Drug Development

Plasma phosphorylated tau (p-tau) has been used as a pre-screening tool in clinical trials. Using plasma p-tau for pre-screening may help identify individuals more likely to be Aβ-PET positive, reducing the number of PET scans needed at enrollment and decreasing recruitment costs. This study aims to evaluate the cost-effectiveness of plasma p-tau using liberal, intermediate, and conservative thresholds as pre-screening tools in clinical trials across the AD spectrum.

We studied 1,673 individuals across the AD spectrum from five cohorts: 707 cognitively unimpaired(CU) and 966 cognitively impaired(CI) with plasma p-tau217 measures and Aβ-PET. We tested three threshold methods:liberal(95% sensitivity), intermediate (Youden index), and conservative(95% specificity). The recruitment sample size was calculated based on p-tau217 assay sensitivity, specificity, and Aβ positivity prevalence for a final sample size of 1,000 based on recent phase 3 trials.

The intermediate threshold was the most cost-effective, followed by the conservative. Surprisingly, the liberal threshold did not outperform the strategy of screening only with Aβ-PET (Figure 1).

In CU, pre-screening with the intermediate threshold increased the initial sample size by 30% compared to the non-pre-screened group (n=5,000), while in CI, it increased by 28%(n=2,128). The number of Aβ-PET scans was reduced by 64% in CU and 46% in CI. Trial Cost: In CU, the intermediate threshold reduced costs by 59% compared to no pre-screening ($24M). In CI, costs were reduced by 39% compared to no pre-screening ($9.4M). We tested the pre-screening method using other p-tau217 assays and epitopes and observed reductions in Aβ-PET scans and recruitment costs for all p-tau assays (Figure 2). Using our models and data, we developed a free online tool that allows users to calculate the sample size needed when using plasma p-tau as a pre-screening method for recruiting individuals for clinical trials. This tool will be released with our publication at AAIC (Figure 3).

Our study demonstrates that plasma p-tau217 is an effective pre-screening tool for clinical trials, with the intermediate threshold being the most cost-effective for both CU and CI. This approach reduces the need for Aβ-PET scans and optimizes recruitment time and costs, potentially enhancing the efficiency of clinical trial designs within the AD spectrum.