Translational lung cancer research最新文献

Background: Oligometastatic disease, i.e., disease with limited metastatic spread (LMS), is increasingly recognised as a distinct clinical entity as it differs from disease with extensive metastatic spread (polymetastatic disease) in terms of treatment and prognosis. This retrospective observational study aimed to characterise a population-based cohort of non-small-cell lung cancer (NSCLC) patients with synchronous metastatic disease, stratified by the extent of metastatic spread.

Methods: NSCLC patients [≥18 years, diagnosed 2016-2020, ICD-10 C34, Union for International Cancer Control (UICC) stage IV at primary diagnosis] were identified from three German population-based cancer registries covering approximately 27% of the national population. The extent of metastatic disease was classified according to the 8^th edition of the tumour-node-metastasis (TNM) system and further subclassified by the number of organ systems involved. We differentiated between patients with TNM-M stage M1a (intra-thoracic LMS), M1b [one single extra-thoracic metastasis or extra-thoracic LMS (ET-LMS)], M1c and ≤3 [limited multi-organ involvement (LMOI)] and >3 organ systems involved or with generalised metastatic disease [extensive multi-organ involvement (EMOI)]. The intent of the statistical analysis was mainly descriptive.

Results: The cohort included 8,033 NSCLC patients with distant metastasis at diagnosis of the primary tumour (1,767 with intra-thoracic LMS, 1,314 with ET-LMS, 4,196 with LMOI and 756 with EMOI). Patients with EMOI were younger (median: 64 vs. 66-70 years), more often female (45.0% vs. 40.6-41.4%), had a higher proportion of adenocarcinomas (83.7% vs. 72.2-78.7%) and showed more frequent N3 lymph node involvement (37.8% vs. 22.9-30.4%) than patients with LMS (subgroups: intra-thoracic LMS, ET-LMS or LMOI). Patients in the ET-LMS subgroup most often presented with one brain, osseous or adrenal metastasis (in descending order), patients in the LMOI subgroup most often with osseous, brain and/or pulmonal metastases, while patients in the EMOI subgroup presented most often with osseous, hepatic and/or pulmonal metastases. Patients were followed-up for a median time of 7 months (interquartile range, 3-16 months), during which every fourth patient with initially LMS experienced a new metastasis with mainly limited spread, usually in another organ than at primary diagnosis. Brain and osseous metastases were most frequently observed during follow-up.

Conclusions: We were able to identify patients with LMS in a population-based real-world dataset. This large data source forms the basis of a study series, in which subsequent analyses will assess survival prospects and optimal treatment strategies for this therapeutically relevant patient group.

Background: Rearranged during transfection (RET) fusion is a key driver in non-small cell lung cancer (NSCLC). Accurate detection is essential for targeted therapeutic strategies, and its reliability varies with the diagnostic methods employed. This study systematically compares the concordance of various molecular profiling techniques for identifying RET fusions in early-stage NSCLC.

Methods: This retrospective study included 40 NSCLC patients with RET fusions (RET+) identified by DNA sequencing (DNA-seq). Comprehensive detection was conducted using fluorescence in situ hybridization (FISH) and RNA sequencing (RNA-seq), incorporating both targeted RNA-seq and whole-transcriptome sequencing (WTS). Clinical and molecular features were evaluated for associations with fusion types.

Results: Patients were predominantly female (67.5%) and never-smokers (87.5%), with a median age of 53 years. All patients underwent FISH, while 39 cases underwent RNA-seq, with one excluded due to RNA quality control failure. The most common fusions were KIF5B::RET and CCDC6::RET (89.7%), alongside noncanonical fusion partners such as ERC1 (5.0%) and CCDC186 (2.5%). WTS achieved a 79.5% (31/39) RET+ detection rate. Targeted RNA-seq uncovered an additional five RET+ cases missed by WTS. Concordance rates for fusion detection were 92.3% between DNA-seq and RNA-seq, 84.6% between RNA-seq and FISH, and 82.5% between DNA-seq and FISH. Interestingly, patients exhibiting nonreciprocal RET translocations were significantly younger (P=0.03) and presented a lower Ki67 proliferation index (P=0.03).

Conclusions: Our findings underscore the complexity of RET fusion characterization and the necessity for a comprehensive diagnostic approach, which enhances the identification of both canonical and noncanonical alterations. This supports precision oncology initiatives aimed at optimizing treatment outcomes for RET+ NSCLC patients.

Background: The phase III PACIFIC trial demonstrated that one year of durvalumab maintenance therapy after concurrent chemoradiotherapy (CCRT) significantly improved overall survival (OS) in unresectable stage III non-small cell lung cancer (NSCLC), reducing the risk of death by 32% compared with observation. Durvalumab therapy administered for up to 12 months is currently the standard of care for patients with locally advanced NSCLC. This study aimed to identify the factors influencing the completion of durvalumab therapy.

Methods: This retrospective study analyzed patients who initiated durvalumab therapy for unresectable NSCLC at Niigata University Medical and Dental Hospital between August 2018 and December 2022. Patients who discontinued the treatment for reasons unrelated to disease progression or adverse events were excluded. The patients were divided into durvalumab completion and non-completion groups, and the clinical factors influencing durvalumab completion were analyzed.

Results: Among the 65 enrolled patients, 46.2% completed one year of durvalumab maintenance therapy. The completion rate was significantly lower in patients who received low-dose carboplatin as part of CCRT (20% vs. 54%, P=0.04), and higher in patients with adenocarcinoma (P=0.03). Older patients were more frequently treated with low-dose carboplatin CCRT (median age: 78 vs. 66 years, P<0.001). Patients receiving low-dose carboplatin had a significantly shorter time to treatment failure (median: 3.8 months vs. not reached, P=0.002) and tended to have a shorter OS (median: 47 months vs. not reached, P=0.06).

Conclusions: Low-dose carboplatin CCRT is associated with lower durvalumab completion rates and worse outcomes, particularly in older patients. Alternative CCRT regimens may warrant consideration, particularly for older patients who are less likely to complete durvalumab maintenance.

Background: Preliminary studies suggest that pan-immune-inflammation value (PIV) has the potential to serve as a prognostic tool for lung cancer. However, existing studies are limited by inconsistent findings regarding the impact of high PIV on patient outcomes. To provide a more comprehensive assessment, we conducted a meta-analysis to clarify the prognostic value of PIV in lung cancer.

Methods: Two researchers independently searched the PubMed, Cochrane, Embase, and Web of Science databases for studies evaluating the associations between PIV and prognoses in lung cancer patients (up to July 15, 2025). Studies were included if they reported high versus low PIV and provided hazard ratios (HRs) with 95% confidence intervals (CIs) for overall survival (OS), progression-free survival (PFS), etc. Study quality was assessed using the Newcastle-Ottawa Scale (NOS). Pooled HRs and 95% CIs were calculated to determine the associations between PIV and patient prognosis.

Results: A total of ten studies comprising 1,969 patients were included. Meta-analyses demonstrated that high PIV was significantly associated with OS (HR =2.86, 95% CI: 2.23-3.65, P<0.001) and PFS (HR =2.06, 95% CI: 1.65-2.59, P<0.001) in lung cancer patients. In non-small cell lung cancer (NSCLC) patients, high PIV was significantly associated with worse OS (HR =2.76, 95% CI: 2.14-3.56, P<0.001) and PFS (HR =1.94, 95% CI: 1.55-2.42, P<0.001). In small cell lung cancer (SCLC) patients, even stronger associations were observed for OS (HR =3.47, 95% CI: 2.21-5.44, P<0.001) and PFS (HR =2.33, 95% CI: 1.63-3.33, P<0.001). Subgroup analyses further confirmed that PIV served as a critical prognostic marker for both OS and PFS. All studies were of high quality according to the NOS.

Conclusions: PIV can serve as an independent prognostic biomarker for survival outcomes in lung cancer patients. Therefore, incorporating PIV into prognostic assessments may provide additional support for individualized treatment decision-making.

Background: Small-cell lung cancer (SCLC) is a highly aggressive disease with limited treatment options beyond first-line platinum-based chemotherapy. Although various second-line regimens such as topotecan and cyclophosphamide-adriamycin-vincristine have been evaluated, clinical outcomes remain poor, particularly in patients with platinum-resistant or refractory disease. The combination of paclitaxel and doxorubicin showed encouraging results in early-phase trials, but data on its use in real-world settings are lacking. This study aimed to assess the efficacy and safety of this chemotherapy regimen in patients with previously treated SCLC in a real-world clinical setting.

Methods: Data were retrospectively collected from all consecutive patients with previously treated SCLC at a single institution in France between 2000 and 2024.

Results: A total of 149 patients were included, of whom 79.9% had platinum-resistant or refractory disease at the time of paclitaxel-doxorubicin initiation. The median progression-free survival (PFS) was 2.4 months [95% confidence interval (CI): 2.1-3.8], and the median overall survival (OS) was 5.1 months (95% CI: 4.7-6.0). The objective response rate was 22.2% in the overall population and 27.7% when excluding patients not evaluable for response. The disease control rate was 41.6% including all patients and 52.1% when excluding non-evaluable cases. In multivariable analysis, an Eastern Cooperative Oncology Group performance status of 2-4 compared to 0-1 was associated with shorter PFS [hazard ratio (HR) =1.9, 95% CI: 1.22-2.8, P=0.004] and OS (HR =2.73, 95% CI: 1.76-4.2, P<0.001). Adverse events led to dose reductions in 54.9% of patients, primarily due to general deterioration, hematologic toxicity, or neuropathy. No treatment-related deaths were reported.

Conclusions: This real-world study suggests that the paclitaxel-doxorubicin combination provides some clinical activity in previously treated SCLC, including in platinum-resistant/refractory disease. Although toxicities were common, they were generally manageable. These findings support further investigation of this regimen with careful patient selection.

Background: Considerable differences exist in the prognosis of patients suffering from advanced non-small cell lung cancer. In fact, they undergo third-generation epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKI) single-agent targeted treatment. This study aimed to recognize the predictive factors influencing the prognosis of such patients, and identify patients at high risk of relapse to instruct clinical therapy and ameliorate prognosis.

Methods: Our research included 255 patients suffering from advanced lung adenocarcinoma (LUAD) treated with third-generation EGFR-TKI monotherapy alone, categorized into the training and validation cohorts. In univariate and multivariate analyses, clinicopathologic features and radiomic features were included. Moreover, statistically significant predictive factors were applied to formulate a nomogram. Area under the curve (AUC) of receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA) were utilized to verify the model.

Results: Neutrophil-lymphocyte ratio (NLR) ≥4.6, EGFRex21 L858R mutation, brain metastasis, and radiomic characteristics worked as independent risk factors for progression-free survival (PFS). Age ≥60 years, EGFRex21 L858R mutation, brain metastasis, monocyte-lymphocyte ratio (MLR) ≥0.3, and radiomic features acted as independent risk factors for overall survival (OS). Incorporating this into the nomogram, the AUC of the nomogram for forecasting 6-, 12-, and 24-month PFS and 1-, 2-, and 3-year OS were 0.810, 0.862, 0.873, and 0.886, 0.881, and 0.839, respectively, in the training cohort, and 0.885, 0.858, 0.847 in the validation cohort, and 0.804, 0.824, 0.806, all showing excellent discrimination. The calibration curves showed great consistency between predicted and actual observations. DCA exhibited a gratifying positive net benefit in most threshold probabilities, indicative of a beneficial clinical result. As claimed by PFS and OS survival analysis, the nomogram could distinguish low-risk group from high-risk group.

Conclusions: Our research formulated and corroborated a nomogram, according to clinicopathologic factors and radiomic features. Such a prediction model could be utilized as a strong tool for forecasting the prognosis of patients.

Background: Malignant pleural effusion (MPE) is a frequent complication of advanced lung cancer, and rapid and accurate diagnosis is critical for timely therapeutic decision-making. Although medical thoracoscopy (MT) provides direct visualization and targeted biopsy, resulting in high diagnostic yield, the clinical utility of its findings is contingent on operator experience and subsequent confirmation via pathological analysis. Recent advances in deep learning have enabled automated image classification in various fields, but its application in thoracoscopic images remains unexplored. The aim of our study was to develop a deep learning-based model to classify pleural malignancy and to evaluate its diagnostic performance.

Methods: We developed a deep learning-based classification model using thoracoscopic images obtained from 426 patients who underwent MT at Incheon St. Mary's Hospital between May 2015 and July 2024. After preprocessing and standardization of 4,932 images (2,093 benign, 2,839 malignant), we trained an InceptionV3-based convolutional neural network using transfer learning and online augmentation during training. Model performance was evaluated according to accuracy, precision, recall, the F1 score, the area under the receiver operating characteristic curve (ROC-AUC), and gradient-weighted class activation mapping (Grad-CAM) visualization.

Results: In total, 4,932 thoracoscopic images from 426 patients were used to train and evaluate the model. In the test set, the InceptionV3-based model achieved a classification accuracy of 81.7% [95% confidence interval (CI): ± 3.5%], with a precision of 82.4% (95% CI: ±4.3%), recall of 86.6% (95% CI: ±3.8%), and F1 score of 84.6% (95% CI: ±3.0%). The AUC was 0.90 (95% CI: ±2.6%), indicating excellent discriminative performance. A confusion matrix indicated 165 true positives, 25 false negatives, 29 false positives, and 262 true negatives. Grad-CAM visualizations confirmed that the model consistently focused on visually relevant pleural abnormalities such as nodularity, thickening, and neovascularization. Notably, the model maintained high accuracy even in cases without overt tumor nodules.

Conclusions: This study presents the first deep learning-based classification model of pleural malignancy using MT images. The model showed excellent diagnostic performance and has the potential to aid real-time clinical decision-making during thoracoscopy by suggesting malignant targets for biopsy or pleurodesis.

Background: Invasive mucinous lung adenocarcinoma (IMA) is challenging to diagnose via bronchoscopy due to its distinct histopathological characteristics, including poorly atypical tumor cells with abundant mucins. Given that cryobiopsy facilitates high-quality and quantity tissue sampling, we hypothesized that it could improve diagnostic success. This retrospective study aimed to determine whether cryobiopsy is a valid diagnostic tool for IMA.

Methods: We retrospectively reviewed consecutive patients who underwent surgical resection for lung cancer at our institution between March 2017 and July 2024. Among them, patients diagnosed with IMA were selected. Those who had undergone a diagnostic biopsy for peripheral pulmonary lesions (PPLs) were included in the study. These patients were divided into two groups: cryo and conventional, based on whether cryobiopsy was performed. The diagnostic yield and safety profiles of both groups were compared.

Results: Of 5,053 surgeries for lung cancer, 201 (4.0%) were diagnosed with IMA. Among them, 106 patients who had undergone bronchoscopy for PPLs were included in the analyses: 46 in the cryo group and 60 in the conventional group. The diagnostic yield was significantly higher in the cryo group than in the conventional group (93.5% vs. 73.3%, P=0.01). Multivariable analysis showed cryobiopsy was significantly associated with higher yield [adjusted odds ratio (OR), 5.07; 95% confidence interval (CI): 1.33-19.40; P=0.02]. Although not fatal, complications like bleeding and pneumonia were more frequent in the cryo group.

Conclusions: Cryobiopsy is a valid technique for making a diagnosis of IMA.