Parkinson’s disease (PD) is a neurodegenerative disorder characterized by the progressive degeneration of dopaminergic neurons. Chronic dopamine (DA) replacement therapy remains the most effective method for treating the cardinal clinical symptoms of PD, including bradykinesia, tremor, and rigidity [1]. However, as PD progresses, the buffering capacity of striatal dopaminergic neurons decreases, and brain DA concentrations increasingly depend on plasma levodopa (LD) concentrations, resulting in “wearing-off” and dyskinesia [2]. Foslevodopa/foscarbidopa is a new soluble formulation of LD and carbidopa (CD) prodrugs delivered as a 24-h/day continuous subcutaneous infusion (CSCI) for the treatment of motor fluctuations in patients with advanced PD. In a previous clinical trial, treatment with foslevodopa/foscarbidopa resulted in a decrease in off-time (-3.5 hours) and an increase in on-time without troublesome dyskinesia (3.8 hours) [3]. The most common adverse event leading to the discontinuation of CSCI was hallucination (n = 42, 17.2%) [3]. Here, we report the case of a 75-year-old man with PD who exhibited psychosis while receiving CSCI and an unexpectedly high plasma LD concentration without a worsening of dyskinesia. The clinical characteristics and course of his PD are summarized in Figure 1.

The patient had been diagnosed with PD at the age of 61 years. His past medical history included mild cardiac hypertrophy without any symptoms. His first PD symptom was right-sidedominant bradykinesia in his upper limbs. Motor symptoms showed a good response to antiparkinsonian medications for 5 years, but the patient’s disease slowly progressed. Wearing-off and dyskinesia emerged at the age of 66 years. He started taking LD five times a day at the age of 68 years. At the age of 75 years, diurnal fluctuations in symptoms and dyskinesia worsened with the emergence of delayed-on states, prompting the switch to deviceaided therapies, including deep brain stimulation (DBS) and LD/CD continuous infusion gel (LCIG) therapy. His advanced age (>70 years) was a contraindication for DBS, and his camptocormia made it difficult to start LCIG therapy. Therefore, CSCI was initiated. At that time, tests scores were normal: MiniMental State Examination, 30; Montreal Cognitive Assessment (Japanese version), 28; and noise pareidolia test, negative. Neuropsychological tests, including Raven’s colored progressive matrices, the digit span task, Ray Auditory Verbal Learning test, Stroop test, and word fluency test, revealed that his cognitive function abilities, including visuospatial cognitive ability and working memory, were retained except for word recall. On the LD challenge test, the improvement rate in the Movement Disorders Society Unified Parkinson Disease Rating Scale (MDS-UPDRS) Part III score was 33.3% (MDS-UPDRS Part III scores of 46 [on state] and 69 [off state]), which was less than expected, apparently because of his camptocormia, but revealed good the responsiveness of his bradykinesia and rigidity to CSCI. The daytime plasma LD concentration determined by high-performance liquid chromatography was 1,000 ng/mL immediately before LD/CD, 1,200 ng/mL one hour after LD/CD, and 1,800 ng/mL two hours after LD/CD. On the basis of the dose of LD/CD and a catecholO-methyltransferase (COMT) inhibitor (opicapone; multiplying the sum of LD equivalents not by 1.33 but rather by 1.5), which corresponds to 1,125 mg as the levodopa equivalent daily dosage (LEDD), foslevodopa/foscarbidopa was introduced at an infusion rate of 0.37 mL/hr (62.9 mg/hr as LD/CD) and at the lower alternative infusion of 0.15 mL/hr according to previous data (Day 1) [4,5]. The concomitant medications used were rotigotine transdermal patches (18 mg), rasagiline (1 mg), zonisamide (25 mg), and istradefylline (20 mg). One week after the initiation of CSCI, the off-time had decreased from 2 hours to 0.5 hours, with an MDS-UPDRS Part III score of 50 (Day 8). Afterward, the patient was hospitalized due to an acute exacerbation of chronic heart failure (Day 32); CSCI became less effective at that point, likely due to subcutaneous edema, with an MDS-UPDRS Part III score of 55. As his symptoms of heart failure were alleviated with medical treatment, the effect of CSCI improved. However, as the effect of the CSCI became obvious, the patient appeared agitated, and visual hallucinations and delusions gradually developed (Day 52). The rate of CSCI was decreased from 0.37 mL/hr to 0.26 mL/hr over two weeks, but the psychiatric symptoms never improved. Interestingly, dyskinesia emerged only when the patient was moving or talking and was mild in that situation. Although the rate of CSCI can be adjusted in increments or decrements of 0.01 mL/hr to achieve more precise LD dosing [4,6], the CSCI rate was adjusted based on the severity of motor and psychiatric symptoms and the dose of dopamine agonists (Figure 1). At the end of this period of tapering (Day 66), it became necessary to stop CSCI altogether because of the persistence of visual hallucinations and delusions. His plasma LD concentrations at the two infusion rates were unexpectedly high, at 5,100 ng/mL (0.37 mL/hr) and 3,600 ng/mL (0.26 mL/hr), respectively, and were likely the source of his psychiatric symptoms.

Foslevodopa/foscarbidopa has been shown to alleviate motor fluctuations, with benefits in both on-time without troublesome dyskinesias and off-time [5]. A previous study characterizing the LD pharmacokinetics of CSCI in PD patients revealed that average steady-state LD concentrations ranged from 747 ng/mL to 4,660 ng/mL across groups receiving four different infusion rates [6]. Although the optimal LD concentration varies widely, this concentration was higher in our patient than in previous LD concentration profiles regarding therapeutic and dyskinetic concentrations [7]. The initial infusion rate calculated via the LEDD corresponding to the LD/CD and COMT inhibitors and increases of more than 0.01 mL/hr can also lead to an unexpectedly high LD concentration. In addition, the difference in the drug delivery route (i.e., subcutaneous infusion rather than oral intake) may lead to unexpectedly elevated LD concentrations even in terms of the LEDD. In the adjustment of the CSCI rate, neurologists should detect dopamine overload by screening patients for symptoms, including headache, insomnia, agitation, hypomania and distractibility, even in the absence of peak-dose dyskinesia. Although there are no reports about heart failure induced by CSCI, LD treatment increases serum homocysteine levels [8], and CSCI may aggravate any comorbid cardiac conditions in advanced-stage PD patients, such as our patient.

In summary, we report the case of a PD patient who exhibited psychosis without a worsening of dyskinesia while on CSCI, accompanied by unexpectedly high plasma LD concentrations. In advanced-stage PD patients, it is safest to start foslevodopa/foscarbidopa at a low infusion rate and adjust the infusion rate carefully.

Citations to this article as recorded by