Development of the SP120 rabbit monoclonal antibody for determining hENT1 status and predicting response to gemcitabine in pancreatic ductal adenocarcinoma.

Abstract

4041 Background: Human equilibrative nucleoside transporter 1 (hENT1) is the primary membrane channel through which gemcitabine (Gem) enters pancreatic tumor cells. Patients whose tumors have low expression of hENT1 may derive little benefit from Gem therapy. Methods: We have developed and analytically validated a rabbit monoclonal antibody, SP120, immunohistochemistry (IHC) in vitro diagnostic (IVD) assay, used on the VENTANA BenchMark ULTRA automated slide staining platform, for assessing hENT1 expression in pancreatic adenocarcinoma. Results: SP120 was shown to be sensitive and specific for membrane expression of hENT1 with variable expression demonstrated across a panel of primary pancreatic adenocarcinoma samples. Using retrospectively collected primary tumor samples (n = 201) from a randomized controlled clinical adjuvant trial (RTOG 97-04), we developed and verified a hENT1 scoring algorithm and cut-off for identifying patients least likely to benefit from Gem (hENT1-low). We confirmed the predictive value of the hENT1 assay by showing there was no association with clinical outcome in the 5FU-treated control group. The distribution of hENT1 expression was similar between these samples and an independent set of 130 pancreatic adenocarcinoma specimens from the AIO-PK0104 study (77 were confirmed metastatic biopsies). Importantly, 100% concordance in hENT1 status (high vs. low) was demonstrated between 16 primary tumors and paired, contemporaneous metastatic lesions. Overall, using this cut-off, approximately two thirds of pancreatic tumors displayed low-hENT1 expression across the different data sets (n = 363). Conclusions: The hENT1 IHC IVD has been developed and is being used in a pivotal trial of a novel lipid-conjugated Gem (CO-101, which enters tumor cells independently of hENT1), vs. Gem in hENT1-low pancreatic cancer patients (NCT01124786). This study is the first prospective test of the hypothesis that gemcitabine is not active in tumors with low hENT1 expression.