Intratumoral Administration of a Novel Cytotoxic Formulation with Strong Tissue Dispersive Properties Regresses Tumor Growth and Elicits Systemic Adaptive Immunity in In Vivo Models

Abstract

The recent development of immune-based therapies has improved the outcome for cancer patients; however, adjuvant therapies remain an important line of treatment for several cancer types. To maximize efficacy, checkpoint inhibitors are often combined with cytotoxic agents. While this approach often leads to increased tumor regression, higher off target toxicity often results in certain patients. This report describes a novel formulation comprising a unique amphiphilic molecule, 8-((2-hydroxybenzoyl)amino)octanoate (SHAO), that non-covalently interacts with payloads to increase drug dispersion and diffusion when dosed intratumorally (IT) into solid tumors. SHAO is co-formulated with cisplatin and vinblastine (referred to as INT230-6). IT dosing of the novel formulation achieved greater tumor growth inhibition and improved survival in in vivo tumor models compared to the same drugs without enhancer given intravenously or IT. INT230-6 treatment increased immune infiltrating cells in injected tumors with 10% to 20% of the animals having complete responses and developing systemic immunity to the cancer. INT230-6 was also shown to be synergistic with programmed cell death protein 1 (PD-1) antibodies at improving survival and increasing complete responses. INT230-6 induced significant tumor necrosis potentially releasing antigens to induce the systemic immune-based anti-cancer attack. This research demonstrates a novel, local treatment approach for cancer that minimizes systemic toxicity while stimulating adaptive immunity.

Keywords: CTLA-4; INT230-6; PD-1; antibody; cell-penetration; checkpoint; cisplatin; cytotoxic; diffusion; dispersion; enhancer; immune activation; intratumoral; vinblastine.

Conflict of interest statement

Ian B. Walters, Franco Abbate and Lewis H. Bender have no conflicts.

Figures

Figure 1

Comparison of dispersion of aqueous drug solutions containing -((2-hydroxybenzoyl)amino)octanoate (SHAO) with India Ink compared to aqueous vehicle with drug (cis) with Ink alone in BxPc3-luc2 pancreatic murine tumor xenografts. The images show unexcised (A) and excised tumors (B), bifurcated along the same plane, dosed with 0.075 mL (1:11) or 0.225 mL (1:4) of the INT230-6 formulation (which contains the enhancer) or drug control administered intratumorally over 90 s to >500-mm3 tumors. (C) Paraffin blocks were made from the injected tumors. Caliper measurements of the longest axis of the stained region were taken to estimate the degree of ink dispersion (INT230-6: mean 8.25 mm vs. drug alone: 2.8 mm p < 0.0002).

Figure 2

In vitro incubation showing cell morphology in the presence or absence of the SHAO molecule. Images showing 24 h of incubation in vitro of Colon-26 cells with SHAO: 0, 1.32 and 4.44 mM.

Figure 3

INT230-6 in vivo treatment of Colon-26 tumors Tumor Growth Inhibition. BALB/c female mice were inoculated with 1 × 106 Colon-26 tumor cells in the right flank (cell injection volume, 0.1 mL/mouse). A total of 10 mice were assessed in each group and treated with intratumoral (IT) doses of INT230-6 or IT Vinblastine + Cisplatin or IV doses of Vinblastine + Cisplatin when mean tumor volume reached 325 mm3 (dose was 100 µL/400 mm3 tumor volume). Cisplatin was administered at 0.5 mg/mL while Vinblastine at 0.1 mg/mL once a day for 5 consecutive days (QDx5). (A) Tumor growth curves represented by the mean tumor volume of each group and error bars represent the SEM. Asterisks are representative of p values < 0.05 in the group’s comparison calculated with two-way ANOVA. (B) Kaplan–Meier survival curves of the Colon-26 tumor bearing mice. Asterisks are representative of p values < 0.05 in the group’s comparison and calculated through Log-Rank (Mantel-Cox) Test.

Figure 4

INT230-6 complete responders show immunological memory upon re-challenge. (A) The individual tumor growth curves of seven naïve BALB/c mice and the two complete responders (CRs) previously treated with INT230-6, inoculated with 1 × 106 Colon-26 tumor cells in the flank opposite to original inoculation. The two groups show statistical differences (p value < 0.05) calculated with two-way ANOVA on the mean tumor volume of each group. (B) Kaplan–Meier curves assessing the survival of mice for 44 days. None of the CR mice showed tumor development for the entire length of the study. No treatment was administered in this study.

Figure 5

IHC analysis of Colon-26 tumors untreated and INT230-6-treated tumors ~10 days post dose.

Figure 6

Tumor growth inhibition in 4T1 tumor bearing mice treated with INT230-6. Tumor growth curves showing BALB/c mice inoculated with 1 × 106 4T1 tumor cells in the right flank and treated with INT230-6 IT (QDx3), INT230-6 IV (QDx3), anti-PD-1 ((Q3Dx2; 3 off) x2) and untreated controls. Due to rapid metastasis formation, mice were randomized at approximately 125 mm3 mean tumor volume. Two mice from the INT230-6 IV group were removed from the study for weight loss. Asterisks represent statistically different groups (p value < 0.05, calculated with two-way ANOVA).

Figure 7

Effects of INT230-6 alone and in combination with checkpoint inhibitors (CTLA-4 and/or PD-1 antibodies) on median tumor volume in mice bearing Colon-26 tumors. (A) There were ten (10) animals per group. The median tumor volume was chosen to better represent the tumor growth curves of BALB/c mice for the full length of the study (100 days). The following animals were found dead in the cages and removed from the study for a better representation of the tumor growth curves: Group 1 (1 Animal), Group 2 (3 animals), Group 3 (2 animals), Group 4 (1 animal), Group 5 (1 animal), Group 6 (1 animal), Group 7 (1 animal), Group 8 (1 animal). Because mice died at different times during the study, statistical analysis was conducted with two-way ANOVA at day 22 (latest available data for the control group 1). At this time, INT230-6-treated groups had statistically significant reduced tumor growth curves when compared to either Group 1 (Control) or Group 8 (anti-PD-1 + anti-CTLA-4) (p value < 0.05). (B) Kaplan–Meier survival curves showing that the addition of anti-PD-1 antibodies to a single cycle (QDx5) of INT230-6 treatment, increased the number of CR animals from 1/7 (14%) (Group 2) to 5/9 (55%) (Group 4). A similar result was obtained in Group 7 (INT230-6 + anti-PD-1 + anti-CTLA-4), where 5/9 mice had CR (55%) considering that CTLA-4 alone added limited benefits in combination with INT230-6; 2/9 CR in Group 6 (22%). The log rank (Mantel-Cox) test showed statistical differences between groups treated with INT230-6 and Group 1 and Group 8.

Figure 8

Molecular structure of 8-((2-hydroxybenzoyl)amino)octanoate, also referred to as SHAO.