Evaluation of SAS1B as a target for antibody-drug conjugate therapy in the treatment of pancreatic cancer

Abstract

Successful therapeutic options remain elusive for pancreatic cancer. The exquisite sensitivity and specificity of humoral and cellular immunity may provide therapeutic approaches if antigens specific for pancreatic cancer cells can be identified. Here we characterize SAS1B (ovastacin, ASTL, astacin-like), a cancer-oocyte antigen, as an attractive immunotoxin target expressed at the surface of human pancreatic cancer cells, with limited expression among normal tissues. Immunohistochemistry shows that most pancreatic cancers are SAS1B^pos (68%), while normal pancreatic ductal epithelium is SAS1B^neg. Pancreatic cancer cell lines developed from patient-derived xenograft models display SAS1B cell surface localization, in addition to cytoplasmic expression, suggesting utility for SAS1B in multiple immunotherapeutic approaches. When pancreatic cancer cells were treated with an anti-SAS1B antibody-drug conjugate, significant cell death was observed at 0.01-0.1 μg/mL, while SAS1B^neg human keratinocytes were resistant. Cytotoxicity was correlated with SAS1B cell surface expression; substantial killing was observed for tumors with low steady state SAS1B expression, suggesting a substantial proportion of SAS1B^pos tumors can be targeted in this manner. These results demonstrate SAS1B is a surface target in pancreatic cancer cells capable of binding monoclonal antibodies, internalization, and delivering cytotoxic drug payloads, supporting further development of SAS1B as a novel target for pancreatic cancer.

Keywords: ASTL/SAS1B/ovastacin; antibody-drug conjugate; pancreatic cancer biomarker; surface cancer-oocyte antigen; targeted immunotherapy.

Figure 1. SAS1B was expressed in a majority of pancreatic cancers and was not detected in normal pancreas ductal epithelium by IHC

TMAs were stained for the expression of SAS1B with 6B1 mAb. SAS1B was not detected in normal pancreatic ductal epithelium (A) and most pancreatic intraepithelial lesions (B). Some stromal cells adjacent to these ducts showed cytoplasmic reactivity, as pictured in A/B. Many ductal carcinomas showed cytoplasmic SAS1B staining (C–E). This ranged from strong, diffuse staining that also included some ill-defined membranous positivity (C) to focal, exclusively cytoplasmic staining (D-E). A minority of ductal carcinomas were negative or showed only trace non-specific staining (F). Images are 400× magnification. SAS1B staining was scored on a 0 (negative) to 3+ positivity scale for each tissue type and result are summarized in the table (G). Percent of samples that were SAS1B positive, for each tissue type, is quantified in the last column (total number of SAS1B positive samples/ total number of samples) (G).

Figure 2. ASTL/SAS1B expression in pancreatic cancer patient derived xenografts

(A) RT-PCR analyses of 15 PDAC (1-15) PDX tumors and 3 normal human pancreas (normal) samples using a c-terminus ASTL specific primer set showed a 309 bp amplicon in 10/15 PDAC samples. Tumors from both males and females, early and late stage disease, as well as primary and metastatic tumors were ASTL^pos (Table). GAPDH was used as a housekeeping control for PCR. (B) Immunohistochemical localization of SAS1B, on representative examples from the same set of 15 PDAC tumors used in (A), labeled with anti-SAS1B mAb, 6B1. Tumor number indicated in bottom right corner of image. Images are 400× magnification. Tumors were scored on a 0 (negative) to 3+ positivity scale; total number of tumors in each group quantified in the table.

Figure 3. SAS1B localized to the cytoplasm and to the cell surface in pancreatic cancer cell lines

(A–D) Fixed and permeabilized indirect immunofluorescence (IIF) using anti-SAS1B mAb, SB2 showed SAS1B localized to the cytoplasm of three pancreatic cancer cell lines (mPanc96, 366, 608) compared to normal keratinocytes. (E–H) IIF on live, non-permeabilized cells using anti-SAS1B mAb, SB2, demonstrated staining of the plasma membrane of mPanc96, 366 and 608 cells but not normal keratinocytes. Data are representative of three independent experiments.

Figure 4. SAS1B surface expression in pancreatic cancer cell lines correlated to anti-SAS1B ADC cell killing in vitro

(A) Median fluorescent intensity of cell surface SAS1B detected by live cell flow cytometry with SB2 monoclonal antibody (blue line) or unrelated control antibody (red line) on PDAC cell lines mPanc96 (left), 608 (left middle), and 366 (right middle) and keratinocytes (right). (B) Cytotoxicity by anti-SAS1B ADC (mAb SB2) titration shown below each flow-cytometric plot for corresponding cell line. SB2- mAb-Duocarmycin immune complexes were generated (ADC) then incubated with cells for 72 hours. Relative cell viability was measured using CellTiter-Glo. Data represent averages of three independent replicates, with 3 technical replicates in each data point.