Abstract P1-08-14: HER2 extracellular domain (ECD): A predictive marker of lapatinib treatment benefit in advanced breast cancer

Abstract

Abstract Pretreatment serum HER2 ECD may be associated with clinical outcome to lapatinib therapy and may also have treatment-independent prognostic value. We performed a meta-analysis to examine the prognostic and predictive role of baseline ECD (bECD) levels for lapatinib therapy. We also compared and contrasted its role with tumor HER2 status. We analyzed bECD and tumor HER2 data in 1902 of 2264 patients (84%) with advanced breast cancer who were randomly assigned to receive lapatinib-containing therapy or control treatment in three clinical trials: EGF30001 (lapatinib and paclitaxel vs paclitaxel), EGF30008 (lapatinib and letrozole vs letrozole), and EGF100151 (lapatinib and capecitabine vs capecitabine). bECD was centrally measured by enzyme linked immunoassay and tumor HER2 was centrally determined by immunohistochemistry and fluorescence in situ hybridization. bECD and tumor HER2 status were examined for associations with objective tumor response (ORR), progression-free survival (PFS) and overall survival (OS) in the treatment groups. Of the patients with both bECD and tumor HER2 data, 31% had HER2 bECD≥15 ng/mL and 28% had HER2+ cancer. The effectiveness of lapatinib-containing therapy was significantly associated with bECD level (treatment-bECD interaction P<0.001 [ORR], P<0.001 [PFS]). In patients with bECD≥15 ng/mL, lapatinib-containing therapy, compared to control, significantly improved ORR (odds ratio [OR] for complete and partial response, 1.88; P = 0.001) and PFS (hazards ratio [HR] for progression or death, 0.69; P<0.001). Among patients with bECD<15 ng/mL, there was no significant difference between treatment groups for ORR (OR, 1.17; P = 0.19) or PFS (HR, 0.93; P = 0.30). The effectiveness of lapatinib-containing therapy was also significantly associated with HER2 tumor status for ORR (OR [HER2+] 2.39, P<0.001; OR [HER2-] 1.10, P = 0.43; treatment-HER2 interaction P = 0.001) and PFS (HR [HER2+] 0.63, P<0.001; HR [HER2-] 0.95, P = 0.46; treatment-HER2 interaction P<0.001). When bECD and HER2 tumor status were examined together, the effectiveness of lapatinib-containing therapy was significantly associated with both bECD and HER2 tumor status for PFS (treatment-biomarker interaction P = 0.02[HER2], P = 0.001[bECD]) but only for bECD for ORR (treatment-biomarker interaction P = 0.07 [HER2], P = 0.003 [bECD]). The effectiveness of lapatinib-containing therapy was not significantly associated with OS for both biomarkers (treatment-biomarker interaction P = 0.30 [bECD] and P = 0.80 [HER2]). In the control groups, bECD≥15 ng/mL (HR 1.35, P = 0.003) and HER2+ (HR 1.67, P<0.001) were associated with shorter PFS after adjusting for other baseline characteristics. bECD≥15 ng/mL (HR 1.89, P<0.001) but not HER2+ (HR 0.97, P = 0.81) was associated with shorter OS after adjusting for other baseline characteristics. Lapatinib therapy was associated with higher ORR and longer PFS in patients with elevated ECD levels. Elevated ECD was also associated with shorter PFS and OS in patients treated with non-lapatinib therapy. ECD provided additional predictive and prognostic information, in addition to tumor HER2 status, that if validated, could aid in treatment decisions. The predictive role of ECD in other anti-HER2 therapies requires further research. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-08-14.