Research

The endoplasmic reticulum (ER) is the major site for protein synthesis and folding. Tumor cells grow rapidly and develop imbalances in nutrition and oxygen supplies, which can lead to perturbation in protein folding in the ER. The adverse microenvironment and accumulation of improperly folded proteins in the ER of cancer cells induces ER stress, which in turn triggers the unfolded protein response (UPR) as an adaptive mechanism to restore ER proteostasis. As this signaling pathway is differentially regulated during tumor growth and progression, modulating UPR signaling is a promising therapeutic strategy for cancer.

In mammalian cells, three major ER transmembrane proteins operate as ER stress sensors: inositol-requiring enzyme 1α (IRE1α), activating transcription factor 6 (ATF6) and PRKR-like ER kinase (PERK). Our lab focuses on the role of IRE1α signaling in regulating cancer progression. IRE1α, an evolutionary conserved ER stress sensor, consists of two enzymatic activities: a serine/threonine kinase and an RNase activity. When misfolded proteins accumulate in the ER, IRE1α dimerizes/oligomerizes and trans-auto phosphorylates, thereby activating its RNases activity (Figure 1). The RNase activity consists of: (i) The non-conventional splicing of the mRNA of XBP1 to allow production of XBP1s (XBP1 spliced), a transcription factor and (ii) RNase-mediated cleavage of target mRNAs, referred to as regulated IRE1-dependent decay (RIDD) of RNA.

Our group was the first to identify a critical role for the IRE1α-XBP1 branch of the UPR in cancer. We showed that XBP1 mediated survival under hypoxia and was essential for tumor growth (Romero-Ramirez et al, Cancer Res 2004). Since demonstrating the therapeutic potential of inhibiting this target, we have focused on developing pharmacologic inhibitors of this pathway. We completed a high-throughput screening strategy and completed a screen of >120,000 compounds.

The Koong Lab was the first to identify a class of compounds that specifically and potently inhibit IRE1α in cell lines, tumor xenografts, and patient-derived specimens (Papandreou et al, Blood 2011). This study demonstrated the feasibility of targeting this pathway pharmacologically, leading to multiple pre-clinical and clinical development paradigms. Since then, we completed a whole genome siRNA screen to identify other genes that regulate this pathway with the overall goal of finding other therapeutic targets in cancer (Yang et al, Molecular Cancer Research 2018).

Ongoing work in the laboratory is focused on developing novel classes of compounds to block the activity of IRE1α and other UPR targets as a therapeutic strategy. We have developed these targeted therapies alone and in combination with radiation therapy in relevant disease models to optimize translation into the clinic.

Ferroptosis, a type of iron-dependent regulated cell death caused by excessive lipid peroxidation within the cell (Lei et al, Nat Rev Cancer 2022), was recently identified to mediate tumor response to radiation therapy (RT). With the support of a U54 grant from the NCI Acquired Resistance to Therapy Network (ARTNet), we focus on elucidating the role of tumor hypoxia in inducing acquired resistance to RT-induced ferroptosis in lung and esophageal cancers and investigate the role of various hypoxia-induced signaling pathways in ferroptosis resistance. We derived tumor cell lines with acquired radioresistance through both in vitro and in vivo approaches (figure on the right). We also developed a syngeneic orthotopic lung cancer model based on a cell line derived from the Kras^G12D/+;p53^R172HΔG/+ (KP^ΔG) model to test tumor response to RT alone or in combination with other treatment modalities. 

The NCI ARTNet focuses on the mechanistic basis of acquired resistance to cancer therapies and disease recurrence. Being a part of this network enables us to utilize the broad-spectrum of expertise and cutting-edge technology platforms among the member centers across the other five ARTNet centers in the country.

Our central focus on ferroptosis has been extended to a collaboration with the Pancreatic Cancer ARTNet Center (University of Oklahoma Health Sciences Center).

Check out our recent publication in Nature Communications, "IRE1α determines ferroptosis sensitivity through regulation of glutathione synthesis."

Pancreatic cancer (mainly pancreatic ductal adenocarcinoma or PDAC) is currently the third-leading cause of cancer-related death in the United States. Despite aggressive combined modality therapies, the five-year overall average survival across all stages remains ~10%. Therapeutic paradigms are hampered by the inability to detect pancreatic cancer until late stages. In addition, pancreatic cancer is relatively resistant to conventional chemotherapy and radiation therapy and to date, no effective targeted therapies or immunotherapies have been approved for clinical application. The Koong Lab focuses on understanding how PDAC develops and designing therapeutic approaches that target molecular pathways mediated by intrinsic and extrinsic cues to improve the treatment outcome of PDAC.

Patients with locally advanced pancreatic cancer (LAPC), that is neither metastatic nor resectable, account for roughly 30-40% of all cases of PDAC. Chemotherapy is modestly effective, but often fails to control the primary tumor and local progression often directly leads to death. For non-resectable PDAC, radiation is the only tested option to achieve local control. Definitive chemoradiation to PDAC is challenging because the nearby stomach and duodenum cannot tolerate high doses of radiation. Therefore, the dose given to most PDACs (usually 50.4Gy or less) does not cause overt toxicity, but also does not control the tumor. A highly conformal technique called stereotactic body radiation therapy (SBRT) uses advanced image guidance to deliver higher doses of radiation to the tumor and is an acceptable first-line treatment option for certain patients with LAPC according to NCCN (v2.2024) and ASTRO guidelines. Unfortunately, dose escalation in SBRT has been difficult due to duodenal toxicity. Consequently, low-dose SBRT has also failed to improve survival in LAPC. Therefore, combination approaches with other therapeutic modalities to improve the treatment efficacy of SBRT are in urgent need.

We utilize the state-of-the-art small animal irradiation system SARRP (by Xstrahl, figure) to perform SBRT on orthotopic mouse PDAC tumors. With the use of tumor cell line derived from the Kras^LSL-G12D/+;Trp53^R172H/+;Pdx1-Cre (KPC, or the other variants) PDAC model of C57BL6 background, we can also evaluate the therapeutic effect of SBRT alone or in combination with other treatment modalities as a syngeneic model with an intact immune system.

Pancreatic ductal adenocarcinoma (PDAC) develops and proliferates under very low oxygen conditions, or hypoxia (Koong et al., Int J Radiat Oncol Biol Phys 2000). These conditions occur within tumors which may have an oxygen concentration of 0.5% or less (for reference, room air contains 21% oxygen).

This intratumoral hypoxia is at least partially due to the fibrotic and desmoplastic stroma (Stylianopoulos et al., Proc Natl Acad Sci USA 2012) that increases intratumoral pressure (Chauhan et al., Cancer Cell 2014), which is thought to block blood flow and the delivery of oxygen, nutrients and even therapeutic drugs (Koay et al., JCI 2014). This property makes the stroma a compelling therapeutic target, with the reasoning that reducing the physical barrier imposed by this desmoplasia would allow therapy to become more effective.

Unfortunately, Phase III clinical studies to disrupt pancreatic cancer stroma by inhibiting Sonic hedgehog protein (De Jesus-Acosta et al., Br J Cancer 2020) or infusion with a recombinant pegylated human hyaluronidase failed to meet survival endpoints (Van Cutsem et al., J Clin Oncol 2020), despite promising Phase II results (Hingorani et al., J Clin Oncol 2018). These clinical data stimulated the pancreatic cancer research field to consider functional inhibition of the stromal elements as a therapeutic strategy. Since hypoxia is a universal feature of pancreatic cancer, we reasoned that hypoxia plays a critical role in cellular homeostasis of the PDAC tumor microenvironment. We focused our efforts on how low oxygen levels may enhance pro-tumor signaling between fibroblasts and the immune system.

We developed a novel dual recombinase system with collaborations with the Kirsch Lab at the Princess Margaret Cancer Centre and the Saur Lab at Technical University of Munich (TUM) and German Cancer Research Center (DKFZ).

For more information on this system, refer to our reviews (Fuentes et al., Cancer Lett 2020; Huang et al., Sci China Life Sci 2017). Ongoing projects in our laboratory indicate that hypoxia signaling within PDAC-associated fibroblasts promotes the immunosuppressive M2 polarization of macrophages.

These newly formed M2 macrophages suppress anti-tumor T cell responses, which leads to favorable conditions for tumor growth.This effect was HIF-2a (HIF2) dependent and could be reversed with a small molecule inhibitor of HIF-2a called PT2399, a preclinical version of belzutifan, an FDA-approved HIF-2a inhibitor for use in VHL-mutant renal cell carcinoma (Garcia et al., Gastroenterology 2022). We are actively working to identify the paracrine factor(s) mediating this polarization effect on macrophages.

We envision that by using our combination of in vivo mouse modeling and in vitro cell/organoid culture systems we will determine the role of hypoxia in various cell types within the PDAC tumor microenvironment, with the goal to identify potentially druggable targets, such as the fibroblast-macrophage signaling axis.