Article Text

Download PDFPDF
The significance of the senescence pathway in breast cancer progression
  1. Rahmawati Pare1,2,
  2. Tao Yang1–3,
  3. Joo-Shik Shin1,2,
  4. Cheok Soon Lee1–6
  1. 1Discipline of Pathology, School of Medicine, University of Western Sydney, Liverpool, New South Wales, Australia
  2. 2Cancer Pathology and Cell Biology Laboratory, Ingham Institute for Applied Medical Research, Liverpool, New South Wales, Australia
  3. 3Department of Anatomical Pathology, Sydney South West Pathology Service, Liverpool Hospital, Liverpool, New South Wales, Australia
  4. 4Department of Tissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital, Camperdown, New South Wales, Australia
  5. 5Cancer Pathology, Bosch Institute, University of Sydney, Sydney, New South Wales, Australia
  6. 6South West Sydney Clinical School, University of New South Wales, Liverpool, New South Wales, Australia
  1. Correspondence to Professor Cheok Soon Lee, Department of Anatomical Pathology, Sydney South West Area Pathology Service, Liverpool Hospital, Locked Mail Bag 7090, Liverpool BC, NSW 1871, Australia; soon.lee{at}


Invasive breast cancer develops through prolonged accumulation of multiple genetic changes. The progression to a malignant phenotype requires overriding of growth inhibition. It is evident that some breast cancers have an inherited basis, and both hereditary and sporadic cancers appear to involve molecular mechanisms that are linked to the cell cycle. Frequently, changes in the molecular pathways with gene deletions, point mutations and/or overexpression of growth factors can be seen in these cancers. Recent evidence also implicates the senescence pathway in breast carcinogenesis. It has a barrier effect towards excessive cellular growth, acting as the regulator of tumour initiation and progression. Later in carcinogenesis, acquisition of the senescence associated secretory phenotype may instead promote tumour progression by stimulating growth and transformation in adjacent cells. This two-edge role of senescence in cancer directs more investigations into the effects of the senescence pathway in the development of malignancy. This review presents the current evidence on the roles of senescence molecular pathways in breast cancer and its progression.


Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Breast cancer is the commonest cancer in women worldwide, although it can also infrequently affect men. In 2008, breast cancer was responsible for 12% of all new cases of diagnosed cancers and 7% of cancer deaths in New South Wales, Australia.1 Statistical analysis has shown a reduction in mortality rates among breast cancer patients, which may reflect advancements in its diagnosis and management.2 Notwithstanding the current therapeutic options for breast cancer, whether by surgery, chemotherapy, radiotherapy and/or hormonal therapy, there is always the possibility of undertreating or overtreating individual patients, and further means of stratifying the treatment groups are needed. The peak incidence for breast cancer occurs in the older age group (age >50).1 However, there is an incremental increase in the prevalence of breast cancer in younger women (age <35), who often have more aggressive disease.2 ,3 Carcinoma of the breast has been seen to evolve in a linear pathway from its associated non-obligate precursor. It has been proposed that the low grade invasive lesions arise from a clustered group of lesions called ‘low grade breast neoplasia family’ and high grade carcinoma develops from high grade precursor. Contrary to the low grade, high grade group is found to be a more heterogenous disease.4 Current scientific data support the premise that cancers involve a multistep disease progression via protracted accumulation of genetic changes.5 The molecular evolutions of breast malignancy from their precursor lesions are less well established. In recent years, progression of malignant invasive tumours is believed to have arisen from the evasion of senescence in preinvasive precursor tumour.6 While cancer therapy had previously targeted cell apoptosis and necrosis, recent interest has focused on discovering the role of cellular senescence on growth inhibition and cancer progression. Therefore, evidence for mechanisms of senescence evasion in breast cancer progression needs further investigation. Such knowledge may help in better risk estimation of tumour development and thus offer more accurate therapy and management options for the breast cancer patient.

Cellular senescence

In 1961, Hayflick and Moorhead convincingly showed that human cells in vitro do not divide indefinitely as previously thought, originating the theory of cellular senescence.7 The latter is defined as the state of irreversible loss of cell replicative ability and appears to be influenced intrinsically by telomere shortening and extrinsically by factors including oncogenic/mitogenic stimuli, DNA damage and tumour suppressor mutation mechanisms via the signal transduction cascade in cancer involving the p53 and p16 pathways.8 One of the critical responses for sustaining cellular integrity is when a cell irreversibly desists to replicate after exposure to influences promoting perpetuated cell replication and environmental stress.9

In conjunction with molecular changes, senescent cells also acquire characteristic morphological features. At the ultrastructural level, senescent cells show protein misfolding in the endoplasmic reticulum.10 Microscopically, the cells are enlarged, flat and irregular11 with vacuolated, granular cytoplasm.9 In addition, a study using human diploid fibroblast cell lines showed phenotypic changes involving the activation of Rb family proteins, with enhancement of actin stress fibres and the redistribution of focal adhesion proteins.12 It has been proposed that the key to regulation of focal adhesion kinase activity and the formation of altered actin stress fibre in the senescent cells is caveolin-1 (Cav-1).11 Cav-1 has been found to play a role in regulating tumourigenesis in a number of human cancers, including breast cancer. It has been shown that Cav-1 suppressed the growth of breast tumours.13

Growth of a senescent cell appears to be arrested at the G1/S interface.14 Though senescent cells are incapable of dividing in the presence of mitogenic stimuli, they are still metabolically active. Senescent cells are capable of expressing senescence-associated β-galactosidase (SA-B-gal) and P16INK4a, and show robust secretion of numerous growth factors, cytokines, proteases and other proteins, as well as nuclear foci containing DNA damage response proteins (DNA-SCARS/TIF) or heterochromatin (senescence-associated heterochromatin foci).15 Production of SA-B-gal at pH 6.0 either in vivo or in vitro by senescent cells is now used as the standard biomarker of senescence activity.16 In addition, expression of matrix metalloproteinases (MMPs), cathepsin B and β4 integrin are elevated in senescent prostatic epithelial cells.17

Senescence subtypes can be categorised as replicative or oncogene induced. Replicative senescence is inducted by the aggregation of shortened telomeres following cell multiplication.18 These repetitive DNA sequences at the chromosomal ends progressively shorten with successive cell division as part of a cellular ageing process, whereby the critically short telomeres are processed not unlike double-stranded DNA breaks. This eventually leads to activation of the p53 and Rb tumour suppressor pathways, which are directed either to senescence or apoptosis.19

On the other hand, oncogene-induced senescence (OIS) is independent of the telomere pathway20 and is triggered by activated oncogenes such as ras,21 DNA damage or other cellular stresses.22 In vitro experiments have shown that premature senescence can result from the lack of conducive culturing conditions which provoke culture shock.23 OIS can also be induced by the loss of a critical tumour suppressor gene or its function. Hence, loss of tumour suppressor B-cell translocation gene 3 in normal cells can divert their proliferative state to senescence through the activation of p16.24 However, escape from senescence occurs when transforming senescent cells acquire new genetic mutations which allow for reversal to their proliferative state.

Senescence in tumourigenesis

Tumourigenesis in human cells arises by the stepwise accumulation of mutations and epigenetic factors that alter gene expression. The process may require a cell to undergo three lifetimes to evade the various neoplastic limiting processes at different check-points of the cell cycle to generate a cancer cell with estimated mutation rate at one in 2×107 per gene cell division.25 While there are repair mechanisms to correct minor genetic events to preserve genome integrity, more deleterious gene alterations will direct cancer cells to undergo either programmed cell death or senescence.26

Senescence opposes cellular transformation in vitro, and OIS is emerging as a potent protective response to oncogenic events in vivo by removing early neoplastic cells from the proliferative pool.27–29 OIS prevents tumour growth via two main pathways: p16INK4A-RB and ARF-p53. The senescent state of neoplastic cells can be induced or maintained by a number of factors via these two pathways. Many human cancers are associated with mutations in these pathways that are linked to senescence or polymorphisms to p21, and the inactivation of p53 or Rb appears to extend the replicative lifespan of cells.30 However, both p16INK4A and p53 proteins are at the beginning of the senescence cascade and tumour cells can acquire downstream mutations that can block senescence despite high levels of these proteins. Activation of p53 may lead to either quiescence (reversible cell arrest) or senescence (irreversible growth arrest) but the divergence between p53-induced quiescence and senescence is determined by the mammalian target of rapamycin (mTOR) pathway in which p53 can suppress cellular senescence and convert to quiescence through the inhibition of mTOR.31

Senescence markers have been identified in a number of human and animal tumours such as melanocytic naevi,32 ,33 murine lung adenomas,34 human dermal neurofibromas,34 human and murine prostatic adenocarcinomas,28 murine pancreatic intraductal neoplasias35 and murine lymphomas.29 Reinforcing the concept that senescence acts as a tumour suppressor, senescent cells in murine tumours induced by oncogenic Ras or inactivation of Pten are enriched in the premalignant but not the malignant tumours.28 ,29

As well as arresting proliferation of pre-neoplastic cells, senescence may also suppress tumour growth by initiating an immune response against the senescent cells. Reactivation of p53 in a murine liver carcinoma model not only induces senescence of the tumour cells, but provokes tumour clearance by the innate immune system.36 This is due, at least in part, to the production of inflammatory cytokines released by the senescent cells.36 Indeed, it has recently been shown that senescent cells derived from different cell types have a similar ‘senescent-associated secretory phenotype’37 that may contribute to the immune response against senescent cells in vivo.

A significant aspect of OIS markers is their role in tumour chemosensitivity. An example is the decoy death receptor DcR2, which suppresses the tumour necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced apoptosis. DcR2 is involved in the regulation of chemosensitivity and its expression has been shown to be important in the progression of a number of human malignancies.38 ,39 The efficacy of some chemotherapeutic drugs is dependent on their ability to trigger senescence at low doses, in contrast to those drugs requiring higher doses by acting via the apoptosis pathway. Hence, senescence-inducing chemotherapeutic agents have the advantage of less toxicity due to lower dosage and may be a useful alternative to treat tumours that are resistant to apoptosis-induced drugs. It has been postulated that mice with senescence defects acquire chemoresistance and have poor prognosis compared with the mice harbouring tumours with intact senescence induction.40 Nevertheless, a recent finding has shown apoptosis resistant breast tumours are as a consequence of p53-induced senescence which in turn obstructs mitotic catastrophe and eventually prevents subsequent cell death.41

Senescence in breast cancer progression

Senescent cells accumulate in ageing tissues and contribute to age-related pathology such as cancer.16 ,42 ,43 Since the incidence of breast cancer increases with age, it is likely that the senescence pathway is involved in its development. Induction of cancer formation is associated with the accumulation of replicative senescence in normal human cells44 and with p21-induced senescence in tumour cells.45 In addition to the activation of growth inhibitor pathways, senescent cells also show increased expression of genes for secreted mitogenic, antiapoptotic and angiogenic factors such as extracellular matrix proteins Cyr61 and prosaposin and transforming growth factor-α associated with paracrine tumour-promoting effects.46 This theory has been supported by a study which shows senescence may predispose hepatocellular carcinoma development from chronic hepatitis C viral infection.47

The senescence response can be negated by inactivation of p53 and/or Rb function by any mutation of these genes, which can lead to continuous cellular proliferation. Mutation of the p53 gene, the guardian of the genome, is the commonest genetic event in primary breast cancer48 and it appears to be more likely to be altered in high grade ductal carcinoma in situ (DCIS) than low grade DCIS.49 p53 Gene mutations are found to occur before the development of invasive breast cancer, specifically in high grade DCIS.50 p53 Mutation is associated with a more aggressive stage of breast cancer with worse survival rates.51 Patients with Li-Fraumeni syndrome possess a high risk of developing breast cancer because they carry the germ-line p53 mutation.52 While mutation in p53 is common in primary breast cancer, mutation of CDKN2 gene which encodes p16 is a rare event, implying that the latter is not a critical genetic change in primary breast cancer formation.53 ,54 However, a study has shown that p16 overexpression in the cytoplasm and nuclei is associated with a highly malignant phenotype of breast cancer.55 The p16 gene activity was found reversely modulates the function of the guardian of the genome, p53 in human mammary epithelial cells, via Rb pathway.56 Additionally, loss of p16INK4a does not abrogate senescence in mouse embryonic fibrobalsts.57 ,58

The tumour suppressor gene p14ARF, which acts as an upstream regulator of p53, induces premature senescence through the activation of p21-dependent pathway. Induction of p21 has led to the formation of enlarged and flattened senescent cells.59 p14ARF is shown to contribute to the tumourigenic process within the breast in a p53-independent pathway via association with human Mdm2. It has been shown that there is overexpression of p14ARF in 47% of invasive and non-invasive breast cancers.60 In addition, p21 is overexpressed in 32% breast cancer patients and is associated with a large tumour size, positive nodal status, high histological grade and high mitotic count.61 Despite its function as a tumour suppressor, p21 is also found to be paradoxically induced cell cycle progression. In a recent finding, progestin, the synthetic progesterone, showed induction of breast tumour progression by inducing expression of p21 via formation of transcriptional complex with stat3, progesterone receptor and ErbB-2.62 Recently, a novel senescence-associated miRNA, miR-22, has been found to play important roles in the progression of cancer of the breast. miR-22 activated senescence in breast carcinoma had restrained tumour enlargement as well as metastasis of the cancer.63

Tumourigenesis of the breast is believed to arise from the overexpression of cyclin D1 through its involvement with pRb inactivation.64 Oncogenic properties of cyclin D1 made it as a target for breast cancer therapy. Surprisingly, kinase-deficient cyclin D1KE/KE in mammary epithelial cells is still able to proliferate in response to ErbB2. This is due to upregulation of autophagy which accounts for reduced premature ErbB2-induced senescence.65 Nevertheless, high activity of autophagy in breast cancer cells also has been associated with retarded tumour growth by virtue of contribution of overexpression of CDK inhibitors induced senescence.66 Breast carcinoma is frequently under persistent oxidative stress caused by the overproduction of reactive oxygen species (ROS).67 ROS are major players downstream of Ras protein in senescence pathway.68 Oxidative stress can induce DNA damage and impel telomere shortening rates. Elevated ROS levels have shown to induce senescence through (i) ARF/p53 pathway via activation of ataxia telangiectasia mutated (ATM)/ataxia telangiectasia & Rad3-related protein (ATR) or (ii) p16/Rb pathway via activation of p38-MAPK.69

The senescence pathway appears to play two contradictory roles in carcinogenesis. At the stage of early carcinogenesis, cellular senescence acts as a protective mechanism against neoplastic transformation. However, later in the evolution of malignancy, senescence appears to stimulate neoplastic growth.29 This is because senescent cells may change the tissue environment that promotes tumour formation in the adjacent cells. Though senescent cell are inactive mitotically, they are still active metabolically in protein expression and secretion, thus acquiring a so-called senescence-associated secretory phenotype.70 Hence, senescent fibroblast cells are found to have the ability to produce growth stimulatory factor that favours the growth of mouse mammary epithelial cells.71 Irradiated fibroblasts which are presumed to be senescent are shown to discompose the mammary epithelial microenvironment and stimulate relevant epithelial cell growth in the mammary gland, thus contributing to breast carcinogenesis.72

Hyperplasia of the breast is inducted by activation of low levels of Ras oncogene. On the other hand, further upregulation of the oncogene promotes permanent cell cycle arrest. However, tumour of the breast eventuated after the cells evade the senescence checkpoints.73 Senescent cells secrete high levels of MMPs.74 High expression of Ras in senescent murine mammary gland is found to be associated with the production of MMP-3 and plasminogen activator inhibitor-1.73 These metabolites are major contributors in breast cancer progression.75 ,76 It has been shown that high expression of MMP-3 and plasminogen activator inhibitor-1 is correlated with malignancy in the mammary gland.77 Moreover, attenuation of transforming growth factor-ß signalling suppresses premature senescence in a p21-dependent manner and promotes oncogenic ras-mediated metastatic transformation in human mammary epithelial cells.78

Senescence de novo marker DCR2 and DEC1 in breast cancer

Senescence de novo markers such as DEC1 and DCR2 were previously identified in cultured cells during tumour progression. Differentiated embryonic chondrocyte, DEC1, is also known as Stra13 or Sharp2. It is a basic helix-loop-helix transcription factor that plays an important role in regulating circadian rhythm, cell division, cell death and malignancy in various cancers.79 ,80 DEC1 is found to be a p53 target in the induction of premature senescence. Overexpression of DEC1 initiates G1 arrest and senescence, and knockdown of DEC1 attenuates DNA damage-induced premature senescence.81 DEC1 mediates premature senescence through the Rb pathway by regulating phosphorylation of p130 (Rb2) protein. In addition, it has been shown that expression of DEC1 and DCR2 is more likely in premalignant lung adenomas rather than in malignant adenocarcinomas.6 It has been observed that there is an increase in the expression of DEC1 during progression from normal to in situ and invasive breast carcinoma.82 These findings suggest that DEC1 may contribute to breast cancer progression to the invasive phenotype. This is in common with the finding of high expression of DEC1 in tumour tissues when compared with normal tissues in human kidney and lung.83

DCR2 or TRAIL-R4 is one of the genes targeted by p53, which acts as an antiapoptotic receptor for the TRAIL. It acts as competitor to DR4 and DR5 for binding to Apo2L/TRAIL in order to inhibit apoptosis induction.84 Studies of DCR2 protein function have shown conflicting roles, with downregulation of its expression in various types of cancers, but high expression in more aggressive types.85 ,86 Cancers with overexpression of DCR2 either within or over the cell surface are more aggressive compared with similar cancers with lower expression. This is due to association with TRAIL resistance.87 Loss of expression of DCR2 in carcinogenesis is directly related to aberrant methylation of its genetic make-up.88 Interestingly, reduced overall survival and disease-free survival have been shown in breast cancer patients with high expression of TRAIL-R4.89 Alleic polymorphism of 2699A/G of DCR2 gene with reduced DCR2 expression appears to be associated with reduced breast cancer risk.90


Cellular senescence is one of the mechanisms that normal cells use to avoid carcinogenesis, and is also used by cancers for their own survival advantage. The significance of senescence during breast cancer progression is not well established. Alteration of the senescence pathway either through p53/ARF or p16/Rb may be important in the progression of breast cancer. Expression of senescence markers of interest may be different in each cancer type and may aid in the early detection of the disease. Examination of human breast cancer samples for the alterations of senescence-related molecules may provide prognostic information including response to treatment. It now appears that by understanding the biology of progression of breast cancer through the senescence pathway, new prognostic factors may allow further optimisation of patient treatment and counselling.


Rahmawati Pare is sponsored by the Ministry of Higher Education, Malaysian Government and the University of Malaysia Sabah.



  • Competing interests None.

  • Provenance and peer review Commissioned; externally peer reviewed.