Loss of HRD functional phenotype impedes immunotherapy and can be reversed by HDAC inhibitor in ovarian cancer

Dong Yang, Fuxue Huang, Wei Wei, Qiaqia Li, Junwan Wu, Yun Huang, Zhi‐Ling Li, Hai‐Liang Zhang, Xuan Li, Qiu-Er Yuan, Qingshan Chen, Gong‐Kan Feng, Rong Deng, Jun‐Dong Li, Xiao‐Feng ZhuInternational Journal of Biological Sciences 2023 · DOI

TL;DR

This study introduces "HRD-EXCUTE," a novel transcriptomic signature that identifies ovarian cancer patients with homologous recombination deficiency (HRD) who are likely to respond to immunotherapy. They found that a loss of this HRD-EXCUTE phenotype renders HRD tumors immunologically "cold" and unresponsive to treatment, but this can be reversed by combining PARP inhibitors (PARPi) with HDAC inhibitors (HDACi), thereby enhancing anti-tumor immunity.

What Problem Does This Paper Address?

Homologous recombination deficiency (HRD) is a hallmark of many cancers, including about 50% of ovarian cancers. HRD is known to make cancer cells susceptible to DNA damage therapies like PARP inhibitors (PARPi). It was also hypothesized that HRD tumors might be more responsive to immunotherapy due to increased genomic instability leading to more mutations and neoantigens. However, clinical trials have shown that HRD status alone isn't a reliable predictor of immunotherapy response, and many HRD patients don't benefit as expected. This suggests that simply having HRD isn't enough; a functional "immune-permissive" phenotype is also crucial. The paper aims to identify a more precise biomarker that captures this functional aspect of HRD and to find therapeutic strategies that can enhance immunotherapy efficacy in HRD patients who lack this favorable phenotype.

Methods

The researchers developed a novel metric called HRD-EXCUTE (HRD-EXpressed Transcriptomic Functional Phenotype). This was derived by:

  1. Differential Gene Expression Analysis: Comparing gene expression between HRD-high and HRD-low ovarian cancer samples from the TCGA dataset.
  2. Identifying Hub Genes: Using protein-protein interaction networks (STRING and Cytoscape) to pinpoint 15 "hub" genes that were consistently upregulated in HRD-high samples and enriched in immune-related pathways.
  3. Calculating HRD-EXCUTE: Defining HRD-EXCUTE as the average expression level of these 15 hub genes.
  4. Prognostic and Predictive Modeling: Employing decision tree analysis to correlate HRD-EXCUTE with patient survival and response to immunotherapy in various clinical cohorts.
  5. Drug Screening: Using the Connectivity Map (CMAP) database to identify drugs that could induce gene expression profiles similar to HRD-EXCUTE.
  6. Mechanism Investigation: Conducting in vitro and in vivo experiments to explore how drug combinations (PARPi + HDACi) influence HRD-EXCUTE, immune cell infiltration, and tumor growth. This involved techniques like Western blotting, RNA sequencing, and flow cytometry.

Datasets

The study utilized several publicly available datasets:

  • TCGA (The Cancer Genome Atlas): OV (ovarian cancer), LUSC (lung squamous cell carcinoma), and CESC (cervical squamous cell carcinoma) cohorts for transcriptomic and clinical data.
  • Affymetrix transcriptomic arrays: Data from an ovarian cancer cohort (GSE18520, GSE30161, GSE19829, GSE63885, GSE26193, GSE9891, GSE14764, GSE23554, GSE26712, GSE31245) to confirm HRD-EXCUTE's prognostic value.
  • IMvigor210 cohort: A well-studied dataset of patients with metastatic urothelial cancer treated with atezolizumab (anti-PD-L1), used to assess immunotherapy efficacy. Since HRD score wasn't available, they used FMOne mutation burden as a proxy for HRD.
  • I-SPY2 trial: A dataset for high-risk HER2-negative breast cancer treated with durvalumab (anti-PD-L1) combined with olaparib and paclitaxel. They used a PARPi7 signature (7 DNA repair genes) as a proxy for HRD.
  • CMAP database: For identifying drugs that induce HRD-EXCUTE-like gene expression signatures.
  • GSE131140 and GSE122153: ChIP-seq data for analyzing histone acetylation.

Key Insights and Ideas

  • HRD is not enough; functional phenotype matters: The study elegantly demonstrates that simply having HRD doesn't guarantee a good response to immunotherapy. A specific functional transcriptomic phenotype (HRD-EXCUTE) is critical for immune engagement.
  • HRD-EXCUTE as a prognostic and predictive biomarker: HRD+ patients with a high HRD-EXCUTE score showed significantly better survival and immune-inflamed tumor microenvironments, while those with low HRD-EXCUTE were more like HRD-negative patients. This suggests HRD-EXCUTE can stratify HRD patients for immunotherapy.
  • Loss of HRD-EXCUTE leads to immune "coldness": Tumors lacking the HRD-EXCUTE phenotype are characterized by lower mutation burden, fewer neoantigens, and reduced immune cell infiltration, making them resistant to immunotherapy.
  • HDAC inhibitors can "reawaken" HRD-EXCUTE: The study identified HDAC inhibitors (like Entinostat) as key agents that can reverse the loss of the HRD-EXCUTE phenotype. This is achieved by promoting histone acetylation and activating the cGAS-STING pathway.
  • Synergy of PARPi and HDACi for immunotherapy: Combining PARPi (like Niraparib or Olaparib) with HDACi (like Entinostat) significantly boosts HRD-EXCUTE, enhances tumor immunogenicity, and leads to improved anti-tumor immunity in vivo, even in immunodeficient models, highlighting the immune-mediated effects.

Key Results

  • HRD-EXCUTE identified 15 hub genes: These genes are upregulated in HRD ovarian cancer and enriched in immune-related pathways.
  • HRD-EXCUTE predicts survival: In TCGA ovarian cancer data, patients with HRD+HRD-EXCUTE+ had significantly better overall survival, disease-specific survival, and progression-free survival compared to HRD+HRD-EXCUTE- or HRD- patients.
  • HRD-EXCUTE predicts immunotherapy response: In the IMvigor210 (urothelial cancer) and I-SPY2 (breast cancer) cohorts, HRD-EXCUTE status was a strong predictor of response to immunotherapy, with HRD+HRD-EXCUTE+ patients showing better outcomes.
  • Combination therapy enhances anti-tumor immunity: In a mouse model of HRD ovarian cancer, the combination of Niraparib (PARPi) and Entinostat (HDACi) significantly delayed ascites formation, reduced tumor growth, increased immune cell infiltration (including CD8+ T cells and M1 macrophages), and promoted the HRD-EXCUTE signature in ascites cells.
  • HDACi-PARPi combination activates cGAS-STING and histone acetylation: Mechanistic studies showed that this combination treatment activates the cGAS-STING pathway and increases histone acetylation (H3K27ac) around HRD-EXCUTE hub genes, driving the immune response.

Limitations and Future Work

  • HRD-EXCUTE correlation with HRD Score: While HRD-EXCUTE was correlated with HRD Score, the correlation coefficient (0.29) was moderate, suggesting that HRD Score alone doesn't fully capture the functional immune phenotype.
  • Clinical validation: While validated in multiple cohorts, further prospective clinical trials are needed to confirm HRD-EXCUTE's utility as a biomarker for predicting immunotherapy response in ovarian cancer.
  • Other HDAC inhibitors: The study focused on Entinostat, but other HDAC inhibitors (e.g., Vorinostat, Givinostat) also showed promise in vitro and warrant further in vivo investigation.
  • Triple therapy: The study briefly explored a triple combination (PARPi + HDACi + a-PD-1) showing enhanced effects, suggesting this could be a promising avenue for overcoming immunotherapy resistance.
  • Drug sensitivity vs. immunogenicity: The authors note that HRD-EXCUTE might not directly increase drug sensitivity of HRD cancer cells but primarily promotes their immunological phenotype.

The findings lay a strong foundation for developing combination therapies involving PARPi and HDACi to improve immunotherapy outcomes in HRD ovarian cancer patients.

Summary updated

Abstract

In recent years, homologous recombination deficiency (HRD) has not achieved the expected substantial promotion of immunotherapeutic efficacy in ovarian cancer. This study aims to explore the role of HRD functional phenotype as a powerful biomarker in identifying HRD patients who may benefit from immunotherapy. HRD functional phenotype, namely HRD-EXCUTE, was defined as the average level of the 15 hub genes upregulated in HRD ovarian cancer. A decision tree was plotted to evaluate the critical role of HRD-EXCUTE in HRD patients. Agents inducing HRD-EXCUTE were identified by CMAP web (Connectivity Map). The mechanisms and immunotherapeutic effect of PARPi and HDACi in promoting HRD-EXCUTE was examined <i>in vitro</i> and <i>in vivo</i>. The decision tree plotted on the basis of HRD and HRD-EXCUTE indicated the HRD patients without the HRD functional phenotype were largely unresponsive to immunotherapy, which was validated by the immunotherapeutic cohorts. Furthermore, loss of HRD-EXCUTE in the HRD patients attenuated immunogenicity and inhibited immune cells in tumor microenvironment. Moreover, Niraparib combined with Entinostat induced HRD-EXCUTE by activating the cGAS-STING pathway and increasing the histone acetylation. The combination therapy could enhance the cytotoxicity of immune cells, and promote pro-immune cells infiltrating into ascites, resulting in inhibited ovarian cancer growth. The HRD functional phenotype HRD-EXCUTE was set up as a potent biomarker to identify whether HRD patients can benefit from immunotherapy. Loss of HRD-EXCUTE in HRD patients were largely insensitive to immunotherapy. The combination of PARPi with HDACi could improve the efficacy of the PARPi-based immunotherapy in ovarian cancer by augmenting the HRD functional phenotype.
BibTeX
@article{yang2023loss,
  author = {Dong Yang and Fuxue Huang and Wei Wei and Qiaqia Li and Junwan Wu and Yun Huang and Zhi‐Ling Li and Hai‐Liang Zhang and Xuan Li and Qiu-Er Yuan and Qingshan Chen and Gong‐Kan Feng and Rong Deng and Jun‐Dong Li and Xiao‐Feng Zhu},
  title = {{Loss of HRD functional phenotype impedes immunotherapy and can be reversed by HDAC inhibitor in ovarian cancer}},
  journal = {International Journal of Biological Sciences},
  year = {2023},
  doi = {10.7150/ijbs.79654},
}
Figure