The canonical PI3K/AKT/mTOR signaling pathway serves as a central hub in cellular signal transduction, governing diverse biological processes such as metabolism, proliferation, and angiogenesis (29–31), with pivotal roles in malignant transformation and progression (32). The elucidation of this pathway began in the mid-1980s, marked by seminal discoveries over nearly 5 decades. A breakthrough emerged from Lewis Cantley’s team during investigations into viral oncogenesis, where they identified a phosphatidylinositol kinase (PI3K) activity linked to cellular transformation, laying the foundation for subsequent signal transduction research. In 1988, Cantley et al. (33) published a landmark study in Nature detailing the catalytic mechanism of class I PI3K: this heterodimeric enzyme, comprising a regulatory subunit (p85α) and a catalytic subunit (p110α), phosphorylates phosphatidylinositol-4,5-bisphosphate (PIP₂) to generate phosphatidylinositol-3,4,5-trisphosphate (PIP₃) at the plasma membrane (34–36). PIP₃ acts as a secondary messenger, recruiting phosphoinositide-dependent kinase 1 (PDK1), AKT, and serum/glucocorticoid-regulated kinase (SGK) to the membrane, thereby initiating downstream signaling cascades (37–39).

During the 1990s, subsequent studies unraveled key downstream regulatory mechanisms. AKT, a central effector kinase, requires dual phosphorylation at Thr308 (by PDK1) and Ser473 (by mTORC2) for full activation (40–43). Activated AKT phosphorylates downstream targets such as tuberous sclerosis complex 2 (TSC2) and forkhead box O (FOXO) proteins, directly or indirectly modulating cell growth and survival (44, 45). mTOR, a highly conserved serine/threonine kinase and critical AKT substrate, orchestrates cellular processes via two distinct complexes: mTORC1 and mTORC2 (46, 47). mTORC1 activation is regulated by TSC phosphorylation: AKT inhibits the TSC1-TSC2 complex (composed of hamartin [TSC1] and tuberin [TSC2]) via TSC2 phosphorylation, relieving its GTPase-activating protein (GAP) suppression on Ras homolog enriched in brain (Rheb), thereby activating mTORC1 (48, 49). Active mTORC1 drives oncogenic progression by phosphorylating ribosomal S6 kinase 1 (S6K1/p70S6K) and eukaryotic translation initiation factor 4E-binding protein 1 (4EBP1), enhancing protein synthesis, ribosome biogenesis, and cell cycle progression (50–52). Additionally, mTORC1 suppresses autophagy, promotes lipid synthesis, and upregulates hypoxia-inducible factor 1α (HIF-1α) to stimulate angiogenesis (53).

Further mechanistic insights revealed PTEN (phosphatase and tensin homolog), a lipid phosphatase, as a critical tumor suppressor that counteracts PI3K signaling by dephosphorylating PIP₃ to PIP₂ (54). By the early 2000s, mTOR’s dual roles through mTORC1 and mTORC2 in regulating protein synthesis, cell growth, and survival were firmly established (55). The advent of CRISPR-Cas9 gene editing (widely adopted post-2013) enabled functional validation of PI3Kα (encoded by PIK3CA) as a driver in oncogenesis, solidifying its status as a therapeutic target (56). Landmark studies published in Nature (circa 2015) resolved the full atomic structure of PI3K, enabling structure-guided design of allosteric inhibitors like alpelisib and accelerating their clinical translation (56). Collectively, these discoveries have constructed a robust framework bridging molecular insights to therapeutic innovation, positioning the PI3K/AKT/mTOR pathway as a cornerstone for targeted therapy in malignancies, including osteosarcoma (Figure 2).

Aberrant activation of the PI3K/AKT/mTOR signaling pathway is a central driver of malignant progression in osteosarcoma (OS). Clinical evidence reveals characteristic expression patterns of key pathway components in OS patients. A study analyzing 130 paired tumor/normal bone tissue samples demonstrated significant downregulation of miR-99a in OS tissues, which inversely correlated with mTOR mRNA overexpression. Low miR-99a or high mTOR expression was associated with advanced Enneking stage (III), increased metastasis, postoperative recurrence, and chemoresistance, underscoring the miR-99a/mTOR axis as a critical regulator of OS aggressiveness (57). Similarly, STEAP2 overexpression in OS cell lines and clinical specimens has been linked to malignant phenotypes and poor prognosis. Mechanistically, STEAP2 activates PI3K/AKT/mTOR signaling to induce epithelial-mesenchymal transition (EMT), thereby enhancing OS cell invasion and migration (58). Immunohistochemical quantification further confirmed elevated phosphorylation levels of downstream mTOR effectors (p-mTOR, p-4E-BP1, and p-S6K1) in OS tissues compared to normal bone (p < 0.001), corroborating sustained pathway activation at the protein level (59).

Molecular investigations highlight multifaceted contributions of PI3K/AKT/mTOR components to OS pathogenesis. Activation of mTOR effectors (e.g., p70S6K and 4EBP1) directly promotes tumor proliferation and survival (57), while additional regulators amplify oncogenic signaling.

• KIF21B upregulation modulates OS cell proliferation and apoptosis via PI3K/AKT signaling (60);

Genomic analyses reveal PIK3CA mutations in ∼80% of OS cases, predominantly clustering in kinase (exon 20) and helical (exon 9) domains. These mutations enhance catalytic activity of the class I PI3K p110α subunit, activating AKT via lipid and protein kinase functions (63, 64). Hyperactivated AKT phosphorylates multiple substrates to regulate proliferation, differentiation, and apoptosis, with its overexpression often correlating with PI3K/HER4 co-amplification—a hallmark of OS metastasis and poor prognosis. Conversely, PTEN loss or downregulation removes inhibitory control over PI3K/AKT signaling, exacerbating malignant phenotypes (65, 66). Additionally, AKT-mediated VEGF-A activation fosters tumor angiogenesis, providing microenvironmental support for OS invasion and metastasis (67).

Targeting this pathway has yielded promising preclinical and clinical advances. PI3K inhibition (e.g., LY294002) reduces p-AKT levels and induces OS apoptosis (68), while modulating non-coding RNAs (e.g., lncRNA MSC-AS1 silencing or miR-191-5p activation) restores chemosensitivity (69). As the pathway’s central integrator, mTOR activation drives cell cycle progression (e.g., G1/S transition) and upregulates pro-proliferative factors (cyclin D1/B1), making it a prime therapeutic target. Preclinical studies demonstrate that rapamycin analogs (e.g., temsirolimus) suppress OS xenograft growth and synergize with cisplatin. Dual-target inhibitors (e.g., NVP-BEZ235 [PI3K/mTOR] and CC-115 [mTOR/DNA-PK]) show superior efficacy in overcoming resistance. Translational evidence supports clinical potential.

• Ridaforolimus monotherapy prolongs PFS in metastatic sarcoma;

Mechanistically, everolimus binds FKBP12 to selectively inhibit mTORC1, reducing phosphorylation of p70S6K and 4E-BP1. This blockade suppresses cell cycle progression, protein synthesis, and VEGF-A-mediated angiogenesis, providing a molecular rationale for precision therapy in OS (20–22, 70). In summary, PI3K/AKT/mTOR dysregulation in OS constitutes a systemic network driving proliferation, invasion, metastasis, and chemoresistance. Current evidence solidifies mTOR-targeted therapy as a cornerstone of OS precision medicine. Future advances in multi-target inhibitors, combination regimens, and biomarker-driven strategies hold promise for developing low-toxicity, high-efficacy therapies for OS patients.

Preclinical evidence shows that EVR inhibits proliferation in diverse human cancer cell lines (e.g., breast, gastric, renal) at subnanomolar concentrations, mainly via multidimensional modulation of the PI3K/AKT/mTOR network. By binding FKBP-12 and selectively inhibiting mTORC1, EVR suppresses S6K1 phosphorylation and releases 4E-BP1–mediated inhibition of oncogenic mRNA translation (Cyclin D1, c-MYC, HIF-1α (71)), leading to G1-phase arrest, activation of mitochondrial apoptosis pathways, and autophagy induction. In parallel, EVR reprograms cancer cell metabolism by downregulating HK2 and LDHA through mTORC1–HIF-1α suppression, thereby inhibiting glycolysis, and blocking SREBP-driven lipid synthesis, which disrupts membrane signaling platforms.

Although EVR is widely known for its immunosuppressive role in transplantation via mTORC1 inhibition in T lymphocytes, emerging evidence suggests it may also modulate the immune microenvironment in OS. Preclinical studies indicate EVR reduces M2 polarization of tumor-associated macrophages, promoting a shift toward the M1 phenotype (72). It enhances dendritic cell antigen presentation (upregulating CD80/CD86), limits expansion of regulatory T cells, and downregulates immunosuppressive cytokines such as IL-10 and TGF-β (EVR has been reported to inhibit IL-10 expression in immune cells (73)). While combination strategies with immune checkpoint inhibitors have shown feasibility in other cancers, OS-specific studies remain limited and often lack immune cell profiling, leaving this a promising but underexplored direction.

In addition, EVR exerts dual anti-angiogenic effects—directly inhibiting endothelial cell proliferation and indirectly suppressing HIF-1α–driven VEGF expression—while prolonged exposure may activate compensatory PI3K/AKT signaling via IRS-1 stabilization. Such resistance mechanisms can be mitigated through combination regimens, for example, with PI3K or CDK4/6 inhibitors, to enhance therapeutic durability.

Recent years have witnessed multidimensional breakthroughs in elucidating the mechanisms of everolimus (EVR) in combinatorial osteosarcoma (OS) therapy. Preclinical investigations have evolved from single-pathway inhibition to cross-network molecular regulation, accumulating robust laboratory evidence. In the realm of metabolic reprogramming and resistance reversal, Moriceau et al. (22) demonstrated that zoledronic acid (ZOL) synergizes with EVR via dual mechanisms: by suppressing the mevalonate pathway, ZOL reduces Ras protein isoprenylation, thereby inhibiting Ras/ERK pro-proliferative signaling while enhancing EVR’s suppression of PI3K/mTOR. In SaOS-2 and U2OS cell models, this combination increased proliferation inhibition by 30% and significantly reduced GTP-Ras activity. Notably, in C57BL/6 osteoblastic and NOD/SCID osteolytic murine models, the EVR + ZOL regimen achieved 60% and 50% tumor volume reduction, respectively, without added toxicity. This study pioneered the link between OS metabolic vulnerability (mevalonate pathway dependency) and targeted therapy resistance, establishing a mechanistic paradigm for subsequent combinatorial strategies.

Building on metabolic insights, Pignochino et al. (21) systematically dissected the synergy between sorafenib and EVR. Using seven OS cell lines (including MNNG-HOS and U-2OS) and NOD/SCID xenografts, they revealed that sorafenib monotherapy inhibits mTORC1 (reduced p-S6) but paradoxically activates mTORC2 (elevated p-mTOR Ser2481), explaining limited clinical responses. EVR combination therapy counteracted this resistance via ROS-mediated AMPK activation (suppressing mTORC1) and conformational interference (blocking mTORC2), achieving complete mTOR pathway suppression. In vivo, this dual inhibition reduced tumor volume by 70%, lung metastases by 50%, and microvessel density by 60%. Higuchi et al. (2019) (75) validated this in doxorubicin (DOX)-resistant patient-derived orthotopic xenograft (PDOX) models (nu/nu mice): EVR + sorafenib emerged as the sole regimen inducing significant tumor regression (p < 0.001 vs control and DOX groups), with histopathology revealing extensive necrosis and stromal remodeling. Sorafenib or EVR monotherapy showed moderate efficacy (p < 0.001 vs control), while DOX was ineffective (p = 0.2 vs control). Crucially, neither study observed cumulative toxicity, underscoring the regimen’s superior therapeutic window.

Advances in cell cycle modulation and apoptosis induction further highlight EVR’s combinatorial potential. Oshiro et al. (24) validated the synergy of palbociclib (CDK4/6 inhibitor) with EVR in DOX-resistant PDOX models. In completely DOX-refractory tumors, combination therapy significantly reduced tumor volume (p = 0.018 vs control and DOX groups), with histology showing coagulative necrosis. EVR monotherapy partially suppressed proliferation (p = 0.04 vs control), while palbociclib and DOX failed. Mechanistically, dual blockade of G1/S transition (via CDK4/6 inhibition) and mTOR-mediated metabolic support synergistically disrupted tumor proliferation. Similarly, Garofalo et al. (74) demonstrated that NT157 (IRS-1/2 inhibitor) degrades IRS-1 to block IGF-1R/AKT/mTOR signaling, synergizing with EVR (combination index <1). In MG-63 and U-2OS cells, NT157-induced G2/M arrest created a time-dependent therapeutic window for EVR’s cytostatic effects, offering a novel non-apoptotic strategy for sequential chemotherapy.

Dual inhibition of proteasome and mTOR pathways exhibits unique therapeutic advantages. Nakamura et al. (23) confirmed that EVR combined with bortezomib activates JNK/p38/ERK MAPK apoptotic pathways while suppressing AKT and c-MYC in HT1080 fibrosarcoma and LM8 OS models, yielding significant synergy in vitro (p < 0.05). In vivo, this combination reduced primary tumor volume (p < 0.05) and lung metastases (p < 0.01). Asanuma et al. (76) mechanistically validated this synergy: in MG-63 cells, bortezomib upregulated JNK/p38 phosphorylation to amplify stress signaling, while EVR inhibited mTORC1 survival pathways, elevating caspase-3/8/9 and PARP cleavage. In 143B xenografts, dose-dependent antitumor efficacy was achieved without overt toxicity, highlighting clinical translatability.

Tumor microenvironment (TME) remodeling has emerged as a novel therapeutic frontier. Sapio et al. (77) reported that AdipoRon (synthetic adiponectin receptor agonist) differentially regulates ERK/mTOR signaling in Saos-2 and U2OS cells: monotherapy activates pro-survival ERK1/2 while inhibiting mTORC1/p70S6K. EVR combination abolished ERK-driven survival, inducing a 2.3-fold increase in G0/G1 arrest in U2OS cells. Oshiro et al. (78) validated “vessel normalization” in lung-metastatic PDOX models: EVR + pazopanib reduced vasculature length by 58% (p < 0.05) and hypoxic regions via mTORC1-VEGF suppression and VEGFR blockade. This dual targeting of tumor-autonomous secretion and TME crosstalk provides a mechanistic foundation for controlling metastatic spread in advanced OS.

Building on preclinical validation of EVR’s multifaceted antitumor effects, such as metabolic modulation, cell cycle arrest, and microenvironment remodeling, its therapeutic potential has now advanced into clinical trials for osteosarcoma.

In adults, Grignani et al. (20) conducted a nonrandomized phase II trial evaluating sorafenib plus EVR as salvage therapy for unresectable or recurrent high-grade OS refractory to standard treatments. Among 38 patients, the regimen achieved a 6-month PFS rate of 45% (95% CI 28–61), narrowly missing the prespecified 50% target but demonstrating modest clinical activity. Dose adjustments or interruptions were required in 66% of cases due to grade 3–4 adverse events—most commonly lymphopenia (16%), hypophosphatemia (16%), and hand–foot syndrome (13%)—though no treatment-related deaths occurred. While mTOR pathway-mediated resistance limited efficacy, the findings underscored the regimen’s potential in refractory OS and highlighted the need for dosing refinement or incorporation of additional targeted agents.

In pediatric populations, an earlier pilot study (n = 14) (79) in heavily pretreated refractory OS showed predominantly stable disease, with median PFS and OS of 4.4 and 6 months, respectively, and a manageable toxicity profile characterized mainly by low-grade skin and hematologic events. Building on these findings, a more recent pediatric trial (n = 12) (80) reported partial responses in 33% of patients and 6-/12-month PFS rates of 54% and 36%, further supporting feasibility in younger patients and underscoring the importance of age-specific optimization. Beyond systemic therapy alone, Anderson et al. (81) pioneered a multimodal regimen integrating α-particle therapy (radium-223) with systemic treatments—including one case combining EVR and a PD-1 inhibitor—and radiotherapy, revealing synergistic potential for bone metastasis control via tumor microenvironment modulation and immune activation.

Precision oncology further illustrates EVR’s versatility. Yi et al. (82) described a case of glioblastoma with secondary OS in which exome sequencing-guided therapy—combining EVR, tumor-treating fields (TTFields), pembrolizumab, and temozolomide—achieved exceptional benefit through dual mechanisms: (1) targeting PTEN loss in glioblastoma to suppress mTORC1-driven proliferation and enhance temozolomide chemosensitivity; and (2) exploiting MSH3/ERCC4 mutation-associated DNA repair defects in secondary OS by disrupting mTOR signaling. The regimen synergistically remodeled the tumor microenvironment (e.g., reduced regulatory T-cell infiltration) and amplified antitumor efficacy through complementary actions, although EVR’s precise contribution and dose-response effects warrant further investigation.