Clinicopathologic association and prognostic value of MELF pattern in invasive endocervical adenocarcinoma (ECA) as Classified by IECC

Microcystic, elongated, and fragmented (MELF) pattern of myometrial invasion is correlated with lymphovascular invasion (LVI) and lymph node metastases in uterine endometrioid carcinoma but has not been described in endocervical adenocarcinoma (ECA). A total of 457 ECAs were collected, and clinical/morphologic parameters correlated with follow-up data. Potential associations between MELF pattern and age, human papillomavirus status, tumor size/grade, LVI, lymph node metastases, Silva pattern were analyzed. Statistical analyses of overall survival (OS), disease-free survival, progression-free survival (PFS) were conducted using Kaplan-Meier analysis, and compared using the Log-rank test. Of 292 ECAs analyzed, 94 (32.19%) showed MELF invasion pattern (MELF-positive). Significant statistical correlation was found between MELF-positive and tumor size (P=0.0017), LVI (P=0.007), Silva pattern (P=0.0005); age, human papillomavirus status, tumor grade, lymph node metastases did not correlate. Fifty-five of 292 patients recurred (18.83%): 18/94 (19.14%) MELF-positive, 37/198 (18.68%) MELF-negative. PFS in MELF-positive: 77.2% and 64.5% at 5 and 10 yr, respectively; PFS in MELF-negative: 82% and 68.5% at 5 and 10 yr, respectively. On multivariate analysis for PFS and other prognostic parameters, only LVI was statistically significant (P=0.001). OS in MELF-positive was 86% and 74.1% at 5 and 10 yr, respectively; OS in MELF-negative, was 89.7% and 86% at 5 and 10 yr, respectively. Median survival was worse in MELF-positive (199.8 mo) versus MELF-negative (226.1 mo); this was not statistically significant. On multivariate analysis for OS and other prognostic parameters, only tumor stage was statistically significant (P=0.002). In ECAs, MELF is not independently associated with survival. Pathologic characteristics of MELF-positive (size, LVI, Silva pattern) versus MELF-negative tumors differ significantly.