DrugMap: A quantitative pan-cancer analysis of cysteine ligandability

Cysteine-focused chemical proteomic platforms have accelerated the clinical development of covalent inhibitors of a wide-range of targets in cancer. However, how different oncogenic contexts influence cysteine targeting remains unknown. To address this question, we have developed DrugMap , an atlas of cysteine ligandability compiled across 416 cancer cell lines. We unexpectedly find that cysteine ligandability varies across cancer cell lines, and we attribute this to differences in cellular redox states, protein conformational changes, and genetic mutations. Leveraging these findings, we identify actionable cysteines in NFκB1 and SOX10 and develop corresponding covalent ligands that block the activity of these transcription factors. We demonstrate that the NFκB1 probe blocks DNA binding, whereas the SOX10 ligand increases SOX10-SOX10 interactions and disrupts melanoma transcriptional signaling. Our findings reveal heterogeneity in cysteine ligandability across cancers, pinpoint cell-intrinsic features driving cysteine targeting, and illustrate the use of covalent probes to disrupt oncogenic transcription factor activity.

Abstract 137: Epigenetic regulation of APOBEC3A mutagenesis and tumor evolution during targeted therapy in non-small cell lung cancer

Acquired drug resistance to even the most effective anti-cancer targeted therapies remains an unsolved clinical problem. Although many drivers of acquired drug resistance have been identified, the underlying molecular mechanisms shaping tumor evolution during treatment are incompletely understood. We recently demonstrated that lung cancer targeted therapies commonly used in the clinic induce the expression of cytidine deaminase APOBEC3A (A3A), leading to sustained mutagenesis in drug tolerant cancer cells persisting during therapy. Preventing therapy-induced A3A mutagenesis by gene deletion decreased the accumulation of chromosomal aberrations and delayed the emergence of drug resistance. Thus, we hypothesize that inhibition of A3A may represent a potential therapeutic strategy to halt the evolution of resistant clones and prevent acquired resistance to lung cancer targeted therapies. Understanding the molecular mechanism of A3A induction during targeted therapy is a crucial step to develop targeting A3A therapies.Using a targeted drug combination screen, we found that DNA methyltransferase (DNMT) inhibitors induce expression of APOBEC3A in non-small cell lung cancer cells, phenocopying the effects of targeted therapies. RNA-seq profiling revealed that both targeted therapies and DNMT inhibitors increase expression of non-coding repeat RNAs. Repeatome analysis identified distinct classes of non-codding RNAs including endogenous retrovirus elements (ERV) in drug tolerant persister cells. Activation of endogenous intracellular viral sensing pathways by exogenous nucleic acids mimicking viral infection induced APOBEC3A, and genetic perturbation of RIG-I or MAVS reduced APOBEC3A induced by targeted therapy. These findings suggest that epigenetic derepression of retroviral repeat elements may underly APOBEC3A mutagenic activity and tumor evolution during lung cancer targeted therapy.

Citation Format: Hideko Isozaki, Ramin Sakhtemani, Naveed Nikpour, Susanna Monroe, Jessica Lin, Lecia Sequist, Zofia Piotrowska, Justin Gainor, Rémi Buisson, Michael Lawrence, Aaron Hata. Epigenetic regulation of APOBEC3A mutagenesis and tumor evolution during targeted therapy in non-small cell lung cancer [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 137.

Abstract 6130: Cancer sensitivity to therapy is constrained by apoptosis regulation in cells of origin

Many chemotherapeutic agents target cellular components or processes that are present in all cancers, yet clinical responses to these agents vary greatly between cancer types and even patient age - the basis for these broad-scale differences are unclear. The vast majority of targeted and cytotoxic cancer therapies including ionizing radiation produce pro-apoptotic signaling in exposed cells, suggesting that the mitochondrial apoptosis sensitivity of cancer cells could act as a central signaling “node” to broadly impact therapy outcomes. To test this, we used BH3 profiling and complementary chemosensitivity assays to analyze hundreds of primary cancer specimens across twelve major cancer types. We find that cancers with typically favorable outcomes including certain hematologic malignancies, testicular cancer, and some pediatric cancers contain mitochondria that are highly primed for apoptosis, which renders them hypersensitive to cytotoxic as well as targeted agents and radiation therapy. Priming levels in many epithelial cancers including ovarian cancer and non-small cell lung cancer are highly heterogeneous, mirroring their variability in clinical outcomes. Finally, many tumor types that are typically chemoresistant including adult soft tissue sarcomas, hepatocellular carcinoma and pancreatic cancer are almost completely resistant to pro-apoptotic signaling. By analyzing in vitro and in vivo pancreatic, ovarian, hepatocellular and sarcoma tumorigenesis models, we find that apoptotic priming generally increases during neoplastic transformation, in part due to consistent upregulation of pro-apoptotic proteins BAX and BAK. However, the level of apoptotic priming in cancer cells is constrained by the baseline apoptosis sensitivity of normal cells prior to transformation. Remarkably, we find that apoptotic priming is dynamically regulated by cell lineage and differentiation state but can also be modulated by oncogenes. For instance, Myc activation typically increases apoptotic priming while activation of mutant Ras signaling decreases it - these changes in priming alter the chemosensitivity of cancer cells. Finally, we use inducible mouse tumor models to demonstrate that neoplastic transformation of cells from developmentally immature tissues yields pediatric tumors that are more primed for apoptosis than equivalent tumors arising in adults. This difference in priming causes pediatric tumors to be more sensitive to front-line therapies and BH3 mimetics targeting pro-survival BCL-2 family proteins in vitro and in vivo. Thus, lineage-determined regulation of apoptosis prior to and during neoplastic transformation leads to broad-scale differences in cancer cell chemosensitivity and can be exploited therapeutically by targeting BCL-2 family proteins.

Citation Format: Cameron Fraser, Xingping Qin, Kenichi Shimada, Johan Spetz, Mary Heather Florido, Rumani Singh, Stacey Yu, Adam Presser, Zintis Inde, Gaurav Joshi, Jennifer Guerriero, Francisco Sanchez-Rivera, Alison Karst, Omar Lopez, Chendi Li, Peter Winter, Ying Yue, Peter Sorger, Jingwei Cheng, Izidore Lossos, Aaron Hata, Ronny Drapkin, Adam Palmer, James Decaprio, Manisha Thakuria, Charles Yoon, Ursula Matulonis, Matthew Meyerson, Elizabeth Stover, Diana Cardona, Kris Wood, Shayna Sarosiek, David Kirsch, Joseph Mancias, Andrew Cherniack, Anthony Letai, Kristopher Sarosiek. Cancer sensitivity to therapy is constrained by apoptosis regulation in cells of origin. [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 6130.

Abstract 1156: GUK1 is a novel metabolic liability in oncogene-driven lung cancer

There is a longstanding desire to take therapeutic advantage of dysregulated metabolic states in cancer. While it has been appreciated that lung tumors rewire their cellular metabolic networks to support unrestrained proliferation, metabolic vulnerabilities have largely not been explored in the context of specific onco-genotypes. This represents a major gap in our understanding of how different oncogenic drivers in non-small cell lung cancer (NSCLC) confer reliance on discrete metabolic networks to sustain tumor growth. The goals of this project are (1) to investigate metabolic dependencies in distinct molecular subtypes of lung cancer and (2) to elucidate how metabolic reprogramming drives resistance to targeted therapy. Using patient-derived cell culture models and tumor specimens collected from patients with ALK-positive (ALK+) NSCLC, we identified that lung tumors with ALK rearrangements harbor a unique metabolic signature marked by reliance on anabolic nucleotide pathways. A phosphoproteomic screen in ALK+ patient-derived cells identified a novel metabolic target of ALK signaling, GUK1, the only known enzyme responsible for GDP synthesis. We show that ALK binds to and phosphorylates GUK1 and that ALK-mediated GUK1 phosphorylation augments GDP/GTP nucleotide biosynthesis. Steady-state and tracing metabolomic studies demonstrate that ALK inhibition and GUK1 phosphomutant are epistatic in guanine nucleotide production. Molecular dynamic modeling suggests that phosphorylation of GUK1 alters the dynamics of active site closure to enhance substrate processivity and protects GUK1 from a non-catalytic confirmation. Introduction of phosphomutant GUK1 into ALK+ patient-derived cell lines results in decreased tumor proliferation in vitro and in vivo in xenograft models. Spatially resolved mass spectrometry imaging of tumor specimens from ALK+ patients demonstrates significant enrichment of guanine nucleotides in ALK+ and phospho-GUK1+ tumor cells. We identified that other oncogenic fusion proteins regulate GUK1 phosphorylation, highlighting the need to further characterize GUK1 as a metabolic liability in NSCLC. Furthermore, a subset of patient-derived cell lines with resistance to ALK tyrosine kinase inhibitors (TKIs) exhibits increased expression and phosphorylation of GUK1, indicating that regulation of this metabolic enzyme may play a role in mediating acquired resistance. We anticipate these studies will pave the way for the development of new therapeutic approaches by exploiting metabolic vulnerabilities in oncogene-driven lung cancers.

Citation Format: Jaime Laurel Schneider, Kiran Kurmi, Ishita Dhiman, Roberta Colapietro, Shakchhi Joshi, Christian Johnson, Satoshi Yoda, Joao Paulo, Daniela Ruiz, Sylwia Stopka, Gerard Baquer, Jessica Lin, Kevin Haigis, Nathalie Agar, Steven Gygi, Aaron Hata, Marcia Haigis. GUK1 is a novel metabolic liability in oncogene-driven lung cancer [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 1156.

Porous Paclitaxel Mesh Reduces Local Recurrence in Patient-Derived Xenograft Resection Model

BACKGROUND

Drug-loaded meshes offer a promising delivery strategy for the prevention of local recurrence. Patient-derived xenograft (PDX) models are representative of individual patient tumors and predictive of clinical outcomes.

METHODS

A PDX model was established in NSG (NOD-scid IL2Rgamma^null) mice using tumor tissue from a patient with aggressive lung adenocarcinoma. Polyglycolic acid (PGA) meshes loaded with paclitaxel (PGA+PTX) were electrospun. Tumor-bearing mice were randomized into 4 groups after macroscopic complete resection: (1) no treatment (n = 10); (2) intraperitoneal PTX at 20 mg/kg (n = 10); (3) PGA mesh without drug (n = 14); and (4) PGA+PTX mesh at 12 mg/kg (n = 14). A 1-cm² mesh was placed onto the tumor resection beds. Groups were observed for local recurrence for 120 postoperative days.

RESULTS

PDX mice treated with PGA+PTX meshes after resection exhibited a >5-fold increase in recurrence-free survival (P < .0001) compared with systemically treated and untreated control groups. Median recurrence-free survival was 24 days for untreated and intraperitoneal PTX groups, 28 days for unloaded PGA mesh group, and undefined for the PGA+PTX mesh group.

CONCLUSIONS

Development of a PDX surgical resection model of non-small cell lung cancer permits robust assessment of postresection local recurrence for preclinical studies of patient-derived tumors. Intraoperative placement of drug-loaded meshes demonstrates superior local disease treatment, suggesting that this approach may improve recurrence-free survival in early-stage non-small cell lung cancer patients undergoing limited resection.

Abstract 1300: Targeted therapies prime lung cancer cells for macrophage-mediated destruction

The CD47/SIRPa axis is an immune checkpoint that regulates macrophage anti-tumor function. Therapies that block CD47 on cancer cells show promise in clinical trials for solid and hematologic malignancies, particularly when combined with other anticancer agents. However, the best combination strategies for using CD47-blocking therapies remain unknown.

In this study, we developed a novel in vitro screening platform to identify drugs that render cancer cells more vulnerable to macrophage attack. We performed an unbiased screen of 800 FDA-approved drugs using primary human macrophages and PC9 cells, a human EGFR mutant lung cancer cell line. We identified EGFR tyrosine kinase inhibitors (TKIs) as drugs that act on the cancer cells and specifically enhance macrophage-mediated cytotoxicity (&gt;4-fold enhancement, p&lt;0.01). In contrast, conventional chemotherapy drugs either showed no significant enhancement or abrogated macrophage activation. The combination of EGFR TKIs with anti-CD47 antibodies elicited maximal phagocytosis across a range of cell lines and conditions. In long-term co-culture assays with macrophages, the combination of EGFR TKIs and anti-CD47 antibodies eliminated persister cells to prevent TKI resistance. Importantly, these findings extended to lung cancers with other RTK-MAPK pathway mutations, such as ALK-rearranged cancers (treated with lorlatinib, alectinib, or crizotinib) or KRAS G12C mutant cancers (treated with sotorasib or adagrasib). In xenograft and syngeneic mouse models, the combination of targeted therapies with CD47 ablation was able to dramatically reduce tumor burden.

To understand the mechanism of synergy, we generated 8 different lung cancer cell lines resistant to EGFR, ALK, or KRAS inhibitors. The resistant lines significantly upregulated CD47 and concomitantly became more sensitive to macrophage attack in vitro and in vivo. By RNA sequencing, we identified multiple mechanisms contributing to vulnerability, including secretion of the cytokine MIP-3 by the cancer cells and alteration of other immunoregulatory molecules. Overall, we have identified a novel therapeutic strategy to enhance the efficacy of RTK-MAPK pathway inhibitors by combining them with anti-CD47 therapies.

Our findings suggest cross-sensitivity occurs such that lung cancer cells that become resistant to targeted therapies also become more sensitive to macrophage attack. Our findings provide rationale for testing this combination approach in patients with lung cancers bearing driver mutations.

Citation Format: Kyle Vacarro, Juliet Allen, Asaf Maoz, Sarah Reeves, Aaron Hata, Kipp Weiskopf. Targeted therapies prime lung cancer cells for macrophage-mediated destruction [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1300.

Abstract 2150: LKB1 loss rewires JNK-induced apoptotic protein dynamics through NUAKs and sensitizes KRAS-mutant non-small cell lung cancers to combined KRAS G12C + MCL-1 blockade

The recent approval of the KRAS G12C inhibitor sotorasib (AMG 510) for non-small cell lung cancer (NSCLC) marked a milestone in the development of targeted therapies for KRAS mutant cancers. While sotorasib and other KRAS G12C inhibitors have demonstrated rapid and durable responses in the clinic, some patients do not achieve responses. The identification of specific vulnerabilities conferred by recurrent co-occurring mutations may enable the development of biomarker-driven combination therapies with enhanced activity in distinct subsets of patients. We screened a panel of KRAS-mutant NSCLC cell lines as well as patient-derived xenograft (PDX) mouse models and observed that loss of the tumor suppressor STK11/LKB1 is associated with increased sensitivity to combined MAPK (either the KRAS G12C inhibitor sotorasib or MEK inhibitor trametinib) and MCL-1 inhibition (AMG 176). Restoration of LKB1 expression in LKB1-deficient cell lines and PDX tumors blunted the apoptotic response to MAPK + MCL-1 inhibition; conversely, deletion of LKB1 in LKB1 wild-type models increased sensitivity. Mitochondrial apoptotic cell death is regulated by interactions between pro- (e.g., BIM) and anti-apoptotic (e.g., MCL-1, BCL-XL) BCL-2 family members. MAPK inhibition increases BIM, while MCL-1 inhibition prevents BIM sequestration by MCL-1, resulting in apoptosis. LKB1 deficient cells exhibit increased association of BIM and MCL-1 upon MAPK inhibition, effectively priming cells for death upon inhibition of MCL-1. Mechanistically, LKB1 deficiency and associated loss of NUAK phosphorylation leads to hyperactivation of the JNK phospho-kinase network. JNK phosphorylates MCL-1 at S64 and T163, which enhances BIM: MCL-1 protein-protein interaction. Conversely, JNK phosphorylates BCL-XL at S62 and prevents sequestration of BIM. This series of phosphorylation events increases MCL-1 dependence and creates a specific vulnerability of KRAS-LKB1 tumors to MAPK + MCL-1 inhibition. Consistent with this mechanism, ex vivo treatment of tumor tissue from a KRAS-LKB1 mutant NSCLC patient with sotorasib or trametinib increased MCL-1 dependent priming. These results reveal a novel link between LKB1 and the regulation of BCL-2 family proteins and provide preclinical rationale for evaluation of combined KRAS G12C + MCL-1 inhibitors for KRAS-LKB1 mutant NSCLC.

Citation Format: Chendi Li, Mohammed Usman Syed, Yi Shen, Audris Oh, Cameron Fraser, Johannes Kreuzer, Christopher Nabel, Kaitlyn Webster, Robert Morris, Sean Caenepeel, Anne Y. Saiki, Karen Rex, J. Russell Lipford, Wilhelm Hass, Kristopher Sarosiek, Paul E. Hughes, Aaron Hata. LKB1 loss rewires JNK-induced apoptotic protein dynamics through NUAKs and sensitizes KRAS-mutant non-small cell lung cancers to combined KRAS G12C + MCL-1 blockade [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2150.

POS0340 VASCULAR IMAGING IN PATIENTS WITH REFRACTORY TAKAYASU ARTERITIS TREATED WITH TOCILIZUMAB: ANALYSIS FROM A RANDOMIZED CONTROLLED TRIAL

Background:

In the TAKT study, a randomized controlled trial of tocilizumab (TCZ) in patients with refractory Takayasu arteritis (TAK) in Japan, the primary end point of time to relapse after induction of remission with glucocorticoid (GC) treatment showed a trend favoring TCZ over placebo (hazard ratio 0.41 [95.41% confidence interval, 0.15-1.10; p=0.0596]),1 but the double-blind period was too short for imaging evaluation.

Objectives:

To independently evaluate vascular imaging in a post hoc analysis of radiographs from the TAKT study.

Methods:

Computed tomography images from patients in the TAKT study were evaluated by three independent radiologists who were not involved in the original trial. Patients who received TCZ and had computed tomography images available (n=28) were included. Assessments were made in 22 arteries for the change from baseline in wall thickness (primary end point), dilatation/aneurysm, stenosis/occlusion, or wall enhancement for at least 96 weeks after the start of tocilizumab treatment. Patient-level assessments were also conducted.

Results:

Among 28 patients who received at least one dose of TCZ and for whom images were available, 86.7% of 22 arteries had improved/stable (no progression) wall thickness at week 96. The proportions of patients with no progressed, partially progressed, or newly progressed lesions were 57.1%, 10.7%, and 28.6% for wall thickness, and the proportions without progressed lesions were 92.9% for dilatation/aneurysm and 85.7% for stenosis/occlusion (Figure 1). Patients with newly progressed lesions, reflecting more refractory disease, were receiving glucocorticoid doses that could not be reduced below 0.1 mg/kg/day at week 96.

Conclusion:

Approximately 60% of patients with TAK treated with tocilizumab did not experience progression in wall thickness. Few patients experienced progressive dilatation/aneurysm or stenosis/occlusion. Wall thickness progression likely resulted from refractory TAK. Patients who experience this should be monitored regularly by imaging, and additional glucocorticoid or immunosuppressive treatment should be considered to avoid vascular progression.

References:

[1]Nakaoka Y et al. Ann Rheum Dis. 2018;77:348-354.

Disclosure of Interests:

Yoshikazu Nakaoka Consultant of: Roche/Genentech, Grant/research support from: Roche/Genentech, Masahiro Yanagawa Consultant of: Roche/Genentech, Akinori Hata Consultant of: Roche/Genentech, Katsuhisa Yamashita Employee of: Chugai Pharmaceutical Co., Ltd., Norihiro Okada Employee of: Chugai Pharmaceutical Co., Ltd., Shinji Yamakido Employee of: Chugai Pharmaceutical Co., Ltd., Hiromitsu Hayashi Consultant of: Roche/Genentech, David Jayne Consultant of: Roche/Genentech, Grant/research support from: Roche/Genentech

Cost-effectiveness analysis for preventing hepatitis B virus reactivation-related death in Japan

Background

There is no worldwide standard recommendation for preventing hepatitis B virus (HBV) reactivation for patients with resolved infection treated with an anti-CD20 antibody for B-cell non-Hodgkin lymphoma. This study aims to compare the cost-effectiveness between two commonly used strategies to prevent HBV reactivation-related death.

Methods

The two strategies compared were prophylactic antiviral therapy (Pro NAT) and HBV DNA monitoring followed by on-demand antiviral therapy (HBV DNA monitoring) using entecavir (Entecavir, a generic drug for Baraclude). Effectiveness was defined as the prevention of death due to HBV reactivation and costs were calculated under the health insurance system of Japan as of April 2018 using Markov model. A cost-minimization analysis, one of the cost-effectiveness analyses, was applied, since the effectiveness was the same between the two strategies according to a meta-analysis. To consider the effect of uncertainty for each parameter, probabilistic sensitivity analysis (PSA) was performed. In the scenario analysis, costs were calculated using lamivudine (Zefix) or tenofovir alafenamide (Vemlidy) instead of entecavir. All analyses were done using TreeAge Pro 2019 (TreeAge Software, Inc., MA, USA).

Results

Estimated costs per patient during the 30 months after initiation of chemotherapy for lymphoma were 1,513 USD with Pro NAT and 1,265 USD with HBV DNA monitoring. A PSA revealed that HBV DNA monitoring was more consistently cost-effective compared with Pro NAT when some parameters were set randomly according to probability distributions. In our scenario analysis, costs of Pro NAT and HBV DNA monitoring were calculated as 2,762 and 1,401 USD using lamivudine, 4,857 and 1,629 USD using tenofovir alafenamide.

Conclusions

Our cost-effectiveness analysis shows that an HBV DNA monitoring strategy using entecavir should be recommended for preventing HBV reactivation-related death in Japan.

Key messages

Cost-effectiveness analysis demonstrated that HBV DNA monitoring was more cost-effective compared to Pro NAT; this result was consistent with PSA. HBV DNA monitoring strategy should be recommended to prevent HBV reactivation-related death for the patients with resolved HBV infection in Japan.

Abstract 2101: Targeting the root of cancer persister cells using an expressed barcode library

Despite a favorable initial response, many cancer patients will experience recurrence of disease within months or years after diagnosis. Recurrence largely arises as a result of the growth of residual cancer cells that remain after treatment. The ability of a subset of cells to survive is attributed frequently to genetic heterogeneity, however recently it was shown that in multiple cancer types relapse can arise due to the presence of persister cells. Persisters are a subpopulation of transiently drug-tolerant cells that are able to survive therapy through reversible, non-mutational mechanisms. Tumor dormancy, stochastic cell state shifts and stem cell-like populations are amongst the mechanisms hypothesized to underlie persister phenotype. However, given the lack of high-throughput methods to concurrently track cell state and lineage, it is not currently feasible to distinguish the relative contribution of each of these factors. To address this need, we generated the Watermelon library. The Watermelon library is a high-complexity expressed barcode library that enables simultaneous tracing of lineage as well as the transcriptional and proliferative state of each cell in the population during drug treatment. We have applied the watermelon system to study the mechanisms underlying the ability of a small population of cells to regain proliferative capacity under constant treatment with EGFR tyrosine kinase inhibitors. We combine time-lapse imaging with single-cell RNA sequencing to show that early drug-cyclers do not acquire a facilitating resistance mutation but rather transition to a new cell state. We find that this non-genetic drug-proliferative state is not restricted to a certain clonal lineage and can be reached by distinct transcriptional paths. We anticipate that this unique library, which can be applied to other systems, would facilitate a better understanding of the cellular and molecular pathways that affect non-inherited drug resistance.

Citation Format: Yaara Oren, Pratiksha Thakore, Mike S. Cuoco, Heidie Frisco Cabanos, Aaron Hata, Joan S. Brugge, Aviv Regev. Targeting the root of cancer persister cells using an expressed barcode library [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2101.

Hepatitis B virus reactivation in patients with rheumatoid arthritis: Analysis of the National Database of Japan

This study aimed to determine the incidence and risk factors for hepatitis B virus (HBV) reactivation in patients with rheumatoid arthritis (RA) undergoing immunosuppressive therapy. The National Database of Japan, in which insurance claim data have been comprehensively accumulated, was utilized. The subjects were 76 641 RA patients who were plausibly initiated on immunosuppressive therapy from April 2013 to March 2014. Laboratory tests of the hepatitis B surface antigen, anti-hepatitis B virus surface antibody, and anti-hepatitis B virus core antibody were performed in 28.23%, 12.52% and 14.63% of patients, respectively, when the therapy was initiated. We found that HBV reactivation and fulminant hepatitis occurred in both the patients with and without HBV DNA monitoring, indicating insufficient monitoring in Japan during the study. The cumulative incidence of HBV reactivation over 24 months was 1.57% (95% confidence interval [CI] = 1.28%-1.92%) in the monitoring group, which consisted of those with resolved HBV infection. Glucocorticoid administration was a potent risk factor for HBV reactivation (hazard ratio [HR]  = 1.70, 95% CI = 1.26-2.29, P = .001 in all subjects, and HR = 1.82, 95% CI = 1.18-2.81, P = .007 in the nonmonitoring group), although it was not statistically significant in the monitoring group (HR = 1.49, 95% CI = 0.99-2.26 and P = .057). No significant risk difference was observed between single administration of methotrexate and biological drugs.

Abstract PR04: Decoding tumor microenvironment to enhance NSCLC targeted therapy

Background: Tyrosine kinase inhibitors (TKI) have yielded promising responses in non-small cell lung cancer (NSCLC) with EGFR mutations and ALK translocations. However, these and other targeted therapies are limited by intrinsic and acquired drug resistance. The previous study from our group investigated tumor autonomous resistance mechanisms by developing patient-derived cancer models (PDCs). In this study, we aimed to decipher the nonautonomous resistance mechanisms via tumor microenvironment by developing patient-derived fibroblast (PDF) models.

Method: Cancer-associated fibroblast cell lines are established directly from individual EGFR mutant NSCLC biopsies. These cell lines, as representative of each patient’s tumor microenvironment, are further subjected to functional analysis. An imaging-based high-throughput platform is developed to screen for nonautonomous resistance by co-culturing PDC and PDF models in vitro. In the parallel, two independent approaches are performed to further identify mechanisms underlying the nonautonomous resistance. These include a drug screen to determine the pathway maintaining the cancer cells’ survival, and a secretomic analysis on the PDFs to identify the plausible cytokine(s) responsible for the resistance.

Result: By co-culturing screening, nonautonomous resistance can be found in a wide spectrum of models. The subsequent drug screen reveals both a canonical HGF-dependent and novel HGF-independent mechanisms contributing to EGFR TKI resistance. Both of these can be explained by the PDF’s variable cytokine secretion and can be overcome by specific therapeutic combinations. Moreover, the microenvironment-driven EGFR TKI resistance has also been validated in vivo. The prevalence of the identified cytokine is further tested in clinical specimens.

Conclusion: PDFs provide a new avenue to explore nonautonomous resistance for targeted therapy. Applying this approach, we identified both the canonical HGF-dependent and novel HGF-independent mechanisms that putatively confer EGFR TKI resistance. Taking EGFR TKI therapy as a paradigm, these findings will be valuable to optimize targeted therapy and to inform the design of personalized pharmaceutical interventions.

This abstract is also being presented as Poster B25.

Citation Format: Haichuan Hu, Lecia Sequist, Zofia Piotrowska, Hillary Mulvey, Sundus Noeen, Patricia Hare, David Kodack, Aaron Hata, Matt Niederst, Cyril Benes, Jeff Engelman. Decoding tumor microenvironment to enhance NSCLC targeted therapy [abstract]. In: Proceedings of the Fifth AACR-IASLC International Joint Conference: Lung Cancer Translational Science from the Bench to the Clinic; Jan 8-11, 2018; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2018;24(17_Suppl):Abstract nr PR04.

Abstract B25: Decoding tumor microenvironment to enhance NSCLC targeted therapy

This abstract is being presented as a short talk in the scientific program. A full abstract is printed in the Proffered Abstracts section (PR04) of the Conference Proceedings.

Citation Format: Haichuan Hu, Lecia Sequist, Zofia Piotrowska, Hillary Mulvey, Sundus Noeen, Patricia Hare, David Kodack, Aaron Hata, Matt Niederst, Cyril Benes, Jeff Engelman. Decoding tumor microenvironment to enhance NSCLC targeted therapy [abstract]. In: Proceedings of the Fifth AACR-IASLC International Joint Conference: Lung Cancer Translational Science from the Bench to the Clinic; Jan 8-11, 2018; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2018;24(17_Suppl):Abstract nr B25.

Abstract 4954: Decoding tumor microenvironment to enhance NSCLC targeted therapy

Background: Tyrosine kinase inhibitors (TKI) have yielded promising responses in non-small-cell lung cancer (NSCLC) with EGFR mutations and ALK translocations. However, these and other targeted therapies are limited by intrinsic and acquired drug resistance. The previous study from our group investigated tumor autonomous resistance mechanisms by developing patient-derived cancer models (PDCs). In this study, we aimed to decipher the non-autonomous resistance mechanisms via tumor microenvironment by developing patient-derived fibroblast (PDF) models. Method: Cancer-associated fibroblast cell lines are established directly from individual EGFR mutant NSCLC biopsies. These cell lines, as representative of each patient's tumor microenvironment, are further subjected to functional analysis. An imaging-based high-throughput platform is developed to screen for non-autonomous resistance by co-culturing PDC and PDF models in vitro. In the parallel, two independent approaches are performed to further identify mechanisms underlying the non-autonomous resistance. These include a drug screen to determine the pathway maintaining the cancer cells' survival, and a secretomic analysis on the PDFs to identify the plausible cytokine(s) responsible for the resistance. Result: By co-culturing screening, non-autonomous resistance can be found in a wide spectrum of models. The subsequent drug screen reveals both a canonical HGF dependent and novel HGF independent mechanisms contributing to EGFR TKI resistance. Both of these can be explained by the PDF's variable cytokine secretion and can be overcome by specific therapeutic combinations. Moreover, the microenvironment-driven EGFR TKI resistance has also been validated in vivo. And the prevalence of the identified cytokine is further tested in clinical specimens. Conclusion: PDFs provide a new avenue to explore non-autonomous resistance for targeted therapy. Applying this approach, we identified both the canonical HGF dependent and novel HGF independent mechanisms that putatively conferring EGFR TKI resistance. Taking EGFR TKI therapy as a paradigm, these findings will be valuable to optimize targeted therapy and to inform the design of personalized pharmaceutical interventions.

Citation Format: Haichuan Hu, Lecia Sequist, Zosia Piotrowska, David Kodack, Aaron Hata, Matt Niederst, Cyril Benes, Jeffrey Engelman. Decoding tumor microenvironment to enhance NSCLC targeted therapy [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 4954.

Abstract A142: Targeting FGFR to overcome EMT-related resistance in EGFR-mutated non-small cell lung cancer

Epithelial-to-mesenchymal transition (EMT) has been identified as a mechanism of both adaptive and acquired resistance to EGFR tyrosine kinase inhibitors (TKIs) in in vitro models and clinical samples of EGFR-mutated non-small cell lung cancer (NSCLC). However, a therapeutic strategy for targeting this mechanism remains unknown. In the setting of adaptive resistance early in the course of EGFR treatment, we observed that FGFR3 upregulation coincided with increased expression of mesenchymal genes, and a pooled shRNA dropout screen identified FGFR3 as a top sensitizer to EGFR inhibitor treatment. Dual FGFR (BGJ398) and EGFR (gefitinib) inhibition suppressed the outgrowth of drug-tolerant persister clones compared to EGFR TKI alone in vitro. Finally, we observed that the combination of BGJ398 and gefitinib prevented the development of in vivo resistance in PC9 xenograft mouse models. These results suggest that dual FGFR-EGFR blockade may be an effective strategy for preventing resistance associated with EMT in EGFR mutant NSCLC.

Citation Format: Sana Raoof, David Ruddy, Daria Timonia, Leah Damon, Jeff Engelman, Aaron Hata. Targeting FGFR to overcome EMT-related resistance in EGFR-mutated non-small cell lung cancer [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2017 Oct 26-30; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2018;17(1 Suppl):Abstract nr A142.

Abstract A145: SHP2 inhibition restores sensitivity to ALK inhibitors in resistant ALK-rearranged NSCLC

Most anaplastic lymphoma kinase (ALK)-rearranged non-small cell lung tumors initially respond to small-molecule ALK inhibitors, but drug resistance often develops. After tumors develop resistance to highly potent 2nd-generation ALK inhibitors, approximately half harbor ALK resistance mutations, while the other half have other mechanisms of resistance. The latter often have activation of at least one of several different tyrosine kinases driving resistance. Such tumors are not expected to respond to the 3rd-generation ALK inhibitor, lorlatinib, which is able to overcome all clinically identified ALK resistance mutations, and further therapeutic options are limited. Herein, we deployed an shRNA screen of 1000 genes in multiple ALK inhibitor-resistant patient-derived cells (PDC) to discover sensitizers to ALK inhibition. This approach identified SHP2, a non-receptor protein tyrosine phosphatase, as a common targetable resistance node in multiple PDCs. SHP2 provides a parallel survival input downstream of multiple tyrosine kinases that promote resistance to ALK inhibitors. The recently discovered small-molecule SHP2 inhibitor, SHP099, in combination with the ALK TKI (tyrosine kinase inhibitor), ceritinib, halted the growth of resistant PDCs by preventing compensatory RAS and ERK1/2 reactivation. These findings suggest that combined ALK and SHP2 inhibition may be a promising therapeutic strategy for resistant cancers driven by several different ALK-independent resistance mechanisms.

Citation Format: Leila Dardaei, Hui Qin Wang, Manrose Singh, Paul Fordjour, Satoshi Yoda, Grainne Kerr, Jinsheng Liang, Yichen Cao, Yan Chen, Justin Gainor, Luc Friboulet, Ibiayi Dagogo-Jack, David Myers, Emma Labrot, David Ruddy, Melissa Parks, Dana Lee, Richard DiCecca, Susan Moody, Huaixiang Hao, Morvarid Mohseni, Matthew LaMarche, Juliet Williams, Keith Hoffmaster, Giordano Caponigro, Alice Shaw, Aaron Hata, Cyril Benes, Fang Li, Jeffrey Engelman. SHP2 inhibition restores sensitivity to ALK inhibitors in resistant ALK-rearranged NSCLC [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2017 Oct 26-30; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2018;17(1 Suppl):Abstract nr A145.