Advanced Parkinson’s Disease: Subcutaneous Levodopa Infusion or One More Round of Oral Optimization
A patient whose oral regimen has been optimized about as far as it can go, weighing a subcutaneous infusion pump against one more bounded round of oral adjustment before committing to it.
Eleanor R., 71, spent thirty-four years cataloguing rare manuscripts at a university library before retiring, and still volunteers two mornings a week helping the acquisitions department digitize its oldest holdings — precise, deliberate work that her hands have grown less able to do reliably over the past year. She was diagnosed with Parkinson’s disease twelve years ago and has been maintained on carbidopa/levodopa since, currently five oral doses a day plus rasagiline. Her hypertension has been controlled on lisinopril for over a decade without incident. What brought her back to clinic wasn’t a new problem so much as an old one that has stopped responding to the usual fix: over the past eight months, her ‘off’ periods — the hour or more between doses when her hands lock and her gait shortens to a shuffle she has to consciously will herself out of — have grown both longer and less predictable, arriving sometimes forty minutes after a dose that used to reliably hold her for three hours. At the same time, the doses that do work are increasingly working too well, producing peak-dose dyskinesia — writhing movements through her shoulders and neck — that make the fine manuscript work she still loves to do essentially impossible during her best hours rather than her worst.
That combination — a narrowing therapeutic window where both undertreatment and overtreatment happen closer together in time — is the signature of advanced levodopa-responsive fluctuation, and it means the usual next oral move, a modest dose increase or a shorter interval, is likely to buy back ‘off’ time only by giving up more of the ‘on’ time she has left. Her team has already added entacapone alongside her levodopa to extend each dose’s effective window and adjusted her rasagiline timing, without meaningfully widening the gap between too little drug and too much. The 12-week randomized M15-736 trial that supported Vyalev’s approval compared this same population — oral levodopa/carbidopa responders whose fluctuations had outgrown oral dosing — against continuous subcutaneous foscarbidopa/foslevodopa infusion, and found a real, measured difference: patients on the infusion gained meaningfully more ‘on’ time without troublesome dyskinesia than oral optimization alone had delivered, and lost real ‘off’ time in the same window. The question in front of the team isn’t whether her disease has progressed — it plainly has — but whether continuous delivery is the next real step or whether there is still oral room left to try before committing her to a device she would wear through every hour of the work she still does with her hands.
Clinic, six months into a shrinking window
Her fluctuation pattern — a narrowing window where undertreatment and overtreatment now sit closer together — is exactly the population M15-736 enrolled: levodopa-responsive patients whose oral regimen had already been pushed as far as it reasonably goes. In that trial, continuous subcutaneous foscarbidopa/foslevodopa infusion produced a real, measured gain in ‘on’ time without troublesome dyskinesia over oral optimization, along with a real reduction in ‘off’ time. I don’t think we’re looking at a trial extrapolated to her case — she’s who the trial was built around.
I’m not disputing the trial result. What I want on the table before we move is what that infusion actually costs her day to day: in the pivotal trial, infusion-site reactions and infections occurred in at least one in ten patients — common enough that it’s not a footnote — and she’s wearing a device continuously, every hour she’s working with her hands, not just during flares.
None of that means the infusion is wrong for her. It means her own stated goal — preserving the fine motor work she still loves — is a real input the trial’s aggregate ‘on time’ endpoint doesn’t fully capture, and I’d rather we knew we’d exhausted the lower-burden options before asking her to take that on.
There’s a real gap worth naming here: ‘her oral doses have been increased’ and ‘every oral mechanism has been tried’ aren’t the same statement. Entacapone extends each levodopa dose’s effective window by blocking peripheral COMT breakdown — that lever is already pulled. Amantadine ER works through an entirely different mechanism, NMDA receptor antagonism, aimed specifically at the dyskinesia itself rather than smoothing delivery, and she hasn’t tried it.
I’d propose a bounded six-week trial of amantadine ER alongside a switch from immediate- to extended-release carbidopa/levodopa, with an explicit reassessment date already on the calendar — and the infusion education visit scheduled now, in parallel, so nothing is lost if the oral trial doesn’t close the gap.
Agreed: a six-week bounded trial of amantadine ER plus a switch to extended-release carbidopa/levodopa, with an explicit date already set to review her dyskinesia and ‘off’-time diaries. The infusion-education visit is scheduled for the same window rather than deferred until the oral trial fails, so no time is lost either way.
Not agreed: how much ‘off’-time reduction should count as reason to convert to the infusion early rather than waiting the full six weeks. The Geriatrician wanted a specific pre-set trigger — no meaningful diary change by week three. The Neurologist preferred not to pre-commit to a numeric threshold given how much Eleanor’s own fluctuations already vary day to day.