The initial therapeutic studies of levodopa in PD were carried out in the 1960s and early 1970s. The subjects were of varying disease durations, some quite advanced with dementia, and standard measures such as the Unified Parkinson's Disease Rating Scale (UPDRS) were not yet devised; however, the results were dramatic (3). In 1967, Cotzias et al. (11) demonstrated the definitive effectiveness of high-dose
L-Dopa (as opposed to D, L-Dopa). These investigators examined 28 patients in an open-label manner with intermittent blinded replacement with placebo and utilized levodopa without a dopa decarboxylase inhibitor. The duration of disease ranged from 1 to 30 years (mean 10 years). All patients responded, with 20 having a marked to dramatic improvement. All motor features improved. Some patients developed fluctuations and dyskinesia quickly and it was suggested that these problems related to duration of disease.
Many studies followed which supported these findings (13,42-44). An example is that by Sweet and McDowell (45) who studied 100 patients treated for up to five years in an open-label fashion. Forty-seven of them completed the whole five years. All signs of PD improved remarkably by six months (60% of patients were more than 50% improved) and despite worsening over the next 4.5 years, the Cornell-weighted scores remained significantly better than baseline. The severity of parkinsonian features at initiation of therapy had little bearing on the ultimate response. Positive results were observed despite the fact that more than half the patients suffered from concomitant dementia. It became clear that levodopa was not a cure for PD as previously hoped, as it did not stop disease progression and was associated with several late complications, including motor fluctuations and dyski-nesia. However, patients with advanced disease were included, dopa decarboxylase inhibitors were not used in a majority of patients, and patients were treated with the maximum tolerated dose, all of which may have increased the likelihood of dyski-nesia and fluctuations. Other studies indicated that lower doses of levodopa result in a similar response with fewer complications (46), whereas others were not in agreement (47).
Several studies reported in the last decade provide more information about the effectiveness of levodopa therapy. They include comparisons of the immediate release (IR) and controlled release (CR) formulations (48), comparisons of levodopa and dopamine agonists in early PD (49-58), and a dose-finding levodopa trial in very early disease (ELLDOPA) (59-60). The populations of patients in these studies were more homogeneous than the earlier trials, as the patients had PD < 5 years and were nonfluctuators at the time of initiation. Recent studies have shown varied frequencies of late complications even in these populations. The variances probably relate to the manner in which they are defined and detected. The CR First study (48) was a five-year, randomized, double-blind study comparing CR and IR carbidopa/lev-odopa in 618 levodopa naive patients (mean duration of disease of 2.3 years). The primary end point was the time until onset of motor fluctuations. Sixty percent of patients completed the five-year study. Mean doses of levodopa in both groups were low (400-500 mg/day). There were no differences between the two formulations with regard to efficacy or frequency of motor fluctuations. Despite low doses, there was a significant improvement in the UPDRS motor score that gradually diminished over time but remained better than the baseline score. Approximately 20% of patients in each group developed wearing-off and dyskinesia, which was far lower than previously reported frequencies.
The CALM-PD study (49-50), a parallel-group, double-blind, randomized trial consisting of both clinical and imaging sub-studies, compared the rates of dopamin-ergic motor complications and dopamine transporter (DAT) ligand uptake after initial treatment of early PD with pramipexole versus levodopa. During the first 10 weeks of the study, patients were titrated with the study medication to one of three possible doses to treat disability, and from week 11 to month 48 were permitted to add open-label levodopa to treat continuing or emerging disability. After month 24, patients were permitted to alter the dosage of the study medication or change open-label levodopa without losing the blind. The two-year data reported that 28% of patients assigned to pramipexole developed motor complications compared with 51% of patients assigned to levodopa (P < 0.001) (50). However, the mean improvement in UPDRS motor score was significantly greater in the levodopa group compared with pramipexole (9.2 vs. 4.5 units, P < 0.001). When extended to four years, slightly more than half (52%) of the patients initially assigned to the pramipexole group developed motor complications compared with 74% of the levodopa-treated patients (49). The mean improvement in UPDRS motor scores from baseline through 48 months was significantly greater in the levodopa group than the pramipexole group, which had a worsening compared to baseline. Fifty-five percent of patients in the pramipexole arm and 67% of the patients in the levodopa arm completed the study. The imaging portion of the study (51) included a subgroup of 82 patients that underwent four sequential I B-CIT single photon emission tomography (SPECT) scans over a 46-month period to compare the rate of loss of DAT binding between the treatment groups. It was assumed that a reduction in striatal I B-CIT uptake is a marker of dopamine neuron degeneration. A 40% relative reduction in the rate of loss of uptake when comparing pramipexole to levodopa was reported.
A similar five-year comparison of ropinirole and levodopa (056 study) in 268 patients was reported (52). One subject was randomized to levodopa for every two that were randomized to ropinirole. Open-label levodopa supplementation was allowed. Approximately half of the patients completed the study. At a mean dose of 16.5mg/day, ropinirole was well tolerated. The primary endpoint was the appearance of dyskinesia, as measured by item 32 on the UPDRS. They were shown to occur earlier and more frequently in patients treated with levodopa than ropinirole. Regardless of levodopa supplementation, 20% of ropinirole subjects experienced dyskinesia by the end of five years versus 45% of levodopa subjects. Prior to the addition of levodopa, 5% of the ropinirole group and 36% of the levodopa group developed dyskinesia. Twenty-three percent of the patients in the ropinirole arm and 34% in the levodopa arm developed wearing-off. The change from baseline in the UPDRS activities of daily living (ADL) score was similar between the two groups; however, there was a significantly greater improvement of 4.5 points in favor of the levodopa group in the UPDRS motor score.
The REAL-PET study (53) was a prospective two-year, double-blind, randomized study comparing ropinirole and levodopa in early untreated PD. Open-label levodopa supplementation was allowed as necessary. The primary outcome measure was reduction in 18F-dopa uptake between baseline and two years, as measured by three-dimensional positron emission tomography (PET) scans. Patients were scanned after four weeks of therapy and at the end of two years. Clinical outcomes were secondary. There were 162 PD patients with less than two years of disease and Hoehn and Yahr stage < 2.5 randomized. There was a lower reduction in 18F-dopa uptake with ropinirole (-13.4%) compared to levodopa (-20.3%), a relative difference of 34% (similar to the CALM-PD results). Clinically, dyskinesia occurred in 4% of ropinirole-treated patients and 27% of levodopa-treated patients, but the UPDRS motor score improved by six points with levodopa and worsened by one point with ropinirole over the two-year period.
The PELMOPET study (Pergolide versus L-Dopa Monotherapy and Positron Emission Tomography trial) (54) was a multicenter, double-blind, randomized, three-year trial comparing pergolide (n = 148) to levodopa (n = 146) without levodopa res cue in early untreated PD patients. The primary outcome measures were clinical efficacy (UPDRS), severity and time to onset of motor complications, and disease progression. During the three years, severity of motor complications was significantly lower and time to onset of dyskinesia was significantly delayed in the group receiving pergolide (3.23mg/day) compared with levodopa (504mg/day). However, time to onset of motor fluctuations was not longer in patients receiving pergolide after three years. UPDRS total, motor and ADL scores, and patient and physician global impressions of severity and improvement were significantly better in patients receiving levodopa. The PET results were similar to REAL-PET favoring pergolide.
Whether these changes in SPECT and PET scans of the CALM-PD, REAL-PET, and PELMOPET studies suggest a protective effect of the dopamine agonist with respect to levodopa or that levodopa may accelerate the rate of loss of uptake or that this is a differential pharmacological effect between drugs is not clear, given the limits of the study designs (55-58). The potential for modification of the imaging outcomes pharmacologically by dopaminergic agents has led to a questioning of the value of such measures under these conditions. It should be noted that in these studies some patients had normal scans. These patients have been referred to as SWEDDs (scans without dopaminergic deficits) and were eliminated from the analyses. The diagnoses in these subjects remain unclear.
The ELLDOPA (Early versus Later Levodopa Therapy in Parkinson's Disease) study (59-60) was a multicenter, placebo-controlled, randomized, double-blind study. The objectives were to examine the impact of levodopa on disease progression. There were 361 PD patients with disease <2 years, Hoehn and Yahr stage <3, and not likely to require symptomatic therapy within the next nine months that were randomized to receive placebo, 50 mg levodopa three times a day (TID), 100 mg TID, or 200 mg TID; 311 completed the study. The study involved 40 weeks of therapy, including a three-day taper and a two-week withdrawal period. The primary outcome was the change in severity of parkinsonism between baseline and week 42, as measured by the total UPDRS score. There was a B-CIT SPECT sub-study of 142 subjects where a scan was completed at baseline and week 40. The clinical results were that levodopa, in a dose-response pattern, reduced the worsening of symptoms from baseline to week 42 (after washout) compared to placebo. This means that the higherdose subjects were less severely impaired after the two-week withdrawal than the other three groups. None of the active treatment groups deteriorated to the level of the placebo group after washout. There was also a strong symptomatic dose-response beginning at week nine. The maximum effect was reached at 24 weeks in the 600 mg/day group. The 150 mg/day group returned to baseline at week 27 and the 300 mg/day group at week 40; however, the 600 mg/day group remained better than baseline until withdrawal of the drug. However, the highest dose caused significantly more dyskinesia than placebo. The SPECT study had the opposite results. The reduction in percent uptake of B-CIT was greater in the levodopa-treated groups than controls, although the difference was not significant. However, when the 19 SWEDDs were removed, as in previous studies, the difference became significant (P = 0.036). The conclusions drawn from the study were that there was no clinical evidence that levodopa accelerated PD progression, in fact, the results suggested that levodopa might instead either slow progression of PD or have a persistent pharmacological effect that goes beyond the two-week washout (beyond the long-duration effect). Two findings that led to questioning the latter include that there was little deterioration of PD after the first week of withdrawal and a small subset of 38
subjects were evaluated four weeks after washout and no further deterioration was detected. The explanation of this end result remains unresolved. The opposite finding on the imaging sub-study may be explained by the potential capability of levodopa to downregulate the DAT. From the standpoint of symptomatic effects of levodopa, the results demonstrated a striking symptomatic dose-response with the 600 mg group maintaining improvement beyond baseline for the entire nine months.
Thirty years of experience and literature have led to several conclusions regarding levodopa therapy in PD. It is currently the most potent symptomatic therapy for PD and we have learned quite a bit about the nuances of treatment. We now treat with the lowest effective dose, not the highest tolerated one; we avoid frequent small doses, which only add to unpredictable responses; we often initiate treatment with a dopamine agonist and add levodopa when necessary; and we have developed adjunctive therapies that complement levodopa, all resulting in reduced long-term complications.
Was this article helpful?