Targeting LKB1/STK11 mutant cancer: distinct metabolism, microenvironment, and therapeutic resistance

Abstract

Despite the development of new classes of therapeutics in oncology, patients with tumors harboring mutations in the tumor suppressor gene STK11/LKB1 continue to exhibit poor clinical response and therapeutic resistance. Recent advances in the understanding of LKB1 mutant tumor biology have illuminated how metabolism and the tumor microenvironment function as effectors of the aggressive nature of this tumor type. New findings have revealed how metabolic reprogramming, a hallmark of LKB1 mutant tumor biology, can be exploited as a potential targetable liability in these tumors. Characterization of the distinctly immunosuppressive LKB1 mutant tumor microenvironment has motivated multiple discoveries of new approaches for rewiring the microenvironment in order to overcome immunotherapy resistance. Indeed, overcoming therapeutic resistance in LKB1 deficient tumors continues to be a major research focus, and some preclinical studies have advanced to clinical trials. Here we critically analyze these findings and discuss therapies currently in development that aim to leverage this new understanding for clinical benefit.

The LKB1 tumor suppressor: distinct tumor biology in the absence of a critical gatekeeper

The serine/threonine kinase LKB1/STK11 has been identified as a central regulator of cell physiology and cancer biology including metabolism, cell polarity, growth control, and tumor suppression [1]. The tumor suppressor function of LKB1 was first described in 1998, when heterozygous LKB1 loss-of-function was identified as the genetic event that causes the inherited cancer disorder Peutz-Jeghers Syndrome [2]. LKB1 has since been identified as a critical tumor suppressor in sporadic cancers including lung, pancreatic, and colorectal cancer [3]. For example, it is estimated that in 2025 there will be approximately 40,000 new cases of LKB1 mutant lung adenocarcinoma (LUAD) in the U.S. alone. LKB1 mutant cancer remains a great unmet clinical need, highlighted by the observation that LKB1 mutation correlates with poor patient prognosis [4,5]. In LUAD, LKB1 mutation frequently co-occurs with KRAS [6] and/or KEAP1 [7,8]. Within the KRAS mutant patient population, concurrent LKB1 mutation was found to correlate with shorter disease-free survival [5]. While new classes of oncology drugs have revolutionized cancer treatment in the last few years, LKB1 mutant cancer patients continue to prove resistant to current standard-of-care including both immunotherapy [9] and KRAS inhibitors [10]. However, identifying targets for overcoming tumor suppressor loss-of-function presents a unique challenge.

Incomplete understanding of LKB1 mutant tumor biology has impeded the identification of successful strategies to treat LKB1 mutant cancer. Encouragingly, recent discoveries have identified new effectors of LKB1’s tumor suppressor function and illuminated vulnerabilities in the metabolism and microenvironment of LKB1 deficient tumors. In this Review, we will highlight these advances primarily from the past four years that have enhanced understanding of the LKB1 kinase pathway in tumor suppression. We emphasize LKB1-specific vulnerabilities in metabolism and the tumor microenvironment and discuss therapies which aim to exploit these distinct characteristics as potential clinical targets for LKB1 deficient tumors.

The LKB1 Kinase Pathway in Tumor Suppression

LKB1/STK11 is a serine/threonine kinase that directly phosphorylates and activates a discrete set of 14 downstream kinases, AMPKα1–2, SIK1–3, MARK1–4, BRSK1–2, NUAK1–2, SNRK, collectively referred to as the AMPK-Related kinases (AMPKRs) (Figure 1). LKB1 and its substrates have been reviewed elsewhere [1,11,12]. In brief, AMP-activated Protein Kinase (AMPK) is activated by LKB1 during energy stress to regulate metabolic pathways including the Mammalian Target of Rapamycin (mTOR) and fatty acid synthesis pathways [13]. The MAP/Microtubule Affinity-Regulating Kinases (MARK) are established regulators of cell polarity, the Salt-inducible kinases (SIKs) regulate gene expression programs, and SNF Related Kinase (SNRK) has emerged as a regulator of skeletal muscle contraction, glucose transport, and cell proliferation [14]. The BR Serine/Threonine Kinases (BRSKs) are primarily characterized as regulators of neuronal polarization [15], while the NUAKs are emerging as regulators of signal transduction, redox homeostasis, and cell motility [16]. Indeed, one study identified LKB1-NUAK signaling as an upstream regulator of Nuclear Factor-κB (NF-kB) activity in ovarian cancer [17]. Each AMPKR sub-family has its own biological functions and set of substrates, such that the signaling network downstream of LKB1 is extensive [14]. However, how LKB1 loss-of-function drives tumorigenesis in tissues where it is a causal tumor suppressor had until recently remained unclear.

Figure 1: The LKB1 kinase pathway and its functions.

LKB1 is a master kinase that phosphorylates and directly regulates a set of 14 discrete kinases, the AMPKRs. Orange (AMPK) and blue (SIKs) highlight AMPKRs which are the subject of new findings discussed herein.

Over 400 LKB1 mutations have been described to date in human disease, the majority of which were characterized as missense or nonsense mutations that result in an inactive catalytic domain [18–20]. The LKB1 protein is comprised of three domains: an N-terminal region, catalytic kinase domain, and C-terminal domain (CTD). Loss of function in the catalytic kinase domain of LKB1 is appreciated to disrupt its ability to suppress tumor cell growth [21]. Mutations that result in the loss of the N-terminal region [22] and mutations in the CTD [23] have also been reported, and functional outcomes were linked to cell polarity phenotypes [22,23]. While the impact of N- and CTD mutations remain incompletely understood, the mechanisms of tumor suppression downstream of LKB1 have recently been clarified.

Recent findings have provided insight into how LKB1 inactivation drives tumor growth. The bulk of these studies have taken place in Non-Small Cell Lung Cancer (NSCLC), where ~20% of cases harbor LKB1 mutation [20,24]. In vivo models have yielded key insight into how LKB1 functions as a tumor suppressor in the lung. Despite AMPK’s established role as a negative regulator of pro-growth metabolic pathways, AMPK was found to support lung tumor growth in vivo by regulating the TFE3 transcription factor and maintaining lysosomal capacity [25]. In contrast, the SIK1 and SIK3 kinases were found to mediate LKB1’s tumor suppressor function through regulation of IL6-JAK-STAT pathway gene expression [26,27], which has been implicated in the recruitment of immunosuppressive neutrophils [28]. Together, these studies have redirected focus toward exploring how the SIKs impart tumor suppression.

Elucidation of pathways dysregulated downstream of the SIKs in LKB1 deficient contexts has become an exciting area of study. The SIK substrates include two families of transcriptional coregulator proteins, the Class IIa Histone Deacetylases (HDACs) and the CREB Regulated Transcription Coactivators (CRTCs), which are excluded from the nucleus when phosphorylated by the SIKs [29]. CRTC-CREB activation has been identified as a critical factor in LKB1 deficient lung tumor growth in vitro [30,31] and in vivo [31]. Interestingly, in fibrolamellar carcinoma, a liver cancer characterized by fusion of the protein kinase A catalytic subunit PRKACA and heat shock protein DNAJB1 (DNAJB1-PRKACA), deregulation of CRTC2 and the histone acetyltransferase p300 as a result of SIK inactivation was shown to drive transcriptional reprogramming and histone acetylation [32]. In mouse embryonic fibroblasts and lung cancer cells, CRTC2-mediated histone acetylation downstream of LKB1 loss was reported to regulate inflammatory gene expression including Interleukin 6 (IL6) [33]. The LKB1-SIK axis has also recently been proposed to restrain KRAS-driven lung tumor growth through transcriptional maintenance of alveolar type II identity which may constrain lineage plasticity [34]. Our understanding of the impactful role of the SIKs in suppressing tumorigenesis downstream of LKB1 continues to expand, and it will be important to identify whether established or novel SIK substrates drive these molecular events.

New downstream substrates involved in LKB1-dependent tumor growth control have recently been discovered. To overcome well-appreciated limitations of 2D culture, an A549 cell spheroid culture system was developed where reintroduction of LKB1 recapitulated its function as a tumor suppressor [35]. A whole-genome CRISPR screen comparing LKB1-null (basal) and LKB1-competent (reconstituted) cells grown in 2D and spheroid conditions led to the discovery that LKB1 promoted the internalization of Epidermal Growth Factor Receptor (EGFR) through the Phosphoinositide Kinase, FYVE-Type Zinc Finger Containing (PIKFYVE) kinase to control tumor cell growth [35]. Separately, LKB1-dependent phosphorylation of the glycolytic enzyme triosephosphate isomerase 1 (TPI1) via the SIKs was reported to confer metabolic flexibility to promote p53 mutant human LUAD cell growth [36]. Another study identified activin A receptor type I (ALK2), which is a BMP6-responsive regulator of iron homeostasis, as a downstream effector in LKB1 mutant lung cancer which could be targeted with clinically viable ALK2 inhibitors [37]. These findings have significantly enhanced the understanding of how the LKB1 signaling pathway regulates disease, and continued efforts in this space are poised to uncover vulnerabilities that may prove actionable for treating LKB1 mutant cancer.

We posit that strategies aimed at drugging novel targets are most likely to succeed if LKB1-specific tumor biology is considered in the approach. Although most studies have focused on lung cancer, evidence continues to support a role for LKB1 in tumor suppression beyond the lung as well [38–45] and, ideally, lessons learned from the lung will provide utility in other tumor contexts. Intervention design would benefit from considering recent work, which has predominantly focused on metabolic vulnerabilities, the tumor microenvironment, and therapeutic resistance.

Metabolic Vulnerabilities in LKB1 Mutant Cancer

Since the central metabolic sensor AMPK is a direct substrate of LKB1, the metabolic state of LKB1 mutant tumors has been the subject of significant research focus (Figure 2). Recent studies along this vein have made progress defining the distinct metabolic profiles of LKB1 deficient tumors, and their sensitivity to redox stress.

Figure 2: LKB1 deficient tumors have distinct metabolic profiles.

Schematic summarizing recent findings discussed in this review. Red arrows indicate direction of change in response to LKB1 deficiency. Cartoon pills indicate where drugs under discussion intersect with the indicated pathways. A) Glucose metabolism. B) Mitochondrial metabolism. C) Reactive oxygen species. D) Combination therapies. Abbreviations: HDAC6i, HDAC6 inhibition; GLSi, Glutaminase inhibition; G6PDi, G6PD inhibition; 8-Cl-Ado, 8-chloro-adenosine; and MitoQ, mitoquinol.

LKB1 deficient tumors have altered metabolic profiles

LKB1 has an established role as a dominant regulator of multiple facets of cell metabolism including glucose, lipid, and protein metabolism as well as autophagy, lysosomal biogenesis, and mitochondrial homeostasis [46]. Studies have reported additional findings about glucose metabolism and mitochondrial function in LKB1 deficient tumors.

Enhanced glucose metabolism has repeatedly been identified as a hallmark of LKB1 deficiency, including in LKB1 mutant cancer [47]. Recent evidence suggests that the downregulation of lipid metabolism in favor of glycolysis may contribute to the immune-excluded phenotype which has been observed in KRAS, LKB1 (KL) mutant lung tumors [48]. Glycolysis thus continues to validate as a relevant target in LKB1 deficient tumors. Histone Deacetylase 6 (HDAC6) inhibition has been shown to impair the activity of glycolytic enzymes in triple-negative breast cancer [49], and this finding was also observed in the KL lung cancer genetically engineered mouse model (GEMM) [50]. Glutaminase inhibition impaired the production of glutamate, which is the precursor for metabolites including TCA cycle intermediates and glutathione [50]. Predicated on the finding that glycolysis may compensate for depleted TCA cycle intermediates and glutathione metabolism, glutaminase inhibition was found to enhance tumor killing in combination with HDAC6 inhibition in KL models [50]. Interestingly, in TRIM27-high glioblastoma, Tripartite Motif Containing 27 (TRIM27) was found to promote glycolysis through ubiquitination of LKB1 [51]. If enhanced glucose dependence of LKB1 mutant cancers can be leveraged for clinical benefit, these findings raise the implication that this dependence could extend to tumor types beyond those with genetic LKB1 mutation. While glucose metabolism has been studied extensively in the context of LKB1 deficient tumors, other metabolic alterations are also of interest.

LKB1 deficiency has been associated with reprogrammed mitochondrial metabolism in cancer [46]. Consistent with prior studies, cancer cell lines and patient-derived xenografts with impaired LKB1 expression and signaling were found to display sensitivity to genetic and pharmacological inhibition of mitochondrial Complex I, including the drugs metformin, rotenone, piericidin A, and IACS-010759 [52,53]. Furthermore, synergy between Complex I inhibitors and radiotherapy was found to induce ferroptosis, presenting another potential therapeutic strategy for LKB1 deficient cancers [52]. While mitochondrial alterations upon LKB1 impairment have been most thoroughly characterized in the context of lung cancer [54], mitochondrial liabilities have also been identified in other disease contexts. In uveal melanoma, LKB1 deficiency was shown to enhance mitochondrial Reactive Oxygen Species (ROS) and SLC8A1-mediated calcium uptake to promote tumor survival [55]. LKB1 deficient uveal melanoma correspondingly displayed vulnerability to co-treatment with the SLC8A1 inhibitor KB-R7943 and a mitochondria- targeted antioxidant, mitoquinol (MitoQ) [55]. Targeting mitochondrial metabolism continues to prove itself as a selective vulnerability of LKB1 deficient tumors, although how to utilize this characteristic for therapeutic benefit remains to be maximally exploited.

LKB1 deficient tumors are vulnerable to redox stress

Another established metabolic characteristic of LKB1 deficient tumors is the presence of elevated basal levels of ROS and reduced antioxidant levels [12,56–59]. Correspondingly, LKB1 deficient tumors often exhibit enhanced vulnerability to redox stress [7]. Oxidative stress was recently shown to promote adeno-to-squamous transition in KL lung cancer, suggesting that ROS accumulation in LKB1 deficient tumors may play an important role in regulating plasticity and lineage fidelity [50]. The increased energetic and redox stress observed in KL tumors reportedly left LKB1 mutant NSCLC cell lines vulnerable to disruption of redox homeostasis by the nucleoside analog 8-chloro-adenosine (8-Cl-Ado) [58]. Interestingly, KL mutant lung tumors were shown to rely on serine-glycine one-carbon metabolism (SGOC) [60]. SGOC produces the antioxidant glutathione through conversion of serine to glycine via Serine Hydroxymethyltransferase (SHMT) enzymes, and was shown to maintain antioxidant defense and redox homeostasis in KL tumors [60]. SHMT inhibition with the small-molecule SHIN2 further synergized with the ROS-inducing chemotherapy paclitaxel to inhibit KL mutant tumor growth in NSCLC xenografts [60]. KL lung tumors were also shown to be dependent on the oxidative pentose phosphate pathway enzyme Glucose-6-Phosphate Dehydrogenase (G6PD), which was attributed to their sensitivity to redox homeostasis [59,61]. These tumors displayed enhanced vulnerability to G6PD inhibition, which was shown to reprogram serine metabolism in KL mutant cells [61]. Targeting redox homeostasis through inhibition of SHMT or G6PD presents a novel therapeutic strategy which has shown promise in preclinical models, but further studies are needed to assess the translational value of these findings.

A significant subset of LKB1 mutant cancers co-occur with mutations inactivating Kelch Like ECH Associated Protein 1 (KEAP1) [6,62]. KEAP1 mutations activate Nuclear Factor Erythroid 2-Related Factor 2 (NRF2), which has been shown to enhance fitness of LKB1 deficient cancer cells under redox stress via induction of antioxidant expression [7]. As such, redox stress is an important facet of LKB1 mutant tumor biology.

LKB1 loss is associated with cachexia

Interestingly, LKB1 loss has recently been associated with cachexia, which is a systemic metabolic deregulation. Cachexia is characterized by muscle and adipose tissue loss resulting in unintended weight loss and is a major clinical concern in cancer treatment and a major cause of cancer-related deaths. Recently, LKB1 loss was identified as a potential driver of cachexia, where transplantation of LKB1 deficient human NSCLC and murine colorectal carcinoma lines was sufficient to confer a cachexic phenotype in murine hosts. An association was also identified between cancer-associated weight loss at diagnosis and LKB1 mutational status in NSCLC patients, further suggesting a role for LKB1 deficiency in promoting cachexia [63]. These findings suggest that LKB1 mutation alters both tumor intrinsic and systemic metabolism. We anticipate that continued study in this area will provide important contributions toward improving patient outcome. LKB1 deficiency drives distinct local and systemic metabolic states, but how this altered metabolism contributes to tumor growth in the context of the LKB1 deficient tumor microenvironment warrants consideration.

LKB1 deficiency produces an immunosuppressive tumor microenvironment

Recent findings have linked LKB1 deficiency to a distinct tumor microenvironment (TME) characterized by extracellular matrix (ECM) remodeling and an immunosuppressive microenvironment (Figure 3). This has spurred significant ongoing interest in and necessity for approaches that can overcome this.

Figure 3: Tumors with LKB1 deficiency have a distinct microenvironment characterized by ECM remodeling and an immunosuppressive TME.

Schematic summarizing recent findings. A) Compared to LKB1 proficient tumors (right), LKB1 deficient tumors (left) exhibit reduced T cell infiltration, enhanced abundance of immunosuppressive cells and inflammatory cytokines, and increased ECM deposition. B) Combination therapies that reverse LKB1-specific TME characteristics to enhance anti-tumor immune response. FAKi, FAK inhibition

LKB1 deficiency promotes ECM remodeling

Increased collagen deposition has been observed and associated with increased invasiveness in the LKB1 deficient TME across several tissue types including lung and breast cancer [64–66]. The increase in collagen deposition in response to LKB1 loss has been attributed to elevated lysyl oxidase (LOX) expression, a key enzyme in collagen cross-linking [65]. Recently, hyperactivation of focal adhesion kinase (FAK) in LKB1 deficient tumors was associated with increased collagen deposition [66]. FAK inhibition was reported to decrease cancer-associated fibroblast (CAF) activation and collagen deposition and increase infiltrating immune cells [66]. This synergized with Programmed cell death protein I (PD-1) blockade to slow primary and metastatic tumor growth in KL lung cancer GEMMs [66]. It is tempting to speculate that strategies targeting the TME could be enhanced in combination with approaches targeting collagen deposition.

The immunosuppressive TME in LKB1 mutant cancer

LKB1 mutant tumors are characterized as immunologically cold with low anti-tumor immune activity [28] associated with a paucity of immunosurveillance-related immune cells and elevated levels of immunosuppressive immune infiltrate.

Imaging and transcriptomic profiling of patient tumors has enabled characterization of the unique immune infiltrate in LKB1 mutant tumors. LKB1 mutant LUADs display decreased intratumoral densities of B cells [67] and T cells [9,67,68], downregulated antigen-processing molecule CD1E expression, and limited differentiation of anti-tumor M1 macrophages [69]. Lung tumors with LKB1 and KEAP1 co-mutation were also reported to exhibit low T cell infiltration [70]. Analysis conducted on a TCGA cohort of gastric cancer patients identified a positive association between LKB1 expression and CD8+ T cell infiltration, and a negative association between LKB1 expression and infiltration of neutrophils and M2 macrophages [71]. Collectively, these results suggest that LKB1 mutation promotes an immunosuppressive microenvironment in patient tumors.

Recent studies employing mouse models have provided insight into the effectors and regulators of the immunosuppressive TME in LKB1 mutant cancer. Transcriptomic and flow cytometry analysis of KL mutant murine lung tumors identified a decrease in B and T cell infiltration compared to KRAS and KRAS, p53 mutant tumors [28,66,72], confirming that patterns of immune infiltration observed in patient data are recapitulated in the KL GEMM. This decrease was phenocopied in a KRAS, AMPK mutant lung tumor model, suggesting that AMPK contributes to LKB1’s effect on immune infiltration [72].

Other studies using the KL GEMM found that tumor cell-intrinsic factors contribute to impaired T cell recruitment through repression of cytokine signaling. Aberrant Poly(ADP-ribose) Polymerase 1 (PARP1) activity upon LKB1 mutation blocked phosphorylation of Signal Transducer and Activator of Transcription 1 (STAT1) which impaired tumor cell-intrinsic type II interferon γ (IFNγ) signaling and T cell recruitment [73]. Inhibition of PARP1 with Olaparib mounted an adaptive immune response and synergized with PD-1 blockade to promote anti-tumor immune activity in the KL GEMM and orthotopic models of LKB1 mutant NSCLC [73]. Another study found that Histone Deacetylase 3 (HDAC3) deletion within tumor cells was sufficient to enhance T cell recruitment into KL GEMM lung tumors, which was attributed to de-repression of chemokine gene expression including the leukocyte chemoattractant Cxcl10 [74]. This response could also be elicited with systemic co-delivery of the HDAC1- and HDAC3-selective inhibitor entinostat together with the Mitogen-activated Protein Kinase I (MEK) inhibitor trametinib, both of which are clinically tolerated drugs [74]. Furthermore, targeting Major Histocompatibility Complex (MHC) class 1-related chain A or B was found to activate Natural Killer (NK) cell-mediated killing of KL lung cancer cells [75]. Thus, there may be yet-untapped tumor cell-intrinsic approaches for engaging and enhancing anti-tumor immunity.

Immunosuppressive cells such as tumor-associated neutrophils, polymorphonuclear myeloid-derived suppressor cells, and tumor-promoting macrophages have also been reported to be enriched in LKB1 deficient tumor models [28,66,76–78]. LKB1 deficient tumor cells have been shown to contribute to polarization and recruitment of immunosuppressive cell types through secretion of cytokines, chemokines, and metabolites. In endometrial epithelial cells, an LKB1-AMPK axis was shown to regulate production of C-C Motif Chemokine Ligand 2 (Ccl2) [76]. Knockout of LKB1 resulted in increased Ccl2 production, which correlated with increased tumor-promoting macrophage density [76]. KL lung cancer cells have repeatedly been shown to express higher levels of cytokines with established roles in immune suppression [79], including immunosuppressive neutrophil recruitment [28] and inflammation [28]. In LKB1 deficient LUAD, increased expression of the lactate transporter Monocarboxylate transporter 4 (MCT4) in tumor cells was shown to contribute to M2 polarization of macrophages [78]. Conditioned media from tumor cells was able to promote M2 polarization, suggesting a role for tumor cell-secreted lactate in mediating immunosuppression [78]. Means to reduce the immunosuppression and enhance the immunosurveillance of LKB1 deficient tumors may both represent exciting areas of target identification for LKB1 deficient tumors. While the optimal approaches for modulating the immune system for clinical benefit remain to be determined, this is an exciting area of study which may hold promise for overcoming resistance to standard-of-care therapies.

Overcoming Therapeutic Resistance and Clinical Trials in LKB1 Mutant Cancer

Clinically, therapeutic resistance remains an enormous challenge for patients with LKB1 mutant tumors, as evidenced by poor clinical outcomes including a mOS ranging from 4.7 – 26 months depending on the clinical cohort [4,8,9,80–87]. First- and second-line standard-of-care therapies for advanced NSCLC include immunotherapies (e.g. anti-PD-1/PD-L1), chemotherapies (e.g. cisplatin), and, for eligible patients with co-occurring KRAS mutations, targeted therapies (e.g. adagrasib/sotorasib). While these treatments were adopted based on net benefit across the clinical cohort tested, recent analyses have revealed that patients within these cohorts who harbor LKB1 mutant tumors experienced comparatively worse outcomes [4,80–82]. LKB1 deficiency, defined by mutational status and gene expression profile, was reported to drive resistance to anti-PD-1 immunotherapy in the clinic [9,88]. Patients with co-occurring KRAS and LKB1 mutations exhibit shorter overall survival and reduced response to anti-PD-1 immunotherapy compared to those with KRAS or KRAS, p53 mutant tumors [9,89]. KEAP1 mutations, with or without concurrent LKB1 mutation, were also reported to contribute to immunotherapy resistance [83–85]. LKB1 mutations were associated with reduced overall response rate and shorter overall survival in patients treated with chemotherapy [5,8] and chemoimmunotherapy [90]. Despite the fact that LKB1 loss was shown to increase sensitivity to inhibition of MEK, a downstream effector of KRAS signaling [91], LKB1 mutation was shown to be associated with poor prognosis in patients treated with KRAS^G12C inhibitors [10].

Considering the severity of illness in LKB1 mutant lung cancer patients, eliciting therapeutic efficacy in this patient population in clinical trials is a distinct challenge. However, we posit that the rigorous mechanistic insights into LKB1 mutant tumor biology, which stands on the foundation of the extensive research effort to date, is poised to uncover successful therapeutic approaches. Considering the new classes of therapeutics developed in recent years, characterizing therapeutic resistance in LKB1 mutant cancer could provide a promising path toward improved therapeutic options in the clinic, particularly if combination treatments are identified that can enhance standard-of-care efficacy.

Overcoming resistance to immunotherapy

Overcoming resistance to immune checkpoint inhibitors in LKB1 mutant tumors has been a major research focus (Table 1), and recent findings have explored approaches for modifying the TME to achieve this goal. One study found that all-trans retinoic acid reduced immunosuppressive myeloid-derived suppressor cells (MDSCs) and sensitized tumors to PD-1 blockade [92]. Another study reported that delivery of dendritic cells expressing C-C Motif Chemokine Ligand 21 (CCL21), a chemokine that recruits effector T cells and dendritic cells, depleted tumor-promoting neutrophils and synergized with PD-1 therapy [93]. Cyclin Dependent Kinase 4/6 (CDK4/6) inhibition through palbociclib was also reported to promote Intercellular Adhesion Molecule 1 (ICAM1)-driven T cell infiltration and sensitize LKB1 deficient lung cancer to immunotherapy [94]. Nanovesicles delivering Tumor Suppressor Candidate 2 (TUSC2), a tumor suppressor that has been reported to play a role in the activation of anti-tumor immune cells, were shown to boost immune activation and chemoimmunotherapy efficacy in humanized KL mouse models of NSCLC [95]. An additional strategy to overcome resistance to immunotherapy includes combining PD-1 inhibitors with anti-Cytotoxic T-Lymphocyte Antigen-4 (CTLA4) therapy, which was shown to increase CD4+ effector cells and inducible nitric oxide synthase (iNOS+) antigen presenting cells [83]. Moreover, the inhibition of Signal Transducer and Activator of Transcription 3 (STAT3) in the TME through an antisense oligonucleotide was reported to enhance the effects of dual checkpoint blockade [96]. Taken together, combination therapies are poised to provide a means for overcoming LKB1-specific resistance to immune checkpoint blockade (ICB). Therapeutic interventions that remodel the TME could enhance the efficacy of immunotherapies such as anti-PD-1.

Table 1:

Treatments that sensitize LKB1 mutant tumors to ICB

Treatment	Effect	Reference #
All-trans retinoic acid	Reduced accumulation of MDSCs	92
Delivery of CCL21-expresssing dendritic cells	Enhanced recruitment of effector T cells	93
Palbociclib (CDK4/6 inhibitor)	Enhanced infiltration of ICAM1-driven T cells	94
Nanovesicle delivery of TUSC2	Increased infiltration of cytotoxic T cells, depletion of MDSCs	95
Anti-CTLA4	Increase iNOS+ antigen presenting cells and CD4+ effector cells	83
STAT3 antisense oligonucleotide	Reverse immunosuppressive inhibition of CD8+ cells by myeloid derived cells	96
MRT68921 (ULK1 inhibitor)	Increased infiltration of CD4+ and CD8+ T cells and enhanced immunoproteasome activity	98
Decitabine and BAY-1217389 (MPS1 inhibitor)	Activation of STING and enhanced T cell mediated antitumor immunity	100
Birinapant (IAP inhibitor)	Activation of STING and increased infiltration of CD8+ T cells	101

Tumor cell-intrinsic approaches for overcoming resistance to anti-PD-1 therapies in LKB1 mutant tumors have also been identified. It has been shown that LKB1 mutant tumors often express PD-1/PD-L1 at lower levels than LKB1 proficient tumors in NSCLC patients [9]. It was reported that LKB1 plays a direct role in regulating PD-1/PD-L1 expression by interacting with the E3 ubiquitin ligase S-phase kinase-associated protein-2 (Skp2) to stabilize these proteins, and loss of LKB1 led to reduced PD-1/PD-L1 levels [97]. In KL mouse models of NSCLC, LKB1 deficiency was shown to increase autophagic flux, which suppressed antigen presentation. This effect was reversed by inhibition of Unc-51-like kinase 1 (ULK1), a kinase involved in regulating autophagy, and this inhibition resulted in enhanced T cell recruitment and PD-1 blockade efficacy [98]. KL tumors have also been reported to epigenetically silence Stimulator of Interferon Genes (STING) [99]. Another study showed that Monopolar Spindle 1 (MPS1), a kinase involved in the spindle assembly checkpoint, further inhibited the cGAS-STING-STAT1 pathway [100]. Combining decitabine, a pyrimidine nucleoside analog that functions as a chemotherapeutic agent, with the MPS1 inhibitor BAY-1217389 was shown to restore STING signaling, promote T cell infiltration, and improve efficacy of PD-1 inhibition [100]. LKB1 loss was also reported to disrupt Janus Kinase (JAK)-mediated STING expression via increased Inhibitor of Apoptosis (IAP) activity [101]. This increase can be reversed with the IAP inhibitor birinapant, which sensitized cells to immune clearance [101]. Stimulation of the type I IFN pathway through activation of STING or IFN-β continues to show promise to influence the TME and sensitize KL tumors to immunotherapy [102]. That combinatorial therapeutic strategies including immunotherapies show promise is encouraging, and continued discovery in this space is anticipated. However, continued efforts toward understanding of resistance mechanisms will be essential for toeing the line between therapeutic efficacy and toxicity.

Therapeutic resistance to targeted therapies

Distinct mechanisms of resistance to targeted therapies have been identified in LKB1 mutant tumors. Fibroblast Growth Factor Receptor I (FGFR1) was determined to be a key HDAC3 target gene that becomes hyperactivated selectively in LKB1 null lung cancer cells upon resistance to the MEK inhibitor trametinib [103]. Leveraging this understanding, combining trametinib with the HDAC1/3 inhibitor entinostat was found to significantly reduce KL tumor growth [103]. Autophagy has also been implicated in trametinib resistance, and combining trametinib with hydroxychloroquine enhanced antiproliferative activity [104]. ULK1/2 inhibition was also shown to synergize with the KRAS^G12C inhibitor sotorasib in KRAS^G12C mutant lung cancer, suggesting that ULK-driven autophagy may be an actionable target in KRAS driven cancers regardless of LKB1 status [105]. Resistance to KRAS inhibitors in LKB1 mutant tumors has been linked to adeno-to-squamous transdifferentiation [106]. Co-mutation of LKB1 and KEAP1 was also reported to correlate with poorer clinical outcomes to adagrasib [10]. Encouragingly, this study also found that combining adagrasib with mTOR inhibition enhanced treatment efficacy in preclinical models [10]. Finally, LKB1 mutation conferred radiotherapy resistance via NRF2 pathway activation and maintenance of redox homeostasis [107]. Correspondingly, the use of glutaminase 1 inhibitors sensitized tumors to radiotherapy through impairment of redox homeostasis [107]. Although still early-days with regard to development of combination therapy approaches, the recent emergence of multiple distinct approaches for LKB1 mutant lung cancer is exciting. Clinical data has repeatedly shown that the LKB1 mutant patient population responds distinctly to treatment and, therefore, it would be wise for future trial design to consider this patient population as a distinct indication.

Clinical trials for LKB1 mutant cancer

Encouragingly, improved understanding of LKB1 mutant tumor biology and mechanisms of therapeutic resistance has led to clinical trials designed to treat cancer patients with LKB1 mutant tumors (Table 2). Several trials where STK11 mutation was a criterion for eligibility have already been completed. Consistent with preclinical evidence that glucose deprivation or caloric restriction can synergize with metformin to restrain LKB1 deficient lung cancer [108], a clinical trial tested standard-of-care platinum-based chemoimmunotherapy in combination with metformin plus/minus fasting-mimicking diet (FMD) in LKB1 inactive LUAD (NCT03709147)^I. While the trial was closed due to low recruitment, the mOS of patients in the arm receiving chemoimmunotherapy, metformin and FMD was double the mOS of patients in the observational arm [109]. Another clinical trial tested glutaminase I inhibition (telaglenastat) in combination with standard-of-care immunochemotherapy in patients with non-squamous NSCLC tumors harboring LKB1, KEAP1 mutations (NCT04265534)^II. This trial was terminated due to lack of clinical benefit, as no improvement in progression free survival was observed in these patients. However, a different ongoing clinical trial is testing the efficacy of telaglenastat in solid tumors, including lung, with LKB1 mutations (NCT03872427)^III. Following promising preclinical evidence supporting PARP inhibition in combination with PD-1 blockade, the efficacy of the PARP inhibitor talazoparib in combination with the PD-1 inhibitor avelumab was also recently tested in patients with LKB1 mutant lung tumors that were refractory to anti-PD-1 therapy (NCT04173507)^IV. Results did not identify efficacy in patients with LKB1 mutant tumors [110]. Similarly, durvalumab, anti-PD-L1 monoclonal antibody, in combination with the PARP inhibitor olaparib failed to achieve clinical benefit for LKB1 mutant NSCLC patients with tumors refractory to standard-of-care therapies (NCT03334617)^V [111]. Despite the lack of clinical benefit reported by these clinical trials, other ongoing trials provide optimism for the LKB1 mutant cancer patient population.

Table 2:

Clinical trials in LKB1/STK11 mutant cancer

ClinicalTrials.gov ID	Drug(s)	Target(s)	Cancer	Status
NCT03709147	Metformin +/− fasting mimicking diet with standard-of-care chemoimmunotherapy	LKB1-specific metabolic vulnerabilities, SOC	LUAD	Terminated due to low recruitment
NCT04265534	Telaglenastat with standard-of-care chemoimmunotherapy	GLS1, SOC	Non-squamous NSCLC	Terminated due to lack of clinical benefit
NCT03872427	Telaglenastat hydrochloride (CB-839 HCl)	GLS1	Solid tumors or malignant peripheral nerve sheath tumors	Active
NCT03334617	Durvalumab and olaparib	PD-1, PARP	NSCLC	Active, results available
NCT04173507	Talazoparib with avelumab	PARP, PD-L1	Non-squamous NSCLC	Completed: Did not meet the threshold for efficacy
NCT05445843	JDQ443	KRASG12C	NSCLC	Active
NCT05276726	JAB-21822	KRASG12C	NSCLC	Active
NCT06495125	Defactinib and avutometinib with nivolumab	FAK, RAK/MEK, PD-1	NSCLC	Recruiting
NCT05469178	Bemcentinib with standard-of-care immunotherapy	AXL, SOC	NSCLC	Terminated
NCT06219174	Difluoromethylornithine with pembrolizumab	ODC, PD-1	NSCLC	Recruiting
NCT05887492	TNG260 with pembrolizumab	CoREST, PD-1	Solid tumors	Recruiting
NCT05807048	Daratumumab	CD38	NSCLC	Recruiting
NCT05704634	Sarilumab with cemiplimab	IL-6R, PD-1	NSCLC	Active
NCT06331650	Carbognilumab with chemotherapy	CTLA-4, PD-1, Chemo	NSCLC	Recruiting
NCT06008093	Durvalumab and tremelimumab with chemotherapy	CTLA-4, PD-1, Chemo	Non-squamous NSCLC	Recruiting
NCT06335355	Adebrelimab and SHR-8068 with chemotherapy	CTLA-4, PD-L1, Chemo	NSCLC	Not yet recruiting
NCT06124963	WX390 and Toripalimab	PI3K, mTOR, PD-1	Gastric-type Endocervical Adenocarcinoma	Recruiting

SOC indicates Standard-of-Care when not detailed in trial description.

Chemo indicates chemotherapy when not further detailed in trial description.

Notably, there are now multiple active clinical trials in lung cancer patients with tumors harboring LKB1 and KRAS^G12C co-mutation. One of these trials is testing the use of single agent JDQ443 (KRAS^G12C inhibitor) as first-line therapy in patients with KRAS^G12C and LKB1 co-mutation (NCT05445843)^VI. Another clinical trial in patients with advanced NSCLC with KEAP1 wild-type tumors harboring KRAS^G12C and LKB1 co-mutation will test the clinical activity of the KRAS^G12C inhibitor JAB-21822 (NCT05276726)^VII. Whether KRAS inhibitors will be able to replace ICB as a safer and more effective first-line option for LKB1 mutant cancer patients remains to be determined.

Several other active clinical trials are investigating combination therapy approaches for enhancing efficacy of PD-1 inhibitors. One such trial aims to utilize the FAK inhibitor defactinib and the RAF/MEK inhibitor avutometinib in combination with the PD-1 inhibitor nivolumab in patients with anti-PD-1-refractory LKB1 mutant tumors (NCT06495125)^VIII. Inhibition of IL-6 was reported to enhance the efficacy of PD-1 inhibitors [112], and an ongoing clinical trial is testing the combination of the IL-6 receptor (IL-6R) antibody sarilumab with the anti-PD-1 antibody cemiplimab in patients with LKB1 mutant NSCLC (NCT05704634)^IX. Due to reports that AXL Receptor Tyrosine Kinase (AXL) inhibition synergizes with anti-PD-1 therapy [113], one clinical trial is testing the efficacy of the AXL kinase inhibitor bemcentinib in combination with standard-of-care chemoimmunotherapy in patients with advanced LKB1 mutant lung cancer (NCT05469178)^X. A different trial will inhibit ornithine decarboxylase (ODC) with the compound difluoromethylornithine in combination with the PD-1 inhibitor pembrolizumab (NCT06219174)^XII. Another trial will combine a novel compound, TNG260, which targets the Co-repressor of Repressor Element-1 Silencing Transcription (CoREST) deacetylase complex, with the anti-PD-1 therapy pembrolizumab in patients with LKB1 mutant solid tumors (NCT05887492)^XII. Based on findings that CD38 inhibition can overcome therapeutic resistance to PD-1 inhibitors in vivo [114], a clinical trial to test this is actively recruiting patients. This trial will test the efficacy of single agent daratumumab, an antibody that targets the CD38 protein, in previously treated LKB1 mutant NSCLC patients (NCT05807048)^XIII. The breadth of combination approaches being tested in clinical trials as a means to overcome PD-1 inhibitor resistance represents the outcome of significant research focus since the initial clinical observation that LKB1 mutant patients fare poorly on ICB. It is exciting that many combination approaches are in clinical trials, and hopefully this will translate to improved patient outcomes in the near future.

There is also evidence that a dual ICB approach can overcome PD-1 inhibitor resistance in LKB1 mutant tumors [83], and means to maximize utility of these approaches continues to be explored. One clinical trial will test the efficacy of the bi-specific antibody carbognilumab (cadonilimab) targeting both CTLA4 and PD-1 in combination with chemotherapy as a first-line treatment in advanced LKB1 mutant NSCLC patients (NCT06331650)^XIV. Similarly, two other clinical trials in advanced NSCLC patients harboring mutations in LKB1, KRAS, KEAP1, or a combination thereof will evaluate the use of dual ICB, anti-CTLA4 and anti-PD-1/PD-L1, in combination with chemotherapy (NCT06008093)^XV (NCT06335355)^XVI. With safety as an important parameter to consider with dual ICB, it will be interesting to learn whether alternating between different combination approaches can extend patient survival.

Although many of the clinical trials are designed for lung cancer patients with LKB1 mutations, therapies for other LKB1 mutant cancer types are also being tested in the clinic. One clinical trial in LKB1 mutant patients with advanced gastric-type endocervical adenocarcinoma will test the efficacy of WX390, a dual inhibitor of Phosphoinositide 3-Kinase (PI3K) and mTOR, in combination with Toripalimab, an anti-PD-1 therapy (NCT06124963)^XVII. Furthermore, two additional studies are recruiting patients with LKB1 mutations across solid tumors (NCT03872427)^III (NCT05887492)^X. These developments represent an exciting step toward leveraging knowledge about LKB1 mutant lung cancer for the benefit of other cancer types. Despite the ongoing challenge of therapeutic resistance in LKB1 mutant tumors, recent improvements in understanding have led to multiple new clinical trials, which provide hope that more effective treatment options for this patient population may be on the horizon.

Concluding Remarks and Future Perspectives

While LKB1 mutant cancer remains a devastating disease, significant progress has been made in recent years, some of which is in the clinic or is poised for clinical translation. The LKB1 mutant cancer population represents a well-defined indication and a large patient population, such that promising treatment approaches should find footing in the transition from bench to bedside. Improved understanding about which arms of the LKB1 signaling pathway drive tumor suppression has been instrumental in moving the field forward, and LKB1-specific cellular metabolism continues to validate as a central tenet of tumor biology. The immunosuppressive nature of LKB1 mutant tumors represents a major leap forward in the understanding of this tumor type, and the clinical finding that this patient population fails to respond to immunotherapy highlights the necessity for continued research in this area. The field now stands on a solid foundational understanding about LKB1 in cancer, and the outline of a unified understanding is beginning to take shape. It is an exciting time for the field, with acquired momentum accelerating discovery; it will be essential to maintain that momentum.

Outstanding questions:

• Which molecular effectors downstream of the SIKs drive LKB1 mutant lung cancer biology?

• What are the molecular mechanisms responsible for the immunosuppressive microenvironment in LKB1 mutant cancer?

• Which combination therapy approaches will succeed in extending patient survival?

• How does LKB1 loss drive cachexia, and can this be reversed?

• Which tumor types or sub-types can benefit from strategies targeting vulnerabilities of LKB1 deficiency?

Highlights:

1. LKB1 is a tumor suppressor that directly phosphorylates and activates a set of 14 discrete kinases to regulate cellular characteristics. The SIK kinases have recently been identified as key mediators of LKB1’s tumor suppressive activity, but the understanding of the substrates involved in tumor growth control downstream of LKB1 remains incomplete.

2. LKB1 deficient tumors have distinct metabolic profiles introducing LKB1-specific metabolic liabilities which may be leveraged as potential therapeutic targets.

3. LKB1 deficiency promotes an immunosuppressive microenvironment.

4. Therapeutic resistance presents a challenge in treating LKB1 deficient tumors. Improved understanding of the unique biology of LKB1 deficient tumors has resulted in novel therapeutic candidates which have and continue to be explored in clinical trials.