Risk Factors Of Developing Pancreatic Cancer Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Pancreatic cancer, like most chronic conditions comes about through the direct interaction of genetic and environmental factors over a prolonged period of time. (1) The aim of this SSC is to examine the various factors, both external (e.g. smoking) or internal (e.g. genetic mutations) that pre dispose a person to develop this condition. This SSC will also underpin the very mechanism of DNA damage within the pancreas, along with other molecular alterations (e.g. deregulation of the cell cycle and apoptosis) and carcinogenic interactions which eventually lead to full blown cancer.

Pancreatic cancer mortality rates are variable amongst the sexes (males: 4.2 to 11.5 deaths per 100.000; females: 2.6 to 7.5 deaths per 100.000), but are higher in men than women

Exogenous Risk Factors

One of the best knows and recognised risk factor of developing pancreatic cancer is the direct inhalation of cigarette smoke. In comparison with non-smokers, the relative risk of people who smoke regularly to develop pancreatic cancer is roughly 2 but some studies claim that the risk is even greater than 2 (2). It is worth noting however, that some have suggested that quitting smoking for a period extending 10 years, may actually lower the relative risk of developing pancreatic cancer back to normal (the same as non smokers) In totality, it is thought that smoking regularly is responsible for 1/3 of all cases (5-6).

Aromatic amines in cigarette smoke are considered the major promoters of DNA damage within human cells (2). DNA adducts with aromatic amines discovered in the pancreas demonstrating that these compounds are able to reach the DNA within the pancreas and cause significant damage to our genome (7,8) Carcinogens within tobacco smoke are thought to alter the genes which control tumour development, e.g. K-ras and Tp53. One of the most definitive piece of work in this area was conducted by Berger who showed a direct correlation between smoking and K-ras mutations in pancreatic cancer development. (9)

There is also an understanding that alcohol consumption may also be a risk factor for developing pancreatic cancer. This does not apply to people who consume moderate amounts of alcohol. Numerous case control studies have shown that the association is only present among people who severely abuse alcohol (2,10) It is interesting to note however that there is an interaction between tobacco smoke and alcohol consumption in K-ras mutations; with regards to people who both smoke and drink. Their risk is thought to be substantially increased in comparison with people who have never smoked or drank. (11)

It is a known fact that caffeine has the ability to modulate cellular processes such as the DNA repair mechanisms, but there is no well established link between high intakes of caffeine containing product e.g. coffee and the subsequent development of pancreatic cancer. It is worth noting however that high intakes of coffee have been shown to increase K-ras mutations. (11)

There is relatively little data with regards to the link between the development of pancreatic cancer and previous microbial infection. There is data that suggests that previous infection with typhoid and paratyphoid could potentially pre-dispose to pancreatic cancer formation. (12) The same may be said for H. Pylori chronic infections. (13 ,14) People with longstanding H. Pylori infections were shown to be twice as likely to develop pancreatic cancer. It is important to note that there is no infection of the pancreas by the organism, so the increased risk is attributed to the high levels of acidity in the duodenum and gastric outlet. This, along with damage caused by N-nitroso compounds and various other genetic factors acting on the host's inflammatory cytokine response may promote pancreatic carcinogenesis (15,16).

Endogenous Risk Factors

There are numerous pre-existing medical conditions that pre dispose individuals to developing pancreatic cancer. One of these conditions is type 2 diabetes. This correlation is verified by all cohort studies that demonstrated an increased chance of developing this cancer in the first 5 years after the diagnosis of diabetes mellitus. When the type of diabetes is taken into account, the increased risk of developing pancreatic cancer is only observed in patients with type 2 diabetes mellitus. Interestingly, the relationship is reversed with regards to type 1 diabetes mellitus. There are numerous explanations for this. Firstly, that the diabetes itself is a consequence of the cancer rather than the cause due to tumour cells invading the pancreas and impairing its normal physiological function. Secondly, both diseases share common risk factors, in this case the major one being smoking tobacco. And thirdly the idea that insulin itself plays a pivotal role in pancreatic carcinogenesis. (2, 17).

Another medical condition associated with developing pancreatic cancer is chronic pancreatitis. The time period of inflammation within the pancreas may be a major risk factor in developing pancreatic cancer. Numerous types of pancreatitis have been implicated in the development of the disease e.g. tropical and hereditary pancreatitis. Alcoholic pancreatitis is also thought to increase the risk of developing pancreatic cancer. There is significant evidence which demonstrates that prolonged pancreatic inflammation is a cause rather than a consequence pancreatic cancer. (18-24)

Other medical conditions are also thought to predispose individuals to developing pancreatic cancer. One of these conditions is peptic ulcer disease, and this is thought to be due to infection with H. Pylori and the subsequent damaging nitric products produced by the microbe. (13, 14) Although the link is not clearly established it is also thought that cholecystitis and cholecystectomies may also produce an increase risk of developing pancreatic cancer. There are thought to be three fundamental reasons for this association. Firstly it is due to increased levels of cholecystokinin (CCK), secondly, increased levels of secondary bile acids, and subsequent reflux of bile or duodenal juice into the pancreatic duct. However, there is the possibility that cholecystitis itself may be a premature indication of pancreatic cancer. (2) There are certain conditions which reduce the chances of a person developing pancreatic cancer. One such condition is allergies, although research does not specify which type of allergy, it is thought that the underlying biological mechanism could be due to involvement of Immunoglobulin E. The study showed that reactions triggered by certain allergens, specifically house dust, cats, mould, and plants, and observed an inverse trend with increasing number of allergies and severity of allergic symptoms. (25, 26) There are numerous reproductive and gynaecological agents that have been linked with an augment of pancreatic cancer. These being: early age of first pregnancy, high parity, endometriosis and ovarian hyperplasia (2, 25). Oestrogen receptors are found in pancreatic tissue in the normal physiological state (26), hence it is plausible that lipid based (steroid) hormones influence pancreatic physiology.

As with most other cancers there is strong scientific evidence to suggest that having a certain genetic makeup will predispose certain individuals to developing pancreatic cancer. Individuals with a family history of pancreatic cancer are 2-3 times more likely to go on to develop pancreatic cancer in their life. It is important to note that this statistic in only referring to high penetrance genes involved in familial cancer syndromes. (2)

A scientific study involving 45,000 sets of twins modelled the involvement of hereditary shared environmental, and non-shared environmental factors to the chief tumor types taking into consideration the history of cancer and whether twins were mono or dizygotic (27). The data demonstrated that 36% of individuals who went onto develop pancreatic cancer largely due to their genetic makeup. The very high number in this instance suggests that perhaps it is the low pentrance genes that are involved in the majority of the normal population. This also shows that there must be numerous other heritable factors may not have been fully considered by epidemiologic studies.

Numerous high penetrance genes have been implicated in the familial pancreatic cancer, the major ones being BRCA2, CDKN2A, STK11, p53, APC, HNPCC, AT, FANCC, and FANCG. (25) Research has also shown that the role of some genes in pancreatic cancer development, e.g. BRCA2, varies from one population to the other. Although the gene itself is a high penetrance gene, it is thought that a combination of genetic and environmental factors eventually control the gene's role in pancreatic carcinogenesis.

Along with the above mentioned cases,there is a fraction of hereditary pancreatic cancer with multiple first-degree relatives afflicted by this tumour type have been recognized following an autosomal dominantly inherited pattern linked to a region located in chromosome 4q32.34 (28). It is thought that smoking and the number of first-degree relatives affected may increase the risk of the members of the family (29); also, genetic anticipation has been thought to occur (29), although this merits further investigation.

Besides the high penetrance genes involved, it is thought that there are a number of polymorphisms occurring in low penetrance genes amongst the general population. It is these polymorphisms, along with exposure to certain environmental factors which may eventually lead to the development of pancreatic cancer. Amongst the analysed genes and polymorphisms/alleles are phase I enzymes such as CYP1A1 (Val/Val and m1/m2/m4), CYP2E1 (c2) and extensive CYP2D6 metabolizers; phase II enzymes such as GSTM1 (null), GSTT1 (null and AB), NAT2 (slow), NAT1 (slow), NQO1 (R139W) and UGT1A7 (*3); and DNA repair enzyme such as XRCC1 (Arg399Gln). It has been found that individuals harbouring the alleles UGT1A7*3, GSTM1B, and NAT1slow had a higher risk of pancreatic cancer. (30-35)

There are researchers across the world who have shown the ability of the pancreas to metabolize aromatic amines by analyzing the amount of ABP-adducts according to both genotypes or phenotypes of several enzymes (13), and that a collaboration between dissimilar subunits of GST enzymes in human pancreas tissue has been demonstrated (36), indicating that low penetrance genes play an important role in the etiology of pancreatic cancer and that this is multifaceted subject should be thoroughly addressed by future epidemiologic studies. In a nutshell, there is clear evidence which suggests that the dysfunction of certain genes may well play a significant role in the development of pancreatic adenocarcinoma. However, it is important to understand that it is the interaction of numerous low penetrance genes and their environment which are most likely to lead to full blown carcinogenesis.

Morphology of Pancreatic Cancer

There are various methods being used to classify neoplasms and the major ones in use today are based on the cell's ability to differentiate from the normal, its phenotypical appearance or anatomical location. However, because of the fact that neoplasms imitate the original cellular architecture of the cell from which they are derived, means that there is substantial debate whether using the phenotype of a cell is indeed the best possible means of analysis. It is also now widely accepted, that certain neoplasms, certainly in the haematopoetic system, are derived from stem cells. It is precisely this reason that many tumour cells are abnormally differentiated and a largely heterogenous bunch. Despite these compounding factors, the cellular phenotype is still the most widely accepted means of classifying neoplasms.

The commonest form of pancreatic cancer is adenocarcinoma of the pancreas (PDAC), which accounts for approximately 90% of all pancreatic cancers. It affects both men and women equally and is most common amongst people over the age of 60. Histopathalogically this carcinoma shows features of ductal cells of the pancreas. The hallmark features of this type of tumour are localization in the head of the pancreas, infiltrating duct-like and tubular structures embedded in a highly desmoplastic stroma. These types of tumour cells produce large amounts of mucin which stimulate intralobular small ductules, this leads to the expression of the marker MUC 1, which itself is a marker for intralobular duct cells. Another marker, which is not found in normal pancreatic tissue, MUC4 can also be found on these abnormal tumour cells, exemplifying some of the genetic and phenotypic anomalies occurring with tumour development and progression. (37-45)

Histologically speaking pancreatic adenocarcinoma are similar to the ductal cells found within the pancreas. The structures found on the luminal side of the cells include microvilli and numerous mucin granules can be found in the cytoplasm on the apical side of the cell. The nuclei of these are rounded with small well defined nucleolus. (42)

Tremendous advances in our ability to understand and analyze large quantities of genetic data have provided evidence that there are three precursor lesions which exist before the development of pancreatic cancer. These pre cancerous lesions have been termed pancreatic intraepithelial neoplasia (PanIN). This has enabled researchers to link morphological changes within the epithelium to specific genetic abnormalities within our genome. Certain mutations have been definitively described as a sequence of events. In about 80% of the cases there is an activation of the K-ras oncogene and the subsequent down regulation CDKN2A/p16, TP53/p53 and SMAD4/DPC4. Secondly, there is the notion that telomerase activity is increased in 95% of all cases. There is substantial loss of alleles at chromosome 9p, 17p and 18q. There are some other genes that have been implicated in development of this carcinoma including the MKK4 gene, the gene for TGF Beta receptors, BRCA 2 gene and the LKB1/STK11 gene. (46-54)

The second major tumour type in pancreatic cancer is intraductal papillary mucinous neoplasm (IPMN's). IPMN's are a relatively uncommon type of exocrine tumour in the pancreas; they do however have a relatively favourable clinical outcome compared to the other types of tumour found in the pancreas. The tumour is derived from columnar epithelial mucin producing cells. The majority of these tumours occur in the head of the pancreas and maintain a consistently cystic appearance and cause clinical signs and symptoms that are consistent with pancreatitis These tumours are slow growing and around 35% of them eventually go onto become aggressive, invade the surrounding tissues and retain the ability to metastasize around the body. The tumour type within this subset which has the best prognosis is the type which arises from within the secondary ducts as opposed to the main duct. (55-59)

There are numerous genetic abnormalities which have been identified in IPMN's; however the relative frequency of these mutations is much less compared to other types of pancreatic tumours. However, it is important to consider the constitution and eventual phenotype of these cells and appreciate that they may arise from the intestinal or pancreatobiliary region. The mutations in the all important K-ras genes have been found, but in differing levels but are significantly lower than pancreatic adenocarcinomas. The same can be said for p53 mutations which are commonly found in cells exhibiting extreme atypia whereas K-ras mutations were present in cells exhibiting minimal atypia. HER-2neu/c-erbB-2 over expression was reported in a large fraction of IPMNs. There was also increased activity of telomerases in cells with extreme atypia. (60-67)

Mucinous cystic neoplasms are slightly more common in women and are found in the tail of the pancreas, have no communication with the ductal system, and may grow up to being 20cm. These tumours are notorious for high recurrence rates and their ability to metastasize. The major deaths in patients are caused by tumours which are deeply invasive; however, if the tumour is picked up early and surgically removed, the prognosis is generally good. (68-70) the expression of markers is variable depending on how invasive the tumour actually was. Non invasive tumours were found to express the markers neither MUC 1 nor MUC 2, but were found to be positive for MUC5AC. The more invasive lesions were found to express MUC1, along with stromal cells expressing oestrogen, progesterone and inhibin receptors. Other scientific studies have concluded that in this type of tumour, K-ras mutations happen early on in tumour development, and the mutations are found to increase exponentially as the tumours become more malignant and invasive. (71-74)

Serous cystic neoplasms of the pancreas involve three subtypes namely, serous microcystic adenoma (SMA), serous oligocystic and ill demarcated adenoma (SOIA) and von Hippel-

Lindau associated cystic neoplasms (VHL-CN). These three subtypes are derived and essentially made up of the same cell type. The cell's cytoplasm is unique as it has a high concentration of glycogen, however each type of serous cystic neoplasms have unique anatomical lopcation within the pancreas, microscopically they appear different and their genetic makeup is different. (75, 76)

Acinar cell carcinomas are relatively rare tumours of the pancreas which, once fully evolved have a tremendous capacity to invade local tissues and metastasize to the liver. In terms of medical intervention, these tumours respond well to chemotherapy. Histologically speaking these tumours have a nodular appearance, with the nodules themselves being quite large and prominent. These tumours differ genetically in their constitution as no mutations in K-ras, p53, p16 or DPC4 have been identified in any othese cancerous cells. However, recent data has shown that these tumours have a high degree of allelic loss. 'Chromosomes 1p, 4q, and 17p show LOH in >70% of cases and chromosomes11q, 13q, 15q, and 16q show allelic loss in 60-70% of cases. The resulting allelotype of PAC is noticeably dissimilar from that of either ductal or endocrine tumors of the pancreas and the involvement of chromosome 4q and 16q seems characteristic of this tumor type. Interestingly, alterations in the APC/β-catenin pathway have been found in 4 of 17 cases of acinar carcinoma studied' (results taken directly from reference). (77, 78)

There is one specific type of tumour that is present that is present in the pancreas of children. That tumour is termed a pancreatoblastoma. It is extremely rare in adults and almost exclusively occurs in children. These tumours contain mainly acinar cells along with some ductal and endocrine cells. But the majority of the cells found are acinar cells. It is for precisely these reasons that histologically speaking these tumours resemble acinar cell carcinomas. Some of the hallmark features that they share are abnormalities within the APC/B-catenin pathway, however these abnormalities are found in a much higher number of pancreatoblastomas, around 80%. Abnormalities and deleteion of alleles has been noted to occur at chromosome 11p, which is the site of the all important wt2 locus, which includes growth and cell cycle regulatory genes that are also distorted in other tumors in children, such as e.g. liver tumours and secondly, nephroblastoma. In addition, innate incongruity at this locus causes the Beckwith-Wiedemann syndrome, which may be associated with pancreatoblastoma. Pancreatoblastomas, similar to acinar cell carcinomas, demonstrate no microsatellite volatility or mutations of the K-ras, p53 or DPC4 genes. (79)

Another subtype of pancreatic tumours is pancreatic endocrine tumours. The first type of tumour in this category is termed a non-functioning PET which very rarely leads to any clinical signs and symptoms. The second subtype is termed functioning pancreatic endocrine tumours as these do produce a range of clinical signs and symptoms due to over secretion of pancreatic hormones by the abnormally growing cells. (80) However, it is worth noting that these tumours arise from histologically different cells, and not from cells within the duct of the pancreas, as is the case with exocrine tumours. (81) The commonest mutation found in these tumours is that of the gene MEN 1, however the one exception to this are insulinomas, where only 7% of the tumours are found to have the MEN 1 mutation. With functioning pancreatic tumours, the incidence of MEN 1 mutation is around 70%, whereas the rate of occurrence within non functioning tumours is much lower, around 27%. This is exemplified by the fact that non functioning pancreatic endocrine tumours are are very common in MEN 1 patients. (80, 81)

Solid pseudo papillary neoplasm are, by definition solid tumours with a pseudo papillary structure. These are malignancies which are not normally very aggressive or invasive and affect female children and younger women. Immunohistochemistry has shown that these tumour cells display a receptor for progesterone. There are no known abnormalities in any of the ras genes, p53, p16 or DPC4. The abnormalities in this case are a result of mutations in the β-catenin gene and subsequently show nuclear β-catenin protein appearance (82, 83). This greatly impairs the cells adhesive capability, leading to cellular degeneration and haemorrhaging, giving rise to their pseudo papillary appearance.

Anatomically speaking, the ampulla of Vater is an extremely unique structure. It is fundamentally the junction between the cells of the pancreaticobillay type and cells which arise from intestinal tissues. This would suggest, that on a molecular level, at least, that these tumours are similar to pancreatic ductal adenocarcinomas. This is true for one subtype of ampullary tumours, which share the genetic mutations that are found in pancreatic ductal adenocarcinomas. However, the second subtype of these tumours actually shares the molecular biology of intestinal carcinomas. There are genes and allelic variations found in ampullary tumours. These include abnormalities in K-ras mutations which are seen in about one-half of cases (84, 85). Inactivation of DPC4 was found in about 50% of cases, as shown by negative staining for the protein by immunohistochemistry. There is, however, no correlation between the lack of expression of Dpc4 and survival (86). However, allelic losses on chromosomal arm 17p (63%) have been previously found to be an independent prognostic factor among ampullary cancers at the same stage (87). In a second group of ampullary carcinomas there were APC gene mutations in a proportion of these cancers (88) or microsatellite instability, a feature that correlated significantly with increased survival. (89) (90).

Pancreatic Precursor Lesions

The commonest type of pancreatic cancer is pancreatic ductal adenocarcinoma. If one is to look at the top 100 cancers prevailing in the world today, it is transparently clear that this tumour type has the worst prognosis and outcome. The reasons for this are multi faceted. The principal one being that this neoplasm is confined to the gland and does not produce any clinical signs or symptoms. The only time it actually produces signs and symptoms is when it invades the local surrounding tissues, lymph nodes and possibly metastasised to the liver or other areas within the retro peritoneum. Another major reason is the anatomical location of the pancreas itself, which is almost completely retoperitoneal and extremely difficult to access. This also means that very little is actually known about the initial stages of the disease. The fundamental question for the medical profession is to identify whether or not there are any pre cursor lesions which can be, somehow identified during their hyper or metaplastic stages so that some form of medical or surgical intervention is possible. We know that these lesions do exist, but the problem remains in identifying that these lesions exist in certain patients, very similar to how PSA is an extremely valuable marker for prostate hyperplasia and subsequent carcinoma development.

The precursor lesions within the pancreas are termed pancreatic intraepithelial neoplasia (PanIN). The WHO in the year 1996 came identified four types of duct lesions, however the lack of scientific basis in naming the lesions made this a relatively unhelpful method of classification. Later, leaders in the field of pancreatic cancer came up with the new PanIN mechanism of classification which was much more standardized and globally acceptable. (91) the principal behind this method was to categorically identify the numerous changes that that seem to take place within the ductal system, but more importantly to identify those changes which were directly relevant to the formation of pancreatic ductal adenocarcinoma. In this way the different grades of PanIN were identified based on the degree of structural dysplacity and the level of atypia within the individual cells themselves, as explained below in Table 1 and Figure 1, overleaf. (92)

Numerous techniques have been used to analyze PanINs, including, microdissection, along with PCR, SAGE and the extensive use of microarrays. Initially, researchers were investigating the prevalence of K-ras mutations in these PanINs as there were an abnormally high levels of these mutations in pancreatic ductal adenocarcinomas. These studies did demonstrate the presence of these mutations, however, their prevalence across the lesions was extremely variable and dependant on the method of analysis used and the different techniques deployed. Those studies which had mainly papillary lesions which were extremely dysplastic, K-ras mutations were found in around 75% of them, and these were most likely to be PanIN-3 lesions. (92) However, in the less invasive and low grade lesions e.g. PanIN-1A and 1B, the frequency was much lower, at around 40%. It is interesting to note as well that in a study done exclusively on PanIN-1A lesions alone, the frequency of K-ras mutations was as low as 20%. K-ras mutations are also found in normal pancreatic duct cells, and therefore it is now generally accepted that they are not a good marker to discriminate between PanIN lesions and grading their level of malignancy. After K-ras the next step was to investigate other genes and molecules which were knows to play a significant role in the development of pancreatic ductal adenocarcinoma. Numerous papers focused on the other genes p16, p53 and DPC4, using LOH analysis to try and identify abnormalities. These studies demonstrated and increasing number of LOH's with increasing grade of dysplasia. (93-100)

Another significant aspect of pancreatic dysplastic lesions was the reduction in the length of telomeres. These were clearly shown in PanIN-1 lesions, all grades, and the perception was that these shortened telomeres actually pre dispose the evolving lesions to gather continuing chromosomal abnormalities as time progressed. Subsequently, p16, p53 and DPC4 are inactivated, and this is thought to occur between PanIN 2 and PanIN 3. (101) data then demonstrated that p53 and DPC 4 mutations occurred mainly PanIN-3 lesions. This provided valuable insight into the genetic events and anomalies leading to pancreatic dysplasia. Based on the premise that LOH 'chromosomal loci 17p (p53) and 18q (DPC4) was already observed in PanIN-2, allelic deletion may precede the mutational event in the biallelic inactivation of these two suppressor genes.' (TAKEN FROM REFERENCE) (102) another important event is the deactivation of the vital tumour suppressor genes BRCA2 and maspin, which are thought to be key players in the development of breast carcinogenesis. It was discovered that PanIN-3 lesions exhibited an allelic loss of the BRCA2 gene and the subsequent transcription of the abnormally mutated maspin gene. The same was true for the extremely invasive adenocarcinomas. (103, 104) Further investigation showed that notch pathway components were present in around half of all pancreatic ductal adenocarcinomas. And certain target genes within this pathway e.g. Hes1 and Notch1 were upregulated in PanIN-1. Along with this it was also discovered that there was an increase in the homeobox transcription factor Pdx1, highlighting the homogeneity the development of the pancreas and the development of these abnormal pre-cursor lesions. It was also shown that genes involved in the hedgehog signalling pathway were found in low grade PanINs. (105, 106)

Detailed analysis has shown that it is possible that low grade PanIN-1 lesions may exist in normal human beings, even early on in life. Interestingly, K-ras mutations can also be seen in PanIN's which may seem harmless and benign, and even so in normal duct epithelium within the pancreas. PanIN's with a certain number of K-ras mutations are thought to be present in the pancreas for a substantial length of time before they transform into a potentially lethal type of neoplasm. (93)

Developmental genes and cancer

Adenocarcinoma of the pancreas, in the developed world, is the 5th leading cause of death. Its prognosis and outcome is extremely poor and the 5 year survival rate is less than 5%. Histologically speaking it has been observed that PanIN-1 lesions show a degree of hyperplasia but not dysplasia. PanIN 2 lesions are essentially dysplastic lesions, and PanIN 3 lesions correspond to developed carcinoma in situ. In this paper so far, I have already discussed the genetic mutations that occur with the development of these types of intraepithelial neoplasia. However, it is also important to consider the genes involved in the development of cancer as these genes may provide insights into where a marker for this cancer may lie. Numerous genes that are involved in the development of this cancer are technically proto-oncogenes e.g. Meis, Pbx and SonicHedgeHog (SHH). Along with these genes, numerous tumour suppressor genes such as Smad 2 and Patched are also involved in pancreatic development. Numerous other factors e.g. certain growth factors like TGF-Beta, FGF, Wnt, EGF and BMP are also thought to be involved in some of the steps which regulate tumour progression.

There is substantial data generated in mice which has shown that the activation of the oncogene K-ras leads to the formation of PanIN 1 lesions, some of which may go onto become highly invasive and metastatic cancers. Similarly, the activation of the notch pathway and the induction of Cox 2 and MMP-7 have been observed in PanIN lesions in humans. It is interesting to note that these mice were generated by 'crossing mice that express a Cre-activated K-ras allele inserted into the endogenous K-ras locus with mice expressing the Cre recombinase under the control of the Pdx1 promoter or the endogenous p48 promoter, thus allowing the expression of the transgene selectively in a pancreatic precursor population. Moreover, when Pdx1-Cre-driven K-ras activation occured in a tissue-specific p16INK4 null background, PanIN formation, tumour progression and metastasis developed in an accelerated form.' (107) These experiments are vital, because they essentially demonstrate, that the progenitor cells for these lesions contain active promoters meaning that PanIN lesions could actually arise from undifferentiated cells within the pancreas. In the past, experiments conducted which c Ras, an oncogene, was artificially expressed in mature acinar and duct cells through the use of elastast of promoters e.g. CK19, eventually resulted in the development of purely acinar or in some cases, mixed acinar and ductal neoplasms along with a degree of tansductal inflammation, but importantly, there was no formation of any PanIN lesions. (108)

Another experiment which provided valuable insight into this complex field was done using mouse models, consisting of TGF alpha caused pancreatic cancer, which demonstrated acinar and ductal cells undergoing metaplasia as a result of massive multiplication of an undifferentiated cellular population, homogenous to cellular expansion within the pancreas during embryonic development. These cells displayed an up-regulation in Pdx1 expression in the epithelium of the duct before it had become malignant along with noticeable focal Pax6 involvement. Also in this model, TGF alpha was shown to induce the activation of the Notch signalling pathway. This is proven by the fact that certain Notch target genes e.g. Hesl and Heyl. Subsequently, pharmacological inhibitors were used to stop the Notch pathway, and this halted the metaplastic sequence of events. Furthermore, Notch target genes and Notch's pathway components were all up regulated in invasive pancreatic cancer along with PanINs. This has provided sufficient evidence that a signalling pathway which is primarily responsible for pancreatic development during embryogenesis is actually activated and functioning during carcinogenesis. Similar data has also been found implicating other developmental pathways e.g. the hedgehog pathway was implicated in the early stages of cancer genesis, and particularly the SHH pathway, once over expressed in mice through the use of a Pdx1 promoter, eventually led to the formation of abnormal tubular lesions which resembled PanIN 1 and 2. These lesions enclosed abnormalities in K-ras and over transcription of HER-2/neu, which happen throughout the sequence of human pancreatic cancer. Hedgehog signalling was established to be activated in cell lines recognized from primary and metastatic pancreatic cancers as well. On the other hand, inhibition of HH signalling with cyclopamine triggered apoptosis and stopped proliferation in a set of pancreatic cell lines both in vitro and in vivo. It remains to be deliberated whether and how these two signalling pathways are united during cancer progression. (109-112)

Ultimately, changes to the rate of expression of transcription factors implicated in pancreatic expansion have been demonstrated in both; in vitro and in vivo situations in where the exocrine pancreas gives rise to the neoplasm. For example, in the azaserine-induced pancreatic carcinogenesis rat model, the DSL6 copied tumors showed an acinar phenotype and displayed p48 when sequentually transplanted. In comparison, the newly derived DSL6 cell lines and neplasms affected by them exhibit a ductal phenotype and do not have p48. Down regulation of p48 expression is also seen in the bulk of human pancreatic adenocarcinomas. Removal of p48 happens extremely rapidly during the acinar-ductal switch process in vitro. Recently, p48 has been known to display an antiproliferative feature that is not confined to cells with acinar roots (113). Hence, it is plausible that changes in p48 expression may work towards uncontrolled growth and tumour formation. Fascinatingly, if the transcriptional activity of another acinar cell specific bHLH Mist1 is inhibited, the acinar-ductal metaplasia process also takes place in vitro and in animal models (114).

As a matter of fact, Mist 1 knockout mice also display a continuous disintegration of the exocrine pancreas that summarizes some hallmark feature of human chronic pancreatitis (115). Hence, in the lack of operational Mist1 and p48, acinar cells do not preserve their distinctiveness. The exact significance of these results in the attainment of neoplastic properties needs to be looked in to. In a nutshell, numerous current studies have brought forward the idea that pancreatic cancer is a condition initiated by the flawed re-activation of signalling pathways that are characteristically down regulated after the conclusion of embryonic maturation. Along with this, want for or abnormal expression of transcription factors that play a part in initial pancreas formation and in later on in the upkeep of cell homeostasis could be a vital process which is a part of this disease.

There is still work to be done with regard to finding the vital genes that play a part in duct cell formation. Due to the fact that these cells enact a fundamental role in the formation of exocrine pathology, the identification of these genes will be extremely valuable to create animal models focusing on this particular cell type. The use of global expression examination as before done for the classification of other embryonic and adult pancreatic cell types will offer additional data (116, 117).

As time progresses, increasing knowledge of the extracellular signals and transcription factors that govern pancreatic cell proliferation, diversification and subsequent demarcation will be extremely helpful in creating in vitro beta cells or acinar cells from adult or embryonic stem cells. In conclusion, it is clear that there are very many reasons why pancreatic adenocarcinoma is one of the world's more deadly malignancies. There are so many risk factors, both exogenous and endogenous which play a vital role in our subsequent risk of developing the cancer, and many of them are simple things e.g. drinking too much coffee. There are perhaps too issues with our genetic structure, and the malfunctioning of certain genes which sometimes may lead to tumour development. I feel that one of the reasons why PDAC is so deadly is that there is no current way of detecting early, any abnormalities in pancreatic function. So far, no singular protein e.g. Prostate specific antigen in prostate cancer has been identified, leaving the cancer to develop undetected and unchecked. There are always ways we can assess someone's risk of developing a specific cancer, but at the same time we need a marker to identify the lesions with regard to pancreatic intraepithelial neoplasia. Until this can happen, discovery of pancreatic adenocarcinoma in patients will be either too little, or unfortunately for some, too late.