177P - HER2 testing in multiple solid tumors: Concordance between 3 scoring algorithms

Background

Trastuzumab deruxtecan is approved in the US for pretreated adult patients with unresectable/metastatic HER2-positive (immunohistochemistry [IHC] 3+) solid tumors without alternative treatments. However, HER2 IHC testing is not routine beyond breast and gastric cancers and there is a lack of consensus regarding the most reliable algorithm for scoring other solid tumors. This exploratory analysis reports concordance between 3 HER2 scoring algorithms in multiple solid tumors.

Methods

Images of HercepTest (Dako AutoStainer)-stained (biliary tract, bladder, cervical, endometrial, ovarian, pancreatic, other [including salivary gland] tumors) and VENTANA HER2 4B5 (Roche)-stained tissue (NSCLC) from a commercial pan-tumor sample set and the DESTINY-PanTumor02 (DP-02) study were assessed by 3 external pathologists using the ASCO/CAP gastric (reference; per DP-02 [all images rescored]), ASCO/CAP breast, and endometrial clinical trial (endometrial tumors only) scoring algorithms. The primary outcome was intra-pathologist concordance between HER2 scoring algorithms by tumor type. The secondary outcome was inter-pathologist concordance.

Results

In total, 488 images were assessed. Positive percent agreement (PPA) between algorithms by tumor type for each pathologist is shown in the table. Between the breast and gastric algorithms, PPA was greater when scoring IHC 3+ and 0 versus IHC 2+ and 1+. PPA between the endometrial and gastric algorithms lacked consistency across HER2 expression levels. Across tumor types and algorithms, the majority (42/51) of inter-pathologist pairwise comparisons had Cohen’s kappa values >0.4 (at least moderate agreement).

Conclusions

Data show the gastric and breast algorithms are comparable in HER2 IHC 3+ identification; lower concordance was observed for IHC 2+ and 1+, and between the gastric and endometrial algorithms. Findings indicate greater awareness of best scoring practice is needed to ensure consistent assessment of HER2 IHC status in solid tumors. Table: 177P

*Salivary gland tumors (n=18) and DP-02 other cohort; ^†PPA was 100% as no cases identified by gastric/breast algorithm

Clinical trial identification

NCT04482309.

Editorial acknowledgement

Thank you to the patients and their caregivers for their participation, and the study site staff for their contributions. Medical writing and editorial support was provided by Zoë Hine, BSc, of Helios Medical Communications, part of Helios Global Group, and was funded by AstraZeneca.

Legal entity responsible for the study

AstraZeneca and Daiichi Sankyo.

Funding

This study is sponsored by AstraZeneca. In March 2019, AstraZeneca entered into a global development and commercialization collaboration agreement with Daiichi Sankyo for trastuzumab deruxtecan (T-DXd; DS-8201).

Disclosure

S.M. Shiller: Non-Financial Interests, Personal, Speaker’s Bureau, Steering Committee/Advisory Board: AstraZeneca; Non-Financial Interests, Personal, Speaker’s Bureau: Lilly; Non-Financial Interests, Personal, Speaker’s Bureau, Advisory Board: Mirati, Astellas, Merck; Non-Financial Interests, Personal, Advisory Board: Pfizer. J. Baumann: Financial Interests, Personal, Full or part-time Employment: AstraZeneca; Financial Interests, Personal, Stocks or ownership: AstraZeneca. G. Kumar: Financial Interests, Personal, Full or part-time Employment, Sr. Director: AstraZeneca; Financial Interests, Personal, Stocks/Shares: AstraZeneca; Financial Interests, Personal, Royalties, 2019 Book. Predictive Biomarkers in Oncology. Applications in Precision Medicine: Springer Nature. A. Brown: Financial Interests, Personal, Full or part-time Employment: Daiichi Sankyo; Financial Interests, Personal, Stocks or ownership: Daiichi Sankyo. L. Luo, H. Bridge: Financial Interests, Personal, Full or part-time Employment: AstraZeneca; Financial Interests, Personal, Stocks/Shares: AstraZeneca. All other authors have declared no conflicts of interest.