Background: Non-classical HER2 FISH results were recently reclassified in the 2018 HER2 guidelines update, and concurrent IHC testing was recommended as part of additional workup to determine the final HER2 status in these groups. In this study, we explored the genomic landscape of HER2 FISH groups using digital droplet PCR (ddPCR) and targeted next-generation sequencing (NGS) on invasive breast carcinomas.

Methods: Fifty-one clinical samples with HER2 FISH and IHC results were included in our analysis and classified into FISH groups based on the updated 2018 ASCO/CAP HER2 testing guideline: (i) Group 1A with ratio ≥2 and signals/cell ≥6, (ii) Group 1B with ratio ≥2 and signals/cell ≥4 and <6, (iii) Group 2 with ratio ≥2 and signals/cell <4, (iv) Group 3 with ratio <2 and signals/cell ≥6, (v) Group 4 with ratio <2 and signals/cell ≥4 and <6, and (vi) Group 5 with ratio <2 and signals/cell <4. Formalin-fixed paraffin-embedded samples were analyzed using two ddPCR assays each targeting an exon in the ERBB2 tyrosine kinase domain (exon 19 and 21, respectively) and a 130-gene NGS-based assay. For ddPCR, ERBB2 amplification status was determined from ddPCR ratios by using a recently published algorithm (Otsuji et al. 2017). For targeted NGS, ERBB2 amplification was called when copy number gains were detected in the majority of exons in ERBB2 (>50% of exons).

Results: Mean ddPCR ratios varied amongst the different FISH groups (P < 0.0001). As expected, patients with Group 1A had the highest mean ddPCR ratios compared to those with other FISH findings (P < 0.0001). Furthermore, there was a correlation between ERBB2 ddPCR ratios and HER2 FISH ratios (ρe19 = 0.4435, P = 0.001 and ρe21 = 0.4644, P = 0.0006). Using ddPCR, ERBB2 amplifications were detected in all classically amplified Group 1A cases (5/5) and in none of the classically non-amplified Group 5 cases (0/12). Interestingly, ddPCR assays called ERBB2 amplification in four cases with non-classical results: one in Group 2 (1/6), two in Group 3 (2/6), and one in Group 4 (1/17), including two cases in Groups 3 and 4 which also showed concomitant HER2 overexpression by IHC (3+). Similarly, targeted NGS revealed ERBB2 amplification in all Group 1A cases (5/5) and in none of the Group 5 cases (0/12). Furthermore, NGS detected amplification in three non-classical cases: one in Group 1B (1/5), one in Group 3 (1/6), and one in Group 4 (1/17), including one case in Group 1B which was not called amplified by ddPCR. Notably, the three cases with amplification by NGS were the only three cases in the non-classical groups with HER2 overexpression by IHC. Overall, there was a strong concordance between ERBB2 amplification status by ddPCR/NGS and HER2 overexpression by IHC (κe19 = 0.79, κe21 = 0.92, κNGS = 1.0).

Conclusion: ERBB2 amplification using ddPCR and NGS is correlated with HER2 overexpression in both classical and non-classical FISH groups, thus providing genomic evidence to support the recent recommendation for concurrent IHC testing in cases with unusual FISH results. Our findings also highlight a potential role of ddPCR and targeted NGS in the workup of challenging HER2 cases.

Citation Format: Yang S-R, Bouhlal Y, De La Vega FM, Ballard M, West RB, Sibley RK, Kuo CJ, Vilborg A, Allison KH. ERBB2 copy number analysis of invasive breast carcinoma using digital droplet PCR and targeted next-generation sequencing: A focus on 'non-classical' HER2 FISH groups using the 2018 ASCO/CAP HER2 testing guideline [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P3-10-12.