Background
Despite advances in the understanding of the disease and new treatment strategies, stage IV GC remains a relevant health issue worldwide. So far, treatment in stage IV GC has been limited to platinum-based chemotherapy. Furthermore, in the case of HER2 expression, the HER2-directed monoclonal antibody trastuzumab is an approved targeted drug for first line treatment. According to the pivotal randomized-controlled phase III registration study, trastuzumab is improving overall survival from 11.1 to 13.8 months. Unfortunately not all patients respond and almost all initial responders eventually develop resistance and experience tumor progression.
The prospective multicenter VARIANZ study aimed to identify resistance biomarkers for HER2-targeted treatment in advanced gastric and esophagogastric junction cancer. More than 500 patients receiving medical treatment for stage IV GC were recruited in 35 German sites and followed for up to 48 months. HER2 status was assessed centrally by immunohistochemistry (IHC) and chromogenic-in-situ-hybridization (ISH). Within the study a HER2 test deviation rate (22.3%) between central and local test was associated with negative impact on patient survival. Patients who received trastuzumab with centrally confirmed HER2+ status (central HER2+/local HER2+) lived significantly longer compared to patients who received trastuzumab for local HER2+ but central HER2 stage IV GC (20.5 months vs. 10.9 months, 95% CI [8.2, 14.4], p<0.001) [1].
In the present analysis, we investigated methodological and biological variables that may promote deviating HER2 test results.
Methods
We analyzed HER2 testing procedures and participation in quality assurance programs of 105 participating local pathology laboratories.
Furthermore, tumor localization, histological subtypes and tested tumor material (for local and central test) were compared between patients with centrally confirmed (central HER2+/local HER2+, n=68) and unconfirmed HER2 status (central HER2-/local HER2+, n=68).
Results
Central confirmation of the local HER2 IHC scores were seen for the majority of locally HER2- IHC 0/1 (172/178; 96.6%), but less frequently for locally IHC3+ (57/124; 46.0%) cases.
Neither use of specific IHC methods nor participation in round robin tests varied between cohorts with confirmed (central HER2+/local HER2+, n=68) or deviating (central HER2-/local HER2+, n=68) HER2 test results.
We found seven IHC antibodies (HercepTest, 4B5, SP3, CB11, UMAB36, A0485, EP3) in routine use in Germany, with HercepTest and 4B5 being the most commonly used.
46 (43.8%) local pathology laboratories participated in quality assurance programs for HER2 testing in GC.
The distribution of the tested tumor material (primary tumor biopsy, surgical specimen, or biopsy from metastasis), related to both local and central HER2 testing, was comparable between the cohorts studied.
Regarding tumor characteristics, deviating test results were more frequently found in GC vs. EGJC (69.1% vs. 39.7%; p=0.001). Within the Laurén histological classification deviating status (central HER2-/local HER2+) was more often found in diffuse type GC (23.5% vs. 5.9% for confirmed HER2+, p=0.004).
Conclusion and Interpretation
VARIANZ study has shown that GC patients with deviating HER2 test results had poor trastuzumab benefit [1], so we dedicated this work to the important question of the underlying causes of HER2 test deviations in GC.
Our analysis demonstrates that neither the antibody platform used for IHC nor participation in round robin tests of local pathology institutes correlated with the deviation rate for HER2 test results. In contrast, we found that tumor characteristics such as primary tumor location and histological phenotype had an impact on test deviations: more HER2 test deviations were seen in distal GC compared to EGJC as well as in the diffuse versus intestinal subtype according to Laurén’s classification.
One main limitation of our study is the so far barely used IHC HER2-antibody CB11 for central testing. However, in our study its highly specific properties [2–4] enabled it to identify patients who benefited from trastuzumab.
In contrast to breast cancer HER2 is very heterogeneously expressed in GC [5, 6]. Due to this different expression, the HER2 testing scheme was adapted for GC [7]. Nevertheless, the success of HER2-targeted therapy in GC lags significantly behind breast cancer therapy [5].
HER2 heterogeneity is a known issue in GC, even in early stages [8]. Although much has been published [6, 9–11] , there is no generally agreed definition for HER2 heterogeneity or diagnostic procedure to identify HER2 heterogeneity. There are different approaches, for example using the relative number of HER2+ stained tumor
cells [10, 12] or the deviation in HER2 status in a set of primary biopsies [13]. HER2 heterogeneity has already been associated with limited trastuzumab benefit and decreased overall survival in trastuzumab treated patients [1, 13, 14].
We conclude that the central confirmation of the HER2 status is a correlate of lower HER2 heterogeneity and may serve as an indicator for better treatment efficacy of trastuzumab.
Further we assume that for distal GC location and for the diffuse subtype where HER2 positivity in general is less common [2, 15–18] weak HER2 expression and intratumoral heterogeneity account for more deviating test results.
In our view, HER2 diagnostic scheme for GC should be adapted. Adaption of HER2 thresholds to identify patients benefiting from trastuzumab treatment has been already postulated [1, 17]. Our data suggests that only patients with low HER2 heterogeneity may benefit from trastuzumab treatment. For the selection of patients with low HER2 heterogeneity, the above tumor characteristics, such as tumor localization and histological subtype, should be reported. EGJC and the intestinal subtype are positive indicators of the presence of low HER2 heterogeneity. The use of highly specific IHC antibodies should be preferred. Furthermore, a positive result in IHC should always be confirmed by the examination of a paired specimen and the percentage of positive stained tumor cells should be reported. This might improve survival outcomes with targeted treatment, prevent overtreatment and associated side effects and costs and may enable successful studies of other promising HER2 targeting drugs.
| Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:90346 |
| Date | 06 March 2024 |
| Creators | Kolbe, Katharina |
| Contributors | Universität Leipzig |
| Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
| Language | English |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, doc-type:doctoralThesis, info:eu-repo/semantics/doctoralThesis, doc-type:Text |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0032 seconds