scholarly journals Deep Learning Analysis of CT Images Reveals High-Grade Pathological Features to Predict Survival in Lung Adenocarcinoma

Cancers ◽  
2021 ◽  
Vol 13 (16) ◽  
pp. 4077
Author(s):  
Yeonu Choi ◽  
Jaehong Aum ◽  
Se-Hoon Lee ◽  
Hong-Kwan Kim ◽  
Jhingook Kim ◽  
...  

We aimed to develop a deep learning (DL) model for predicting high-grade patterns in lung adenocarcinomas (ADC) and to assess the prognostic performance of model in advanced lung cancer patients who underwent neoadjuvant or definitive concurrent chemoradiation therapy (CCRT). We included 275 patients with 290 early lung ADCs from an ongoing prospective clinical trial in the training dataset, which we split into internal–training and internal–validation datasets. We constructed a diagnostic DL model of high-grade patterns of lung ADC considering both morphologic view of the tumor and context view of the area surrounding the tumor (MC3DN; morphologic-view context-view 3D network). Validation was performed on an independent dataset of 417 patients with advanced non-small cell lung cancer who underwent neoadjuvant or definitive CCRT. The area under the curve value of the DL model was 0.8 for the prediction of high-grade histologic patterns such as micropapillary and solid patterns (MPSol). When our model was applied to the validation set, a high probability of MPSol was associated with worse overall survival (probability of MPSol >0.5 vs. <0.5; 5-year OS rate 56.1% vs. 70.7%), indicating that our model could predict the clinical outcomes of advanced lung cancer patients. The subgroup with a high probability of MPSol estimated by the DL model showed a 1.76-fold higher risk of death (HR 1.76, 95% CI 1.16–2.68). Our DL model can be useful in estimating high-grade histologic patterns in lung ADCs and predicting clinical outcomes of patients with advanced lung cancer who underwent neoadjuvant or definitive CCRT.

Author(s):  
Justyna Błach ◽  
Paweł Krawczyk ◽  
Juliusz Pankowski ◽  
Jarosław Buczkowski ◽  
Izabela Chmielewska ◽  
...  

IntroductionThe importance of modern treatments for the extension of overall survival in advanced lung cancer (LC) patients is rarely reported in clinical trials (crossover effect). Recent clinical trials have compared experimental treatment methods and shown that chemotherapy is no longer a comparator. We studied the relevance of innovative treatment to the extension of overall survival in Polish lung cancer patients.Material and methodsWe described the outcome in 1463 patients diagnosed and treated for advanced LC. The study included patients receiving all available forms of treatment, i.e. chemotherapy, immunotherapy, EGFR tyrosine kinase inhibitors, ALK inhibitors, and best supportive care (BSC).ResultsMedian OS (mOS) for the whole group of patients was 6.5 months. mOS was significantly higher in patients with SCC (8.0 months) and AC (7.0 months) compared to patients with SCLC (6 months) and NSCLC NOS (3.5 months). mOS was 30 months for EGFR TKI-treated patients, 34 months for patients receiving second-line immunotherapy, 8.5 months for chemotherapy patients, and 1.0 month for patients who received BSC. mOS for patients treated with ALK inhibitors and first-line immunotherapy was not reached. The use of targeted therapies or immunotherapies significantly (p < 0.0001) reduced the risk of death compared to chemotherapy (HR = 0.373, 95% CI: 0.288–0.484 and HR = 0.313, 95% CI: 0.255–0.385).ConclusionsThe use of modern therapies in one of the treatment lines compared to chemotherapy significantly increased the long-term survival of advanced LC patients (34.5 vs. 8.5 months, HR = 0.336, 95% CI: 0.284– 0.397, p < 0.0001). Correct and early LC diagnosis is required, because patients with late diagnosis have a particularly poor prognosis.


2013 ◽  
Vol 31 (15_suppl) ◽  
pp. 6611-6611
Author(s):  
Hyunseok Kang ◽  
Sungjin Kim ◽  
Zhengjia Chen ◽  
Bassel F. El-Rayes ◽  
Johann Christoph Brandes ◽  
...  

6611 Background: The administration of chemotherapy to patients with limited performance status and within 6 weeks of death is considered an indicator of poor quality care. We assessed predictors of inpatient chemotherapy use and the risk of death in hospitalized lung cancer patients treated with chemotherapy in the US. Methods: Data were obtained from all US states that contributed to the Nationwide Inpatient Sample (NIS) by Agency for Health Care Research and Quality (AHRQ) in 2006 and 2010. Lung cancer diagnoses and inpatient chemotherapy were identified using Clinical Classification Software (CCS) code which is based on ICD9 and CPT codes. Univariate and multivariate analyses were performed to compare patients based on chemotherapy administration using ANOVA, chi-square test, and logistic regression. Initial analysis in the 2006 NIS data was validated in the 2010 NIS data. Results: 24,025 and 24,323 eligible hospitalized lung cancer patients including 1,005 (4.2 %) and 869 (3.6 %) patients treated with chemotherapy were identified in 2006 and 2010 respectively. Female gender, radiation use, urban location and longer length of stay (LOS) were significantly associated with receipt of chemotherapy. Chemotherapy administration was associated with prolonged hospital stay (14.1 ± 9.8 vs. 8.8 ± 8.1 days, p<.001) and increased odds of death in unadjusted analyses. Adjusted analysis showed significant increased odds of death in chemotherapy-treated patients with metastatic disease (vs. no metastasis); poor performance status indicated by severe loss of function (vs. minor/moderate loss of function) and increased LOS (Table). Conclusions: Inpatient administration of chemotherapy to hospitalized US lung cancer patients is associated with higher mortality and can be explained by treatment given to patients with high co-morbidity and disease burden. [Table: see text]


2018 ◽  
Author(s):  
Yutao Liu ◽  
Fang Xu ◽  
Yubo Wang ◽  
Qingchen Wu ◽  
Buhai Wang ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document