Author of

• 11866

  +

  P2.18 - Poster Session 2 - Pathology (ID 176)

  • Event: WCLC 2013
  • Type: Poster Session
  • Track: Pathology
  • Presentations: 2
  • Moderators:
  • Coordinates: 10/29/2013, 09:30 - 16:30, Exhibit Hall, Ground Level

  • 11885

    +

    P2.18-019 - Clinicopathological features in non-small cell lung cancer patients with EGFR and KRAS mutations. (ID 2985)

    09:30 - 09:30  |  Author(s): N. Imamura
    • Abstract

    Background
    Some of molecular pathways have been shown to have prognostic impact in non-small cell lung cancer (NSCLC). Epidermal growth factor receptor (EGFR) mutations predict the effect of EGFR tyrosine kinase inhibitors. KRAS is also critical oncogene, and it has been reported that KRAS pathway might interact with EGFR. But, the role of KRAS in NSCLC is unclear. We investigated the relationship between EGFR and KRAS mutation status and clinicopathological features in NSCLC.

    Methods
    A total of 383 consecutive patients with NSCLC underwent complete resection from 2006 to 2008 were examined retrospectively. The expression of EGFR and KRAS were evaluated by tissue microarray.

    Results
    The mutations of EGFR and KRAS were detected in 181/383 (47.3%) and 32/383 (8.4%) patients, respectively. On analysis of EGFR mutations, female were 107/181 (59.1%) and 51/202 (25.2%), adenocarcinoma were 177/181 (97.8%) and 123/202 (60.9%), no vascular invasion were 147/181 (81.2%) and 110/202 (54.5%), and non-smoker were 99/181 (54.7%) and 41/202 (20.3%) patients in EGFR mutation and wild type patients, respectively. As a result, EGFR mutation was found more frequently in female, adenocarcinoma, no vascular invasion, and non-smoker. The number of patients with pathological T1a were 49/181 (27.0%) and 42/202 (20.8%), T1b were 63/181 (34.8%) and 41/202 (20.3%) in EGFR mutation and wild type patients, respectively. Moreover, average tumor diameter was smaller in patients with EGFR mutation (2.68 cm±0.92) than wild type (3.34cm±1.70) (P<0.001). There were no differences in clinicopathological characteristics between exon19 and 21 EGFR mutations. In contrast, there were no significant differences between KRAS mutation and gender, histopathological type, vascular invasion and.smoking. Although KRAS status was not correlated with pathological T factors, average tumor diameter was larger in patients with KRAS mutation (3.49 cm±2.00) than wild type (2.98 cm±1.35) (P<0.001).

    Conclusion
    Our results suggest that EGFR mutation may suppress vascular invasion, and tumor growth, on the other hand, KRAS mutation may correlate with activation of tumor growth.

  • 11886

    +

    P2.18-020 - Appropriateness Evaluation of Handling Method for the Small Lung Adenocarcinoma in the Frozen Section Diagnosis by Radiological-Pathological Correlation (ID 3007)

    09:30 - 09:30  |  Author(s): N. Imamura
    • Abstract

    Background
    The frozen section diagnosis is often performed in the sublobar resection of lung tumor. As no standard of preparation method for the frozen section is proposed, its methodology differs depending on institutions. In this study, we examine appropriateness of our preparation method for a resected specimen with a small adenocarcinoma by comparing between radiological and pathological tumor size.

    Methods
    We retrospectively reviewed the records of 59 resected lung specimens for the frozen section diagnoses (54 wedges and 5 segmentectomies) of lung adenocarcninomas from January to December 2008. After the specimen was well inflated with saline using injector, the pathologist cut it into segments with a width of 3-5mm and immersed them in saline. Taking the segment with maximum diameter of tumor as a sample, the pathological tumor sizes were measured (I) macroscopically by using metal straight ruler, (II) microscopically on the frozen section, and (III) microscopically on the permanent paraffin section. For obtaining the stereoscopic tumor size (Ⅱ and Ⅲ), we used a stereoscopic image analysis software, Leica Application Suite (Leica Microsystems; Tokyo, Japan). CT tumor size was measured by using 1-2mm thin-section CT (X-Vigor/Real or Aquillion, Toshiba Medical Systems, Tokyo, Japan). We obtained the tumor shadow disappearance rate (TDR) by comparing tumor size on the lung and mediastinal window image, to classify 59 cases into two groups according to TDR; TDR≧50% defined as the air-containing type (Group A, n=44) and TDR＜50% as the solid-containing type (Group S, n=15). We also calculated the diremption rate (DR%) between the pathological and the CT tumor size (DR% = |CT tumor size － each pathological tumor size|/CT tumor size×100(%)) and compared Group A and Group S.

    Results
    Mean CT tumor size and its standard deviation(SD) were 18.36±5.23mm, and mean pathological tumor sizes and SD of Ⅰ, Ⅱ, and Ⅲ were 17.17±6.12, 14.29±3.66, and 14.23±4.38mm, respectively. Mean CT tumor size was statistically larger than that of Ⅱ and Ⅲ (p<0.001 using Paired t-test). All the three pathological tumor sizes were correlated to the CT tumor size by Pearson’s correlation analysis (correlation coefficient were 0.766, 0.700, and 0.682, respectively). DR% of Ⅱ and Ⅲ were significantly higher in Group A than Group S by Mann-Whitney U-test (Mean DR% of group A / S (p-values) of Ⅰ, Ⅱ, and Ⅲ were 17.0/13.8% (p=0.196), 25.8/19.3% (p=0.093), and 27.3/15.5% (p=0.032) , respectively).

    Conclusion
    There was a strong correlation between CT tumor size and each pathological tumor size, which shows that our preparation method of the specimen for the frozen section is appropriate to obtain sufficient information about the lung tumor. Furthermore, we found that the pathological tumor size is considerably underestimated by measuring tumor size on the frozen or permanent paraffin section, especially the tumor classified as “air-containing type” including adenocarcinoma with good prognosis. It is therefore important to inflate the lung specimen sufficiently and to transfer it to microscopical examination without tissue shrinking.