Reproducibility Assessment of Hormonal Receptor Status and HER2 in Cytology Specimens by Image Analysis: A Pilot Study
Sudeshna Bandyopadhyay, LaQuita King, Dongping Shi, Neelofur Nazeer, Tatyana Kalinicheva, Shashi Madan, Rouba Ali-Fehmi, Paul Tranchida, Pam Tabaczka, Vinod Shidham, Tamar Giorgadze. WSU/KCC/DMC, Detroit, MI
Background: Estrogen receptor (ER), Progesterone receptor (PR) and HER2 status evaluation in primary and metastatic breast cancer in cytology specimens has known prognostic and therapeutic implications. The aim of this study is to assess the concordance of ER/PR/Her2 results in cytology specimens by recently implemented in our laboratory Image Analyzer (Ventana, Tucson, AZ) and semiquantitative scoring by pathologists. To our knowledge, this is the first study of this kind.
Design: Twenty three consecutive ER/PR/HER2 immunostained cytology specimen cell block preparation slides from primary and metastatic breast carcinoma for the years 2009-2011 were retrieved, blindly reviewed by 5 cytopathologists, and scored per CAP/ASCO guidelines. The Image Analyzer (IA) results of these cases were retrieved and the scores compared with blind review (BR). ER/PR/HER2 status of metastatic tumors was also compared with that of primary tumor (PT) when available. Fluorescence in situ hybridization (FISH) results of equivocal HER2 cases were documented.
Results: Out of 23 cases (7 FNA from metastatic lesions and 16 effusion fluids), 1 was insufficient for analysis. Four cases were discordant between the IA and BR scores. Two cases showed discordance in the Her2 analysis: IA: 2+, BR: 1+. Her-2/neu was not amplified by FISH in both cases. In 2 other cases, IA picked up weak positivity in PR which was missed by BR.
The ER/PR/Her2 status of PT was available in 20 cases. Discordance between the PT and the IA scores was seen in 8 cases. In 4 cases, IA scored the Her2 as 2+, while the Her2 in PT was negative. Her2 was not amplified by FISH in these cases. In 2 cases of metastatic effusions, IA scored ER as positive and weakly positive, while in PT they were reported as negative. In 1 case of metastatic effusion, PR was scored as positive while PT was negative. Lastly, in 1 case of metastatic effusion, PR was scored as negative while PT was reported as positive.
Conclusions: IA appears to be more sensitive in assessing weak hormone receptor positivity than BR. One of the frequent discrepancies between both comparative groups (IA vs BR and IA versus PT) was the evaluation of Her2 score. It appeared to be overcalled by IA. Discrepant cases between IA and PT status may represent transformation of the receptor status, inadequate sampling, or limitations due to scant cellularity in cell blocks. Overall, IA appears to be an efficient method for evaluation of ER/PR/HER2 status in cytology specimens; however, larger studies should be done to assess accuracy of this method in cytology samples.
Tuesday, March 20, 2012 9:30 AM
Poster Session III # 81, Tuesday Morning