Assistant Professor of Neuroscience and Neurology, Feil Family Brain and Mind Research Institute, Department of Neurology, Weill Cornell Medical College

Dr. Ishii reports no financial relationships relevant to this field of study.

SYNOPSIS: In this multicenter Phase IIa clinical trial in mild Alzheimer’s disease dementia, the tyrosine kinase Fyn inhibitor AZD0540 had no significant effects after 52 weeks of treatment.

SOURCE: van Dyck CH, Nygaard HB, Chen K, et al. Effect of AZD0530 on cerebral metabolic decline in Alzheimer disease: A randomized clinical trial. JAMA Neurol published online July 22, 2019.

In Alzheimer’s disease (AD), the abnormal accumulation of amyloid-beta peptides is believed to be the major pathogenic event leading to synaptic dysfunction, pathological tau accumulation, cognitive impairment, and neurodegeneration. Recent intervention trials in AD have focused on developing therapies to limit the production of amyloid-beta by inhibiting secretase enzymes or by promoting clearance of amyloid-beta through immunotherapy with antibodies against amyloid-beta. Despite several promising drug candidates, all clinical trials specifically targeting amyloid-beta have failed. An alternative approach would be to target downstream effects of amyloid-beta pathology. Although several candidates are being investigated, one promising target is the tyrosine kinase Fyn. Amyloid-beta binds to prion protein on the cell surface of neurons to activate intracellular Fyn kinase, which subsequently mediates synaptotoxicity and tau pathology. In a transgenic mouse model of amyloid-beta pathology, inhibition of Fyn kinase by AZD0530 (saracatinib) results in sustained rescue of memory function. A previous Phase Ib clinical study of AZD0530 demonstrated the safety, tolerability, and central nervous system availability of orally administered AZD0530.

In this multicenter, Phase IIa, placebo-controlled, randomized clinical trial, participants age 55 to 85 years with mild AD and evidence of amyloid pathology by ¹⁸F-florbetapir positron emission tomography (PET) scan were enrolled to evaluate the safety, tolerability, and the effects of AZD0530 after 52 weeks of treatment. The primary outcome measure was ¹⁸F-fluorodeoxyglucose (FDG) PET measurement of the reduction in relative regional cerebral metabolic rate for glucose (CMRgI). Standard clinical assessments including Alzheimer’s Disease Assessment Scale-Cognitive Subscale, Mini-Mental Status Examination (MMSE), Clinical Dementia Rating-Sum of Boxes (CDR-SB), Neuropsychiatric Inventory, and magnetic resonance image (MRI) scans for volumetric analysis were also collected. A subset of participants had cerebrospinal fluid (CSF) collected at baseline and 52 weeks to assess the effects of AZD0530 on CSF tau levels and to measure CSF AZD0530 levels.

A total of 293 participants were screened, and 159 were randomized with 79 to the AZD0530 group and 80 to the placebo. Although the study drug was relatively well tolerated, there were significant number of adverse events, which most frequently were gastrointestinal in nature (48.1% in AZD0530 and 28.7% in placebo). A total of 128 (80.5%) subjects completed the study with early discontinuations owing primarily to these adverse events (26.5% in AZD0530 and 13.8% in placebo group). AZD0530 treatment had no significant effect on ¹⁸F-FDG PET-measured reduction in relative CMRgl at 52 weeks. Additionally, there were no differences in all clinical assessments measured. There was a trend toward slowing the decline in hippocampal volume and entorhinal thickness in the treatment group, but these and the other predetermined volumetric measures by MRI did not reach statistical significance. Although the number of participants who had CSF collected was small (34 participants at both baseline and 52 weeks), there were no significant treatment differences for rates of changes in either CSF total tau or phosphorylated tau.

COMMENTARY

The overall results from this trial were disappointing, as there were no statistically significant effects of AZD0530 treatment on any of the primary or secondary outcome measures. However, there are important limitations that may have contributed to study failure. First, there was a larger number of participants that discontinued the trial than expected, leading to a significant reduction in the statistical power of the study and the inability to detect all but large effect sizes. Second, in the limited number (13) of study participants receiving active treatment who had a week-52 lumbar puncture, the CSF AZD0530 level was slightly below the predetermined efficacy threshold. The dose of AZD0530 could be increased in future trials to ensure maximum efficacy, but this may be difficult due to the high number of participants with adverse effects from the drug at the present dose.

The results from this trial highlight the challenges in developing an AD therapy based on inhibition of Fyn kinase. Since Fyn kinase has important physiological functions outside of the central nervous system, off target effects due to inhibition of Fyn kinase may be difficult to avoid. Additionally, none of the Fyn kinase inhibitors are completely selective and will inhibit other members of the Src family of non-receptor tyrosine kinases, which will further increase the likelihood of off target effects. Importantly, inhibiting Fyn kinase may be too late even in mild AD dementia, where there is already significant neurodegeneration. Although Fyn kinase is downstream of amyloid-beta, it is still an early event in the pathogenesis of AD and may need to be targeted at earlier stages of AD such as mild cognitive impairment, where the effects of Fyn kinase may still be reversible. Alternatively, the downstream effects of Fyn kinase activation could be targeted.

Despite the disappointing results from this trial and the challenges in developing Fyn kinase inhibitors as a disease modifying therapy for AD, there were significant advances made in this trial. Notably, the study investigators provided strong evidence for the use of ¹⁸F-FDG PET measurement of CMRgI as an outcome measure that was well correlated with traditional clinical measures but with greater precision. Therefore, the use of CMRgl should be considered in future AD intervention trials. Finally, with the increasing number of drug trial failures directly targeting amyloid-beta, especially once dementia develops, this study highlights a need to continue to investigate additional downstream targets of amyloid-beta pathology to develop an effective disease modifying therapy for symptomatic AD.