Innotest

Amyloid-β (Aβ₄₂) levels were measured using ELISA (Innotest^®, Innogenetics, Ghent, Belgium). This assay was conducted at the Sahlgrenska Institute in Sweden.

CSF analytes of total tau and beta-amyloid_1-42 were obtained using previously reported procedures and evaluated with either a sandwich ELISA 2 (INNOTEST, Innogenetics, Ghent, Belgium) or a LUMINEX xMAP platform (INNO-BIA AlzBio3, Innogenetics). A ratio of total tau to beta-amyloid (t-tau:Aβ) was generated across platforms using an autopsy-validated conversion factor that has been cross-validated across two independent series (Irwin et al., 2012 (link)). Specifically, it has been demonstrated that a t-tau:Aβ ratio above threshold (>0.34) is 95.5% accurate across two autopsy series (Irwin et al., 2012 (link)). In this study cohort 11 patients had autopsy or a genetic mutation consistent with FTLD pathology and all of them were correctly classified with CSF as having FTLD pathology. Using this threshold we identified 72 patients with a CSF profile not consistent with AD, which we presume is FTLD, and 21 patients had a CSF profile consistent with AD. Our cohort that contains 22.5% AD cases provides a representative sample that is consistent with previous reports suggesting that approximately 20–30% of clinical FTD cases have AD pathology (Harris et al., 2013 (link); Lee et al., 2011 (link)).

Samples were analyzed using a Luminex xMAP platform (INNO-BIA AlzBio3™, Innogenetics, Ghent, Belgium) (n=52) or an ELISA assay (INNOTEST^®, Innogenetics, Ghent, Belgium) (n=13), as described⁴¹ . Briefly, the xMAP platform utilized capture monoclonal antibodies (MAbs) 4D7A3 (Aβ_1–42), AT120 (ttau), and AT270 (ptau) bound to color-specific beads. We used an assay sensitive to phosphorylation at threonine-181 since this is the Alzheimer’s Disease Neuroimaging Initiative standard for which the highly reliable Luminex method is available⁴¹ . Biomarker analytes were detected using reporting MAbs 3D6 (Aβ_1–42) and HT7 (ttau, ptau). Some older samples were analyzed with an ELISA method, where MAbs for capturing and reporting ttau and ptau were AT120/HT7 and BT2, HT7/AT270, respectively. As described previously^{20 (link)}, ELISA values for Aβ also were measured using an “in house” method with the capturing MAb BAN-50, and the reporting MAb BC-05. Using an autopsy-validated formula^{40 (link)}, a linear regression model converted natural-log transformed raw CSF values from ELISA to an xMAP equivalent.

Human CSF samples were obtained from 15 AD (according to ICD-10 criteria [27 ], mean age ± SD = 67.87 ± 10.232), 20 mild cognitive impairment (MCI according to Winblad criteria [28 (link)], mean age ± SD = 65.65 ± 10.373) and 21 subjective cognitive impairment (SCI, no objective cognitive impairment, mean age ± SD = 57.48 ± 5.409) subjects from the Memory Unit, Geriatric Clinic, Karolinska University Hospital, Huddinge. As part of the diagnostic procedure these subjects were assessed by the mini-mental state examination (MMSE) test, and levels of phosphorylated tau (phosphorylation site threonine 181) was measured by enzyme-linked immunosorbent assay (ELISA) kits (INNOTEST®, Innogenetics, Ghent, Belgium). Brain tissue samples were obtained from 10 AD-patients (9 Braak stage 5–6 definite AD [29 (link)], and 1 Braak stage 3–4 probable AD), and 10 non-demented control subjects, all from the Brain Bank at Karolinska Institutet. There was no statistical difference in age or post mortem interval (PMI) between the AD and control group (see Table 1). Half of each brain was fixed in formalin and tissue samples embedded in paraffin. The other half was dissected according to region, frozen and stored at −80°C. The study has been approved by the regional ethics committee of Stockholm.