Efficacy of levodopa/carbidopa/entacapone versus levodopa/carbidopa in patients with early Parkinson’s disease experiencing mild wearing-off: a randomised, double-blind trial
- First Online:
- Cite this article as:
- Tolosa, E., Hernández, B., Linazasoro, G. et al. J Neural Transm (2014) 121: 357. doi:10.1007/s00702-013-1114-x
- 767 Views
To compare the efficacy and safety of levodopa/carbidopa/entacapone (LCE) with levodopa/carbidopa (LC) on Parkinson’s disease (PD) patients with mild, or only minimally disabling motor complications. A prospective 3-month, multicentre, parallel-group, double-blind, and randomised phase IV study was performed. The primary endpoint was to assess the efficacy of LCE compared to LC on ADLs using the UPDRS part II. Secondary endpoints were assessed by the UPDRS (I, III and IV) scores, QUICK and PDQ-39 questionnaires, and patient and investigator clinical global impression (CGI). Ninety-five patients were randomly assigned to treatment with LCE (100/25/200 or 150/37.5/200 mg tablets, n = 46) or LC (100/25 mg tablets, n = 49), at the same levodopa dose that were administered before randomization. Treatment with LCE resulted in significantly greater improvement in UPDRS part II (ADLs) scores compared to treatment with LC (adjusted mean difference between groups of −1.5 points) (p = 0.0288). Amelioration was also observed in UPDRS part III scores (p = 0.010), and CGI (patient and investigator) scores (p = 0.015, and p = 0.028, respectively). LCE and LC were generally well tolerated with 78 % of subjects completing the study. Most AEs (50 % in LCE and 71.4 % in LC) were classified as mild. No serious AEs were related to the treatment. Treatment with LCE results in improved efficacy compared to LC in PD patients with mild, or minimally disabling motor fluctuations, maintaining a good safety and tolerability profile.
KeywordsParkinson diseaseWearing-offUPDRSQuick questionnaire
Levodopa remains the most effective therapy for the treatment of Parkinson’s disease (PD) and it is considered as the “gold standard” of symptomatic treatment since the end of the 1960’s (Yahr et al. 1969).
It is currently administered with a dopa-decarboxylase inhibitor (DDCI) which shifts the peripheral metabolism of levodopa to an alternative pathway of catabolism-methylation [catechol-O-methyltransferase (COMT)] (Boiko et al. 2008). COMT inhibitors block peripheral l-dopa metabolism, prolong the half-life of levodopa and increase the delivery of levodopa to the brain (Ruottinen and Rinne 1998). Entacapone is one such peripheral COMT inhibitor that extends the half-life of levodopa elimination by 85 % (Nutt et al. 1994; Schrag 2005). The addition of entacapone to levodopa in patients with moderate/severe PD reduces the dose and the frequency of administration of levodopa, decreasing “off” time and increasing “on” time which together result in an improved and prolonged clinical response to levodopa (Brooks 2008; Hauser et al. 2009; Ruottinen and Rinne 1998). Moreover, several studies demonstrated that entacapone was well tolerated, with an excellent risk/benefit ratio (Larsen et al. 2003; Piccini et al. 2000).
The aim of the present study was to investigate whether treatment with levodopa/carbidopa/entacapone (LCE) when compared with levodopa/carbidopa (LC) improves functional disability as measured by the UPDRS part II score in PD patients with mild, minimally disabling motor complications.
Materials and methods
This was a 3-month, multicentre, parallel-group, double-blind, randomised phase IV study to compare the efficacy and safety of LCE versus LC on PD patients with mild, or only minimally disabling motor complications. An institutional review board approved the protocol at each site, and all subjects provided written consent prior to the study participation.
Eligible patients were male or female patients aged 30–80 years with a previous diagnosis of idiopathic PD according to the UK Parkinson’s Disease Society Brain Bank criteria. Additional eligibility criteria included stable levodopa treatment for at least 1 month prior to entry at the study. Patients would be receiving, at least 1 month before baseline, stable doses of LC 100/25 mg immediate release (IR) (a minimum of one whole tablet and a maximum of one tablet and a half, per individual dose), or LC 100/25 mg slow release (SR) (one tablet per individual dose) or LC 200/50 mg SR (one tablet per individual dose), at different daily doses (total daily levodopa dose of 300 minimum and 600 maximum). Patients were required to acknowledge experiencing wearing-off (WO) diagnosed by the QUICK questionnaire (at least two positive answers), impaired ADLs according to the UPDRS part II (9–16 score out of maximum score of 52), and either absent or mild dyskinesia. Mild dyskinesia was defined as: maximum UPDRS score of 1 to item 35 (25 % of day with dyskinesia), 0–1 to items 36 (not disabling or mildly disabling dyskinesia) and 37 (not painful dyskinesia or slightly painful dyskinesia), and 0 to item 38 (no presence of early morning dystonia). Women in fertile age should be negative with a urine pregnancy test before baseline visit. Main exclusion criteria included patients previously or currently treated with entacapone; symptoms, signs or history of atypical or secondary Parkinsonism; hallucinations or psychiatric disorders related to dopaminergic treatments; major depression; current treatment with neuroleptics, rotigotine or monoaminooxidase inhibitors (with the exception of 10 mg of selegiline/day or 1 mg of rasagiline per day) during the 60 days prior to the screening visit; history of neuroleptic malignant syndrome and/or nontraumatic rhabdomyolysis.
Patients were enrolled in the study during routine control visits or through their medical records by the team of investigators. Patients who fulfilled the inclusion/exclusion criteria were randomly assigned in a 1:1 ratio to LCE or LC treatment groups.
Randomization sequence was created using the SAS program, version 9.1 (SAS Institute Inc., Cary, NC, USA) statistical software, using random block sizes of 4.
Study visits were conducted at screening (visit 1), baseline (visit 2), 1 month (visit 3), 2 months (visit 4), and 3 months (final visit).
The study was approved by the local Clinical Ethics Committee at each participating centre and conducted in accordance with the Declaration of Helsinki guidelines.
LCE 100/25/200 mg (Stalevo® 100) and LCE 150/37.5/200 mg (Stalevo® 150) was manufactured by Orion Corporation Orion Pharma, Finland. IR LC 100/25 (Sinemet®) was manufactured by Bristol Myers Squibb, Spain. Study medication was over encapsulated by Farmasierra (Spain) to prevent unblinding of the treatment.
If the dose of levodopa needed to be increased, consideration was given to increase the dose frequency and/or the use of another strength of LCE. If levodopa had to be decreased, time between doses was extended or the strenght of LCE was reduced.
The primary efficacy endpoint was the mean differences from baseline to 3-month visit in the UPDRS part II score. Secondary efficacy variables included the UPDRS parts I, III, and IV. Assessments were performed by raters blinded at all study visits, except at screening visit (only part II of the UPDRS was evaluated at screening visit). Other secondary efficacy variables included the 39-item PD QoL questionnaire (PDQ-39) and the 19-item QUICK questionnaire. They were assessed at all study visits, except the PDQ-39 that was not evaluated at screening. The validated QUICK questionnaire was used as a tool for screening of WO. It consists of a 19-item instrument where the patient answers whether symptoms present in the questionnaire improve after the following dose of medication. The presence of two positive items is considered for diagnosing WO (Martinez-Martin et al. 2008).
The clinical global impression (CGI) of change (patient and investigator) scores were completed at 3-month (final visit).
Safety was evaluated via monitoring for adverse events (AEs), standard clinical laboratory testing (haematology, blood chemistry, and urinalysis), ECG recordings, routine physical examinations, and regular recording of vital signs. All AEs, serious AEs (SAEs), and vital signs were recorded at each study visit. AEs and SAEs were classified according to the medical dictionary for regulatory activities (MedDRA).
Data analyses were performed with the SAS program version 9.1. All statistical tests were conducted against a two-sided alternative hypothesis employing a significance level of 0.05 unless otherwise noted.
The analysis of the clinical efficacy was performed in the full analysis set (FAS)–last observation carried forward (LOCF) populations. The LOCF technique was used to impute missing data. To ensure the robustness of the main analysis conclusion, an additional analysis was carried out using the per-protocol (PP)-LOCF population.
All randomized patients who took at least one dose of study medication were included in the safety population, and all patients from safety data set who presented at least one post-baseline measure of UPDRS part II questionnaire were included in the FAS-LOCF population. Patients who completed at least 52 days of treatment without protocol deviations were included in the PP-LOCF population.
All the variables were described by treatment group and visit. For categorical variables, % relative and n of each category were obtained while for continuous variables, the following descriptive statistics were calculated: mean, standard deviation (SD), and 95 % CI.
The primary endpoint was to determine the mean difference from baseline to 3-month visit (final visit) in the UPDRS part II score using the covariance (ANCOVA) analysis to determine the mean differences in these measurements, by adjusting the baseline score. Secondary endpoints included the mean differences from baseline to 3-month visit in UPDRS (part I, III, and IV), and PDQ-39 scores.
Other secondary endpoints included the CGI assessment and the QUICK questionnaire. For CGI assessment (investigator and patient) the proportion of patients in each category (very much improved, much improved, minimally improved, no change, minimally worse, much worse, and very much worse) was analysed. Treatment group differences observed by physician and patient CGI assessment and QUICK questionnaire were analysed using the Mann–Whitney U test for non-paired data.
The sample size was calculated under the assumption of obtaining a confidence interval of 95 % for two-tailed test with a difference (delta) of 1.5 points in Part II of the UPDRS between treatment groups. Sample size calculation assumed a standard deviation of 2.8 (Reichmann et al. 2005), and a 80 % of power. Under these assumptions, it was estimated that the sample size should be 43 patients per treatment group (86 evaluable patients in total). Anticipating a 15 % loss, it would be necessary to randomize 50 patients per group (100 patients in total).
The study was conducted from October 2006 to March 2008 at 27 centres in Spain.
Baseline demographics and characteristics (safety population)
LCE (n = 46)
LC (n = 49)
Female, n (%)
Age (years), mean ± SD
66.4 ± 8.2
66.5 ± 9.0
Hoehn and Yahr stage, n (%)
Time since diagnosis (years), mean ± SD
4.7 ± 4.0
4.4 ± 3.8
EQD (dopamine agonists, mg), mean ± SD
293.5 ± 172.2
318.9 ± 215.5
EQD (levodopa with decarboxylase inhibitor, mg), mean ± SD
390.0 ± 100.9
410.2 ± 96.8
EQD (total, mg), mean ± SD
585.7 ± 211.8
611.9 ± 241.8
The mean (SD) daily dose of levodopa (LCE vs LC) during the study was: 395.6 (102.7) mg vs 414.3 (98.4) mg (at baseline), 396.3 (101.5) mg vs 420.7 (101.8) mg (visit 3), 394.6 (105.3) mg vs 420 (103.0) mg (visit 4) and 394.2 (101.3) mg vs 414.9 (100.5) mg (visit 5), respectively.
UPDRS (part I, II, IIII IV), PDQ-39 and Quick questionnaires scores along the study (FAS-LOCF population)
Total UPDRS score, mean ± SD
34.1 ± 9.1
35.1 ± 8.1
28.4 ± 9.6
33.7 ± 8.4
26.8 ± 10.8
33.5 ± 9.3
26.8 ± 10.7
33.3 ± 11.2
2.1 ± 1.9
2.1 ± 1.8
1.5 ± 1.6
2.3 ± 2.0
1.5 ± 1.7
2.2 ± 1.9
1.6 ± 1.6
2.0 ± 1.8
11.3 ± 2.0
11.6 ± 2.0
9.6 ± 3.4
11.1 ± 2.6
8.9 ± 3.8
11.1 ± 3.5
9.1 ± 3.6
11.2 ± 4.2
17.8 ± 6.5
18.6 ± 5.5
14.9 ± 5.8
17.3 ± 5.0
14.0 ± 6.3
17.4 ± 4.7
13.9 ± 6.4
17.5 ± 6.1
2.9 ± 1.8
2.7 ± 1.7
2.4 ± 2.0
2.9 ± 1.5
2.4 ± 2.2
2.9 ± 1.6
2.3 ± 2.3
2.6 ± 1.8
PDQ-39 score, mean ± SD
140.1 ± 23.5
139.6 ± 24.8
147.5 ± 20.9
140.2 ± 24.0
146.8 ± 24.5
140.6 ± 24.5
146.3 ± 24.4
139.4 ± 25.9
Quick questionnaire, mean ± SD
4.8 ± 1.3
4.8 ± 1.1
3.8 ± 2.3
4.5 ± 1.8
3.9 ± 2.3
4.5 ± 2.3
4.0 ± 2.6
4.3 ± 2.8
The analysis in the PP-LOCF population yielded similar results; significant differences in the adjusted mean differences between groups were observed [−1.85 (95 % CI −3.19 to −0.50), p = 0.0078].
Significant greater improvement in the UPDRS part III with LCE compared to LC was observed between baseline and final visit (mean difference of −3.6 ± 4.7 LCE group vs −1.2 ± 5.3 LC group, p = 0.010).
The global UPDRS score showed significant higher mean difference (baseline vs final visit) −6.8 ± 8.0 in the LCE group and −1.9 ± 8.4 in the LC group (p = 0.0074).
CGI assessments (patient and investigator) were performed at the final visit. The patient score was of −0.8 ± 1.1 in the LCE group and −0.2 ± 1.4 in the LC group, p = 0.0208. The investigator evaluation was also favourable to LCE group (−0.8 ± 1.0 LCE and −0.1 ± 1.0 LC, p = 0.0015).
The mean differences in the UPDRS part I between baseline and final visit showed a higher score reduction in the LCE group (−0.5 ± 1.6 LCE group vs −0.1 ± 1.2 LC group), without significant differences. In the UPDRS part IV, the mean differences between baseline and final visit in both treatments exhibited a higher score reduction in the LCE treatment group (−0.6 ± 1.8 in the LCE and −0.1 ± 1.5 in the LC group). Nevertheless, these differences were not statistically significant.
Adverse events were documented in 44.4 and 38.8 % of patients in the LCE and LC group, respectively. Most of the events were of mild intensity (50.0 % LCE and 71.4 % LC group). Twenty-seven AEs (35.6 % of patients) in the LCE group and 16 AEs (18.4 % of patients) in the LCE group were related to the treatment. Three serious AEs (2.2 % in LCE vs 4.1 % LC) were reported which were not considered related to the treatment.
Three (6.7 %) patients in the LCE group and 1 (2.0 %) patient in the LC group discontinued the study due to an AE.
There were no deaths during the study. All patients showed laboratory haematological and biochemical parameters within normal range, or only minor abnormalities were reported. There were no significant or clinically relevant abnormalities in liver function in any treatment group according to transaminase tests.
Adverse events (safety population)
LCE (n = 45)
LC (n = 49)
Total (n = 94)
Most frequent AEs (≥4 %)
Dyskinesia and dystonia
These are the results of a short-term study showing that the triple therapy combination (LCE) (100/25/200 or 150/37.5/200 mg) had greater efficacy than LC treatment in PD patients with only minimally disabling motor complications, as shown by the improvement in UPDRS part II (ADLs) scores. This improvement could be explained in part by the significant amelioration in severity of motor symptoms as observed in the UPDRS III. Similar improvements in ADLs in patients treated with entacapone have been published in the FIRST-STEP study (mean change at week 39 of −1.7, p = 0.045), SENSE study (mean change at week 6 of −2.1, p < 0.0001), and ComQol study (mean change at weeks 5 and 13 of −1.6, p = 0.0001) (Eggert et al. 2010; Hauser et al. 2009; Reichmann et al. 2005; Stocchi et al. 2010b). Fung et al. (2009) also found clinically relevant improvements in global status in patients with mild, non disabling WO.
In our study differences between study groups, favouring LCE, was also observed in UPRDS Part III (motor impairments) scores, global UPRDS scores, and subject-patient CGI scores. Moreover, a trend towards less WO (assessed by QUICK questionnaire) was observed in the LCE versus LC group, consistent with the approved indication of entacapone as an adjunct to levodopa as previously described (Parkinson Study Group 1997; Rinne et al. 1998).
The greater clinical efficacy with LCE noticed in this study is consistent with pharmacokinetic studies that show that the addition of entacapone causes a 1.32 to 1.39-fold increase in the plasma levodopa area under the curve (Nutt et al. 1994) and the observation that dual inhibition of levodopa peripheral metabolism with both carbidopa and entacapone increases central bioavailability of levodopa to a greater extent than that with carbidopa alone (Gordin et al. 2004; Solla et al. 2010).
The assessment of QoL shows a higher but not significant QoL score in the LCE group. This trend was observed in several studies using PDQ-39, PDQ-8 and VAS assessment, respectively (Hauser et al. 2009; Stocchi et al. 2010a). However, Stocchi et al. (2010a) reported that LCE patients had a slightly worsening in PDQ-39 scores when comparing LCE with LC. One potential explanation for not finding significant differences between groups using the PDQ-39 could be the relative insensitivity to change of this measure, as has been suggested in early PD patients (Hauser et al. 2009).
Patient and investigator CGI scores were significantly improved in the LCE compared with the LC treatment group. These results agree with previous studies comparing LCE with LE treated patients that reported a significant CGI improvement by patient (69.6 %) and by investigator (80.4 %) (Eggert et al. 2010).
One limitation of this study could be the sample size. The study was powered to detect differences in the primary endpoint, but a larger sample size could have confirmed the existence of differences in secondary end-points such as quality of life.
In our study, LCE and LC were generally well tolerated with 78 % of subjects completing the study. Most AEs (50.0 % in LCE and 71.4 % in LC) were classified as mild. No serious AEs were related to the treatment. We did not observe, as previously reported (Hauser et al. 2009), an increase in the incidence of AEs, including nausea and diarrhoea, compared with LC. We have detected dyskinesia in 18 % of patients with LCE treatment compared to 6 % in LC group. An increase in dyskinesia may be expected with the co-administration of a COMT inhibitor to LC (Fahn 2005; Fahn et al. 2004; Stocchi 2006). The STRIDE-PD study performed in early PD patients found that those treated with LCE had an increased risk to develop dyskinesia compared to those receiving LC. The authors suggested that the shorter time to onset and increased frequency of dyskinesia in the LCE group were likely due to the higher levodopa dose equivalents administered in this group combined with a failure to achieve continuous dopamine stimulation (CDS) (Stocchi et al. 2010b). The higher incidence of dyskinesia observed in LCE-treated patients in our study could similarly be explained on the basis of the higher dopaminergic load in the LCE compared to the LC patients. In most studies, an initial increase in dyskinesia occurred when adding entacapone, but it was ameliorated by reducing the levodopa dose (Rinne et al. 1998; Stocchi 2006).
In conclusion, PD patients without or only minimally disabling motor complication showed improvement in motor and activities of daily living scores when treated with LCE compared to LC. Adding entacapone to LC in the early stages of dopaminergic treatment lessens functional disability maintaining a good safety and tolerability profile.
This study was supported by Novartis, S.A. Editorial assistance was supported by Montserrat Sabaté and biostatistics by Emma Albacar from Trial Form Support.
This study has been conducted by the above-signed authors and the DERBI study group (in alphabetical order): Adolfo Minguez, Àngels Bayés, Carlos Leiva, Ernest Balaguer, Esther Cubo, Francesc Miquel, Francisco Vivancos, Jose Juni, Juan Andrés Bruguera, Lydia Vela, Matilde Calopa, Maria José Catalán, Mariateresa Buongiorno, Miguel Aguilar, and Oriol de Fàbregues.
Conflict of interest
Hernández B is employee of Novartis S.A. This study was supported by Novartis, S.A.