The possibility that early treatment with rasagiline (Azilect) might change the course of Parkinson disease (PD) has tantalized the PD community since results from an 18-month trial were announced in 2008. But the trial, which suggested that low-dose, but not high-dose, treatment might slow disease progression, has raised more questions than it answered. Those questions have become even more acute now that the final report on the trial is in print in the Sept. 24 New England Journal of Medicine.
The trial results “don't prove disease modification, but it's as close as we've gotten,” according to lead study author Warren Olanow, MD, chair of the department of neurology and neuroscience at Mount Sinai School of Medicine, in New York. But independent experts are cautious, and are questioning exactly what disease modification means in this context.
DELAYED START TRIAL
Rasagiline is an inhibitor of monoamine oxidase B, and trial results in animal models of the disease have suggested that MAO-B inhibitors can protect dopamine neurons from harm. But testing that proposition in humans is complicated by the drug's effects on symptoms. Since the only current measures of PD disease progression are clinical, does improvement mean the disease itself has been slowed, or simply that symptoms have improved?
To address that problem, PD researchers have turned to the “delayed-start” trial design, in which, during the first phase of the trial, some patients receive active treatment while others receive placebo. In the second phase, the placebo-treated patients also take active treatment, without breaking the blind. By the end of the trial, symptomatic benefits in both groups would presumably be equal, so that differences in disability might be attributable to differences in the duration of treatment.
In the rasagiline trial, 1,176 newly diagnosed PD patients were randomized to rasagiline at either 1 mg or 2 mg per day, or placebo, for 36 weeks (phase 1). In phase 2, placebo-treated patients were switched to one or the other doses of rasagiline for an additional 36 weeks.
Parkinsonian disability was measured by the Unified Parkinson's Disease Rating Scale (UPDRS). To achieve a positive outcome, rasagiline at either dose had to meet each of three hierarchical endpoints. First, the slope of change in UPDRS for patients on active treatment had to be superior to that of patients on placebo during phase 1. A disease-modifying drug would be expected to slow the rate of decline versus placebo.
The second endpoint compared the total change in UPDRS from baseline to 72 weeks in the early-start and delayed-start groups by dose. Longer treatment with a disease-modifying agent would be expected to have a greater overall long-term benefit. Finally, the slope of change in UPDRS for delayed-start patients during phase 2 should not be superior to that for early-start patients —there should be no evidence of catching up.
The results showed that early treatment with 1 mg per day rasagiline met all three endpoints — early-start patients declined more slowly in phase 1, were overall better off at the end of the trial, and declined at the same rate as delayed-start patients in phase 2. The differences were small, though — the UPDRS score increased (worsened) by 2.8 points in the early-start group, and by 4.5 points in the delayed-start group, a difference of 1.7 points over about 18 months.
Early treatment with 2 mg per day rasagiline met the first endpoint, but, surprisingly, given the results of the low dose, failed on the second — there was no significant difference between the groups in the overall change in UPDRS, which might be expected from a higher dose of a disease-modifying drug.
The explanation, Dr. Olanow said, may be in symptomatic effects of rasagiline, which may have masked any difference in disease progression in these very mildly disabled patients. “The UPDRS has a floor. You reach a point where it doesn't get better.”
Post-hoc analysis of patients most disabled at baseline seemed to support this explanation — in that group, high-dose rasagiline met all three endpoints. But, Dr. Olanow said, it is possible the positive result of the low-dose group is a false-positive, rather than a false-negative in the high-dose group.
If the effect is real, the mechanism is unknown, Dr. Olanow noted — it may be that rather than protecting dopamine neurons, early treatment is preserving some beneficial compensatory mechanism the brain already has in place for making the best use of remaining neurons.
Whatever the mechanism, Dr. Olanow thinks the data show that early treatment with rasagiline “provides a benefit that can't be achieved by later introduction,” which he suggested might be a new indication for the drug. “Other physicians need to make up their own mind,” he said, “but I do use rasagiline —the data support it. I would treat patients as soon as they diagnosed.”
“But the bottom line is that we are really looking to get a reduction in cumulative disability. The goal needs to be an effect in the longer term — five to ten years. But such studies may be too difficult for industry to support,” he said.
“I think treatment probably is changing the disease,” said Caroline Tanner, MD, director of Clinical Research at the Parkinson's Institute in Sunnyvale, CA, “but whether that is through stopping some primary etiologic mechanism, or by giving the dopamine system a bit of a ‘breather,’ I don't think we can determine in these short-term observations.”
Lisa Shulman, MD, professor of neurology at the University of Maryland School of Medicine, said: “There could be a host of reasons why earlier treatment could be beneficial,” including the effects of prolonging mobility. “The fact that you start medication nine months later doesn't mean you leap back in time as though those nine months never occurred,” even if treatment had no effect on dopamine neurons at all. Depression, anxiety, social engagement — all may be affected by symptomatic therapy, whatever the agent, and all may have lasting effects.
“I tell all my patients that there may be a negative effect in delaying start of any therapy, because they themselves may not be aware of how much ground they are losing,” Dr. Shulman said, but her research has suggested that the difference of 1.7 UPDRS points between the groups is “not clinically meaningful,” and is “generally too small for patients to perceive.”
One beneficiary of the rasagiline trial is likely to be selegiline, another MAO-B inhibitor approved for PD. The drugs are “strikingly similar,” Dr. Tanner said, and selegiline is available as a generic. In early disease, putting patients on one or the other drug “is not a bad thing to do,” she said, but there are other early treatment options, including enrolling in clinical trials of other potentially disease-modifying agents.