HER2 assessment using fluorescence in situ hybridization compared with OncotypeDX and association with risk of breast cancer death

Objective

Guidelines mandate demonstration of 95% concordance to another laboratory or method before reporting patient HER2 results. In this analysis, HER2 results obtained with Oncotype DX, a validated 21-gene recurrence score assay utilizing RT-PCR, were compared with standard fluorescence in situ hybridization (FISH) results. The Oncotype DX HER2 results were then associated with relative risk of breast cancer death.

Methods

Breast cancer specimens from the Kaiser Oncotype DX study were evaluated for HER2 by FISH with positive >2.2, equivocal 1.8 to 2.2, and negative <1.8. HER2 was also assessed with Oncotype DX (RT-PCR) with positive ≥ 11.5 units, equivocal >10.7 to <11.5 units, and negative ≤ 10.7 units (each unit = twofold change in expression). Concordance analyses were conducted following ASCO/CAP guidelines that mandate 95% concordance. Logistic regression was used to estimate association between Oncotype DX HER2 results and risk of breast cancer death.

Results

Of 568 patients, 12% (67 patients) were HER2⁺ by Oncotype DX and 11% (60) by FISH. Fifty-five patients were HER2⁺ by both methods. Of the 12 patients Oncotype DX HER2⁺ but FISH negative, using FISH 11 patients were HER2^- and one patient was HER2 equivocal. The positive and negative HER2 concordance by FISH and Oncotype DX was 97% (95% CI = 96% to 99%). Risk of breast cancer death was significantly greater in patients with HER2 ≥ 11.5 (OR = 1.84, 95% CI = 1.13 to 2.99) compared with patients with HER2 ≤ 10.7. However, in patients with HER2 >10.7 to <11.5 compared with HER2 ≤ 10.7, there was no greater risk (OR = 0.75, 95% CI = 0.46 to 1.20). There were 71 (12.5%) polysomy cases.

Conclusion

There is a high degree of concordance between RT-PCR using Oncotype DX and central laboratory FISH assessment of HER2 status.

Disclosure

FLB is an employee and shareholder of Genomic Health, Inc.