Gastric Biopsies Are Appropriate for Assessment of HER2: A Correlation Study with Resection Specimens
Andrea Grin, Eugene T Hsieh, Christine Brezden-Masley, Catherine J Streutker. St. Michael's Hospital, Toronto, Canada; Sunnybrook Health Sciences Centre, Toronto, Canada; University of Toronto, Toronto, Canada
Background: The survival benefit of trastuzumab in advanced HER2 positive gastric cancer has been demonstrated in a recent large prospective randomized multicentre trial (ToGA trial). Accurate assessment of HER2 status is essential to select patients who will benefit from this therapy. Biopsies are often tested, particularly in a neoadjuvant setting or when patients are inoperable, but little is known on the correlation of HER2 status in biopsy versus resection specimens. While biopsies may have more optimal tissue fixation than resections, gastric cancer has been shown to display significant tumor heterogeneity, which could lead to false negatives in small biopsy specimens. In this study, we sought to examine the diagnostic accuracy of HER2 testing in biopsy specimens.
Design: Forty patients with gastric carcinoma where both initial biopsy and subsequent resection specimen were available were evaluated for the expression of HER2 by immunohistochemistry (4B5, Ventana) by the criteria used in the ToGA trail. FISH (Abbott Molecular) was performed on cases equivocal (2+) by immunohistochemistry and on discrepant cases.
Results: The average number of biopsy fragments was 5.9 with an average of 3.2 fragments positive for carcinoma. Most tumors were intestinal (55%), 15% were diffuse and 30% mixed by the Lauren classification. Overall, 4 of 40 (10%) tumors were HER2 positive, defined as either 3+ by IHC or FISH HER2:CEP17 ratio ≥2.0. Seventy percent (28/40) of patient-paired biopsy and resection specimens showed concordant IHC scores. When 0+ and 1+ scores were grouped and defined as negative, 2+ as equivocal and 3+ as positive, concordance increased to 85% (34/40). Incorporating FISH results in equivocal cases further improved concordance to 95% (38/40). Two cases showed major discrepancies (positive vs. negative overall result), one of which would have resulted in a patient being denied trastuzumab if HER2 was evaluated on the biopsy alone.
Conclusions: Despite tumor heterogeneity in gastric cancer, biopsy specimens are appropriate for evaluation of HER2 when IHC is used along with FISH following the criteria used in the ToGA trial. In our study, there was excellent correlation for HER2 status between biopsy and resection specimens.
Tuesday, March 20, 2012 1:00 PM
Poster Session IV # 60, Tuesday Afternoon