The Rheumatoid Arthritis That Wasn't: Chronic CPPD Needing an Off-Label DMARD
A single patient spent three years treated for rheumatoid arthritis he never had. The disagreement, once the real diagnosis is confirmed, is which off-label drug to reach for when no agent is actually approved for it.
Walter G., a 74-year-old man, has spent nine years building furniture in a garage workshop, a hobby that started as a way to fill the first months after his wife's death and has since become the thing his grandchildren most associate with him, each one now sleeping in a bed he made by hand. That hobby has grown slower and more frustrating over the past three years as symmetric stiffness and swelling in his wrists and finger joints settled in and never fully let up, forcing him to relearn grip techniques he'd used without thinking for decades. He retired from the postal service around the same time the joint symptoms began, and was diagnosed with seronegative rheumatoid arthritis at another clinic three years ago on the strength of that symmetric pattern and an hour of morning stiffness, cycling through hydroxychloroquine alone, then a brief methotrexate trial he stopped due to nausea, without ever quite bringing his symptoms under control. A referral for persistent disease activity finally prompted repeat serologies and imaging that changed the diagnosis entirely: rheumatoid factor and anti-CCP antibodies both negative, again, and wrist films showing unmistakable chondrocalcinosis — calcium deposition along the triangular fibrocartilage and both wrist joint spaces, a finding never previously imaged for.
Chronic CPPD can present exactly this way, a genuinely symmetric, inflammatory, RF/anti-CCP-negative polyarthritis with morning stiffness that mimics rheumatoid arthritis closely enough to be missed for years — the "pseudo-rheumatoid" phenotype. No drug carries an FDA-approved indication for this presentation specifically, and the evidence for the two agents usually considered points in opposite directions. Hydroxychloroquine holds the only positive randomized result — Rothschild and Yakubov's small double-blind trial reported at least a 30% reduction in flares in 76% of treated patients against 32% on placebo, in a cohort enriched for destructive arthropathy. Methotrexate's support is uncontrolled, open series of five and ten patients, while the one rigorous test it has faced — Finckh's randomized double-blind crossover trial in twenty-six patients — found no significant effect on disease activity at all. Walter's own history complicates the choice further: he's already tried and stopped both once, though under the wrong diagnosis and, in methotrexate's case, possibly the wrong dose or without adequate anti-nausea support.
Two drugs that already failed once, under the wrong diagnosis
His prior hydroxychloroquine trial happened while he was being treated for the wrong disease, which makes it hard to say it really tested the drug against what he actually has. The one randomized trial of hydroxychloroquine in CPPD was small, but it was positive — Rothschild and Yakubov got a response in roughly three quarters of treated patients against a third on placebo. That is more than methotrexate has ever produced under controlled conditions, and given his renal function I'd rather re-try the safer agent properly before reaching for the other one.
I'm not going to pretend the evidence favors methotrexate — Finckh is a negative trial and I'd rather say so plainly. But it enrolled recurrent and persistent CPPD arthritis broadly, at a fixed 15mg weekly for three months a period; it tested the general CPPD population, not the pseudo-rheumatoid phenotype specifically, which is the presentation the uncontrolled series reporting benefit were actually describing. Walter is that phenotype. And I don't think his own prior methotrexate trial was a real test either: six weeks, stopped for nausea, with no confirmation folic acid was ever given or the dose adjusted. A properly supported re-trial is a genuinely different intervention from what happened before.
I take the renal-safety point seriously, but his renal function today is mild to moderate, not a hard contraindication — that's a reason to adjust the methotrexate dose, not to rule it out entirely. What I'm arguing for is a fair test of the one option his record has never actually given a fair test to, not a claim that the literature is on my side.
I think the deciding factor is which of his two prior failures is actually fixable. His nausea on methotrexate is a manageable side effect — dose adjustment, antiemetics, a switch to subcutaneous administration if oral nausea persists. His renal function is a fixed physiologic fact that limits how confidently we can dose methotrexate regardless of how well the nausea itself is managed. That makes hydroxychloroquine, properly re-trialed at an adequate dose and duration this time, the more durable choice, even though neither drug has strong evidence behind it specifically for CPPD.
Agreed: a proper hydroxychloroquine re-trial at confirmed weight-based dosing, with baseline ophthalmologic screening and follow-up in three months to assess whether it controls his symptoms this time now that the correct diagnosis is being treated.
Not agreed: whether a well-managed methotrexate re-trial would have been at least as reasonable a first choice. The second rheumatologist maintains it likely would have, on the argument that Finckh's negative result was generated in a broader CPPD population than his; the group's actual choice rested on the renal-function argument being more durable than the disputed point about whether his original methotrexate trial was ever a fair test. If hydroxychloroquine doesn't control his symptoms, that disagreement will resurface directly.