PRC2-mediated repression of SMARCA2 predicts EZH2 inhibitor activity in SWI/SNF mutant tumors Thomas Januarioa,1, Xiaofen Yea,1, Russell Bainerb, Bruno Alickec, Tunde Smitha, Benjamin Haleyd, Zora Modrusand, Stephen Gouldc, and Robert L. Yaucha,2 aDepartment of Discovery Oncology, Genentech, Inc., South San Francisco, CA 94080; bDepartment of Bioinformatics, Genentech, Inc., South San Francisco, CA 94080; cDepartment of Translational Oncology, Genentech, Inc., South San Francisco, CA 94080; and dDepartment of Molecular Biology, Genentech, Inc., South San Francisco, CA 94080 Edited by Joan S. Brugge, Harvard Medical School, Boston, MA, and approved October 3, 2017 (received for review March 8, 2017) Subunits of the SWI/SNF chromatin remodeling complex are fre- of the ovary, hypercalcemic-type (SCCOHT) (3, 7–9). Although the quently mutated in human cancers leading to epigenetic dependen- mechanisms underlying tumorigenesis in these specific contexts cies that are therapeutically targetable. The dependency on the have yet to be fully elucidated, data are further supportive of a polycomb repressive complex (PRC2) and EZH2 represents one such tumor-suppressive function (10, 11). vulnerability in tumors with mutations in the SWI/SNF complex Efforts to therapeutically target SWI/SNF-defective cancers subunit, SNF5; however, whether this vulnerability extends to other have focused on identifying novel vulnerabilities that may be a SWI/SNF subunit mutations is not well understood. Here we show consequence of the altered chromatin state caused by muta- that a subset of cancers harboring mutations in the SWI/SNF ATPase, tions in BAF complex subunits. One such described vulnera- bility was based on the initial discovery in Drosophila of an SMARCA4, is sensitive to EZH2 inhibition. EZH2 inhibition results in a opposing and antagonistic role of BAF and polycomb com- heterogenous phenotypic response characterized by senescence plexes in regulating gene expression (12). Subsequent studies and/or apoptosis in different models, and also leads to tumor revealed that this antagonistic relationship may result in human growth inhibition in vivo. Lower expression of the SMARCA2 cancers with specific defects in BAF subunits to become de- paralog was associated with cellular sensitivity to EZH2 inhibition pendent on the activity of the polycomb repressive group 2 in SMARCA4 mutant cancer models, independent of tissue deriva- (PRC2) complex. This is best exemplified in MRTs, as loss of tion. SMARCA2 is suppressed by PRC2 in sensitive models, and SNF5 results in the altered genomic occupancy of the repressive induced SMARCA2 expression can compensate for SMARCA4 and chromatin mark deposited by PRC2 at histone H3 lysine 27 resi- antagonize PRC2 targets. The induction of SMARCA2 in response to dues (H3K27me3), leading to the repression of lineage-specific EZH2 inhibition is required for apoptosis, but not for growth arrest, targets (13). In these models, disruption of the histone methyl- through a mechanism involving the derepression of the lysomal transferase activity of EZH2, the catalytic subunit of PRC2, im- protease cathepsin B. Expression of SMARCA2 also delineates EZH2 paired tumor growth, thereby demonstrating that SNF5 mutant inhibitor sensitivity for other SWI/SNF complex subunit mutant tumors depend on EZH2 activity (13, 14). More recently, a similar tumors, including SNF5 and ARID1A mutant cancers. Our data support dependency on EZH2 was described in the context of ARID1A monitoring SMARCA2 expression as a predictive biomarker for EZH2- mutant cancers, suggesting that PRC2 activity may be a common targeted therapies in the context of SWI/SNF mutant cancers. vulnerability in SWI/SNF-defective lesions (15). However, whether targeting EZH2 will be effective in all cancers harboring these EZH2 | SMARCA4 | SMARCA2 | PRC2 | ARID1A specific mutations or in other SWI/SNF subunit mutant contexts remains an open question. WI/SNF, or BAF (Brg/Brahma-associated factors), com- Significance Splexes compose a family of ATP-dependent chromatin remodeling complexes that play critical roles in controlling gene transcription and DNA repair through their ability to regulate Targeting epigenetic dependencies caused by mutations in MEDICAL SCIENCES the accessibility of DNA in the nucleus (1, 2). In mammals, these chromatin-modifying enzymes represents a novel therapeutic complexes are composed of approximately 15 subunits, including approach in cancer. Notably, cancers harboring mutations in the either the SMARCA4 or SMARCA2 ATPase and additional SNF5 subunit of the SWI/SNF chromatin remodeling complex core and accessory subunits whose specialized protein domains have been shown to be susceptible to small-molecule inhibitors enable interactions with chromatin and/or DNA substrates. SWI/ of the EZH2 histone methyltransferase that are currently in clin- SNF complexes notably control lineage specification and dif- ical development. We demonstrate that EZH2 inhibition can be ferentiation programs in tissues, and these highly specific func- effective in SMARCA4 mutant cancers that concurrently tran- tions can be achieved through combinatorial subunit assembly, as scriptionally silence the paralog helicase SMARCA2. SMARCA2 is well as tissue/lineage-specific expression of some subunits. directly suppressed by EZH2, and SMARCA2 expression levels Large-scale cancer genome sequencing efforts have revealed that predict EZH2 inhibitor activity in other SWI/SNF mutant contexts, multiple SWI/SNF complex proteins are recurrently mutated in including ARID1A mutant tumors. These data provide insight into ∼20% of human cancers (3). Mutations often render the encoded the utility of EZH2 inhibitors in SWI/SNF mutant tumors and have protein nonfunctional and associate with unique tumor spectra, important implications regarding predictive diagnostics. suggesting distinct tumor-suppressor functions across these cancers. For example, the SWI/SNF complex subunit SNF5 is subject to Author contributions: T.J., X.Y., R.B., S.G., and R.L.Y. designed research; T.J., X.Y., B.A., T.S., B.H., Z.M., S.G., and R.L.Y. performed research; T.J., X.Y., R.B., B.A., T.S., and R.L.Y. biallelic mutational inactivation in nearly all malignant rhabdoid analyzed data; and R.L.Y. wrote the paper. tumors (MRTs), and studies in genetically engineered mouse models of SNF5 inactivation have supported its role as a bona fide The authors declare no conflict of interest. tumor suppressor (3–6). Similarly, recurrent inactivating mutations This article is a PNAS Direct Submission. in additional BAF subunits have been identified in other indica- Published under the PNAS license. tions, including ARID1A mutations in ovarian clear-cell and 1T.J and X.Y. contributed equally to this work. endometroid cancers, as well as hepatocellular and gastric cancers. 2To whom correspondence should be addressed. Email: [email protected]. SMARCA4 mutations likewise are observed across a spectrum of This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. cancers of the lung and bladder, and nearly all small cell carcinomas 1073/pnas.1703966114/-/DCSupplemental. www.pnas.org/cgi/doi/10.1073/pnas.1703966114 PNAS | November 14, 2017 | vol. 114 | no. 46 | 12249–12254 Downloaded by guest on September 30, 2021 Multiple inhibitors targeting the enzymatic activity of EZH2 are AB currently in clinical development, with EPZ-6438 (tazemontstat) 20 20 representing the most clinically advanced molecule. Early clinical 15 7 days 15 13 days 10 10 data presented with EPZ-6438 has shown promise, as objective 5 5 4 4 responses have been observed in a subset of SNF5 mutant and H1299 TOV-112D TOV-112D 3 3 SMARCA4 mutant patients treated with EPZ-6438 as a single 2 2 1 1 Caspase 3/7 activity (fc) agent. These data not only begin to provide early clinical proof of 0 0 NA EPZ-6438: concept, but indicate that EZH2 inhibition may be effective in the A549 SNU-484 A549 A549 HCC15 context of SMARCA4 mutant cancers, a preclinical finding that has HCC15 COV434 COV434 SNU-484 SNU-484 UM-UC-3 UM-UC-3 NCI-H522 NCI-H661 NCI-H522 NCI-H661 TOV-112D TOV-112D NCI-H1703 NCI-H1299 NCI-H1568 yet to be published. Notably, not all patients with tumors harboring NCI-H1703 NCI-H1299 NCI-H1568 EPZ-6438 EPZ-6438 EPZ-6438 EPZ-6438 SNF5 or SMARCA4 defects responded to therapy, suggesting that sensitive resistant sensitive resistant identifying a biomarker predictive of response to EZH2 inhibition NCI-H1568 could provide significant benefit. In the present study, we demon- NCI-H1703 C control EPZ-6438 strate that EZH2 inhibition is effective in a subset of SMARCA4 mutant cancer models, and that the PRC2-mediated transcriptional UM-UC-3 NCI-H522 suppression of the paralog ATPase, SMARCA2, can predict the Cov-434 preclinical activity of EZH2 inhibitors. Importantly, we show that the level of SMARCA2 expression may be a global predictive HCC-15 COV434 biomarker of EZH2 activity in other BAF mutant cancers. NCI-H522 Results A Subset of SMARCA4 Mutant Cancers Is Responsive to EZH2 Inhibition. NCI-H661 We evaluated the effect of EZH2 inhibition using the EZH2- NCI-H1568 targeting histone methyltransferase inhibitor, EPZ-6438, on clo- SNF5mut control EPZ-6438 ( M): D vehicle nogenic growth across a panel of 11 SMARCA4 mutated cancer EPZ-6438 (30mg/kg) cell lines derived from different tumor types (Dataset S1). A dose- 500 EPZ-6438 (100mg/kg) 0 .06 .18 ) 3 EPZ-6438 (450mg/kg) dependent inhibition of clonogenic growth independent of tissue 400 derivation was observed in a subset of these SMARCA4 mutant cell .56 1.67 5 lines (Fig. 1A). The degree of growth inhibition on EPZ-6438 300 treatment was similar to that previously described in models 200 100 characterized by mutations in SNF5 (G401). No activity was ob- (mm volume tumor = 0 served in a panel of SWI/SNF wild-type models (n 8). The dif- 0 102030 ferential sensitivity to EPZ-6438 was not due to differences in Day target engagement, as a similar dose-dependent inhibition of H3K27 methylation was observed in sensitive and resistant models Fig.