Subscribe to eTOC

No Harm in Using Levodopa Early in Parkinson's for Symptom Relief, but the Drug Does Not Provide a Disease-Modifying Effect, Study Finds

Article In Brief


Unified Parkinsons Disease Rating Scale (UPDRS) and Parkinsons Disease Questionnaire-39 scores during the trial: The mean UPDRS total scores (Panel A) and Parkinsons Disease Questionnaire-39 (PDQ-39) scores (Panel B) for each trial group are shown. UPDRS scores range from 0 to 176, with higher scores indicating more severe disease; the scale includes subscales of mental function, activities of daily living, and motor function. PDQ-39 scores range from 0 to 100, with higher scores indicating a lower disease-related quality of life.

A large randomized study found levodopa, in combination with carbidopa, did not have a disease-modifying effect on Parkinson's disease, either beneficial or detrimental, over the 80 weeks of the trial.

While early initiation of levodopa does not appear to slow progression of Parkinson's disease (PD), there is no downside to using it in patients with symptoms that could be helped by the drug, suggests a paper published January 24 in the New England of Medicine (NEJM).

The new findings might just settle the long-running debate over whether it is wise to start using levodopa early in the course of Parkinson's disease (PD), the study authors and independent experts told Neurology Today.

There has been a lot of angst about the early use of levodopa, with some patients and neurologists firmly against it, fearing side effects and other complications that might make things worse for patients in the long run. Previous research on the early use of levodopa for early PD produced conflicting and, in some cases, ambiguous results, helping to create two camps of opinion, the second one seeing a clear benefit to using levodopa as early as needed for symptom management.

Now, a randomized, multicenter, delayed-start trial of levodopa, conducted in the Netherlands, arrived at two key points: Levodopa, in combination with carbidopa, did not have a disease-modifying effect, either beneficial or detrimental, over the 80 weeks of the trial. But at the same time, study patients started on levodopa (100 mg three times a day) at the beginning did not experience more dyskinesia and levodopa-related fluctuations in motor response compared with patients who were delayed in starting the drug, a finding that suggests that there is no harm in starting the drug as soon as it is needed.

“I think the study will have an impact on clinical practice,” said study coauthor Rob M.A. de Bie, PhD, a neurologist who specializes in movement disorders at the University of Amsterdam.

“Physicians can more assuredly tell the patient that there is no long-term detrimental effect of starting levodopa early,” Dr. de Bie said in an email to Neurology Today. He said patients leery of the drug may be heartened to hear that, “starting earlier with levodopa may improve their quality of life right at the time of diagnosis. “

Dr. de Bie said that “apart from nausea, which was slightly more frequent in the early-start group, we did not find any downsides of starting levodopa early.”

The study authors cited the ELLDOPA (Earlier Versus Later Levodopa Therapy in Parkinson's Disease), published in NEJM in 2004, that found that after 40 weeks and a two-week drug washout period, patients who were given levodopa had not deteriorated as much as patients assigned to placebo. That “suggested that the drug has slowed the progression of Parkinson's disease or that the drug had had a prolonged effect on symptoms that was interpreted as ameliorating the underlying disease,” they wrote.

Neuroimaging data from that same trial, however, pointed to a possible neurotoxic effect from levodopa, suggesting the drug may accelerate the loss of dopamine nerve terminals.

Two-Phase Study Design

For the current so-called LEAP trial, researchers from Amsterdam University Medical Centers designed a delayed-start trial of levodopa to determine if they could separate any possible disease-modifying effect of the drug from any effect on symptoms.

Patients were recruited from 50 community hospitals and seven academic hospitals in the Netherlands. Patients were eligible to take part in the trial if they had received a diagnosis for PD within the previous two years from an experienced neurologist using standard criteria, if they had insufficient disability to warrant treatment with antiparkinsonian medication, if they were at least 30 years old, and if they had a life expectancy of at least two years. Patients previously treated with PD medication, including levodopa and dopamine agonists, were excluded. Also excluded were patients for whom tremor was their most prominent symptom, anyone with dementia, or patients whose symptoms were characteristic of atypical or secondary parkinsonism.

After a baselines assessment, 445 patients with early PD were randomly assigned—222 to the early-start group and 223 to the delayed-start group. Early start consisted of levodopa (100 mg three times a day) plus carbidopa (25 mg three times per day) for 80 weeks. Delayed start consisted of placebo for 40 weeks followed by levodopa plus carbidopa for 40 weeks at the same dosing used for the early start.

The primary outcome was the between-group difference in the mean change from baseline to week 80 in the total score on the Unified Parkinson's Disease Rating Scale (UPDRS)—scores range from 0-176, with higher scores indicating more disability. The mean UPDRS score in the early-treatment group was 28.1 at baseline, compared to 29.3 for the delayed-start group.

The study found that levodopa had a positive effect on symptoms of Parkinson's disease in the first 40 weeks of the trial when those on the drug were compared to those on placebo. But there was “no significant difference in the UPDRS score at week 80 (with all patients receiving active treatment from week 40 on), which indicates that the severity of parkinsonian symptoms at the end of the trial did not differ significantly between patients who received early initiation of the drug and those who received delayed initiation,” the study reported.

The mean change in UPDRS from baseline to week 80 was -1.0 for the early initiation group and -2.0 for the delayed group, a non-significant difference.

“These findings imply that levodopa had no disease-modifying effect on Parkinson's disease over the period of the trial,” the study concluded. The researchers noted that “whether higher doses of the drug, longer periods of administration, or initiation of the drug at later stages of the disease could alter the course of Parkinson's disease warrants evaluation in future trials.”

An accompanying editorial in the NEJM, by Susan Bressman, MD, and Rachel Saunders-Pullman, MD, MPH, from the Icahn School of Medicine at Mount Sinai, said the study had limitations, including the fact that 39 percent of patients in the delayed-start group ended up on levodopa during the first 40 weeks because of symptoms (cross-over was allowed in the trial design), which could have affected the final outcome. But the editorialists said the findings nonetheless support the practice that “there is no reason to delay therapy when it is clinically indicated.”

“The results of the current trial, taken together with those of other trials, support treatment that is guided by clinical need and that uses the lowest dose that provides a satisfactory clinical effect,” Drs. Bressman and Saunders-Pullman wrote.

Expert Commentary

Melissa Nirenberg, MD, PhD, FAAN, a movement disorders specialist and adjunct professor of neurology at NYU School of Medicine, said she hoped the new findings would counter what has been described as “levodopa phobia” in the medical literature and clinical practice.

“Levodopa phobia has been a major problem, because it has led patients and providers to be afraid to initiate therapy with the most effective treatment for Parkinson's disease,” she said, putting off levodopa therapy for as long as possible, even amid bothersome symptoms or worsening disability that could interfere with work and other activities. “Many patients have been under-dosed as a result and suffered significant impairment and reduced quality of life for no reason.”

Dr. Nirenberg said fears that levodopa might be toxic to the brain led some neurologists to turn to dopamine agonists for early treatment of Parkinson's disease, medications that she said can have troubling side effects such as impulse control disorders, including hypersexuality and compulsive eating, shopping and gambling. In other cases, doctors may use a dose of levodopa that is insufficient to effectively control motor symptoms such as rigidity, bradykinesia, and shuffling gait.

“Another myth about levodopa is that it stops working after several years,” said Dr. Nirenberg, who specializes in Parkinson's disease and other movement disorders. “In fact, it is the progression of the disease rather than the timing of initiation of levodopa therapy that affects the response to medications. When levodopa is started later, patients may just miss the initial ‘honeymoon period,’ in which there is usually an excellent response to treatment without motor complications such as fluctuations or dyskinesias,” she said.

She said she spends considerable time dispelling inaccurate information with Parkinson's patients who come in with the mindset that levodopa is not for them.

“Based on the findings of this study, levodopa should not be started before it is needed,” Dr. Nirenberg said. But “once dopaminergic therapy is indicated to control the symptoms of Parkinson's disease,” levodopa has considerable benefit, and there is “no reason to wait to initiate treatment with this safe and highly effective medication.”

Roy Alcalay, MD, assistant professor of neurology at Columbia University, said he thought the study results bring levodopa back to where it was at the start: “Levodopa should be used for what it is designed for: to treat the symptoms of Parkinson's disease.” Symptom control can help keep people active in work and leisure and prevent falls due to balance and gait issues, he said.

He said the delayed-start study was well designed to sort out the question of symptom relief versus longer-term disease modification, for which it found none. Dr. Alcalay, who studies the genetics and biomarkers of PD, said research in the field is focused on other therapeutic approaches to slowing the progression of PD. He said he believed “the answer is not in dopamine-related intervention.”

John G. Nutt, MD, FAAN, professor of neurology at Oregon Health & Science University, said the new study is informative, but not a “game changer by any means,” noting that “the beneficial effects in the ELLDOPA study were well explained by the long-duration response to levodopa.”

“I think I do what most clinicians do. If patients have symptoms that interfere with their life and they have symptoms that are bothering them, then I would start the drug. This study suggests that you aren't going to lose by doing that.”

Dr. Nutt said he thinks the study would have been more interesting if it had introduced exercise into the equation to see if patients in the early-start group were able to maintain an exercise regime better than the delayed-start group and that may have translated into a better clinical state at the end of the trial.

There is some research to suggest that exercise may have a neuroprotective effect in Parkinson's disease, Dr. Nutt said, as well as a practical one in keeping patients active and more mobile.

Dr. Nutt said he is involved in a gene therapy trial for Parkinson's that is investigating the effects of AADC (aromatic L-amino acid decarboxylase) on levodopa pharmacokinetics and pharmacodynamics.


Dr. de Bie received grant support from GE Health and Medtronic. Drs. Nirenberg, Alcalay, and Nutt had no disclosures.

Link Up for More Information

•. Verschuur CVM, Suwijn SR, Boel JA, et al; for the LEAP Study Group. Randomized delayed-start trial of levodopa in Parkinson's disease N Engl Med 2019; 380(4):315–324.
    •. Bressman S, Saunders-Pullman R. When to start levodopa therapy for Parkinson's disease N Engl J Med 2019; 380(4):389–390.