Pharmacokinetics, Safety, and Tolerability of Orally Administered Ursodeoxycholic Acid in Patients With Parkinson’s Disease—A Pilot Study

Abstract

Mitochondrial dysfunction is implicated in the pathogenesis of Parkinson’s disease. Preliminary data have shown lower brain adenosine triphosphate (ATP) levels in Parkinson’s disease versus age-matched healthy controls. Ursodeoxycholic acid (UDCA) may improve impaired mitochondrial function. Our objective was to evaluate UDCA tolerability, pharmacokinetics, and its effect on brain bioenergetics in individuals with Parkinson’s disease. An open-label, prospective, multiple-ascending-dose study of oral UDCA in 5 individuals with Parkinson’s disease was completed. A blood safety panel, plasma concentrations of UDCA and UDCA conjugates, and brain ATP levels were measured before and after therapy (week 1: 15 mg/kg/day; week 2: 30 mg/kg/day; and weeks 3–6: 50 mg/kg/day). UDCA and conjugates were measured using liquid chromatography–mass spectrometry. ATP levels and ATPase activity were measured using 7-Tesla ³¹P magnetic resonance spectroscopy. Secondary measures included the Unified Parkinson’s Disease Rating Scale and Montreal Cognitive Assessment. UDCA was generally well tolerated. The most frequent adverse event was gastrointestinal discomfort, rated by subjects as mild to moderate. Noncompartmental pharmacokinetic analysis resulted in (mean ± standard deviation) a maximum concentration of 8749 ± 2840 ng/mL and half-life of 2.1 ± 0.71 hr. Magnetic resonance spectroscopy data were obtained in 3 individuals with Parkinson’s disease and showed modest increases in ATP and decreases in ATPase activity. Changes in Unified Parkinson’s Disease Rating Scale (parts I-IV) and Montreal Cognitive Assessment scores (mean ± standard deviation) were –4.6 ± 6.4 and 2 ± 1.7,respectively. This is the first report of UDCA use in individuals with Parkinson’s disease. Its pharmacokinetics are variable,and at high doses it appears reasonably well tolerated. Our findings warrant additional studies of its effect on brain bioenergetics.

Mitochondrial dysfunction leading to increased oxidative stress has been implicated in the pathogenesis of Parkinson’s disease.^1–4 Specifically, the presence of mitochondrial complex I dysfunction has been supported by postmortem brain studies in individuals with Parkinson’s disease and Parkinson’s disease animal models.^5–8 Additionally, mitochondrial impairments have been noted in the peripheral blood, well beyond the substantia nigra where the classic pathological changes are found in Parkinson’s disease.⁹ Other researchers using positron emission tomography scanning have reported metabolic changes in posterior regions of the brain in Parkinson’s disease supporting extranigral involvement.¹⁰ This suggests that both nigral and extranigral brain regions may have altered cellular function that is potentially restorable with treatment interventions. Our preliminary work demonstrated metabolic impairment, that is, a reduction in adenosine triphosphate (ATP) levels in the occipital cortex, as compared to age- and sex-matched healthy controls.¹¹ From this initial proof-of-concept study using a noninvasive, high-field magnetic resonance spectroscopy (MRS) method, we sought to determine if it is possible to alter these metabolic changes with ursodeoxycholic acid (UDCA), a drug that is used for other conditions and has been suggested to improve mitochondrial function.^12–17

UDCA is a naturally occurring hydrophilic bile acid formed by gut flora from chenodeoxycholic acid and makes up approximately 2% to 3% of the total bile acid pool in humans.^12,18 It is approved by the US Food and Drug Administration (FDA) for the management of primary biliary cholangitis at doses of 15 to 20 mg/kg/day.^12,19 Several extrahepatic effects of UDCA and its taurine conjugate (TUDCA) have been reported across multiple in vitro and preclinical animal model–based studies.^17,20–23 UDCA may stabilize impairments in the mitochondrial membrane by decreasing the translocation of pro-apoptotic bax from cytosol to the mitochondrial membrane, thus preventing perturbation of lipid and protein order.²² By decreasing translocation of bax, UDCA inhibits the mitochondrial transition, collapse of the transmembrane potential, cytochrome c release, and production of reactive oxygen species. Furthermore, studies conducted in animal models of ischemic stroke²⁴ and Huntington disease²⁵ indicate that UDCA reaches the central nervous system, where it exerts a neuroprotective effect and spares cognitive and sensorimotor decline in animal models.

Although UDCA and its conjugates have been studied extensively in animal models, there is a limited number of clinical studies evaluating UDCA pharmacokinetics and its neurological effects, particularly for neurodegenerative disorders.²⁶ UDCA is currently being evaluated in a clinical trial for Parkinson’s disease (NCT03840005). Previously, colleagues at our institution published the only report of UDCA in amyotrophic lateral sclerosis (Lou Gehrig’s disease), which appeared to be safe and well tolerated in 18 subjects at oral doses between 15 and 50 mg/kg/day.²⁶ The objective of our current study was to characterize the pharmacokinetics and evaluate the safety and tolerability after repeated oral administration of UDCA in individuals with Parkinson’s disease. A secondary aim of this study was to measure the effect of UDCA therapy on brain bioenergetics using high-field MRS.

Methods

This was an open-label, prospective, multiple-ascending-oral-dose study of UDCA in individuals with Parkinson’s disease over 6 weeks. The study was approved by the University of Minnesota Human Research Protection Program and the FDA (IND131477) and was listed on ClinicalTrials.gov (NCT02967250). The procedures followed were in accordance with the University of Minnesota Human Research Protection Program, FDA, and the World Medical Association Declaration of Helsinki. The entire study consisted of 4 clinical visits (Table 1). A written informed consent was obtained from all participants prior to enrollment. The inclusion criteria for the study were individuals aged 18 years with medically stable, mild to moderate Parkinson’s disease without dementia. Individuals who could not undergo a 7-Tesla ³¹P MRS scan (eg, due to metal in the body) were excluded from the study. Also excluded were those with unstable conditions or other neurological disorders, pregnant or lactating women, and those unable to adhere to study protocol for any other reason.

Table 1.

Schedule of Events for the Study

	Visit 1	Visit 2 Baseline	Visit 3 ~14-21 Days	Visit 4 ~Day 42
Informed consent	X
Inclusion/Exclusion criteria	X	X
MRI safety screen	X	X		X
Medical history	X	X
Concomitant medication(s)	X	X		X
Physical and neurologic exams	Xa	Xa		X
Vital signs		X	X	X
Montreal Cognitive Assessment	Xa	Xa		X
Unified Parkinson’s Disease Rating Scale	Xa	Xa		X
Modified Hoehn and Yahr	Xa	Xa		X
Modified Schwab and England Scale	Xa	Xa		X
Baseline blood collection		X
Oral UDCA administration		X		X
7-Tesla MRS		X		X
Blood PK collection (1 at baseline; ~8 at visit 4)		X		X
Blood safety panel			X	X
Postscan safety monitoring (30–60 min)		X		X
Adverse event querying			X	X

MRI,magnetic resonance imaging;MRS,magnetic resonance spectroscopy;PK, pharmacokinetic;UDCA, ursodeoxycholic acid.

X^a:completed on visit 1 or 2.

Baseline Measurements

Age, sex, ethnicity, weight, the Movement Disorder Society’s Unified Parkinson’s Disease Rating Scale (UPDRS parts I-IV) scores, Montreal Cognitive Assessment (MoCA) scores, medications, supplements, and vital signs were recorded at baseline. Each subject underwent a 7-Tesla ³¹P MRS scan. Blood samples were collected to measure endogenous levels of UDCA, TUDCA, and the glycine conjugate of UDCA (GUDCA) and for blood safety assessment panels that consisted of complete blood count with platelets, and tests for calcium, glucose, electrolytes, total protein, creatine phosphokinase, phosphorus, liver function, and kidney function for safety assessment. Subjects were instructed to withhold their Parkinson’s disease medications on the day of study visits 2 and 4 until after the MRS scan.

UDCA Dosing

UDCA tablets Urso (250 mg) and Urso Forte (500 mg) were purchased from Axcan Pharmaceuticals, Quebec City, Quebec, Canada. The total daily dose divided into 3 approximately equal doses was progressively elevated in the following manner as UDCA is known to increase gastrointestinal motility: 15 mg/kg/day in week 1, 30 mg/kg/day in week 2, and 50 mg/kg/day from week 3 through week 6. To enhance compliance, each subject was provided with only 2 to 3 weeks’ worth of UDCA tablets and was instructed to take them according to their individualized dosing schedule after meals thrice a day. Additional tablets for the remaining dosing period were provided at visit 3 or via mail for out-of-state residents. Also, a weekly telephone survey was conducted to assess compliance and document any adverse events. In case of adverse events resulting from escalated dose, subjects were allowed to continue with the tolerated dose for the remainder of the study.

Post-UDCA Measurements

Subjects returned after 2 weeks of UDCA therapy for visit 3. A blood sample was drawn for blood safety panel assessments, and any information about adverse events was recorded. Upon completion of 42 days of treatment, subjects returned for the final study visit. Upon arrival, an indwelling catheter was placed in participants’ arms for blood collections. A blood sample was collected to measure steady-state concentrations of UDCA and its conjugates and for blood safety panel assessments. The participants then received their final dose of UDCA in the clinic and approximately 5 cm³ of blood was obtained at up to 8 collection times bracketing 30 minutes and 8-hour postdosing (example collection schedule: 30, 60, 120, 180, 240, 300, 360, 420 minutes) for pharmacokinetic assessment. Participants also underwent a 7-Tesla ³¹P MRS scan within 3 hours of when they had their baseline scan. MoCA and UPDRS assessments were conducted. While the study was open-label, the post-UDCA reviewer was blinded to whether treatment had been received.

Magnetic Resonance Spectroscopy Methodology

All magnetic resonance measurements were conducted with a 7 Tesla/90 cm actively shielded human scanner (Siemens MAGNETOM, Erlangen, Germany) and a home-built ¹H/³¹P surface coil probe was used for collecting anatomic imaging and ³¹P MRS data in the human brain. Specifically, we applied the ³¹P MRS technique in combination with the magnetization transfer preparation to assess the metabolic activity of the ATP synthesis reactions and the absolute metabolites quantification approach to determine the ATP concentration in the human brain.^27–29

Measurement of UDCA, TUDCA, and GUDCA

Blood samples were placed on ice immediately, centrifuged for 10 minutes to obtain plasma, and then stored at –80°C until analysis. Concentrations of UDCA and its conjugates were measured using liquid chromatography–mass spectrometry (LC-MS). LC-MS–grade water, acetonitrile and ammonium acetate were purchased from Fisher Scientific (Houston, Texas). The standards of UDCA and its conjugates were purchased from Alfa Aesar (Devens, Massachusetts) and Acros Organics (Morris, New Jersey), respectively.

Briefly, plasma samples were mixed with 2 volumes of aqueous acetonitrile containing 5 μM of oleanolic acid as the internal standard and centrifuged at 18 000 × g for 10 minutes to remove the proteins and particles before LC-MS analysis. A 5-μL aliquot of diluted plasma sample was then injected into an ultraperformance liquid chromatography system (Waters, Milford, Massachusetts) and separated in a BEH C18 column. Water with 10 mM of ammonium acetate (pH = 9) (A), 95% acetonitrile and 5% water with 10 mM of ammonium acetate (pH = 9) (B) were used as mobile phase. The eluent from liquid chromatography was introduced into a SYNAPT G2-Si mass spectrometer (Waters) for accurate mass measurement and ion counting. For electrospray ionization, capillary voltage and cone voltage were maintained at 3 kV and 35 V for negative-mode detection. Source temperature and desolvation temperature were set at 120°C and 350°C, respectively. Nitrogen was used as both cone gas (50 L/h) and desolvation gas (600 L/h), with argon used as the collision gas. Sodium formate solution (mass-tocharge [m/z] ratio: 50–1000) was used to calibrate the mass spectrometer and was monitored by the intermittent injection of the lock mass leucine enkephalin ([M + H]⁻= 554.2615 m/z) in real time. Mass chromatograms and mass spectral data were acquired and processed by MassLynx software in centroided format. The concentrations of UDCA and its conjugates were determined by calculating the ratio between the peak area of the metabolite and the peak area of internal standard, and fitting with a standard curve using QuanLynx software (Waters).

Pharmacokinetic and Statistical Analysis

Using steady-state noncompartmental analysis, pharmacokinetic parameter values for UDCA were estimated for each subject (Phoenix WinNonlin 6.3; Pharsight Corp., Cary, North Carolina). Maximum concentration of UDCA (C_max), time at which C_max was observed, minimum concentration of UDCA (C_min) at steady state and area under the concentration-time curve (AUC) of UDCA from time 0 to tau were obtained. Trapezoidal rule using the linear up–log down method was used to calculate the AUC. Terminal-phase elimination rate constant (k_el) was obtained using the last 4 to 5 UDCA concentrations assuming a first-order elimination process and the terminal half-life was calculated as 0.693/k_el. Apparent clearance at steady state (CL_ss/F) was calculated as dose divided by AUC from time 0 to tau and apparent volume of distribution as CL_ss divided by k_el. Accumulation index was calculated as 1/|1-e^−kel*tau| and assumes that k_el is linear with respect to dose and time. The fluctuation percentage was calculated as 100*(C_max-C_min)/C_avg between dose time and tau.

Results

Demographics and Clinical Data

Five individuals (4 men, 1 woman) with mild to moderate Parkinson’s disease (Hoehn and Yahr stage 2) were enrolled and completed the study. The mean age was 65 years (range, 54–74), and mean weight was 78.4 kg (range, 70–88). All participants identified themselves as white and were taking at least 1 antiparkinsonian medication from among mirapex (0.75 mg/day), gabapentin (200 mg/day), rasagiline (1 mg/day), amantadine (100 mg/day), carbodopa/levodopa (75/300 mg/day or 150/600 mg/day), and macuna L-dopa (natural L-dopa, 300 mg/day).

Safety and Tolerability

Following 6 weeks of treatment, UDCA was found to be safe and well tolerated in all 5 individuals with Parkinson’s disease. No serious adverse events were reported. Most common adverse events were described as mild to moderate and related to intestinal discomfort (3) or increased tremor (2). Of these, 1 subject experienced increased tremor and headache at the highest dose (50 mg/kg/day). These symptoms resolved upon back titration to the 30 mg/kg/day dose, which was continued for the remainder of the study. No abnormal laboratory values from the safety panel assessments were reported for any participant during the course of the study.

Pharmacokinetics of UDCA and Conjugates

Baseline concentrations of endogenous UDCA at visit 1 for subjects 1 through 5 were found to be 191, 75, 166, 296, and 20 ng/mL, respectively, with a mean of 150 ng/mL. Figure 1 shows the steady-state concentration-time profiles for UDCA and GUDCA for each participant. Concentrations of TUDCA were found to be below the limit of detection (50 ng/mL). GUDCA concentrations generally paralleled those of UDCA. Average predose UDCA concentration at steady state prior to the last dose was found to be 2288 ng/mL with a range of 1024 to 5714 ng/mL indicating accumulation of the drug over time. Absorption appears to be erratic and highly variable. Multiple UDCA peaks, most likely due to enterohepatic circulation, were seen in some subjects. Pharmacokinetic parameter estimates were obtained after adjusting for the baseline concentrations using a noncompartmental analysis approach and are reported in Table 2. High between-subject variability in the parameter estimates, particularly for the time at which C_max was observed and apparent volume of distribution, was observed. Subject 3 had a lower maximum drug concentration as compared to other subjects, which could be due to reduced absorption, higher clearance, or differences in sampling schedule.

Figure 1.

Steady-state concentration-time profiles for ursodeoxycholic acid (UDCA) and its glycine conjugate (GUDCA). Taurine UDCA concentrations were below detection limit (50 ng/mL); concentration at time 0 is the predose steady-state concentration.

Table 2.

Pharmacokinetic Parameter Estimates for UDCA Obtained Using Noncompartmental Analysis

Subject ID	Dose (mg/ kg)	tmax (h)	Cmax (ng/mL)	Cmin (ng/mL)	AUC(0-tau) (μg • h/mL)	Clss/F (mL/min/kg)	Vz/F (mL/kg)	Half-life (t1/2) (h)	Accum. index	Fluctuation %
1	30	2.5	5441	613	21.0	9.0	1556	2.0	1.07	184
2	50	1	10 377	2603	38.4	7.7	1438	2.1	1.08	162
3	50	1	5166	902	19.5	16.6	4445	3.1	1.20	175
4	50	2	11 523	780	28.0	10.2	845	1.0	1.00	307
5	50	2.9	11 237	934	23.5	15.2	1466	1.1	1.01	351
Mean	…	1.9	8749	1166	27.4	11.8	1950	1.9	1.1	236
Standard deviation	…	0.8	2840	727	7.1	3.5	1273	0.8	0.1	77.6

AUC_(0-tau), area under the concentration-time curve from time 0 to infinity, CL_ss/F, apparent clearance at steady state, C_max, maximum plasma concentration; C_min, minimum plasma concentration; T_max, time at maximum concentration; V_z/F, apparent volume of distribution.

Magnetic Resonance Spectroscopy

³¹P MRS data were obtained from the occipital lobe of 3 participants enrolled in the study. Due to the hardware problem of the scanner, the data from the other 2 participants were either not acquired or unusable. The ATP concentrations in mM unit measured at pre/post-UDCA conditions are 2.68/2.73, 2.76/2.79, and 2.72/2.75 in subjects 1, 2 and 4, respectively. In the first 2 subjects, we also observed 10% and 48% reduction in the metabolic rate of ATPase reaction, which was accompanied by 5% and 8% increase of the metabolic rate of creatine kinase reaction in the same brains, respectively. These observations could be attributed to the correction effect of the UDCA treatment.

MoCA and UPDRS assessment

Table 3 shows MoCA and UPDRS scores for the 5 study participants before and after 6 weeks of UDCA therapy. Change in UPDRS and MoCA scores (mean ± standard deviation) was found to be –4.6 ± 6.4 and 2 ± 1.7, respectively.

Table 3.

Montreal Cognitive Assessment (MoCA) and Unified Parkinson’s Disease Rating Scale (UPDRS parts I-IV) Scores Before and After Ursodeoxycholic Acid (UDCA) Treatment

Subject	MoCA Score (Baseline)	MoCA Score (Post-UDCA Treatment)	MoCA Change From Baseline	UPDRS Score (Baseline)	UPDRS Score (Post-UDCA Treatment)	UPDRS Change From Baseline
1	25	27	2	45	32	−13
2	29	30	1	29	28	−1
3	26	28	2	50	46	−4
4	28	28	0	46	51	5
5	25	30	5	54	44	−10
Mean	26.6	28.6	2	44.8	40.2	−4.6
Standard Deviation	1.8	1.3	1.9	9.5	9.8	7.2

Discussion

This is the first report of UDCA pharmacokinetics in individuals with Parkinson’s disease. Based on the elimination half-life between 1 and 3 hours, we can assume that steady state had been achieved within 6 weeks. Comparison of the baseline and predose UDCA concentrations showed a steady accumulation over time. The dose-adjusted C_max values obtained were comparable with those previously reported in patients with amyotrophic lateral sclerosis.²⁶ UDCA pharmacokinetics, particularly the apparent volume of distribution, in the 5 individuals with Parkinson’s disease were found to be highly variable, potentially due to its erratic absorption and enterohepatic circulation. This is not surprising, as UDCA is a bile acid and upon oral administration is extracted by the liver, converted to its glycine and taurine conjugates, and secreted as bile into the gallbladder. Upon stimulation by food intake, bile is released from the gallbladder back into the intestine. About 95% of bile acids are reabsorbed from the intestine and return to the liver via the portal vein to complete 1 enterohepatic cycle.^18,30 Bile acids can be recycled 4 to 12 times per day between hepatocytes and enterocytes.¹⁸ This recycling of UDCA may lead to increased duration of exposure as characterized by double peaks seen in some concentration-time profiles. Hence, a larger study with additional concentration measures over a longer time period is warranted to fully characterize the pharmacokinetics of UDCA and its conjugates in individuals with Parkinson’s disease.

Our results showed that UDCA was generally safe and well tolerated at doses higher than those approved by the FDA for the treatment of primary biliary cholangitis. Safety panel assessment values were within normal limits for all participants, and most adverse events were minor. Gastrointestinal discomfort was the most frequently reported adverse event. These findings were in agreement with earlier reports of UDCA and information provided in the product label.^12,26 Constipation is one of the most common nonmotor symptoms associated with Parkinson’s disease and occurs in approximately 66% of individuals with Parkinson’s disease.^31–33 While the increased gastrointestinal motility caused by UDCA therapy may lead to intestinal discomfort in some individuals, it may alleviate constipation in some.

While preliminary, we observed increased occipital cortical ATP levels in 3 participants for whom we had baseline and post-UDCA MRS measurements. These findings support the hypothesis that UDCA can improve mitochondrial function in Parkinson’s disease as seen in preclinical studies. For example, UDCA has been shown to improve mitochondrial function as verified by elevated ATP levels in a rat rotenone model of Parkinson’s disease.¹³ Further, treatment with TUDCA prevented dopaminergic cell death induced by 1methyl-4-phenyl-1,2,3,6-tetrahydropyridine in a mouse model of Parkinson’s disease.²⁰ Also, previous work by Mortiboys et al¹⁵ has shown that UDCA rescues mitochondrial dysfunction in parkin-mutant fibroblasts obtained from patients with Parkinson’s disease.

The small number of subjects and relatively short observation period limit the interpretation of our results. While UDCA was generally well tolerated, the frequency and intensity of adverse events increased with dose. A randomized, double-blind, placebocontrolled trial of UDCA (28–30 mg/kg/day) in 150 patients with primary sclerosing cholangitis found that after 6 years of treatment, the UDCA group showed significantly improved results from liver tests but was associated with significantly higher risk of death and serious adverse events.³⁴ An additional limitation was that factors such as food intake and concomitant medications that may impact the pharmacokinetics of UDCA could not be evaluated due to the small sample size. Small improvements were noted in MoCA and UPDRS scores, but their significance is uncertain given the small sample size and the unblinded nature of this study.

UDCA appears to be a potential treatment for Parkinson’s disease given its efficacy in animal models and its safety profile in humans. Thus, UDCA is currently being evaluated for its effect on disease progression in a randomized, double-blind, placebocontrolled, proof-of-concept study (NCT03840005) at a dose of 30 mg/kg/day over 48 weeks, and results from this study are pending. It is expected that this trial will provide useful data on safety and tolerability; however, the challenge remains as to establishing whether UDCA has a disease altering effect given the lack of therapeutic or theragnostic biomarker, and thus a larger clinical trial will be needed. Meanwhile, 7-Tesla ³¹P MRS holds some promise as such a biomarker, but it remains a method limited to a few centers due to its technical challenges.

Conclusions

Based on our results in 5 individuals with Parkinson’s disease, UDCA appears to be reasonably safe and tolerable and provides the first report of UDCA pharmacokinetics in individuals with Parkinson’s disease. In addition, observed effects of UDCA on ATP concentration and ATP synthesis reactions in the brains of individuals with Parkinson’s disease provides a basis for further studies. Future studies relating UDCA exposure to its response would be useful for optimizing dosing for large, well-controlled trials evaluating its safety and efficacy.