Preoperative identification of clinicopathological prognostic factors for relapse-free survival in clinical N1 non-small cell lung cancer: a retrospective single center-based study

A comparison of 18 F-FDG PET-based radiomics and deep learning in predicting regional lymph node metastasis in patients with resectable lung adenocarcinoma: a cross-scanner and temporal validation study

OBJECTIVE

The performance of 18 F-FDG PET-based radiomics and deep learning in detecting pathological regional nodal metastasis (pN+) in resectable lung adenocarcinoma varies, and their use across different generations of PET machines has not been thoroughly investigated. We compared handcrafted radiomics and deep learning using different PET scanners to predict pN+ in resectable lung adenocarcinoma.

METHODS

We retrospectively analyzed pretreatment 18 F-FDG PET from 148 lung adenocarcinoma patients who underwent curative surgery. Patients were separated into analog (n = 131) and digital (n = 17) PET cohorts. Handcrafted radiomics and a ResNet-50 deep-learning model of the primary tumor were used to predict pN+ status. Models were trained in the analog PET cohort, and the digital PET cohort was used for cross-scanner validation.

RESULTS

In the analog PET cohort, entropy, a handcrafted radiomics, independently predicted pN+. However, the areas under the receiver-operating-characteristic curves (AUCs) and accuracy for entropy were only 0.676 and 62.6%, respectively. The ResNet-50 model demonstrated a better AUC and accuracy of 0.929 and 94.7%, respectively. In the digital PET validation cohort, the ResNet-50 model also demonstrated better AUC (0.871 versus 0.697) and accuracy (88.2% versus 64.7%) than entropy. The ResNet-50 model achieved comparable specificity to visual interpretation but with superior sensitivity (83.3% versus 66.7%) in the digital PET cohort.

CONCLUSION

Applying deep learning across different generations of PET scanners may be feasible and better predict pN+ than handcrafted radiomics. Deep learning may complement visual interpretation and facilitate tailored therapeutic strategies for resectable lung adenocarcinoma.

Glucose metabolic heterogeneity correlates with pathological features and improves survival stratification of resectable lung adenocarcinoma

OBJECTIVE

We investigated whether glycolytic heterogeneity correlated with histopathology, and further stratified the survival outcomes pertaining to resectable lung adenocarcinoma.

METHODS

We retrospectively analyzed the ¹⁸F-fluorodeoxyglucose positron emission tomography-derived entropy and histopathology from 128 patients who had undergone curative surgery for lung adenocarcinoma. Disease-free survival (DFS) and overall survival (OS) were analyzed using univariate and multivariate Cox regression models. Independent predictors were used to construct survival prediction models.

RESULTS

Entropy significantly correlated with histopathology, including tumor grades, lympho-vascular invasion, and visceral pleural invasion. Furthermore, entropy was an independent predictor of unfavorable DFS (p = 0.031) and OS (p = 0.004), while pathological nodal metastasis independently predicted DFS (p = 0.009). Our entropy-based models outperformed the traditional staging system (c-index = 0.694 versus 0.636, p = 0.010 for DFS; c-index = 0.704 versus 0.630, p = 0.233 for OS). The models provided further survival stratification in subgroups comprising different tumor grades (DFS: HR = 2.065, 1.315, and 1.408 for grade 1-3, p = 0.004, 0.001, and 0.039, respectively; OS: HR = 25.557, 6.484, and 2.570, for grade 1-3, p = 0.006, < 0.001, and = 0.224, respectively).

CONCLUSION

The glycolytic heterogeneity portrayed by entropy is associated with aggressive histopathological characteristics. The proposed entropy-based models may provide more sophisticated survival stratification in addition to histopathology and may enable personalized treatment strategies for resectable lung cancer.

Impact of TP53 Mutations on EGFR-Tyrosine Kinase Inhibitor Efficacy and Potential Treatment Strategy

BACKGROUND

We investigated the impact of factors that influence TP53 mutations on the efficacy of EGFR-tyrosine kinase inhibitors and potential treatment strategies.

MATERIALS AND METHODS

Tumor samples were collected to screen gene mutations by next-generation sequencing, as well as the patients' baseline characteristics. The overall response to treatment with TKIs was evaluated based on interval computed tomography scans at each follow-up time point. A Fisher's exact test and log-rank test were used to determine the statistical differences in this study.

RESULTS

A total of 1134 clinical samples were collected from NSCLC patients, and TP53^mut was identified in 644 cases and EGFR^mut in 622 cases. A low frequency of TP53^mut or more than 50% EGFR co-mutation rate were related to the prognosis of TKI-treated patients. In addition, TP53^mut in the region outside of the DB domain had the strongest correlation with TKI resistance, whereas various types of mutations in the DB domain only had an impact on PFS. A grouping study of EGFR-TKI-based treatment revealed that EGFR-TKIs with chemotherapy were associated with more significant survival benefits for patients with prognostic TP53^mut, whereas EGFR-TKI therapy was favorable for TP53^wt patients. Furthermore, TP53^mut could shorten the time to the relapse of postoperative patients, who will also likely respond well to EGFR-TKIs with chemotherapy.

CONCLUSION

Various characteristics of TP53^mut affect the prognosis of TKI-treated patients to varying degrees. EGFR-TKIs with chemotherapy were benefit for patients' survival with prognostic TP53^mut, which provides an important reference for treatment management of EGFR^mut patients.

Tumor glycolytic heterogeneity improves detection of regional nodal metastasis in patients with lung adenocarcinoma

OBJECTIVE

The diagnostic performance of ¹⁸F-FDG PET for detecting regional lymph node metastasis in resectable lung cancer is variable, and its sensitivity for adenocarcinoma is even lower. We aimed to evaluate the value of ¹⁸F-FDG PET-derived features in predicting pathological lymph node metastasis in patients with lung adenocarcinoma.

METHODS

We retrospectively analyzed pretreatment ¹⁸F-FDG PET-derived features of 126 lung adenocarcinoma patients who underwent curative surgery. A logistic regression model was used to analyze the association between study variables and pathological regional lymph node status obtained from the curative surgery. Furthermore, Cox regression analysis was used to test the effect of the study variables on survival outcomes, including disease-free survival (DFS) and overall survival (OS).

RESULTS

The primary tumor entropy (OR = 1.7, p = 0.014) and visual interpretation of regional nodes via ¹⁸F-FDG PET (OR = 2.5, p = 0.026) independently predicted pathological regional lymph node metastasis. The areas under the receiver-operating-characteristic curves were 0.631, 0.671, and 0.711 for visual interpretation, primary tumor entropy, and their combination, respectively. Based on visual interpretation, a primary tumor entropy ≥ 3.0 improved the positive predictive value of positive visual interpretation from 51.2% to 63.0%, whereas an entropy < 3.0 improved the negative predictive value of negative visual interpretation from 75.3% to 82.6%. In cases with positive visual interpretation and low entropy, or negative visual interpretation and high entropy, the nodal metastasis rates were approximately 30%. In the survival analyses, the primary tumor entropy was also independently associated with DFS (HR = 2.7, p = 0.001) and OS (HR = 4.8, p = 0.001).

CONCLUSIONS

Our preliminary results show that the primary tumor entropy may improve ¹⁸F-FDG PET visual interpretation in predicting pathological nodal metastasis in lung adenocarcinoma, and may also show a survival prognostic value. This versatile biomarker may facilitate tailored therapeutic strategies for patients with resectable lung adenocarcinoma.

Prognostic Value of Combing Primary Tumor and Nodal Glycolytic-Volumetric Parameters of 18F-FDG PET in Patients with Non-Small Cell Lung Cancer and Regional Lymph Node Metastasis

We investigated whether the combination of primary tumor and nodal ¹⁸F-FDG PET parameters predict survival outcomes in patients with nodal metastatic non-small cell lung cancer (NSCLC) without distant metastasis. We retrospectively extracted pre-treatment ¹⁸F-FDG PET parameters from 89 nodal-positive NSCLC patients (stage IIB–IIIC). The Cox proportional hazard model was used to identify independent prognosticators of overall survival (OS) and progression-free survival (PFS). We devised survival stratification models based on the independent prognosticators and compared the model to the American Joint Committee on Cancer (AJCC) staging system using Harrell’s concordance index (c-index). Our results demonstrated that total TLG (the combination of primary tumor and nodal total lesion glycolysis) and age were independent risk factors for unfavorable OS (p < 0.001 and p = 0.001) and PFS (both p < 0.001), while the Eastern Cooperative Oncology Group scale independently predicted poor OS (p = 0.022). Our models based on the independent prognosticators outperformed the AJCC staging system (c-index = 0.732 versus 0.544 for OS and c-index = 0.672 versus 0.521 for PFS, both p < 0.001). Our results indicate that incorporating total TLG with clinical factors may refine risk stratification in nodal metastatic NSCLC patients and may facilitate tailored therapeutic strategies in this patient group.