Pancreatic cancer (PC) is a devastating disease. Gemcitabine has been the standard therapy for experimental regimens in patients with advanced PC for over a decade, but recently, the overall survival has been significantly prolonged using combination therapies, such as gemcitabine plus erlotinib or a combination of oxaliplatin, irinotecan, fluorouracil and leucovorin (FOLFIRINOX) [
1‐
3]. Despite some recent progress, however, the overall survival rate of patients with PC is still less than 5% [
4]. The model explaining the progression of PC is influenced by multiple genetic alterations. During early genetic events, such as activating point mutations in the
K-
ras oncogene and the overexpression of the
HER-
2/
neu gene, pancreatic duct lesions show minimal cytological and architectural atypia. The inactivation of the
p16 tumor suppressor gene appears to occur at a later stage, followed by the loss of the
p53,
SMAD4, and
BRCA2 tumor suppressor genes [
5‐
8]. For instance, the
HER-
2/
neu gene is not expressed in the epithelium lining of normal pancreatic duct, but it is highly expressed in pancreatic intraepithelial neoplasia [
9]. However, two clinical trials assessing anit-HER2 trastuzumab therapy in patients with PC overexpressing HER2 have produced disappointing results [
10,
11]. Although such recent breakthroughs in the molecular biology of PC have assisted in translational research, creating hope for individualized therapy and better disease management, the inhibition of epidermal growth factor receptor using erlotinib is, to date, the only targeted approach that has been demonstrated to result in a survival [
1]. Therefore, further understanding of the molecular biology of PC is needed.
The transforming growth factor, beta (TGFB) receptor II (
TGFBR2) and
SMAD4 genes are commonly inactivated in several types of cancer, providing evidence that the TGFB signal functions as a tumor suppressor [
12,
13]. Thirty percent of colorectal cancers are thought to contain a mutation in the
TGFBR2 gene. The human locus 18q21, which encodes the
SMAD2 and
SMAD4 genes, is often mutated or lost completely in several cancers. The loss of the
SMAD4 gene eliminates the classic SMAD2/3/4 heteromeric complexes that have been implicated in a large number of TGFB-dependent transcriptional regulatory complexes. As a result, TGFB-mediated growth inhibition is lost. The
SMAD4 gene is inactivated in 55% of PC tumors, and numerous studies on TGFB signal in PC have been reported. The loss of the
SMAD4 gene is correlated with both a poor prognosis and the development of widespread metastases in patients. The
TGFBR2 gene is also altered in a smaller subset of PC tumors [
5‐
7,
14,
15]. In addition, pancreatic-specific
TGFBR2 or
SMAD4-knockout mice with active
K-ras expression developed PC [
16,
17]. However, the roles of defects other than those in the
SMAD4 and
TGFBR2 genes in PC remain unclear, and few studies regarding the activin signal, which also belongs to the TGFB superfamily, have been reported [
18‐
20]. Defects in several genes involved in the activin signal pathway have been characterized in several cancers. For instance, two 8-bp polyadenine tracts in the activin A receptor, type IIA (
ACVR2A) gene were reported to be targets for frameshift mutations in gastrointestinal cancers with microsatellite instability [
21]. Similarly, the activin signal induces growth inhibition and apoptosis mainly through SMAD-dependent pathways in many other cancers [
22‐
27]. Thus, the dysregulation of the activin signal is directly involved in carcinogenesis. In contrast, however, a recent study has demonstrated that Nodal/Activin signal is associated with self-renewal and the tumorigenicity of PC stem cells [
20]; thus, the role of activin signal in pancreatic carcinogenesis remains controversial. In the present study, we identified a homozygous deletion of the activin A receptor, type IB (
ACVR1B) gene in PC cell lines using array-comparative genomic hybridization (array-CGH). Furthermore, we investigated the role of this homozygous deletion in PC cell lines and the status of the
ACVR1B gene in clinical samples of PC.