Abstract
Parkinsonism-dominant multiple system atrophy (MSA-P) is typically a progressive disorder with poor responsiveness to levodopa and an unfavorable prognosis. However, in certain cases, the response to levodopa can be as robust as in Parkinson’s disease (PD), with severe motor fluctuations developing during treatment. Unlike PD, no established therapy exists to maintain activities of daily living (ADLs) in such patients. We present three cases of young-onset MSA-P who demonstrated sustained levodopa responsiveness and were treated with levodopa–carbidopa intestinal gel (LCIG) following the emergence of disabling motor fluctuations. In all three patients, parkinsonism was the predominant symptom from onset until LCIG initiation, with only mild autonomic or cerebellar symptoms. Prior to LCIG introduction, their motor complications closely resembled those of advanced PD. LCIG therapy successfully reduced “off” time and dyskinesia in all cases. However, long-term follow-up revealed a gradual decline in ADLs due to disease progression. These cases suggest that LCIG may be a valuable treatment option for selected MSA-P patients with preserved levodopa responsiveness.
Keywords
Introduction
Parkinsonism-predominant multiple system atrophy (MSA-P) is a fatal neurodegenerative disease with very poor prognosis that commonly develops in middle age. It generally presents with severe autonomic symptoms, parkinsonism with poor levodopa responsiveness, and/or cerebellar symptoms. 1
As for the details of the response of MSA-P to levodopa, it has been reported that the response to levodopa is constant in approximately 30% of patients in the very early phase of the disease and then promptly diminishes. Although >80% of patients with young-onset MSA (YOMSA) respond to levodopa, >50% of them experience wearing-off phenomenon and levodopa-induced dyskinesia (LID) similar to patients with advanced Parkinson disease (PD). 2 YOMSA is indistinguishable from PD with concerning levodopa responsiveness. It can be difficult to diagnose MSA-P in the initial stages, in the case that few cerebellar or autonomic symptoms other than Parkinsonism have appeared. Device-aided therapies (DATs) such as deep brain stimulation (DBS) and levodopa–carbidopa intestinal gel (LCIG) are considered for PD patients with motor complications for which control with oral medications is insufficient. However, DATs are not currently a positive choice in MSA similar to advanced stage PD, as DBS treatment studies for MSA have reported negative conclusions. 3
In this study, the authors report three cases of long-term levodopa-responsive MSA-P in which motor fluctuations such as wearing off are effectively treated with LCIG, and discuss advantages of LCIG treatment for MSA.
Material and methods
Out of a total of 59 patients who underwent LCIG at our hospital from 2016 to 2021, we selected three patients who were finally diagnosed with MSA. The diagnosis of MSA was based on the MSA diagnostic criteria of the movement disorder society, 2022 edition. 4 We included the MDS 2015 PD diagnostic criteria, 5 because in these cases, the diagnosis or important differential diagnosis was PD from the onset to the induction phase of LCIG treatment.
Case presentations
Characteristics of MSA is shown in Table 1.
Characteristics of MSA patients with LCIG.
HY, Hoehn–Yahr; LCIG, levodopa–carbidopa intestinal gel; LEDD, levodopa equivalent daily dose; MDS-UPDRS, Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale; MSA, multiple system atrophy; PD, Parkinson’s disease.
Case 1
A 45-year-old man developed subtle action tremors and bradykinesia of his right arm. He was diagnosed with “clinically established PD” and prescribed levodopa. Brain magnetic resonance imaging (MRI) showed no abnormalities. 123I-metaiodobenzylguanidine (MIBG) myocardial scintigraphy revealed no cardiac sympathetic neuronal loss. He experienced wearing-off and LID in his right arm and neck 4 years later. Neurogenic bladder that required urinary catheterization occurred at the age of 50. His physicians considered the neurogenic bladder was caused by diabetes mellitus that had been treated with insulin injections for more than 10 years. At the age of 51, he experienced frequent wearing-off symptoms despite levodopa–carbidopa treatment administered more than six times per day. In addition, both selegiline and entacapone had been introduced to optimize oral therapy and achieve continuous dopaminergic stimulation (CDS), but these adjustments did not sufficiently alleviate the motor fluctuations. He was able to walk independently in an “on” state, though not in an “off” state. In consideration of the burden imposed by the above wearing-off regarding activities of daily living (ADLs), DATs was discussed by the patient and his physicians.
During the detailed evaluation that was required for considering the indication for DATs, intravenous administration of levodopa (200 mg) dramatically improved his parkinsonism, and his score on the Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) Part III was improved from 47 to 25 points (47% improvement). His finger-nose test revealed bilaterally mild decomposition, although there was no nystagmus, ataxic dysarthria, pyramidal tract signs, or orthostatic hypotension at all. Score on the Odor Stick Identification Test for Japanese (OSIT-J) was decreased to 6 (mean number of correct answers was 9.50 for a male in his 50s). Although brain MRI revealed slight cerebellar atrophy, there was no brainstem atrophy and no signal changes in the cerebellar legs or putamen (Figure 1(a)). Cerebral blood flow (CBF) measured with [123I] N-isopropyl-p-iodoamphetamine (123I-IMP) in single-photon emission computed tomography (SPECT) showed no hypoperfusion in the cerebellum, brainstem, basal ganglia, or occipital lobes, although there were areas of hypoperfusion in the superior frontal gyrus (Figure 1(b)). Genetic testing for hereditary spinocerebellar ataxia (SCA) associated with triplet repeats expansion, that is, SCA1, 2, 3, 6, 7, 8, 12, 17, 31, and DRPLA, was negative. Further analyses of genes associated with hereditary PD revealed no pathological significance.
(a) T2 weighted images (T2WI) of Case 1 taken at the introduction of LCIG. No MSA-specific imaging changes were noted. (b) 123IMP-SPECT images at the same time as (a) show no evidence of hypoperfusion in the cerebellum, brainstem, or basal ganglia. (c) T2WI showed marked pons atrophy and hot cross bun’s sign. There is also atrophy of the left putamen. (d) T2WI of Case 2 already showed a hot cross bun’s sign and high intensity area of middle cerebellar peduncle on T2WI taken around the time of LCIG introduction. There was no atrophy of the basal ganglia. (e) T2WI of Case 3 showed no MSA-specific changes. Atrophy of the brainstem was not seen. There was iron deposition in putamen, though it was not accompanied by atrophy and did not meet the definition of a slit sign. (f) 123IMP-SPECT images of Case 3 show no evidence of hypoperfusion in the cerebellum and brainstem.
His diagnosis at this point was Parkinson’s disease, based on levodopa responsiveness and imaging findings consistent with PD. Urinary retention was attributed to diabetic autonomic neuropathy. The most significant obstacle in his life was the fluctuation of motor symptoms, which were the off state. Therefore, continuous levodopa treatment was expected to dramatically improve his ADLs, which was severely impaired by prolonged “off” time. LCIG treatment was initiated by the strong will of the patient and his family. LCIG eventually reduced off time to less than 2 h per day. The levodopa equivalent daily dose (LEDD) was reduced from 2042 to 1447 mg/day after LCIG initiation. At the age of 54, his trunk ataxia symptoms progressed, which made him unable to gait even under LCIG treatment. Thus, LCIG was discontinued and replaced with oral levodopa therapy. His brain MRI taken around that time demonstrated marked atrophy of the cerebellum and brainstem, and these findings were consistent with MSA (Figure 1(c)). At the age of 55, the patient died of candidemia. He was definitively diagnosed as MSA based on an autopsy.
Case 2
A woman developed gait disturbance due to bradykinesia and dystonia of both legs, and muscle rigidity of her extremities at the age of 36. These symptoms were relieved with levodopa treatment, with no apparent asymmetry between the left and right sides. Her initial diagnosis was consistent with “clinically probable PD.” At the age of 39, motor fluctuations such as end-of-dose akinesia and peak-dose LID with choreic movements of the four limbs appeared. At the age of 41, her motor fluctuations worsened and her off hours were more than 4 h per day, though she was treated with levodopa preparations ⩾ 5 times/day. In addition, oral therapy had been optimized with adjunctive use of both selegiline and entacapon, but these were insufficient to control her motor complications. During the evaluation for indications of DATs, besides the presence of parkinsonism, she exhibited a positive Babinski sign in the right leg and mild limb-kinetic ataxia, without truncal ataxia or dysarthria. She also demonstrated no autonomic dysfunction. Brain MRI showed hot cross bun’s sign in her pons and high intensity area of middle cerebellar peduncle (Figure 1(d)). 123I-MIBG myocardial scintigraphy was performed and revealed a normal H/M ratio (early: 2.87; delayed: 3.43), suggesting preserved cardiac sympathetic innervation. According to the diagnostic criteria for PD, her disease violated the absolute exclusion criteria. 5 On the other hand, as MSA, her disease was categorized into prodromal MSA based on the 2022 diagnostic criteria as for levodopa responsiveness. 4 The physicians assessed that continuous administration of levodopa could significantly improve her ADLs, even if her diagnosis was MSA-P. Therefore, they decided to adopt LCIG therapy. At the age of 42, LCIG infusion was initiated and reduced the severe “off” state symptoms. At the age of 45, since her balance disorder worsened, she came to need assistance with gait despite of LCIG treatment being continued. Now aged 47, progression of the primary disease made it difficult for the patient to gait, and she is now wheelchair-bound. The levodopa response has been decreasing though it still is effective for swallowing. Thus, her diagnosis was changed to “clinically established MSA.”
Case 3
A man developed resting tremor in right upper limb at the age of 44. He had right-predominant muscle rigidity and akinesia, and these symptoms responded well to levodopa administration. He was diagnosed with “clinically established PD.” At the age of 48, wearing off and peak-dose dyskinesia appeared. The “off” time became gradually prolonged, and he became unable to gait without assistance in “off” state. He also had severe diphasic dyskinesia in both legs and difficulty walking due to it during the 30 min before turning “off” state. He was prescribed amantadine for his dyskinesia. However, the amantadine was discontinued as it caused drug-induced hallucinations such as mistaking lint for insects and seeing strange men in the room. To optimize oral therapy, entacapone had been added to his regimen, but it failed to adequately address his motor complications. Since the above motor fluctuations severely interfered with his independent living, his physicians considered the indication for DATs. At the age of 49, he complained of myoclonus on both upper extremities in addition to resting tremor and occasional urinary retention that required catheterization, though orthostatic hypotension and cerebellar ataxia symptoms were not present. Brain MRI revealed no abnormalities suggestive of MSA (Figure 1(e)). Cerebral blood flow in evaluated by [123I] IMP scintigraphy was reduced in the bilateral posterior lobe, though there was no reduction in cerebellar or basal ganglia blood flow (Figure 1(f)). 123I-MIBG myocardial scintigraphy showed a normal H/M ratio (early: 2.70; delayed: 2.60), indicating no evidence of sympathetic denervation. Overall, he was diagnosed with possible MSA. LCIG treatment reduced the “off” time and the diphasic LID symptoms improved remarkably, which enabled him to walk continuously and urinate unassisted with abdominal pressure. ADLs assessment with the Barthel index showed a marked improvement from 55 to 90 points.
Discussion
All three MSA patients in our department received LCIG treatment. Their onset age was relatively young, and all three were initially diagnosed with Parkinson’s disease based on the presence of limb bradykinesia and muscle rigidity; in one case, resting tremor was also observed. Their parkinsonian symptoms demonstrated a marked response to levodopa, and a pronounced wearing-off phenomenon appeared 5–6 years after the onset of the disease. Compared to typical multiple system atrophy, Parkinson’s disease mimics such as our patients have been reported to have good levodopa responses and a high frequency of visual hallucinations. 6 LIDs were also observed in all cases. A high degree of dopaminergic neuron loss and postsynaptic cell survival in the putamen is possibly a prerequisite for levodopa responsiveness and dyskinesia in MSA. 7 Even when these patients were treated with high oral doses of levodopa (>1000 mg), they were unable to maintain activities of daily living due to prolonged off-time during which they were unable to walk on their own. However, LCIG treatment with continuous levodopa administration improved the wearing-off phenomenon in all patients and significantly prolonged on time. Both CGI (clinical global impression) and PGI (patient global impression) were markedly improved.
Previous reports have also described the use of LCIG in patients later diagnosed with MSA. In a case series by Nilsson et al., two patients who were initially diagnosed with Parkinson’s disease received LCIG for more than 13 months before their diagnosis was revised to MSA. 8 More recently, a study involving seven MSA patients treated with LCIG was published, in which the average duration of LCIG treatment was 29.6 months (range: 4–55 months). 9 In that study, younger patients tended to continue LCIG for longer periods, while patients over 60 years of age had a poor prognosis and discontinued within 20 months. In all cases, treatment cessation was attributed to progression of the underlying disease. Together with our cases, these reports suggest that MSA can initially mimic PD, with good levodopa responsiveness, and that LCIG may provide temporary benefit before atypical features emerge and limit therapeutic efficacy.
Our cases did not exhibit poor prognostic factors for MSA, such as severe autonomic symptoms in the early disease course or older age at onset, and they were able to continue LCIG treatment for a relatively long period compared to previous reports. 10 However, treatment was not continued beyond 5 years in either the previous cases or ours. LCIG treatment may be considered in carefully selected MSA patients with good levodopa responsiveness and without clear poor prognostic factors. Based on our experience, several clinical features may help identify MSA-P patients who could benefit from LCIG despite their atypical diagnosis. These include relatively young age at onset, a gradual disease progression compared to typical MSA, and a marked responsiveness to levodopa—often indistinguishable from that seen in idiopathic Parkinson’s disease. Furthermore, in cases where oral dopaminergic therapies, including combination regimens aiming for CDS, fail to sufficiently control motor fluctuations or dyskinesias, LCIG may be considered. Even in the presence of mild autonomic or cerebellar symptoms, if these are not significantly impairing the patient’s daily functioning, LCIG can offer a meaningful improvement in quality of life during a specific therapeutic window. In addition, although evidence remains limited, continuous dopaminergic stimulation may have a stabilizing effect on blood pressure regulation in MSA patients with autonomic dysfunction, and this possibility warrants further investigation.
As MSA progresses and cerebellar ataxia and autonomic dysfunction become more prominent, the effectiveness of levodopa in terms of ADLs is obviously reduced. Therefore, it is necessary to select MSA patients for whom LCIG treatment is appropriate and to initiate it at the best timing so as to maximize and prolong the potential benefit. Additionally, the presence of PEG-J access may serve as a future advantage in patients who are likely to develop dysphagia, as it can facilitate enteral nutrition and hydration. However, given that the clinical benefits of LCIG in MSA may be limited in duration due to disease progression, the cost–benefit balance of this therapy should be carefully considered on an individual basis.
Nevertheless, several limitations of this study should be acknowledged. First, this is a retrospective case series with a small sample size, and the findings may not be generalizable to the broader population of patients with MSA. Motor improvement was evaluated based on clinical records and physician impressions, without consistent use of standardized measures such as MDS-UPDRS Part III or IV. Second, while our findings suggest that LCIG can be beneficial in selected MSA-P patients, only a small subset of such patients—those with relatively slow disease progression, good levodopa responsiveness, and limited nonmotor symptoms—are likely to benefit. Furthermore, longitudinal evidence supporting the efficacy and durability of LCIG in MSA remains lacking. Future prospective studies are needed to validate our observations and to better define patient selection criteria.
In this study, LCIG therapy improved motor fluctuations of three levodopa-responsive patients with YOMSA. Since LCIG treatment for patients with MSA is off-label, patient selection and the timing of LCIG initiation need to be thoroughly discussed. Despite these limitations, LCIG may represent a viable treatment option for maintaining ADLs and quality of life in a carefully selected subset of MSA patients.
