Sphingomyelin-derived epacadostat nanovesicle enhances IDO1 inhibition for improved melanoma combination immunotherapy

Epacadostat (EPA), the most advanced inhibitor targeting the indoleamine 2,3-dioxygenase 1 (IDO1), is a critical endogenous metabolic immune regulator. In preclinical melanoma models¹, EPA potentiated the antitumor efficacy of immune checkpoint blockers (ICB) in comparison to ICB monotherapy. Phase I/II clinical trials proved the excellent safety and significant IDO1 targeting ability of EPA². Much to the disappointment, combing EPA with Keytruda (anti-PD-1 inhibitor) exerted zero numerical progression-free survival benefit versus Keytruda alone in a subsequent Phase III randomized clinical study (NCT02752074) in unresectable or metastatic melanoma patients³. Despite many reasons were blamed for this blockbuster failure such as dosing regimen and experimental designs, the poor pharmacokinetics of EPA is neglected¹.

Piggybacked by our previous success using sphingomyelin (SM)-derived drug nanotherapeutic strategy⁴, recently, we developed a SM-derived EPA nanovesicle. The SM-EPA conjugate was bridged by an oxime-ester bond that is highly responsive to hydrolase cleavage in tumors⁵. Following the spontaneous self-assembly into liposomal nanovesicle (Epacasome), EPA was securely packaged in the lipid bilayer. We showed that Epacasome strengthened IDO1 suppression by reducing the production of kynurenine, and promoted T cell proliferation in a co-culture setting. Also, it markedly increased the intracellular uptake via clathrin-mediated endocytosis with highly efficient drug release inside cancer cells. Systematic pharmacokinetic and biodistribution studies demonstrated that free EPA was eliminated from the blood rapidly with a T_1/2 = 0.15 h. However, Epacasome significantly extended the circulation T_1/2 to ~4.72 h, drastically increasing area under curve (AUC: 1,196.55 vs 7.81 μg/ml*h), mean residence time (MRT, 6.82 vs 0.22 h), and dramatically decreased volume of distribution (1.1 vs 5.97 μg/(μg/ml)). The improved pharmacokinetics allowed Epacasome a higher chance for tumor uptake based on the enhanced permeability and retention effect, resulting in 46.3-fold more EPA delivery to tumors compared to free EPA. More strikingly, Epacasome deepened tumor penetration and distribution, which are vital for improved IDO1 inhibition in tumors. Epacasome was superior to free EPA in reducing the tumor growth and prolonging mouse survival. It also synergized with anti-PD-1 inhibitor to further fortify the IFN-γ and granzyme-B-mediated cytotoxic T lymphocytes (CTLs) immunity. Meanwhile, this combination regimen repolarized macrophage to inflammatory M1 phenotype, attenuated granulocyte myeloid-derived suppressor cells and the Foxp3⁺ Tregs, and enhanced CD80⁺/CD86⁺ and CD103⁺ dendritic cells in B16-F10 melanoma model. These improved anti-melanoma efficacy and immune responses were found to be CD8 T cell dependent and attributed to the boosted IDO1 inhibition as reflected by the greatly increased tryptophan/kynurenine ratios in both plasma and tumor.

Noteworthily, in preclinical mouse models, EPA enhanced ICB response often when combined with chemotherapeutic agents. Hence, lacking a chemodrug could be one of other reasons for the failure of EPA + anti-PD-1 combination. To fill in the missing piece in the combination regimen and further boost the likelihood of clinical success, we used Epacasome to co-deliver DTIC, an approved chemotherapy for treating metastatic melanoma. DTIC has shown immunogenic potential by triggering natural killer (NK) and CD8⁺ T-cell immunity by upregulating NKG2D ligands on melanoma cells. Via systematic anticancer activity screening at various drug ratios in B16-F10 cells, DTIC/SM-EPA (3.66/1, molar ratio) had the smallest combination index, indicating the strongest synergy achieved. Thus, this drug ratio was implemented into Epacasome using thin film hydration. However, the poor aqueous solubility of DTIC limited its drug loading capacity (DLC) to only 0.24% when loaded into the Epacasome. To tackle this bottleneck, we creatively converted free DTIC into DTIC·HCl salt (Fig. 1b), which substantially increased DTIC DLC up to 40.13% in DTIC/Epacasome. Compared to co-administration of DTIC/Liposome + Epacasome, co-delivery DTIC/Epacasome not only improved the pharmacokinetics and tumor accumulation, but also synchronized the therapeutic delivery for both DTIC and EPA and enabled drug co-localization in tumor. The parental EPA can be readily released from the conjugate in tumors as the oxime-ester bond is cleaved by the intratumoral hydrolase, while endowing Epacasome with high degree of integrity during circulation. Furthermore, clathrin-mediated endocytosis allowed Epacasome to bypass the drug effluxion triggered by p-glycoprotein, enhancing intracellular uptake and drug retention inside cancer cells, avoiding drug resistance in the long run.

Fig. 1. The development of DTIC laden SM-derived EPA liposomal nanovesicles (DTIC/Epacasome) and the mediated immune antitumor effects.

a, Conjugation of sphingomyelin (SM) and epacadostat (EPA) bridged by an oxime-ester bond, which forms Epacasome b, Schematic depicting the co-encapsulation of DTIC using Epacasome. c, Schematic diagram of the antitumor immunity triggered by DTIC/Epacasome.

In the late-stage metastatic B16-F10-Luc2 melanoma model, we revealed that co-delivery DTIC/Epacasome outperformed co-administration of DTIC/Liposome and Epacasome or monotherapy in antitumor effects and inhibition of tumor metastasis with extended survival. The co-delivery nanotherapeutic further bolstered the efficacy of α-PD-1 and shrunk primary tumors volume by ~75% and eliminated tumor metastasis. In addition, DTIC/Epacasome significantly enhanced the expression of NKG2D ligands (Rae-1 and Mult-1) and MHC-I on tumor cells, boosted tumour infiltrating NKG2D⁺ (receptor), CD69⁺, IFNγ⁺, Perforin⁺, and Granzyme B⁺ NK⁺ cells or CD8⁺ T cells.

Through systemically depletion of CD8α, NK1.1, IFN-γ, and NKG2D, we corroborated that the antitumor effects of DTIC/Epacasome were CD8⁺ T cells and IFN-γ indispensable and partially dependent on NKs and NKG2D. To delve deeper into the therapeutic potential, DTIC/Epacasome was investigated in a clinically relevant post-surgical melanoma model. Co-delivery DTIC/Epacasome performed much better than that of co-administration of DTIC/Liposome and Epacasome in circumventing tumor relapse and extending mouse survival, especially when combined with PD-1 blockade.

In summary, SM-derived Epacasome nanotechnology can substantially fortify the clinical success of EPA by further enhancing the ICB immunotherapies, boasting considerable potential to rescue EPA and the whole IDO1 drug discovery pipeline.