7 FDA Approved CSF Biomarkers Demonstrate Strong Predictive Value for Identification of Alzheimer Disease Based on PET Scan in Patients with Standalone Alzheimer Cognitive Screening Tests.

Abstract Background Alzheimer’s disease, a progressive neurological disease, affects an estimated 6.7 million Americans aged 65 and older. Although positron emission tomography (PET) scan is the emerging gold standard for diagnosis of Alzheimer’s disease, Elecsys amyloid beta (Aβ42), total tau (t-Tau) and phosphorylated tau (p-Tau at position 181) in cerebral spinal fluid (CSF) have recently been approved by the FDA as surrogate biomarkers for Alzheimer’s disease. We implemented these Elecsys CSF assays in our laboratory using Cobas e801 analyzer and established a cut-off of > 0.024 for pTau181/Ab 42 ratio for indication of Alzheimer disease. We studied correlation between pTau181/Ab 42 ratio and actual diagnosis of Alzheimer’s ­disease based on PET scan in 143 patients. Materials and Methods All Elecsys assays were obtained from Roche and CSF specimens were collected using specialized Sarstedt tube supplied by Roche. Imaging studies were performed using fluorodeoxyglucose PET scans. Screening for cognitive function (SCF) in these patients was performed using either MOCA (Montreal Cognitive Assessment), SLUMS (Saint Louis University Mental Status Examination), STMS (The Short Test of Mental Status) or MMSE (Mini-Mental State Exam). Results There was no statistically significant difference in age of patients with the CSF biomarker ratio above the cut off limit of 0.024 (80 patients, mean age 71.3 years, SD: 7.9) and ratio below the cut-off (63 patients, mean age 66.3 years, SD :10.0) using Levene’s test. However, older patients were more likely to show an elevated pTau181/Ab 42 ratio. We observed no racial bias (Caucasian, African American, Hispanic and others) in elevated CSF biomarker results or positive PET scans using Chi-Square test. Chi-square analysis also showed a statistically significant association between positive CSF tests and PET scans (p < 0.01). The mean pTau181/Ab42 ratio was 0.055 (SD: 0.05, range: 0.025-0.12) in patients resulted positive for the CSF biomarkers, but the ratio was 0.0138 (SD: 0.0046, range: 0.007-0.024) in patients who tested negative (differences were also statistically significant by t-test). The sensitivity of pTau181/Ab42 was 76.2%, specificity was 84.2% and positive predictive value was 84.2% when compared to PET scan findings. We observed poor correlation between standalone SCF and PET scan result using Chi-Square analysis. For example, MOCA showed a sensitivity of 50.0%, specificity of 66.7% and positive predictive value of only 33.3%. However, binary logistic regression demonstrated a combination of abnormal SCF, and a positive CSF screen significantly predicted a positive imaging result (N = 40, p<.001). Conclusions CSF pTau181/Ab42 ratio is a diagnostic marker for Alzheimer disease correlated well with PET scan compared to various standalone cognitive screening assessment. A negative CSF biomarker test result may help rule out Alzheimer disease. Further, a combination approach inclusive of SCF and CSF biomarkers was significant for positive PET imaging.