Despite high initial efficacy, targeted therapies eventually fail in advanced cancers, as tumors develop resistance and relapse. In contrast to the substantial body of research on the molecular mechanisms of resistance, understanding of how resistance evolves remains limited. Using an experimental model of ALK positive NSCLC, we explored the evolution of resistance to different clinical ALK inhibitors. We found that resistance can originate from heterogeneous, weakly resistant subpopulations with variable sensitivity to different ALK inhibitors. Instead of the commonly assumed stochastic single hit (epi) mutational transition, or drug-induced reprogramming, we found evidence for a hybrid scenario involving the gradual, multifactorial adaptation to the inhibitors through acquisition of multiple cooperating genetic and epigenetic adaptive changes. Additionally, we found that during this adaptation tumor cells might present unique, temporally restricted collateral sensitivities, absent in therapy naïve or fully resistant cells, suggesting the potential for new therapeutic interventions, directed against evolving resistance.

While initially highly successful, targeted therapies eventually fail as populations of tumor cells evolve mechanisms of resistance, leading to resumption of tumor growth. Historically, cell-intrinsic mutational changes have been the major focus of experimental and clinical studies to decipher origins of therapy resistance. While the importance of these mutational changes is undeniable, a growing body of evidence suggests that non-cell autonomous interactions between sub-populations of tumor cells, as well as with non-tumor cells within tumor microenvironment, might have a profound impact on both short term sensitivity of cancer cells to therapies, as well as on the evolutionary dynamics of emergent resistance. In contrast to well established tools to interrogate the functional impact of cell-intrinsic mutational changes, methodologies to understand non-cell autonomous interactions are largely lacking.Evolutionary Game Theory (EGT) is one of the main frameworks to understand the dynamics that drive frequency changes in interacting competing populations with different phenotypic strategies. However, despite a few notable exceptions, the use of EGT to understand evolutionary dynamics in the context of evolving tumors has been largely confined to theoretical studies. In order to apply EGT towards advancing our understanding of evolving tumor populations, we decided to focus on the context of the emergence of resistance to targeted therapies, directed against EML4-ALK fusion gene in lung cancers, as clinical responses to ALK inhibitors represent a poster child of limitations, posed by evolving resistance. To this end, we have examined competitive dynamics between differentially labelled therapy-naïve tumor cells, cells with cell-intrinsic resistance mechanisms, and cells with cell-extrinsic resistance, mediated by paracrine action of hepatocyte growth factor (HGF), within in vitro game assays in the presence or absence of front-line ALK inhibitor alectinib. We found that producers of HGF were the fittest in every pairwise game, while also supporting the proliferation of therapy-naïve cells. Both selective advantage of these producer cells and their impact on total population growth was a linearly increasing function of the initial frequency of producers until eventually reaching a plateau. Resistant cells did not significantly interact with the other two phenotypes. These results provide insights on reconciling selection driven emergence of subpopulations with cell non-cell autonomous resistance mechanisms, with lack of evidence of clonal dominance of these subpopulations. Further, our studies elucidate mechanisms for co-existence of multiple resistance strategies within evolving tumors. This manuscript serves as a technical report and will be followed up with a research paper in a different journal.

Despite high initial efficacy, therapies that target oncogenic kinases eventually fail in advanced, metastatic cancers. This failure in initially responsive tumors is the direct result of the evolution of drug resistance under therapy-imposed selective pressures. In contrast to the massive body of experimental research on the molecular mechanisms of resistance, understanding of its evolutionary origins and dynamics remains fragmented. Using a combination of experimental studies and mathematical modeling, we sought to dissect the evolution of resistance to different clinical ALK inhibitors in an experimental model of ALK positive NSCLC. We found that resistance can originate from heterogeneous, weakly resistant, sub-populations with variable sensitivity to different ALK inhibitors. Instead of the commonly assumed stochastic single hit (epi) mutational transition, or drug-induced reprogramming, we found evidence of a hybrid scenario, of gradual, multifactorial development through acquisition of multiple cooperating genetic and epigenetic adaptive changes, amplified by selection. Additionally, we found that intermediate resistance phenotypes might present unique, temporally restricted collateral sensitivities, absent in therapy naïve or fully resistant cells, suggesting new opportunities for therapeutic interference.

Despite inducing strong and durable remissions, inhibitors of mutant ALK (ALKi) are not curative for advanced ALK+ lung cancers, as residual tumors eventually develop resistance and relapse. Apart from tumor cells’ intrinsic ability to escape from therapy, there is a growing body of evidence suggesting the contribution of extrinsic factors, produced by cancer-associated fibroblasts (CAFs). Despite the multitude of studies that demonstrated the ability of multiple CAF-produced factors to reduce the sensitivity of tumor cells to ALKi, the contribution of these effects toward responses in vivo remains unresolved. To study the impact of stroma on the ability of tumor cells to survive and develop resistance to ALKi, we derived multiple isolates of fibroblasts from clinical samples. Consistent with previous reports, we found that co-culture with CAFs or CAF-conditioned media protects tumor cells against ALKi. We observed that the degree of protection varies between different CAF isolates. This variability in the extent of protection could be attributed to variability in the levels of secreted hepatocyte growth factor (HGF) a known paracrine mediator of environmental resistance. Moreover, exogenous HGF phenocopied the effect of CAFs while blocking HGF-cMET signaling with neutralizing antibodies or pharmacologically abrogated the protective effect. To test the relevance of these findings in vivo, we took advantage of the inability of murine HGF to activate human cMET by comparing the response of ALK+ xenograft tumors to front-line ALKi alectinib between NSG mice and NSG-derivative strain with humanized HGF. In contrast to the in vitro data, HGF status had only a minimal impact on the remission-relapse dynamics of xenograft tumors. This lack of differences reflected a strong HGF-independent sheltering effect of the stromal niche. Our histological analyses of samples at different points of remission-relapse response revealed that while alectinib potently suppressed tumor cell proliferation, proximity to stroma reduced this cytostatic effect, without impacting cell proliferation in the absence of therapy. To gain insights into the mechanistic underpinning of this HGF-independent effect, we used spatial transcriptomics, comparing stroma-proximal and stroma-distant tumor regions. These analyses, as well as functional validation studies, indicate that stroma-sheltering effects are mediated by multiple mechanisms, acting in an additive fashion. In summary, our studies indicate that therapy resistance of tumor cells reflects the combined action of both intrinsic and microenvironmental factors. Our findings indicate that focusing on a single resistance mechanism at a time is unlikely to induce strong, durable responses. Instead, tackling the issue of therapy resistance necessitates the consideration of multiple resistance mechanisms as well as the moving target nature of tumors under therapy. Citation Format: Bina Desai, Tatiana Miti, Daria Miroshnychenko, Viktoriya Marusyk, Chandler Gatenbee, Menkara Henry, Uwe Rix, Alexander Anderson, Eric Haura, David Basanta, Andriy Marusyk. Stromal facilitated multifactorial resistance to tumor cells against targeted therapies in ALK+ NSCLC [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 552.

Background: Brain metastasis (BM) develops in ~10-30% of breast cancer (BC) patients; triple negative breast cancer (TNBC), the most aggressive subtype of BC, exhibits the highest incidence of BM. Treatment options for BCBM are extremely limited due to the poor biological understanding. Hence, there is an urgent need to elucidate molecular mechanisms driving BCBM. We aim to study BCBM pathogenesis by dissecting the role of tumorigenic fucosylated proteins secreted by BM-associated fibroblasts (bmCAFs). Methods: We assessed global fucosylation (post-translational protein modification by L-fucose) patterns by fucose-binding lectin pulldown and immunoblot (IB) analysis. Conditioned media (CM) derived from CAFs that were depleted or not of fucosylated proteins were used to treat BC cells to assess motility and invasion. Fuco-proteomics and phosphoproteomic profiling identified bmCAF-secreted fucosylated (sf) proteins in bmCAF-derived CM and associated global signaling changes induced in BC cells, respectively. qRT-PCR was performed to analyze FUT11 levels under normoxia and hypoxia. Immunofluorescence and IB analyses of BC cells treated ± with bmCAF-derived CM was performed to validate PVR downstream signaling changes. Stereotactic intracranial implantation of BC cells alone or with ctrl/PVR knocked-down- bmCAFs was used for the BCBM mouse model. Results: We discovered that fucosylated proteins secreted uniquely by bmCAFs but not by normal-breast fibroblasts (NBF) or primary tumor-CAFs (tCAF), potently drives BC proliferation and invasiveness. Fucosylated proteomic profiling of the bmCAF secretome identified a soluble Polio Virus Receptor (PVR) isoform as uniquely upregulated, fucosylated, and secreted by bmCAFs. Of the 13 fucosyltransferases (FUTs), we found that bmCAFs upregulate FUT11, a key hypoxia-related gene. FUT11 and secreted fucosylated PVR (sfPVR) are significantly increased in bmCAFs exposed to hypoxia, consistent with the hypoxic environment of the brain. PVR can exist as transmembrane and secreted isoforms. Whereas transmembrane PVR is known to contribute to poor prognosis in a number of cancers by facilitating tumorigenic cell:matrix interactions and attenuating anti-tumor immunity, roles of secreted fucosylated PVR (sfPVR) in cancer are unknown. Our phosphoproteomics analyses of BC cells identified cellular adhesion/cytoskeletal and EPHA2 signaling as significantly modulated by bmCAF sfPVR to drive BC cell motility and invasiveness. Indeed, PVR knockdown abrogates the ability of bmCAFs to enhance BC growth/spread in the brain in a mouse model. Conclusion: Our data demonstrate that pathological hypoxia drives FUT11-mediated fucosylation and secretion of PVR by bmCAFs, which potently drives BC cells motility and invasiveness, representing a mechanism by which bmCAFs can promote BCBM. We expect our ongoing and further studies to advance our understanding of how sfPVR from bmCAFs promotes BCBM and to establish a basis for therapeutic targeting of PVR and/or use of fucosylated PVR and its signaling effectors as biomarkers. Citation Format: Emma Adhikari, Qian Liu, Viktoriya Marusyk, Victoria Lzumi, John M. Koomen, Andriy Marusyk, Eric Lau. Hypoxia-induced secretion of fucosylated PVR/CD155 from brain met-associated fibroblasts drives breast cancer invasive capacity by altering cell-cell contacts & focal adhesion [abstract]. In: Proceedings of the AACR Special Conference: Cancer Metastasis; 2022 Nov 14-17; Portland, OR. Philadelphia (PA): AACR; Cancer Res 2022;83(2 Suppl_2):Abstract nr B003.