Priming agents transiently reduce the clearance of cell-free DNA to improve liquid biopsies

Abstract

Liquid biopsies enable early detection and monitoring of diseases such as cancer, but their sensitivity remains limited by the scarcity of analytes such as cell-free DNA (cfDNA) in blood. Improvements to sensitivity have primarily relied on enhancing sequencing technology ex vivo. We sought to transiently augment the level of circulating tumor DNA (ctDNA) in a blood draw by attenuating its clearance in vivo. We report two intravenous priming agents given 1 to 2 hours before a blood draw to recover more ctDNA. Our priming agents consist of nanoparticles that act on the cells responsible for cfDNA clearance and DNA-binding antibodies that protect cfDNA. In tumor-bearing mice, they greatly increase the recovery of ctDNA and improve the sensitivity for detecting small tumors.

Fig. 1.. Priming agents reduce clearance of cfDNA and improve the recovery of ctDNA.

(A) Priming agents are injected 1 to 2 hours prior to a blood draw and improve the recovery of ctDNA by >10-fold. (B) (Left) In the absence of a priming agent, cfDNA (mostly in the form of mononucleosomes) is (i) rapidly taken up by macrophages of the MPS in the liver and (ii) degraded by circulating nucleases, yielding little ctDNA molecules in a blood draw. (Center) Following intravenous administration of a nanoparticle (NP) priming agent, (iii) cellular uptake is attenuated through MPS saturation. (Right) Intravenous administration of an antibody priming agent (iv) extends the half-life of cfDNA in circulation and (v) protects it from nuclease digestion. Both priming strategies enhance ctDNA recovery and improve mutation detection from a blood draw.

Fig. 2.. SPE liposomes inhibit the uptake of mononucleosomes by macrophages in vitro and increase the recovery of cfDNA through decreased clearance in healthy mice.

(A) Schematic of in vitro macrophage uptake inhibition assay. (B) Representative images of uptake of Cy5-labeled mononucleosomes (Cy5-MN) following incubation of J774A.1 with different liposomes at 5 mg/ml, without liposomes or Cy5-MN (negative control, NC), or with Cy5-MN only (positive control, PC). Scale bars, 100 nm. (C) Quantification of Cy5-MN uptake by J774A.1 cells from epifluorescence images after liposome pre-treatment (mean ± SEM, n = 3 to 4 wells per condition, N = 2). *P < 0.05; **P < 0.01; ***P < 0.001; one-way ANOVA. (D) (Left) Experimental approach to determine the plasma bioavailability of W601-mononucleosomes (W601-MN) following SPE liposome priming. (Right) Percentage of W601 remaining in plasma 60 min after administration of different SPE liposome doses (median, n = 3 to 4 mice per group, N = 2). *P < 0.05, two-tailed Mann-Whitney test. (E) (Left) Experimental approach to determine plasma cfDNA yields and liposome organ biodistribution. (Right) Plasma cfDNA concentration following Cy7-SPE administration (mean ± SEM, n = 3 mice per group). The largest elevation relative to the PBS group was at 30 min with the dose of 100 mg/kg liposome (10.3-fold, *P = 0.034) and at 3 hours with the dose of 300 mg/kg liposome (78.0-fold, **P = 0.005) (unpaired two-tailed t test; n = 3 mice per group, N = 1). (Insert) Organ biodistribution of Cy7-SPE liposomes 1 hour after administration. Images from a representative mouse are shown (n = 4 mice, N = 3).

Fig. 3.. Antibody priming agent binds cfDNA and attenuates its clearance in healthy mice.

(A) EMSA of free and histone-bound dsDNA (4 ng/μL per lane) with varying concentrations of DNA-binding mAb 35I9 in PBS (N = 3). (B) Fluorescence signal from the digestion of a DNA substrate carrying a hexachlorofluorescein dye on one end and a dark quencher on the other, with or without 0.1 U of DNase I and mAb 35I9. Points indicate mean and lines indicate SEM of three technical replicates. Fluorescence signals across the whole experiment were compared by using mixed models with replicates as random effects (N = 2). (C) (Left) Experimental approach to evaluate the effect of priming mAb on dsDNA clearance. (Right) Concentration of W601 in plasma 60 min after injection of W601 only, coinjection with an unrelated IgG2a antibody, with 40 μg of DNA-binding antibody 35I9, or with 40 μg of 35I9 together with anti-FcγRI (20 μg), anti-Fc(γ)RII, and anti-Fc(γ)RIII (40 μg) (n = 3 mice per group, N = 2). (D) (Left) Overview of the DNA binding and FcγR binding properties of engineered mAb aST3 versus IgG2a control mAb and DNA-binding mAb 35I9. (Right) Concentration of W601 in plasma 60 min after coinjection of W601 with an unrelated IgG2a antibody or with the Fc-mutant aST3 DNA-binding antibody (n = 3 mice per group, N = 3). (E) (Left) Experimental approach to quantify pharmacokinetics of 3519 and aST3 labeled with AQuora750 in plasma. (Right) Plasma clearance of antibodies over time (mean ±SEM, n = 5 mice per group, N = 1). (F) Biodistribution and quantification of 3519 (Fc-WT antibody) or aST3 concentration in liver and spleen 1 hour after administration (n = 5 mice per group, N = 1). [(C), (D), (F)] ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; one-way ANOVA.

Fig. 4.. Liposome priming improves ctDNA recovery and enables detection of smaller tumors in a murine lung metastasis model.

(A) Experimental approach for the detection of mutations from the plasma of Luc-MC26 tumor-bearing mice using the liposome priming agent. For each mouse, blood was drawn prior to and 1 hour after i.v. administration of PBS or SPE liposomes (100 mg/kg) at 1 week (W1), 2 weeks (W2), and 3 weeks (W3) after tumor inoculation. (B) Plasma cfDNA concentrations, (C) concentration of mutant molecules detected, and (D) tumor fractions 1 hour after PBS (white) or SPE (blue) administration at W1, W2, or W3 (n = 6 to 12 mice per group). (E) (Left) Mutational fingerprints showing distinct SNVs detected pre– and post–SPE administration for mice with high tumor burden (burden > 1.5e8 p/s total flux, as measured by IVIS). Each vertical band corresponds to a SNV in our 1822-SNV panel and is colored blue if detected at least once in the plasma sample. (Right) Quantification of these distinct SNVs. (F) Sensitivity of ctDNA tests versus SNV threshold for tumor detection in mice with low tumor burden after administration of PBS or SPE liposomes (burden < 1.5e7 p/s total flux, as measured by IVIS). (G) Sensitivity of ctDNA tests for different tumor burdens after PBS or SPE administration (Low, burden < 1.5e7 p/s; Medium, 1.5e7 p/s ≤ burden ≤ 1.5e7 p/s; High, burden > 1.5e8 p/s). Sensitivity was calculated as the fraction of samples for which the number of SNVs detected in a blood sample was ≥ 2 (n = 6 to 12 mice per group; *P < 0.05, Chi-squared test) (fig. S20, independent replicate at week 2). Boxplots in (B), (C), (D), and (E) show median and interquartile range. ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; one-way ANOVA.

Fig. 5.. Antibody priming agent improves ctDNA recovery in murine lung metastasis model.

(A) Experimental approach for the detection of mutations from the plasma of Luc-MC26 tumor-bearing mice with the antibody priming agent aST3. (B) Plasma cfDNA concentrations, (C) concentration of mutant molecules detected, and (D) tumor fractions detected 2 hours after administration of IgG2a control mAb or various doses of aST3 (n = 6 mice per group) (fig. S29, independent replicate at aST3 4.0 mg/kg). (E) Percentage of distinct SNVs from an 1822-SNV panel detected in plasma with control mAb or various doses of aST3. (F and G) Estimation of sensitivity for detection of ctDNA upon administration of 8 mg/kg of IgG2a control or 4 mg/kg of aST3 versus (F) panel size (G) or tumor fraction based on binomial down-sampling of mutant molecules, with a detection threshold of ≥ 2 SNVs (mean ± SEM, n = 100 replicates). Boxplots in (B) to (E) show median and interquartile range. ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001; one-way ANOVA.