Therapy-Induced Senescence In Triple-Negative Breast Cancer: Biomarkers And Reversal Strategies

Therapy induced senescence (TIS) is a state of stable cell cycle arrest, often arising as a consequence of DNA damage, caused by chemotherapy drugs like doxorubicin. TIS is characterised by the senescence associated secretory phenotype (SASP), a heterogeneous collection of pro-inflammatory factors that initially promotes immune surveillance and clearance of damaged cells. However, persistent SASP signalling impairs this response, enabling senescent cell accumulation. Persistent SASP secretion creates a pro-inflammatory and pro-tumorigenic microenvironment, in which SASP components drive angiogenesis, microenvironment remodelling, tumour invasion and metastasis. Breast cancer (BC) is the most common cancer in the UK, of which triple negative BC (TNBC) is a clinically challenging subtype, due to its aggressive behaviour and high relapse rates. TNBC accounts for approximately 15% of BC cases but nearly 40% of BC deaths. As doxorubicin is a first line TNBC treatment and a potent inducer of TIS, it represents a clinically relevant model for investigating TIS. Given that TIS can arise as a consequence of chemotherapy and contribute to tumour progression, inflammation, and treatment response, it is essential to both be able to identify and mitigate its detrimental effects. One aim of this research was to identify predictive or prognostic biomarkers of TIS in TNBC. TIS models were developed, using doxorubicin, in two TNBC cell lines (MDA-MB-231 and HCC1806) and one non-tumorigenic breast cell line (MCF10A). Senescence was confirmed using multiple hallmarks, including senescence associated β-galactosidase staining, nuclear area changes, γ-H2AX staining, and altered expression of key senescence associated genes. A literature review identified several candidate biomarkers associated with senescence, cancer and the cytoskeleton, whilst a bioinformatics approach assessed their potential predictive and prognostic significance. Candidate biomarkers were evaluated for their presence in TIS TNBC models. Although several genes showed comparable expression patterns in TIS induced non-malignant cells, these trends were often not conserved in TIS TNBC cell lines, highlighting the complexity of defining a universal TIS biomarker. Among the candidates assessed, Lamin B1 emerged as a reliable marker of TIS and was therefore used in subsequent experiments. Metformin was assessed for its senotherapeutic potential in TIS models, to mitigate the detrimental consequences of TIS, based on its established tolerability and reported anti-senescence effects, extending beyond SASP attenuation. As existing literature predominantly uses metformin at supra- pharmacological concentrations, this study compared a clinically relevant concentration (50 μM) with a supra-pharmacological concentration (500 μM). Treatment of TIS models with metformin did not result in clear attenuation of senescence hallmarks. To determine whether this was due to insufficient uptake or pathway activation, expression of key metformin transporters were assessed using qPCR and immunostaining, and pathway activation (AMPK and mTOR phosphorylation) was assessed using western blot. Key transporters were present in cell models, and preliminary data indicated pathway activation was consistent, to some extent, with metformin’s expected mechanism of action. To investigate the paracrine effects of TIS and their modulation by metformin, indirect co-culture models were developed using patient derived fibroblasts (a key component of the tumour microenvironment), and TIS HCC1806. Fibroblasts were assessed for markers of senescence (senescence associated β-galactosidase staining and inflammatory gene expression) and fibroblast activation (scratch assays and morphological analysis), both of which contribute to cancer progression. Fibroblasts developed a pro-inflammatory gene expression profile when co-cultured with TNBC cells, which was further enhanced by TIS. These findings indicate that SASP signalling may promote pro-tumorigenic alterations within the tumour microenvironment. Treatment with supra-pharmacological metformin concentrations (500 μM) reduced the expression of several inflammatory factors which were not attenuated by physiological concentrations (50 μM), suggesting that clinically relevant metformin concentrations may be insufficient to effectively suppress SASP signalling. Metformin treatment simultaneously increased expression of the inflammatory factor interleukin 8, underlining the complexity of metformin’s role in senescence attenuation. In summary, gene expression analysis of candidate biomarkers, identified through bioinformatics analysis, revealed that that TIS associated gene expression is context dependent and emphasised the difficulty of identifying a universal TIS biomarker. SASP signalling was shown to drive an inflammatory gene expression profile in fibroblasts, potentially contributing to pro-tumorigenic changes within the TME. Although metformin did not directly attenuate TIS in the cell models used, it modulated inflammatory signalling at supra-physiological concentrations, indicating that it may have senotherapeutic potential in certain contexts, but is unlikely to be effective at clinically relevant doses.