Virtual Library
Start Your Search
Kyongmin Sarah Beck
Author of
-
+
MA13 - Interventional Pulmonology (ID 914)
- Event: WCLC 2018
- Type: Mini Oral Abstract Session
- Track: Interventional Diagnostics/Pulmonology
- Presentations: 1
- Moderators:
- Coordinates: 9/25/2018, 10:30 - 12:00, Room 206 AC
-
+
MA13.01 - CT-Guided Transthoracic Needle Biopsy for Evaluation of PD-L1 Expression: Comparison of 22C3 and SP263 Assays (ID 11312)
10:30 - 10:35 | Presenting Author(s): Kyongmin Sarah Beck
- Abstract
- Presentation
Background
Although there are a few studies about concordance of different assays testing PD-L1 expression using surgical specimens, there hasn’t been any such concordance study using real-world biopsy specimens. However, many of the patients requiring immunotherapy and thus PD-L1 testing have unresectable lung cancer and have to rely on small biopsy results. Although phase 2 of Blueprint phase 2 does include core biopsy specimens, they are mixed with bronchial biopsy specimens and the absolute number is very small (n=20). We sought to evaluate the concordance of 22C3 and SP263 assays in a larger number CT-guided transthoracic needle biopsy (TNB) specimens.
The purpose of this study was to assess the concordance of two commercially available diagnostic assays (22C3 and SP263) in evaluating programmed cell death ligand-1 (PD-L1) expression using specimens from CT-guided TNB in a routine clinical setting.
a9ded1e5ce5d75814730bb4caaf49419 Method
This retrospective analysis reviewed 202 non-small cell lung cancer (NSCLC) patients who underwent CT-guided TNB at our institution from April 2017 to February 2018. Among these, biopsy specimens tested with both 22C3 and SP263 assays were included for review. Concordance of PD-L1 expression levels determined by the two assays was assessed using intraclass correlation coefficient, and the agreement of dichotomized values at various cut-offs (1%, 25%, and 50%) were assessed using Cohen’s κ coefficient of agreement. Clinical characteristics and biopsy-related factors were also assessed for the association of concordance of PD-L1 expression detected by different assays
4c3880bb027f159e801041b1021e88e8 Result
In total, 80 patients (M:F =47:33, mean age: 68.0 years) were included in the study. Concordance of PD-L1 expression levels was high (intraclass coefficient: 0.892) between 22C3 and SP263 assays. Agreements at cut-off levels of 1%, 25%, and 50% were also good, with κ values of 0.878, 0.698, and 0.790 respectively. Positive percent agreement was 93.2%, 100.0%, and 95.2% for agreements at 1%, 25%, and 50%. At multivariate analysis, the presence of emphysema was significantly related to discordant PD-L1 results (odds ratio: 0.059, p= 0.005).
8eea62084ca7e541d918e823422bd82e Conclusion
There is a high concordance of PD-L1 expression evaluated with 22C3 and SP263 assays using CT-guided TNB specimens.
6f8b794f3246b0c1e1780bb4d4d5dc53Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.
-
+
P2.16 - Treatment of Early Stage/Localized Disease (Not CME Accredited Session) (ID 965)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Track:
- Presentations: 1
- Moderators:
- Coordinates: 9/25/2018, 16:45 - 18:00, Exhibit Hall
-
+
P2.16-02 - Predicting Pathological Noninvasiveness in T1 Non-Small Cell Lung Cancer on Chest CT Scan Using Deep Learning Algorithm (ID 13752)
16:45 - 18:00 | Presenting Author(s): Kyongmin Sarah Beck
- Abstract
Background
Because of the risk of recurrence, lobectomy is indicated even for small lung cancers. If we could accurately predict and classify noninvasive small lung cancers on CT images using deep learning algorithm prior to surgery, a more refined selection of candidates who would benefit from limited resection without increasing the risk of recurrence would be possible. The purpose of this study was to evaluate the ability of an artificial intelligence (deep learning algorithm) to predict pathological noninvasiveness of T1 non-small cell lung cancer (NSCLC) using computed tomography (CT) images.
a9ded1e5ce5d75814730bb4caaf49419 Method
468 preoperative CT images of NSCLC smaller than 3 cm, resected at our institution from 2008 to 2015 were used to train and validate the deep learning algorithm. Noninvasiveness was defined as the absence of nodal involvement, vascular invasion, and lymphatic invasion, and NSCLCs were classified as either noninvasive or invasive according to the pathological reports. A deep 3D convolutional neural network (CNN) was trained using 5-fold cross-validation method. To normalize input data, we rescaled CT image to let one voxel size be represented as (1mm,1mm,1mm). Input cube size was (32mm, 32mm,32mm). Horizontal flip and vertical flip augmentation were applied, and max-pooling and dropout layer were used to avoid overfitting. Receiver operating characteristic (ROC) curves and areas under the curve (AUCs), accuracy, and sensitivity/specificity were used to assess the performance of 3D CNN. We also added tumor size as an external feature to the 3D CNN model and evaluated the performance of the combined model using Xgboost classifier.
4c3880bb027f159e801041b1021e88e8 Result
157 out of 486 samples (33.5%) were invasive. A subsample group composed of 10% of the data (32 noninvasive and 16 invasive samples) was retained for the validation set.
The 3D CNN showed an AUC of 0.826, 77.08% accuracy, 68.7% sensitivity, and 81.2 % specificity. Adding tumor size to the 3D CNN model showed an AUC of 0.855, 81.25% accuracy, 87.5% sensitivity, and 78.1% specificity.
8eea62084ca7e541d918e823422bd82e Conclusion
Artificial intelligence can accurately predict pathological noninvasiveness in T1 size NSCLC on CT images.
6f8b794f3246b0c1e1780bb4d4d5dc53
