Evaluation of Exploratory Fluid Biomarker Results from a Phase 1 Senolytic Trial in Mild Alzheimer's Disease

Abstract

Senescent cell accumulation contributes to the progression of age-related disorders including Alzheimer's disease (AD). Clinical trials evaluating senolytics, drugs that clear senescent cells, are underway, but lack standardized outcome measures. Our team recently published data from the first open-label trial to evaluate senolytics (dasatinib plus quercetin) in AD. After 12-weeks of intermittent treatment, we reported brain exposure to dasatinib, favorable safety and tolerability, and modest post-treatment changes in cerebrospinal fluid (CSF) inflammatory and AD biomarkers using commercially available assays. Herein, we present more comprehensive exploratory analyses of senolytic associated changes in AD relevant proteins, metabolites, lipids, and transcripts measured across blood, CSF, and urine. These analyses included mass spectrometry for precise quantification of amyloid beta (Aß) and tau in CSF; immunoassays to assess senescence associated secretory factors in plasma, CSF, and urine; mass spectrometry analysis of urinary metabolites and lipids in blood and CSF; and transcriptomic analyses relevant to chronic stress measured in peripheral blood cells. Levels of Aß and tau species remained stable. Targeted cytokine and chemokine analyses revealed treatment-associated increases in inflammatory plasma fractalkine and MMP-7 and CSF IL-6. Urinary metabolites remained unchanged. Modest treatment-associated lipid profile changes suggestive of decreased inflammation were observed both peripherally and centrally. Blood transcriptomic analysis indicated downregulation of inflammatory genes including FOS, FOSB, IL1β, IL8, JUN, JUNB, PTGS2. These data provide a foundation for developing standardized outcome measures across senolytic studies and indicate distinct biofluid-specific signatures that will require validation in future studies. ClinicalTrials.gov: NCT04063124.

Figure 1.

Correlation plot of cerebrospinal fluid (CSF) dasatinib (D) versus neurofilament light chain (NfL) levels. Post-treatment dasatinib (D) level correlation with baseline cerebrospinal fluid neurofilament light chain (NfL) derived from simple linear regression. R² = 0.7373; P = 0.0624.

Figure 2.

Inflammatory protein levels altered by dasatinib plus quercetin (D+Q) treatment measured by Luminex^® protein platform. a-d, Effect of dasatinib plus quercetin (D+Q) on plasma and cerebrospinal fluid (CSF) inflammatory markers. Mean difference (95% CI): a, plasma fractalkine, 629 (549.90 to 705.60); b, plasma MMP-7, 226 (0.198 to 452.90); c, CSF IL-6, 1.06 (0.500 to 1.616). Baseline to post-treatment changes were assessed using two-sided paired sample t-tests, P<0.05, N = 3–5, color coded by participant. Mean difference = post-treatment - baseline; 95% CI, for the post versus baseline mean difference. No correction for multiple comparisons was made due to small sample size.

Figure 3:

Effects of dasatinib plus quercetin (D+Q) treatment on the circulating plasma lipidome normalized to total protein concentration. a-d, Effects of dasatinib plus quercetin (D+Q) treatment on the circulating plasma lipidome normalized to total protein content. Plasma lipidome was assessed using multidimensional mass spectrometry-based shotgun lipidomics. a, MetaboAnalyst unsupervised PCA plot reducing all plasma lipid species data into three dimensions. Baseline and post-treatment groups are color-coded in gray and orange respectively, subjects are color coded to match color code assignments across all figures. b, All 11 lipid classes assessed in plasma samples. Paired samples are connected with a line. c, Volcano plot comparing all 194 plasma lipid species at baseline and post-treatment. d, Plot of the nine differentially abundant lipids (DALs) lipid species significantly decreased from baseline to post-treatment. Paired samples are connected with a line, each color represents a different subject (N=5). Only P < 0.1 are shown.

Figure 4.

Baseline and post-treatment significantly differentially expressed Conserved Transcriptional Response to Adversity (CTRA) gene counts in peripheral blood mononuclear cell samples measured with nanoString nCounter XT CodeSet custom CTRA gene expression panel. a-g, Effects of dasatinib plus quercetin (D+Q) on the expression of Conserved Transcriptional Response to Adversity (CTRA) gene counts measured in peripheral blood mononuclear cell (PBMC) samples. Seven inflammatory genes were significantly decreased post-treatment. Mean difference (B-statistic): a, FOSB, −218.87 (−0.713); b, PTGS2, −377.76 (−1.177); c, IL-8, −675.93 (−1.215) (d) FOS, −1579.32 (−1.669); (e) IL-1B, −152.94 (−1.922), (f) JUNB, −1267.29 (−3.546) (g) JUN, −505.57 (−3.754). Baseline to post-treatment changes were assessed using two-sided paired sample t-tests, P < 0.05, N = 4, color coded by participant. Paired baseline and post-treatment measures existed for all but one of the participants (blue) for whom only a post-treatment sample was collected. Mean difference = post-treatment - baseline. No correction for multiple comparisons was made due to small sample size.