Virtual Library

Start Your Search

Chang Chen



Author of

  • +

    MA03 - Lung Cancer Screening - Next Step (ID 896)

    • Event: WCLC 2018
    • Type: Mini Oral Abstract Session
    • Track: Screening and Early Detection
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/24/2018, 10:30 - 12:00, Room 206 AC
    • +

      MA03.07 - Development and Validation of Deep Learning Model for Recognition of Histologic Subtype of Lung Adenocarcinoma from CT Images (ID 14412)

      11:10 - 11:15  |  Author(s): Chang Chen

      • Abstract
      • Presentation
      • Slides

      Background

      The clinical decision to either follow-up or resection from radiologic features for lung adenocarcinoma (atypical adenomatous hyperplasia [AAH], adenocarcinoma in situ [AIS], minimally invasive adenocarcinoma [MIA] and invasive adenocarcinoma [IA]) appearing as Sub-solid nodules (SSNs) is still challenge, and currently more relies on measures of diameter, solid component ratio. With the successful application of deep learning neuro-network (DLNN) for the classification of skin or common treatable blinding retinal diseases, we hypothesized that DLNN might help the histologic subtype classification of SSNs from CT images. The purpose of this study is to develop and validate a deep neuro-network model to classify AAH, AIS, MIA and IA or define a feasible classification for follow-up or treatment decision.

      a9ded1e5ce5d75814730bb4caaf49419 Method

      A total of 869 patients with 1344 pathologic confirmed nodules (AAH: 75, AIS: 340, MIA:321, IA: 608) were enrolled into this study. Two 3D mixed-scale dense-connected convolutional neuro network models (3D MS-DenseNet) were developed for 2 classification tasks: 4-class (AAH, AIS, MIA, IA), 3-class (AAH, AIS/MIA, IA). Eighty percent of whole datasets were randomly selected for training set, while other 20% were used for testing set. The nodules were firstly selected using a bounding box in 3D Slicer, and then cropped into 128 x 128 x 128 matrix size as the input to MS-DenseNet, and the output layer from the network was a 4-node or 3-node softmax classifier. Confusion matrix were used for the performance evaluation of both models and the classification accuracy for each class were reported.

      4c3880bb027f159e801041b1021e88e8 Result

      The classification accuracy of AAH, AIS, MIA, IA in testing set were 0.75, 0.45, 0.52, 0.85 respectively by 4-class, suggesting that the differentiation between AIS and MIA from CT images by neuro-network is challenge. While in the 3-class classification task with purpose of decision supporting for treatment, the classification accuracy of AAH, AIS/MIA, IA were 0.70, 0.73, 0.88 in the same testing set.

      8eea62084ca7e541d918e823422bd82e Conclusion

      The DLNN showed potential capability in differentiating AAH, IA from other adenocarcinoma subtypes, while failed to differentiate AIS and MIA. When combing AIS and MIA for reclassify adenocarcinoma subtypes from the perspective of treatment, the DLNN achieved reasonable performance, suggesting that DLNN might be useful in supporting clinical treatment decision whether to follow-up or take different resection for SSNs.

      6f8b794f3246b0c1e1780bb4d4d5dc53

      Only 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.

      Only Active 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 or select "Add to Cart" and proceed to checkout.

  • +

    MA09 - Lung Cancer Surgical and Molecular Pathology (ID 908)

    • Event: WCLC 2018
    • Type: Mini Oral Abstract Session
    • Track: Pathology
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/24/2018, 15:15 - 16:45, Room 202 BD
    • +

      MA09.02 - Tumor Size and Frozen Section Should Be Considered Jointly to Predict the Final Pathology for Lung Adenocarcinoma (ID 13365)

      15:20 - 15:25  |  Author(s): Chang Chen

      • Abstract
      • Slides

      Background

      Invasive adenocarcinoma intraoperatively misdiagnosed as adenocarcinoma in situ or minimally invasive adenocarcinoma is more likely to undergo potentially insufficient resection. The purpose of our study was to evaluate the diagnostic accuracy of frozen section.

      a9ded1e5ce5d75814730bb4caaf49419 Method

      We retrospectively reviewed 1,111 lung adenocarcinomas to evaluate the diagnostic performance of frozen section. A derivation cohort consisting of 436 cases of AIS or MIA diagnosed by frozen section in the same period were analyzed to find predictive factors for invasive adenocarcinoma as the final diagnosis. Validation cohorts were included to confirm the results.

      4c3880bb027f159e801041b1021e88e8 Result

      Intraoperatively measured tumor size was the only independent factor for invasive adenocarcinoma as the final diagnosis (P = 0.001) in the derivation cohort, and was confirmed by validation cohorts. Fifty-nine misdiagnosed invasive adenocarcinomas in the three cohorts consisted of 54 lepidic predominant type, 1 papillary and 4 acinar predominant type. There were no positive N1, N2 node, pleural, lymphatic and vascular invasion cases found. Thirty-seven (37/59, 63%) cases of misdiagnosis were attributed to sampling error, which was the main reason.figure1.jpgfigure3.jpg

      8eea62084ca7e541d918e823422bd82e Conclusion

      Adenocarcinoma in situ or minimally invasive adenocarcinoma ≥ 1 cm by frozen section were more likely to be invasive adenocarcinoma because of sampling error.

      6f8b794f3246b0c1e1780bb4d4d5dc53

      Only Active 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 or select "Add to Cart" and proceed to checkout.

  • +

    P1.16 - Treatment of Early Stage/Localized Disease (Not CME Accredited Session) (ID 948)

    • Event: WCLC 2018
    • Type: Poster Viewing in the Exhibit Hall
    • Track:
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/24/2018, 16:45 - 18:00, Exhibit Hall
    • +

      P1.16-19 - Neither Maximum Tumor Size nor Solid Component Size Was the Best Prognosticator for Subsolid Nodule (ID 13662)

      16:45 - 18:00  |  Author(s): Chang Chen

      • Abstract

      Background

      Solid component size is used to define the T stage of subsolid nodule in the eighth edition TNM stage classification. Our study aimed to explore whether solid component size was the best parameter for T staging.

      a9ded1e5ce5d75814730bb4caaf49419 Method

      We retrospectively reviewed the clinical data of 431 cTis-T3N0M0 subsolid nodule from Shanghai Pulmonary Hospital. Maximum tumor size, solid component size and tumor size in mediastinal window were carefully recorded. Prognostic ability of different turmor size was compared by time-dependent receiver operating curve.

      4c3880bb027f159e801041b1021e88e8 Result

      Survival revealed maximum tumor size, solid component size and tumor size in mediastinal window were statistical significant predictors. However, solid component size performed the worst of them, relatively.wclc 3.tiffwclc 2.jpg

      8eea62084ca7e541d918e823422bd82e Conclusion

      Tumor size remains a common used parameter for nodule evaluation. Solid component size maybe not the best parameter for T staging.

      6f8b794f3246b0c1e1780bb4d4d5dc53

  • +

    P1.17 - Treatment of Locoregional Disease - NSCLC (Not CME Accredited Session) (ID 949)

    • Event: WCLC 2018
    • Type: Poster Viewing in the Exhibit Hall
    • Track:
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/24/2018, 16:45 - 18:00, Exhibit Hall
    • +

      P1.17-18 - Treatment for Patients with T4 Superior Sulcus Non-Small Cell Lung Cancer: A Propensity-Matched Analysis of SEER Database (ID 12319)

      16:45 - 18:00  |  Author(s): Chang Chen

      • Abstract
      • Slides

      Background

      Superior sulcus tumors (SSTs), a unique subgroup of locally advanced non–small-cell lung carcinoma (NSCLC), remain a great challenge for clinicians. T4 SSTs used to be a contraindication for operations, and the optimal treatment modality for T4 SS NSCLCs remains uncertain. The aim of our study is to evaluate the roles of surgical treatment and radiotherapy for patients with T4 SSTs.

      a9ded1e5ce5d75814730bb4caaf49419 Method

      We used the SEER database [1973-2015] to identify patients diagnosed with T4 stage SS NSCLC (according to 7th edition AJCC staging system) between 2004 and 2015, those with M1 disease were excluded. Propensity score matching with Kaplan-Meier and Cox proportional hazards model were performed to estimate prognosis.

      4c3880bb027f159e801041b1021e88e8 Result

      A total of 384 patients were included (mean 66.4±11.71 years-of-age). Among them, the majority was male (59.4%) with lesions located in the left lung (52.3%) and diagnosed with IIIB stage (56.6%). 47 patients underwent cancer-directed surgery, and radiotherapy was received by 66.9% of patients. Median overall survival (OS) and lung cancer specific survival (LCSS) was 12 and 17 months, and 5-year OS, LCSS was 15.8%, 25.4%, respectively. In the matched population, the median survival outcomes were better with receipt of surgery (OS: 51.3 vs 35.1 months; p=0.049 LCSS: 67.1 vs 36.3 months; p=0.003). Multivariate Cox analysis showed that age ≧ 66 years (hazard ratio [HR] = 1.639, 95% confidence interval [CI] 1.214-2.213, p=0.001), unmarried status (HR = 1.356, 95% CI 1.023-1.798, p=0.034), tumor sized ≧ 6.0 cm (HR = 1.694, 95% CI 1.263-2.273, p<0.001) were associated with inferior OS. Cancer-directed surgery (HR = 0.537, 95% CI 0.337-0.855, p=0.009) and radiotherapy (HR = 0.644, 95% CI 0.472-0.878, p=0.006) were independent protective factors for patients with T4 superior sulcus NSCLC. However, neither adjuvant nor neoadjuvant radiotherapy was independent prognostic factor for those received surgery (p>0.05). Conversely, in the subgroup analysis, favorable impacts of radiotherapy were observed for non-surgery patients (OS: HR = 0.58, 95% CI 0.42-0.79, p<0.001; LCSS: HR = 0.55, 95% CI 0.37-0.75, p<0.001).

      8eea62084ca7e541d918e823422bd82e Conclusion

      Our study shows superior sulcus NSCLC patients with T4 stage have dismal prognosis. Surgical resection remains the optimal option for those with resectable disease. Moreover, for non-surgery tumors, the use of radiotherapy should be considered.

      6f8b794f3246b0c1e1780bb4d4d5dc53

      Only Active 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 or select "Add to Cart" and proceed to checkout.

  • +

    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
    • +

      P2.16-13 - A Proposal of Classification for Subsolid Nodule: Prognostic Impact of T Descriptor (ID 13377)

      16:45 - 18:00  |  Author(s): Chang Chen

      • Abstract

      Background

      Prognosis of lung cancer presenting as subsolid nodule is satisfying, and prognostic impact of tumor size for subsolid nodules is not so clear as pure-solid nodule. Our study aimed to find whether T staging is suitable for subsolid nodule.

      a9ded1e5ce5d75814730bb4caaf49419 Method

      We retrospectively reviewed 431 cTis-T3N0M0 subsolid nodules in our hospital. Maximum tumor size, solid component size, consolidation/tumor ratio (CTR) were measured.

      4c3880bb027f159e801041b1021e88e8 Result

      Five year recurrence-free survival (RFS) and overall survival (OS) were 94.2% and 97.0%. Nodules with maximum tumor size < 2 cm & CTR ≤ 0.5, maximum tumor size < 2 cm & CTR > 0.5 and maximum tumor size > 2 cm & CTR ≤ 0.5 were defined as low-risk group. Nodules with maximum tumor size > 2 and CTR > 0.5 were defined as high-risk group. Five year RFS and OS were both 99.0% for low-risk group, 81.0% and 91.4% for high-risk group. Maximum tumor size, solid component size and CTR were not the prognosticator for low-risk nodules but were significant for high-risk nodules.wclc.jpg

      8eea62084ca7e541d918e823422bd82e Conclusion

      We suggest using T descriptor only for clinical high-risk subsolid nodule. Prognosis of low-risk nodules is excellent, no clear relationship with tumor size or CTR.

      6f8b794f3246b0c1e1780bb4d4d5dc53