Osteosarcoma (OS) in pediatric, adolescent, and young adult (AYA) patients is an aggressive bone cancer with limited treatment options. Dysregulation of the CDK4/6-cyclin D axis and the PI3K/mTOR pathway contributes to OS pathogenesis, providing a biological rationale for co-targeting these signaling nodes. However, pharmacologic CDK4/6 inhibition can trigger compensatory activation of the PI3K/mTOR pathway, restoring D-type cyclin expression and partially reactivating CDK4/6 signaling. Thus, dual inhibition of the CDK4/6 and PI3K/mTOR pathways not only addresses two parallel oncogenic drivers but may also prevent potential CDK4/6 inhibitor resistance mediated by feedback activation of PI3K/mTOR. In this study, we tested the hypothesis that coordinated targeting of these pathways would improve tumor control in preclinical OS models. In vitro sensitivity analyses using palbociclib and voxtalisib demonstrated additive to synergistic OS growth suppression, with palbociclib inducing G1 arrest and senescence, and the combination enhancing autophagy. Furthermore, the efficacy, tolerability, and mechanisms of palbociclib and voxtalisib, alone or in combination, were evaluated in molecularly defined primary treatment-naïve, and relapsed/metastatic OS models. In the relapsed/metastatic PDX77-TT2 model, short-term palbociclib exposure activated PI3K/mTOR signaling, whereas the combination of palbociclib and voxtalisib in long-term studies produced marked tumor suppression and extended survival. In the primary treatment-naïve PDX96 model, long-term palbociclib exposure generated a robust CDK4/6 pharmacodynamic response. The addition of voxtalisib reinforced autophagy, sustained CDK pathway inhibition, and improved overall tumor control. In an OS lung-colonization model, CDK4/6 inhibition alone markedly reduced OS lung nodules, with combination therapy providing comparable suppression. Dual CDK4/6-PI3K/mTOR inhibition achieves tumor control across various OS models, supporting the use of genomically guided, pathway-targeted strategies for pediatric and AYA OS.

The retinoblastoma (RB) protein was the first tumor suppressor discovered and has been extensively studied for its canonical role in cell-cycle regulation. However, RB has broader noncanonical roles in DNA damage repair, chromosomal stability, apoptosis control, lineage commitment, cell differentiation and broad transcriptional regulation. Historically, RB inactivation has been associated with tumorigenesis, as well as resistance to cyclin-dependent kinase 4/6 inhibitors (CDK4/6i), leading to its investigation as a potential predictive biomarker. However, clinical data have not demonstrated that RB function reliably predicts response to CDK4/6i consistently. These discrepancies highlight the need to reconsider RB’s role in therapeutic response, as RB loss can promote replication stress, induce chromosomal instability, and lead to transcriptional reprograming, potentially generating context-specific therapeutic vulnerabilities. In this review, we examine the multifaceted biology of RB and evaluate how its loss influences responses to chemotherapy and targeted therapies. We highlight emerging strategies that exploit RB-deficient states using rational monotherapy and combination approaches. Reframing RB dysfunction from a binary biomarker to a driver of exploitable cellular vulnerabilities may inform and expand precision oncology strategies for aggressive and treatment-resistant cancers.