ADVERTISEMENT
Evaluation of HER2 status in equivocal gastric cancer tissue samples using surrogate DNA methylation markers
Background
Her2 (ERBB2) proto-oncogene amplification and overexpression are observed in 20-30% of gastric cancer (GC) patients and play a crucial role in invasive GC development and treatment. To date, fluorescent in situ hybridization (FISH) has been the gold standard to identify HER2 amplification status in equivocal cases of GC (IHC2+). However, a certain group of GC patients cannot be evaluated by FISH because of poor tissue quality, which may be due to inappropriate sample handling. Besides, FISH is time-consuming, expensive, and technically demanding compared to other alternative technologies like IHC and qPCR. Additionally, FISH results can only be interpreted by well-trained personnel, which makes the FISH test only accessible to large laboratories/institutions. Therefore, it is essential to develop cost-effective alternative technologies to increase the capacities of evaluating HER2 amplification status in equivocal GC cases. In this study, we developed a methylation-specific quantitative real-time PCR (qMSP) assay using surrogate DNA methylation markers for the accurate evaluation of HER2 status in HER2 equivocal GC patients.
Methods
This multicenter, retrospective cohort study enrolled 292 patients (44 HER2 IHC 0/1+, 218 IHC 2+ with FISH results, and 30 HER2 IHC 3+ cases) from five hospitals in China. Resected or needle-biopsied tumor samples were fixed in formalin and embedded in paraffin following a standard procedure. Sixteen samples were excluded due to QC failures. 72 samples were analyzed by targeted bisulfite sequencing using an enriched panel with pre-selected gastric cancer-associated informative CpG sites. Top DNA methylation markers were selected for HER2 status stratification by comparing IHC 0/1+ to IHC 3+ samples and then qMSP assays were designed for these markers. Next, we tested the markers’ performances in 202 HER2 IHC2+ samples (110 FISH- and 92 FISH+) and developed diagnostic models by either logistic regression or ΔCt cut-off. qMSP assay performance was compared with previously reported performance of a standard Her2 copy number qPCR assay (targeting intron 14 of HER2).
Results
105 HER2-overexpression-specific methylation markers were identified by DNA methylation targeted sequencing. 69 of them were selected based on genomic region and performance. We have developed and validated two models for distinguishing HER2 status in HER2 IHC2+ samples (training set n = 98; validation set n = 104): 1) A 4-marker logistic regression model achieved a ROC-AUC of 0.95 in the validation set; 2) A 3-marker model using ΔCt cut-off modeling method achieved sensitivity of 94.5% and specificity of 92.9% in the validation set; The overall concordance between the ΔCt cut-off model and FISH results was 93.2%, compared to the standard HER2 CNV qPCR assay with 79.8% accuracy in this cohort. Furthermore, an integrated model comprising the methylation and standard CNV assays demonstrated improved accuracy of 96.2%.
Conclusions
We have developed a simple, sensitive, and cost-effective methylation-based qPCR assay for evaluating HER2 amplification status in gastric cancer, especially in the HER2 equivocal cases, which may complement the conventional FISH assay to increase the capacities of investigating HER2 status in clinical practice.
Legal entity responsible for the study
The author.
Funding
This study was supported by Scheme of Guangzhou Economic and Technological Development District for Leading Talents in Innovation and Entrepreneurship (Grant NO.2017-L152); Scheme of Guangzhou for Leading Talents in Innovation and Entrepreneurship (Grant NO.2016007); Scheme of Guangzhou for Leading Team in Innovation (Grant NO.201909010010); Science and Technology Planning Project of Guangdong Province, China (Grant NO.2017B020226005).
Disclosures
M. Bibikova: Shareholder / Stockholder / Stock options: AnchorDx, Illumina; Full / Part-time employment: AnchorDx, Illumina; Officer / Board of Directors: AnchorDx. Z. Chen: Full / Part-time employment: AnchorDx Medical Co., Ltd., AnchorDx, Inc. All other authors have declared no conflicts of interest.
