Modeling pharmacodynamic response to the poly(ADP-Ribose) polymerase inhibitor ABT-888 in human peripheral blood mononuclear cells

Abstract

Background: Poly(ADP-ribose) polymerase (PARP) facilitates DNA repair and PARP inhibitors may potentiate the effect of DNA-damaging chemotherapeutic agents in patients with cancer. Collection of peripheral blood mononuclear cells (PBMCs) as a surrogate tissue to monitor PARP inhibitor pharmacodynamic effects has several advantages over tumor biopsy collection, including minimally invasive sample collection and the ability to collect multiple samples for longitudinal assessment of drug effect.

Methodology/principal findings: Using our previously validated immunoassay for measuring poly(ADP-ribose) (PAR), a product of PARP, in tumor biopsies, we validated a method to quantify PAR levels in PBMCs to monitor the pharmacodynamic effects of the PARP inhibitor ABT-888 in clinical trials. The inter-individual variation in PAR levels was large. No significant difference (P = 0.67) was measured between median baseline PAR levels in 144 healthy volunteers (131.7 pg/1×10(7) PBMCs [interquartile range, 79.5-241.6]) and 49 patients with cancer (149.2 pg/1×10(7) PBMCs [interquartile range, 83.2-249.3]). In addition, PAR levels monitored in healthy volunteers over 3 weeks had considerable intra- and inter-individual variation (range, 44-1073 pg PAR/1×10(7) PBMCs). As a pharmacodynamic model, we quantified changes in PAR levels in human PBMCs treated ex vivo with clinically relevant concentrations of ABT-888. Of 40 healthy volunteer PBMC samples treated with ABT-888, 47.5% had greater than 50% PAR reduction compared to vehicle-treated controls. Considerable inter-sample heterogeneity in PAR levels was measured, and several ABT-888-insensitive samples were identified.

Conclusions/significance: Our results emphasize the importance of using a validated method to measure PAR levels, and support further investigation into the role of PARP in PBMCs. To this end, the PAR immunoassay has been validated for use with PBMCs and incorporated into clinical trials to assess PBMCs as a potential pharmacodynamic surrogate for tumor biopsies in clinical trials of PARP inhibitors.

Figure 1. Dilution linearity of the PAR polymer standards and protein content from isolated PBMC samples.

(A) Dilution linearity of the PAR polymer standards was determined during quantitative validation of the PAR immunoassay for PBMCs. Concentrations of PAR standards ranged from 7.8 to 1000 pg PAR/mL. (B) Total protein content (bicinchoninic acid protein assay) for isolated PBMCs (counted by hemocytometer) from 11 healthy volunteers. Triangles indicate samples where plasma protein contamination skewed the total protein readout when compared to cell number.

Figure 2. Baseline PAR levels in PBMCs from healthy volunteers and patients with cancer.

(A) PAR levels in PBMC samples from 135 healthy volunteers and 47 patients with cancer. Box plot represents the interquartile range with median indicated; whiskers represent the 10^th and 90^th percentile. (B) PAR levels in PBMCs collected from eight healthy volunteers (HV) once per week for 3 consecutive weeks.

Figure 3. PBMC PAR levels in healthy volunteers and patients with cancer after ex vivo ABT-888 treatment.

(A) Pooled PBMCs from healthy volunteers were treated ex vivo for 2 h with increasing concentrations of ABT-888. PAR levels were then determined by PAR immunoassay and normalized (100%) to the vehicle-treated control. Error bars represent standard deviations from three separate experiments. PAR levels were compared between (B) PBMCs from four healthy volunteers (HV) and four patients (Pt) with cancer and (C) 40 individual healthy volunteers. PBMC samples were treated ex vivo with 0.21 µM ABT-888 (the target clinical exposure) for 2 h and PAR levels are reported relative to vehicle-treated controls (100%). Dashed line, 50% reduction.