Fingertip Sweat Patch Tracks Levodopa Levels in PD

A soft, wearable fingertip patch continuously monitored levodopa in passively collected sweat without requiring exercise or external power, with results that closely correlated with gold-standard blood testing in Parkinson’s disease (PD).The pilot study included five healthy participants who consumed fava beans, three healthy participants who took a single dose of immediate-release levodopa/carbidopa, and four participants with PD who took levodopa as prescribed. In participants with PD, patch-measured levodopa levels aligned with improving motor symptom scores and captured individual differences in drug clearance.“In real time, a patient is not able to know when the dosage is going down, when it’s suboptimal, when it’s optimal,” first author Tamoghna Saha, PhD, a postdoctoral scholar in the Aiiso Yufeng Li Family Department of Chemical and Nanoengineering at UC San Diego in La Jolla, California, told Medscape Medical News. “And if they want to know, they have to give their blood samples to a clinician who would take a week or two to give back results.”The study was published online on July 27 in the Proceedings of the National Academy of Sciences.Managing a Moving TargetLevodopa remains the most effective treatment for PD motor symptoms, but its management grows increasingly difficult as the disease progresses. Early on, patients respond well to small doses, and the margin between too little drug and too much — the therapeutic window — is wide.Over time, dopamine-producing neurons die, reducing the brain’s capacity to store and buffer the drug.“As the disease progresses, the therapeutic window gets smaller and smaller,” study co-investigator Irene Litvan, MD, Tasch Endowed Professor of Neurology and director of Parkinson and Other Movement Disorders Center at UC San Diego, told Medscape Medical News. “It gets to a point in which the amount that you provide may be enough for the patient to do well, but suddenly they get off, or they have an excess of medication.”Clinicians currently adjust levodopa therapy based on patients’ self-reported symptoms and brief office observations, an approach Litvan called “challenging,” in part because patients themselves often struggle to describe their symptoms accurately.“Many people don’t know how to differentiate one from the other,” she said, referring to the distinction between undertreated tremor and overtreated dyskinesia.Blood-based monitoring via high-performance liquid chromatography (HPLC) is accurate but impractical in routine care.Levodopa Monitoring via the FingertipThe patch, called the S-CDM, sits on the fingertip. The finger was targeted by the researchers due to its high density of sweat gland. A soft hydrogel draws sweat passively from the skin through osmosis, eliminating the need for physical activity or electrical stimulation.The sweat flows through a serpentine paper channel past a two-electrode biosensor made of a nickel anode and a tyrosinase-coated carbon cathode. The sensor generates a voltage signal proportional to levodopa concentration without consuming external power. A flexible circuit board transmits data wirelessly via Bluetooth.“When we are dealing with Parkinson’s, we are dealing with older people,” Saha said. “The point is to minimize exertion, to minimize any sort of external stimulation as much as possible.”To convert sweat signals to estimated blood concentrations, the researchers developed a personalized calibration model requiring one initial blood draw. The resulting coefficient was stable across repeated tests in the same individual, suggesting it could eventually be stored in a mobile app and eliminate the need for ongoing blood draws. Participants rated comfort at 4.5 out of 5 on a standardized scale.Drug Levels Reflect Motor SymptomsLevodopa levels estimated from sweat correlated strongly with HPLC-measured blood concentrations, with Pearson coefficients of 0.85 in healthy participants and 0.89 in those with PD.In the four PD participants, motor symptom scores — assessed every 10 minutes using a modified MDS-UPDRS Part III — decreased as levodopa levels rose, with the lowest scores occurring within 10-30 minutes of peak drug concentrations.Levodopa cleared from the blood faster in PD participants than in healthy individuals, though overall bioavailability was similar between groups. Blood pressure showed predominantly hypotensive trends during drug absorption, consistent with levodopa’s known cardiovascular effects.A support vector machine algorithm trained on sweat-derived concentrations, blood pressure, and motor scores predicted blood levodopa levels with an R2 of 0.85 and a mean absolute error of 2.02 µM against HPLC measurements. The model identified sweat readings as the most influential predictor, followed by diastolic blood pressure.Both investigators emphasized that the results are preliminary. “This is all pilot data. I want to make sure that is not construed as we have already resolved the problem,” said Litvan.She said validation in a larger, more diverse population across multiple visits and conditions will be needed before the device can reach the clinic.Saha said the group has received approval to test the patch with extended-release formulations and multiple daily doses and is working to miniaturize the platform into a ring form factor.An ‘Insulin Pump’ for PD?The technology draws a clear parallel to advances in diabetes management, said Alberto Espay, MD, MSc, professor and endowed chair of the James J. and Joan A. Gardner Family Center for Parkinson’s Disease and Movement Disorders at the University of Cincinnati in Cincinnati. Espay was not part of the study.“We’ve always said that levodopa is to Parkinson’s what insulin is to diabetes. We now have infusion systems that mimic the insulin pump, but we don’t do closed-loop pumping, in part because there hasn’t been a way to continuously measure levodopa. This is the first step in that direction,” Espay told Medscape Medical News.However, Espay cautioned that blood levodopa must still cross into the brain and undergo conversion to dopamine before patients experience relief. That conversion is not captured by the peripheral monitoring this sensor performs. “The changes in function are not tightly connected necessarily to blood levels. They are connected to brain levels, and not even brain levels of levodopa, but brain levels of dopamine,” he said.Litvan agreed that a closed-loop system — pairing continuous monitoring with a drug infusion pump — remains a distant goal but said her team is in active discussions with European pump manufacturers to explore the possibility. “That’s what we call the closed loop. That will take a long time.”Espay said the proof of principle was encouraging but that the harder validation work lies ahead. “What is left to do is the hardest work: Showing that what you can do in four people is ultimately a good reflection of brain levels and, with that, of the response,” he said.The study was supported by the UC San Diego Center for Wearable Sensors and the National Institutes of Health (1R01NS141451-01). HPLC analysis was performed by the Emory HPLC Bioanalytical Core. The authors reported no competing interests. Espay reported having no relevant financial relationships.