Cancer is fundamentally a genetic disease that develops when alterations in multiple genes result in dysregulation of a select number of cellular pathways and processes. In the last decade, rapid advances in nucleotide sequencing and bioinformatic technology have led to an expansion of information obtained from the sequencing of cancer genomes. Sequencing of pancreatic cancer (PC) genomes as well as premalignant lesions of the pancreas has deepened our understanding of tumorigenesis and identified potential novel therapeutic targets. Although PC has generally been associated with acquired genetic mutations, a small proportion of PCs may be due to inherited PC syndromes. In this chapter, we focus on the current understanding of proposed critical genes and affected pathways that have been shown to be important in sporadic and hereditary PCs.

With the rapid advancement of next generation sequencing technology, several cancer sequencing initiatives have focused on characterizing the genomic landscape of PC. The first genome-wide analysis of PC was performed in 2008, based on 24 pancreatic tumors and xenografts using capillary sequencing and single nucleotide polymorphism arrays.¹ This study found that PCs harbor, on average, 48 genetic mutations and confirmed that KRAS, TP53, CDKN2A, and SMAD4 were the most commonly mutated genes in this disease. Despite the large number of mutations, if one excludes KRAS, TP53, CDKN2A, and SMAD4,² none of the remaining mutated genes occurred with a prevalence greater than 5%. However, the authors were able to classify these genes into 12 core signaling pathways, which they suggested could be potential targets for therapeutic interventions. A subsequent study that utilized whole exome sequencing and copy-number analysis of 99 human PCs confirmed the vast heterogeneity of mutated genes. This study was remarkable for the use of laser capture microdissection of the pancreatic tumors to minimize the incorporation of stromal elements and maximize the enrichment of the sample for adenocarcinoma. This study identified 2,016 genes with nonsilent mutations and 1,628 copy-number variations, with an average of 26 mutations per patient. In addition to KRAS, TP53, CDKN2A, and SMAD4, mutations in several novel genes including ZIM2, MAP2K4, NALCN, SLC16A4, MAGEA6, EPC1, and ATM were identified. These mutations identified novel pathways not originally identified in the previous 12 core signaling pathways, which affect chromatin modification and axon-guidance. Similar to previous reports, the investigators also demonstrated that the majority of mutated genes had a prevalence of less than 2%. Although PC has generally been associated with acquired genetic mutations, a small proportion of PCs may be due to inherited PC syndromes. In this chapter, we focus on the current understanding of proposed critical genes and affected pathways that have been shown to be important in sporadic and hereditary PCs.³ The integrated genomic and transcriptional analysis further refined the biological insights into PC pathogenesis and may help to identify potential therapeutic targets.

Additional next-generation sequencing studies of human PCs have defined structural rearrangements and explored clonal relationships between metastases.^4,5 In a study of 13 human PCs (including 10 primary tumors and 3 low-passage cell lines), significant interpatient heterogeneity was observed with different types of chromosomal rearrangements and different frequencies of rearrangements.⁵ In contrast to other solid tumors, a distinctive pattern of structural rearrangement called “fold back inversion” was identified involving frequent breakage-fusion-bridge cycles. This form of rearrangement was found to be an early event in the development of PC and may play a role in the amplification of cancer genes. This study demonstrated that genomic instability persisted after tumor dissemination, resulting in the ongoing genetic evolution in distant metastases. In another analysis, the clonal relationships among metastases were examined in seven PC patients who had undergone a rapid autopsy. This study identified clonal populations within the primary tumor that were felt to give rise to distant metastases; however, the metastatic clones continued to evolve genetically when compared to the parental clone in the primary tumor. A quantitative analysis of the timing of genetic evolution of PCs suggested that the time from the occurrence of the initiating mutation in the pancreas to the acquisition of metastatic potential may be approximately 15 years. Taken together, these sequencing studies have reinforced two fundamental concepts: there are a consistent small number of mutations that are likely required to drive the initiation of PCs, and the vast genetic heterogeneity of PCs suggests that additional mutations may be required for initiation of the primary pancreatic tumor.

KRAS Signaling

The most common genetic alteration in PC is the activating point mutation at codon 12 of the KRAS gene and is found in approximately 95% of PCs.⁶ The Ras protein plays an important role in regulating cell proliferation, differentiation, and survival. KRAS point mutations have been found in 36% of PanIN-1, 44% of PanIN-2, and 87% of PanIN-3 lesions.⁷ This suggests that KRAS mutations may be one of the earliest events in the malignant progression of PC. Indeed, in a transgenic model of PC with an inducible KRAS G12D mutation, oncogenic KRAS was required for the maintenance of PanIN lesions. However, when the inducible KRAS mutation was silenced, PanIN lesions reversed and there was a decrease in stromal proliferation, suggesting that PC tumorigenesis is Ras dependent.⁸ Due to the high prevalence of KRAS mutations, it is an ideal target for PC therapy; however, previous clinical trials using pharmacologic inhibitors targeting Ras through posttranslational modification have failed to demonstrate clinical benefit.⁹ For several decades, the Ras protein has been deemed an “undruggable” protein due to its affinity for its substrate at picomolar concentrations. As an alternative, attempts to abrogate downstream signaling of Ras with pathway-specific inhibitors have been made; these have also achieved limited success due to the redundancy of many of these pathways. However, recognizing the oncologic importance of targeting Ras for both PC and other solid tumors, a Ras Project, sponsored by the National Institutes of Health and the National Cancer Institute, has been initiated to further characterize mutant Ras and develop novel targeted agents.

DNA Damage Control

TP53 has been shown to play multiple complex roles in cancer. The protein product of TP53 binds to specific sites of DNA and activates the transcription of certain genes that control cell division and apoptosis. Tp53 protein is stabilized after DNA damage and cellular stress, and serves as a DNA checkpoint regulator. TP53 mutations are the most common, single-point mutations in human cancers and inactivating mutations are observed in up to 75% of human PCs.¹⁰ Loss of TP53 expression has been observed in PanIN lesions with dysplasia but not in early PanIN lesions.¹¹ In transgenic models of PC, the addition of a mutant TP53 allele results in accelerated tumor progression and induction of invasive and metastatic disease.¹² Similar to KRAS, TP53 mutant proteins have been considered “undruggable” and alternative therapeutic modalities including RNAi or re-expression of wild-type TP53 have not demonstrated any efficacy in clinical studies.¹³

Cell Cycle Regulation

The p16/RB1 pathway is important for cell cycle division. The retinoblastoma protein (Rb1) is a transcriptional regulator and regulates the entry of cells into S phase. A complex of cyclin D and a cyclin-dependent kinase (Cdk4 and Cdk6) regulate Rb1 via phosphorylation. The p16 protein is a Cdk inhibitor that binds Cdk4 and Cdk6. Virtually all PCs have loss of p16 function through homozygous deletions, mutation, and loss of heterozygosity, or promoter methylation of the p16/CDKN2A gene.^14,15 The silencing of the coding regions of Ink4a (p16/CDKN2A) in transgenic mice with oncogenic KRAS is associated with rapid PC tumor formation and death.¹² In addition, inherited mutations of the p16/CDKN2A gene cause a familial melanoma/PC syndrome known as familial atypical multiple mole melanoma. As such, p16 inactivation is thought to be an early event in PC progression, and loss of expression has been associated with larger tumors, increased risk of early metastasis, and poor survival.¹⁶

Transforming Growth Factor b (TGF-b) and SMAD4 Signaling

The Smad pathway mediates signals initiated upon the binding of the extracellular proteins TGF-b and activin to their receptors. These signals are transmitted to the nucleus by the Smad family of genes, including SMAD4. SMAD4 mutations have been found in only 22% of localized PCs but in up to 75% of metastatic PCs; therefore, SMAD4 mutations are thought to be a late mutation in PC tumorigenesis.¹⁷ Among intraductal papillary mucinous neoplasms (IPMN), SMAD4 expression is preserved in the majority of noninvasive IPMN but is frequently lost with the development of invasive carcinoma.¹⁸ The relatively high frequency of SMAD4 mutations, including homozygous deletions and intragenic mutations combined with loss of heterozygosity, suggests that this protein is a tumor suppressor important in PC pathogenesis.

Pancreatic Cancer Predisposition Genes

Up to 10% of PCs are thought to be a result of inherited susceptibility loci.¹⁹ Recognition of high-risk families is important for understanding PC biology and for recommending risk reduction strategies. Familial PC (FPC) is defined as two or more first-degree relatives with PC; some authors are more inclusive considering three or more relatives of any degree in this classification.^20–22 Current genetic testing strategies identify a germline mutation in less than 20% of suspected hereditary cases (Table 138-1).^23–25 The remaining 80% are considered to have a familial form of the disease. The most important genes which impart increasing risk include BRCA1, BRCA2, ATM, CDKN2A, APC, MLH1, MSH2, MSH6, PMS2 PRSS1, and STK11. A significant proportion of susceptibility genes are still unknown.

There is some controversy regarding age of onset in FPC. Some reports have concluded that the familial forms of PC do not show an earlier age of onset when compared to sporadic disease.²⁶ This observation challenges the current understanding of accelerated phenotype in hereditary cancer, as is evident, for example, with breast cancer in those with a BRCA1 mutation.²⁷ Other researchers have observed a phenomenon of anticipation in FPC, where subsequent generations develop the disease at an earlier age.^28,29