Abstract 3986: Digital droplet PCR measurement for plasma HER2 amplification in patients with AGC

Abstract

Abstract Introduction: With the recent introduction of human epidermal growth factor receptor 2 (HER2) targeted therapies for patients with advanced gastric cancer (AGC), determining HER2 status is essential to select the patients who may benefit from this treatment. The current standard method assessing HER2 positivity is immunohistochemistry (IHC) or in situ hybridization assays, but this way of HER2 assessment has weaknesses especially when considering tumor heterogeneity. Using digital droplet polymerase chain reaction (ddPCR) technique, we evaluated HER2 amplification in both tissue and plasma samples collected from AGC patients and analyzed clinical implication of HER2 amplification in circulating tumor DNA. Method: Biopsied tissue and plasma samples were collected from patients with metastatic or recurrent AGC who were enrolled in AGC biomarker study conducted at Asan Medical Center. Samples were obtained before the start of anti-cancer treatment. HER2 amplification in DNA from tissue and cell-free DNA from plasma were determined by ddPCR, performed in Kindai University. Results: Samples of 63 patients with HER2 positive GC and 37 patients with HER2 negative GC enrolled between April 2014 and October 2017 were included in the analysis. As expected, the median HER2 copy number (CN) was higher in patients diagnosed as HER2 positive than negative, both for tissue (4.54 vs 0.95, P < 0.001) and plasma (1.49 vs 1.07, P < 0.001). ROC curve analysis showed 83.8% specificity and 69.8% sensitivity for tissue HER2 positivity at 1.17 CN of plasma HER2. In patients receiving anti-HER2 therapy (n=57), there was a trend for worse progression-free survival (PFS) outcome in patients with higher plasma HER2 CN (>1.5 vs ≤1.5), although no statistical significance was seen (median PFS 6.67 vs 8.19 months, P = 0.172). In multivariate Cox regression analysis including the age (>60 vs ≤60), HER2 IHC intensities (2+ vs 3+), and risk groups determined by Koo et al.’s prognostic model (Good vs Moderate vs Poor), higher plasma HER2 CN (>1.5 vs ≤1.5) was significantly associated with poor PFS (HR 2.22, 95% CI 1.14-4.32, P = 0.019). However, when tumor burden (sum of largest diameter of all measurable lesions) was combined as another factor in patients with measurable disease (n=39), plasma HER2 CN was no longer an independent factor (HR 1.38, 95% CI 0.58-3.27, P = 0.466) and larger tumor burden (>6.5cm vs ≤6.5cm) was found to be an independent prognostic factor for poor PFS (HR 2.63, 95% CI 1.12-6.21, P = 0.027), instead. Of note, plasma HER2 CN showed a positive relationship with tumor burden multiplied by tissue HER2 CN in linear regression (β=0.32, P = 0.002). Conclusion: Using the ddPCR technique, we found that plasma HER2 CN was significantly increased in HER2 positive GC patients compared to negative patients. However, plasma HER2 CN did not provide more information on predicting therapeutic effects than the tumor burden in patients receiving anti-HER2 therapy. Citation Format: Kyoungmin Lee, Kazuko Sakai, Min-Hee Ryu, Jae-Joon Kim, Young Soo Park, Young-Soon Na, Jungeun Ma, Hana Na, Kazuto Nishio, Yoon-Koo Kang. Digital droplet PCR measurement for plasma HER2 amplification in patients with AGC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3986.