Levodopa Carbidopa Intestinal Gel in Advanced Parkinson’s Disease: DUOGLOBE Final 3-Year Results

Journal of Parkinson’s Disease 13 (2023) 769–783 DOI 10.3233/JPD-225105 IOS Press 769 Clinical Research Levodopa Carbidopa Intestinal Gel in Advanced Parkinson’s Disease: DUOGLOBE Final 3-Year Results K. Ray Chaudhuria,∗, Norbert Kov´ acsb, Francesco E. Pontieric,d, Jason Aldrede, Paul Bourgeoisf, Thomas L. Davisg, Esther Cuboh, Marieta Anca-Herschkovitschi, Robert Iansekj, Mustafa S. Siddiquik, Mihaela Simul, Lars Bergmannm, Mayra Ballinam, Pavnit Kukrejam, Omar Ladhanim, Jia Jiamand David G. Standaertn aParkinson’s Foundation International Centre of Excellence, King’s College Hospital, and King’s College Institute of Psychiatry, Biomedical Research Centre, Psychology & Neuroscience, London, United Kingdom bDepartment of Neurology, University of P´ecs, P´ecs, Hungary cDepartment of Neuroscience, Mental Health and Sensory Organs, Sapienza University of Rome, Rome, Italy dSanta Lucia Foundation, IRCCS, Rome, Italy eSelkirk Neurology, Spokane, WA, USA fDepartment of Neurology AZ Groeninge, Kortrijk, Belgium gDepartment of Neurology, Vanderbilt University Medical Center, Nashville, TN, USA hNeurology Department, Hospital Universitario Burgos, Burgos, Spain iDepartment of Neurology, Edith Wolfson Medical Center, Holon, Israel jKingston Centre, Monash Health, Melbourne, Victoria, Australia kDepartment of Neurology, Wake Forest School of Medicine, Winston Salem, NC, USA lDepartment of Neurology, Victor Babes University of Medicine and Pharmacy, Timisoara, Romania mAbbVie Inc., North Chicago, IL, USA nDepartment of Neurology, University of Alabama at Birmingham, Birmingham, AL, USA Accepted 14 May 2023 Pre-press 5 June 2023 Published 25 July 2023 Abstract. Background: Levodopa-carbidopa intestinal gel (LCIG) improves motor and non-motor symptoms in patients with advanced Parkinson’s disease (aPD). Objective: To present the final 36-month efficacy and safety results from DUOGLOBE (DUOdopa/Duopa in Patients with Advanced Parkinson’s Disease – a GLobal OBservational Study Evaluating Long-Term Effectiveness; NCT02611713). Methods: DUOGLOBE was an international, prospective, long-term, real-world, observational study of patients with aPD initiating LCIG in routine clinical care. The primary endpoint was change in patient-reported “Off” time to Month 36. Safety was assessed by monitoring serious adverse events (SAEs). ∗Correspondence to: Dr. K. Ray Chaudhuri, Department of Basic and Clinical Neuroscience, The Maurice Wohl Clinical Neuroscience Institute, King’s College London, 5 Cutcombe Road, London SE5 9RT, UK. Tel.: +44 203 299 7154; E-mail: Ray[email protected].; ORCiD: 0000-0003-2815-0505. ISSN 1877-7171 © 2023 – The authors. Published by IOS Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0). 770 K.R. Chaudhuri et al. / Final 3-year Results from the DUOGLOBE Study Results: Significant improvements in “Off” time were maintained over 3 years (mean [SD]: –3.3 hours [3.7]; p< 0.001). There were significant improvements to Month 36 in total scores of the Unified Dyskinesia Rating Scale (–5.9 [23.7]; p= 0.044), Non-Motor Symptoms Scale (–14.3 [40.5]; p= 0.002), Parkinson’s Disease Sleep Scale-2 (–5.8 [12.9]; p< 0.001), and Epworth Sleepiness Scale (–1.8 [6.0]; p= 0.008). Health-related quality of life and caregiver burden significantly improved through Months 24 and 30, respectively (Month 24, 8-item Parkinson’s Disease Questionnaire Summary Index, –6.0 [22.5]; p= 0.006; Month 30, Modified Caregiver Strain Index, –2.3 [7.6]; p= 0.026). Safety was consistent with the well-established LCIG profile (SAEs: 54.9% of patients; discontinuations: 54.4%; discontinuations due to an adverse event: 27.2%). Of 106 study discontinuations, 32 patients (30.2%) continued LCIG outside the study. Conclusion: DUOGLOBE demonstrates real-world, long-term, reductions in motor and non-motor symptoms in patients with aPD treated with LCIG. Keywords: DUOGLOBE, Parkinson’s disease, levodopa-carbidopa intestinal gel, dyskinesia, real-world data INTRODUCTION Levodopa is considered the “gold standard” for the treatment of Parkinson’s disease (PD) [1]. With disease progression, the benefit of oral levodopa diminishes as the therapeutic window narrows due to the short half-life of levodopa and progressive denervation of the striatum and subsequent postsynaptic plasticity [2]. In addition, erratic gastric emptying (a common symptom with advancing PD) leads to irregular gastrointestinal absorption of oral levodopa and unstable plasma levodopa concentrations, collectively resulting in pulsatile stimulation [2, 3]. Thus, motor and non-motor symptoms become increasingly difficult to manage with oral levodopa, which can cause patients to experience predictable and unpredictable fluctuations between “On” periods with potentially disabling dyskinesias and “Off” periods when the patient experiences a return of their parkinsonian symptoms and may even be “frozen” and akinetic [2, 4, 5]. These symptoms progressively worsen over time and greatly impact patients’ functional capacity and health-related quality of life (HRQoL) [6, 7]. Levodopa carbidopa intestinal gel (LCIG; also known as carbidopa-levodopa enteral suspension [CLES]) is a stable gel suspension of levodopacarbidopa (20 mg/mL and 5 mg/mL, respectively) for continuous daytime infusion in patients with advanced PD (aPD) [8, 9]. Continuous infusion enables levodopa concentrations to be kept at a constant level within the individual’s optimal therapeutic window, making LCIG a meaningful option for managing aPD [3, 10]. Results from controlled clinical trials have demonstrated beneficial effects of LCIG therapy on motor symptoms, including reductions in “Off” time, increases in “On” time without troublesome dyskinesia, and improvements in HRQoL and activities of daily living [11–14]. Notably, results from a recent randomized clinical trial demonstrated an improved reduction in dyskinesia as measured by the Unified Dyskinesia Rating Scale (UDysRS) following treatment with LCIG vs. optimized medical treatment [15]. LCIG has also demonstrated beneficial effects on motor and non-motor symptoms in observational studies [16–21], and systematic literature reviews and meta-analyses [22, 23]. DUOdopa/Duopa in Patients with Advanced Parkinson’s Disease, a GLobal OBservational Study Evaluating Long-Term Effectiveness (DUOGLOBE), was the first international, fully prospective, long-term, non-interventional, postmarketing, observational study of patients with aPD treated with LCIG in a routine clinical setting. One-year interim results indicated significant improvements in motor symptoms (including “Off” time and dyskinesia), non-motor symptoms (including sleep), HRQoL, and caregiver burden, with safety events consistent with those noted in previous controlled clinical trials and observational studies [24]. This report presents the final 36-month results from the DUOGLOBE study. METHODS Study design and treatment DUOGLOBE was a global multicenter, singlearm, non-interventional, post-marketing, observational study (NCT02611713) conducted in 55 sites across 10 countries (Australia, Belgium, Hungary, Israel, Italy, Romania, Slovenia, Spain, United Kingdom, and the United States) [24]. Detailed methods for this study have been published [24]. In brief, patients enrolled in this 36-month real-world study had aPD for whom their physi- K.R. Chaudhuri et al. / Final 3-year Results from the DUOGLOBE Study 771 cians decided to start LCIG treatment according to the local product label and specific reimbursement criteria. LCIG dosage was individually optimized for clinical response and concomitant PD medications were permitted at the discretion of the treating physician. National and/or local independent ethics committees, institutional review boards, and/or health authorities in all countries approved the protocol, patient information, and informed consent requirements according to the applicable national regulatory requirements. Patients In addition to patient inclusion being based on local LCIG label and reimbursement criteria, key eligibility criteria included patients having no prior exposure to LCIG; a Mini-Mental State Examination score ≥24; no prior surgery for PD including, but not limited to, deep brain stimulation or cell transplantation (for non-US centers); and no current subcutaneous apomorphine infusion (with ≥4 weeks required between drug discontinuation and study inclusion). All patients and caregivers provided informed consent. Assessments Assessment occurred before starting LCIG (baseline) therapy; at Day 1 (start of LCIG treatment via percutaneous endoscopic gastrostomy with jejunal extension for those patients who participated in the preceding nasojejunal test phase only); and at routinely scheduled visits, which were closest to Months 3, 6, 12, 18, 24, 30, and 36 (±14 days each), or at the time of premature discontinuation. The primary endpoint was the change in the number of hours of “Off” time as reported by the patient for the day before the clinical visit (baseline) compared with the same measure at Month 36. Secondary endpoints included mean change from baseline to the end of the study in the Unified Parkinson’s Disease Rating Scale (UPDRS) Part II (activities of daily living); Part III (motor examination performed in the “On” state); and the following items in Part IV: item 33 (dyskinesia-related disability), item 34 (dyskinesia-related pain), and item 35 (early morning dystonia). Secondary endpoints also included the UDysRS total score (signs/symptoms of dyskinesia) and subdomain scores, Non-motor Symptom Scale (NMSS) total and subdomain scores, Parkinson’s Disease Sleep Scale-2 (PDSS-2) total score (sleep quality), Epworth Sleepiness Scale (ESS) total score (daytime somnolence), 8-item PD Questionnaire (PDQ-8) summary index (HRQoL), and caregiver burden (Modified Caregiver Strain Index). Only serious adverse events (SAEs) and adverse events (AEs) leading to premature discontinuation were reported as part of this observational study from initiation of LCIG treatment to 30 days after the last study visit. Statistical analysis Planned enrollment was approximately 200 patients. It was assumed that 60% of patients would complete the 36-month follow-up period and that the mean (SD) decrease from baseline to Month 36 in the number of hours in “Off” time would be 4 hours. Therefore, the distance from the lower limit of the 95% confidence interval (CI) to the mean decrease would be 0.72 hours (i.e., the lower limit of the 95% CI of the mean decrease from baseline to Month 36 would be 3.28 hours). Significance for all efficacy measures was determined using a one-sample ttest compared with baseline efficacy assessments. Safety assessments were performed with the safety population, which included all patients who had nasojejunal and/or percutaneous endoscopic gastrostomy with jejunal extension placement, irrespective of whether patients withdrew prematurely or not. Efficacy assessments were performed with the full analysis population, which included all patients in the safety population who had at least one post-baseline effectiveness assessment after undergoing percutaneous endoscopic gastrostomy with jejunal extension placement. Data sharing Clinical trial data can be requested by any qualified researchers who engage in rigorous, independent scientific research, and will be provided following review and approval of a research proposal and statistical analysis plan and execution of a data sharing agreement. Data requests can be submitted at any time and the data will be accessible for 12 months, with possible extensions considered. For more information on the process, or to submit a request, visit the following link: https://www.abbvieclinicaltrials.com/hcp/datasharing/. 772 K.R. Chaudhuri et al. / Final 3-year Results from the DUOGLOBE Study RESULTS Patients Patient demographics and baseline characteristics of the 195 patients included in the analysis have been reported previously [24]. Most patients were male (61.5%), mean (SD) age was 70.2 (8.2) years, and mean duration of PD was 11.2 (4.8) years (Table 1). Two patients previously underwent deep brain stimulation. Mean (SD) LCIG treatment duration was 923.5 (367.2) days, with a median daily duration of LCIG infusion of 16 hours from Day 1 through Month 36. At each timepoint, between nine and 12 patients were receiving 24-hour LCIG therapy. Of note, mean (SD) daily dose of LCIG remained fairly stable from Day 1 (1241.2 [501.6] mg/day) through Month 36 (1365.4 [499.1] mg/day) (Fig. 1A). Within the first 6 months of LCIG use, there were decreases in concomitant use of oral levodopa derivatives; monoamine oxidase B inhibitors; and, most prominently, catechol-Omethyltransferase inhibitors that remained steady throughout the next 2.5 years (Fig. 1B). At Month 3, 18.1% of patients were receiving LCIG monotherapy and 13.3% were receiving LCIG in combination with oral levodopa only (“levodopa monotherapy”); these percentages remained relatively stable throughout the study (Month 36 : 15.5% and 17.9%, respectively) (Fig. 1C). Total levodopa equivalent dose also remained stable throughout the study, regardless of whether patients were receiving LCIG as a monotherapy or in combination with other PD medications (Fig. 1D). Of 195 enrolled patients, 106 (54.4%) discontinued the study prematurely with AEs being the primary reason (patients could have multiple reasons for discontinuation) in 48 of these patients (Supplementary Figures 1 and 2). Importantly, of the 106 patients who discontinued the study, 32 (30.2%) continued treatment with LCIG outside the study (Supplementary Figure 2). Motor complications Mean (SD) daily hours spent in the “Off” state significantly decreased from baseline to Day 1 and remained decreased through Month 36 (Month 36 [primary endpoint]: –3.3 [3.7]; p< 0.001) (Fig. 2A). Changes from baseline at Month 36 were statistically significant in all age groups (Fig. 2B). Table 1 Baseline demographics and clinical characteristics Characteristic Total N= 195 Sex, n(%) Male 120 (61.5) Female 75 (38.5) Age (y); mean ±SD 70.2 ±8.2 <65 y, n(%) 44 (22.6) 65–75 y, n(%) 95 (48.7) >75 y, n(%) 56 (28.7) BMI; mean ±SD BMI, kg/m225.9 ±4.1a PD duration, y: mean ±SD 11.2 ±4.8 <10 y, n(%) 94 (48.5) ≥10 y, n(%) 100 (51.5) Time to LCIG initiation, y; mean ±SD from: PD symptoms 12.2 ±5.0 Start of motor fluctuations 5.6 ±4.7 MMSE total scoreb; mean ±SD 27.7 ±2.2 Hoehn and Yahr stage; n(%) During “On” 1 4 (2.1) 1.5 0 2 33 (17.6) 2.5 21 (11.2) 3 80 (42.6) 4 43 (22.9) 5 7 (3.7) Missing 7 During “Off” 10 1.5 2 (1.1) 2 6 (3.2) 2.5 15 (8.1) 3 59 (31.7) 4 82 (44.1) 5 22 (11.8) Missing 9 Daily “Off” time (h); mean ±SD 6.0 ±3.4 UPDRS Part II (ADL); mean ±SD 14.8 ±7.8 UPDRS Part III (motor function); mean ±SD 27.6 ±13.2 UDysRS total score; mean ±SD 33.7 ±21.1 NMSS total score; mean ±SD 88.2 ±51.1 PDSS-2 total score (sleep quality); mean ±SD 26.6 ±11.7 ESS total score (daytime sleepiness); mean ±SD 9.8 ±5.3 PDQ-8 summary index (HRQoL); mean ±SD 45.1 ±18.1 MCSI total score (caregiver burden); mean ±SD 10.9 ±6.4 an=182. bPatient MMSE total score at baseline must be ≥24 for inclusion. ADL, activities of daily living; BMI, body mass index; ESS, Epworth Sleepiness Scale; HRQoL, health-related quality of life; LCIG, levodopa-carbidopa intestinal gel; MCSI, Modified Caregiver Strain Index; MMSE, Mini-Mental State Examination; NMSS, Non-Motor Symptoms Scale; PD, Parkinson’s disease; PDSS-2, Parkinson’s Disease Sleep Scale-2; PDQ-8, 8-item Parkinson’s Disease Questionnaire; SD, standard deviation; UDysRS, Unified Dyskinesia Rating Scale, UPDRS, Unified Parkinson’s Disease Rating Scale. For UDysRS total scores, significant reductions from baseline were observed at all time points through Month 36 (Month 36 mean [SD]: –5.9 [23.7]; p= 0.044) (Fig. 2C). Exploratory analysis K.R. Chaudhuri et al. / Final 3-year Results from the DUOGLOBE Study 773 Fig. 1. Stability of therapy over time. (A) LCIG dose, (B) anti-PD comedications, (C) monotherapy vs. combination therapy, and (D) total levodopa equivalent dose. aOnly patients who participated in the preceding nasojejunal test phase were assessed at D1. BL, baseline; COMT, catechol-O-methyltransferase; D, day; LCIG, levodopa-carbidopa intestinal gel; LED, levodopa equivalent dose; M, month; MAO-B, monoamine oxidase-B; PD, Parkinson’s disease; SD, standard deviation. was conducted to evaluate the effect of LCIG on UDysRS at Month 36 based on age; there was no significant difference at any of the ages evaluated (Fig. 2D). The composite historical score was significantly reduced from baseline through Month 36 (Month 36 p= 0.012). The subdomain of “On” dyskinesia (Part I) was significantly reduced from baseline until Month 24 (Month 24 p= 0.001), with “Off” dystonia (Part II) significantly reduced through Month 36 (Month 36 p< 0.001); both these improvements exceeded the level of clinical relevance [25]. Significant reductions from baseline during the study were seen for other components of the UDysRS, including the objective score through Month 18 (Month 18 p= 0.045), the impairment subdomain (Part III) through Month 6 (Month 6 p< 0.001), and the disability subdomain (Part IV) through Month 24 (Month 24 p= 0.033) (Supplementary Figure 3). Additional support for the reduction of the presence and symptoms of dyskinesia and dystonia through Month 36 was seen in the UPDRS scale, including dyskinesiarelated disability (mean [SD], –0.4 [1.3]; p= 0.016), dyskinesia-related pain (–0.4 [1.0]; p< 0.001), and early morning dystonia (–0.2 [0.6]; p< 0.001) (Supplementary Figure 4). UPDRS Part II and III scores, after initial improvements (significant for UPDRS III until Month 3), demonstrated significant worsening from baseline to Month 36 in the overall group (Supplementary Figure 4). Patients aged ≥65 years showed significant worsening in UPDRS Part II and III scores at Month 36, while patients aged younger than 65 years had nominal, but not statistically significant, improvements in UPDRS Part III scores throughout the study (Supplementary Figure 4). A summary of baseline and change-from-baseline values for patients with Month 36 data can be found in Table 2. 774 K.R. Chaudhuri et al. / Final 3-year Results from the DUOGLOBE Study Fig. 2. Change from baseline to Month 36 in (A) patient reported “Off” time, (B) “Off” time by age subgroups, (C) dyskinesia as measured by UDysRS total score, and (D) UDysRS total score by age subgroups. Significance level for change from baseline was determined using the one-sample ttest. *p<0.05; **p<0.01; ***p<0.001. aOnly patients who participated in the preceding nasojejunal test phase were assessed at D1. bAs reported in Hauser et al [26]. BL, baseline; D, day; LCIG, levodopa-carbidopa intestinal gel; M, month; MCID, minimal clinically important difference; SD, standard deviation; UDysRS, Unified Dyskinesia Rating Scale. Table 2 Change from baseline to month 36 in motor and non-motor endpoints among patients with 36-month data Endpoints nBaseline Change from baseline Daily “Off” time (h) 80 5.8 (3.1) –3.3 (3.7)*** UDysRS total score 67 35.2 (21.4) –5.9 (23.7)* UPDRS Part II (ADL) 85 13.4 (8.5) 4.3 (8.3)***a UPDRS Part III (motor function) 83 24.9 (13.6) 5.8 (13.9)***a NMSS total score 79 83.7 (46.5) –14.3 (40.5)** PDSS-2 total score (sleep quality) 85 27.7 (12.3) –5.8 (12.9)*** ESS total score (daytime sleepiness) 84 9.6 (5.3) –1.8 (6.0)** PDQ-8 summary index (HRQoL) 84 45.2 (18.6) –2.5 (19.6) MCSI total score (caregiver burden) 52 12.3 (6.8) –1.3 (7.8) Table includes patients with Month 36 data only. All data are presented as mean ±SD. aReflects worsening from baseline. *p<0.05; **p<0.01; ***p<0.001. ADL, activities of daily living; ESS, Epworth Sleepiness Scale; HRQoL, health-related quality of life; MCSI, Modified Caregiver Strain Index; NMSS, Non-Motor Symptoms Scale; PDSS-2, Parkinson’s Disease Sleep Scale-2; PDQ-8, 8-item Parkinson’s Disease Questionnaire; SD, standard deviation; UDysRS, Unified Dyskinesia Rating Scale, UPDRS, Unified Parkinson’s Disease Rating Scale. K.R. Chaudhuri et al. / Final 3-year Results from the DUOGLOBE Study 775 Fig. 3. Change from baseline in (A) NMSS total score and (B) NMSS total score by age subgroups. Significance level for change from baseline was determined using the one-sample t test. *p<0.05; **p<0.01; ***p<0.001. aOnly patients who participated in the preceding nasojejunal test phase were assessed at D1. bAs reported in Martinez-Martin et al [31]. BL, baseline; D, day; M, month; MCID, minimal clinically important difference; NMSS, Non-Motor Symptom Scale; SD, standard deviation. Non-motor symptoms NMSS total scores demonstrated significant and sustained improvement from baseline at Month 36 (mean [SD], –14.3 [40.5]; p= 0.002) and all time points (Fig. 3A,B). In addition, three of nine NMSS subdomains were significantly improved through Month 36: sleep/fatigue (mean [SD], –3.7 [12.3]; p= 0.007), gastrointestinal tract (–2.5 [6.1]; p< 0.001), and miscellaneous (–3.9 [10.3]; p< 0.001) (Supplementary Figure 5). Of note, the last statistically significant improvement was measured at Month 12 for one domain (cardiovascular, including falls), and Month 24 for three domains (mood/cognition, attention/memory, and sexual function) (Supplementary Figure 5). In addition to improvement in the NMSS sleep/fatigue subdomain, stable and significant improvement in sleep quality (Fig. 4A,B) and daytime sleepiness (Fig. 4C) was seen with LCIG. Significant improvement from baseline at Month 36 was observed for PDSS-2 total scores (mean [SD], –5.8 [12.9]; p< 0.001) and ESS total scores (mean [SD], –1.8 [6.0]; p= 0.008). Significant improvement was seen from Day 1 (PDSS-2) and Month 3 (ESS) onward. Patient HRQoL and caregiver burden Patient HRQoL and caregiver burden improved from baseline, with significant and consistent improvement in PDQ-8 through Month 24 (mean [SD]: –6.0 [22.5]; p= 0.006) (Fig. 5A,B) and Modified Caregiver Strain Index through Month 30 (–2.3 [7.6]; p= 0.026) (Fig. 5C,D). Both scales continued to show numerical, but not statistically significant, improvements at Month 36. Safety A total of 107 (54.9%) patients experienced SAEs (Table 3), with 31 SAEs considered as having a reasonable possibility of being related to LCIG treatment. The most common SAEs were fall (n= 8), (worsening of) PD (n= 8), and urinary tract infection (n= 7) (Table 3). One patient experienced an SAE of polyneuropathy, and one experienced an SAE of chronic inflammatory demyelinating polyradiculoneuropathy; both SAEs were adjudicated by the investigator as having no reasonable possibility of relationship to study drug. A total of 53 (27.2%) patients discontinued the study owing to an SAE. However, this includes 34/195 (17.4%) fatal AEs, with all reported as having no reasonable possibility of relationship to study treatment with the exception of one patient with an intestinal obstruction and a medical history of diverticulitis that was adjudicated as possibly related to treatment (Supplementary Table 1). Most fatal AEs were related to complications of aPD, cardiovascular disease, and complications from non-treatment-related infections. A total of six patients (3.1%) discontinued the study due to COVID-related infections, restrictions, or fears of infection. These discontinuations included two patients (1.0%) who had an SAE of COVID- 776 K.R. Chaudhuri et al. / Final 3-year Results from the DUOGLOBE Study Fig. 4. Change from baseline in (A) PDSS-2 total score, (B) PDSS-2 total score by age subgroups, and (C) ESS total score. Significance level for change from baseline was determined using the one-sample ttest. *p<0.05; **p<0.01; ***p<0.001. aOnly patients who participated in the preceding nasojejunal test phase were assessed at D1. bAs reported in Horvath et al [32]. BL, baseline; D, day; ESS, Epworth Sleepiness Scale; M, month; MCID, minimal clinically important difference; PDSS-2, Parkinson’s Disease Sleep Scale-2; SD, standard deviation. 19 and/or COVID-related pneumonia, both of which resulted in death. In general, body weight and body mass index (BMI) remained relatively stable throughout the study. At baseline, mean body weight was 73.2 kg and mean BMI was 25.7 kg/m2; at Month 36, mean weight and BMI were 71.2 kg and 24.9 kg/m2, respectively, with both decreasing throughout the study (Supplementary Figure 6A). Mean (SD) decrease from baseline to Month 36 was –3.1 (8.0) kg for weight and –1.2 (2.8) kg/m2for BMI. At last visit, 8.2% of patients demonstrated an increase in baseline weight ≥7%, while 22.1% demonstrated ≥7% decrease. Most patients remained within the same BMI category at the end of the study that they were in at baseline (Supplementary Figure 6B). Looking at safety findings across age, patients in the 65to 75-year age group experienced higher rates of SAEs than did those in the younger than 65-year age group and older than 75-year age groups (whose rates were similar to each other), with rates of severe AEs lowest in the younger than 65-year age group and similar in the 65 to 75 year and older than 75year age groups (Table 3). Rates of severe AEs were lowest in patients aged younger than 65 years and were similar in patients aged 65 to 75 years and older than 75 years. DISCUSSION We report the final results of DUOGLOBE, a 3year, fully prospective study designed to evaluate LCIG use in routine clinical practice in a large realworld population with aPD. Treatment with LCIG demonstrated “Off” time improvements that were maintained over 3 years and remained above the min- K.R. Chaudhuri et al. / Final 3-year Results from the DUOGLOBE Study 777 Fig. 5. Change from baseline in (A) PDQ-8 summary index, (B) PDQ-8 summary index by age subgroups, (C) MCSI total score, and (D) MCSI total score by age subgroups. Significance level for change from baseline was determined using the one-sample ttest. *p<0.05; **p<0.01; ***p<0.001. aOnly patients who participated in the preceding nasojejunal test phase were assessed at D1. bAs reported in Horvath et al [36]. BL, baseline; D, day; MCID, minimal clinically important difference; MCSI, Modified Caregiver Strain Index; PDQ-8, 8-item Parkinson’s Disease Questionnaire; SD, standard deviation. imal clinically important difference (MCID) of 1 hour in this population, despite the progressive nature of PD [26]. The beneficial effects of LCIG use were also reflected in improvements in UDysRS scores and subscores. Improvements in “On” dyskinesia (Part I) were above the MCID of –2.1 through Month 24 and improvements in “Off” dystonia (Part II) were above the MCID of –1.8 through Month 36 [25]. Reductions in UDysRS scores are particularly noteworthy, as this scale (which includes both patient subjective and clinician objective dyskinesia ratings) has demonstrated a high sensitivity for detecting treatment-related changes [27]. The current findings of significant decreases in “Off” time coupled with significant improvements in “On” dyskinesia with LCIG treatment are consistent with those reported in previous studies [15, 28, 29]. Parallel improvements in “Off” time and “On” time with dyskinesia are possible with LCIG, likely due to continuous delivery of stable plasma levodopa levels that can remain in the therapeutic window [10, 15, 28]. Non-motor symptoms of PD are often unrecognized and untreated [30]. Significant improvements were observed in the NMSS total scores, which remained above the 13.9-point MCID throughout the 3 years [31]. Improvements in sleep and daytime sleepiness (as measured by the PDSS-2 and ESS, respectively) were also observed throughout the study, with PDSS-2 changes exceeding the –3.4 point MCID threshold at all timepoints [32]. These findings are particularly relevant, as non-motor symptoms and sleep disturbances have been directly linked to deterioration of HRQoL [33–35]. This study supports this relationship, with both non-motor symptoms and HRQoL improving after LCIG treatment. Signifi-