way, but she's also female, which is a named factor in the wrong direction — not a hypothetical risk, a documented one from the same literature we're both citing.
And I'd flag directly: one of the actual randomized trials in that same evidence base, Yates et al., failed to show non-inferiority for dose reduction at all — about half of tapered patients maintained response against eighty percent on standard dose. "The guideline supports tapering" is true and also sits next to a real negative trial inside its own reference list. I'll concede the obvious rejoinder before you make it: Yates didn't require a fixed duration of remission before randomizing, which is the leading explanation for why it dissents from the rest of the set. That cuts in her favor. It doesn't make the negative-predictor list go away.
I don't think this needs to resolve to "taper" or "don't." She has roughly a year before she wants to conceive, which is exactly the kind of runway that makes a cautious, monitored taper reasonable even granting the negative-predictor concern. Start by extending her dosing interval, not stopping the drug outright; recheck BASDAI and CRP at each extension; and set a plan to reverse the taper immediately at the first sign of flare, well before she's actually trying to conceive. That gives us real information about whether she's one of the successful tapers or one of the Yates-pattern failures, on a timeline that still leaves room to correct course.
Agreed: begin an extended-interval taper of adalimumab with BASDAI/CRP reassessment at each step, and revert to standard dosing immediately at any sign of flare, well ahead of any conception attempt.
Not agreed: how much the negative-predictor data should have weighed against attempting the taper at all. The clinical pharmacologist would have preferred waiting for a longer stretch of documented remission, given her named risk factor, before starting; the rheumatologist and maternal-fetal medicine specialist judged the year-long runway and close monitoring plan sufficient to proceed now, leaving the disagreement about threshold, not method, on the record.