HDAC inhibition turns cancer-cell quiescence into a druggable vulnerability against chemotherapy-resistant ovarian cancer
核心洞察
A University of Pittsburgh and Magee-Womens Research Institute study shows that inhibiting the NuRD complex, via MBD3 (搜索)/CHD4 (搜索) knockdown or HDAC inhibitors, drives ovarian cancer (搜索) cells into a reversible quiescent (G0) state.
Quiescent cells become dependent on proteasome and autophagy activity, so combining HDAC inhibitors with carfilzomib or chloroquine produces synergistic cell death, with synergy indices of 25–73 in vitro.
In an ovarian cancer (搜索) PDX model, vorinostat plus carfilzomib markedly limited tumor growth, with tumors undetectable in two of nine mice, and valproic acid plus chloroquine induced mild tumor regression.
A preclinical study from the University of Pittsburgh and Magee-Womens Research Institute & Foundation (搜索) identifies a two-drug strategy for killing quiescent, treatment-resistant cancer cells that can seed relapse after chemotherapy, according to a paper published August 13 in Cancer Gene Therapy. The work, led by corresponding author Ronald J. Buckanovich, proposes exploiting a vulnerability created by cancer-cell quiescence rather than waiting for residual cells to resume proliferation.
Most cytotoxic regimens preferentially target actively dividing cells, allowing slow-cycling or non-dividing cancer cells to survive treatment and potentially drive recurrence. The problem is particularly relevant in ovarian cancer (搜索), where high initial response rates to platinum-based chemotherapy are frequently followed by relapse. Approximately 324,400 new cases of ovarian cancer were diagnosed worldwide in 2022, and although most patients will have an initial response to surgery and chemotherapy, roughly 70% will relapse and develop chemotherapy-resistant disease.
Mechanism: NuRD complex downregulation drives quiescence
The researchers found that downregulation of the nucleosome remodeling and deacetylase (NuRD) complex — specifically its components MBD3 (搜索) and CHD4 (搜索) — promotes entry into a quiescent, drug-tolerant state. Single-cell RNA sequencing of primary quiescent ovarian cancer (搜索) cells showed downregulated mRNAs for MBD3 and CHD4, a finding confirmed by Western blot in three ovarian cancer cell lines (PT340, PT412, and HEY1). Genetic knockdown of either MBD3 or CHD4 decreased cell proliferation without changing cell viability and increased the proportion of cells in the G0 phase of the cell cycle, as measured by the Fucci fluorescent cell cycle reporter system.
Histone deacetylase inhibitors (HDACi) phenocopied this effect pharmacologically. Treatment with the pan-HDAC inhibitors vorinostat (SAHA), belinostat, or valproic acid (VPA), or the HDAC1/2/3/11-selective inhibitor mocetinostat, induced a quiescent state, with more than 90% of cells arrested in G0. The arrest was reversible, as vorinostat washout allowed cells to resume proliferation within approximately 24 hours. Mechanistically, HDAC inhibition altered chromatin accessibility at RHO/RAC GTPase-associated loci, suppressed the MRTFA (搜索)/SRF (搜索) pro-proliferative transcriptional axis, and drove cytoplasmic localization of MRTFA — a transcription factor that cooperates with SRF to drive cellular proliferation.
Quiescence creates a proteostasis dependency
Rather than directly killing these cells, HDAC inhibition helps establish a quiescent state that creates a second vulnerability: increased dependence on proteasome and autophagy activity to maintain protein homeostasis. Vorinostat-treated ovarian cancer (搜索) cells showed increased proteasome activity compared with DMSO controls, and the autophagy marker ATG5 was upregulated in both serum-starved and vorinostat-induced quiescent cells.
Combining HDACi with the proteasome inhibitor carfilzomib or the autophagy inhibitors hydroxychloroquine or chloroquine exploited that dependency and produced synergistic cell death. The HDACi-carfilzomib combination produced synergy indices of 25–73 under the highest single agent (HSA) model in vitro, despite neither drug being cytotoxic as a single agent. The chloroquine and valproic acid combination resulted in a 50% reduction in cell viability across three ovarian cancer (搜索) cell lines (PT340, PT412, and A2780).
Preclinical efficacy across models
In an ovarian cancer (搜索) patient-derived xenograft (PDX) mouse model, vorinostat plus carfilzomib produced greater tumor growth inhibition than either monotherapy or vehicle control across groups of 8–10 tumors, with tumors undetectable in two of nine mice. Immunohistological analysis showed that more than 90% of cells in control and carfilzomib groups were proliferative, while dual therapy-treated tumors showed a profound reduction in Ki67 staining and large areas of necrosis. In a separate in vivo study, valproic acid plus chloroquine prevented tumor growth and induced mild tumor regression.
HDACi-induced quiescence was also demonstrated in triple-negative breast cancer (搜索) patient-derived organoids and breast and lung cancer cell lines, suggesting the underlying biology may extend beyond ovarian cancer (搜索). In two patient-derived triple-negative breast cancer organoid models, the vorinostat-carfilzomib combination reduced viable cell number by 85–90%. Activity was additionally evaluated in an ex vivo co-culture model incorporating primary murine hepatocytes to approximate aspects of the liver microenvironment.
A core quiescence signature
By overlaying RNA-seq datasets from quiescent cancer cells with essential quiescence genes identified in yeast, the researchers identified 25 genes — including upregulated genes such as SLC7A8, WIPI1, and SERINC2, and downregulated genes such as CCNF, CCNB1, HSPA14, and IRAK1 — that they propose represent a conserved quiescent cell core signature. This signature was validated in quiescent cells across ovarian, lung, and breast cancer cell lines.
Clinical context and translational outlook
HDAC inhibitors including vorinostat and belinostat have previously been evaluated as single agents in ovarian cancer (搜索) with limited efficacy, consistent with the study's finding that HDAC inhibition can induce arrest without killing the cells. As the authors note, induction of quiescence would be expected to slow tumor growth or induce stable disease but would be unable to reduce tumor volume, which may explain why single-agent HDACi trials in solid tumors have not led to tumor regression. There is also clinical precedent for combining HDAC and proteasome inhibition: panobinostat plus bortezomib was previously FDA-approved for multiple myeloma (搜索), although panobinostat was subsequently withdrawn from the US market for commercial reasons.
The most immediately translatable strategy pairs the HDACi valproic acid with chloroquine. Both are repurposed drugs with extensive human clinical experience, potentially shortening the translational path compared with new molecular entities. The authors said they are planning a window-of-opportunity clinical study of the combination in ovarian cancer (搜索).
The approach remains preclinical, with relatively small in vivo groups, and established safety profiles for the individual drugs do not establish the safety or optimal dosing of the combination in ovarian cancer (搜索). The authors also acknowledge several limitations, including reliance on pharmacologic HDAC inhibition with broad genome-wide effects, a single PDX model with a relatively short treatment and follow-up period, and a core quiescence signature derived from transcriptomic overlap that has not been functionally verified.
If translated successfully, the strategy would turn a conventional obstacle in cancer treatment into a therapeutic vulnerability: rather than waiting for dormant residual cells to resume proliferation, it seeks to exploit the biological dependencies created by quiescence itself.