Study design
VARIANZ is a prospective, non-interventional multicenter study which recruited 548 patients at 35 German sites from 2014 to 2018. VARIANZ was funded by the German Federal Ministry of Education and Research (grant number: BMBF 01ZX1610E). Main results were published previously (Haffner et al. 2021). Briefly, adult patients receiving first-line chemotherapy with or without trastuzumab for histologically confirmed advanced GC or EGJC were recruited. The primary endpoint was survival (OS), defined as time from the beginning of first-line chemotherapy until death from disease related cause. VARIANZ was conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki. All patients had given written informed consent. Approvals for the study protocol were obtained from ethics committees of all study sites.
HER2 testing
The first HER2 diagnostic test was performed in local pathology institutes to determine the HER2 status and decide for the indication for trastuzumab therapy. The same or another tumor block was sent to the central pathology institute (Leipzig University Medical Center, Department of Pathology, Leipzig, Germany). HER2 expression was assessed centrally on fresh cut slides of the provided tumor block originating from primary tumor biopsies, surgical specimen or metastatic lesions using IHC (CB11 antibody, DCS, HI608C0I, Hamburg, Germany) and chromogenic in-situ-hybridization (ISH) (Zytomed Systems, C-3022-40, Berlin, Germany) according to established scoring criteria (Hofmann et al. 2008). The central laboratory did not receive the locally stained tissue sections. The IHC assessment was performed on fresh cut and newly stained slides without knowledge of the ISH results (and vice versa). ISH was performed and analyzed for all specimens regardless of the IHC test results. For IHC all tumor cells per slide were analyzed and for ISH 20 cells were scored. For borderline cases (HER2/CEP17-ratio of 1.8–2.2) 40 more cells were analyzed. For HER2 evaluation, done by two dedicated GI pathologists at the central study pathology institute, no technical support or image analysis system was used.
Survey on HER2 test methodology
Tumor specimens from 105 pathology institutes were sent to the central pathology. All participating pathologists were asked to complete a questionnaire containing information about the tumor tissue (biopsy versus resection specimen; primary tumor versus metastatic site), their participation in round robin tests and the applied HER2 test method. We asked specifically for the following antibodies for IHC: HercepTest and A0485 (Dako, Glostrup, Denmark), SP3 (Labvision; Thermo Fisher Scientific, Fremont, CA, United States), 4B5 (Ventana Medical Systems, Tucson, AZ, United States), CB11 (Novocastra, Newcastle upon Tyne, Great Britain), UMAB36 (OriGene Technologies, Rockville, United States) and EP3 (Cell Marque, Rocklin, United States).
Statistical methods
Stratification into groups was done according to central and local HER2 test results: HER2+/HER2+ (central HER2+/local HER2+), HER2+/HER2- (central HER2+/local HER2-), HER2-/HER2+ (central HER2-/local HER2+) and HER2-/HER2- (central HER2-/local HER2-). Comparative analysis such as sex, initial tumor stage, prior gastric surgery, tumor location and histopathological subtype according to Laurén’s classification of the different patient cohorts (HER2+/HER2 + vs. HER2-/HER2+) was carried out using the Chi² test. P < 0.05 was considered statistically significant. Analyses were done with IBM SPSS version 24.0 (IBM Corp., Armonk, NY).