BCL-2 and BCL-xL in Cancer: Regulation, Function, and Therapeutic Targeting.

1. Introduction
B-cell lymphoma (BCL)-2 and BCL-xL are two antiapoptotic members of the BCL-2 family that play central roles in regulating the intrinsic apoptotic pathway. Apoptosis is the most common type of programmed cell death in human tissues and, together with cell proliferation, contributes to tissue homeostasis, development, and remodeling, such as during the formation of the aortic arch [1,2]. There are two major pathways mediating apoptosis: the extrinsic and intrinsic pathways (Figure 1) [3].
The extrinsic pathway is initiated by extracellular death signals. Cell membranes express death receptors, such as Fas, TRAIL-R1/2, and TNF-R, that, upon binding to their respective ligands, recruit adaptor proteins like FADD to form the death-inducing signaling complex (DISC), resulting in the activation of initiator caspase-8 and caspase-10 [3,4]. In contrast, the intrinsic pathway is triggered by intracellular stress signals, including DNA damage, oncogenic activation, oxidative stress, or endoplasmic reticulum (ER) stress, which converge on the mitochondria and lead to mitochondrial outer membrane permeabilization (MOMP) [3]. Members of the BCL-2 family tightly regulate this pathway [5]. These proteins are organized into three major functional groups: (i) antiapoptotic proteins (BCL-2, BCL-xL, myeloid cell leukemia-1 (MCL-1), BCL-w, BCL-2-related gene A1 (A1/BFL-1), and BCL-B), (ii) pro-apoptotic effectors (BCL-2-associated X protein (BAX), BCL-2 homologous antagonist/killer (BAK), and BCL-2-related ovarian killer (BOK)), and (iii) BH3-only activators and sensitizers (such as BCL-2-interacting mediator of cell death (BIM), BH3-interacting domain death agonist (BID), p53 upregulated modulator of apoptosis (PUMA), NOXA, BCL-2-modifying factor (BMF), BCL-2 interacting killer (BIK), BCL-2 agonistic of cell death (BAD), and Harakiri (HRK)). Their interactions are mediated through conserved BCL-2 homology (BH) domains, which allow antiapoptotic proteins to sequester BH3-only factors or directly inhibit BAX/BAK oligomerization, thereby preventing MOMP. For instance, BIK binds and inhibits pro-survival proteins, including BCL-xL, thereby promoting MOMP; given BIK’s ER localization, the interaction is thought to occur at ER–mitochondria contact sites [6]. MOMP results in the release of pro-apoptotic factors, most notably cytochrome c, from the mitochondrial intermembrane space into the cytosol, where cytochrome c associates with APAF-1 and procaspase-9 to form the apoptosome [7]. This complex activates caspase-9 (intrinsic pathway), or, in the extrinsic route, caspase-8 and caspase-10 directly activate downstream effectors. Ultimately, both pathways converge on the activation of executioner caspases, such as caspase-3 and caspase-7, which cleave essential structural and regulatory proteins, leading to the morphological and biochemical hallmarks of apoptotic cell death [8]. While the BCL-2 family comprises several proteins with pro- or antiapoptotic functions, BCL-2 and BCL-xL have been particularly implicated in tumor cell survival and therapeutic response pathways in cancer [9]. Both BCL-2 and BCL-xL are among the most studied antiapoptotic regulators of the intrinsic apoptotic pathway, acting as central control points for MOMP, the critical step that determines whether a cell will commit to apoptosis [10]. MOMP is tightly coupled to mitochondrial metabolic status. Mitochondria integrate bioenergetic flux, redox balance, and calcium signaling, all of which critically influence apoptotic susceptibility. Antiapoptotic proteins such as BCL-2 and BCL-xL not only prevent MOMP by inhibiting BAX/BAK activation but also modulate mitochondrial metabolism by regulating oxidative phosphorylation efficiency, mitochondrial membrane potential, and ATP production [11,12]. In particular, BCL-xL has been shown to interact with components of the F1F0-ATP synthase, enhancing mitochondrial energetic efficiency and promoting cell survival under metabolic stress [12]. Moreover, BCL-2 family members localize to mitochondria-associated membranes, where they regulate ER–mitochondria calcium transfer, thereby linking metabolic signaling to apoptotic priming [13]. Metabolic stress conditions, such as nutrient deprivation or altered mitochondrial respiration, can sensitize cells to apoptosis by shifting the balance of BCL-2 family interactions, influencing cytochrome c release and caspase activation [14]. These metabolic–apoptotic interconnections are particularly relevant in cancer cells, where metabolic reprogramming contributes to apoptotic resistance and impacts the therapeutic efficacy of BCL-2 and BCL-xL inhibitors [14].
Because apoptosis is a fundamental tumor-suppressive mechanism, its evasion is considered a hallmark of cancer. BCL-2 and BCL-xL are frequently dysregulated across malignancies, where their overexpression supports tumor cell survival, enhances chemoresistance, and facilitates disease progression [15]. Functional assays such as BH3 profiling have shown that cancers can exhibit a specific dependency on BCL-2 or BCL-xL for survival, reflecting a measurable shift in apoptotic threshold that correlates with therapeutic sensitivity or resistance in different tumor types, a pattern less consistently observed for other antiapoptotic family members [16,17]. This dependency is not uniform across all antiapoptotic family members but is especially evident for BCL-2 and BCL-xL in various hematologic and solid tumor contexts, where their relative expression and binding affinities dictate apoptotic control and drug responsiveness [18].
Beyond apoptosis suppression, these proteins also influence metabolic adaptation, mitochondrial dynamics, calcium signaling, angiogenesis, and cell migration, highlighting their multifaceted roles in tumor biology [19]. Clinically, the therapeutic relevance of the BCL-2 family is underscored by the success of BH3 mimetics such as Venetoclax, approved for hematologic malignancies, although resistance mechanisms, particularly involving MCL-1 and BCL-xL, remain important challenges. In this review, we summarize the structural and functional characteristics of BCL-2 and BCL-xL, highlight their contribution to tumor biology, and discuss clinical trials exploring current therapeutic strategies targeting these key survival proteins.

2. BCL-2 and BCL-xL Genes and Regulation
The BCL2 gene is located on chromosome 18 (18q21.33) and is constituted by three exons (Figure 2). The BH domains are encoded by the first two exons, while the third encodes the transmembrane domain [20,21].
Transcription of BCL2 is promoted in response to signaling from diverse cytokines, such as interleukin (IL)-2, IL-3, IL-4, IL-6 and IL-7, or upon activation of antigen receptors which leads to the stimulation of several transcription factors [22]. In the endometrium and glandular cells, BCL2 gene transcription seems to be regulated by c-Jun. It is possible that, in the endometrium, this regulation depends on the interaction of c-Jun with estrogen–estrogen receptor α [23]. PR/SET domain 10, an epigenetic regulator involved in development and cell differentiation, also promotes BCL-2 mRNA expression by binding to BCL2 gene promoter [24]. Furthermore, the transcription factor GATA-1 is also involved in BCL2 transcription regulation, while GATA-4 was found to promote BCL-2 expression in both normal ovarian cells and ovarian granulosa cell tumors [25,26]. Moreover, NF-κB overexpression was associated with a six-fold increase in BCL2 promoter transcriptional activity, whereas mutations in the promoter eliminated this effect [27]. The BCL2 promoter also presents a CRE site where the transcriptional factor CREB, when phosphorylated, can bind and promote BCL-2 expression [28]. Sec6 and Sec8 were found to regulate BCL2 transcription by modulating CREB and NF-κB activity in malignant peripheral nerve sheath tumor cells. Depletion of Sec6 and Sec8 decreases BCL-2 expression [29].
In colon and liver cancer, methionine adenosyltransferase α2 was found to bind to the BCL2 promoter and induce its expression, but also to the BCL-2 protein stabilizing it [30].
On the other hand, c-Myc can repress BCL-2 expression to promote apoptosis. This regulation appears to require c-Myc binding to the transcription factor MIZ-1, leading to its functional inactivation [31].
At the post-transcriptional level several proteins have been reported to promote stability or to destabilize BCL-2 mRNA. In glioblastoma, human antigen R was found to bind to the 3′ untranslated region of both BCL-2 and BCL-xL mRNAs, stabilizing them and preventing their degradation [32]. Similarly, nucleolin has also been shown to bind to 3′ untranslated region in BCL-2 and BCL-xL mRNAs, stabilizing them [33]. Furthermore, ζ-crystallin, transformer 2β (TRA2β), and La-related protein 1 also act as stabilizers of BCL-2 in different types of cancer [34,35,36]. On the other hand, adenylate-uridylate-rich element RNA-binding protein 1 and zinc finger protein 36, C3H1 type-like 1 have been found to destabilize BCL-2 mRNA, leading to its degradation and consequently decreasing BCL-2 protein levels [35,37,38]. The following miRNAs have also been reported to regulate BCL-2 mRNA levels: miR-15/16, miR-21, miR-24, miR-29a, miR-34a, miR-124-3p, miR-125a, miR-125b, miR-153, miR-155, miR-181a, miR-195, miR-202, miR-204, miR-205, miR-206, miR-214, miR-223, miR-338-3p, miR-365, miR-370-3p, miR-383, miR-432-5p, miR-448, miR-497, miR-503-5p, miR-744, miR-1290, and miR-1915-3p [39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54].
Moreover, BCL-2 has two protein isoforms: BCL-2α, an antiapoptotic isoform with 239 amino acids, and BCL-2β, with 205 amino acids and no known function [21].
The BCL-X gene, also known as BCL2L1, localizes on chromosome 20 (20q11.21) [55]. The expression of BCL-X is induced by IL-2, IL-3, IL-6, granulocyte-macrophage colony-stimulating factor, colony-stimulating factor-1, leukemia inhibitory factor, erythropoietin, and also by the activation of antigen receptors (Figure 3) [22]. The BCL-X gene transcription is regulated by several transcription factor families such as STAT, NF-κB, E26 transformation specific sequence, and activator protein 1 complex [56]. Its expression is also stimulated by integrin, vitronectin, hepatocyte growth factor, and the activated RAS/mitogen-activated protein kinase (MAPK) pathway [22]. In prostate cancer cells, hypoxia-inducible factor (HIF)-1α was found to directly bind to a region of BCL-X promoter known as hypoxia-responsive element promoting BCL-xL transcription [57]. Furthermore, type 2 calreticulin mutations in myeloproliferative neoplasms were associated with activation of the transcriptional factor activating transcription factor 6, leading to increased transcription of BCL-xL [58]. Upon CD40 stimulation the NF-κB subunits p65 and p52 in chronic lymphocytic leukemia (CLL) cells bind to the BCL-X promoter, inducing BCL-xL expression [59]. MAPK-mediated phosphorylation of GATA-1 can also promote BCL-xL transcription. Nonetheless, GATA-1 activity is antagonized by Gfi1B [26].
GATA-3 and GATA-4 have also been reported to regulate transcription of BCL-xL [26]. Nonetheless, BCL-xL expression can be repressed by c-Myc to promote apoptosis [31].
Due to alternative splicing two major mRNA isoforms exist: BCL-xL, with 780 base pairs, and BCL-xS, with 591 base pairs [55]. The splicing is highly regulated involving several proteins like SRC associated with mitosis, of 68 kDa (SAM68), alternative splicing factor (ASF) 1/Serine/arginine-rich splicing factor (SRSF) 1, heterogeneous nuclear ribonucleoproteins (hnRNPs), RNA binding motif protein (RBM) 25, and RBM4 [56,60].
For instance, when SAM68 is overexpressed and associated with hnRNP A1, which is regulated by Fyn kinase, it promotes BCL-xS production, while when depleted BCL-xL production is induced [60]. However, factor that binds to inducer of short transcripts 1, a transcription factor, SRSF1 and SRSF10 can inhibit SAM68, promoting BCL-xL expression [60,61]. In lung cancer, the serine/threonine kinase CK1ε was found to phosphorylate SRSF10, which is suggested to be necessary for SRSF10 to bind to BCL-X pre-mRNA, leading to BCL-xL expression [62]. Nonetheless, SRSF10 can also promote BCL-xS in association with SAM68 and hnRNP A1/A2 in response to DNA damage. SRSF1 is positively regulated by NEK2 and SR protein kinase 1, but negatively regulated by PTBP1, also known as hnRNP I, and RBM4, which compete for the same binding sites [63]. Moreover, other proteins that can bind to RNA G-quadruplexes, usually found in introns, and G-quadruplex stabilizing small molecules, such as GQC-05, can also modulate BCL-X splicing. For instance, GQC-05 was shown to shift splicing to express BCL-xS [64]. U1 snRNP, hnRNP F, hnRNP H, RBM25, RBM10, RBM4, SRSF2, SRSF3, TRA2β, and staurosporine also promote BCL-xS expression [60,63,65,66,67,68]. Nonetheless, in glioma cells, SRSF1 and SRSF2 were shown to promote BCL-xL expression, whereas SRSF6 favored BCL-xS expression. The truncated form of dual-specificity and tyrosine phosphorylation-regulated protein kinase 1A, involved in Alzheimer’s disease, however, inhibits SRSF1, enhancing production of the BCL-xS isoform [69]. Similarly, RBM11 also antagonizes SRSF1 [70]. Ceramide, a component lipid that regulates growth pathways and cell stress responses, has also been implicated in the splicing of BCL-X by promoting BCL-xS production [63].
On the other hand, hnRNP K, SRSF7, splicing-factor-3B-subunit-1 (SF3B1), and SRSF9 are involved in the expression of BCL-xL [60,66,71]. Core and auxiliary proteins, like Y14, eukaryotic translation initiation factor 4A-III, RNA-binding protein with SR domain 1 (RNPS1), acinus, and Sin3A-associated protein, 18 kDa, part of the exon junction complex, were also shown to promote BCL-xL production since their depletion shifted the alternative splicing in favor of BCL-xS [72]. In addition, the SB1 region in the BCL-X pre-mRNA can be bound by a repressor of BCL-xS isoform. A slower RNA polymerase II (RNAPII) rate of elongation can allow the repressor to bind to this region and induce BCL-xL production. However, the splicing-related factor transcription elongation regulator 1 was shown to modulate the RNAPII rate of elongation shifting the production of BCL-xL to BCL-xS [73]. In multiple myeloma, N-acetyltransferase 10 was found to acetylate BCL-xL mRNA, stabilizing it and allowing upregulation of BCL-xL which is essential for the activation of the phosphatidylinositol 3-kinase (PI3K)/AKT pathway promoting cancer proliferation and progression [74].
At the post-transcriptional level, miRNAs regulate BCL-xL expression. For instance, the miRNAs miR-5-5p, Let-7b-5p, Let-7c/g, miR-34a, miR-125b, miR-133a/b, miR140-5p, miR-203a-3p, miR-203b-3p, miR-326, miR-377, miR-491, miR-608, miR-4270, and miR4300 all target BCL-xL [22,39,60,75,76,77]. In pancreatic cancer, the long non-coding RNA MIR4435-2HG was found to act as ceRNA that binds to miR-513a-5p, promoting BCL-X expression inducing cell proliferation [78].
Besides BCL-xL, antiapoptotic form with 233 amino acids, and BCL-xS, pro-apoptotic form with 170 amino acids, a third protein isoform, BCL-xβ, with 277 amino acids and no known function exists [55,56].

3. BCL-2 Family Members Domains and Functions
BCL-2 family members can be divided in three categories: prosurvival, which include BCL-2 and BCL-xL, containing four BH domains (BH1 to BH4); pro-apoptotic, such as BAX and BAK, containing three BH domains (BH1 to BH3); and another category comprising proteins that only contain the BH3 domain, like BAD and BIM (Figure 4) [5,79,80]. However, several other BCL-2 family proteins, such as BCL-G, BCL-2 family kin (BFK), and BCL-RAMBO, do not fit in these categories.
Nonetheless, several members have a hydrophobic C-terminal region suggested to function as a transmembrane anchor [5,79].
The BH1–3 domains form a hydrophobic loop which is needed for BCL-2 and BCL-xL to interact with apoptotic regulators, including BH3-only proteins, and perform their antiapoptotic activity [80].
The BH4 domain is essential for several functions of BCL-2 and BCL-xL. For instance, this domain is necessary for the heterodimerization of BCL-2 and BAX which leads to BAX inhibition [81]. Furthermore, phosphorylation of BCL-2 at serine (Ser) 87 is important for this interaction since it decreases BCL-2’s affinity to BAX [82].
The BH4 domain is also necessary for BCL-2-dependent recruitment of rapidly accelerated fibrosarcoma-1 to the mitochondria, where it serves as a scaffold for BAD and protein kinase–theta interaction, leading to BAD phosphorylation, inhibiting it and consequently apoptosis [83,84]. The antiapoptotic function of BCL-2 depends on the phosphorylation of its Ser70 residue that is negatively regulated by protein phosphatase 2A-B56δ when cells are subjected to oxidative stress (Figure 5 and Table 1) [85,86].
The phosphorylation of Ser70 can be carried out by MAPK and protein kinase Cα [87,88]. In addition, p38 MAPK phosphorylates the residues threonine (Thr) 56 and Ser87 of BCL-2, decreasing its antiapoptotic activity [89]. During oxidative stress, tumor necrosis factor-α induces the dephosphorylation of these residues and leads to BCL-2 degradation [90]. Cyclin-dependent kinase (CDK) 1 was also found to phosphorylate BCL-2 at Thr56, during G2/M, leading to cell cycle inhibition, even in normal cells [91]. CDK1 in association with cyclin B1 also phosphorylates BCL-2 at Thr69, Ser70, Thr74, and Ser87, leading to increased affinity to BAK and BIM [92]. Furthermore, it has been reported that in BCL-2 and BCL-xL, the cleavage of the BH4 region, performed by caspase-1 or -3, can lead these proteins to promote apoptosis instead of their usual role of inhibiting it [80]. Interestingly, during exposure to Cisplatin, CDK2 was shown to phosphorylate BCL-xL at Ser73, leading to BCL-xL behaving like BAX and promoting apoptosis (Table 2) [93]. In addition, phosphorylation of the Ser62 residue or the deamidation of the asparagine residues on positions 52 and 66 in the intrinsically disordered region (IDR), also known as flexible loop domain, of BCL-xL promotes a structural rearrangement decreasing BCL-xL affinity for pro-apoptotic BH3 domains [94]. Deamidation will also promote BCL-xL degradation. It has been shown that the rate of deamidation can increase in the presence of DNA-damaging agents [95]. Two histidines close to the asparagine residues seem to be crucial for the promotion of BCL-xL deamidation by sensing pH increases caused by DNA-damaging agents [96].
The region of BCL-xL that the phosphorylated and/or deamidated IDR interacts with is also the region that interacts with p53. The interaction with p53 suppresses apoptosis dependent on BAX. Furthermore, PUMA can also bind to this region competing with p53 and BH3-only proteins, releasing them from BCL-xL [94,97]. Free p53 can then activate BAX and subsequently apoptosis [98].
BCL-xL needs to be present in the mitochondria’s outer membrane to promote its integrity. This is achieved by the interaction of BCL-xL with vacuolar protein sorting (VPS)35 and VPS26, which are components of retromer, a protein complex part of the endosomal protein sorting machinery. MICAL-like protein 1, a protein associated with retromer, is also important for BCL-xL mitochondrial localization [10]. Mitofusin 2 and mitofusin 1, two dynamin-related GTPases, also interact with BCL-xL, but only mitofusin 2 is necessary for BCL-xL-induced mitochondrial aggregation [99].
To prevent MOMP, BCL-xL inhibits BAX activity. Nonetheless, this process is still poorly understood. BAX is a pro-apoptotic BCL-2 family member that has a crucial role in the formation of pores in the mitochondrial membrane that leads to MOMP [2].
There are several proposed models for BAX and BAK, which is also involved in MOMP promotion, repression, and activation. In the direct model, BH3-only proteins can be divided in activators, such as BID and BIM, and sensitizers like BIK and PUMA [100]. For instance, when BID is cleaved by caspase-8 it converts into its active form, cBID, which contains p7 and p15 (tBID), and it translocates to the outer membrane of the mitochondria where it can activate BAX and BAK and also promote BCL-xL insertion on the membrane. However, BCL-xL inhibits BAX activation by binding to tBID or to BAX, blocking their interaction [101,102,103]. The complex formed by BCL-xL and BAX is stabilized by the interaction of vaccinia-related kinase-2A, a nuclear envelope kinase, and BCL-xL, hindering BAX dissociation while BAD binds to BCL-xL, allowing BAX and tBID interaction [104,105]. Thus, in the direct model, proteins like BAD and BIK inhibit the interaction of BCL-xL and BAX, consequently promoting apoptosis [100].
On the other hand, the indirect model proposes that BAX and BAK do not interact directly with BH3-only proteins but are indirectly activated by the inhibition of the prosurvival BCL-2 members carried by BH3-only proteins. The difference between the direct and indirect models is that in the indirect one all BH3-only proteins act as sensitizers [100]. A more recently proposed model suggests a combination of both models where prosurvival proteins can both sequester the activators and BAX and BAK [56].
BCL-xL is also involved in the retrotranslocation of BAX from mitochondria to cytosol by leading to the establishment of weak inhibitory mitochondrial complexes [106]. It is suggested that in non-apoptotic cells BAX localization in the mitochondria or cytosol follows a dynamic equilibrium that can be shifted in favor of a more cytosolic one by BCL-xL [2]. It seems that BAD plays a role in this retrotranslocation by binding to BCL-xL, promoting the release of BAX, and that AKT-mediated phosphorylation of BAD regulates this process [107]. Paradoxically, it was also found that BCL-xL overexpression led to higher concentration of BAX in the mitochondria which can possibly be explained by the fact that BCL-xL is also involved in the translocation of BAX from the cytosol to the mitochondria [2]. More recently, it was reported that binding of BCL-xL with BH3-only members inhibits its retrotranslocation, causing their accumulation in the mitochondria. It is suggested that this process might recruit pro-apoptotic proteins to the mitochondria whilst inhibiting them [108]. Additionally, it was shown that increased levels of BCL-2 and MCL-1 accelerated retrotranslocation of BAX [109]. Nonetheless, for BAK, only MCL-1 and BCL-xL increased expression led to a similar effect [110]. E2F1, a transcriptional factor, has been shown to inhibit BCL-xL retrotranslocation which seems to be essential for BCL-xL’s role in BAK inhibition [111].
Ras, a GTPase, is also involved in both pro-apoptotic and antiapoptotic signaling. In apoptosis induced by the cell membrane death receptor Fas signaling, Ras is activated. Nonetheless, active mitochondrial Ras is then regulated by BCL-2 activity, leading to the repression of Ras-mediated apoptotic signaling [112].
Both BCL-xL and BCL-2 regulate autophagy by interacting with Beclin-1, inhibiting its activity [113,114]. Beclin-1 forms, alongside other proteins, class III PI3K complexes that play a role in the formation of the autophagosome [115]. The interaction between BCL-2 and BCL-xL with BAK, BAX, and Beclin-1 can be inhibited by the JNK-mediated phosphorylation of BCL-2 and BCL-xL, leading to the promotion of apoptosis and autophagy [116]. Furthermore, in osteoclasts precursors, autophagy is induced by the receptor activator of NF-κB ligand-facilitated phosphorylation of BCL-2 Ser70 which decreases interaction between BCL-2 and Beclin-1 [82]. On the other hand, PARK2-mediated mono-ubiquitination of BCL-2 seems to enhance the interaction between BCL-2 and Beclin-1 [117].
BCL-2 is also involved in the inhibition of GABARAP, part of the GABARAP subfamily that is involved in phagophore and autophagosome formation, through the BH4 domain, affecting GABARAP lipidation and consequently inhibiting autophagy [118,119].
Other than the mitochondria, BCL-xL and BCL-2 also play a role in the ER where they are involved in the repression of inositol 1,4,5-trisphosphate receptor (IP3R) activity [120]. IP3R is a Ca2+ channel that is involved in important cellular events such as proliferation and apoptosis [121]. Thus, BCL-2 and BCL-xL-mediated inhibition regulates Ca2+ pro-apoptotic release. BCL-2 inhibits IP3R through the interaction of the BH4 domain with the receptor. Additionally, the transmembrane domain is essential for BCL-2 localization near the IP3R transmembrane domain [120,122]. However, BCL-xL-mediated inhibition of IP3R does not seem to be dependent on the BH4 domain but rather on BH3 domain since, when lysine (Lys) 87 was mutated, BCL-xL lost the capacity to interact with IP3R [120,123]. The transmembrane domain and BCL-xL capacity to dimerize are also suggested to play a role in this interaction [100]. Nonetheless, it seems that BCL-xL can both activate and inhibit IP3R, and the role is dependent on BCL-xL concentration. At low concentrations BCL-xL activates IP3R, while at higher concentrations BCL-xL inhibits it [124]. Furthermore, BCL-2 and BCL-xL inhibit ryanodine receptors in a similar way. Ryanodine receptors are Ca2+ channels that can be found in several types of cells, including hippocampal neurons, and it is suggested that they also play a role in apoptosis [125,126]. BCL-xL also inhibits voltage-dependent anion channel 1 (VDAC1), preventing Ca2+ pro-apoptotic signals from entering the mitochondria [127]. Accumulation of misfolded proteins in the ER leads to ER stress which when unresolved triggers apoptosis. It is suggested that the promotion of apoptosis during ER stress occurs through the RING finger (RNF) 183-mediated ubiquitination of BCL-xL and its consequent degradation [128].
In hippocampal neurons, BCL-xL has been shown to increase both the energy metabolism and synaptic activity while also promoting synapse formation through dynamin-related protein 1 (Drp1), a GTPase involved in neurite growth [129]. For instance, in rat hippocampal neurons, BCL-xL was shown to promote synapse formation through the regulation of caspase-3 by inhibiting Drp1-dependent mitochondrial fission [130]. In addition, BCL-xL interaction with Drp1 and clathrin is required for vesicle endocytosis regulation. This function is dependent on calmodulin which promotes the translocation of both BCL-xL and Drp1 to synaptic vesicles [131]. BCL-xL also plays a role in neurite growth by inhibiting death receptor 6 activity both under normal and hypoxic conditions [129].
BCL-2 and BCL-xL have also been shown to play a role in hematopoietic differentiation. In IL-3-deprived factor-dependent cell-Patersen Mix multipotent progenitor cells, BCL-2 led to their differentiation into granulocytes and monocytes/macrophages while the ones expressing BCL-xL differentiated into erythroid cells [132].
A role in the cell cycle has also been proposed for BCL-2 since increased levels of BCL-2 are associated with delays in the transition from G0/G1 to S phase. This delay is promoted by the regulation of reactive oxygen species (ROS) and ATP levels. It is suggested that BCL-2 can regulate mitochondrial metabolic pathways, leading to a decrease in both ATP and ROS levels while increasing p27kip1, a CDK inhibitor, expression [133]. It is also suggested that BCL-2 increases p130 expression and that p27 leads to CDK2 inhibition. CDK2 is involved in the degradation of both p27 and p130. p130 in turn forms a repressive complex with E2F4 that, as proposed in the model, should repress transcription of genes that are involved in cell cycle entry [134]. Moreover, BCL-2’s antiapoptotic function is inactivated through phosphorylation of Ser70 at G2/M through the ASK1/JNK pathway [135]. As previously referred, CDK1 also phosphorylates BCL-2 at G2/M. This phosphorylation occurs at Thr56, inducing G2/M arrest [91]. However, it has been found that BCL-2 phosphorylated at Thr56 accumulates in nuclear structures in early prophase, while in late prophase it localizes around mitotic chromosomes. In metaphase, BCL-2 remains localized around mitotic chromosomes. It seems that BCL-2 forms complexes with CDK1, nucleolin, and PP1 and that dynamic phosphorylation of BCL-2 might regulate its function during mitosis [136].
BCL-xL was also shown to play a role during the cell cycle where it undergoes dynamic phosphorylation/dephosphorylation events at both Ser49 and Ser62. A pool of Ser49-phosphorylated BCL-xL is found at centromeres during G2 checkpoint, a result of DNA damage. Further, another pool of phospho-BCL-xL Ser49 occurs through telophase where it seems to play a role in cytokinesis. The phosphorylation of BCL-xL is performed by polo-like kinase (PLK) 3, which is consistent with the proposed role of BCL-xL in the cell cycle since PLK3 is involved in cell cycle progression [137]. The phosphorylation at Ser62 has also been implicated in G2 arrest and is promoted by both PLK1 and JNK2, leading to BCL-xL recruitment to nucleolar structures during the stabilization of G2 arrest [138]. Additionally, PLK1 and MAPK14/p38α-mediated BCL-xL phosphorylation of Ser62 during prometaphase and metaphase led to BCL-xL’s association with the spindle assembly checkpoint silencing complexes and also to its localization at centromeres. It is suggested that BCL-xL might be involved in spindle assembly and chromosome segregation [139]. In human diploid fibroblasts this dynamic was shown to promote chromosome stability and prevent aneuploidy [140]. Furthermore, in Rat1 fibroblasts, BCL-xL was also shown to delay the cell cycle, while BAD has been shown to counter the cell cycle arrest promoted by both BCL-2 and BCL-xL [141].
BCL-xL has also been implicated in DNA damage response. JNK is activated in response to genotoxic agents, leading to its translocation to the mitochondria where it phosphorylates BCL-xL in both Thr47 and Thr115, promoting apoptosis [142]. Additionally, in rat cardiac myocyte, response to oxidative stress leads to the activation of K-Ras which consequently promotes the Ras association domain family (RASSF) 1A-mediated activation of the mammalian sterile 20-like kinase (MST) 1. MST1 in turn phosphorylates BCL-xL at Ser14, repressing its interaction with BAX, thus inducing apoptosis [143].
BCL-xL also plays a role in the regulation of mitophagy. It was found that BCL-xL interacts with PARK2, also known as Parkin, preventing its translocation to the mitochondria but also with PINK1, blocking the binding of PARK2 to PINK1, repressing mitophagy [144]. PINK1 was also found to phosphorylate BCL-xL at Ser62, promoting its antiapoptotic function [145].

4. Function of BCL-2 and BCL-xL in Cancer

4.1. Invasiveness and Progression
Besides their roles in non-transformed cells, BCL-2 and BCL-xL have also been reported to play roles in pro-tumoral events.
For instance, in cancer cells, BCL-2 pro-survival family members can promote cell invasion and migration by increasing ROS levels, although not high enough to stimulate cell death [152].
It has been proposed that BCL-2 regulates ROS production by modulating mitochondrial respiration in cancer cells. BCL-2 interacts with the cytochrome c oxidase (COX) Va subunit through its BH2 domain and its C-terminal region, promoting the localization of COX Va to the mitochondria. High BCL-2 expression also correlates with increased mitochondrial localization of COX Vb localization at the mitochondria, although this effect is thought to occur indirectly as a consequence of COX Va enrichment in the mitochondria. This way, BCL-2 increases COX activity and ROS production. However, under oxidative stress, BCL-2 decreases COX Vb presence in the mitochondria and stabilizes COX Va, leading to decreased COX activity and lower ROS levels [153]. In breast cancer, BCL-xL-mediated increase in ROS was shown to be dependent on BCL-xL interaction with VDAC1 [154]. Furthermore, it has been reported that BCL-xL promotion of metastasis is independent of its apoptotic activity and that the metastatic function is only observed for BCL-xL localized in the nucleus where it increased tri-methylation of histone 3 in Lys4, a marker of transcriptional activation [155]. BCL-xL translocation to the nucleus seems to be carried out by the transcriptional regulator C-terminal binding protein 2 [156].
In breast cancer, BCL-xL overexpression induces lymph node metastasis by preventing cytokine-induced cell death and promoting the ability of cells to proliferate in an anchorage-independent manner [157]. ER-positive breast cancer cells present higher expression of the lncRNA BC200. BC200 participates in breast cancer pathogenesis since it binds to BCL-X pre-mRNA and recruits hnRNP A2/B1 which in turn blocks SAM68 from binding to the pre-mRNA. This way, BC200 promotes the production of BCL-xL [158]. Furthermore, in triple-negative breast cancer (TNBC), high levels of the kinase Aurora A and BCL-xL were linked to promotion of metastasis [159]. In addition, BCL-2 overexpression leads to NF-κB activation, increasing matrix metalloproteinase (MMP)-9 expression in breast cancer. MMP-9 is associated with tumor metastasis and invasion [160]. It also increases MMP-2, it too associated with invasiveness, in cooperation with N-MYC, a transcription factor involved in cell proliferation, in neuroblastoma and lung cancer [161,162].
In lung cancer, BCL-xL expression is associated with migration and invasion promotion since increased expression of Let-7a-5p led to downregulation of BCL-xL and repression of these processes. Moreover, Let-7a-5p expression induces toxic autophagy by hindering the PI3K signaling pathway [163]. γ-irradiation also promoted migration and invasion in lung cancer by promoting STAT3 activity. STAT3 then induced BCL-xL transcription. Increased expression of BCL-xL was associated with augmented expression of MMP-2 and vimentin, enhanced phosphorylation of p38 and AKT, and downregulation of E-cadherin [164].
C-X-C chemokine receptor type 4 (CXCR4), a chemokine receptor, has been shown to promote tumor growth and survival by modulating microRNAs: in neuroblastoma, it downregulates miR-15a/16-1, leading to increased BCL-2 and cyclin D1 expression, while in ovarian cancer, it suppresses let-7a, resulting in the upregulation of BCL-xL [165].
Similarly, in melanoma cells, the overexpression of miR-365 led to inhibition of cell proliferation by downregulating BCL-2 and cyclin D1 [166]. BCL-2 is also involved in progression in melanoma by promoting stability of Semaphorin 5A, an axon regulator, at the mRNA and protein levels. Semaphorin 5A promotes migration through the activation of the MEK/ERK pathway [167]. The lncRNA LHFPL3-AS1-long also plays a role in melanoma stem cells tumorigenesis by binding to miR-181a-5p, preventing BCL-2 mRNA degradation [168]. Moreover, metastatic melanoma patient samples showed higher expression of BCL-2 and BCL-xL than primary melanoma, benign nevi, and normal skin samples, suggesting a role for these proteins in melanoma progression [169]. Accordingly, BCL-2 overexpression was found to be associated with higher expression of MMP-2 and MMP-7 [170]. In melanoma and glioblastoma cells, BCL-xL overexpression was found to promote migration, invasion, and angiogenesis [171].
Furthermore, glioblastoma progression is promoted in part by increased expression of BCL-xL caused by downregulation of tumor suppressor candidate 2 due to neural precursor cell-expressed developmentally downregulated 4-mediated polyubiquitination [172]. Accordingly, inhibition of the splicing factor SF3B1 reduced migration, tumorigenesis, and vascular endothelial growth factor (VEGF) secretion by shifting BCL-X pre-mRNA for BCL-xS production and repressing the AKT/mTOR/β-catenin pathways [71]. Phosphatase and tensin homolog (PTEN), in normal cells, presents tumor suppressive activity. However, in glioblastoma cells with mutant p53, PTEN was shown to interact with the complex formed by mutant p53, acetyl-CBP, and NFYA, promoting their binding to BCL-xL and c-Myc promoter regions upregulating these proteins. Consequently, the upregulation of BCL-xL and c-Myc promoted cell invasion, proliferation, and tumor progression [173]. Moreover, BCL-2 expression in glioma cells increases Furin and transforming growth factor-β (TGF-β) expression which in turn leads to the upregulation of MMPs, promoting glioma cells invasiveness [174]. Similarly, BCL-xL upregulation induced TGF-β2, MMP-2, and MT1-MMP, promoting invasiveness of malignant glioma cells [175].
BCL-2 overexpression also promotes migration through modulation of urokinase-type plasminogen activator receptor (uPAR). uPAR is the receptor of uPA, a protease that converts the extracellular zymogen plasminogen to plasmin. Plasmin is involved in invasion and metastasis. By activating ERK leading to increased Sp1, a transcriptional factor, activity, BCL-2 promotes uPAR expression and consequently invasion [176]. BCL-2 was also shown to promote cell migration through interference in the Hippo pathway since it reduces MST2 protein levels. MST2 regulates Yes-associated protein (YAP), a transcription coregulator, by activating large tumor suppressor kinases (LATS) 1/2. LATS1/2 activation leads to the maintenance of YAP in the cytoplasm, inhibiting its transcriptional activity [177]. Microphthalmia-associated transcription factor (MITF) is a transcription factor involved in the expression of miR-211, a microRNA, associated with migration and invasion suppression, that is regulated by BCL-2. BCL-2 represses MITF nuclear activity probably through the interaction with heat shock protein (HSP) 90 [178]. In addition, both BCL-2 and BCL-xL can bind to a tumor suppressor protein known as SUFU, repressing its binding to GLI proteins. GLI proteins regulate some cell proliferation genes’ expression, and their binding to SUFU suppresses this activity, decreasing cell proliferation [179].
Circular RNAs have also been reported to contribute to cancer progression since, in osteosarcoma, circ_0000376 was found to bind to miR-432-5p, allowing the expression of BCL-2. Inhibition of circ_0000376 led to the repression of osteosarcoma cells migratory and invasive capabilities and increased cell death [45].
The knockdown of BCL-2, BCL-xL, or MCL-1 hindered migration and invasion in colorectal cancer (CRC) cells [180]. Moreover, BCL-2 was found to be associated with colon carcinogenesis since overexpression of miR-15a, which regulates BCL-2 expression, led to lower cell proliferation and reduced cell invasion properties [181]. Additionally, BCL-2 expression has been associated with early stages of carcinogenesis in colorectal neoplasias [182]. It has also been reported that BCL-2 was upregulated in CRC metastatic cells when compared with nonmetastatic ones [183]. Furthermore, dual specificity phosphatase 4 (DUSP4) is also associated with carcinogenesis in CRC. DUSP4 prevents JNK-mediated phosphorylation of BCL-2, while its silencing blocks the interaction of BCL-2 with Beclin-1 or BAX, increasing autophagy and apoptosis and repressing migration and invasiveness [184]. Depletion of circDUSP16 also affected CRC cells’ migration, invasion, and proliferation due to decreased BCL-2 expression. circDUSP16 targets miR-432-5p, and its depletion allows miR-432-5p to downregulate E2F6. E2F6 overexpression was previously reported to be associated with c-Src/ERK and BCL-2 upregulation. Thus it seems that E2F6 promotes BCL-2 expression by inducing c-Src/ERK signaling [185,186]. Besides BCL-2 overexpression in CRC, BCL-2 Ser70 phosphorylation status was also related with tumor aggressiveness since tumors in more advanced stages showed lower expression of phosphorylated Ser70 [187]. BCL-xL is also involved in CRC cells’ invasion, proliferation, and clonogenic formation since its inhibition impaired all of these processes [188]. The protease-activated receptor 2 (PAR2) is also involved in cancer progression and in the regulation of the immune microenvironment. PAR2 stabilizes BCL-xL by promoting Ser145 phosphorylation. The phosphorylation of this residue prevents interaction with RNF152 and consequently BCL-xL polyubiquitination and degradation. Expression of BCL-xL inhibits type I IFN secretion, preventing recruitment of CD8+ T cells to metastatic sites [150,189]. Additionally, the lncRNA HEIH is involved in CRC tumorigenesis by targeting miR-939, preventing repression of BCL-xL transcription since miR-939 binds to NF-κB, blocking its binding to BCL-X promoter [190].
In oral cancer several studies have shown that BCL-2 is associated with differentiation since BCL-2 expression is increased in sequentially progressing epithelial dysplasia and in poorly differentiated carcinomas when compared to well-differentiated ones [191,192,193,194,195]. In accordance, Niedzielska et al. showed that in patients with squamous cell carcinoma (SCC), BCL-2 is more expressed than in patients with hyperplasia and patients with neoplasm in situ malignancy [196]. Nevertheless, tissue from the incision line, close to the tumor, showed higher BCL-2 expression in the three groups assessed. Similar results were reported by Juneja et al. [197]. However, in some studies this relation was not observed, with BCL-2 being sporadic in oral premalignant tissue [198,199]. On the other hand, it has been reported that there is no significant difference regarding BCL-xL expression between poorly differentiated oral squamous cell carcinoma (OSCC) and basaloid SCC [200]. Nonetheless, BCL-xL expression in OSCC is correlated with progression and resistance to Cisplatin treatment [201]. Overexpression of BCL-2 in OSCC was also found to lead to increased expression of MMP-9 and enhanced migration and invasion behavior [202]. Moreover, overexpression of BCL-2 and p53 seems to be mutually exclusive, suggesting that both genes can induce carcinogenesis in OSCC, independently [203]. Nonetheless, tumors expressing both proteins showed higher probability of presenting unfavorable characteristics [204]. Another study found higher mRNA ratios of BCL-2/BAX mostly in poorly differentiated oral carcinomas, while BAX at the protein level was downregulated in these poorly differentiated carcinomas [205]. The downregulation of BCL-2 by miR-34a was also found to inhibit sinonasal SCC migration and invasion capabilities [43].
The BCL-2 antiapoptotic family members have been found to contribute to leukemogenesis promoted by human T-cell leukemia virus type 1 and bovine leukemia virus. These viruses upregulate BCL-2, BCL-xL, MCL-1, and BFL-1 while they downregulate BAX, BIM, and BID. This way infected lymphocytes can survive and proliferate, increasing genomic instability and promoting leukemogenesis [206].
In acute myeloid leukemia (AML), spastic paraplegia 6 protein, a dominant autosomal HSP, is associated with disease progression since it regulates the BMPR2-SMAD-BCL-2/BCL-xL pathway [207]. Furthermore, coexpression of MYC and BCL-xL or BCL-2 can drive AML tumorigenesis [208].
In cervical cancer, tumorigenesis was found to be promoted by the lncRNA RUSC1-AS1 since it acted as a CeRNA, inhibiting miRNA-744 and leading to increased expression of BCL-2 [41].
In retinoblastoma, a correlation between BCL-2 expression and tumor invasiveness and poor differentiation was found [209].
In hepatocellular carcinoma (HCC), follistatin-like protein 5 was shown to inhibit cancer progression by downregulating BCL-2 and upregulating BAX, BAD, and PUMA [210]. Similarly, both miR-202 and miR-448 were also found to repress HCC progression and cell growth by targeting BCL-2 [42,44].
BCL-xL is also involved in prostate cancer progression since a study showed that the overexpression of miR-608 which targets BCL-xL mRNA represses cancer progression, while another study demonstrated that overexpression of BCL-xL increases cancer progression by repressing senescence and apoptosis, reducing survival in vivo [211,212].
BCL-xL expression was also found to be necessary to promote islet tumor cells and pancreatic neuroendocrine cancer invasiveness [213,214].
Progression in cholangiocarcinoma is associated with aberrant alternative splicing. For instance, high expression of serine/arginine protein kinase (SRPK) 1 and SRPK2 is usually found on this type of cancer, leading to phosphorylation of SRSFs that promote antiapoptotic splicing isoforms such as MCL-1 and BCL-xL. Consequently, inhibition of SRPK1 and SRPK2 led to increased expression of MCL-1S and BCL-xS, promoting cell death [215].
PARK2 is a tumor suppressor that regulates the cell cycle and programmed cell death. In several cancers, PARK2 is usually dysregulated, resulting in increased proliferation and repression of apoptosis. It was found that PARK2 can ubiquitinate BCL-xL, and this regulation is essential for PARK2 tumor suppression activity. Since PARK2 promotes BCL-xL degradation, the dysregulation of PARK2 leads to high expression of BCL-xL and to cancer cell death prevention [117].
Both BCL-2 and BCL-xL play critical roles in tumor proliferation and invasiveness which are related with more aggressive tumors. Thus, BCL-2 and BCL-xL can be important targets in the treatment of more advanced tumors.

4.2. Angiogenesis
Angiogenesis is an important process for cancer cells since it is essential for cell growth and metastasis, and BCL-2 and BCL-xL have been found to regulate this process in several types of cancer. For instance, overexpression of BCL-xL leads to increased levels of IL-8, through NF-κB activation, and consequently to the promotion of angiogenesis in melanoma and glioblastoma cells [216,217]. Similarly, in a zebrafish melanoma xenograft model, BCL-xL was also found to promote angiogenesis through the regulation of IL-8 [218]. Additionally, BCL-2 overexpression enhances VEGF expression by promoting PI3K and MAPK pathways signaling in cancer cells subjected to hypoxic conditions [219,220]. Similarly, in melanoma and breast cancer cells, increased BCL-2 expression also led to induction of angiogenesis by modulation of VEGF expression through HIF-1 and by stabilization of VEGF mRNA [221,222]. In lymphoma, overexpression of BCL-2 isoform β was shown to promote angiogenesis by interacting with the chaperonin T-complex protein ring complex essential for the secretion of VEGF-A and vessel tube formation [223]. In colon cancer, BCL-2 expression was reported to be correlated with VEGF expression and microvessel density which potentially suggests a role for BCL-2 in angiogenesis in this type of cancer [224].
In HCC, BCL-2 was found to interact with the transcription factor Twist1 through both its BH2 and transmembrane domains. Under hypoxic conditions these proteins are coexpressed. Their interaction increases Twist1 nuclear presence, inducing the expression of several genes such as VEGFR1, VEGFR2, MMP-2, and MMP-9 involved in processes like angiogenesis and epithelial–mesenchymal transition [225].
Angiogenesis is essential to provide oxygen and nutrients to cancer cells needed for their growth but also allows cells to escape to the bloodstream and invade other tissues. Thus, it is not surprising that BCL-2 and BCL-xL play a role in the regulation of angiogenesis since they are so important for cell invasiveness.

4.3. BCL-2 and BCL-xL Effects on Drug Responses
BCL-2 and BCL-xL are not only involved in processes vital for cancer cell survival and tumor progression but also in the resistance to treatment. For instance, BCL-2 and BCL-xL overexpression has been shown to lead to radioresistance and chemoresistance to several drugs, including Cytosine arabinoside, Cisplatin, Topotecan, Gemcitabine, Docetaxel, and Paclitaxel [226,227,228,229,230,231,232,233,234,235,236,237,238,239,240,241,242,243].
In advanced bladder cancer patients previously treated with radiotherapy, lower expression of BCL-2 led to better survival after treatment with Cisplatin than in patients with high BCL-2 expression [244]. Additionally, BCL-2 expression was associated with worse response to treatment with concurrent radiotherapy and platinum therapy in advanced oropharyngeal SCC. BCL-2, but not BCL-xL, expression was also found to induce resistance in vitro [245].
In gallbladder cancer and osteosarcoma cell lines, miR-125b was found to increase sensitivity to Cisplatin by downregulating BCL-2 [246,247]. Similarly, miR-204 also conferred sensibility to Cisplatin in neuroblastoma [248]. On the other hand, nicotine was found to increase resistance to Cisplatin by increasing BCL-2 expression in oral cancer [249]. In ovarian cancer, inter-α-trypsin inhibitor heavy chain 3, involved in the stabilization of the extracellular matrix, downregulation increases BCL-2, BCL-xL, and MCL-1 expression after Cisplatin treatment, leading to resistance to this drug [250]. Similarly, in bladder cancer, the downregulation of genes associated with retinoid-interferon-induced mortality-19 also induced Cisplatin resistance by reducing BCL-xL polyubiquitination and degradation [151]. Suppression of BCL-xL deamidation has also been found to lead to resistance to DNA-damaging agents such as Cisplatin [251,252]. Cisplatin treatment in gastric cancer cells increased expression of CDK1 which in turn activated DNA methyltransferase 1, silencing miR-145. miR-145 suppression increases SRY-box transcription factor 9 (SOX9) expression and consequently BCL-xL expression since BCL-xL is a direct transcriptional target of SOX9. Increased expression of BCL-xL decreases gastric cancer cells sensitivity to Cisplatin [253]. Additionally, overexpression of miR-193a-3p in CD44+ gastric cancer cells leads to increased expression of BCL-xL and resistance to Cisplatin by targeting SRSF2 [254].
In mesothelioma cell lines, inhibition of both BCL-xL and BCL-2 increased cell sensitivity to Cisplatin [255]. However, expression of BCL-xL was found to only confer resistance to Cisplatin in head and neck squamous cell carcinoma (HNSCC) cells with wild type p53 since HNSCC cells with mutant p53 were sensitive to Cisplatin regardless of BCL-xL expression [235].
Moreover, knockdown of BCL-2, BCL-xL, or MCL-1 enhanced CRC cells’ sensitivity to Oxaliplatin [180]. Nonetheless, Oxaliplatin apoptosis induction occurs due to a shift in splicing from BCL-xL to BCL-xS. Oxaliplatin weakens 14-3-3ε binding to SRSF10 which dissociates from hnRNP F/H, but not hnRNP K, resulting in its dissociation from the pre-mRNA. 14-3-3ε continuous interaction with hnRNP A1 and Oxaliplatin-induced dephosphorylation of SAM68 enhance the affinity of SAM68 to hnRNPA1 which then represses RNPS1, thus promoting BCL-xS production [256]. Gfi1, a transcriptional repressor, was also shown to prevent DNA damage-induced apoptosis by repressing the transcription factor PU.1, averting Hemgn degradation. Hemgn then activates the BCL-X promoter, upregulating BCL-xL [257].
Breast cancer cell lines overexpressing BCL-2 showed increased resistance to Cisplatin and Bischloroethylnitrosourea while showing higher sensitivity to Doxorubicin, Vincristine, Vinblastine, and Actinomycin D [258].
On the other hand, in a non-Hodgkin lymphoma cell line, increased expression of HIF-1α led to resistance to treatment with Cisplatin and Doxorubicin by promoting BCL-xL expression [259]. Likewise, high expression of BCL-xL in HCC cells induces resistance to Doxorubicin. Interestingly, Pyrrolidine dithiocarbamate, an antioxidant that can also inhibit NF-κB, overcame this resistance by inducing paraptosis [260]. A chronic myeloid leukemia (CML) CD44v16 cell line was also found to be resistant to Doxorubicin through the activation of the NF-κB/Snail/BCL-2 pathway [261].
Furthermore, BCL-2 positive non-small cell lung cancer (NSCLC) patients were found to be less responsive to treatment with Cisplatin combined with Gemcitabine than BCL-2 negative ones [232]. Similarly, in resistant non-Hodgkin lymphoma, low expression of BCL-2 was predictive of a better response to treatment with Gemcitabine in combination with Cisplatin and Dexamethasone [262].
BCL-xL has been associated with Gemcitabine resistance and the addition of DT2216, a BCL-xL degrader, overcame Gemcitabine resistance both in vitro and in PDTX pancreatic cancer models [233]. Moreover, pancreatic carcinoma cells with higher BCL-2 expression showed a higher Gemcitabine 50% lethal dose than cell lines with lower BCL-2 expression [263].
One of the most common TNBC treatment options is the combination of Doxorubicin, Cisplatin, and 5-fluorouracil (5-FU). Nonetheless, in a TNBC cell line, resistance to this combinatorial approach arose from the expression of BCL-xL [264].
In CRC, paxillin, an adapter protein, was found to phosphorylate BCL-2 at Ser87, promoting its stability and leading to resistance to 5-FU [148]. BCL-xL expression was also found to confer resistance to 5-FU and radiotherapy in CRC, while inhibition of BCL-xL combined with 5-FU or radiotherapy led to synergistic effects [265]. In gastric carcinoma, miR-383 was shown to increase cell sensitivity to 5-FU by targeting BCL-2 mRNA [50]. Additionally, high expression of phosphatase phosphoglycerate mutase family member 5 (PGAM5) in HCC patients leads to resistance to treatment with 5-FU due to PGAM5 and BCL-xL interaction that promotes BCL-xL stabilization [266].
The mechanism behind BCL-2-induced resistance to DNA-damaging drugs such as Cisplatin and 5-FU is related to the Ser70 phosphorylation of BCL-2. The phosphorylation of this residue decreases the affinity of BCL-2 to the mitochondrial complex-IV subunit-5A. This interaction is essential for mitochondrial complex-IV activity and ROS production. Since Etoposide and Doxorubicin led to the reduction in Ser70 phosphorylation, resistance to these drugs was not observed [267].
Furthermore, glycochenodeoxycholate, the principal compound in the bile, also promotes BCL-2 Ser70 phosphorylation, increasing HCC cell survival and promoting chemoresistance [87]. Active GTPase-Rac1 was also found to promote BCL-2 phosphorylation at Ser70, preventing apoptosis of cancer cells [268].
On the other hand, in breast cancer, PARK2 expression confers sensitivity to antimicrotubule agents such as Docetaxel and Vinorelbine by interacting with BCL-2 phosphorylated in Ser70, promoting BCL-2 polyubiquitination and consequently its degradation. Treatment with these drugs leads to PARK2 upregulation which can then lead to degradation of BCL-2, inducing BAX activation and apoptosis [269]. Nonetheless, Docetaxel resistance in prostate cancer was shown to arise from TGF-β induction of Krüppel-like factor 5, a transcription factor, acetylation which in turn leads to BCL-2 upregulation. Moreover, TGF-β also prevents BCL-2 ubiquitination induced by Docetaxel [270].
It was also reported that phosphorylation of Thr69, Ser70, and Ser87 of BCL-2 occurs after Paclitaxel treatment. However, contrarily to the effects observed for other drugs, it is suggested that at least phosphorylation of both Ser70 and Ser87 is essential for Paclitaxel to fully induce cell death [147]. Furthermore, in breast and ovarian cancer cells, overexpression of miR-203b-3p and miR-203a-3p increased sensitivity to Paclitaxel by downregulating BCL-xL. Interestingly, c-Myc was found to promote the transcription of miR-203b-3p and miR-203a-3p in breast cancer cells [75]. Moreover, antimicrotubule agents cause cell death by inducing prolonged mitotic arrest. Cell fate of mitotic arrested cells is defined by the duration of the arrest, cyclin B1 degradation, and apoptotic signaling. For instance, high expression of BCL-xL allows for cyclin B1 degradation to reach it threshold, and the cell exits mitosis, while low expression of BCL-xL will lead to the apoptotic signaling threshold to be reached and cell death induced. An intermediate level of BCL-xL might let the cell exit mitosis but die after. Thus, BCL-xL expression can influence the sensitivity to drugs that act by causing mitotic arrest [271,272].
In cancer cells treated with Vinblastine, PGAM5 dephosphorylates BCL-xL at Ser62, increasing BCL-xL’s affinity to BAX and BAK to prevent cell death [149].
In CRC, PAR2/BCL-xL axis is involved in epidermal growth factor receptor (EGFR) targeting resistance [150]. In addition, in EGFR-mutant lung cancer, deficiency of RBM10 was found to reduce sensitivity to EGFR targeting by reducing the BCL-xS/BCL-xL ratio [273]. Similarly, Gefitinib-resistant lung cancer cells were shown to inhibit autophagy through SRSF1 activity, which promotes production of the BCL-xL isoform. BCL-xL binds to Beclin-1 and prevents autophagy. On the other hand, under starvation conditions, SRSF1 is repressed increasing the BCL-xS/BCL-xL ratio. Beclin-1 is then free to interact with PIK3C3 and induce autophagy [274].
Furthermore, abnormal splicing of BCL-X was found to confer resistance to Imatinib in CML cells by decreasing BCL-xS/BCL-xL ratio. Restoring BCL-X splicing sensitized CML cells to Imatinib both in vitro and in vivo [275]. CML cells resistant to Imatinib were also found to overexpress methyltransferase-like 14 (METTL14). METTL14 increases the m6A level at the A1001 site of the BCL-X mRNA which is then recognized by hnRNP C to induce BCL-xL expression, promoting CML progression and Imatinib resistance. Additionally, METTL4 overexpression also leads to upregulation of BCL-2 and downregulation of BAX and caspase-3 [276].
Cancer cells’ sensitivity to the splicing modulator E7107, which targets SF3b, decreased in the presence of BCL-xL. Nonetheless, no effect was observed for splicing modulator targeting SRPK or RBM39/DCAF15 potentially due to the fact that these proteins do not play a role in the splicing regulation of BCL-X mRNA [277].
In a multiple myeloma cell line with antisense p53, increased expression of BCL-2 was correlated with Dexamethasone resistance [278]. Furthermore, the expression of BCL-2 in diffuse large B-cell lymphoma (DLBCL) patients treated with Cyclophosphamide, Doxorubicin, Vincristine, and Prednisone is associated with worse prognosis. However, addition of Rituximab to this treatment approach overcame the association of BCL-2 expression to worse prognosis [279]. In multiple myeloma, BCL-xL expression seems to be associated with resistance to treatment with Melphalan and Prednisone or Vincristine, Adriamycin, and Dexamethasone [280].
In nonmuscle invasive bladder cancer, it was shown that protein S100A16 promotes resistance to Mitomycin C by increasing AKT/BCL-2 pathway signaling [281].
In AML, induction of resistance to Cytarabine is suggested to be promoted by CXCR4-facilitated repression of Let-7a expression, which induces Yin Yang 1-mediated transcriptional promotion of MYC and BCL-xL [282]. Paradoxically, a different study showed that activation of CXCR4 led to downregulation of BCL-xL and upregulation of NOXA and BAK [283].
GEX1A is a splicing modulator that in leukemic cells led to cell death by shifting MCL-1 splicing towards the pro-apoptotic isoform MCL-1S. Nonetheless, cells with high levels of BCL-xL were less responsive to GEX1A treatment [284]. Moreover, in HCC cell lines, increased expression of Let-7c, a miRNA targeting BCL-xL, potentiated the apoptotic effect of Sorafenib, a MCL-1 inhibitor [76].
In AML, administration of Flavopiridol, a CDK inhibitor, increases BCL-2 expression, and it is suggested that BCL-2 inhibition could enhance Flavopiridol efficacy [285].
High expression of BCL-xL in glioma stem cells has also been associated with increased resistance to treatment with Volasertib, a PLK1 inhibitor [286].
In TNBC, inhibition of BCL-xL synergized with CDK1/2/4 inhibitors, but not with inhibitors of the transcription factor Forkhead box M1, CDK4/6, Aurora A, and Aurora B [287]. Nonetheless, in small cell lung cancer (SCLC), resistance to the Aurora B inhibitor AZD2811 was overcome by the inhibition of BCL-2 [288]. Moreover, BCL-2 and BCL-xL were found to prevent death of MYC overexpressing cells prompted by the pan-Aurora inhibitor VX-680, but not polyploidy induction. This is achieved by the interaction of BCL-2 and BCL-xL with Beclin-1 and autophagy-related gene 6 blocking autophagy induction [289].
High expression of BCL-xL has also been associated with resistance to v-Raf murine sarcoma viral oncogene homolog B (BRAF) inhibitors. BRAF is involved in the regulation of cell growth, survival, and differentiation, and its inhibition in metastatic melanoma leads to increased expression of several BCL-2 family proteins such as BCL-xL and BCL-w. Furthermore, high BCL-2 expression before treatment was inversely correlated with response to BRAF inhibition [290].
In mantle cell lymphoma, depletion of BAX and overexpression of BCL-xL both alone and in conjugation render cells resistant to treatment with the proteasome inhibitor Bortezomib [291].
Argininosuccinate synthetase 1 silencing is common in several types of cancer, resulting in dependency on extracellular arginine. In this sense, arginine deprivation therapies have been explored but have shown disappointing anticancer effects. This can be explained by the fact that BCL-xL prevents apoptosis induced by this type of therapies [292].
In myeloma, high expression of BCL-2 is associated with interferon therapy resistance [293]. Moreover, lower expression of BCL-2 predicts worse prognosis in adjuvant endocrine therapy-treated ER-positive breast cancer patients [294].
ER stress inducers, in HCC, led to the upregulation of GOLGA2P10, a lncRNA, and consequently to increased BCL-xL expression and BAX phosphorylation preventing cell death [295].
Furthermore, in PDAC, collagen XI/αI is associated with resistance to treatment by inducing the AKT/CREB/BCL-2 pathway, which leads to increased expression of BCL-2 and decreased activity of BAX, thus inhibiting apoptosis [296].
The RNA helicase DHX33 in association with activating protein 2β was shown to promote cancer cell survival by increasing BCL-2 mRNA expression [297].
In CRC, RASSF4 expression is usually downregulated. RASSF4 regulates BCL-2 expression through YAP and its lower expression in CRC induces cell proliferation and drug resistance [298].
CML cells expressing BCR-ABL were shown to repress apoptosis induced by chemotherapy through the promotion of STAT5 activity that leads to increased expression of BCL-xL [299,300].
Moreover, in melanoma, expression of BCL-2 was associated with resistance to biochemotherapy by repressing apoptosis induction [301].
In patients with operable carcinoma of the breast, low BCL-2 expression is predictive of pathological complete response after preoperative chemotherapy [302].
However, BCL-2 expression can also confer sensitivity to some drugs. For instance, high expression of BCL-2 in non-germinal-center B-cell-like DLBCL was also associated with better response to Zanubrutinib, a Bruton tyrosine kinase inhibitor [303]. Moreover, postmastectomy radiotherapy in breast cancer patients led to a better outcome in high BCL-2 expression patients [304]. Similarly, in HNSCC, BCL-2-positive patients also had better response to radiotherapy than BCL-2 negative patients [305].
However, BCL-2 positive prostate cancer patients showed increased failure to radiotherapy treatment. Additionally, the group who showed worse failure to treatment was the BCL-2 positive group with abnormal expression of BAX [306]. In human glioma, miR-153-3p was found to increase radiosensitivity by targeting BCL-2 expression [307]. Treatment with the BCL-xL inhibitor A1331852 also radiosensitized mesothelioma cells [308]. Accordingly, in laryngeal cancer, expression of both BCL-2 and BCL-xL was associated with radioresistance [309]. In CRC overexpression of tumor necrosis factor receptor-associated factor 4, a E3 ligase, led to radioresistance through the activation of JNK/c-Jun and consequent increase in BCL-xL expression [310]. Moreover, in malignant glioma and pancreatic cells, BCL-xL expression was also associated with radioresistance [311,312]. In prostate cancer patients, BCL-2 was found to be overexpressed in patients who failed brachytherapy compared to patients who responded to treatment [313].
Prostate cancer patients BCL-2 negative and with normal BAX showed high response to treatment with androgen deprivation combined with radiotherapy. The predictive value of negative BCL-2 and normal BAX was more pronounced for short-term androgen deprivation than for long-term [314]. On the other hand, BCL-2-positive prostate cancer patients treated with neoadjuvant androgen deprivation and radical radiotherapy showed better prognosis than BCL-2-negative ones [315].
Furthermore, prostate cancer cells overexpressing BCL-2 are susceptible to Poly (ADP-Ribose) polymerase (PARP) inhibition combined with radiotherapy. Overexpression of BCL-2 blocks Ku80 from entering the nucleus. Ku80 is essential for the non-homologous end joining DNA repair pathway. Thus, BCL-2 overexpressing cells depend on the alternative PARP1-dependent end-joining pathway to repair DNA double-strand breaks, making them susceptible to PARP inhibition [316].
Another common mechanism that allows cancer cells to survive therapy is through the induction of senescence. For instance, in melanoma cells, a combination of an inhibitor of Aurora A, which promotes senescence, and a BCL-xL inhibitor led to enhanced treatment efficacy. BCL-xL represses BAX, maintaining p53 activation which in turn promotes p21-mediated senescence. When BCL-xL is inhibited p21 is degraded by caspases and the apoptotic pathway is induced [317]. Accordingly, BCL-xL has been described to promote senescence in several types of cancer [56].
Treatment with Palbociclib or bromodomain and extra-terminal (BET) protein inhibitors can induce senescence, leading to resistance to these drugs. For instance, Palbociclib-induced melanoma senescent cells showed a reduction in HRK and BIM expression and increased BCL-xL:BAK affinity, preventing apoptosis [318]. Similarly, TNBC BET inhibitors-induced senescent cells present higher levels of BCL-xL, and its inhibition leads to increased sensitivity to BET inhibitors [319].
Moreover, in PDAC, BCL-xL was found to protect cells exposed to a microenvironment scarce in oxygen and nutrients by repressing cell cycle progression [320]. An acidic microenvironment also leads to the upregulation of BCL-2 and BCL-xL by MEK/ERK activity promoted by G protein-coupled receptor 65 which acts as an acid sensor [321]. Overexpression of BCL-2 and BCL-xL is largely linked to increased resistance to chemotherapy and radiotherapy. Therefore, combining standard treatments with BCL-2 and BCL-xL inhibitors may help overcome this resistance and improve drug efficacy and patient outcomes.

5. BCL-2 and BCL-xL Expression in Cancer Across TCGA and CPTAC Datasets Using UALCAN Analysis
Given the central roles of BCL-2 and BCL-xL in cancer-related processes and drug resistance, we used the UALCAN webtool to assess BCL-2 and BCL-xL mRNA and protein expression levels across multiple cancer types compared with normal tissues (Table 3 and Table 4). At the mRNA level, BCL-2 upregulation was observed in all kidney cancer types analyzed, as well as in cholangiocarcinoma and HCC, but no upregulation was observed at protein level. Conversely, BCL-2 mRNA was downregulated in bladder urothelial carcinoma, breast invasive carcinoma, cervical SCC, colon adenocarcinoma, uterine corpus endometrial carcinoma, lung SCC, prostate adenocarcinoma, rectum adenocarcinoma, stomach adenocarcinoma, and thyroid carcinoma, while at the protein level it was downregulated in breast cancer and HCC.
On the other hand, BCL-xL was upregulated in bladder urothelial carcinoma, breast invasive carcinoma, cervical SCC, cholangiocarcinoma, colon adenocarcinoma, uterine corpus endometrial carcinoma, esophageal carcinoma, HNSCC, kidney chromophobe, renal papillary cell carcinoma, HCC, prostate adenocarcinoma, rectum adenocarcinoma, stomach adenocarcinoma, and thyroid carcinoma at the mRNA level and in ovarian and colon cancers, uterine corpus endometrial adenocarcinoma, lung adenocarcinoma, and pancreatic adenocarcinoma at the protein level. Conversely, BCL-xL was downregulated at the mRNA level in lung SCC, and at the protein level in clear cell renal cell carcinoma, lung SCC, and HNSCC.
Nonetheless, it is important to note that the UALCAN webtool does not distinguish between BCL-xL and BCL-xS isoforms, as the search is based on the gene (BCL2L1) and not the protein.

6. Co-Targeting of BCL-2 and BCL-xL in Clinical Trials
Due to their roles in cancer progression and also in resistance to chemo- and radiotherapy, several dual BCL-2/BCL-xL inhibitors have been designed and assessed in clinical trials across a wide spectrum of malignancies, including HCC, lung cancer, melanoma, and ovarian cancer (Table 5) [322].
Navitoclax (ABT-263), a BH3-mimetic that targets BCL-2, BCL-xL, and BCL-w, has been the most extensively investigated agent of this class. Early phase I trials in hematologic and solid tumors demonstrated antitumor activity with manageable toxicity profiles [323,324,325].
In SCLC and other solid tumors, Navitoclax induced partial responses or stable disease in a minority of patients, though severe adverse events (AEs) such as fatal respiratory failure, left ventricular systolic dysfunction, and asymptomatic lipase elevation were occasionally reported [308]. In relapsed or refractory CLL, partial responses were observed, but hematologic toxicities, especially thrombocytopenia and neutropenia, were frequent, establishing a maximum tolerated dose (MTD) of 200 mg/day with intermittent dosing [309].
A phase 2 study in patients with relapsed/refractory lymphoid malignancies treated with Navitoclax corroborated these findings even though clinical activity was only observed in a minority of these patients [326]. Nonetheless, two phase 2 trials showed that Navitoclax alone had limited activity against advanced and recurrent SCLC and ovarian cancer [327,328].
Combinatorial approaches with Navitoclax have also been investigated, and in patients with solid tumors the combination of Gemcitabine and Navitoclax was deemed safe but produced mostly stable disease, while the combination of Navitoclax and Docetaxel was also tolerable and showed clinical activity with partial responses in a subset of patients [329,330].
Navitoclax plus Trametinib, a MEK inhibitor, was also deemed safe and led to durable responses, and a recommended phase II dose was established [331].
The phase II recommended dose could not be achieved with the combination of Navitoclax and Erlotinib while a study investigating Navitoclax with Cisplatin and Paclitaxel was discontinued due to high toxicity and low clinical activity [332,333]. Nonetheless, Navitoclax with Paclitaxel showed moderate activity. The addition of irinotecan to Navitoclax resulted in partial responses. However, grade ≥ 3 AEs occurred in 77.4% of patients [334].
In NSCLC, Navitoclax combined with Osimertinib, an EGFR inhibitor, was also tolerable with clinical efficacy [335]. Early studies pairing Navitoclax with Vistusertib, an mTOR inhibitor, or Sorafenib, a multi-kinase inhibitor, also indicated manageable safety but minimal objective responses [336,337].
In a phase 2 study with patients with myelofibrosis, Navitoclax combined to Ruxolitinib, an inhibitor of Janus kinase 1 and 2, led to significantly improved spleen volume and symptom burden even though the median overall survival (OS) was not met [338]. A phase I trial in relapsed/refractory CD20+ lymphoid malignancies exploring the combination of Navitoclax with Rituximab, a CD20 inhibitor, demonstrated multiple complete and partial responses, even though grade 4 thrombocytopenia occurred in 17% of patients [339]. In B-cell CLL, the same combination was well tolerated and led to higher response rates and prolonged progression-free survival [340]. Navitoclax with Venetoclax and chemotherapy in relapsed/refractory acute lymphoblastic leukemia and lymphoblastic lymphoma achieved objective responses but were limited by grade 3/4 myelosuppression [341].
Additional trials in different cancer types evaluating Navitoclax with other agents, including Ketoconazole (NCT01021358), Etoposide plus Cisplatin (NCT00878449), Olaparib (NCT05358639), Rifampin (NCT01121133), Venetoclax plus cladribine-based salvage therapy (NCT06007911), Venetoclax with Decitabine (NCT05455294, NCT05222984, NCT05740449), Venetoclax (NCT05215405, NCT05192889, NCT05054465), Ruxolitinib with or without Mivebresib (NCT04041050, NCT04480086), Ruxolitinib plus ABBV-744 (NCT04454658), Ruxolitinib (NCT04472598, NCT04468984), Fludarabine plus Cyclophosphamide and Rituximab or Bendamustine plus Rituximab (NCT00868413), Dabrafenib plus Trametinib (NCT01989585), Venetoclax plus Ibrutinib and Rituximab (NCT05864742), Abiraterone acetate with or without Hydroxychloroquine (NCT01828476), and Bendamustine plus Rituximab (NCT01423539), were terminated with no results published, ongoing or were withdrawn.
Obatoclax mesylate, also known as GX15-070, is an inhibitor of BCL-2, BCL-xL, BCL-w, and MCL-1. This drug has also been tested in multiple clinical contexts. Phase I studies in hematologic malignancies and solid tumors demonstrated limited efficacy but acceptable safety, characterized mainly by neurological and psychiatric AEs, including somnolence and dizziness [342,343,344]. In AML, a few patients achieved stable disease while in myelodysplastic syndromes, no objective response was observed [345,346]. In myelofibrosis, hematologic improvement was observed in only one patient, whereas in Hodgkin’s lymphoma, the insufficient clinical responses led to a decision against further enrollment [347,348]. Other trials exploring Obatoclax mesylate in hematologic malignancies (NCT00438178) and systemic mastocytosis (NCT00918931) were completed with no published results.
Combination strategies with Obatoclax improved outcomes in certain settings. When added to Carboplatin/Etoposide in extensive-stage SCLC, it increased objective response rates to 62% vs. 53% for chemotherapy alone, without introducing unexpected toxicity [349,350]. In CLL, combination with Fludarabine and Rituximab produced complete and partial responses; however, neuropsychiatric effects were frequent [351]. In solid tumors and relapsed SCLC, the association with Topotecan achieved stable disease or partial responses, with hematologic and neurologic toxicity as primary limitations [352,353]. The combination with Bortezomib in mantle cell lymphoma yielded complete or partial responses in some patients, while association with Docetaxel in NSCLC achieved partial and stable responses, with frequent grade 3/4 neutropenia [354,355]. Other trials combining Obatoclax with Bortezomib (NCT00538187, NCT00719901), Vincristine/Doxorubicin/Dexrazoxane (NCT00933985), Rituximab with or without Bendamustine (NCT01238146, NCT00427856), and Carboplatin plus Etoposide (NCT01563601) were withdrawn, or completed without published results.
AT-101, also known as oral gossypol, is an orally bioavailable pan-BCL-2 inhibitor with activity against BCL-2 and BCL-xL that has also undergone extensive evaluation. Administration of AT-101 to refractory metastatic breast cancer patients led to a minor response and two patients achieving stable disease. Two out of the three patients receiving 50 mg/day had grade III dermatologic toxicity which was dose limiting. The MTD for this drug was 40 mg/day [356].
In castration-resistant prostate cancer, AT-101 demonstrated limited activity accompanied by gastrointestinal AEs, while in castration-sensitive metastatic prostate cancer, nearly one-third of patients achieved undetectable PSA levels [357,358]. Studies in glioblastoma multiforme yielded mostly stable disease, and activity was minimal in SCLC and adrenocortical carcinoma [359,360,361]. Trials in B-cell non-Hodgkin’s lymphoma (NCT05338931) remain ongoing, whereas one in relapsed or refractory B-cell malignancies (NCT00275431) was completed with no published results.
Combination regimens have shown comparatively better outcomes. AT-101 with Paclitaxel/Carboplatin produced objective responses in advanced solid tumors, and with Cisplatin/Etoposide achieved partial responses in approximately 15% of patients [362,363]. In gastroesophageal carcinoma, AT-101 combined with Docetaxel, 5-FU, and radiotherapy yielded high complete response rates [364]. In relapsed/refractory SCLC, combination with Topotecan induced partial and stable responses despite hematologic toxicity [365]. In head and neck cancer, AT-101 plus Docetaxel produced partial and stable responses, with lymphopenia as the main grade 3/4 AE [366]. In NSCLC, combinations with Docetaxel alone or Docetaxel plus Cisplatin resulted in disease stabilization in most patients, with neutropenia and anemia as the most common grade ≥ 3 AEs [367,368]. In laryngeal cancer, AT-101 with Docetaxel and platinum agents achieved partial responses in over half of treated patients [369]. The administration of AT-101, Docetaxel, and Prednisone in metastatic castration-resistant prostate cancer resulted in a median OS of 18.1 months; however, significant myelosuppression was observed [370]. A trial combining AT-101 with Erlotinib in NSCLC harboring EGFR mutations (NCT00988169) showed limited activity. Additionally, studies testing AT-101 with Temozolomide with or without radiotherapy (NCT00390403), Erlotinib (NCT00934076), Lenalidomide (NCT01003769 [296]), and Rituximab (NCT00286780, NCT00440388) were withdrawn or terminated with no published results.
Pelcitoclax, also known as APG-1252, is a dual BCL-2 and BCL-xL inhibitor that is being explored in several clinical trials. The results of the first clinical trial exploring Pelcitoclax in locally advanced or metastatic solid tumors led to three partial responses, one in SCLC, one in ovarian cancer, and another in a patient with neuroendocrine prostate cancer. Moreover, 11 patients achieved stable disease, leading to an overall disease control rate (DCR) of 30.4%. The recommended dose for further trials was a weekly dose of 240 mg of Pelcitoclax which in this trial led to a DCR of 50%. The regime was also found to be tolerable with the most common AEs being transaminase elevations and thrombocytopenia [371]. There are currently three active trials investigating Pelcitoclax: its combination with Cobimetinib in recurrent ovarian and endometrial cancers (NCT05691504), with Osimertinib in EGFR-TKI–resistant NSCLC (NCT04001777), and its use alone or with Chidamide in relapsed or refractory non-Hodgkin lymphoma (NCT05186012). In contrast, three monotherapy trials in SCLC or advanced solid tumors (NCT03387332), advanced neuroendocrine tumors (NCT04893759), and myelofibrosis after prior therapy (NCT04354727), as well as one evaluating Pelcitoclax plus Paclitaxel in relapsed/refractory SCLC (NCT04210037), were terminated or withdrawn without any published results.
More recently, AZD0466, a novel dual BCL-2/BCL-xL inhibitor which consists of a drug conjugate of AZD4320 and a DEP® G5 poly-L-lysine dendrimer, entered early-phase clinical development. Trials in advanced hematologic or solid tumors (NCT04214093) were initiated, as well as studies evaluating AZD0466 combined with Voriconazole (NCT04865419) and with other anticancer agents in non-Hodgkin lymphoma (NCT05205161). One clinical trial was terminated with no published results, and the other two were terminated based on benefit–risk profile assessment.
Besides the limited efficacy observed with dual BCL-2 and BCL-xL inhibition as monotherapy, toxicity and the possibility of cancer cells to develop resistance are other major concerns for this type of inhibitors. Navitoclax, for example, is associated with neutropenia and dose-limiting thrombocytopenia since neutrophil progenitors and platelets are dependent on BCL-2 and BCL-xL, respectively, for their survival [372,373]. To mitigate thrombocytopenia, AZD4320 was developed; however, it exhibited dose-limiting cardiovascular toxicity in preclinical studies [372,374]. The drug-dendrimer conjugate AZD0466 was subsequently designed to overcome this limitation and showed comparable efficacy to AZD4320 with reduced toxicity, but clinical trials evaluating AZD0466 were terminated without published results [374].
Moreover, cancer cells can acquire resistance to BCL-2 and BCL-xL inhibitors through compensatory survival pathways. For instance, increased expression of MCL-1 and BFL-1 was found to be associated with resistance to ABT-737, a dual BCL-2/BCL-xL inhibitor [375]. In lymphoid and leukemic cells, Navitoclax resistance arises from high MCL-1 expression, while cells with high BCL-2 and NOXA mRNA levels show higher sensitivity to this drug [376].
Furthermore, BIM has been shown to interact with BCL-2 and BCL-xL through both its BH3 domain and C-terminal region. When both regions are connected to BCL-2 or BCL-xL, a mechanism known as double-bolt locking, displacement of these proteins is hindered, leading to resistance to Navitoclax and potentially to other BH3 mimetics [377].
Despite these limitations, clinical findings collectively indicate that these agents can enhance apoptotic signaling and overcome tumor cells’ resistance to several drugs [227,233,234,235,378]. Thus, not only could the targeting of BCL-2 pro-survival family members potentially promote cancer cells death through increased apoptotic signaling, but it also could potentially overcome resistance and increase the therapeutic potential of other drugs. Nevertheless, continued development of novel BCL-2 and BCL-xL–targeting strategies is essential to try and minimize toxicity, circumvent resistance mechanisms, and fully explore their therapeutic potential, particularly in combination regimens.

7. Conclusions
BCL-2 and BCL-xL play a central role in apoptosis regulation, but their functions extend far beyond preventing cell death. They contribute to tumor progression, invasiveness, angiogenesis, chemotherapy resistance, and cellular metabolism, highlighting their multifaceted role in cancer biology. Although selective inhibitors, such as BH3 mimetics, show promise, challenges remain regarding efficacy and toxicity. Future research should continue to focus on developing combination therapies that target BCL-2 and BCL-xL alongside other oncogenic pathways, exploring context-specific inhibitors that minimize off-target effects, and identifying biomarkers to predict patient response. Additionally, deeper investigation into their non-apoptotic roles, including regulation of metabolism and mitochondrial dynamics, could uncover novel therapeutic opportunities and improve precision oncology approaches.