Radiologic-Pathologic Correlation before Signout Significantly Reduces Overdiagnosis of Pulmonary Adenocarcinoma In Situ and Minimally Invasive Adenocarcinoma in Surgically Resected Lung Nodules
Saul Harari, Jane Ko, Harvey Pass, David Naidich, James Suh. NYU Langone Medical Center, New York, NY
Background: The 2011 IASLC/ATS/ERS multidisciplinary classification of lung adenocarcinoma defined the entities of adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA) as early-stage tumors that should have 100% and near 100% 5-year disease-specific survivals, respectively. However, it can be challenging to distinguish between areas of lepidic (in situ) growth and invasion even on permanent slides. Our aim is to investigate whether radiologic-pathologic correlation is good clinical practice for diagnosis of AIS and MIA.
Design: The radiologic and pathologic characteristics of 103 surgically resected lung adenocarcinomas at one institution (2006-10) were reviewed independently. Each tumor was re-classified using the 2011 terminology while blinded to imaging studies. All cases of possible AIS and MIA with radiologic-pathologic discrepancies underwent additional review. Next, all cases of surgically resected AIS and MIA at the same institution (2011-12) that were diagnosed following correlation with radiologic findings were collected prospectively and reviewed for discrepancies. The numbers of cases that required a change in diagnosis from AIS and MIA between the two sets were compared.
Results: 11 possible cases of AIS and 19 possible cases of MIA were identified on retrospective review, of which 4 AIS and 13 MIA were analyzed for discrepancies. 3 cases of AIS were changed to acinar predominant adenocarcinoma (APA), 3 cases of MIA were changed to lepidic predominant adenocarcinoma (LPA) and 3 cases of MIA were changed to APA. Overall, 21/30 (70%) cases were confirmed: 8/11 (73%) AIS and 13/19 (68%) MIA. In contrast, 6 cases of AIS and 6 cases of MIA were identified prospectively, of which 1 AIS and 2 MIA were analyzed for discrepancies. All 12 cases (100%) were confirmed. The difference in the numbers of cases requiring a change in diagnosis to more invasive adenocarcinomas (LPA or APA) between the two study groups was statistically significant (p-value = 0.0413).
Conclusions: Review of the CT appearance of lung nodules before signout reduces overdiagnosis of AIS and MIA in surgical resections by nearly one-third. Although it requires additional time and effort to access imaging studies and reports, radiologic-pathologic correlation should be performed when considering a final pathologic diagnosis of either tumor.
Monday, March 4, 2013 1:00 PM
Poster Session II # 291, Monday Afternoon