In this study, 143 of 1463 (9.8%) of GC and GEJ adenocarcinoma cases were HER2-positive (3+) by IHC in one of the largest Chinese studies to date. In the ToGA trial, the percentage of HER2-positive (IHC 3+ or IHC 2+/FISH positive) GC or GEJ cancer patients was 22.1% overall and around 10.4% of IHC 3+ in resected samples , similar to the present result. Recent studies also reported a range of 8.5% to 10.3% for HER2 overexpression in GC [20, 21, 24]. Recently, three studies in Chinese GC cases applied the same FDA-approved reagents and scoring criteria and reported HER2 IHC 3+ rates of 9.0% (77/860) , 6.9% (10/145) , and 5.8% (4/69) , respectively, all slightly lower than the present result. This variation may be partly explained by different sample sets. In addition, HER2-positivity varied by tumor site, with higher rates of HER2-positivity in GEJ adenocarcinoma than in stomach cancer in this study (14.6% vs. 7.0% respectively; P<0.01), which is consistent with the results of other studies [11, 20]. In the ToGA trial, the countries with the highest ratio of GEJ:stomach cancer were found to have above-average HER2-positivity rates, regardless of sample size . Another explanation is the different ratio of intestinal:diffuse/mixed cancer among the studies. A positive association between HER2-positivity and intestinal-type cancers was identified in this and other studies [28–33]. In the ToGA trial, for example, countries with higher ratios of intestinal:diffuse/mixed cancer had increased HER2-positivity rates .
No correlation was found between HER2 overexpression and pT, N, or M factors, or TNM stages, in the present study. Previous studies including all pathological stages have also reported no correlation between pathological stage and overexpression [20, 21, 24]. A correlation of HER2-positivity with well or moderately differentiated carcinomas was found in the present study (P<0.01). However, other studies demonstrated both an association and no association between HER2 overexpression and tumor differentiation [25, 32, 34, 35]. These conflicting data may be due to different sample sizes and the low prevalence of HER2 in GC and GEJ adenocarcinoma. In addition, varying methods of evaluation and scoring schemes with different cut-points were used before the establishment of a standard guideline for assessing HER2 expression. Perhaps with the current consistent guidelines for HER2 assessment in this disease, future studies will provide more clarity regarding this issue.
The present study revealed a high concordance rate, 98.5%, between IHC and FISH, which is similar to another study of surgical resections from 145 Chinese GC cases, with a concordance rate of 94.5% . In contrast, in the ToGA study, the concordance rate of IHC and FISH was 87.2%, as 7.5% of IHC 1+/0 samples and 54.6% of IHC 2+ cases were found to be FISH-positive . One of the possible explanations for this discrepancy might be different specimen types and reagents. In the ToGA study, both biopsies and surgically-resected specimens were included; the HER2-positivity rate was higher in biopsies than in surgically-resected specimens (23.1% vs. 19.9%; P=0.03), and biopsy samples were also more likely to be HER2-amplified than surgical samples when analyzed by FISH (P=0.01) than by IHC . The majority of current studies using a standard guideline for HER2 expression assessment used biopsies and tissue microarrays (TMAs), which were tested analogous to biopsies, to interrogate HER status in GC. The concordance rate of IHC and FISH in these studies varied from 88.6% to 97.7% [20, 32, 36]. A specific study evaluating HER2 protein expression on whole-tissue sections and TMAs was conducted, indicating a discrepancy in HER2 expression on whole-tissue sections from TMAs. For instance, the proportion of IHC 2+ was 4.7% in whole-tissue sections, whereas it was 8.6% in TMAs, which might be partially explained by heterogeneous HER2 expression . Another possibility, which was revealed by a study aiming to validate the guideline for HER2-expression assessment in GC, was different interpretations of staining of TMAs among observers, although a uniform assessment guideline was followed . Therefore, reliable separation of IHC 1+/0 and IHC 2+ may be difficult in biopsy samples, and FISH analysis should be used for definitive classification.
In conclusion, a thorough analysis of 1463 Chinese GC/GEJ cancer cases using FDA-approved reagents showed HER2 overexpression in 9.8% of carcinomas, with a strong preference for GEJ location, intestinal cancer subtype, and well or moderately differentiated carcinomas. Approximately 29% of IHC 2+ cases showed HER2 gene amplification, and there was a high concordance rate (98.5%) between IHC and FISH in GC and GEJ adenocarcinoma.