Autophagy and VPS34 as targets in anti-cancer therapy

Abstract

Autophagy (Greek for self-eating) is a cellular pathway that breaks down cytoplasmic components to fuel metabolism. In this way, autophagy maintains cellular homeostasis and sustains survival during stress conditions. In cancer, autophagy is often dysregulated and has been described to contribute to malignancy and therapy resistance. Recently, autophagy has also been identified to play a key role in immunosuppression. Strategies to target autophagy for anti-cancer treatment have therefore generated renewed interest. The general aim of this thesis was to evaluate whether inhibiting VPS34, an essential protein in the autophagic pathway, impacts cancer growth and survival and represents a promising target for anti-cancer therapy.

Autophagy has been consistently reported as a mechanism to enable cancer progression and to drive treatment resistance. In Paper I, we, therefore, set out to identify cancer drugs that may trigger autophagy induction in cancer cells. Using the classical cellular GFP-LC3 reporter assay, we performed a high-throughput screening of the FIMM compound library containing approved or clinically relevant anti-cancer drugs. Among the discovered autophagy inducers in the screen, we selected two receptor tyrosine kinase inhibitors, Sunitinib and Erlotinib, to evaluate whether simultaneous autophagy inhibition may increase their anti-cancer efficacy. As a potent autophagy inhibitor, we used SB02024, a small molecule VPS34 inhibitor (VPS34i) developed using a fragment-based drug design approach by Sprint Bioscience. We demonstrate that SB02024 significantly decreased tumor growth in two breast cancer xenograft models. Combination treatment with SB02024 further increased sensitivity to both RTK inhibitors in breast cancer cells in vitro. Based on these results, VPS34 represents a promising target for anti-cancer therapy and as a combination therapy may improve the clinical efficacy of autophagy-inducing drugs.

Immunotherapies such as immune checkpoint blockade (ICB) have revolutionized cancer medicine but many cancer patients are either not eligible for treatment or fail to respond. Autophagy by controlling inflammation and immunity has been proposed as a critical immune evasion mechanism. In Paper II, we, thus, investigated the impact of inhibiting VPS34 in immunocompetent mice bearing syngeneic mouse tumors. Using melanoma or colorectal cancer (CRC) tumor-bearing mice, we showed that either genetic or pharmacological VPS34 inhibition significantly decreased tumor growth in a T- and NK-cell dependent manner. The anti-tumor response was further characterized by an increase of the chemokines CCL5, CXCL10, and the cytokine IFN in the tumor microenvironment resulting in the recruitment and activation of immune effector cells. Mechanistically, we demonstrate that VPS34i treatment increased secretion of CCL5 and CXL10 by tumor cells through activation of IRF7/STAT1 signaling. Finally, VPS34i treatment increased sensitivity to anti-PD-1/-PD-L1 therapy resulting in decreased tumor growth in vivo and prolonged survival of mice. These data demonstrate that VPS34 plays a crucial role in cancer immune evasion and targeting VPS34 may improve clinical responses to ICB in melanoma and CRC cancer.

Deepening the knowledge of how autophagy inhibition unleashes anti-tumor immune responses may enable patient stratification and identify rational treatment combinations with immunotherapeutic agents. Based on the discovery of a VPS34i-mediated anti-tumor immune response in Paper II, we sought out to identify underlying signaling mechanisms of the observed chemokine release in Paper III. To do this, we treated Renca tumor-bearing mice, a syngeneic mouse model for renal cell carcinoma (RCC), with VPS34i and profiled tumor mRNA expression using a NanoString panel focused on immune-oncology gene signatures. As previously shown for melanoma and CRC tumors, VPS34i treatment induced immune effector cell infiltration and chemokine secretion in Renca tumors. We further identified that VPS34i treatment triggered type I Interferon signaling, an important innate immunity pathway to trigger inflammation in the TME. Using both mouse and human RCC and melanoma cells, we demonstrated that VPS34i-mediated type I interferon signaling is dependent on cGAS-STING pathway activation. Combination treatment of VPS34i with the STING agonist ADU-S100 further increased type I interferon signaling in vitro. Finally, SB02024 + ADU-S100 combination treatment potently reduced melanoma tumor growth and increased mice survival in vivo.

In summary, these data identify VPS34 as a key regulator of cGAS-STING signaling in cancer cells. Pharmacological VPS34 inhibition thereby presents an attractive approach to increase immune cell infiltration in the TME and may provide clinical benefits for patients treated with immunotherapies including emerging drugs targeting the STING pathway.