Lung cancer remains a major cause of cancer-related mortality, including among patients diagnosed with potentially curable disease. Current decisions regarding perioperative treatment and follow-up rely on a limited number of established prognostic factors. Further knowledge is needed regarding the clinical, histopathological, and molecular factors associated with subsequent metastatic disease, particularly brain metastasis (BM), which is associated with poor prognosis. We investigated these factors in a large cohort of patients with surgically resected stage I–III lung cancer followed for more than five years. BM was associated with higher pathological tumour and lymph-node stage and younger age at surgery. Lung adenocarcinoma was associated with a higher risk of BM than lung squamous cell carcinoma. Several molecular alterations were associated with metastatic disease or BM, including KRAS mutations in lung adenocarcinoma and NFE2L2 mutations in lung squamous cell carcinoma. Additional associations involving specific EGFR and KRAS mutation subtypes and rarer alterations were observed, but were based on small numbers of cases and should be considered exploratory. Several of the associated genes are involved in the RAS/MAPK, PI3K/AKT/mTOR, or NRF2-related pathways. However, pathway activation and causality could not be established in this observational study. These findings support further investigation of clinical, histopathological, and molecular factors associated with metastatic progression after surgical resection. Validation in larger, independent, stage-stratified and contemporary cohorts with comprehensive treatment data is required before the identified molecular alterations can be considered for clinical risk stratification or surveillance.
Comparative evidence for hepatic resection and stereotactic body radiotherapy (SBRT) in breast cancer liver oligometastases remains limited. We compared survival, progression, and recovery after these treatments. This three-center retrospective study included adults with breast cancer and ≤ 3 liver metastases undergoing hepatic resection or SBRT between 2010 and 2023. Patients with uncontrolled extrahepatic disease were excluded. Propensity score matching generated 1:1 treatment pairs. Overall survival (OS) was the primary endpoint. Adjusted Cox models evaluated OS and progression-free survival (PFS); progression outcomes were analyzed with death as a competing event. Among 1,055 eligible patients, 708 were matched (354 per group). During a median follow-up of 54.8 months, 432 deaths occurred. Median OS was 48.3 months after surgery and 38.8 months after SBRT (adjusted hazard ratio [HR] for SBRT versus surgery, 1.24; 95
Lung cancer brain metastasis develops through reciprocal interactions between metastatic tumor cells and a brain-specific vascular, glial, and immune microenvironment. Tumor-derived extracellular vesicles (EVs) can participate in this process by altering blood–brain barrier (BBB) properties, influencing brain-directed distribution, and transferring regulatory cargo to endothelial and myeloid recipient cells. This review integrates direct evidence from non-small cell lung cancer (NSCLC) brain or central nervous system metastasis with related lung cancer EV studies, while distinguishing these findings from cross-cancer mechanistic evidence. We first describe the transition from the BBB to the heterogeneous blood–tumor barrier and distinguish the contributions of astrocytes, resident microglia, recruited monocyte-derived macrophages, and T cells. We then examine how defined EV cargo, including lnc-MMP2-2, LINC00482, miR-21, L1CAM, and miR-374a-5p, affects barrier integrity, myeloid state, or brain-metastatic behavior. Myeloid responses are interpreted according to cellular origin and function rather than by a fixed M1/M2 binary. Exercise-related myokines and exerkines are considered separately as a hypothesis-generating systemic layer that could influence tumor-cell EV production, deliver signals through muscle-derived EVs, or alter recipient-cell responsiveness; none of these routes has yet been established as a causal mechanism in lung cancer brain metastasis. Together, current evidence supports a role for EV-mediated vascular and immune/glial communication in lung cancer brain metastasis, while leaving the contribution of exercise-related signals unresolved.
Stereotactic radiosurgery (SRS) is the treatment of choice for limited brain metastases (BM), and its combination with PD-1/PD-L1 inhibitors is hypothesized to offer synergistic therapeutic benefits. However, there are concerns regarding augmented risk of radiation-induced brain necrosis (RIBN). This systematic review and meta-analysis evaluate the efficacy and toxicity profile of PD-1/PD-L1 inhibitors combined with SRS or fractionated stereotactic radiotherapy (FSRT). Four databases were screened, Embase, Web of Science, PubMed and Scopus from inception to November 2025. Primary outcomes were overall survival (OS), extracranial and distant intracranial progression free survival (PFS), and local control (LC). A subgroup analysis was performed based on the primary metastasis origin and radiotherapy treatment modalities. We extracted RIBN events and grade ≥ 3 neurological events. A single‐arm meta‐analysis was performed using the random-effects modeling. This review protocol was registered on November 2025 on PROSPERO (ID CRD420251171879). Thirty-four studies were included, published between 2016 and 2025 comprising 2,587 patients with BM. The pooled one-year OS was 65
To explore if adding local radiotherapy to systemic treatment could enhance the survival of newly-diagnosed stage IVB cervical cancer (CC) patients and compare the impact of different radiotherapy strategies on survival. We included patients diagnosed from 2004 to 2022. Local radiotherapy strategies included external beam radiotherapy (EBRT) and/or brachytherapy (BT). Multivariable Cox regression models were used to assess the link between treatment modality and survival. A total of 4257 patients were involved, all receiving chemotherapy. There were 1280 (30.1
