Lung cancer can be divided in two main types: non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC). Smoking is the major risk factor for both subtypes and symptoms are similar, including shortness of breath and a persistent cough, but therapeutic intervention differs. SCLC is particularly recalcitrant to chemotherapeutic treatment: chemoresistance rapidly occurs after an initial chemosensitivity. No targeted therapies are currently available, and the low frequency of oncogenic driver mutations limits the druggable targets. However, SCLC exhibits frequent mutations in chromatin-regulating genes, suggesting that targeting the epigenetic remodeling is a promising strategy. Among these targets, Lysine-specific histone demethylase 1A (LSD1) is highly expressed in SCLC and is known to silence the transcription of target genes. LSD1 inhibitors showed efficacy in preclinical models of SCLC, but the underlying mechanism is unknown. The MacPherson lab, Human Biology Division, recently proposed in Science Signaling that the modulation of NOTCH signaling by LSD1 inhibition is key to the therapeutic efficacy.
As the objective of the study was to understand the efficacy of LSD1 inhibition, Dr. Arnaud Augert and colleagues started by screening SCLC cell lines for the antiproliferative activity of ORY-1001, a selective inhibitor of LSD1. RNAseq analysis of these cell lines after treatment revealed that the NOTCH signaling pathway was one of the most dysregulated by treatment; many components of the pathway were upregulated upon LSD1 inhibition. In accordance with these results, Achaete-scute homolog 1 (ASCL1), transcriptionally suppressed by NOTCH signaling, was downregulated after LSD1 inhibition. Interestingly, ASCL1 is a transcription factor that promotes neuroendocrine transcriptional programs, which are reactivated in SCLC.
However, it is well known that tumor cell lines rapidly diverge from the originating tumor when cultured for a long period of time. Thus, when the authors started to dig further into the mechanism, they decided to use a model in which cells would be closer to their initial status in the tumor. They isolated tumor cells from SCLC patient-derived xenograft (PDX) samples, in which patient tumor cells are passaged in vivo in mice, thereby limiting selection pressures and maintaining heterogeneity. Exposure of these cells to different concentrations of ORY-1001 allowed them to determine the IC50 of ORY-1001 for each and classify them as sensitive or resistant. Co-first author of the study Dr. Emily Eastwood explains: “When the therapeutic potential of LSD1 inhibition was examined across a panel of PDX models, a complete and durable response was observed in one exceptionally responding PDX model. These findings were exciting, because I generated this exceptionally responding model from a chemoresistant patient – a patient with few therapeutic options.” When analyzing these cells by Western Blot (WB) and RNAseq, the authors demonstrated that sensitive cells robustly overexpress NOTCH and downregulate ASCL1 after LSD1 inhibition, whereas resistant cells respond to LSD1 inhibition with modest changes in NOTCH/ASCL1. Sequencing after chromatin immunoprecipitation (ChIP-seq) revealed that LSD1 binds to the NOTCH locus and that LSD1 inhibition results in increased acetylation of the lysine 27 in histone 3 (H3K27Ac), a landmark for transcriptional activation.
In order to assess the importance of each of the components of the LSD1-NOTCH-ASCL1 axis in ORY-1001 activity, the researchers used small hairpin RNA (shRNA) against LSD1 and ASCL1 in sensitive cells. Whereas both shLSD1 and shASCL1 cells exhibited reduced proliferation and viability compared to controls, only shLSD1 led to an increase NOTCH signaling as assessed by WB. Similarly, inducible expression of the NOTCH intracellular domain (NICD), independent of extracellular signaling, was sufficient to repress ASCL1 and prevent tumor cell proliferation. As an ultimate demonstration of the robustness of the proposed mechanism, pharmacological inhibition of the NOTCH signaling using gamma-secretase inhibitors (gamma-secretases cleave and release the NICD upon NOTCH receptor activation) partially rescued the anti-tumor effect of ORY-1001. Dr. Augert concludes: “We show that a selective LSD1 inhibitor ORY-1001 leads to the activation of NOTCH signaling, a tumor suppressive pathway in SCLC. Activation of NOTCH signaling results in the suppression of the transcription factor ASCL1, a master regulator of neuroendocrine cell fate. The suppression of ASCL1 ultimately leads to the repression of SCLC tumorigenesis”.
Furthermore, NOTCH signaling activation and ASCL1 inhibition was also observed in vivo when PDX were treated with ORY-1001, leading to increased apoptosis and slower tumor growth. However, the differential response between responders to ORY-1001 and non-responders is still unclear. This is an ongoing work in the MacPherson lab, Dr. Emily Eastwood says: “We still need to understand the underlying reason why some SCLC models robustly activate NOTCH and suppress ASCL1. I hope to find additional exceptional responders by screening through other PDX models that are still in the pipeline and link exceptional response with other genetic mutations”.
This work was supported by National Institutes of Health, a Seattle Translational Tumor Research Grant, the Lung Cancer Research Foundation, Roche and a University of Washington Cancer Consortium Support Grant.
Cancer Consortium faculty members Drs David MacPherson, Renato Martins, Keith Eaton and Denny Liggitt contributed to this research.
Augert A, Eastwood E, Ibrahim AH, Wu N, Grunblatt E, Basom R, Liggitt D, Eaton KD, Martins R, Poirier JT, Rudin CM, Milletti F, Cheng WY, Mack F, MacPherson D. 2019. Targeting NOTCH activation in small cell lung cancer through LSD1 inhibition. Sci Signal. 12(567). doi: 10.1126/scisignal.aau2922.