Nilotinib Effects in Parkinson’s disease and Dementia with Lewy bodies

Study design

Twelve PDD or DLB participants with parkinsonian Hoehn and Yahr stage 3–5 were randomized into 150 mg (n = 5) or 300 mg (n = 7) groups and received oral daily doses of Nilotinib for 24 weeks with 12 weeks follow up (Fig. 1). Diagnosis of PD was according to the UK Brain Bank Diagnostic Criteria. Diagnostic criteria for probable DLB included the presence of dementia and at least two of three other symptoms, including fluctuating attention and concentration, visual hallucinations and parkinsonian motor signs. Participants were stable on L-Dopa treatment and other PD medications for four weeks prior to study enrollment as indicated in Table 1. Participants on stable treatment (>4 weeks) with monoamine oxidase (MAO)-B inhibitor Selegeline up to 10 mg/day or Rasagiline up to 1 mg/day were enrolled. Montreal Cognitive Assessment (MoCA) in the range of 18–26 was used at screening to determine cognitive impairment (MCI), moderate dementia (17–10) and severe dementia (<10) in PD and DLB patients, respectively. This is the first clinical study to examine the effects of the cancer drug Nilotinib in the PD population. We selected PDD or DLB patients to evaluate safety and tolerability in this population and excluded co-morbidities including myocardial, hematological and hepatic abnormalities, resulting in a high number of screen failures. PDD or DLB participants who were enrolled also had cognitive symptoms, and this was not an exclusion criterion as our preclinical data suggest that Nilotinib may improve cognitive performance in animal models.

The primary objective was to determine the safety and tolerability of daily Nilotinib for 6 months. Electrocardiograms (EKGs), physical and neurological exams and blood draws for laboratory chemistry were performed every two weeks for the first two months, and every month thereafter. Patients with cardiovascular disease, myocardial infarcts, QTc interval >450 ms and those who are on drugs that prolong QTc or taking exclusionary medications (i.e. coumadin) were not enrolled. Thirty patients were screened and twelve were enrolled and most screen failures were subjects who had QTc>450 ms. The secondary objectives were to determine whether Nilotinib crosses the BBB and whether it affects phosphorylation of Abl in CSF samples. Nilotinib concentration was measured in plasma and CSF at baseline (before treatment) at two months (interim) and six months treatment. Pharmacodynamic analyses were performed to determine target engagement via measurement of CSF phosphorylated Abl. Other exploratory markers associated with neurodegeneration, including CSF α-synuclein, DJ-1, DA metabolite HVA, total tau and p-tau181, Aβ42 and Aβ40, Neuron Specific Enolase (NSE) and glial and neuronal marker S100B were also measured. Tertiary/exploratory clinical outcomes included measurement of motor and non-motor symptoms using Unified Parkinson’s Disease Rating Scale (UPDRS) I-IV. Nilotinib effects on progression of dementia were evaluated using the Mini Mental State Examination (MMSE) and Scales for Outcomes in Parkinson’s disease-Cognition (SCOPA-Cog). The Columbia Suicide Severity Rating Scale (CSSRS) was performed at every visit.

Screening and enrollment

Participants and study partners were consented prior to any study-related procedures. Demographics, concurrent medications, and medical history were gathered from the participant and study partner (Table 1). Patients who could not give consent were consented by a legally authorized representative. Vital signs were recorded with comprehensive physical and neurological examinations. Randomization was performed via an internet based randomization module and subjects were assigned to one of the 2 treatment groups. The chance for randomization to the groups was 1:1 for 150 mg:300 mg Nilotinib.

Safety monitoring

Physical and neurological exams and EKGs were performed with laboratory blood chemistry, including total cholesterol, low density lipoprotein (LDL), high density lipoprotein (HDL), triglyceride, phosphate, total bilirubin, calcium, chloride, sodium, potassium, magnesium, lipase, α-amylase, alanine aminotransferase (ALT) and aspartate aminotransferase (AST). A standard hematology panel included complete blood count with differential for white blood cells (WBC), hemoglobin, hematocrit and platelets.

Baseline

This visit was scheduled 2–4 weeks after screening and results from all screening procedures were reviewed for inclusion/exclusion criteria. CSF was collected via lumbar puncture and blood draws were performed for biomarker measurements. Blood biomarker samples were collected 30 minutes before CSF collection. Vital signs, brief physical and neurological examination along with documentation of concurrent medications and reviewed AEs were performed at every visit.

Plasma and CSF collection

Blood draw (15 ml) and lumbar puncture (∼15 ml CSF) were performed on all patients at baseline, 2 and 6-months treatment. Collection was obtained at 1, 2, 3, 4 or 5 hrs after Nilotinib administration by mouth. Plasma was isolated immediately after blood draw and aliquoted and stored at –80°C. CSF was aliquoted and stored at –80°C. Freeze/thaw cycles were avoided. To avoid CSF contamination with blood, the first 1 mL of CSF collection was discarded and all samples were centrifuged at 1000 g for 15 minutes. Samples that contained >25 ng/ml Hemoglobin were eliminated and were not tested for biomarkers. The hemoglobin levels in CSF samples were measured using human hemoglobin ELISA Quantitation (Cat # E80-136) Kit (Bethyl Lab Inc, Montgomery, TX, USA) according to the manufacturer’s instructions.

Mass spectrometry to evaluate Nilotinib pharmacokinetics

Plasma and CSF samples (20 μl) were thawed initially on ice at room temperature and transferred to Eppendorf tubes containing 100 μl of water. Extraction solvent (500 μl) Acetonitrile/Methanol (50:50) containing the internal standard (5 ng/mL of Nilotinib_13C_2H3) was added to the sample. The mixture was vortexed and incubated for 20 min on ice to accelerate protein precipitation and dialysis through 25 μm membranes was performed to obtain unbound Nilotinib. After incubation, the samples were vortexed and centrifuged at 13,000 rpm for 20 min at 4°C. The supernatant containing unbound Nilotinib was freeze-dried using speed vacuum and reconstituted in 200 μL of Methanol: Water (50:50) and processed by mass spectrometry.

The samples were resolved on an Acquity UPLC BEH C18 1.7 μm, 2.1×50 mm column online with a triple quadrupole mass spectrometer (Xevo-TQ-S, Waters Corporation, USA) operating in the multiple reaction monitoring (MRM) mode. The instrument parameters were optimized to gain maximum specificity and sensitivity of ionization for the parent [m/z = 530.27 (Nilotinib), 438.25 and daughter ions [m/z = 289.01 (Nilotinib) using “IntelliStart” feature of MassLynx software (Waters Corporation, USA). The metabolite ratios were calculated by normalizing the peak area of endogenous metabolites within tissue samples normalized to the internal standard Nilotinib_13C_2H3.

Phospho-Abl (pan-tyrosine) enzyme-linked immunosorbent assay (ELISA)

PathScan^® pan-tyrosine phospho-Abl solid phase sandwich ELISA was performed on the same CSF and plasma samples that were used for Nilotinib measurement in CSF. Abl rabbit antibody was coated on the microwells and 100 μl CSF or plasma was incubated for 2 hrs at 37°C. Bcr-Abl and Abl protein (phospho and nonphospho) were captured by the coated antibody. All samples at baseline, 2 months and 6 months were analyzed side-by-side using same reagents. After washing, a pan-tyrosine phosphorylated Abl detection antibody was added to each well to detect captured tyrosine-phosphorylated Bcr-Abl and Abl protein. Samples were incubated with detection antibody for 1 hr at 37°C. After washing, anti-mouse IgG, horseradish peroxidase (HRP)-linked antibody was added and incubated for 10 min at 37°C to recognize the bound detection antibody. The magnitude of the absorbance for developed color is proportional to the quantity of tyrosine-phosphorylated Bcr-Abl and Abl protein in the samples.

α-synuclein measurement

Solid phase α-synuclein sandwich ELISA (Cat#SIG38974, Biolegend) was performed on CSF and plasma. To avoid freeze-thaw cycles, immediately after LP and blood draws, 15 mL CSF and 5 mL plasma were aliquoted on ice into 0.5 mL tubes and stored at –80°C. Fresh aliquots were used to perform ELISA. All samples at baseline, 2 months and 6 months were analyzed side-by-side using same reagents. Total α-synuclein rabbit monoclonal antibody (amino acids 118–123) was coated on the microwells and 200 μl CSF or plasma was added to designated wells. CSF samples were diluted 1:10 while plasma samples were diluted 1:50. After overnight sample incubation at 4°C, α-synuclein was captured by the coated antibody. After washing, a biotinylated mouse monoclonal α-synuclein (amino acids 103–107) detection antibody was added to each well to detect the captured α-synuclein (amino acids 118–123). Samples were incubated with 50 μl of detection antibody for 2 hrs at room temperature. After washing, 200 μl of streptavidin HRP was added and incubated for 1 hr at room temperature to recognize the bound biotinylated detection antibody. Samples were then washed and incubated with 100 μl of chemiluminescent substrates. Plates were shaken for 10–15 seconds and read immediately by a luminometer.

HVA measurement

A 100 μl CSF and plasma samples were incubated with 100 μl HRP-conjugate reagent and incubated for 1 hr at 37°C using solid phase sandwich ELISA (MyBioSource, Cat# MBS064661). All samples at baseline, 2 months and 6 months were analyzed side-by-side using same reagents. After washing, 50 μl of chromogen solution A and 50 μl of chromogen solution B were added to the solution and incubated for 15 min at 37°C. The reaction was stopped with 50 μl stop solution and the optical density was read at 450 nM. The magnitude of the absorbance is proportional to the quantity of CSF HVA.

Human neurodegenerative disease magnetic bead panels

A multiplex Xmap technology that uses magnetic microspheres internally coded with two fluorescent dyes was used to measure markers of neurodegeneration. All samples at baseline, 2 months and 6 months were analyzed side-by-side using same reagents. Through precise combinations of these two dyes, multiple proteins are measured within the sample. Each of these spheres is coated with a specific capture antibody. The capture antibody binds to the detection antibody and a reporter molecule, completing the reaction on the surface of the bead. 25 μL of CSF or plasma was incubated overnight at 4°C with 25 μL of a mixed bead solution, containing human NSE and DJ-1 (Millipore, Cat#: HND1MAG-39K) or S100B, total tau, p-tau181, Aβ42, and Aβ40 (CSF Aβ40 is diluted 1:10). After washing, samples were incubated with 25 μL detection antibody solution for 1.5 hrs at room temperature (Millipore, CAT#: HND4MAG-36K), 25 μL of Streptavidin-Phycoerythrin was added to each well containing the 25 μl of detection antibody solution. Samples were then washed and suspended in 100 μl of sheath fluid. Samples were then run on MAGPIX with Xponent software. The Median Fluorescent Intensity (MFI) data was analyzed using a 5- parameter logistic or spline curve-fitting method for calculating analyte concentrations in samples.

Statistical data analysis

Data are summarized as raw values, range as appropriate, and mean±SD for n = 5 in 150 mg group and n = 7 in 300 mg group, unless otherwise indicated. All graphs and statistical analyses were performed in Graph Pad Prism Software version 5.01 (Graph Pad Prism Software, Inc. CA. USA). To account for large differences between groups at baseline, all estimates used for means and SD were based on mean of relative changes- not the mean value changes. As each group has less than 10 subjects and the normality of biomarkers is unclear, non-parametric tests such as Wilcoxon signed test and Mann-Whitney test were performed to compare between baseline, two and six months treatment with Nilotinib for both 150 mg and 300 mg treatment groups for biomarkers. As non-parametric tests may be underpowered, a two tailed paired t-test was also used to examine differences between baseline and six months treatment with Nilotinib for both 150 mg and 300 mg treatment groups for biomarkers and clinical outcomes. The significance level was adjusted using the Bonferroni correction. The sample size n and p-values from each test are indicated in each table or figure. No comparison is made in this pilot study between 150 mg and 300 mg due to large baseline differences and randomization of most DLB patients into the 150 mg group and PDD patients into 300 mg group.

Study approvals

The protocol was approved by the Institutional Review Board (IRB #2014-0006) at Georgetown University Medical Center as well as by Georgetown-Howard Universities Center for Clinical and Translational Science (GHUCCTS) scientific review board. This study was performed in consultation with the FDA. The study was listed on clinicaltrials.gov (NCT02281474). All participants were recruited from the movement disorders clinic at MedStar Georgetown University Hospital and were consented at the GHUCCTS Clinical Research Unit (CRU). Eligible participants were computer randomized to either 150 mg or 300 mg groups and enrolled in the study. Patients were de-identified and a Pad-Id was included (e.g. NIL) in the patient’s clinical chart. An independent safety and monitoring board (DSMB) that included a cardiologist, hematologist, clinical pharmacologist, neurologist and biostatistician monitored safety signals in patients in the study.

Screening, enrollment and retention

Thirty participants were screened and twelve PDD or DLB participants (Table 1) were randomized into 150 mg (n = 5) or 300 mg (n = 7) groups and received oral daily doses of Nilotinib for 24 weeks with an additional 12 weeks follow up (Fig. 1). Eighteen patients failed screening due to cardiovascular disease, infarcts, QTc interval >450 ms or excluded medications (i.e. Coumadin). One patient (NIL 9) was withdrawn from the study due to serious adverse events (SAEs) and another (NIL 8) discontinued at 5 months due to increased caregiver burden. Ten patients completed the 24-week study and returned for a follow up visit (week 36). One patient (NIL 12) did not follow up at week 36.

Serious Adverse Events (SAEs)

Adverse Events (AEs) that required or resulted in hospitalization were classified as an SAE and were listed in Supplementary Table 1. One participant in the 300 mg Nilotinib group (NIL 9) was withdrawn at week 4 and diagnosed with myocardial infarct and left bundle block. One participant in the 300 mg Nilotinib group was hospitalized with urinary tract infection (UTI) and one participant in the 150 mg group was hospitalized with pneumonia and UTI (Supplementary Table 1).

Non-serious AEs

Non-serious AEs, which did not require hospitalization were classified per system and/or preferred term and summarized in Supplementary Table 2 as number of events and number of affected participants in each group. One participant in the 150 mg group and one in the 300 mg Nilotinib group had transient QTc prolongation >450 ms but returned to a normal QTc range in subsequent visits (detailed in Supplementary Table 3). Participants in the 150 mg group reported three UTIs, two cases of pneumonia, one cold virus, one mild back pain, one mild headache, one mild dysgraphia, one mild left foot drag, one mild confusion, one mild hallucination, one mild paranoia, one mild agitation, one moderate anxiety, one mild incontinence, one moderate itching and one skin irritation. Participants in the 300 mg group reported one incident of blurry vision, one diarrhea, one nausea, one mild fatigue, two generalized weakness, three UTIs, one pneumonia, one weight loss, one tooth extraction, one dizziness, two mild hallucinations, one mild paranoia, one mild crying episode, one mild urinary urgency, one mild cough and one eczematous lesion.

QTc prolongation

One participant (NIL 9) in the 300 mg was withdrawn due to moderate cardiac ischemia and myocardial infarct and was double-stented for a left bundle block (Supplementary Table 3) and this was classified as an SAE in Table 2. NIL 8 had slight QTc prolongation >450 ms at week 8, and NIL 15 had two episodes of QTc prolongation >450 ms at week 2 and week 8 but QTc returned to normal range in subsequent visits. NIL 8 withdrew from the study at week 24 due to increased caregiver burden, but returned for follow up (week 36) and NIL 12 did not return for follow up at week 36.

Blood chemistry

Laboratory tests (Supplementary Table 4) were performed throughout eleven visits according to protocol. Laboratory results of some participants were outside the normal range at baseline (less than 100% (<1 fold) difference from baseline) due to comorbidities that were not exclusionary. A total of three AEs below normal range of alanine aminotransferase (ALT) level (<100%) were observed in the 150 mg Nilotinib group. Two AEs above normal range (>100%), including one ALT and one bilirubin elevation were observed in the 150 mg Nilotinib group. Two AEs with transient increase of ALT and one AST above normal range (>100%) were observed in the 300 mg Nilotinib group but returned to normal by end of the study.

Nilotinib pharmacokinetics

Blood and CSF were collected from twelve participants at baseline prior to treatment (n = 12, control) and at 2 months treatment (n = 12). Blood and CSF were collected from nine patients at six months (n = 9) of Nilotinib treatment (total n = 33 samples) and collection time via lumbar puncture (LP) was staggered at 0, 1, 2, 3, 4 or 5 hrs after oral administration of Nilotinib. Blood was collected 30 min prior to LP. Two CSF samples were not tested because they had >25 ng/ml hemoglobin. Pharmacokinetic analysis demonstrated that Nilotinib (Fig. 2A and E) peaked in the plasma 2 hrs (T_max) after dosing at 55 nM and 86 nM (C_max) in the 150 mg and 300 mg Nilotinib groups, respectively. Nilotinib (Fig. 2B and E) also peaked in the CSF 2 hrs (T_max) after dosing at 5 nM and 11 nM (C_max) in the 150 mg and 300 mg groups, respectively. Area under the curve (AUC) was 157 and 232 ng/ml*h for plasma and 8 and 14 ng/ml*h for CSF in 150 mg and 300 mg groups, respectively. Nilotinib was quantified in reference to ¹³C Nilotinib standard and samples were precipitated and dialyzed to obtain unbound Nilotinib. Nilotinib was not detected in the CSF (t_1/2 ∼1.8) 3 hrs after dosing suggesting short bioavailability of unbound or free Nilotinib compared to plasma t_1/2 = 12.4 (Fig. 2E). The CSF:plasma ratio of Nilotinib (Fig. 2C) is 12% and 5% with 300 mg and 150 mg Nilotinib, respectively.

Target engagement

To demonstrate whether Nilotinib inhibits Abl via de-phosphorylation, we measured pan-tyrosine CSF Abl relative to total Abl via ELISA. Phosphorylated Abl was reduced (30%) with 150 mg and 300 mg Nilotinib (Fig. 2D), suggesting that Nilotinib may inhibit CNS Abl phosphorylation. The absorbance level of Abl phosphorylation in the CSF was maximal (100%) at baseline prior to Nilotinib administration (Fig. 2F and G). However, pan-tyrosine Abl absorbance levels were reduced in all patients in the 150 mg group (Fig. 2F) and 300 mg groups(Fig. 2G).

Unified Parkinson Disease Rating Scale (UPDRS)

Motor symptoms were monitored with UPDRS at all clinic visits (Table 2). The UPDRS scores were obtained at 1–4 hrs after oral dosing with Nilotinib in each participant. An average decrease of 3.4 points and 3.6 points in UPDRS-III (motor) was observed at six months (week 24) compared to baseline with 150 mg and 300 mg Nilotinib, respectively. These effects were reversed at the 36-week follow up visit in all participants. An average decrease of 7 points and 11.1 points in UPRDS I-IV was also observed at 6 months in the 150 mg and 300 mg groups, respectively. The change in UPDRS was progressive throughout the 6 months (24-week) trial, and was reversed in participants in the follow up visit at 36 weeks. Most participants experienced increased psychotic symptoms (hallucination, paranoia, agitation) and some dyskinesia whilst on Nilotinib suggesting an increase in CNS DA levels, which required lowering or discontinuation of monoamine oxidase (MAO)-B inhibitor, Azilect or Rasagiline(Table 1).

Mini Mental Status Examination (MMSE)

Cognitive performance was monitored with MMSE at all clinic visits (Table 3). An average increase of 3.85 and 3.5 points in MMSE was observed at six months (24-week) compared to baseline with 150 mg and 300 mg Nilotinib, respectively. The increase in MMSE scores was incremental during the 24-week treatment, and returned to baseline at the 36-week follow up visit. The Scales for Outcomes in Parkinson’s Disease-Cognition (SCOPA-Cog) test also show an average increase of 1.85 and 2 points at six months compared to baseline with 150 mg and 300 mg Nilotinib, respectively.

PD-related biomarkers

Because of the small sample size (n < 10), normal distribution is uncertain, thus a non-parametric Wilcoxon signed and Mann-Whitney tests were performed to compare biomarker changes between baseline and two and six months treatment with 150 mg and 300 mg Nilotinib. However, as non-parametric tests may be underpowered, a two tailed paired t-test was also used to examine differences between groups at baseline and after Nilotinib treatment. CSF homovanillic acid (HVA), a DA metabolite, was significantly increased (Supplementary Table 5) in the 150 mg group at 2 months (t-test) and remained elevated (non-significant) at 6 months compared to baseline. CSF HVA also significantly increased in the 300 mg group at 6 months compared to baseline (Supplementary Table 5). Abl activation as well as DJ-1 are up-regulated in response to oxidative stress [25–27] but CSF DJ-1 trended towards a decrease in the 150 mg group, and was significantly reduced in the 300 mg group at six months compared to baseline and at 2 months compared to 6 months (Supplementary Table 5).

CSF α-synuclein (Supplementary Table 5) was slightly reduced at 2 and 6 months compared to baseline in the 150 mg Nilotinib group, but the level of α-synuclein was significantly (t-test) lower at 2 months compared to 6 months. However, α-synuclein (Supplementary Table 5) was unaltered in the 300 mg Nilotinib group between baseline, 2 and 6 months. Data from the Parkinson’s Progression Markers Initiative (PPMI) suggest that CSF α-synuclein does not change over six months in de novo or mid-stage PD. However, the slight decrease in CSF α-synuclein in the 150 mg Nilotinib group over six months in the current PDD and DLB cohort suggests that a placebo group is needed to interpret the data definitively. Plasma α-synuclein was slightly reduced at 2 months compared to baseline in the 150 mg group (Supplementary Figure 1A) but at 6 months it became significantly higher compared to 2 months. Interestingly, plasma levels of α-synuclein in some subjects were elevated 3 months after termination of treatment. Similarly, plasma α-synuclein was significantly reduced at 2 months compared to baseline in the 300 mg group (Supplementary Figure 1B) but returned to baseline at 6 months. Plasma levels of α-synuclein in some subjects were elevated 3 months after termination of treatment. Taken together these data suggest that Nilotinib may stabilize α-synuclein levels in PD, but a larger sample with a placebo group is needed to interpret theseresults.

Surrogate biomarkers

CSF S100B is a marker of glial death in the brain [28, 29] and neuron specific enolase (NSE) is elevated after stroke and brain trauma [30, 31], suggesting cell death. The level of S100B was gradually reduced in the 150 mg and 300 mg groups. However, NSE level was significantly reduced at 6 months compared to baseline in the CSF in both 150 mg and 300 mg groups (Supplementary Table 5). CSF total tau and p-tau are increased in PD and DLB compared to control [32–34]. Total tau and the ratio of total tau/p-tau 181 were significantly reduced at 2 months compared to baseline in the 150 mg group, while total tau was significantly reduced at 6 months compared to baseline in the 300 mg group. No changes in CSF Aβ40 or Aβ42 were detected in 150 mg group, but CSF Aβ40 was significantly reduced at 6 months compared to baseline in the 300 mg Nilotinib group.