Of the 254 records initially identified through the database search, 50 were immediately excluded for reasons related to the study population (preclinical or animal model studies, n = 12) or publication type (review, pre-print, commentary, n = 38). Thus, 204 articles were screened, and among these, 192 were excluded as they did not focus on the levels of pTau217 in blood and/or CSF (n = 188) or did not deal with the AD patients population (n = 4). Therefore, 12 studies were included in the systematic review [14, 15, 22,23,24,25,26,27,28,29,30,31]. However, two studies were excluded because it was not possible to obtain the necessary data [14, 26]. Consequently, ten studies were included in the single group and head-to-head meta-analysis. Figure 1 shows an overview of the research process, while Table 1 outlines the key data extracted from each study (see Fig. 1 and Table 1).

PRISMA flow diagram outlining literature review and study selection

Overall, we included 819 participants classified as A+ and 1055 as A−, based on biomarkers [23, 28, 31] or biomarkers in combination with clinical diagnostic criteria [15, 22, 24, 25, 27, 29, 30]. Regarding laboratory methods used to assess pTau levels, the Meso Scale Discovery (MSD) platform was the most widely used [15, 22, 23, 25, 27,28,29], closely followed by the Single Molecule Array for Protein Detection (Simoa) [15, 22, 25, 30, 31]. Interestingly, two studies employed a novel approach, implementing a mass spectrometry assay technique either alone (namely, WashU) [15] or in combination with immunoprecipitation (IP-MS) [24], respectively. The methodological details of the included cohort and cross-sectional studies are outlined in Table 2 (see Tables 2 and 3). In brief, for the cohort studies, patient selection, outcome(s) ascertainment, and the assessment of subject comparability were optimal or adequate across nearly all investigations. Similarly, among the included cross-sectional studies, the same items proved optimal or adequate in all cases.

Table 4 provides an overview of the results of the ten studies included in the single-group meta-analysis (see Table 4).

About MSD, i.e., the most common laboratory technique (seven studies), pTau217 concentrations were highest in CSF (weighted mean [WM], 24.4; 95% CI, 18.1 to 30.7) and, to a lesser extent, in the blood (WM, 0.54; 95% CI, 0.42 to 0.65) in A+ subjects, while the A− group was characterized by sharply lower levels in both CSF (WM, 8.06; 95% CI, 4.50 to 11.6) and, particularly, blood (WM, 0.18; 95% CI, 0.16 to 0.20). Regarding Simoa, i.e., the second most widely used assessment method (six studies), the analysis highlighted a distribution overlapping with what was documented with MSD. Specifically, the A+ group showed higher levels in both CSF (WM, 23.3; 95% CI, 13.6 to 33.1) and blood (WM, 0.22; 95% CI, 0.15 to 0.29) as compared to A− participants who, similarly, displayed higher levels on the CSF (WM, 5.22; 95% CI, 2.99 to 7.45) than on blood (WM, 0.06; 95% CI, 0.03 to 0.09). Moreover, the only research that employed a mass spectrometry method reported a similar distribution of results, i.e., with the highest levels in A+ subjects, particularly in CSF (WM, 10.5; 95% CI, 9.89 to 11.1) and, to a lesser extent, in the blood (WM, 2.72; 95% CI, 2.44 to 3.0). In contrast, concentrations were lower in A− subjects and, again, higher in CSF (WM, 2.90; 95% CI, 2.74 to 3.06) than in blood (WM, 0.75; 95% CI, 0.69 to 0.81) [15]. Finally, one study evaluated an original IP-MS method in two different cohorts and documented a similar trend, i.e., the A+ group showed higher levels in the CSF (WM, 224; 95% CI, 202 to 246) as compared to blood (WM, 0.44; 95% CI, 0.16 to 0.71), whereas the A− participants were characterized by overall lower values than the previous group and, however, higher concentrations in the CSF (WM, 50.2; 95% CI, 37.8 to 62.7) than in blood (WM, 0.11; 95% CI, 0.05 to 0.16) [24]. In summary, all methods consistently demonstrated higher mean pTau217 levels in the CSF, representing the CNS compartment and, crucially, concentrations in the A+ group invariably exceeded those in the A− subjects, regardless of the biological substrate analyzed (i.e., CSF or blood). Notably, the IP-MS assessment method highlighted the highest CSF pTau217 levels in both A+ and A− groups [24]. At the same time, the mass spectrometry technique (i.e., WashU) showed the highest plasma pTau217 concentrations across both groups [15].

While the previous meta-analysis provided a descriptive overview of mean pTau217 levels in CSF and blood within the A+ and A− groups, stratified by assessment method, this section will show a direct comparison of these measurements. The aim is to evaluate whether pTau217 concentrations can reliably distinguish between individuals in the two groups. Table 5 and Figs. 2 and 3 provide an overview of the results of the included studies in the head-to-head meta-analysis, stratified by assessment method (see Table 5 and Figs. 2 and 3).

Results of the meta-analyses comparing the mean CSF levels (in pg/mL) among A+ versus A− subjects, stratified by assessment method (A: Immunoprecipitation; B: Lilly (MSD); C: Simoa). Abbreviations: A+/− Amyloid-positive/negative; CSF: cerebrospinal fluid; MSD: Meso Scale Discovery; Simoa: Single Molecule Array for Protein Detection

Results of the meta-analyses comparing the mean blood levels (in pg/mL) among A+ versus A− subjects, stratified by assessment method (A: Immunoprecipitation; B: Lilly (MSD); C: Simoa). Abbreviations: A+/− Amyloid-positive/negative; CSF: cerebrospinal fluid; MSD: Meso Scale Discovery; Simoa: Single Molecule Array for Protein Detection

Table 5 (first section) summarises the results about CSF pTau217 levels in A+ and A− groups (see Table 5). A direct comparison of pTau217 concentrations in CSF highlights that participants in the A+ group consistently exhibit, in a statistically significant way, higher values than those in the A− group, regardless of the assessment method employed. Specifically, about the MSD technique, mean pTau217 values were significantly higher in the A+ than in the A− group (mean difference [MD], 16.1; 95% CI, 10.8 to 21.4; p <0.001) (see Fig. 2, section B). Accordingly, Simoa also highlighted higher pTau levels in A+ subjects than in A− ones (MD, 18.1; 95% CI, 9.93 to 26.3; p <0.001) (see Fig. 2, section C). Finally, although assessed in only two patient cohorts, the IP-MS method also demonstrated its ability to differentiate between A+ and A− participants, yielding significantly higher values in the former compared to the latter (MD, 173.6; 95% CI, 146.3 to 201.0; p <0.001) (see Fig. 2, section A). Therefore, these findings underscore that evaluating pTau217 on CSF is a robust and reliable marker for distinguishing patients with amyloid pathology from those lacking this neuropathological characteristic.

Table 5 (second section) provides an overview of the results of blood pTau217 values in the A+ and A− groups (see Table 5). Notably, the analysis showed results similar to those observed on CSF. Specifically, irrespective of the laboratory method employed, pTau217 assessments on blood demonstrated that they can reliably differentiate A+ from A− participants in a statistically significant way. In particular, the MSD technique showed that A+ participants are characterised by higher values than A− ones (MD, 0.17; 95% CI, 0.08 to 0.25; p <0.001) (see Fig. 3, section B), and the same was highlighted for the Simoa (MD, 0.16; 95% CI, 0.10 to 0.22; p <0.001) (see Fig. 3, section C). Finally, also on blood, IP-MS showed a statistically significant difference between A+ and A− subjects (MD, 0.33; 95% CI, 0.10 to 0.56; p = 0.005) (see Fig. 3, section A). Thus, crucially, blood pTau217 measurements have also demonstrated their ability to differentiate between A+ and A− individuals, reinforcing the distinction observed in CSF assessments and highlighting their potential as reliable tools for identifying amyloid pathology.