Article In Brief
Researchers have expressed concern about nilotinib's effect on safety in Parkinson's disease, despite a more recent study suggesting its promise for treating certain symptoms.
Two phase 2 studies—one published in JAMA Neurology and the other in press— recently reached completely different conclusions about the potential efficacy of nilotinib, a tyrosine kinase inhibitor that demonstrated some early promise in treating Parkinson's disease (PD) symptoms.
Approved in 2007 for a specific type of chronic myelogenous leukemia, nilotinib (Tasigna) inhibits the activity of non-receptor tyrosine kinase Abelson (c-Abl), a protein involved in cell growth and morphology in blood cancers. Some research has also linked c-Abl to sporadic and familial PD, where it is involved in regulating aggregation of alpha-synuclein protein and deactivation of Parkin protein.
Preliminary data from a small 2016 open-label phase 1 trial of nilotinib reported some improved motor symptoms in advanced PD patients treated safely with nilotinib; however, it was not powered for efficacy.
Because of its ability to inhibit formation of toxic proteins involved in neurodegeneration, it has been of interest to both PD and Alzheimer's disease investigators in recent years. Nilotinib has also been associated with adverse events (AEs) and currently carries a black box warning of increased cardiovascular risks and other AEs in leukemia patients.
Several cell and animal models of neurodegeneration studies suggest that nilotinib may reduce alpha-synuclein pathology in PD, and a small open label study tested the safety/tolerability of nilotinib in PD patients for the first time and reported in 2016 that there were positive, though exploratory, signs of efficacy.
There has also been preclinical evidence that the drug can cross the blood-brain barrier and degrade alpha-synuclein and tau in the brains of some animal models of neurodegeneration. Additional studies suggested that it might improve clinical outcomes of PD by bolstering dopamine metabolism.
JAMA Neurology Study
In the first study, published online late last year and in print in March in JAMA Neurology, researchers at Georgetown University conducted a randomized trial in 75 moderately severe PD patients who received either a daily dose of 150 mg or 300 mg nilotinib, or placebo, for 12 months. They reported that treatment was safe and well-tolerated, and biomarkers indicated an increase in dopamine metabolites in plasma and cerebrospinal fluid (CSF). There was also a significant decrease in CSF levels of hyper-phosphorylated tau and a significant decrease in oligomeric alpha-synuclein (but only with the 150 mg dose).
After 12 months, CSF homovanillic acid levels were 40 percent higher in patients given150 mg nilotinib, and 22 percent higher in the 300 mg group, than in the placebo group. CSF 3,4-dihydroxyphenylacetic acid levels were 85 percent higher in the nilotinib 150-mg group and 131 percent higher in the 300-mg group than in the placebo group.
“These data collectively suggest that the 150 mg–300 mg dosage of nilotinib is a safe and acceptable range to be used in a larger phase 3 study,” lead author Charbel Moussa, PhD, associate professor of neurology and director of the Laboratory for Dementia and Parkinsonism told Neurology Today.
However, other researchers have expressed concern about the drug and believe proceeding to a phase 3 trial is premature, and that further research is needed before taking that step.
Phase 2 NILO-PD Study
In mid-February, researchers with the Parkinson Study Group presented early results from another phase 2 NILO-PD trial at the third Pan American Parkinson's Disease and Movement Disorders Congress in Miami.
The results, which are in press for publication, show little evidence of nilotinib altering CSF dopamine or its metabolites, at either the 150 mg or 300 mg dose, in moderate PD patients at six months.
Principal researcher Tanya Simuni, MD, professor of neurology and head of the movement disorders division at Northwestern University Feinberg School of Medicine, told Neurology Today that while the researchers found nilotinib to be safe and tolerable, it demonstrated no clinically or biologically meaningful beneficial effect. She also said that because of the study's strict exclusion criteria, 40 percent of PD patients were considered ineligible to participate in the trial, compared with 25 percent in the Georgetown study.
“We were not powered to assess efficacy, but we found nilotinib had low CSF exposure, indicating poor brain penetration, and it failed to change dopamine or its metabolite levels. Although our findings do not refute some potential neuroprotection with other molecules targeting the c-Abl pathway, it is not an optimal molecule for assessing the therapeutic potential of c-Abl inhibition in PD,” she said
“Unfortunately, the results we observed did not support testing nilotinib in a larger study. This is not the outcome we hoped for, but we remain dedicated to pursuing better treatments by advancing other potential therapies in today's robust Parkinson's pipeline.”
Dr. Simuni told Neurology Today that anyone considering adding nilotinib to their PD treatment regimen should be made aware of the lack of reliable efficacy data on nilotinib. While nilotinib was seen to be safe and tolerable in the study population, the study also had strict inclusion criteria, excluding anyone with cardiac issues or other health challenges. Therefore, its safety in the wider population of PD patients remains unknown.
“The conclusion that nilotinib is reasonably safe holds poorly to scrutiny,” according to an editorial that accompanied the JAMA Neurology paper by Alberto J. Espay, MD, FAAN, neurology professor and director and chair of the University of Cincinnati's Gardner PD center, and colleagues at two other university centers.
“Selected safety endpoints underscore a dose-related increase in the incidence of serious adverse events in the study's end points: 4 AEs in 3 patients (17 percent) receiving placebo; 6 in 5 patients (27 percent) in the 150 mg nilotinib cohort, and 12 in 9 patients, or 57 percent, in the 300 mg group,” they wrote.
“Overall, SAEs [serious adverse effects] were significantly more common with nilotinib, 300 mg, compared with placebo and there was a nonsignificant but dose-dependent increased incidence in falls in the treated groups (placebo, 39 percent; nilotinib, 150mg, 54 percent; nilotinib, 300mg, 61 percent).”
They also noted that the study was underpowered to assess risk of sudden cardiac death, for which nilotinib has a black box warning at higher daily dosages. There were also no reported differences in motor and nonmotor secondary and exploratory outcomes between nilotinib and placebo patients.
“Of concern, the nilotinib 300 mg group experienced worsening of activities of daily living from baseline to 12 months as assessed by Movement Disorders Society–Unified Parkinson's Disease Rating Scale (MDS-UPDRS) part II scores, now favored by regulatory agencies for assessing disease modification.”
Finally, they observed that because all patients received a single, random dose of nilotinib at baseline, the reported improved levels of PD biomarkers in CSF need to be interpreted “in light of the nilotinib groups receiving higher daily dosages of levodopa”—14 percent higher in the nilotinib 150-mg group and 4.4 percent higher in the nilotinib 300-mg group—and the potential influence of variability of timing of lumbar puncture relative to levodopa administration.
Georgetown University's Dr. Moussa dismissed the reservations expressed in the editorial. “I am not concerned about the views expressed in the editorial as all AEs were not related to nilotinib, although more work is needed,” he told Neurology Today.
“I disagree with the highly idealistic, unrealistic and theoretical approach of PD patient selection as mentioned in the editorial. Clinical or symptomatic variability of idiopathic non-familial and non-genetically linked PD does not negate or eliminate the fundamental fact that the disease is associated with loss of CSF dopamine and misfolding of proteins, including alpha-synuclein oligomerization and tau hyper-phosphorylation.”
Since the study showed a positive effect on a mix of disease-relevant CSF biomarkers, he said patient selection should narrow inclusion to biomarker-confirmed idiopathic PD. “Therefore, we should be able to appropriately examine nilotinib effects on the most likely clinical manifestations that are associated with the disease stage and medication and/or treatment regimen.”
Aparna Wagle Shukla, MD, associate professor and director of clinical trials at the Fixel Institute for Neurological Diseases at the University of Florida (UF) in Gainesville, told Neurology Today that the editorial provides many legitimate arguments against consideration of a phase 3 trial.
“Although the investigators of this randomized clinical trial conclude that nilotinib is a reasonably safe drug, the editorial urges the readers to exercise caution as nearly one-fourth of the patient population assigned to the 150 mg dose-group, and one half of the 300 mg dose-group developed serious adverse events.
“While one cannot ignore the increased incidence of falls reported frequently by the higher dosage group, the study was not powered to address the potential danger of sudden cardiac death, a known black box warning for this cancer drug,” she said. “Also, the Parkinson's population enrolled for the trial was carefully screened for pre-existing cardiac disease.”
In addition, although clinical outcomes were not the primary outcomes, worsening activities of daily living in patients receiving 300 mg of nilotinib drug is discouraging, Dr. Shukla said.
“The investigators also claimed that nilotinib is capable of influencing biomarkers as reflected by the CSF levels of homovanillic acid and oligomeric α-synuclein, however, these claims may not hold water due to various reasons,” she continued. “The data comparisons were possibly driven by the outliers in each group, and comparisons of final CSF levels were not performed concerning baseline levels.”
Finally, because CSF levels did not get correlated with clinical outcomes, these biomarkers may not be appropriate “surrogate” measures for future clinical trials, she said
Dr. Shukla's colleague Adolfo Ramirez-Zamora, MD, chief of the UF Movement Disorders Division, noted that the NILO-PD study, presented at the Pan American Parkinson's Disease and Movement Disorders Congress in February, showed that the drug was overall safe and tolerable in this population but did not exert a clinically meaningful benefit or biological effect in PD patients.
The upcoming challenges include defining the right outcome measures (clinical vs laboratory), he said, and the appropriate population to consider for additional trials, particularly with medications with similar mechanism of action but increased potency.
“It is important for patients and the entire Parkinson's community to take a close look at the results of this and similar studies,” Dr. Ramirez-Zamora said. “As we aim to develop and identify potential therapies to slow down PD, it is equally as important to recognize drugs with limited or lack of benefit in well-designed clinical trials.”