Profile of study; tid indicates 3 times daily
Change in "off" time (top) and "on" time (bottom) between baseline and week 6 of the study. Data are means. Intention-to-treat population, last-observation-carried-forward analysis. Asterisk indicates P<.001 for pairwise comparison with placebo after adjustment for multiple comparisons; tid, 3 times daily.
Change in levodopa dosage between baseline and week 6 of the study. Top, Total daily levodopa dose. Bottom, Number of daily intakes. Intention-to-treat population, last-observation-carried-forward analysis. Asterisk indicates P<.001; dagger, P<.01 for pairwise comparison with placebo after adjustment for multiple comparisons; and tid, 3 times daily.
Percentage of patients reporting new onset or worsening of their initial dyskinesia at each assessment point plotted over time. Adjustments to dopaminergic medications resulted in a reduction in this side effect. tid indicates 3 times daily.
Adler CH, Singer C, O'Brien C, Hauser RA, Lew MF, Marek KL, Dorflinger E, Pedder S, Deptula D, Yoo K, for the Tolcapone Fluctuator Study Group III. Randomized, Placebo-Controlled Study of Tolcapone in Patients With Fluctuating Parkinson Disease Treated With Levodopa-Carbidopa. Arch Neurol. 1998;55(8):1089–1095. doi:10.1001/archneur.55.8.1089
To assess the efficacy and tolerability of the catechol-O-methyltransferase inhibitor tolcapone in reducing "off/on" fluctuations in levodopa-treated parkinsonian patients.
A randomized, double-blind, placebo-controlled, parallel-group study.
Fifteen Parkinson disease clinics.
Two hundred fifteen referred outpatients with Parkinson disease who showed predictable end-of-dose motor fluctuations that were not controlled by a stable levodopa-carbidopa (Sinemet) regimen of at least 4 weeks' duration.
In addition to their usual levodopa-carbidopa regimen, patients received placebo or tolcapone, 100 or 200 mg, 3 times daily orally for 6 weeks.
Primary Outcome Measure
Change in daily off/on time.
Tolcapone, 100 and 200 mg 3 times daily, reduced off time by 2.0 and 2.5 hours per day, respectively, and increased on time by 2.1 and 2.3 hours per day, respectively (P<.001 vs placebo). Investigators' global measures of disease severity indicated that significantly more tolcapone-treated patients had reduced wearing off and symptom severity (P<.001 vs placebo). No significant change in quality-of-life measures occurred. Clinical improvements occurred despite a reduction in total daily levodopa dose of 185.5 mg (23%) in the tolcapone, 100 mg 3 times daily, group and 251.5 mg (29%) in the 200 mg 3 times daily group. Principal adverse events (mainly dyskinesia and nausea) were levodopa related, were not treatment limiting, and were seldom reported as reasons for withdrawal. The frequency of withdrawals because of adverse events was similar in all groups (3% to 7%).
Tolcapone was well tolerated and substantially increased on time and reduced off time in patients with fluctuating Parkinson disease. Additionally, levodopa requirements were significantly decreased.
TOLCAPONE (3,4-dihydrox-4‘-methyl-5-nitrobenzo-phenone) is a selective and reversible catechol-O-methyltransferase inhibitor1 that is being developed as adjunctive therapy to levodopa plus a peripheral dopa decarboxylase inhibitor (benserazide or carbidopa) in the treatment of Parkinson disease.
In pharmacokinetic studies in healthy volunteers receiving combination treatment with levodopa plus carbidopa or benserazide, the area under the plasma concentration–time curve and the elimination half-life of levodopa were increased when tolcapone was added to the regimen. There was no effect on peak concentrations. At the same time, the area under the curve for the methylated derivative of levodopa was reduced.2 These results suggest that tolcapone should stabilize plasma levodopa concentrations in parkinsonian patients who are experiencing "wearing off."3
Preliminary clinical trials showed that tolcapone reduced "off" time and increased "on" time in parkinsonian patients with the wearing-off phenomenon.4- 8 These results were subsequently confirmed in a larger trial, in which tolcapone was shown to be well tolerated in patients with Parkinson disease.9 The present study provides further confirmation of the efficacy and tolerability of tolcapone (100 and 200 mg given 3 times daily [tid]) in increasing on time and reducing off time in levodopa-treated parkinsonian patients with wearing off.
This double-blind, placebo-controlled, randomized, parallel-group trial was conducted at 15 centers in the United States and Canada. All patients provided signed informed consent that was approved by each site's institutional review board.
Patients were eligible for the study if they were at least 30 years old, had at least 2 of the cardinal signs of idiopathic Parkinson disease (rigidity, resting tremor, or bradykinesia), and had been treated with levodopa-carbidopa for at least 1 year, with clear clinical improvement. They were required to be taking at least 4 daily doses of levodopa-carbidopa (Sinemet), or 3 doses if at least 2 were controlled release, and to show predictable end-of-dose wearing off that could not be eliminated by adjusting their existing antiparkinsonian medications. They had to be able to keep reliable diaries of off time and on time, alone or with the help of a family member. The minimum acceptable dose of carbidopa was 20 mg with each dose of levodopa or a total daily dose of 70 mg. Women had to be sterile or using effective contraception. Patients were required to have been taking a stable regimen of levodopa (either immediate-release or controlled-release formulations), dopamine agonists, selegiline hydrochloride, or other antiparkinsonian drugs for at least 4 weeks before randomization.
The principal exclusion criteria were nonidiopathic Parkinson disease or parkinsonian variants, sudden and unpredictable off/on fluctuations, or a diphasic pattern of dyskinesias. Other exclusion criteria included treatment with centrally acting dopamine antagonists or monoamine oxidase inhibitors (other than selegiline) within the previous 2 months, drug or alcohol abuse within the previous 2 years, psychotic illness or major depression within the previous 6 months, and any other clinically significant medical or neurological abnormalities.
Eligible patients underwent an evaluation with medical history and examination, laboratory tests, and an electrocardiogram. Patients received instructions from the investigator on completing the diary forms (see below) and had to complete at least 3 diaries each week during the 3- to 4-week screening period (only the last 3 diaries completed each week were assessed). The diaries had to reveal at least 2 separate off periods after the first on period of the day, or at least 3 hours of off time after the first on period of the day. Patients who qualified were then assessed by means of the Unified Parkinson's Disease Rating Scale (UPDRS)10 in the on state only, and completed the Sickness Impact Profile (SIP)11 on the day before randomization.
Patients who completed the screening period and satisfied the criteria for both inclusion and exclusion were allocated randomly to treatment groups. Patient randomization numbers were generated in blocks of 6 for each center by the sponsor's drug-packaging department and incorporated into double-blind labeling. These randomization numbers were then assigned sequentially in each center in the order in which the patients were enrolled.
During the study, both the patients and the investigators were blind to the treatment. Blinding was effected by the use of matching placebo tablets. No open key to the blinding code was available to investigators, study monitors, or the sponsor's employees. The investigators received medications bearing sealed coded labels (containing the identity of the treatment dispensed) for each patient. At the final monitoring visit, all labels were counted and checked by the study monitor for evidence of tampering before being returned to the sponsor. No labels were opened during the study.
In addition to their usual levodopa-carbidopa regimen, patients randomly received placebo or tolcapone, 100 or 200 mg tid. The first daily dose of tolcapone or placebo was given with the first dose of levodopa-carbidopa; the second and third doses of tolcapone or placebo were taken at 6-hour intervals thereafter, regardless of meals or subsequent levodopa-carbidopa intakes. Treatment was continued for 6 weeks.
After the first day of study medication, the dosage of levodopa could be modified either by increasing the dosage interval or by reducing the dose at each time point, if the patient had dopaminergic-related side effects. If this occurred, the dosage could be increased at a later date, but the total levodopa dose or dosage frequency was not to exceed that at baseline. If dopaminergic-related adverse events did not resolve after reduction of the levodopa dose, adjustments in concomitant antiparkinsonian medication were allowed. These medications could be decreased or discontinued if an adverse event was believed to be specific to that agent.
Efficacy and safety of the study drug were assessed between baseline (the day before study medication was first administered) and week 2 of treatment and at weeks 3 (±3 days) and 6 (±3 days). Patients were examined by the same investigator at all visits and at the same time of day as the baseline evaluation. The primary efficacy measure was the change from baseline in off/on time recorded on the patients' diaries. For each 30-minute period during the day, the patients were asked to rate their mobility as off (poor mobility or complete immobility), on (good to excellent mobility), intermediate (neither off nor on), or asleep. In addition to the diaries completed during the screening period, patients had to complete diaries on at least 3 days during the week before each visit; again, only the last 3 diaries of each week were used to assess the patient's off/on status to avoid bias.
Secondary efficacy measures were used to assess the patients' clinical status. The change in UPDRS subscales I (mentation, behavior, and mood), II (activities of daily living) (on and off states), III (motor function) (on state only), IVa (dyskinesia), and IVb (clinical fluctuations) scores were evaluated. Investigator's Global Assessments (IGAs) of change were made for symptom severity, the wearing-off phenomenon, and overall efficacy and tolerability; a 9-point scale, ranging from 4 (very marked improvement) to−4 (very marked deterioration), was used in these assessments. A 7-point scale, ranging from 3 (marked reduction) to−3 (marked increase), was used to evaluate dyskinesia. The SIP was assessed at baseline and after 6 weeks to evaluate quality of life; UPDRS subscale II was also used as a measure of quality of life.
The planned total sample size for this study was 180 patients (60 per treatment group) because 30% of the patients were expected to have unevaluable data or to be unavailable because of withdrawal. This would provide a sample of 126 patients (42 patients per treatment group). This sample size allows a power of 80% at the 2-sided α level of 0.05 in detecting a 30% difference (which we judged to be a clinically relevant change) between an active treatment group and placebo. On the basis of the results of previous tolcapone studies,4- 9 the standardized effect size (mean divided by SD) of treatment difference in reduction of average percentage on or off time from baseline to week 6 was assumed to be approximately 0.6.
Information on adverse events was collected at each visit. Checklists were used to elicit information on dopaminergic symptoms associated with levodopa treatment. An adverse event was defined as the worsening of a previous symptom or the emergence of a new symptom. Vital signs (heart rate and standing and supine blood pressures) were recorded at each visit. Laboratory tests were performed at baseline and after 3 and 6 weeks. A 12-lead electrocardiogram was recorded at baseline and after 6 weeks.
Efficacy was assessed in part by means of an intention-to-treat analysis. Patients were included in this analysis if they had been randomized to treatment, had taken at least 1 dose of study medication, and had at least 1 follow-up visit. Any missing data were replaced by the last available observation after baseline. The safety analysis included all patients who were randomized to treatment, received at least 1 dose of study medication, and had at least 1 safety evaluation.
Off/on time, UPDRS and SIP scores, and levodopa dosage were regarded as continuous variables and analyzed by analysis of covariance, with Bonferroni-Holm adjustment to correct for multiple comparisons. The 2-way analysis of covariance model included main effect terms of center and treatment, center-by-treatment interaction term, and the baseline value as covariate. The IGAs of change in symptom severity, the wearing-off phenomenon, and overall efficacy were regarded as categorical variables and analyzed by the Cochran-Mantel-Haenszel test, with Bonferroni-Holm adjustment. For each dose of tolcapone, the null hypothesis (that mean changes are the same for placebo and the dose of tolcapone) was tested against the alternative hypothesis (that mean changes are different between placebo and the dose of tolcapone). For each treatment, least-squares estimates of the actual treatment mean (least-squares means) were reported. The difference between estimates for placebo and each tolcapone treatment group was also calculated. The 95% confidence intervals of treatment differences were constructed on the basis of the least-squares means. Data were entered and analyzed by statisticians at Hoffmann-La Roche, Nutley, NJ.
In total, 215 patients correctly completed their off/on self-rating charts according to the protocol during the screening period and were randomized to treatment (Figure 1). The 3 groups were well matched, on average, in terms of patient characteristics and disease history (Table 1). Patients entered into the study had suffered from Parkinson disease for approximately 10.5 years and had been taking levodopa for approximately 8.5 years. Most patients (194 of the 215) had a Hoehn and Yahr score of 2 to 3 during on time, and 163 of the 215 (76%) experienced "on" dyskinesias before randomization.
The effects of tolcapone treatment on levodopa dosage, off/on time, UPDRS scores, IGAs, and SIP scores are summarized in Table 2.
Analysis of the patients' diaries showed that off time was significantly reduced and on time significantly increased in both tolcapone groups, compared with the placebo group (P<.001; Table 2). The onset of response was rapid and marked: improvement was observed after 2 weeks of treatment (first assessment performed after baseline; Figure 2). By week 6, on time had increased by more than 2 hours, and off time had decreased to a similar extent in both tolcapone groups. This represents a reduction in off time of up to 38%.
A significant reduction was seen in total levodopa dose and number of intakes of levodopa per day in both tolcapone groups compared with the placebo group (Table 2; Figure 3). The total daily dose of levodopa decreased in the tolcapone groups by 23% in those receiving 100 mg tid (P<.01 vs placebo) and by 29% in those receiving 200 mg tid (P<.001 vs placebo). The mean number of daily levodopa intakes decreased by 0.5 in the 100-mg tid group (P<.01 vs placebo) and by 1.2 in the 200-mg tid group (P<.001 vs placebo).
Tolcapone-treated patients had significant improvements on IGAs of change compared with placebo (Table 2). Investigators judged 50 (72%) of 69 patients who received tolcapone, 100 mg tid, and 57 (77%) of 74 who received tolcapone, 200 mg tid, to be improved, compared with only 19 (26%) of 72 patients taking placebo (P<.001). The IGAs of change in wearing off and symptom severity also improved significantly compared with placebo (P<.001). The score for the UPDRS subscale III (motor function in the on state) was reduced in tolcapone-treated patients, but the reductions were not significantly different from those in placebo-treated patients. The UPDRS subscale II (activities of daily living), total UPDRS subscales I to III, and SIP scores showed no significant difference between the tolcapone and placebo groups.
Of the 215 patients entered into the study, 13 withdrew, 11 because of adverse events (5  in the placebo group and, in the tolcapone groups, 2  of those receiving 100 mg tid and 4  of those receiving 200 mg tid). In the placebo group, most withdrawals occurred within 4 weeks of starting treatment, whereas in the tolcapone groups, most withdrawals occurred after 4 to 6 weeks.
Adverse events were reported by 53 (74%) of 72 patients in the placebo group, compared with 59 (86%) of 69 receiving tolcapone, 100 mg tid, and 72 (97%) of 74 receiving tolcapone, 200 mg tid. Most adverse events were mild or moderate; of the 658 events reported overall, only 95 (14%) were rated as severe (an adverse event was classified as severe even if it was usually mild but reached a peak of severity at some point during the patient's treatment). The frequency of severe adverse events was 11% in the placebo group and 11% in those receiving 100 mg tid and 19% in those receiving 200 mg tid.
Twelve adverse events were reported at an incidence of at least 5% greater in 1 or both tolcapone groups than in the placebo group (Table 3). The most frequent adverse events in tolcapone-treated patients were dopaminergic-related events: dyskinesia, nausea, dystonia, anorexia, confusion, hallucination, and excessive dreaming. The most frequent nondopaminergic adverse events were dizziness (nonorthostatic), increased sweating, and dry mouth (Table 3). Dopaminergic-related events tended to occur more often in patients receiving tolcapone, 200 mg tid; most resolved with decreases in levodopa dosage.
Dyskinesia was by far the most frequent adverse event, occurring at some point during the study in 14 (19%) of 72 patients in the placebo group; 43 (62%) of 69 in the tolcapone, 100-mg tid group; and 49 (66%) of 74 in the tolcapone, 200-mg tid group. However, all 3 treatment groups showed similar incidences at baseline and week 6 in duration of dyskinesia and painful dyskinesia according to the UPDRS subscale IVa (dyskinesia ratings). The degree of disability resulting from dyskinesia was also similar in all 3 groups. No difference in early-morning dystonia was observed between the treatment groups.
Worsening of dyskinesia, as with most dopaminergic symptoms, occurred most often at the start of treatment. The prevalence diminished with time, so that by the end of the study 5 (7%) of 72 patients in the placebo group; 23 (33%) of 69 in the tolcapone, 100-mg tid group; and 25 (34%) of 74 in the tolcapone, 200-mg tid group had developed dyskinesia or had worse dyskinesia than at baseline (Figure 4). Dyskinesia was not, however, cited as the reason for withdrawal by any patient who withdrew because of an adverse event and was therefore not regarded as treatment limiting.
The IGA of overall tolerability showed that, in the tolcapone groups, 19 (26%) of 69 patients receiving 100 mg tid and 27 (36%) of 74 receiving 200 mg tid showed improvement, compared with 11 (15%) of 72 in the placebo group. However, the proportion of patients who showed deterioration was also higher in the tolcapone treatment groups (15/69  in the 100-mg tid group, 19/74  in the 200-mg tid group) than in the placebo group (7/72 ). No consistent changes in vital signs, electrocardiogram, or laboratory tests were associated with tolcapone treatment.
The present study showed significant reductions in wearing-off effects in patients with fluctuating Parkinson disease who were treated with tolcapone, and findings were consistent with previous studies.4- 9 The results obtained with the 2 dosages of tolcapone were similar, although both the therapeutic effect and the incidence of adverse events tended to be somewhat greater in the group receiving tolcapone, 200 mg tid.
In the present study, treatment with tolcapone was associated with rapid decreases in off time and increases in on time, assessed by patient diaries and IGAs; indeed, improvement was noticeable within the first 2 weeks of treatment. These data indicate a rapid onset of response to tolcapone in patients with fluctuating disease receiving levodopa therapy. No significant treatment effect was seen on the UPDRS subscale II (activities of daily living during on time) or the SIP scores during the short duration of the study. The reduction in off time in itself might be considered as an improvement in patients' quality of life.
The clinical improvements associated with tolcapone treatment occurred despite a mean decrease in total daily levodopa dose. Both the total levodopa dose and the number of daily intakes were reduced. This is consistent with previous findings that treatment with tolcapone increases levodopa bioavailability2,12 and, hence, leads to reduced levodopa requirements in parkinsonian patients. These patients had already undergone levodopa adjustments, so it is believed that these benefits could not be obtained by further levodopa dose manipulation. The mechanism of action for tolcapone appears to be its inhibition of catechol-O-methyltransferase,1 which may occur in both the peripheral and the central nervous system, because the drug penetrates the blood-brain barrier.13
In general, tolcapone was well tolerated; the principal adverse events were dopaminergic-related and may be secondary to increased levodopa bioavailability or to tolcapone itself. New-onset or an increase in dyskinesia occurred in more than 65% of tolcapone-treated patients, compared with 19% in the placebo group. Nausea was the second most frequent adverse event, reported by approximately 28% of patients in both active treatment groups, compared with 11% in the placebo group. This is consistent with studies of other dopaminergic agents used as adjunctive therapy with levodopa in patients having motor fluctuations.14,15 These dopaminergic adverse events were managed by adjusting the levodopa dosage: the prevalence of dyskinesia decreased from week 2 of treatment onward. They were not reported as a cause of withdrawal from treatment, indicating that they were not treatment limiting. The increased prevalence of dyskinesias in tolcapone-treated patients might conceivably have led to unblinding; however, dyskinesias also occurred in one fifth of the patients who received placebo, indicating that these adverse events were not exclusive to tolcapone-treated patients. Therefore, the extent of potential unblinding and any consequent introduction of bias would have been limited.
In conclusion, the present study shows that adjunctive treatment with tolcapone was well tolerated, rapidly and significantly reduced off time and increased on time in patients with fluctuating Parkinson disease, and resulted in reduced daily levodopa requirements. No definite advantage in terms of efficacy was observed with the 200-mg tid dose, which was also associated with a slightly higher incidence of adverse events, suggesting that treatment should be initiated with the lower dose. Whether adjunct therapy with tolcapone has advantages over a dopamine agonist is unclear, although 1 study has shown that tolcapone is more effective than bromocriptine in reducing off time and levodopa dose and was better tolerated.16
Accepted for publication January 22, 1998.
This work was supported by F. Hoffmann-La Roche, Basel, Switzerland.
We thank all the site coordinators and patients for their dedication and persistence.
Mayo Clinic Scottsdale, Scottsdale, Ariz: Charles H. Adler, MD, PhD; Manfred D. Muenter, MD; John N. Caviness, MD. Vanderbilt University, Nashville, Tenn: Thomas Davis, D. A. Shearon. Albany Medical College, Albany, NY: Stewart Factor, DO, Eric S. Molho, MD. Tampa General Hospital, Tampa, Fla: Robert A. Hauser, MD; Theresa Zesiewicz, MD. Barrow Neurological Institute, Phoenix, Ariz: Matt Kurth, MD; Abraham Lieberman, MD; Susan Imke, RN. Yale Movement Disorders, New Haven, Conn: Kenneth L. Marek, MD; J. Cellar; B. Fassell. Colorado Neurological Institute Movement Disorders Center, Englewood: Christopher O'Brien, MD; Lauren C. Seeberger, MD. Washington University School of Medicine, St Louis, Mo: Joel Perlmutter, MD. Boston University School of Medicine, Boston, Mass: Marie-Helene Saint-Hilaire, MD; Robert G. Feldman, MD. The Parkinson's Institute, Sunnyvale, Calif: Caroline Tanner, MD; James W. Tetrud, MD; Heidi Shale, MD. Sinai Clinical Neuroscience Center, West Bloomfield, Mo: Richard Trosch, MD; Peter A. LeWitt, MD; R. C. Berchou; K. L. Mistura. University of Southern California, Los Angeles: Cheryl Waters, MD; Mark Lew, MD. University of Miami School of Medicine, Miami, Fla: William Weiner, MD; Carlos Singer, MD; Lisa M. Shulman, MD. University of Ottawa, Ottawa, Ontario: Lynn Barclay, MD; J. David Grimes, MD; T. Mandis. Markham Stouffville Hospital, Markham, Ontario: Mark Guttman, MD.
Reprints: Charles H. Adler, MD, PhD, Parkinson's Disease and Movement Disorders Center, Mayo Clinic Scottsdale, 13400 E Shea Blvd, Scottsdale, AZ 85259.