Characterization of CPH:SA microparticle‐based delivery of interleukin‐1 alpha for cancer immunotherapy

Abstract Background Interleukin‐1 alpha (IL‐1α) is a pro‐inflammatory cytokine that can activate immune effector cells and trigger anti‐tumor immune responses. However, dose‐limiting toxicities including cytokine storm and hypotension has limited its use in the clinic as a cancer therapy. We propose that polymeric microparticle (MP)‐based delivery of IL‐1α will suppress the acute pro‐inflammatory side effects by allowing for slow and controlled release of IL‐1α systemically, while simultaneously triggering an anti‐tumor immune response. Methods Polyanhydride copolymers composed of 1,6‐bis‐(p‐carboxyphenoxy)‐hexane:sebacic 20:80 (CPH:SA 20:80) was utilized to fabricate MPs. Recombinant IL‐1α (rIL‐1α) was encapsulated into CPH:SA 20:80 MPs (IL‐1α‐MPs) and the MPs were characterized by size, charge, loading efficiency, and in‐vitro release and activity of IL‐1α. IL‐1α‐MPs were injected intraperitonially into head and neck squamous cell carcinoma (HNSCC)‐bearing C57Bl/6 mice and monitored for changes in weight, tumor growth, circulating cytokines/chemokines, hepatic and kidney enzymes, blood pressure, heart rate, and tumor‐infiltrating immune cells. Results CPH:SA IL‐1α‐MPs demonstrated sustained release kinetics of IL‐1α (100% protein released over 8–10 days) accompanied by minimal weight loss and systemic inflammation compared to rIL‐1α‐treated mice. Blood pressure measured by radiotelemetry in conscious mice demonstrates that rIL‐1α‐induced hypotension was prevented in IL‐1α‐MP‐treated mice. Liver and kidney enzymes were within normal range for all control and cytokine‐treated mice. Both rIL‐1α and IL‐1α‐MP‐treated mice showed similar delays in tumor growth and similar increases in tumor‐infiltrating CD3+ T cells, macrophages, and dendritic cells. Conclusions CPH:SA‐based IL‐1α‐MPs generated a slow and sustained systemic release of IL‐1α resulting in reduced weight loss, systemic inflammation, and hypotension accompanied by an adequate anti‐tumor immune response in HNSCC‐tumor bearing mice. Therefore, MPs based on CPH:SA formulations may be promising as delivery vehicles for IL‐1α to achieve safe, effective, and durable antitumor responses for HNSCC patients.


| BACKGROUND
Cytokines are small cell-signaling proteins that regulate innate and adaptive immune responses. Cytokines are highly important for tumor immunosurveillance and the use of cytokine therapy to activate the immune system of cancer patients is an important treatment modality. 1,2 Various cytokines have been tested as anti-cancer agents in preclinical and clinical studies. [3][4][5][6][7] These studies led to the approval of interferon-alpha (IFNα) for hairy cell leukemia in 1986 and it is now also approved for use in follicular lymphoma, melanoma, renal cell carcinoma, cervical intraperitoneal neoplasms and AIDS-related Kaposi's sarcoma. [8][9][10][11] Additionally, high-dose interleukin-2 (IL-2) was approved for the treatment of metastatic renal cell carcinoma in 1992, and metastatic melanoma (MM) in 1998. [12][13][14][15][16] Clinical use of these cytokines for the treatment of cancer was among the first lines of evidence that the immune system can be modulated to mount significant anti-tumor immune responses.
Unfortunately, the use of IL-1α or IL-1β as anti-cancer agents in clinical trials have not lived up to the initial excitement caused by the observed preclinical anti-tumor immune responses. Clinical studies were conducted in the late 1980s and early 1990s in cancer patients.
Recombinant IL-1 ligands (marketed as Dainippon/Immunex  and Syntex [IL-1β]) were administered to cancer patients with the most common symptoms being flu-like symptoms including fever, chills, rigors and nausea. 17 However, hypotension was a problematic dose-limiting toxicity with 3 of 5 patients experiencing hypotension at the maximum tolerated dose (MTD) of 0.1 and 0.3 μg/kg (with pressors for blood pressure support). 17 IL-1 ligand-induced hypotension resulted from a significant drop in systemic vascular resistance (SVR) leading to a rise in heart rate, and cardiac output rose to compensate for the drop in blood pressure and SVR. 17 It is reported that the cardiovascular effects of IL-1α were similar to that of septic shock and further plans for clinical trials for cancer therapy were discontinued. 17 Therefore, managing toxicity of cytokine therapy is critically important for progressing from bench to bedside.
Controlled release systems using polymeric nano/micro-particles may provide an appropriate strategy to infuse low levels of IL-1 ligands (and other cytokines) over a longer period of time to cancer patients and also protect from the deleterious side effects associated with large bolus doses. Synthetic polymeric microparticles (MPs) such as poly(lactide-co-glycolide) (PLGA), polyanhydrides, and poly-(diaminosulfide)s (PNSN) are examples of drug delivery vehicles that have grown in interest in the cancer immunotherapy field. Polymeric MPs have been investigated for the delivery of checkpoint inhibitors, engineered T cells, co-stimulatory receptor agonists and cancer vaccines. [38][39][40][41] Polyanhydrides are an example of surface eroding polymers that are highly preferred for drug delivery strategies because of the accurate predictability of the erosion process. [42][43][44] This surface erosion process contributes to a constant release rate of the drug as well as enhanced stability of the drug in the polymer. 42,43 More importantly, polyanhydrides are the only class of surface eroding polymers that are Food and Drug Administration (FDA) approved. Polyanhydrides can also be used for cytokine therapy with a goal for reduced toxicities while retaining therapeutic activity. Polyanhydride copolymers composed of 1,6-bis-(p-carboxyphenoxy)-hexane: sebacic acid 20:80 (CPH:SA 20:80) were used in this study to encapsulate IL-1α. Here we evaluated the safety profile of CPH:SA-based IL-1α-MPs compared to free (unencapsulated) IL-1α in experimental animals. Results showed that CPH:SA-based IL-1α-MPs generated a slow and sustained systemic release of IL-1α resulting in reduced toxicity (weight loss, cytokine storm, and hypotension) compared to free IL-1α in tumor-bearing mice. Therefore CPH:SA formulations may be promising as delivery vehicles for IL-1α to achieve safe and effective antitumor responses for cancer patients.

| Fabrication of microparticles loaded with IL-1α
Interleukin-1α-microparticles (IL-1α-MPs) were fabricated by a waterin-oil-in-water (w/o/w) double emulsion solvent evaporation method that has been described previously with some modifications. 40  twice with nanopure water. The obtained particles were resuspended in 2 ml of 10% sucrose solution. The particle suspensions were frozen at À80 C and then lyophilized using a FreeZone 4.5 freeze dry system (Labconco Corporation) at À53 C and 0.045 mbar pressure. The lyophilized microparticles were stored in sealed containers at À20 C.
Blank microparticles were fabricated using the same process. For the second dose escalation study, CPH:SA microparticles with increased IL-1α loading were prepared following the above procedure with some modifications, where 550 μg of IL-1α was dissolved in 100 μl of 1% PVA solution to form W1, and the final particles were resuspended in 2 ml of 5% sucrose instead of 10% sucrose. Characterization data is shown in Table 1.

| Characterization of CPH:SA IL-1α-MPs
The MPs were resuspended in nanopure water for characterization.
Size distribution and zeta potential of the MPs were measured using a Zetasizer Nano-ZS (Malvern) by dynamic light scattering. Particle suspension was added to a polystyrene cuvette (Sarstedt Inc.) and folded capillary cell (Malvern) to determine size and zeta potential, respectively. The shape and surface morphology of MPs were determined by scanning electron microscopy (SEM) (Hitachi High Technologies). A drop of particle suspension was transferred onto a silicon wafer mounted on a SEM stub and dried completely at room temperature.
Samples were coated with gold palladium using an argon beam K550 sputter coater (Emitech Ltd.). SEM images were taken using a Hitachi

| In vitro drug treatment and immune cell activation
Recombinant human IL-1α (rIL-1α) was purchased from BioLegend (San Diego, California) and used at a concentration of 10-100 ng/ml for 24 h. Human peripheral blood mononuclear cells (PBMCs) were T A B L E 1 Characteristics of IL-1α-loaded CPH:SA microparticles.  monocytes, and CD40+ mDCs were considered activated. The gating strategies are shown in Figure S1. Percentage of positive cells were then quantified and plotted as fold change compared to control.

| In vivo tumor cell implantation and drug treatment
Four-six-week-old male C57BL/6J mice were purchased from The Jackson Laboratory. All experimental animals were housed in the Animal Care Facility at the University of Iowa and handled using aseptic procedures. Mice were allowed at least 5 days to acclimate prior to handling, and food and water were made freely available. 2.8 | Assessment of blood pressure in conscious mice by radiotelemetry C57Bl/6 mice (n = 6 mice/treatment group) were anesthetized with ketamine (91 μg/g, IP) and xylazine (9.1 μg/g, IP) and radiotelemetry probes (PC10, DSI) that allowed for measurement of arterial blood pressure and heart rate were implanted into the thoracic aorta through the left common carotid artery, as described previously. 45,46 The mice were allowed a week to recover. Each animal was housed in individual cages in the Animal Care Facility and monitored daily for signs of ill-health. Mice were then inoculated with HNSCC (mEERL) cells and treated with rIL-1α and IL-1α-MPs (equivalent to 7.5 μg IL-1α on Days 10) as already described above. Changes in blood pressure (mean arterial pressure) and heart rate (HR) were measured over the entire experimental period (18 days) at a sampling rate of 500 Hz for 10 s once every 5 min using the DSI Acquisition software.

| Statistical analyses
Statistical analysis was carried out using GraphPad PrismV

| IL-1α evoked hypotension is prevented by CPH:SA IL-1α-MPs
One of the adverse effects of various cytokine therapies is the marked reduction in blood pressure. Therefore, we measured blood pressure and heart rate in a separate cohort of tumor-bearing conscious mice before and after cytokine delivery via radiotelemetry. Mean arterial pressure reduced significantly in mice treated with rIL-1α and tumor regression in 80% of treated mice. 47 However, significant weight loss was observed in the IL-1α-NP-treated mice (compared to the empty NP-treated mice) in the initial 5 days after treatment, 47 which was likely due to undesirable particle burst release kinetics with 90% of IL-1α being released in the first hour. Nevertheless, the observed antitumor activity of these NPs led to the further identification of particle formulations with improved release kinetics and toxicity profiles.
As shown here, CPH:SA-based IL-1α-MPs showed a sustained release profile accompanied by minimal weight loss. There are many reasons for the difference in release kinetics between the CPH:SAbased IL-1α-MPs and the previously studied CPTEG:CPH IL-1α-NPs which include the size difference between MP and NP systems, different preparation procedures, and differences in interaction/packaging between rIL-1α and polymer. However, CPH:SA polymers are more hydrophilic than CPTEG:CPH polymers and are likely more compatible for loading with water soluble cytokines 40 which may explain the differences in release kinetics between these three polymers.
PLGA polymers as a delivery vehicle for IL-1α have been previously tested. 48 Unlike polyanhydrides, PLGA polymers undergo bulk erosion and not surface erosion. 49 Therefore, these polymers do not show well-defined drug release kinetics, particularly for water-labile low molecular weight agents like cytokines/chemokines. 49  life-threatening events. 53 One of the life threating conditions that can arise from CRS is hypotension, which is of high concern for cytokines like IL-1α. 17 All clinical studies reported hypotension from rIl-1α and rIL-1β therapy resulting in patient withdrawal from clinical trials and early termination of the trials. Given the lack of cytokine storm ( Figure 3c) and hypotension (Figure 5a) observed in IL-1α-MP-treated mice, it is worth following up on these studies using multiple doses and dosing schedules to confirm.
Lastly, it was surprising that that the three different dosing strategies of rIL-1α tested here (3.75 μg Â 2 doses, 7.5 μg once and 15 μg once) showed no major differences in weight loss (Figures 3b and 4).

CONFLICT OF INTERESTS
No conflicts of interest to disclose.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.