Expression of therapy-induced senescence markers in breast cancer samples upon incomplete response to neoadjuvant chemotherapy

Abstract Senescence is a cell stress response induced by replicative, oxidative, oncogenic, and genotoxic stresses. Tumor cells undergo senescence in response to several cancer therapeutics in vitro (Therapy-Induced Senescence, TIS), including agents utilized as neoadjuvant chemotherapy (NAC) in the treatment of invasive breast cancer. TIS has been proposed to contribute to adverse therapy outcomes including relapse. However, there is limited evidence on the induction of senescence in response to NAC in clinical cancer and its contribution to disease outcomes. In this work, the expression of three senescence-associated markers (p21CIP1, H3K9Me3 (histone H3 lysine 9 trimethylation), and Lamin B1) was investigated in breast cancer samples that developed partial or incomplete pathological response to NAC (n=37). Accordingly, 40.54% of all samples showed marker expression consistent with a senescence-like phenotype, while the remainders were either negative or inconclusive for senescence (2.70 and 56.8%, respectively). Moreover, analysis of core-needle biopsies revealed minimal changes in p21CIP1 and H3K9Me3, but significant changes in Lamin B1 expression levels following NAC, highlighting a more predictive role of Lamin B1 in senescence detection. However, our analysis did not establish an association between TIS and cancer relapse as only three patients (8.1%) with a senescence-like profile developed short-term recurrent disease. Our analysis indicates that identification of TIS in tumor samples requires large-scale transcriptomic and protein marker analyses and extended clinical follow-up. Better understanding of in vivo senescence should elucidate its contribution to therapy outcomes and pave the way for the utilization of senolytic approaches as potential adjuvant cancer therapy.


Introduction
Breast cancer is the most commonly diagnosed female malignancy in the United States (1), and the second leading cause of cancer-related deaths in women worldwide (2). Since the standardof-care for invasive breast cancer patients has improved dramatically, the number of breast cancer survivors has increased; however, this increase in the surviving population is associated with increasing rates of cancer recurrence (3). For example, in Dutch women, the regional recurrence and distant metastasis rates over the first 10 years post-therapy ranged from 7.5-26% with luminal invasive carcinoma (4), while, in a population of breast cancer female patients in the United Kingdom, the 20-year recurrence rate was 10-17%, depending on the biological subtype and stage (5). In parallel, there is an increased utilization of neoadjuvant chemotherapy (NAC) for the treatment of breast cancer as around 17% to 79% of patients receive a regimen of NAC depending on their biologic subtype (6,7). Interestingly, the recurrence rate is higher in certain breast cancer patients receiving neoadjuvant chemotherapy (mainly, anthracycline-based therapy) in contrast to patients that did not receive the same treatment, strongly suggesting the possible contribution of cellular responses to chemotherapy as mechanisms of cancer recurrence (8). Thus, there is an avid need for the identification of novel mechanisms that might contribute to disease recurrence, especially in cancers that develop partial response to therapy; a better understanding of these mechanisms should pave the way for a more effective anticancer therapy and higher survival rates.
Cellular senescence is a cell state characterized by a stable growth arrest accompanied by transcriptomic and epigenetic alterations (9,10), macromolecular and metabolic changes (11,12), increased lysosomal biogenesis (senescence-associated beta galactosidase, SA-β-gal) (13,14), as well as the secretion of a spectrum of chemokines and cytokines collectively named the Senescence-Associated Secretory Phenotype (SASP) (15). Senescence is an established cell stress response to DNA damaging, targeted and hormonal cancer therapies in preclinical tumor models, hence the name Therapy-Induced Senescence (TIS) (16). While TIS has been extensively established in many preclinical models, the induction and potential role of senescence in tumor cells of cancer patients receiving anticancer treatment is still under investigated. This largey because the canonical marker of senescence, SA-β-gal, is difficult to detect in fixed tumor samples and typically requires flash frozen, unfixed samples. For example, the seminal report by Poele et al. showed that senescence is detected in approximately 40% of tumor samples collected from 36 patients with breast cancer receiving a single regimen of NAC (namely, cyclophosphamide, doxorubicin and 5-fluorouracil) utilizing archived frozen tumor samples (rather than flash frozen, fresh samples) (17). Another report by Roberson et al. showed evidence of senescence induction based on SA-β-gal staining in freshly-frozen tumor samples derived from three lung cancer patients receiving platinum-based/taxane therapy (18). While the detection of the SA-β-gal staining is a classical approach to identify senescent cells, its sole use as proof of senescence induction has been received with multiple concerns, including nonspecificity (19). Furthermore, it is challenging to obtain fresh human samples to carry out histochemical staining for SA-β-gal on a routine basis. Moreover, while suggestive evidence on senescence induction in cancer patients is provided in the literature, the contribution of TIS to the overall therapeutic outcome is yet to be determined, especially that senescence has been proposed as a mechanism of tumor dormancy and cancer recurrence (20). In this work, we hypothesized that TIS might be a component of incomplete disease response to NAC, in that tumor cells may respond to therapy primarily by senescence rather than cell death, which more classically reflects complete pathological response (pCR). Further, we aimed to establish whether the assessment of multiple markers other than SA-β -gal might prove useful for determining senescence in archived tumor samples. The stability of the senescent growth arrest is enforced by the activation of several cell cycle-regulating proteins (21), including cyclindependent kinase inhibitors (CDKIs) such as p21 CIP1 or p16 INK4a (22,23). p21 CIP1 mediates the growth arrest occurring in replicative exhaustion-induced senescence (24). Moreover, several studies have established the relationship between p21 CIP1 and senescence in tumor cells, as increased expression of p21 CIP1 was shown to be sufficient to drive tumor cells into a stable growth arrest (25). Importantly, many studies have suggested that p21 CIP1 is key in the induction of senescence after exposure to anti-cancer agents, i.e., TIS (26). Accordingly, p21 CIP1 is a frequently used senescence marker in preclinical models (27). In addition to the cell cycle regulators, senescent cells undergo structural changes such as nuclear envelop remodeling marked by degradation of nuclear laminar proteins such as Lamin B1; thus, Lamin B1 loss is an established biomarker of senescence (28). Also, senescent cells develop epigenetic signatures collectively called the Senescence-associated Heterochromatic Foci (SAHF) that contribute in the regulation of expression of proliferation-associated genes (29). SAHF can include several epigenetic signatures such as histone H3 lysine 9 tri-methylation (H3K9Me3) and the Heterochromatin Protein 1 (HP1) (30). SAHF represent areas of transcriptionally silent and compacted chromatin that result from the presence of repressive H3K9Me3 and absence of activating H3K4Me3 (31). Accordingly, the increased expression level of H3K9Me3 has been considered as an indicator for the occurrence of senescence. These features are manifested variably in therapy-induced senescent tumor cells in preclinical models (32). Therefore, we examined the expression of these three proteins that reflect the previous senescence hallmarks simultaneously. Here, we provide evidence on the expression of these three TIS-associated markers, namely p21 CIP1 , H3K9Me3 and Lamin B1, in breast tumor samples following NAC and developed partial or incomplete response to therapy. Lastly, we include a correlative analysis of the expression of TIS-associated markers and evidence of cancer recurrence in the studied population.

Sample
All samples were obtained from patients diagnosed with non-metastatic, invasive breast fluorouracil, epirubicin and cyclophosphamide (FEC) or 5-fluorouracil, epirubicin, cyclophosphamide followed by docetaxel (FEC+D). The criteria for exclusion for the histopathological/biochemical analysis were (i) patients with inoperable, metastatic disease; (ii) patients who were deceased throughout the treatment period and are no longer followed up in the oncology clinic; (iii) patients who also received radiation or hormonal neoadjuvant therapy (e.g., tamoxifen); (iv) patients who had complete response to NAC determined by surgical and radiological assessment.
In this work, our sample included a total of 89 patients diagnosed with a subtype of breast cancer and treatment of the indicated NAC regimens with variable pathological responses (n=89). Of those, 10 patients were deceased while receiving treatment and were excluded from the histopathological/biochemical analysis. Furthermore, a total of 37 patients who had a partial or no response to NAC as determined by intraoperative staging were only considered for immunohistochemical staining (n=37). Formalin-fixed paraffin-embedded (FFPE) breast tumor blocks for all included patients were collected (n=37). The diagnosis of a subtype of invasive breast carcinoma has been confirmed histopathologically using hematoxylin-and eosin-stained sections of mastectomy specimens following the administration of NAC by three specialized pathologists at JRMS, PHH and Jordan University Hospital (JUH). All patients underwent Modified Radical Mastectomy (MRM) within a period of 20-30 days following the completion of the last cycle of NAC, which represents the timepoint of sample collection after exposure to chemotherapy. Lastly, core needle biopsy samples prior to receiving NAC of 11 out of 37 patients were available and collected for further histopathological analysis/biochemical staining (n=11). Immunohistochemical staining was performed in 37 resected tumor specimens of patients receiving NAC and 11 core needle biopsy FFPE samples to assess the protein expression of three senescence-associated markers: Lamin B1, H3K9Me3 and p21 CIP1 as follows. 5-um-thick tissue sections were cut, using a microtome (LEICA RM2125RT) and placed on clean, charged glass slides. The tissue sections were dried by placing in the oven at 70°C for 20 minutes. Sections were dewaxed in two changes of xylene for 5 minutes, rehydrated in ethanol (100%) for one minute, then immersed in ethanol (95%) for one minute, and washed twice with distilled water for 5 minutes to remove residual alcohol. Antigen retrieval was performed by placing the container of the slides in 10 mM sodium citrate buffer solution, pH 6, for 1 hr at 95°C in the water bath. The slides were then cooled at room temperature for 30 minutes.

Immunohistochemistry (IHC)
The slides were washed with phosphate buffered saline (PBS) (pH 7.3± 0.10 diluted to 1L; 64123666, BIO-RAD) for five minutes and then treated with 3% hydrogen peroxide (DQ400-60KE, BioGenex) for 10 minutes. After that, sections were washed in PBS and treated with 0.1 % Triton X-100 in PBS to permeabilize cell membrane/nuclear envelope for 15 minutes. The slides were washed in PBS and then incubated for 15 minutes with power block reagent (Catalogue Number: DQ400-60KE, BioGenex). The slides were then incubated with mouse monoclonal antibody specific for human Lamin B1 and mouse monoclonal antibody against H3K9Me3 for 1.5 hours at room temperature, and mouse monoclonal antibody against p21 CIP1 overnight at 4°C. After incubation, slides were then rinsed in PBS before being treated with Super Sensitive polymer -HRP IHC Detection System (DQ400-60KE, BioGenex) followed by incubation with secondary antibody for 45 minutes at room temperature. Slides were then rinsed in PBS and incubated with polymer -HRP reagent for 30 minutes at room temperature and rinsed in PBS. After that, the substrate solution containing Diaminobezidine Chromogen (DAB) was used for 10 minutes and the slides were washed in distilled water. For the staining step, slides were then lightly counterstained with hematoxylin for 3 minutes and then washed with tap water. After hematoxylin staining, slides were treated with lithium carbonate solution for 30 seconds and then washed using tap water. Finally, the slides were dehydrated through ascending concentration of ethanol (95%, 100%) and rinsed in xylene. The slides were mounted using dibutyl phthalate in xylene (DPX).
Each staining series had positive control slides (for Lamin B1, normal colon epithelium, for H3K9Me3, human colon carcinoma, for p21 CIP1 , human bladder carcinoma) and the negative control slides. Negative control slides were performed by omitting the specific primary antibody (replaced by PBS) from the staining procedure on the same tissue samples that were utilized as positive controls.

Antibodies and expression evaluation
All antibodies were stored at either 4°C or -20°C as per the manufacturer's instructions.  Nuclear staining of Lamin B1, H3K9Me3 and p21 CIP1 was scored semi-quantitatively in the most prominently stained area of the tissue slides measuring stained cells and/or area ratio by two independent pathologists using a light microscope (Olympus BX 25, Olympus, Tokyo, Japan) under 20x and 40X objective lenses. The expression assessment of the Lamin B1, H3K9Me3 and p21 CIP1 markers and the cut-offs used based on previous studies (33)(34)(35). For Lamin B1, any nuclear staining in the tumor samples was considered positive, while tumor samples with <10 % positively stained tissue were considered negative. For H3K9Me3 and p21 CIP1 , tumor samples with >50 % positively stained tissue were considered positive while tumor samples with ≤50% positively stained tissue were considered negative.
Determining senescence induction was based on the evaluation of all three tested biomarkers combined. In that, only samples that show positive expression for H3K9Me3 and p21 CIP1 and negative expression of Lamin B1 were considered positive for senescence. Consequently, samples that were negative for H3K9Me3 and p21 CIP1 and positive expression of Lamin B1 were considered negative for senescence. Lastly, all other expression possibilities were considered as inconclusive for senescence.

Patient Follow-up
Using patient databases of the Departments of Surgery at JRMS and PHH, postoperative patient follow-up status was evaluated. All patients (n=55) were followed-up for evidence of secondary disease (i.e., recurrence) until December 1 st , 2020 (median follow-up period, 18 months following the time of operation). Evidence of cancer recurrence/metastasis following therapy was also confirmed through the Oncology Department at JRMS and Department of Surgery at PHH. Breast cancer recurrence/metastasis was confirmed using clinical assessment,

Statistical analysis
The relationship between Lamin B1, p21 CIP1 and H3K9Me3 expression were derived using Fisher's exact test. Correlations between the different variables were computed using the Chisquare (χ 2 ) test and Fisher's exact test. Fisher's exact test was used for data set comparisons of n<5. Wilcoxon signed-rank and McNemar's tests were used to evaluate the difference between the pre-and post-neoadjuvant chemotherapy samples. The results were considered as statistically significant with p-values of ≤0.05 whereas p-values of >0.05 were taken as non-significant.
Data analysis was performed using IBM SPSS Statistics Version 24.

Evaluation of breast tumors based on response to neoadjuvant chemotherapy (NAC).
Our initial analysis identified a total number of 89 breast cancer patients who were diagnosed with breast cancer and receiving one form of NAC in two centers. Of those, 10 patients were  Figure 1). All patients were females with an average age of 48.65 years ( Table   1). All patients' disease was confirmed as invasive breast carcinoma through core-needle biopsy prior to receiving any form of therapy and around 92.4 % of the patient samples were diagnosed with Invasive Ductal Carcinoma (IDC), while only 7.6 % were diagnosed with Invasive Lobular Carcinoma (ILC) of stages I-III ( Table 1). Most tumors were graded as G2 (65.8 %) ( Table 1).
The receptor status of all 79 tumors was determined through core-needle biopsy prior to onset of therapy. Of those, 64.6 % were hormone receptor positive (Estrogen Receptor positive, ER+, and Progesterone Receptor positive, PR+), 29.1 % were Human Epidermal Growth Factor Receptor 2 positive (HER2+) and 13.9 % had triple-negative disease ( Table 1).
Of the remaining 79 patients, 55 patients received NAC only, while 24 patients were concomitantly treated with radiation and/or hormonal (e.g., tamoxifen) therapies prior to surgical resection and were excluded from the biochemical staining (Supplementary Figure 1). The percentage of patients that developed complete pathologic response (pCR, ypT0/is, ypN0) to NAC combined with either radiation or hormonal therapies was 18.9 %, while the remaining 81.1 % had partial or incomplete pathological response to therapy. In comparison, the percentage of patients who developed pCR to NAC alone was 27.3 %, while the remaining 72.7% had partial or incomplete pathologic response to sole NAC. Interestingly, 33.3 % of patients who developed pCR following NAC were positive for ER and PR and 40.0 % were negative for HER2 receptor ( Table 2). On the contrary, 67.5 % of patients who developed partial or incomplete response to NAC were positive for ER and PR and 27.5 % were positive for HER2 receptor ( Lamin B1, the loss of which is reflective of nuclear envelope remodeling associated with senescence induction (40).
The expression of p21 CIP1 , H3K9Me3 and Lamin B1 was immunohistochemically determined in paraffin-embedded tumor samples obtained intraoperatively following the completion of NAC treatment and scoring for these markers was performed as described previously (33)(34)(35).
Immunoreactivity of p21 CIP1 , H3K9Me3, Lamin B1 was localized in the nuclei of tumor cells in consistence with their typical expression (Supplementary Figure 2). Our analysis showed that the expression of p21 CIP1 , H3K9Me3, and Lamin B1 in tumor samples of patients who developed partial or incomplete response to NAC was 94.6 %, 75.7 %, and 48.6 %, respectively ( Figure   1A). There was no correlation between Lamin B1, H3K9Me3 and p21 CIP1 when their individual expressions where compared (Fisher's exact test, p=0.714, p=1.000 and p=0.054) (Figure 1B).
For example, samples that were positive for p21 CIP1 or H3K9Me3 were not necessarily negative for Lamin B1 ( Figure 1B). Furthermore, we found no significant relationship between the expressions of Lamin B1, H3K9Me3, p21 CIP1 , and clinicopathologic parameters including the stage of the tumor, histologic tumor grade and hormone receptor status in patients who received NAC (Table 3). However, expression level of Lamin B1 is significantly correlated with tumor grade ( Table 3). To our knowledge, there is limited investigation of on the H3K9Me3 and Lamin B1 expression in human breast cancer and this is the first study to indicate the levels of H3K9Me3 expression and Lamin B1expression in breast cancer samples following exposure to NAC.
To further investigate if the changes in expression of the three TIS-associated markers were due to exposure to NAC, immunohistochemical staining of Lamin B1, p21 CIP1 , and H3K9Me3 was performed on available core-needle biopsy samples (n=11) which were collected prior to receiving NAC (Figure 2). Interestingly, and to our surprise, we observed high expression levels of p21 CIP1 and H3K9Me3 in the core-needle biopsy samples prior to the exposure of breast tissue to NAC, and a decrease in the expression level of p21 CIP1 and H3K9Me3, albeit non-significant,  (Table 4). In order to eliminate the possibility of a staining artifact, which could have accounted for the high expression level of p21 CIP1 and H3K9Me3 in the biopsy specimens, we stained 3 breast tumor samples from patients  Next, we wanted to examine whether there is a connection between the induction of a senescence-like phenotype in patients who received NAC and did not develop pCR and the development of secondary metastasis (local or distant recurrent breast cancer disease) following NAC and radical mastectomy. The number of patients that confirmed evidence of local or distant recurrence following surgical resection was 8 (14.5 %) ( Table 6). As expected, the number of patients who had a pCR to NAC had a lower incidence of developing recurrent disease in comparison to patients whose tumors poorly responded to NAC (Table 6). Moreover, up to 10.9 % of patients developed recurrent disease and were also positive for hormone receptors (ER and PR), while 3.6 % of patients developed recurrent disease and were HER2-positive (Table 6).
Lastly, it was evident that of the 8 patients with partial or incomplete response to NAC that conducted studies related to the comparison of neoadjuvant to adjuvant systemic therapy in women with operable breast cancer (50). During those studies, it was concluded that there was no difference in overall survival, progression-free survival or time to local-regional recurrence following preoperative chemotherapy (50).
Accordingly, local and distant recurrence of breast cancer following anticancer therapy remains to be a major challenge to successful, eradicating cancer treatment. Unfortunately, the underlying molecular and pharmacological mechanisms of tumor dormancy followed by cancer recurrence are not fully elucidated. Interestingly, the recurrence rate was higher in certain breast cancer populations receiving anthracycline-based NAC compared to patients that did not receive the same treatment, highly suggesting the potential association between cellular responses to chemotherapy and cancer relapse (8).
Molecularly, responses of tumor cells to cancer therapy include apoptosis, necroptosis, autophagy, mitotic catastrophe and several forms of growth arrest such as cellular senescence (51). While the ideal response to therapy is apoptosis, evidence has shown that it is not always the case, and cells can alternatively undergo senescence (52)(53)(54)(55). Senescence has traditionally been defined as an "irreversible" form of growth arrest, and the characteristic cytostatic nature of senescence encouraged the utilization of senescence-inducing therapy (traditional DNA damaging chemotherapy) (56). However, several reports recently have confirmed that TIS in tumor cells is not obligatorily irreversible and that some tumor cells can escape the stable senescent growth arrest (reviewed in (16). Moreover, senescent tumor cells have been shown to acquire stem-cell like characteristics (57), leading to more aggressive phenotypes (58), and can contribute to adverse outcomes of cancer therapy (59), which led to the proposal of TIS as a mechanism of tumor dormancy and cancer recurrence (20). Accordingly, our analysis aimed to investigate the induction of TIS in breast cancer samples that underwent partial or incomplete pathological response to NAC.
Several markers are commonly used to identify senescence activation in vitro (60). However, senescence-associated markers independently are not specific to assay establishment of senescence both under in vitro and in vivo conditions until now (61). Accordingly, identifying senescence in vivo should be based on the use of multiple senescence-associated markers (27).
Numerous in vitro studies showed that increased expression of p21 CIP1 , H3K9Me3, and reduced expression of Lamin B1 as evidence of senescence induction (26, 28,30,62,63). In our study, we investigated the expression of p21 CIP1 , H3K9Me3, and Lamin B1 using IHC to evaluate TIS induction in breast cancer patients who received NAC and developed partial or poor response to therapy.
In the current study, we observed a slight, non-significant decrease in the expression levels of p21 CIP1 , and H3K9Me3 (although both were positively expressed in most samples) and a significant decline in the expression of Lamin B1 in patients who received NAC and that, based on staining for all three markers, senescence-like phenotype was identified within advanced breast cancers in response to NAC. Upregulation of p21 CIP1 protein leads to cell cycle arrest by suppressing transition from G1 phase into S-phase (64), and is an established feature of senescence in different breast cancer cell lines in vitro (65,66). In the present study, we found that expression of p21 CIP1 expression in breast cancer tissues after NAC was high, despite no in patients who received NAC. These findings are in line with a previous study showing a significant reduction in p21 CIP1 protein expression levels in invasive breast cancer who received NAC regimens (docetaxel in combination with epirubicin) compared to the core biopsy specimens from same breast cancer patients before NAC (67). In contrast, another study showed an increased expression of p21 CIP1 in breast cancer in response to NAC (62). These previous observations, and our results, suggest that p21 CIP1 might not be a useful senescence-associated marker in the evaluation of TIS in invasive breast cancer due to its variable expression following exposure to DNA damaging NAC.
Similarly, we found that high nuclear expression of H3K9Me3 in patients who received NAC and developed a partial response to therapy (75.7 %). A Previous report showed that there is H3K9Me3 positive expression in 71.8% in human breast cancer (34). Moreover, our statistical analysis suggested that expression of H3K9Me3 is not significantly correlated with pathologic TNM stage, tumor grade and status of ER, PR and HER2 in patients who received NAC in a similar fashion to our observations on p21 CIP1 . The expression levels of H3K9Me3 also have not significantly changed between samples obtained prior to NAC (by core-needle biopsy) or following NAC (as resected during radical mastectomy). Again, indicating that changes in H3K9Me3 expression levels might not be critical in identifying the induction of TIS in breast cancer samples in response to NAC. On the other hand, our analysis suggested that expression level of Lamin B1 is significantly correlated with tumor grade, but not significantly correlated with pathologic TNM stage or receptor status. Moreover, we observed a significant decline in Lamin B1 expression in patients who received NAC consistent to changes in expression that occurs in in vitro senescence (28,40).
Some reports indicated that Lamin B1 expression is reduced in normal human fibroblast and mouse cell lines by various stimuli such as DNA damaging agents, replicative exhaustion, or oncogenic signaling (68); however, unfortunately, the evidence in clinical breast cancer samples is scarce. Its noteworthy that loss of Lamin B1 expression was identified in different malignant tissue such as prostate, breast and esophageal carcinoma (69), indicating that the decline of Lamin B1 expression might be considered a malignancy biomarker. Overall, our data suggest that Lamin B1 displays a significant change in expression following exposure to NAC and can be considered as a marker for future testing batteries utilized to identify TIS in vivo.
Our observations indicate that based on the concomitant expression of three senescenceassociated markers, senescent-like phenotypes exist within human breast cancers who received NAC (40.54 %). Previous reports suggest that senescence can indeed be induced in vivo in response to CAF regimen (cyclophosphamide, doxorubicin, and 5-fluorouracil) based on SA-βgal, p53 and p16 INK4a , and showed that 41% of tumors stained positive for SA-β-gal marker while patients with high nuclear p16 INK4a (but low nuclear p53 expression) (17). Moreover, a previous study has shown that the TIS marked by IHC staining for Ki-67, plasminogen activator inhibitor-1, and SA-β-gal was detected in 30 samples of colorectal cancer (70). These studies should be combined with our data in order to develop better predictive models for the identification of TIS in vivo. Finally, several previous studies reported that the expression of senescence biomarkers correlated with poor outcomes in different kinds of cancers such as breast, colon, and bone cancer (35,(71)(72)(73). In addition to determining the extent of senescence induction in vivo, the exact role of senescence in determining the outcome of therapy is still debatable (74). Accordingly, there is uncertainty to the contribution of senescence to the effectiveness of therapy. In that, although chemotherapy-induced senescence has been studied for decades, the impact of senescence on disease control remains uncertain. Senescence can be perceived as a favorable outcome of cancer treatment due to the fact that senescent tumor cells are in a growth-abrogated phase that halts tumor proliferation (75). On the other hand, the fact that senescent cells are resistant to apoptosis can be interpreted to indicate that senescence-mediated growth arrest serves as a strategy for tumor cells to evade the cytotoxicity of chemotherapy and radiation. The extensive genetic heterogeneity of tumor cells, and specifically, of a senescent tumor population, would, however, suggest that there are subpopulations of these metabolically viable cells that would not obligatorily persist in a permanent senescent-like state of growth arrest (76). This largely invites for the development of better testing approaches to identify senescent tumor cells in vivo.
A limitation to our study is that the number of biomarkers tested is low and the association of their in vitro expression does not necessarily correlate with in vivo senescence. A better approach shall implement wide-transcriptomic analysis of senescence-associated pathways in order to identify an in vivo signature. Such readily available biomarkers can be implemented in carefully investigating the contribution of TIS to therapy outcomes, recurrence rates and the validity of the recent proposition of using senolytic therapy as adjuvant cancer treatment (77). Senolytic agents are drugs that selectively kill senescent cells, that have been shown to exert remarkable potential to clear senescent cells in vitro and in vivo in preclinical studies of aging-related pathologies and