ARID1A Governs Genomic Stability and Proliferation in SCLC via c-MYC/PARP1 Suppression Driving Vulnerability to BET Inhibitors

Authors

Cao G

Ma L

Dai X

Hou P

Yao X

Li G

Zhang J

Chen C

Lin W

Doi

10.34133/research.0908

PMID: 22037554 · DOI: 10.34133/research.0908 · Journal: Research (2025)

TL;DR

This study establishes ARID1A as a tumor suppressor in SCLC that transcriptionally represses MYC and PARP1, acting as both a gatekeeper of cell proliferation and a caretaker of genome stability. ARID1A loss triggers replication stress, DNA double-strand breaks, and PI3K/AKT pathway activation. The authors demonstrate that ARID1A deficiency creates a therapeutic vulnerability to the BET inhibitor JQ1, and identify the ARID1A-targeting compound BRD-K98645985, which shows single-agent antitumor activity and synergizes with JQ1 to suppress SCLC progression in vitro and in vivo.

Cohort & data

  • Somatic mutation profiles of ARID1A and ARID1B analyzed in 249 SCLC clinical specimens from cBioPortal (Gardner 2017, CLCGP, JHU, UCOLOGNE 2015 datasets).
  • Mutation spectrum of ARID1A and ARID1B assessed in 52 SCLC cell lines.
  • RNA-seq data from 50 SCLC and 75 LUAD cell lines from CCLE (DepMap).
  • Expression data from 81 and 69 human primary SCLC tissues and 7 nontumor lung tissues (George et al. 2015; GSE60052).
  • Microarray data of 18 primary human SCLC and adjacent noncancerous tissues (GSE149507).
  • Transcriptomic profiles of 107 SCLC cases with paired normal tissues (GSA-Human accession HRA003419).
  • Protein expression profiles from 112 SCLC subjects with matched controls (OMIX002489).
  • Kaplan-Meier survival analysis: n=41 for OS and n=33 for PFS.
  • GDSC1 dataset for JQ1 IC50 correlation analysis.
  • In vivo xenograft models using DMS273 and H446 SCLC cell lines in nude mice.

Key findings

  • ARID1A exhibits a low somatic mutation rate (2.9%) in SCLC, unlike ovarian clear cell carcinoma and gastric carcinoma where it is frequently mutated.
  • SCLC cell lines display markedly higher ARID1A expression compared to LUAD cell lines (P < 0.0001, CCLE data).
  • High ARID1A expression correlates with better OS in SCLC patients (median 48 vs 20 months, P = 0.008, n=41, log-rank test) and trends toward better PFS (median 22 vs 12 months, P = 0.078, n=33).
  • ARID1A knockdown promotes cell proliferation, colony formation, and tumor growth in vivo; overexpression suppresses these phenotypes in DMS273, DMS53, and H446 cell lines.
  • ARID1A transcriptionally represses MYC and PARP1 by binding their promoters, confirmed by ChIP-PCR and dual-luciferase reporter assays.
  • ARID1A depletion triggers elevated gamma-H2AX foci (DNA double-strand breaks), increased RAD51 expression, phosphorylation of CHEK1 and RPA2, and activation of the PI3K/AKT1 pathway.
  • Positive correlation between JQ1 IC50 values and ARID1A expression levels in SCLC (r = 0.368, P = 0.032, GDSC1 dataset), indicating enhanced JQ1 sensitivity upon ARID1A loss.
  • BRD-K98645985 (targeting the ARID1A-containing BAF complex) shows potent single-agent antitumor activity and Bliss synergy with JQ1 in vitro and in vivo.
  • Combination of JQ1 and BRD-K98645985 achieves superior tumor suppression compared to either agent alone in xenograft models (P < 0.0001 for combination vs PBS).

Genes & alterations

  • ARID1A — Tumor suppressor in SCLC; low mutation rate (2.9%); high expression correlates with survival advantage; transcriptionally represses MYC and PARP1 via promoter binding and chromatin remodeling (H3K27me3 enrichment, H3K27Ac reduction).
  • MYC — Transcriptionally upregulated upon ARID1A loss; promotes cell proliferation and genome instability; c-MYC overexpression partially rescues ARID1A-mediated growth suppression.
  • PARP1 — Transcriptionally upregulated upon ARID1A loss; PARP1 overexpression partially rescues ARID1A-mediated growth suppression; elevated PARP1 counteracts MYC-induced genome instability.
  • RAD51 — Induced upon ARID1A depletion as part of the DNA damage response; suppressed upon ARID1A overexpression; JQ1 treatment dose-dependently reduces RAD51 expression.
  • CHEK1 — Phospho-CHK1 robustly activated upon ARID1A depletion; reversed by ARID1A, MYC, or PARP1 re-expression; JQ1 suppresses p-CHK1 levels.
  • ARID1B — Low mutation rate (3.0%) in SCLC, characterized alongside ARID1A.
  • BRD4 — BET family member; BRD4-binding sites increase in ARID1A-knockout cells; therapeutic target of JQ1.
  • AKT1 / PIK3CA — PI3K/AKT pathway robustly activated (elevated p-AKT, p-S6RP) upon ARID1A silencing; reversed by ARID1A overexpression.

Clinical implications

  • ARID1A expression may serve as a prognostic biomarker for SCLC: high ARID1A expression associates with significantly improved OS (HR not explicitly stated; median OS 48 vs 20 months, P = 0.008).
  • ARID1A-low/deficient SCLC tumors may represent a therapeutically targetable subgroup with enhanced sensitivity to BET inhibitors (JQ1), establishing a potential synthetic lethal strategy.
  • The combination of BRD-K98645985 (ARID1A-targeting BAF complex inhibitor) and JQ1 demonstrates synergistic antitumor activity, suggesting a novel combinatorial therapeutic strategy for SCLC.
  • No significant adverse effects on normal mouse growth were observed with single-agent or combination therapy, supporting a favorable safety profile for translational development.

Limitations & open questions

  • All in vivo efficacy data are from subcutaneous xenograft models in nude mice, which do not recapitulate the immune microenvironment or metastatic biology of SCLC.
  • The survival analysis cohort is small (n=41 for OS, n=33 for PFS), and PFS did not reach statistical significance (P = 0.078).
  • JQ1 is a tool compound with known pharmacokinetic limitations; clinical BET inhibitors (e.g., molibresib, birabresib) would need to be tested for translational relevance.
  • BRD-K98645985 is a recently described compound and lacks clinical pharmacology data; its specificity and off-target effects in human tissues remain to be characterized.
  • The study does not address SCLC molecular subtypes (ASCL1, NEUROD1, POU2F3, YAP1) and whether ARID1A effects are subtype-dependent.
  • The mechanism by which ARID1A loss activates the PI3K/AKT pathway was not fully elucidated; the authors note this warrants further investigation.

Citations from this paper used in the wiki

  • “Patients with high ARID1A expression had a better overall survival (OS) (median survival of 48 months of the high ARID1A group versus 20 months of the low ARID1A group, P = 0.008, n = 41, log-rank test) and progression-free survival (PFS) (median survival of 22 months of the high ARID1A group versus 12 months of the low ARID1A group, P = 0.078, n = 33, log-rank test).”
  • “ARID1A transcriptionally represses c-MYC and PARP1 expression. ARID1A depletion triggered replication stress response (RSR), DNA double-strand breaks (DSBs), and PI3K/AKT pathway activation.”
  • “A positive correlation was observed between the IC50 values of JQ1 and ARID1A expression levels in SCLC (r = 0.368, P = 0.032, GDSC1 dataset), suggesting that cellular sensitivity to JQ1 treatment may be enhanced by ARID1A KD.”
  • “Combination therapy with BRD and JQ1 demonstrated superior therapeutic efficacy compared to JQ1 monotherapy, with Bliss independence analysis confirming a robust synergistic effect in vitro.”

This page was processed by paper-compiler on 2026-05-06.