STINGing leukemia stem cells through PSTK inhibition

In this issue of Blood, He et al provide compelling evidence that PSTK inhibition triggers ferroptotic cell death and links ferroptosis to cyclic GMP-AMP synthase (cGAS)–stimulator of interferon genes (STING) activation.¹ These findings pave the way for future drug development in acute myeloid leukemia (AML), particularly in overcoming resistance to apoptosis.

Targeting mitochondria has shown increasing promise in AML treatment, exemplified by the discovery and widespread use of BH3 mimetics like venetoclax, which induce intrinsic apoptosis by targeting BCL2 dependency in AML blasts.² However, both primary and acquired resistance to BH3 mimetics remain significant therapeutic challenges. Mechanisms of resistance include metabolic adaptations, clonal selection of signaling mutations, and evasion of apoptotic pathways, such as the emergence of BAX gene mutations.³ Ferroptosis, a regulated form of cell death driven by iron-dependent lipid peroxidation, is influenced by metabolic enzymes that regulate lipid metabolism, iron homeostasis, and antioxidant defenses.⁴ One of its key regulators, the selenoprotein GPX4, plays a critical role in maintaining antioxidant defenses. However, to date, direct inhibition of GPX4 has been challenging in malignant hematology.⁵ 

In their study, He and colleagues performed a genome-wide clustered regularly interspaced short palindromic repeats (CRISPR)-based loss-of-function screen and identified PSTK as a genomic determinant of myeloid differentiation. PSTK, a key enzyme in the synthesis of selenoproteins, including glutathione peroxidases and thioredoxin reductases, emerged as a critical regulator of both differentiation and cell viability. PSTK inactivation significantly reduced the frequency of leukemia-initiating cells, which are known to drive relapse. Moreover, genetic and pharmacological inhibition of PSTK demonstrated potent synergistic activity with conventional chemotherapy and venetoclax-based nonintensive therapies in preclinical models, including patient-derived xenografts, while sparing normal hematopoietic cells.

Interestingly, the PSTK inhibitor punicalin was particularly effective when administered immediately after chemotherapy, with no additional cardiac toxicities in mouse models. This effect was attributed to high mitochondrial reactive oxygen species generation induced by cytotoxic agents and the absence of detoxification due to reduced expression of detoxifying enzymes after PSTK inactivation. Cell death triggered by PSTK inhibition exhibited hallmark features of ferroptosis and was dependent on cGAS-STING activation. The proposed model suggests that PSTK inhibition induces mitochondrial oxidative stress, leading to the release of mitochondrial DNA into the cytosol, which activates cGAS-STING signaling. This activation further amplifies mitochondrial reactive oxygen species generation, creating a positive feedback loop that enhances ferroptosis (see figure).

Current treatment options remain limited for AML patients who fail intensive chemotherapy or venetoclax-based regimens. Novel agents aimed at enhancing first-line therapy efficacy are under active investigation. Strategies combining small molecules or immunotherapies with chemotherapy or venetoclax-based backbones in frontline settings are promising. However, the optimal timing of drug administration (sequential vs combined) remains a critical area of research to maximize response rates while minimizing cumulative hematologic and nonhematologic toxicities.⁶ 

He et al's study provides significant preclinical insights into the effects of metabolic reprogramming immediately after cytotoxic therapy. This metabolic rewiring renders AML blasts highly sensitive to oxidative agents and metabolic drugs. Furthermore, the authors underscore the potential to overcome intrinsic apoptosis resistance in AML by activating alternative cell death pathways.⁷ Notably, this strategy warrants further exploration in the context of TP53-mutated AML, where primary apoptosis resistance is a well-documented challenge. CRISPR screens have demonstrated that TP53 mutations are associated with resistance to apoptosis,^8,9 and clinical studies have validated these findings in large cohorts of patients with AML and myelodysplastic syndrome who harbor TP53 mutations.

Finally, the authors highlight the efficacy of ferroptosis in overcoming apoptosis resistance in AML and elucidate the intricate interplay between metabolism, cell death, and immunogenicity through the cGAS-STING pathway. PSTK inhibition thus emerges as a promising therapeutic target in cancer treatment. Although STING agonists are already under development for oncology and lymphoma,¹⁰ the potential of ferroptosis induction via direct or indirect PSTK inhibitors warrants further investigation as a novel approach in AML therapy.

Conflict-of-interest disclosure: S.G. declares no competing financial interests.