Lineage Plasticity in SCLC Generates Non-Neuroendocrine Cells Primed for Vasculogenic Mimicry

INTRODUCTION

Vasculogenic mimicry (VM), the process of tumor cell transdifferentiation to endow endothelial-like characteristics supporting de novo vessel formation, is associated with poor prognosis in several tumor types, including SCLC. In genetically engineered mouse models (GEMMs) of SCLC, NOTCH, and MYC co-operate to drive a neuroendocrine (NE) to non-NE phenotypic switch, and co-operation between NE and non-NE cells is required for metastasis. Here, we define the phenotype of VM-competent cells and molecular mechanisms underpinning SCLC VM using circulating tumor cell-derived explant (CDX) models and GEMMs.

METHODS

We analyzed perfusion within VM vessels and their association with NE and non-NE phenotypes using multiplex immunohistochemistry in CDX, GEMMs, and patient biopsies. We evaluated their three-dimensional structure and defined collagen-integrin interactions.

RESULTS

We found that VM vessels are present in 23/25 CDX models, 2 GEMMs, and in 20 patient biopsies of SCLC. Perfused VM vessels support tumor growth and only NOTCH-active non-NE cells are VM-competent in vivo and ex vivo, expressing pseudohypoxia, blood vessel development, and extracellular matrix organization signatures. On Matrigel, VM-primed non-NE cells remodel extracellular matrix into hollow tubules in an integrin β1-dependent process.

CONCLUSIONS

We identified VM as an exemplar of functional heterogeneity and plasticity in SCLC and these findings take considerable steps toward understanding the molecular events that enable VM. These results support therapeutic co-targeting of both NE and non-NE cells to curtail SCLC progression and to improve the outcomes of patients with SCLC in the future.

FOXC2 promotes vasculogenic mimicry and resistance to anti-angiogenic therapy

Vasculogenic mimicry (VM) describes the formation of pseudo blood vessels constructed of tumor cells that have acquired endothelial-like properties. VM channels endow the tumor with a tumor-derived vascular system that directly connects to host blood vessels, and their presence is generally associated with poor patient prognosis. Here we show that the transcription factor, Foxc2, promotes VM in diverse solid tumor types by driving ectopic expression of endothelial genes in tumor cells, a process that is stimulated by hypoxia. VM-proficient tumors are resistant to anti-angiogenic therapy, and suppression of Foxc2 augments response. This work establishes co-option of an embryonic endothelial transcription factor by tumor cells as a key mechanism driving VM proclivity and motivates the search for VM-inhibitory agents that could form the basis of combination therapies with anti-angiogenics.

Abstract 2653: Lineage plasticity in small cell lung cancer generates non- neuroendocrine cells primed for vascular mimicry

Small cell lung cancer (SCLC) is a highly aggressive neuroendocrine (NE) cancer with rapid and widespread metastasis and a dismal prognosis measurable only in months. The inherent lack of tumor biopsies has limited our understanding of SCLC phenotypic heterogeneity, and except for the addition of Immune Checkpoint Inhibitors the therapeutic strategy has remained largely unchanged for decades. We are investigating Vascular Mimicry (VM) in SCLC, which is linked to poor prognosis and may provide a novel therapeutic opportunity1. VM describes the ability of aggressive tumor cells to undergo trans-endothelial differentiation, enabling angiogenesis-independent tumor vascularization. We hypothesize that, via extracellular matrix (ECM) remodeling, VM provides a favorable tumor microenvironment (TME) to support tumor progression and metastasis.

We explored SCLC VM competency and molecular mechanisms underpinning this using histopathology and transcriptomic and proteomic analyses of NE and non-NE SCLC cells in patient faithful Circulating Tumor Cell (CTC)-Derived eXplant (CDX)2,3 models and genetically-engineered mouse models (GEMMs).

We demonstrate that within human and murine SCLC tumors in vivo and ex vivo the non-NE cells, unlike NE cells, are VM competent. NE SCLC cells in an ASCL1-expressing CDX model undergo a Notch-driven transition to a non-NE phenotype. We show that non-NE cells are transcriptionally primed to undergo VM and express gene signatures consistent with pseudohypoxia, blood vessel development and extracellular matrix (ECM) organization. These cells undergo post-translational modification of proteins involved in cell-cell and cell-ECM adhesion including a VM-specific glycosylation of integrin β1. On Matrigel VM-primed non-NE cells remodeled the ECM forming tubular networks, an established surrogate assay for VM competence. Integrin β1 blockade impedes VM in vitro by inhibiting non-NE cell collagen adhesion and thus ECM remodeling.

This previously unrecognized functional intratumoral heterogeneity in SCLC supports therapeutic co-targeting of NE and non-NE cells to curtail VM-supported tumor growth and metastasis. The ultimate aim is to identify targetable therapeutic options with which to inhibit VM to improve outcomes for patients with SCLC.

References

1S. C. Williamson et al., Vasculogenic mimicry in small cell lung cancer. Nat Commun 7, 1 (2016).

2C. L. Hodgkinson et al., Tumorigenicity and genetic profiling of circulating tumor cells in small-cell lung cancer. Nature medicine 20, 897 (2014).

3K. L. Simpson et al., A biobank of small cell lung cancer CDX models elucidates inter-and intratumoral phenotypic heterogeneity. Nature Cancer 1, 4 (2020).

Citation Format: Sarah M. Pearsall, Stuart C. Williamson, Fernando García Marqués, Sam Humphrey, Ellyn Hughes, Ian Cannell, Kristopher K. Frese, Melanie Galvin, Mathew Carter, Lynsey Priest, Alastair Kerr, Martin J. Humphries, Jonathon D. Humphries, Fiona Blackhall, Gregory Hannon, Julien Sage, Sharon J. Pitteri, Kathryn L. Simpson, Caroline Dive. Lineage plasticity in small cell lung cancer generates non- neuroendocrine cells primed for vascular mimicry [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2653.

The Rare YAP1 Subtype of SCLC Revisited in a Biobank of 39 Circulating Tumor Cell Patient Derived Explant Models: A Brief Report

INTRODUCTION

Recent consensus defines four SCLC subtypes on the basis of transcription factor expression: ASCL1, NEUROD1, POU2F3, and YAP1. The rare YAP1 subtype is associated with "neuroendocrine (NE)-low" cells among SCLC cell lines and patient samples. We evaluated YAP1 in 39 patients with phenotypically diverse circulating tumor cell-derived explant (CDX) models and revisited YAP1 in terms of prevalence, cell phenotype, and intertumor and intratumor heterogeneity.

METHODS

YAP1 transcript and protein expression were assessed by RNA sequencing and immunohistochemistry or multiplexed immunofluorescence of NE and non-NE CDX subpopulations. Physically separated NE and non-NE CDX ex vivo culture lysates were Western blotted for YAP1, NE marker SYP, and AXL.

RESULTS

RNA sequencing normalized for the four subtype transcription factors identified YAP1 expression in 14 of 39 CDX. A total of 10 CDX expressed YAP1 protein, and eight had strong YAP1 expression confined to rare non-NE cell clusters. This was confirmed in ex vivo CDX cultures in which adherent non-NE cells lacking SYP expression expressed YAP1. However, in two CDX, weaker cellular YAP1 expression was observed, widely dispersed in SYP-positive NE cells.

CONCLUSIONS

YAP1 was predominantly expressed in non-NE cell clusters in SCLC CDX, but two of 39 CDX expressed YAP1 in NE cells. CDX22P, with relatively high YAP1 expression, is an ASCL1 NE subtype with a low NE score and an outlier within this subtype in our CDX biobank. These descriptive data reveal subtly different YAP1 expression profiles, paving the way for functional studies to compare YAP1 signaling in non-NE and low NE cell contexts for potentially personalized therapeutic approaches.

Assessment of compliance with home cardiorespiratory monitoring in infants at risk of sudden infant death syndrome. Collaborative Home Infant Monitoring Evaluation (CHIME)

OBJECTIVES

Documented monitoring was used to evaluate prospectively (1) the level of compliance among infants in whom cardiorespiratory monitoring was clinically indicated and (2) factors that might influence compliance: diagnosis, socioeconomic status, maternal age and education, and alarms.

STUDY DESIGN

Sixty-seven infants (51% female, 49% term) were sequentially enrolled, and monitoring was prescribed for the following indications: siblings of sudden infant death syndrome victims (16%), apnea of prematurity (45%), and apparent life-threatening events or apnea of infancy (39%). Demographic data, alarm and event data, and a summary report of monitor use from the first monitor download were obtained.

RESULTS

Maternal age, education, and insurance status did not differ significantly by indication for monitoring. The median number of monitor alarms per 10 hours of use was 0.7 for apnea or bradycardia and 0.6 for loose lead alarms. Monitors were available for use in the home from 2 to 106 days (median, 11 days). Median hours of monitor use per full day in the home was 15.5 hours. Of 67 infants, 58 used the monitor for at least part of every day in the home. The number of hours of monitor use per day did not differ significantly by diagnostic category, chronologic age, alarms, maternal age, education, or insurance type. This study population of infants at increased risk of sudden infant death syndrome had excellent compliance; 75% of the infants were monitored more than 10.5 hours per day, and 25% were monitored more than 21 hours per day.

CONCLUSIONS

Documented monitoring provides an objective measure of compliance. These data provide a potential goal for level of compliance with home cardiorespiratory monitoring.

Polysomnography in obese children with a history of sleep-associated breathing disorders

We hypothesized that obese children with a history of breathing difficulty during sleep would demonstrate (1) evidence of complete and partial obstructive sleep apnea (OSA) with hypercarbia and/or hypoxemia; and (2) correlation between symptoms, degree of obesity, adenoid and tonsil size, and polysomnography (PSG) results. We evaluated 32 obese children [% ideal body weight (IBW), 196 +/- 45%] with a sleep history questionnaire, airway radiographs, electrocardiograms (ECG), and PSG. By history, we found snoring (100%), difficulty breathing (59%), sweating (44%), restlessness (53%), arousals (41%), apnea (50%), worsening with upper respiratory infection (URI) (81%), hypersomnolence (59%), and mouth breathing (59%). We found adenoid and/or tonsil enlargement on 75% of airway x-ray pictures. ECGs were abnormal in 5 patients. Among all patients, mean sleep study oxyhemoglobin saturation (SaO2) was 85 +/- 16% and mean end-tidal CO2 (PetCO2) was 51 +/- 7 torr; 84% had paradoxical inward movement of the chest on inspiration, 59% had OSA, and 66% had partial OSA. In those with > or = 200% IBW and adenotonsillar enlargement, elevated PetCO2 and the presence of hypoxemia (SaO2 < 90%) for > or = 5% of the total sleep time (TST) were correlated, unlike in patients of similar weight but without adenotonsillar enlargement. Individuals symptoms did not correlate with the severity of PSG abnormalities. By discriminant analysis, using three variables (IBW, presence of adenotonsillar tissue, and presence of > or = 5 symptoms), we could predict PSG abnormalities with up to 81% reliability. Our findings indicate that in obese children, particularly those with %IBW > or = 200 and adenotonsillar hypertrophy, with sleep-disordered breathing evaluation by polysomnography should be considered.