PD05-06: Determination of HER2 Status with Analysis of Plasma DNA by Digital PCR in Patients with Metastatic Breast Cancer.

Abstract

Abstract Background. A small proportion of originally HER2 negative primary breast cancers relapse with HER2 amplified disease. Identification of these cancers is important as these patients may benefit from HER2 targeting therapies. A proportion of free plasma DNA originates from the cancer, and accurate determination of HER2 copy number in plasma DNA has the potential to non-invasively determine HER2 status in patients with metastatic breast cancer. Digital PCR can differentiate small increases in DNA concentration with a much higher degree of accuracy than conventional quantitative PCR, and here we assess the potential of plasma DNA digital PCR to determine HER2 status. Methods. We examined a cohort of 44 patients with metastatic breast cancer treated at the Royal Marsden Hospital who had received a median of 1.5 prior courses of chemotherapy for recurrent disease; 34 patients had ER positive and 9 HER2 positive (all 3+ positive HercepTest) disease as determined on biopsy of recurrent disease in 16 patients and original primary cancer in 28 patients. Following informed consent, plasma samples were taken, free plasma DNA was extracted and concentration determined with LINE1 quantitative PCR. We designed a custom HER2 copy number TaqMan MBG-probe (Applied Biosystems) and two chromosome 17 peri-centromeric control probes TUFMP1 and UBBP4, avoiding known SNPs and regions of normal copy number variation. Digital PCR was performed in 384 well format on a 7900HT Fast RT-PCR system, with copy number ratio determined using the Poisson distribution. Results. HER2 Digital PCR assay was initially tested on DNA from 9 HER2 amplified cell lines. HER2/UPPB4 ratio was increased in 9/9 HER2 amplified cell lines, and HER2/TUFMP1 in 7/9 due to co-amplification of the 17q peri-centromeric DNA in 2 cell lines. Dilution experiments with SKBR3 DNA in non-amplified DNA demonstrated sensitivity down to ∼1% levels. We selected HER2/UBBP4 digital PCR for examination of patient samples. Plasma DNA Digital PCR HER2/UBBP4 ratios differed substantially between HER2 non-amplified (mean 1.047, SD 0.149, median 1.046, range 0.70−1.264), and HER2 amplified cancers (median 2.231, range 1.138−7.89, p<0.001). Digital PCR HER2 ratios had a high diagnostic accuracy with ROC curve AUC 0.967 (95% CI 0.585−0.997). With a threshold Digital PCR HER2 ratio of 1.3 the test had 100% specificity and 89% sensitivity, with negative predictive value 97%. The single miscalled HER2 amplified patient had a low tumour burden. Results of a test set using sequential probability ratio test analysis will be presented at the conference. Conclusions: Plasma DNA Digital PCR accurately determined HER2 status in patients with metastatic breast cancer. Analyses of plasma DNA have the potential to replace biopsy sampling of metastatic disease to guided targeted therapy. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr PD05-06.