Anti-PD-L1 F(ab) Conjugated PEG-PLGA Nanoparticle Enhances Immune Checkpoint Therapy

Abstract

Background: Immune checkpoint therapies are effective in the treatment of a subset of patients in many different cancers. Immunotherapy offers limited efficacy in part because of rapid drug clearance and off-target associated toxicity. PEG-PLGA is a FDA approved, safe, biodegradable polymer with flexible size control. The delivery of immune checkpoint inhibitors such as anti-PD-L1 (α-PD-L1) via PEG-PLGA polymer has the potential to increase bioavailability and reduce immune clearance to enhance clinical efficacy and reduce toxicity. Methods: The Fc truncated F(ab) portion of α-PD-L1 monoclonal antibody (α-PD-L1 mAb) was attached to a PEG-PLGA polymer. α-PD-L1 F(ab)-PEG-PLGA polymers were incubated in oil-in-water emulsion to form a α-PD-L1 F(ab)-PEG-PLGA nanoparticle (α-PD-L1 NP). α-PD-L1 NP was characterized for size, polarity, toxicity and stability. The relative efficacy of α-PD-L1 NP to α-PD-L1 mAb was measured when delivered either intraperitoneally (IP) or intravenously (IV) in a subcutaneous mouse colon cancer model (MC38). Antibody retention was measured using fluorescence imaging. Immune profile in mice was examined by flow cytometry and immunohistochemistry. Results: Engineered α-PD-L1 NP was found to have pharmacological properties that are potentially advantageous compared to α-PD-L1 mAb. The surface charge of α-PD-L1 NP was optimal for both tumor cell uptake and reduced self-aggregation. The modified size of α-PD-L1 NP reduced renal excretion and mononuclear phagocyte uptake, which allowed the NP to be retained in the host system longer. α-PD-L1 NP was non-toxic in vitro and in vivo. α-PD-L1 NP comparably suppressed MC38 tumor growth. α-PD-L1 NP appeared to elicit an increased immune response as measured by increase in germinal center area in the spleen and in innate immune cell activation in the tumor. Finally, we observed that generally, for both α-PD-L1 NP and α-PD-L1 mAb, the IP route was more effective than IV route for tumor reduction. Conclusion: α-PD-L1 NP is a non-toxic, biocompatible synthetic polymer that can extend α-PD-L1 antibody circulation and reduce renal clearance while retaining anti-cancer activity and potentially enhancing immune activation.

Keywords: Immunotherapy; MC38; PD-L1; PEG-PLGA.

Figure 1

Nanoparticle characterization. (A) Schematics of α-PD-L1 F(ab)-PEG-PLGA synthesis. The Fc portion of the antibody is detached to reduce immune clearance and the remaining F(ab) portion is attached to PEG-PLGA to increase nanoparticle size in order to prevent antibody from premature clearance. PLGA is used to control the degradation rates or drug release rates. Water-soluble synthetic polymer, PEG, is used as a protein carrier, to reduce the immunogenicity of the conjugated proteins. (B) MALDI-TOF of α-PD-L1 Fragmentation. MALDI-TOF of α-PD-L1 Fragmentation shows presence of both F(ab) and F(ab)₂. The presence of F(ab)₂ is not relevant in the next antibody-to-polymer conjugation step since F(ab)₂ lacks the free thiol groups exposed on F(ab) that is necessary for conjugation with maleimide linker on PEG-PLGA polymer. The filtration step following the conjugation step eliminates any free antibody fragments that did not bind to polymers. (C) Size and surface charge of the empty NP (n=16) and α-PD-L1 F(ab)-PEG-PLGA (n=27) measured by 90Plus Particle Size Analyzer, Zetasizer Nano ZS90 and 90Plus Zeta Potential Analyzer.

Figure 1

Nanoparticle characterization. (A) Schematics of α-PD-L1 F(ab)-PEG-PLGA synthesis. The Fc portion of the antibody is detached to reduce immune clearance and the remaining F(ab) portion is attached to PEG-PLGA to increase nanoparticle size in order to prevent antibody from premature clearance. PLGA is used to control the degradation rates or drug release rates. Water-soluble synthetic polymer, PEG, is used as a protein carrier, to reduce the immunogenicity of the conjugated proteins. (B) MALDI-TOF of α-PD-L1 Fragmentation. MALDI-TOF of α-PD-L1 Fragmentation shows presence of both F(ab) and F(ab)₂. The presence of F(ab)₂ is not relevant in the next antibody-to-polymer conjugation step since F(ab)₂ lacks the free thiol groups exposed on F(ab) that is necessary for conjugation with maleimide linker on PEG-PLGA polymer. The filtration step following the conjugation step eliminates any free antibody fragments that did not bind to polymers. (C) Size and surface charge of the empty NP (n=16) and α-PD-L1 F(ab)-PEG-PLGA (n=27) measured by 90Plus Particle Size Analyzer, Zetasizer Nano ZS90 and 90Plus Zeta Potential Analyzer.

Figure 2

Nanoparticle Stability and Toxicity. (A) Size of empty NP and α-PD-L1 NP were tracked over time using Zetasizer Nano ZS90. (n=3 in each group). (B) Cell number of MC38 were counted 26hours after culture with 10µg/ml α-PD-L1 mAb, α-PD-L1 NP, or empty NP. (C) Body weight of healthy mice that received control IgG, α-PD-L1 mAb, or α-PD-L1 NP were recorded every 3-4 days (n=3; 200µg/mice, 3 injections over 9 days; left). 20% weight loss is a criterion for euthanasia (dotted line above and below). Spleen weight of healthy mice that received either PBS or empty NP was measured at 4 weeks post treatment (n=3; right).

Figure 3

In vivo retention of intravenously injected NP compared to α-PD-L1 mAb. Cy5 dye tagged α-PD-L1 mAb, F(ab)2, F(ab), empty NP, and α-PD-L1 NP were injected into NSG mice IV and fluorescence in different organs was traced 24hours post injection (500 µg of mAb and an equivalent amount of NPs were injected to mice).

Figure 4

In vivo effects of injected NP based on route of administration. (A) Tumor volumes were measured in mice that received α-PD-L1 NP IP (n=5) versus IV (n=4) until the first mouse reached endpoint as dictated by protocol. Control IgG (n=3) was injected IP as standard practice (200µg/mice, 3 injections over 9 days). (B) Tumor volumes between mice that received control IgG, α-PD-L1 mAb, or α-PD-L1 NP were monitored every 3-4 days post MC38 injection for 3 weeks (n=5 in each group; 200µg/mice, 3 injections over 9 days). (C) Body weight of mice that received control IgG, α-PD-L1 mAb, or α-PD-L1 NP IP were monitored every 3-4 days post MC38 injection for 24-26days (n=5 in each group; left). Spleen weight was measured at the time mice were euthanized (right).

Figure 5

Change in immune profile over time. (A) IHC of CD4 and CD19 were used to identify the GC region and quantify the number of GC in the spleen at 3 weeks post MC38 injection (n=3 in each group). Flow cytometry was used for immune profiling of spleen (B) and tumor (C) of mice that received α-PD-L1 mAb or α-PD-L1 NP 4 weeks post MC38 injection.

Figure 5

Change in immune profile over time. (A) IHC of CD4 and CD19 were used to identify the GC region and quantify the number of GC in the spleen at 3 weeks post MC38 injection (n=3 in each group). Flow cytometry was used for immune profiling of spleen (B) and tumor (C) of mice that received α-PD-L1 mAb or α-PD-L1 NP 4 weeks post MC38 injection.