Unraveling Non-Canonical Senescence Mechanisms and Their Immunomodulatory Impact in Cancer Therapy

Alisertib induced apoptosis, DNA damage, and ICD in vitro, and notably caused selective depletion of immunosuppressive myeloid-derived suppressor cells (MDSCs). In vivo, combining alisertib with anti-CTLA-4 significantly improved survival, enhanced CD8+ T cell infiltration, and reduced MDSC levels in the TME—effects not observed with alisertib or ICIs alone. These findings highlight a synergistic interaction between Aurora A inhibition and CTLA-4 blockade [58]. Yet, senescence is not limited to canonical p53–Rb signaling. When these tumor suppressor pathways are compromised, cells engage alternative stress responses and epigenetic programs to sustain a senescent phenotype—mechanisms discussed in the following section.

While the canonical p53–p21–Rb axis is a primary enforcer of TIS in many cancer cells, senescence can still occur in the absence of functional TP53 and RB1, indicating the presence of alternative, compensatory signaling cascades. These alternative pathways involve both transcriptional and epigenetic regulatory mechanisms that engage CDK inhibitors, stress kinases, and pro-inflammatory signaling, collectively leading to a senescent phenotype even in p53-incompetent contexts [39, 59]. One of the central alternative mediators in p53-deficient cells is p16 ^INK4a (CDKN2A), which is frequently upregulated in response to oncogenic stress and various anticancer treatments. p16 acts independently of p53 by directly inhibiting CDK4 and CDK6, preventing phosphorylation of Rb and enforcing G1 arrest [39]. In Rb-deficient cells, however, the role of p16 must be compensated by other mechanisms, given that Rb is the main effector of CDK4/6 inhibition. In such cases, senescence can be sustained through Rb family members such as p107 (RBL1) and p130 (RBL2), which can substitute for Rb in repressing E2F transcription factors and halting the cell cycle [59, 60]. In cells lacking both p53 and Rb1, a non-canonical senescence program can still be initiated via persistent DDR signaling. In these cells, DNA damage from chemotherapy, radiation, or targeted therapies activates ATM/ATR kinases, which phosphorylate histone H2AX (γH2AX), CHK2, and 53BP1, forming DDF. These foci persist due to the inability of mutant cells to efficiently repair damage, leading to continued activation of stress-activated kinases such as p38 MAPK and JNK [13]. p38 MAPK, in particular, plays a central role in establishing senescence in the absence of p53. Activation of p38 in response to oxidative stress or unresolved DDR enhances transcriptional activity of C/EBPβ and NF-κB, which together drive the SASP. This inflammatory secretome reinforces growth arrest in an autocrine/paracrine fashion and recruits immune cells to eliminate damaged cells. Notably, this process can occur without classical p53-mediated gene expression, suggesting that SASP-associated transcription factors can compensate for the loss of tumor suppressors by enforcing senescence through inflammatory reprogramming [61, 62]. Further supporting this, studies in MYC-driven tumors—which often harbor p53 and Rb1 deletions—have shown that treatment with CDK inhibitors (e.g., CDK4/6 or CDK9 inhibitors) can induce a non-canonical senescence-like state characterized by SASP induction, immune activation, and durable growth arrest. Even without functional p53 and Rb, these tumors exhibit upregulation of cell cycle repressors, such as p27 ^Kip1 (CDKN1B), and increased expression of ISGs, facilitating immune recognition [63, 64].

Another important adaptation involves epigenetic reprogramming. In the absence of p53/Rb signaling, senescence can be enforced by chromatin remodeling through histone methylation (e.g., H3K9me3, H3K27me3) and recruitment of transcriptional repressors such as heterochromatin protein 1 alpha (HP1α) and enhancer of zeste homolog 2 (EZH2). A study investigated why TIS (via MEK inhibition with trametinib and CDK4/6 inhibition with palbociclib) elicits strong NK cell–mediated tumor control in KRAS-mutant lung cancer but not in pancreatic ductal adenocarcinoma (PDAC). In PDAC, the pancreatic TME suppressed SASP and NK cell immunity through EZH2-mediated epigenetic repression of key proinflammatory SASP factors and NK cell–activating cytokines (e.g., IL-15, IL-18) and chemokines. Genetic or pharmacological inhibition of EZH2 restored SASP activity, enhanced NK cell recruitment, and enabled tumor cytotoxicity in murine and human PDAC models. Combining EZH2 blockade with TIS even produced complete tumor regressions in some PDAC-bearing mice, an effect dependent on NK cells and SASP signaling. Transcriptomic data from human PDAC samples supported this mechanism, linking EZH2 repression signatures with improved NK infiltration and SASP induction [65, 66]. Lastly, emerging evidence shows that cells lacking TP53 and RB1 can enter a polyploidy-associated senescence state. Following mitotic slippage or cytokinesis failure—often triggered by Aurora kinase or microtubule inhibitors—cells become polyploid and undergo metabolic and transcriptomic reprogramming characteristic of senescence, including increased mitochondrial ROS, lipofuscin accumulation, and SASP secretion [67, 68]. Regardless of the initiating trigger, senescent tumor cells profoundly remodel their surroundings through SASP secretion. The next section explores how these secreted mediators recruit and activate innate and adaptive immune effectors within the TME.

Senescent cells produce a variety of chemokines, including CCL2 (MCP-1), CCL5 (RANTES), CXCL1, CXCL2, and CXCL10, which attract immune cells such as NK cells, CD8+ CTLs, DCs, and macrophages [69, 70]. For instance, in melanoma models, therapy-induced senescence via AURKA or CDK4/6 inhibition led to the upregulation of CCL5, a chemokine known to recruit T cells and other leukocytes to the tumor site. This recruitment was essential for effective tumor control, as depletion of CCL5 or immune effector cells diminished the therapeutic benefit of AURKA inhibition [48]. In addition to chemokines, TIS triggers the secretion of cytokines such as IL-6, IL-8, IL-1α, and TNF-α, which act not only as inflammatory signals but also as immunomodulatory agents that activate innate and adaptive immune cells [4]. IL-6, for example, enhances antigen presentation by increasing MHC class I expression on tumor cells and promotes the differentiation of Th17 cells, which are involved in tumor surveillance. Meanwhile, IL-8 (CXCL8) recruits neutrophils and NK cells and supports their activation within the TME [5, 71]. Importantly, the effectiveness of TIS-induced immune recruitment is context-dependent. In immunologically “cold” tumors, such as pancreatic cancer, senescence alone may be insufficient to stimulate immune infiltration due to the suppressive TME. However, modulation of SASP composition or combination with agents that alter the epigenetic regulation of SASP genes (e.g., EZH2 inhibitors) can restore cytokine production and enhance NK cell recruitment and cytotoxicity, as demonstrated in models of KRAS-driven pancreatic cancer [72, 73]. Furthermore, TIS can upregulate stress-induced ligands such as MICA/B and ULBPs, which are recognized by the NKG2D receptor on NK cells and γδ T cells. These ligands promote the direct killing of senescent tumor cells, providing an important mechanism for immune-mediated clearance of senescent populations. The interplay between SASP factors and these ligands amplifies the immune visibility of senescent cells, particularly when combined with ICIs or adoptive immune cell therapies [74, 75].

TIS has long been recognized as a powerful tumor-suppressive mechanism and a therapeutic target . By halting the proliferation of damaged or stressed tumor cells, TIS prevents the propagation of potentially malignant clones and promotes a state of durable growth arrest. However, accumulating evidence has revealed a Janus-faced nature of TIS, where the same processes that initially suppress tumor progression can paradoxically contribute to tumor relapse, progression, and metastasis [6].