Article Text
PDF
Abstract
Background Determination of HER2 protein expression by immunohistochemistry (IHC) and genomic status by fluorescent in situ hybridisation (FISH) are important in identifying a subset of high HER2-expressing gastric cancers that might respond to trastuzumab. Although FISH is considered the standard for determination of HER2 genomic status, brightfield ISH is being increasingly recognised as a viable alternative. Also, the impact of HER2 protein expression/genomic heterogeneity on the accuracy of HER2 testing has not been well studied in the context of gastric biopsy samples.
Aims/methods To study the utility of brightfield ISH in the evaluation of HER2 genomic status, the correlation coefficient between dual-colour HER2/Chromosome 17 chromogenic in situ hybridisation (CISH) and FISH was ascertained. To study the impact of intratumoral heterogeneity on the accuracy of HER2 testing, the concordance rate of HER2 protein expression/genomic status between matched biopsies and surgical resection specimens of high HER2-expressing gastric cancers was ascertained.
Results The dual-colour CISH assay showed a 100% concordance rate with FISH results in 119 samples (Pearson correlation coefficient 0.987, p<0.001). Five of the 11 high-HER2 expressors (defined as IHC 3+ or IHC 2+/FISH-amplified according to Trastuzumab for Gastric Cancer trial criteria) showed an IHC 3+ score on matched biopsies (concordance rate 45.5%). Nine of these 11 cases showed HER2 amplification on matched biopsies (concordance rate 81.8%).
Conclusion Dual-colour CISH is an excellent alternative for the evaluation of HER2 genomic status in gastric cancers. Determination of HER2 status by HER2 IHC alone in limited gastric biopsy samples results in a high false-negative rate, and diagnostic accuracy appears to be improved if HER2 genomic testing, either alone or concurrently with IHC, is performed for HER2 testing.
- Gastric cancer
- HER2
- fluorescent in situ hybridisation
- brightfield
- dual-colour
- chromogenic in situ hybridisation
- molecular pathology
- molecular oncology
- cancer research
- gastrointestinal disease
- molecular genetics
- colon
- haematopathology
Statistics from Altmetric.com
- Gastric cancer
- HER2
- fluorescent in situ hybridisation
- brightfield
- dual-colour
- chromogenic in situ hybridisation
- molecular pathology
- molecular oncology
- cancer research
- gastrointestinal disease
- molecular genetics
- colon
- haematopathology
Introduction
The recent ToGA (Trastuzumab for Gastric Cancer) trial showed that addition of trastuzumab to chemotherapy improved survival in patients with advanced HER2-positive gastric cancer. This survival benefit was most evident in patients with high expression of HER2, defined as immunohistochemistry (IHC) 2+ and fluorescent in situ hybridisation (FISH) positive, or IHC 3+.1 Determination of HER2 genomic status thus appears to be an important component of the treatment algorithm for trastuzumab use in gastric cancer, similar to that in breast cancer.2
Our group has been interested in studying the utility of brightfield microscopy to evaluate HER2 genomic status in gastric cancer. Previously, we reported that single-colour chromogenic in situ hybridisation (CISH) was comparable with FISH in the examination of HER2 genomic status in gastric cancers using a tissue microarray (TMA) platform.3 We hypothesise that a dual-colour HER2/Chromosome 17 (Chr17) centromere CISH assay should increase the robustness of the brightfield microscopy analysis compared with a monochrome, HER2-only probe. In this study we sought to determine the concordance rate between a dual-colour CISH assay and FISH, the accepted standard of HER2 genomic analysis to date.
Another important issue that has not been addressed in detail is the impact of intratumoral heterogeneity on accurate HER2 testing in gastric cancers in the context of gastric biopsies. For breast cancer, ASCO-CAP guidelines recommend using either IHC assays for initial evaluation of HER2 status, followed by reflex FISH testing of equivocal IHC categories, or initial FISH testing.2 HER2 IHC expression in gastric cancer is known to be heterogeneous,4 and a certain false-negative rate would be predicted if the HER2 status is determined by IHC alone on gastric biopsy samples. Based on the limited number of studies thus far, HER2 genomic amplification appears to be more homogeneous.3 5 We therefore also hypothesised that the analysis of HER2 genomic status, either alone or concurrently with IHC, might be more reliable in determining accurate HER2 status than IHC alone in gastric biopsy samples. To ascertain this, we studied the concordance rate of HER2 protein expression/genomic status between matched biopsies and surgical resection specimens of HER2-amplified gastric cancer cases.
Materials and methods
The study population, TMA construction, IHC and FISH were as previously described.3 Briefly, IHC was performed using the automated slide stainer, BenchMark XT (Ventana Medical Systems, Roche Diagnostics, Arizona, USA), anti-HER2 antibody (clone 4B5) (Ventana Medical Systems) and ultraView™ DAB detection kit (Ventana Medical Systems). The IHC scoring criteria were: 0, no reactivity, or membranous reactivity in <10% of cells; 1+, faint/barely perceptible membranous reactivity in >10% of cells—cells are reactive only in part of their membrane; 2+, weak to moderate complete or basolateral membranous reactivity in >10% of tumour cells; 3+, moderate to strong complete or basolateral membranous reactivity in >10% of tumour cells.
FISH was performed according to the PathVysion HER2 DNA Probe Kit and Paraffin Pretreatment Kit II (Vysis, Abbott Molecular, Illinois, USA). A HER2/CEP-17 ratio greater than 2.2 and less than 1.8 indicated amplification and non-amplification, respectively. A ratio between 1.8 and 2.2 was interpreted as borderline.
In our previous study, 128 samples could be analysed by both IHC and FISH. Fifteen cases displayed HER2 genomic amplification. Of these 15 HER2-amplified cases, 12 displayed HER2 IHC 3+, one displayed IHC 2+, and two displayed IHC 1+.3
In addition, 12 matched biopsy specimens of HER2-amplified gastric cancer cases were retrieved and studied by IHC, FISH and dual-colour HER2/Chr17 CISH. Ethics approval was obtained for this study (NUS-IRB 06-063).
HER2 dual-colour CISH assay
Four-micrometre sections were stained using the Ventana BenchMark XT (Ventana Medical Systems) using a fully automated INFORM HER2 Dual ISH DNA Probe Cocktail assay (Ventana Medical Systems). The sections were deparaffinised and pretreated with CC2 buffer (Ventana Medical Systems) at pH 6.0 using high heat, followed by enzyme digestion of protein using ISH Protease 2 (Ventana Medical Systems) for 8 min. A cocktail of dinitrophenyl (DNP)-labelled HER2 DNA probe and digoxigenin (DIG)-labelled Chr17 centromere probe (Ventana Medical Systems) was then added to the sections for hybridisation for 6 h.
The HER2 DNP-labelled probe was detected using the ultraView SISH DNP detection kit (Ventana Medical Systems) and visualised as a black signal. Following development of the SISH signal, the Chr17 probe was detected using the ultraView Red ISH DIG Detection Kit (Ventana Medical Systems) and visualised as a red signal. Slides were then counterstained with Haematoxylin II (Ventana Medical Systems) and Bluing reagent (Ventana Medical Systems).
The HER2 and Chr17 copy numbers were counted in 20 nuclei independently by two observers (BY and BP). The Pearson correlation coefficient was determined for both HER2/Chr17 scores; as the scores were significantly correlated, further analyses utilised the scores from one observer (BY). HER2 genomic status was defined as non-amplified if the mean HER2/Chr17 ratio was <2.0 and amplified if the HER2/Chr17 ratio was ≥2.0. This criterion is similar to that reported in the ToGA trial.1
Data analysis
The correlation between the mean HER2/CEP-17 scores by FISH and mean HER2/Chr17 scores by dual-colour CISH was analysed by scatter-plot format, and statistical significance was ascertained using the Pearson correlation coefficient. The correlation of HER2-status between biopsy and surgical resection specimens was evaluated using the Pearson χ2 coefficient.
Results
Our IHC and FISH results were available from the previous study.3 In this current analysis, only 119 out of 128 samples contained sufficient tumour to allow accurate assessment of HER2 status. The dual-colour HER2/Chr17 CISH assay gave optimal signals in 115 cores (figure 1); the remaining four samples were successfully analysed on full sections. There was a significant correlation between the scores of the two observers (Pearson correlation coefficient 0.89, p=0.01), and hence subsequent analyses utilised the score from one observer as described in the previous section. There was a significant correlation between the FISH and dual-colour HER2/Chr17 CISH results (Pearson correlation coefficient 0.987, p<0.001) (table 1, figure 2).
Representative examples of dual-color HER2/Chr17 CISH, FISH and IHC in gastric cancer. A–C, Sample with HER2 genomic non-amplification and IHC score 0; D–F, Sample with HER2 genomic amplification and IHC score 3+. Black and red signals in dual-color CISH represent HER2 and Chr17 respectively; orange and green signals in FISH represent HER2 and CEP-17 respectively. Original magnification (A, B, D, E: 1000×, C, F: 400×).
Concordance between HER2 fluorescence in situ hybridisation and dual-colour chromogenic in situ hybridisation results
Scatter-plot of correlation between FISH and dual-color CISH HER2/Chr17 scores.
Of the 15 HER2-amplified gastric cancer surgical resection specimens, 12 matched biopsies could be retrieved. The HER2 protein expression/genomic statuses of the matched biopsy samples and surgical resection specimens are shown in tables 2, 3. Five of the 11 cases defined as high-HER2 expressors according to the ToGA trial criteria (IHC 3+ or IHC 2+/FISH-amplified) showed an IHC 3+ score on matched biopsies (concordance rate 45.5%). Nine of these 11 cases showed HER2 amplification on matched biopsies (concordance rate 81.8%) (table 2). There was no significant correlation of the HER2-status between biopsy and surgical resection specimens (p=0.1) (table 3).
HER2 genomic and protein expression statuses of matched biopsies and surgical resection specimens
Concordance of HER2 status* between biopsy and surgical resection specimens
The two cases that did not show HER2 amplification on matched biopsies were analysed for HER2 genomic status on full sections of the surgical resection specimens. HER2 amplification was present in the majority of tumour cells (>90%) for both cases, but a minor population (<10%) did not display HER2 amplification. Intratumoral heterogeneity thus accounted for the discrepant findings between the matched gastric biopsies and surgical resection specimens.
Discussion
Although FISH remains the gold standard for assessing HER2 genomic amplification at present, the merits of brightfield ISH such as ease of use and longer-lasting records are well known,6 and many investigators have reported brightfield ISH as a highly comparable alternative to FISH (reviewed by Gruver et al7). Furthermore, brightfield ISH eliminates the requirement for, and attendant costs associated with, fluorescent microscopy, and thus appears to be a more cost-effective option. Our results suggest that the dual-colour HER2/Chr17 CISH assay is an excellent alternative for the evaluation of HER2 genomic status in gastric cancers, having a perfect correlation with the HER2 FISH results. Furthermore, the dual-colour HER2/Chr17 CISH assay also allowed for direct correlation of HER2 genomic status with histomorphological features, which enabled accurate determination of intratumoral genomic heterogeneity in two cases subjected to full section analysis. In our experience with gastric cancer HER2 testing, the major advantage of CISH over FISH lies primarily in morphological evaluation, particularly of small-volume tumours—definite localisation of neoplastic cells is relatively straightforward by CISH and exceedingly difficult by FISH.
The ASCO-CAP guidelines allow for IHC in initial evaluation of HER2 status followed by reflex FISH testing of equivocal IHC categories in breast cancers.2 In view of the known heterogeneous IHC staining pattern in gastric cancers,4 the analysis of HER2 status by IHC alone on biopsy material in the diagnostic algorithm will be challenging, given a predicted significant false-negative rate. In our series, the concordance rate between biopsies analysed by IHC and surgical resection specimens of 11 high HER2-expressors (defined as IHC 3+ or IHC 2+/FISH-amplified cases in the ToGA trial) was 45.5%. This implies that if IHC was employed as the primary triaging assay in the diagnostic workup of limited gastric biopsy samples, at least 50% of patients would be falsely classified as HER2-negative and would be denied standard of care treatment. If HER2 genomic status is analysed simultaneously in gastric biopsy samples, the concordance rate increases to 81.8% (9/11).
Although our biopsy series is small, our results suggest that clinicians should be cautious when interpreting negative HER2 IHC results obtained from limited gastric biopsy samples and that analysis of HER2 genomic status (either alone or concurrently with IHC) appears to increase the accuracy of HER2 testing on such biopsy samples. Besides validating these findings in larger series, future studies should also investigate if increasing the number of biopsies decreases the false-negative rate and, if so, what number of biopsies required for optimal evaluation is.
Take-home messages
Dual-colour HER2/Chr17 CISH is an excellent alternative to FISH for evaluation of HER2 genomic status in gastric-cancer specimens.
Determination of HER2 status by HER2 IHC alone in limited gastric biopsy samples results in a high false-negative rate, and diagnostic accuracy appears to be improved if HER2 genomic testing, either alone or concurrently with IHC, is performed for HER2 testing.
Interactive multiple choice questions
This JCP article has an accompanying set of multiple choice questions (MCQs). To access the questions, click on BMJ Learning: take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to: http://jcp.bmj.com/education. Please note: the MCQs are hosted on BMJ Learning the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into JCP with their journal's username and password. All users must also complete a onetime registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.
Footnotes
Competing interests MS-T is a member of the Key Opinion Leaders Group of Ventana-Roche.
Ethics approval Ethics approval was obtained from the relevant Institutional Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.