KRAS mutations are high prevalence oncogenic drivers for multiple cancers. With the advent of new classes of KRAS inhibitors that are showing meaningful clinical activity, research is now turning to questions of optimal combinations of therapies for specific indications, as many patients with KRAS G12C mutations do not respond and/or develop resistance to single-agent treatment. Here, we investigate combination therapies that may overcome resistance to KRAS G12C inhibitors. We found that pemigatinib, a potent and selective FGFR1-3 inhibitor, had a significantly high Bliss synergy score in combination with KRAS G12C inhibitors, and FGFR1 activity was shown to decrease KRAS G12C-dependency conferring inherent resistance in mesenchymal-like cell lines. Knockdown experiments verified the importance of FGFR1, but not FGFR2-4, for the synergistic effect with KRAS G12C inhibitors. Additionally, human lung cancer xenograft and patient-derived xenograft models with a mesenchymal phenotype and high FGFR1 expression were sensitive to the combination of G12C inhibitors and pemigatinib. In short, we demonstrate that pemigatinib and KRAS G12C inhibitors are promising agents for combination therapy in non-small cell lung cancer with a mesenchymal-like phenotype harboring high FGFR1 expression and KRAS G12C mutations to broaden patient response.

CDK2 is a critical regulator of the cell cycle. For a variety of human cancers, the dysregulation of CDK2/cyclin E1 can lead to tumor growth and proliferation. Historically, early efforts to develop CDK2 inhibitors with clinical applications proved unsuccessful due to challenges in achieving selectivity over off-target CDK isoforms with associated toxicity. In this report, we describe the discovery of (4-pyrazolyl)-2-aminopyrimidines as a potent class of CDK2 inhibitors that display selectivity over CDKs 1, 4, 6, 7, and 9. SAR studies led to the identification of compound 17, a kinase selective and highly potent CDK2 inhibitor (IC50 = 0.29 nM). The evaluation of 17 in CCNE1-amplified mouse models shows the pharmacodynamic inhibition of CDK2, measured by reduced Rb phosphorylation, and antitumor activity.

Transforming growth factor-β (TGFβ) signaling is common in many solid tumors and is initiated by binding of the high affinity canonical ligands TGFβ1, 2, οr 3 to TGFβR2, which forms a heteromeric receptor complex with TGFβR1 (Derynck et al, Nature Review Clinical Oncology 2020). Activation of the pathway results in potent suppression of immune cell-mediated anti-tumor immunity and has been reported to predict poor response to PD-(L)1 targeted therapy in patients (Mariathasan et al, Nature 2018; Kieffer et al, Cancer Discovery 2020). However, TGFβ drug development has been hampered by the occurrence of adverse events. INCA33890 is a dual PD-1 and TGFβR2 binding bispecific Biclonics® antibody, developed to antagonize the TGFβ signaling pathway specifically in cells co-expressing PD-1 and TGFBR2. Additionally, it potently antagonizes the PD-1 axis independently of TGFβR2 co-expression. The cell-selective action of INCA33890 was designed to mitigate risks of the known adverse effects associated with TGFβ-pathway inhibition in tissues requiring active TGFβ signaling. ΙΝCΑ33890 mediates its specificity through a PD-1 binding arm with a >10-fold higher affinity relative to the TGFβR2 binding arm. Consistent with this profile, in isogenic Jurkat cells expressing TGFβR2 ± PD-1, INCA33890 potently inhibits TGFβ1-induced pSMAD activation in a PD-1-correlated manner. Additionally, in two independent PD-1 reporter assays, INCA33890 inhibited SHP recruitment and enhanced NFAT activation with a potency within an order of magnitude to that of pembrolizumab. In mixed lymphocyte reaction assays with exhausted primary human T-cells, INCA33890 was found to induce a similar level of anti-tumor cytokine production as the combination of pembrolizumab and an anti-TGFβR2 antagonist mAb. Treatment of primary ovarian ascites with INCA33890 ex vivo induced IFNγ production in all donors tested, while pembrolizumab had no activity. Similarly, in human CD34+ cell-engrafted NSG mice, INCA33890 significantly inhibited the growth of human MDA-MB-231 and A375 subcutaneous xenograft tumors, whereas pembrolizumab or an anti-TGFβR2 antibody had little or no monotherapy activity. INCA33890 had a balanced pharmacokinetic and potency profile and was well tolerated in NHPs at exposures required for pharmacodynamic and tumor growth inhibiting activity in rodents. Encouragingly, there was no evidence of adverse effects in NHPs due to TGFβ-pathway blockade. Collectively, these results provide compelling data for an effective and specific approach to simultaneously antagonizing TGFβ and PD-1 signaling in tumors. Clinical development of INCA33890 in checkpoint inhibitor-resistant and other cancers has been initiated. Citation Format: Liang-Chuan S. Wang, Rinse Klooster, Ashwini Kulkarni, Amaya Garcia de Vinuesa, Maxim Soloviev, Linda JA Hendriks, Lu Huo, Michael Weber, Arpita Mondal, Yonghong Zhao, Shane Harvey, Xin He, Hong Chang, April Horsey, Alla Volgina, Yue Zhang, Veethika Pandey, Yan-Ou Yang, Jonathan Rios-Doria, Evgeniy Eruslanov, Daniel J. Powell, Steven M. Albelda, John de Kruif, Horacio Nastri, Cecile Geuijen, Patrick A. Mayes. INCA33890, a novel PD-1×TGFꞵR2 bispecific antibody conditionally antagonizes TGFꞵ signaling in primary immune cells co-expressing PD-1 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2936.

KRAS is one of the most frequently mutated oncogenes. Clinical studies with recently developed covalent KRASG12C inhibitors have shown promising anticancer activity in patients with KRASG12C tumors (substitution of glycine to cysteine at amino acid 12). Not all KRASG12C patients respond to single-agent treatment or, despite initial responses, develop drug resistance. Here, we investigate therapeutic options to overcome resistance to KRASG12C inhibitor therapy in non-small cell lung cancer (NSCLC). Cancer cell lines and in vivo tumor xenograft models representing high fibroblast growth factor receptor-1 expression (FGFR1hi) and KRASG12C tumors were treated with FGFR or KRASG12C inhibitors. The impact of single-agent versus combination treatment was measured in cell signaling, functional, and in vivo tumor growth assays. Analysis of a genome-wide genetic screen and corresponding mutation and expression data from DepMap identified a subset of NSCLC cells that harbor high FGFR1 expression (FGFR1hi) and KRASG12C mutations. FGFR1hi cancer cells exhibited mesenchymal-like features including high levels of vimentin and low levels of E-cadherin. To assess the functional role of high FGFR1 expression in KRASG12C-mutant cancer cells, pemigatinib, a potent and selective inhibitor of FGFR1-3, was tested alone or in combination with KRASG12C inhibitors. The combination of pemigatinib and KRASG12C inhibitors was synergistic in mesenchymal-like lung cancer cells with high FGFR1 expression, whereas no synergy was observed in cells with low FGFR1 expression. Furthermore, inhibition of FGFR1 activity was essential as a FGFR2-3 specific inhibitor demonstrated only modest activity in combination with KRASG12C inhibitors. Notably, treatment of FGFR1hi KRASG12C LU99 cells with covalent KRASG12C inhibitors resulted in an increase in FRS2 phosphorylation, a marker of FGFR pathway activation, which was suppressed by pemigatinib. To determine whether increased FGFR1 activity may be an acquired resistance mechanism, KRASG12C-mutant Mia-Paca-2 clones resistant to KRASG12C inhibitors were generated. Subsequent protein analysis identified high levels of FGFR1 expression in a subset of resistant clones. In vivo studies with mesenchymal KRASG12C-mutant xenografts confirmed increased antitumor efficacy and inhibition of pERK with the combination of KRASG12C inhibitors and pemigatinib, compared with single-agent treatment. In contrast, in vivo combination activity was not observed in NSCLC tumors possessing an epithelial-like phenotype. We demonstrate that NSCLC with a mesenchymal-like phenotype and harboring high FGFR1 expression and KRASG12C mutations may uniquely benefit from combination treatment with current KRASG12C-covalent inhibitors and blockade of FGFR1-mediated activity. Our results support pemigatinib as a promising agent for combination therapy with KRAS inhibitors. Citation Format: Margaret Favata, Michael Weber, Angela Abdollahi, Valerie Dostalik Roman, Matt Farren, Aidan Gilmartin, Sunkyu Kim, Susan Wee, Jonathan Rios-Doria. Pemigatinib, an FGFR inhibitor, overcomes resistance to KRASG12C inhibitors in mesenchymal-like NSCLC tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 430.

Abstract The primary cause of cancer associated mortality is tumor metastasis. The concept of the tumor pre-metastatic niche is supported by evidence of changes at distal pre-metastatic sites that create a permissive environment to allow disseminated tumor cells to seed. Myeloid-derived suppressor cells (MDSCs) remodel the tumor microenvironment and function as immunosuppressive cells to promote tumor growth. Previously, we demonstrated that the clinical stage LSD1 specific inhibitor, INCB059872 significantly reshaped the myeloid compartment in the murine 4T1 syngeneic murine model of breast cancer. Treatment with INCB059872 significantly reduced the population of MDSCs in the tumor microenvironment. Since it has been reported that MDSCs promote establishment of a pre-metastatic niche, we hypothesized that INCB059872 could suppress or delay metastatic processes in the 4T1 model and thereby could impact spontaneous metastases to the lung. In vitro, INCB059872 significantly suppressed cancer cell migration of triple negative breast cancer cells, SUM145PT. In vivo, the effect of INCB059872 on forming the metastatic niche using the 4T1 mouse breast tumor model was explored. Vehicle treated animals exhibited a significant infiltration of MDSCs to the primary tumor and lungs prior to cancer cells metastasizing. In contrast, INCB059872 administration significantly suppressed the infiltration of MDSCs in primary tumor and lung tissues. Histological analyses further demonstrated the reduction of metastatic loci in lung with INCB059872 treatment. Plasma levels of CCL2, a cytokine which is required for the recruitment and functional specialization of MDSCs, were significantly reduced in animals treated with INCB059872. These data suggest a possible mechanism to reduce infiltration of MDSCs into lung tissues. Notably, analyses of molecular pathways using RNA-Seq identified that components of the EMT associated pathway are also downregulated in tumors treated with INCB059872, which further supports the role of INCB059872 in the inhibition of metastasis. Taken together, these preclinical data suggest that inhibition of LSD1 with INCB059872 can suppress metastasis through multiple molecular and cellular mechanisms, notably by inhibition of the formation of the pre-metastatic niche by modulating the population of MDSCs in the primary tumor and distal tissues. Citation Format: Sang Hyun Lee, Melody Diamond, Antony Chadderton, Huiqing Liu, Alla Volgina, Valerie Roman, Michael Weber, Chunhong He, Rebecca Stewart, Denise Hertel, Phillip Liu, Liangxing Wu, Julian Oliver, Swamy Yeleswaram, Alan Roberts, Wenqing Yao, Gregory Hollis, Reid Huber, Peggy Scherle, Bruce Ruggeri. The FAD-directed LSD1 specific inhibitor, INCB059872, inhibits cell migration and metastasis by suppressing premetastatic niche formation in a spontaneous metastasis mouse model [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3929.

T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive hematological tumor that is derived from the clonal expansion of immature T-cell progenitors. Multiple genetic and epigenetic alterations are attributed to the development of malignant T cell transformation. Among these, there is supporting evidence for a role of lysine specific demethylase (LSD1) in T-ALL. Oncogenic transcription factors, such as TAL-1, Notch, and ZEB2, form a complex with LSD1 to alter gene expression in T-ALL cells. In addition, LSD1 is aberrantly expressed in ALL, including B-ALL and T-ALL. Furthermore, the overexpression of LSD1 under control of the Sca-1 promoter in transgenic mice triggered T leukemogenesis via acquisition of self-renewal activity and alteration in the differentiation program to T-cell lineages. Together with the known function of LSD1 in regulating the activity of self-renewal in hematological malignancies, these studies prompted evaluation of the efficacy of the potent, selective, and orally bioavailable FAD-directed LSD1 inhibitor, INCB059872, in preclinical models of T-ALL. Expression of LSD1 was abundant in human-T-ALL cell lines as detected by immunoblotting. In vitro, INCB059872 treatment significantly inhibited the proliferation of a subset of human T-ALL cell lines. In vivo, once daily oral administration of INCB059872 inhibited tumor growth significantly in multiple human T-ALL subcutaneous xenograft models including Molt-4, RPMI-8402, CCRF-HSB-2, and CCRF-CEM, but was ineffective against DND-41 xenografts. The anti-tumor efficacy observed with INCB059872 had no clear genetic correlation with Notch mutation status of T-ALL tumors. Combination efficacy studies of INCB059872 with standard care of agents or targeted therapeutic agents in T-ALL models are currently being evaluated. These data suggest exploring the potential clinical development of INCB059872 as a therapy for T-ALL patients. Citation Format: Melody Diamond, Yvonne Lo, Antony Chadderton, Min Ye, Valerie Roman, Michael Weber, Chunhong He, Liangxing Wu, Swamy Yeleswaram, Alan Roberts, Wenqing Yao, Gregory Hollis, Reid Huber, Peggy Scherle, Bruce Ruggeri, Sang Hyun Lee. The evaluation of INCB059872, an FAD-directed inhibitor of LSD1, in preclinical models of T-ALL [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1893.

Abstract Numerous studies have elucidated that the most pivotal functions of lysine specific demethylase-1 (LSD1) are associated with regulating normal or malignant hematopoiesis by maintaining stem cell self-renewal and regulating myeloid differentiation. In preclinical models, studies with either pharmacological inhibition or genetic knockdown of LSD1 demonstrated that LSD1 is essential for differentiation of progenitor cells during normal hematopoiesis. In the clinic, AML manifests itself via clonal expansion of abnormal differentiation and proliferation of myeloid cells and, therefore, the inhibition of LSD1 activity with small molecule inhibitors could be a promising therapeutic approach for AML. Previously, we reported upon the identification of a flavin adenine dinucleotide (FAD) directed LSD1 specific inhibitor, INCB059872, which is efficacious in preclinical mouse models utilizing human AML cell lines and primary AML cells by inducing cell differentiation as indicated by the induction of CD11b and CD86 markers. Using a larger panel of myeloid and HSC flow cytometry markers, our currents efforts expanded upon these observations to ascertain whether INCB059872 enhanced lineage commitment at hematopoietic stem cell (HSC) and/ or promoted monocytic/granulocytic differentiation of human primary AML cells ex vivo and in human systemic AML PDX models. In both human AML PDX models and human primary AML samples, INCB059872 increased myeloid differentiation with increasing populations of monocytes (CD14+) and granulocytes (CD15+). Furthermore, INCB059872 induced the differentiation of early hematopoietic progenitors, CD34+/CD38- to more committed CD34+/CD38+ multipotent/oligopotent progenitors, which in turn gave rise to lineage specific progenitors in the human AML PDX models. These studies support further exploration of INCB059872 as a promising novel epigenetic agent for AML therapy whose mechanism of action lies in part through the induction of differentiation of leukemic stem/progenitor cells to more committed hematopoietic lineages. Citation Format: Antony Chadderton, Min Ye, Melody Diamond, Valerie Roman, Michael Weber, Chunhong He, Liangxing Wu, Swamy Yeleswaram, Alan Roberts, Wenqing Yao, Gregory Hollis, Reid Huber, Peggy Scherle, Bruce Ruggeri, Sang Hyun Lee. The FAD-directed LSD1 specific inhibitor, INCB059872, is a promising epigenetic agent for AML therapy by inducing differentiation of leukemic stem/progenitor cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1888.