A Fourth Flare and Two New Diagnoses of Its Own: What a Modest Trial Margin Actually Buys Him
A single patient, on his fourth glucocorticoid taper attempt for polymyalgia rheumatica. The disagreement isn't whether re-escalating steroids would work again — it's whether 'it's worked three times' is still the right reason to do it a fourth.
Douglas F., 66, builds sets for the community theater's mainstage productions, backstage carpentry work that has him overhead with a drill more evenings than not — work that's gotten harder every time his polymyalgia rheumatica has flared, which by now is four times in ten months. There has never been a hint of giant cell arteritis alongside it: no headache, no jaw claudication, no visual symptoms, at diagnosis or at any of the three prior flares — this has stayed, honestly and consistently, polymyalgia and nothing more. Each of the first three flares answered the same way: re-escalate the prednisone, wait for the girdle stiffness and the ESR to settle, taper again, slower. It has worked three times. It is also, each time, cost him something — new-onset hypertension the team now attributes squarely to cumulative glucocorticoid exposure, and cataract surgery scheduled for next month that his ophthalmologist linked directly to the same. He is flaring again now, at 6mg, shoulder and hip stiffness back and his ESR climbing from a taper-nadir of 12 to 54.
Sarilumab was approved for exactly this scenario — relapsing PMR during a glucocorticoid taper — on the strength of a trial that randomized patients who had already done what Douglas has just done a fourth time: flared while tapering. The margin it showed wasn't dramatic. Just under three in ten sarilumab patients reached sustained remission at week 52 against about one in ten on placebo, a real but modest difference, and one that leaves roughly seven in ten sarilumab patients not meeting that trial's strict definition either. What makes the comparison harder to wave off is what re-escalating prednisone a fourth time actually commits him to: not a clean, cost-free return to the familiar, but another several months at a dose that has already produced two separate, documented steroid toxicities in under a year. A modest trial margin against a treatment that keeps working but keeps leaving something behind is not the same choice as a modest margin against a genuinely free option.
The fourth flare, read two ways
Re-escalating prednisone has worked three times. It's familiar, it's cheap, and there's no uncertainty about how he'll respond — we already know. Sarilumab is a newer biologic with its own real risks, infection and lipid changes among them, for a trial margin that wasn't especially large. If it's worked three times, I don't think a fourth flare is obviously the moment to switch course.
I'd weigh 'it's worked three times' differently, because each of those three times also cost him something real — he now has steroid-induced hypertension and is having cataract surgery next month. This isn't his first flare during a taper, it's his fourth, which is precisely the population SAPHYR enrolled: patients who had already relapsed on a taper attempt. That's not an extrapolation, that's a direct population match.
I understand the margin looks modest on paper, but the comparator group in that trial wasn't a placebo doing nothing — it was placebo plus another full year of tapering steroids, which is functionally the same path you're proposing for him now.
The honest read of the trial data is that sarilumab's absolute benefit is real but not large — roughly an eighteen-point difference in sustained remission, not a dramatic one. But the choice in front of Douglas isn't sarilumab against a cost-free re-taper; it's sarilumab against a fourth round of exactly the exposure that has already given him two documented toxicities. Weighed against that specific comparator, rather than against steroids in the abstract, the modest margin is doing real work. I'd start sarilumab, keep a short prednisone bridge only until it takes effect, and leave methotrexate off the table for now rather than adding a third agent before seeing how he does on two.
Agreed: start sarilumab with a short prednisone bridge, aiming to complete the glucocorticoid taper faster than his prior three attempts.
Not agreed: whether methotrexate should be added now as a second steroid-sparing agent or held in reserve — the rheumatologist would rather see sarilumab's own effect first before layering on a second agent with weaker independent evidence in PMR.