Non-Small Cell Lung Cancer with Resistance to EGFR-TKI Therapy: CT Characteristics of T790M Mutation-positive Cancer

Machine learning-based radiomics strategy for prediction of acquired EGFR T790M mutation following treatment with EGFR-TKI in NSCLC

The epidermal growth factor receptor (EGFR) Thr790 Met (T790M) mutation is responsible for approximately half of the acquired resistance to EGFR-tyrosine kinase inhibitor (TKI) in non-small-cell lung cancer (NSCLC) patients. Identifying patients at diagnosis who are likely to develop this mutation after first- or second-generation EGFR-TKI treatment is crucial for better treatment outcomes. This study aims to develop and validate a radiomics-based machine learning (ML) approach to predict the T790M mutation in NSCLC patients at diagnosis. We collected retrospective data from 210 positive EGFR mutation NSCLC patients, extracting 1316 radiomics features from CT images. Using the LASSO algorithm, we selected 10 radiomics features and 2 clinical features most relevant to the mutations. We built models with 7 ML approaches and assessed their performance through the receiver operating characteristic (ROC) curve. The radiomics model and combined model, which integrated radiomics features and relevant clinical factors, achieved an area under the curve (AUC) of 0.80 (95% confidence interval [CI] 0.79-0.81) and 0.86 (0.87-0.88), respectively, in predicting the T790M mutation. Our study presents a convenient and noninvasive radiomics-based ML model for predicting this mutation at the time of diagnosis, aiding in targeted treatment planning for NSCLC patients with EGFR mutations.

CT Radiomics Predict EGFR-T790M Resistance Mutation in Advanced Non-Small Cell Lung Cancer Patients After Progression on First-line EGFR-TKI

RATIONALE AND OBJECTIVES

We aim to explore the value of chest CT radiomics in predicting the epidermal growth factor receptor (EGFR)-T790M resistance mutation of advanced non-small cell lung cancer (NSCLC) patients after the failure of first-line EGFR-tyrosine kinase inhibitor (EGFR-TKI).

MATERIALS AND METHODS

A total of 211 and 135 advanced NSCLC patients with tumor tissue-based (Cohort-1) or circulating tumor DNA (ctDNA)-based (Cohort-2) EGFR-T790M testing were included, respectively. Cohort-1 was used for modeling and Cohort-2 was for models' validation. Radiomic features were extracted from tumor lesions on chest nonenhanced CT (NECT) and/or contrast-enhanced CT (CECT). We used eight feature selectors and eight classifier algorithms to establish radiomic models. Models were evaluated by area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis (DCA).

RESULTS

CT morphological manifestations of peripheral location and pleural indentation sign were associated with EGFR-T790M. For NECT, CECT, and NECT+CECT radiomic features, the feature selector and classifier algorithms of LASSO and Stepwise logistic regression, Boruta and SVM, and LASSO and SVM were chosen to develop the optimal model, respectively (AUC: 0.844, 0.811, and 0.897). All models performed well in calibration curves and DCA. Independent validation of models in Cohort-2 revealed that both NECT and CECT models individually had limited power for predicting EGFR-T790M mutation detected by ctDNA (AUC: 0.649, 0.675), while the NECT+CECT radiomic model had a satisfactory AUC (0.760).

CONCLUSION

This study proved the feasibility of using CT radiomic features to predict the EGFR-T790M resistance mutation, which could be helpful in guiding personalized therapeutic strategies.

Analysis of the Diagnosis Model of Peripheral Non-Small-Cell Lung Cancer under Computed Tomography Images

This study aimed to explore the effect of deep learning models on lung CT image lung parenchymal segmentation (LPS) and the application value of CT image texture features in the diagnosis of peripheral non-small-cell lung cancer (NSCLC). Data of peripheral lung cancer (PLC) patients was collected retrospectively and was divided into peripheral SCLC group and peripheral NSCLC group according to the pathological examination results, ResNet50 model and feature pyramid network (FPN) algorithm were undertaken to improve the Mask-RCNN model, and after the MaZda software extracted the texture features of the CT images of PLC patients, the Fisher coefficient was used to reduce the dimensionality, and the texture features of the CT images were analyzed and compared. The results showed that the average Dice coefficients of the 2D CH algorithm, Faster-RCNN, Mask-RCNN, and the algorithm proposed in the validation set were 0.882, 0.953, 0.961, and 0.986, respectively. The accuracy rates were 88.3%, 93.5%, 94.4%, and 97.2%. The average segmentation speeds in lung CT images were 0.289 s/sheet, 0.115 s/sheet, 0.108 s/sheet, and 0.089 s/sheet. The improved deep learning model showed higher accuracy, better robustness, and faster speed than other algorithms in the LPS of CT images. In summary, deep learning can achieve the LPS of CT images and show excellent segmentation efficiency. The texture parameters of GLCM in CT images have excellent differential diagnosis performance for NSCLC and SCLC and potential clinical application value.

Radiomics and Radiogenomics in Evaluation of Colorectal Cancer Liver Metastasis

Colorectal cancer is one common digestive malignancy, and the most common approach of blood metastasis of colorectal cancer is through the portal vein system to the liver. Early detection and treatment of liver metastasis is the key to improving the prognosis of the patients. Radiomics and radiogenomics use non-invasive methods to evaluate the biological properties of tumors by deeply mining the texture features of images and quantifying the heterogeneity of metastatic tumors. Radiomics and radiogenomics have been applied widely in the detection, treatment, and prognostic evaluation of colorectal cancer liver metastases. Based on the imaging features of the liver, this paper reviews the current application of radiomics and radiogenomics in the diagnosis, treatment, monitor of disease progression, and prognosis of patients with colorectal cancer liver metastases.

Primary tumor location in lung cancer: the evaluation and administration

ABSTRACT

Lung cancer continues to be the leading cause of cancer-related death in the world, which is classically subgrouped into two major histological types: Non-small cell lung cancer (NSCLC) (85% of patients) and small-cell lung cancer (SCLC) (15%). Tumor location has been reported to be associated with the prognosis of various solid tumors. Several types of cancer often occur in a specific region and are more prone to spread to predilection locations, including colorectal cancer, prostate cancer, gastric cancer, ovarian cancer, cervical cancer, bladder cancer, lung tumor, and so on. Besides, tumor location is also considered as a risk factor for lung neoplasm with chronic obstructive pulmonary disease/emphysema. Additionally, the primary lung cancer location is associated with specific lymph node metastasis. And the recent analysis has shown that the primary location may affect metastasis pattern in metastatic NSCLC based on a large population. Numerous studies have enrolled the "location" factor in the risk model. Anatomy location and lobe-specific location are both important in prognosis. Therefore, it is important for us to clarify the characteristics about tumor location according to various definitions. However, the inconsistent definitions about tumor location among different articles are controversial. It is also a significant guidance in multimode therapy in the present time. In this review, we mainly aim to provide a new insight about tumor location, including anatomy, clinicopathology, and prognosis in patients with lung neoplasm.

Comparison of bronchoscopy and computed tomography-guided needle biopsy for re-biopsy in non-small cell lung cancer patients

BACKGROUND

New therapeutic drugs have been developed for non-small cell lung cancer (NSCLC), and the prognosis of advanced NSCLC patients has improved. However, resistance to these drugs is a concern, and re-biopsy is necessary to determine the mechanism of drug resistance. There are many reports about the protocols for re-biopsy, including techniques such as bronchoscopy and computed tomography-guided needle biopsy (CTNB); however, there is no consensus on which method is optimal. Therefore, we retrospectively reviewed the bronchoscopy and CTNB re-biopsies conducted at our hospital.

METHODS

We retrospectively analyzed 79 cases of re-biopsies with bronchoscopy or CTNB in patients with NSCLC from January 2014 to December 2016 at our institute.

RESULTS

Forty-nine cases of bronchoscopy and 30 cases of CTNB were taken for re-biopsy. The diagnostic rates of bronchoscopy and CTNB were 83.7% and 100%, respectively (p = 0.023). The complication rates of bronchoscopy and CTNB were 18.4% and 36.7%, respectively (p = 0.11), with a statistically significant difference in the incidence of pneumothorax (0% vs. 23.3%, respectively; p < 0.01). Pneumothorax required drainage in 6.7% of all CTNB cases. There were no fatalities in either group.

CONCLUSIONS

CTNB showed a higher diagnostic rate; however, it was associated with a higher rate of complications such as pneumothorax. Hence, the optimal modality must be determined individually for each patient.

Radiogenomics in Interventional Oncology

PURPOSE OF REVIEW

Radiogenomics is a growing field that has garnered immense interest over the past decade, owing to its numerous applications in the field of oncology and its potential value in improving patient outcomes. Current applications have only begun to delve into the potential of radiogenomics, and particularly in interventional oncology, there is room for development and increased value of these applications.

RECENT FINDINGS

The field of interventional oncology (IO) has seen valuable radiogenomic applications, from prediction of response to locoregional therapies in hepatocellular carcinoma to identification of genetic mutations in non-small cell lung cancer. Future directions that can increase the value of radiogenomics include applications that address tumor heterogeneity, predict immune responsiveness of tumors, and differentiate between oligoprogression and early widespread progression, among others. Radiogenomics, whether in terms of methodologies or applications, is still in the early stages of development and far from maturation. Future applications, particularly in the field of interventional oncology, will allow realization of its full potential.

Clinical implication and usefulness of de novo EGFR T790M mutation in lung adenocarcinoma with EGFR-tyrosine kinase inhibitor sensitizing mutation

In total, 102 cases diagnosed as lung adenocarcinoma with EGFR-tyrosine kinase inhibitor (TKI) sensitizing mutations (mEGFR) and had been treated with 1^st ~ 2^nd generation EGFR-TKI alone were enrolled for this study. De novo T790 M status was tested using the tissues at the initial diagnosis and positivity was defined as the ratio of T790 M/wild-type copies over 0.00294 by ddPCR.

Seventy patients (68.6%) harbored the de novo T790 M. De novo T790 M was more frequently detected in cases with EGFR L858 R mutation than those with EGFR exon 19 deletion (E19d) mutations (P = 0.024). Forty-three patients underwent rebiopsy due to disease progression. The cases who experienced progression due to acquired T790 M were more likely to have E19d at initial diagnosis and the presence of de novo T790 M and the ratio of T790 M/wild-type copies did not relate to the emergence of acquired T790 M. On the other hand, the cases with a longer duration of disease-control by EGFR-TKI had higher change to get acquired T790 M mutation (P-value = 0.040).

The presence of de novo T790 M has limitation in predicting disease progression by acquired T790 M, suggesting that identifying de novo T790 M through the ultrasensitive methods may not be necessary identifying patients who would be beneficial by 3rd-generation EGFR-TKI as the 1st line treatment.

A CT-derived deep neural network predicts for programmed death ligand-1 expression status in advanced lung adenocarcinomas

Background

Programmed death ligand-1 (PD-L1) expression remains a crucial predictor in selecting patients for immunotherapy. The current study aimed to non-invasively predict PD-L1 expression based on chest computed tomography (CT) images in advanced lung adenocarcinomas (LUAD), thus help select optimal patients who can potentially benefit from immunotherapy.

Methods

A total of 127 patients with stage III and IV LUAD were enrolled into this study. Pretreatment enhanced thin-section CT images were available for all patients and were analyzed in terms of both morphologic characteristics by radiologists and deep learning (DL), so to further determine the association between CT features and PD-L1 expression status. Univariate analysis and multivariate logical regression analysis were applied to evaluate significant variables. For DL, the 3D DenseNet model was built and validated. The study cohort were grouped by PD-L1 Tumor Proportion Scores (TPS) cutoff value of 1% (positive/negative expression) and 50% respectively.

Results

Among 127 LUAD patients, 46 (36.2%) patients were PD-L1-positive and 38 (29.9%) patients expressed PD-L1-TPS ≥50%. For morphologic characteristics, univariate and multivariate analysis revealed that only lung metastasis was significantly associated with PD-L1 expression status despite of different PD-L1 TPS cutoff values, and its Area under the receiver operating characteristic curve (AUC) for predicting PD-L1 expression were less than 0.700. On the other hand, the predictive value of DL-3D DenseNet model was higher than that of the morphologic characteristics, with AUC more than 0.750.

Conclusions

The traditional morphologic CT characteristics analyzed by radiologists show limited prediction efficacy for PD-L1 expression. By contrast, CT-derived deep neural network improves the prediction efficacy, it may serve as an important alternative marker for clinical PD-L1 detection.

Diffuse Lung Metastases in EGFR-Mutant Non-Small Cell Lung Cancer

Diffuse lung metastases have been reported in non-small cell lung cancer (NSCLC) harboring epidermal growth factor receptor (EGFR) mutations. The purpose of our study was to compare the incidence of diffuse lung metastases in EGFR-mutant NSCLC and EGFR-wild type NSCLC and to assess other imaging features that may be associated with diffuse lung metastases in EGFR-mutant NSCLC. Two radiologists retrospectively reviewed pre-treatment imaging of metastatic NSCLC cases with known EGFR mutation status. We assessed the imaging features of the primary tumor and patterns of metastases. The cohort consisted of 217 patients (117 EGFR-mutant, 100 EGFR wild-type). Diffuse lung metastasis was significantly more common in EGFR-mutant NSCLC compared with wild-type (18% vs. 3%, p < 0.01). Among the EGFR-mutant group, diffuse lung metastases were inversely correlated with the presence of a nodule greater than 6 mm other than the primary lung lesion (OR: 0.13, 95% CI: 0.04–0.41, p < 0.01). EGFR mutations in NSCLC are associated with increased frequency of diffuse lung metastases. The presence of diffuse lung metastases in EGFR-mutant NSCLC is also associated with a decreased presence of other larger discrete lung metastases. EGFR mutations in NSCLC should be suspected in the setting of a dominant primary lung mass associated with diffuse lung metastases.

Clinical Features and Progression Pattern of Acquired T790M-positive Compared With T790M-negative EGFR Mutant Non-small-cell Lung Cancer: Catching Tumor and Clinical Heterogeneity Over Time Through Liquid Biopsy

BACKGROUND

Clinical-pathologic predictors of acquired T790M epidermal growth factor receptor (EGFR) mutation in Caucasian patients with non-small-cell lung cancer (NSCLC) progressing after first-/second-generation tyrosine kinase inhibitors (TKIs) is an open field for research. Similarly, the best time point for T790M detection by liquid or tissue biopsy after disease progression is currently matter of debate.

PATIENTS AND METHODS

This is an observational study at 7 Italian centers enrolling patients with EGFR-mutant NSCLC progressing after first-/second-generation EGFR TKIs, between 2014 and 2018, aiming at comparing baseline clinical-pathologic features and progression patterns in acquired T790M-positive compared with T790M-negative cases.

RESULTS

A total of 235 patients received first-line treatment with gefitinib (N = 126; 53%), erlotinib (N = 51; 22%), or afatinib (N = 58; 25%). In 120 (51%) cases, T790M was detected in liquid biopsy, tissue biopsy, or both. Age younger than 65 years (P = .037), the presence of common mutations (P = .004), and better response to first-line TKI (P = .023) were correlated with T790M positivity. T790M detection was associated with higher number of new progressing sites (P = .04), liver progression (P = .002), and a lower frequency of lung metastases (P = .027). When serial liquid biopsies were performed (N = 15), an oligoprogressive disease was correlated with a negative test outcome, whereas systemic progression was observed at the time of T790M positivity.

CONCLUSION

This study on a Caucasian population showed that age, type of EGFR mutation at diagnosis, response to first-line treatment, and peculiar progression pattern are associated with T790M status. Serial liquid biopsy might be useful for treatment selection, especially when tissue rebiopsy is not feasible.

A Radiologist's Guide to the Changing Treatment Paradigm of Advanced Non-Small Cell Lung Cancer: The ASCO 2018 Molecular Testing Guidelines and Targeted Therapies

OBJECTIVE. The purpose of this article is to provide an imaging-based guide of the modern genomic classifications and targeted therapies for advanced non-small cell lung cancer (NSCLC) with an emphasis on the relevance of the 2018 American Society of Clinical Oncology molecular testing guidelines for radiologists. CONCLUSION. Knowledge of the radiologic relevance of lung cancer driver mutations and modern targeted agents is essential for imaging interpretation of advanced NSCLC in the modern age of precision medicine.

Frequency and clinical features of BRAF mutations among patients with stage III/IV lung adenocarcinoma without EGFR/ALK aberrations

Purpose

BRAF mutations are found in 1-5% of non-small cell lung cancers, particularly adenocarcinomas. However, information regarding this mutation is limited in patients without EGFR/ALK aberrations, who have limited treatment options.

Patients and methods

The medical records of 224 stage III/IV adenocarcinoma patients without EGFR/ALK aberrations and with available pathologic tissue, were retrospectively reviewed. BRAF mutations were evaluated using a PNAClamp^TM BRAF mutation detection kit (Panagene, Daejeon, Korea). The outcomes in the study population were compared with stage III/IV adenocarcinoma patients harboring an EGFR mutation. A case report of targeted therapy against BRAF mutations was also presented.

Results

A cohort of 222 adenocarcinoma patients with adequate pathologic tissue samples was analyzed. The median patient age was 63 years, 68.8% of the patients were male and 68.7% were ever-smokers. The V600E BRAF mutation was detected in 4 patients (1.8%). The 222 study patients had a poorer survival outcome compared to stage III/IV adenocarcinoma patients with an EGFR mutation (median, 12 vs 67 months, P<0.001) from a recent previous study. Moreover, a 47-year-old female with a recurrent adenocarcinoma and a BRAF V600E mutation exhibited tumor regression after a fourth line therapy with dabrafenib and trametinib, targeting agents against BRAF mutations.

Conclusion

Although BRAF mutations are found in 1.8% of advanced adenocarcinoma patients without EGFR/ALK aberration, they may be able to serve as a treatment target in those patients.

Nondiagnostic Percutaneous Transthoracic Needle Biopsy of Lung Lesions: A Multicenter Study of Malignancy Risk

Purpose To evaluate the malignancy risk of lung lesions that show nondiagnostic results at transthoracic needle biopsy (PTNB) of the lung and to identify any malignancy-associated risk factors in each nondiagnostic category. Materials and Methods In this retrospective study, 9384 initial PTNBs (9239 patients [mean age, 65 years; age range, 20-99 years] consisting of 5729 men [mean age, 66 years; age range, 20-99 years] and 3510 women [mean age, 63 years; age range, 20-94 years]) were performed in eight institutions between January 2010 and December 2014. PTNB results were categorized as diagnostic (malignant or specifically benign) or nondiagnostic (nonspecific benign pathologic findings, atypical cells, or insufficient specimen), and the proportion of final malignant diagnoses per nondiagnostic category was obtained. Malignancy-associated factors were determined by using multivariable analyses. Results Nondiagnostic results were present in 27.6% (2590 of 9384) of PTNBs. Proportions of final malignant diagnoses were 21.3% (339 of 1592) for nonspecific benignities, 90.1% (503 of 558) for atypical cells, and 46.6% (205 of 440) for insufficient specimens. In the nonspecific benign category, granulomatous inflammation (odds ratio [OR], 0.04; 95% confidence interval [CI]: 0.02, 0.12; P < .001), abscess (OR, 0.04; 95% CI: 0.01, 0.28; P = .001), and organizing pneumonia (OR, 0.05; 95% CI: 0.01, 0.23; P < .001) were demonstrated to be important factors negating malignancy. Atypical cells suspicious for malignancy were more associated with malignancy (OR, 6.3; 95% CI: 1.9, 21.0; P = .003) than were atypical cells of indeterminate malignancy. All 130 lesions with atypical cells suggestive of malignancy were finally malignant. Conclusion After nondiagnostic lung biopsies, lesions categorized as atypical cell lesions have a high likelihood of malignancy, with somewhat lower likelihood for lesions with insufficient specimens and nonspecific benign categories. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Elicker in this issue.