BDIAP: Symposium on Intestinal Pathology
Abstract:
Problems in Appendiceal Pathology
Mucinous and carcinoid tumours represent the two broad categories of neoplasm that occur in the appendix. The mucinous tumours of appendix can be viewed as forming a pathological spectrum of neoplasms, surrounding which there has previously been some diagnostic confusion, largely due to terminological inconsistencies. The term “mucocoele” (essentially referring to an appendix dilated by mucin) is one such source of confusion as it is a descriptive term and not a pathological diagnosis, that has been used to include situations with accumulation or oversecretion of mucin: (1) appendiceal obstruction with accumulation of mucin (with no neoplastic epithelial abnormality), usually <1cm diameter; (2) diffuse mucosal hyperplasia or hyperplastic polyp; (3) mucinous cystadenoma (neoplastic epithelium with low-grade atypia); (4) mucinous cystadenocarcinoma (with evidence of wall invasion).
Benign epithelial neoplasms of the appendix, usually known as mucinous adenomas or cystadenomas, can be diffuse or focal (when forming small polyps or sessile polyps). They can show various patterns including: (1) dilatation of the appendix with a single layer of neoplastic epithelium showing low-grade atypia; (2) an undulating morphology, often associated with mucin secretion and appendiceal dilatation; (3) villous or tubulo-villous adenoma, as typically seen elsewhere in the colon and rectum; (4) serrated adenoma with at least minimal or mild atypia, as cytological atypia is required to distinguish this from diffuse mucosal hyperplasia or hyperplastic polyp which show no atypia. Such benign mucinous tumours can be associated with dilatation of the appendix, sometimes with mucin dissection of the wall with or without rupture leading to release of mucin into the peritoneum to produce so-called “pseudomyxoma peritonei”.
Malignant mucinous tumours or adenocarcinomas of the appendix are diagnosed when there is evidence of invasion of the wall and are staged using a TNM staging system as for colorectal cancers. Lymph node metastasis tends to occur late, whereas peritoneal spread is more likely. However, well differentiated adenocarcinomas usually grow slowly and can invade on a broad front. Such invasion on a broad front by very well differentiated neoplasms showing minimal to mild atypia can sometimes be difficult to distinguish from an adenoma with appendiceal dilatation and wall thinning with rupture. Often the clinical significance of the finding of an appendiceal adenocarcinoma is due to the peritoneal spread with formation of “pseudomyxoma peritonei”. It is interesting to note that around 10% of appendiceal adenocarcinomas are associated with another primary colorectal cancer. The rare signet ring cell carcinoma of the appendix is diagnosed when greater than 50% of the tumour consists of malignant signet ring cells.
As well as the more clear cut benign and malignant mucinous tumours of appendix, there are those where it is difficult to distinguish well differentiated adenocarcinoma with broad-front invasion from adenoma with mucin dissection and wall rupture, hence cases falling into this “grey area” can be indicated by use of the term “mucinous tumour of uncertain malignant potential” [described in the chapter on adenocarcinomas of the appendix in the WHO classification of tumours] (Carr NJ, Arends MJ, Deans GT, & Sobin LH., 2000). However, to clarify this situation, Robert Young’s group (Misdraji, Young et al, 2003) have suggested a new classification system for appendiceal mucinous neoplasms, involving use of 2 major categories: low-grade appendiceal mucinous neoplasm (LAMN) and mucinous adenocarcinoma (MACA). The low-grade appendiceal mucinous neoplasms show low-grade atypia or low-grade architectural complexity and no evidence of destructive invasion (thus including well differentiated neoplasms with broad-front invasion). Whereas, the mucinous adenocarcinomas show high-grade atypia or high-grade architectural complexity or destructive invasion of the wall of the appendix. Follow-up studies demonstrated a difference in survival when comparing LAMN with extra-appendiceal spread versus mucinous adenocarcinoma and when comparing LAMN with peritoneal spread versus non-invasive mucinous adenocarcinoma with peritoneal spread (Misdraji et al, 2003).
Spread of mucinous tumours to the peritoneal cavity has traditionally been called “pseudomyxoma peritonei”, referring to the presence of mucinous material on peritoneal surfaces. However, “pseudmyxoma peritonei” is not a definitive pathological diagnosis, but a description of a clinico-pathological entity. The current view is that a case of “pseudomyxoma peritonei” can be more accurately described as falling into one of four types: (1) mucinous ascites, where acellular mucin is present in the peritoneum; (2) organising mucinous fluid, where the mucin has excited a peritoneal response of granulation tissue and inflammatory cells; (3) disseminated peritoneal adenomucinosis, where there is abundant intraperitoneal mucin but only scanty epithelium showing low-grade atypia; and (4) peritoneal mucinous carcinomatosis, in which the mucin is accompanied by abundant malignant epithelium showing high-grade atypia (Ronnett et al, 1995). The presence of mucinous carcinomatosis confers a poorer prognosis relative to the more common pattern of disseminated peritoneal adenomucinosis. For disseminated peritoneal adenomucinosis, the presence of mucin outside the right lower quadrant of the abdomen also associates with a poorer prognosis. To make the diagnosis accurately requires thorough sampling by both surgeon and pathologist (Ludeman & Shepherd, 2005).
Traditional theories of pathogenesis of “pseudmyxoma peritonei” include rupture of appendiceal or ovarian mucinous tumours, or neoplastic spread of mucinous tumours of appendix or ovary (or other sites), or metaplasia of the mesothelium. However, over the last decade or so, accumulation of immunohistochemical and molecular evidence (mainly loss of heterozygosity studies) has suggested that “pseudomyxoma peritonei” is almost certainly due to neoplastic spread or rupture of appendiceal mucinous tumours (even if synchronous ovarian mucinous tumours are present) to the peritoneum and that the ovarian mucinous tumours (when present) are also the result of neoplastic spread from an appendiceal primary mucinous tumour. Hence, appendicectomy is required in the presence of either “pseudomyxoma peritonei” or ovarian mucinous tumours and such specimens of appendix must be assessed histologically in their entirety for presence of mucinous tumours (LAMN or MACA can be associated with either “pseudomyxoma peritonei” (of any of the 4 types) or ovarian mucinous tumours, usually regarded as showing the pattern of borderline mucinous tumours of ovary).
Increased susceptibility to development of appendiceal mucinous tumours is seen in situations of increased predisposition to colorectal cancers, including patients with long-standing Ulcerative Colitis, Familial Adenomatous Polyposis (FAP), both of which are well documented, and also in Hereditary Non-Polyposis Colorectal Cancer (HNPCC) syndrome, which has only recently been described. In Familial Adenomatous Polyposis, there is inheritance of a germline mutation in one allele of the APC gene and the tumour somatically acquires a deletion or mutation in the second APC allele. Both appendiceal adenomas/LAMNs and adenocarcinomas are well recognised in FAP. In HNPCC, there is inheritance of a germline mutation in one allele of one of the DNA mismatch repair (MMR) genes MSH2, MLH1, MSH6. Subsequently, a founder cell for a tumour somatically acquires an inactivating alteration in the second allele of the same MMR gene, resulting in defective DNA mismatch repair, which can be identified by testing tumour DNA for the presence of microsatellite instability (MSI – variations in length of repetitive sequences such as AAAAAA….. or CACACACACA…). Alterations in expression of the DNA mismatch repair proteins MSH2, MLH1, MSH6 and PMS2 can be detected immunohistochemically in tumour cells in HNPCC-associated tumours.
Carcinoid or endocrine cell tumours of appendix are the most common (~50-85%) of all appendiceal tumours and are found in around ~1% of all appendicectomy specimens. Approximately ~20% of all carcinoids are found in the in appendix. All age groups can be affected, with carcinoid tumours commonest around 30-40years. Females have a higher incidence than males of appendiceal carcinoid tumours, but this may partially reflect the higher frequency of appendicectomy in females. Most carcinoid tumours are small in size (usually <1cm and rarely >2cm) and thus are mostly discovered as incidental findings. Liver metastasis is rare for appendiceal carcinoid tumours and hence the carcinoid syndrome is also rare for carcinoid tumours of the appendix. Most are serotonin-producing enterochromaffin (EC) cell carcinoid tumours. Around ~70% occur near the tip of appendix as ovoid tumours, but the mixed endocrine/exocrine carcinoid and goblet cell carcinoid tumours can occur in any part of the appendix and can be diffuse in growth pattern. Carcinoid tumours appear grey/white when fresh and yellow when formalin-fixed.
The origin of carcinoid tumours has been suggested to be either (1) EC cells scattered in the lamina propria (clustered with nerves & sustentacular cells) - EC carcinoid tumour cells can be seen to associate with small nerves and S100-positive sustentacular cells; or (2) EC cells in crypts - mixed endocrine/exocrine or goblet cell carcinoids have been suggested to arise from these cells, which would explain their appearance in any part of the appendix.
Carcinoid tumours of the appendix show 3 patterns: (1) argentaffin or enterochromaffin (EC) cell carcinoid tumours, which are the most frequent type, that usually produce seronin; (2) non-argentaffin (L-cell) carcinoid tumours, which are much less common and may produce glucagon-like peptide & PP/PYY peptides; and (3) mixed endocrine/exocrine tumours, including goblet cell carcinoid (mucinous or adenocarcinoid) and tubular carcinoid tumours (including microacinar variant of goblet carcinoid) – and this third group tends to behave more aggressively in terms of metastatic spread. The EC cell or argentaffin carcinoid tumour has solid & packeted nests of tumour cells with some peripheral palisading. The cytoplasm is NSE+, with granules that are Chromogranin A+ (CrgA+) and contain 5-hydroxytryptamine / serotonin. Some contain peptide hormones such as somatostatin, neurotensin, motilin, pancreatic polypeptide and substance P. The L-cell or non-argentaffin carcinoid tumours are usually smaller and appear similar to rectal carcinoids with columns, ribbons, trabeculae, and acini/tubules. They are positive for a range of peptides, are NSE+, synaptophysin+, chromogranin B+, but usually negative for CrgA. The third group of mixed endocrine / exocrine tumours is mostly represented by goblet cell carcinoid tumours. They often form clusters and strands of epithelial cells distended with mucin or signet-ring-like in appearance, which may resemble normal goblet cells but with nuclear compression. They are NSE+, CrgA+ and PGP9.5+.
In terms of prognosis, the first two groups of argentaffin & non-argentaffin carcinoid tumours are similar, in that carcinoid tumours showing the presence of invasion into muscle, perineural & lymphatic invasion, and penetration into the serosa including some localised involvement of the mesoappendix can still be cured by appendicectomy. Metastatic spread to the peritoneum, lymph nodes and liver is rare (and when present is usually seen at primary surgery). The most important prognostic parameter is appendiceal primary tumour size, as carcinoid tumours of less than 1cm almost never metastasise, those of 1-2cm rarely metastasise, whereas carcinoid tumours greater than 2cm have an approximate ~ 30% risk of metastasis. In addition, there is a poorer prognosis for goblet cell carcinoid tumours, as they seem to behave more aggressively with ~20% showing extra-appendiceal spread, including spread to peritoneal surfaces and ovaries and some showing involvement of lymph nodes or liver. These aggressive goblet cell carcinoid tumours tend to demonstrate high-grade atypia, a mitotic rate of >2/10HPF, and a “carcinomatous growth” pattern with fused or cribriform glands or sheets.
In conclusion, mucinous tumours of the appendix form a pathological spectrum, for which diagnostic categorisation has previously been dogged by non-uniform and inconsistent use of terminology. However, a new classification system has been proposed splitting them into two major groups: low-grade appendiceal mucinous neoplasms (LAMN) and mucinous adenocarcinomas. “Pseudomyxoma peritonei” is not a definitive pathological diagnosis, but includes 4 types which should be specified following thorough sampling. Increased susceptibility to mucinous tumours of the appendix occurs in long-standing Ulcerative Colitis, FAP and HNPCC, the latter should be considered if the patient is of young age, or has a positive family history, or there are multiple tumours at different sites. Carcinoid tumours are the commonest appendiceal tumours, can be classified into 3 groups and the potentially aggressive features to look for include primary appendiceal tumour size of greater than 2cm, and goblet cell carcinoid tumour pattern.
References:
1. Appendiceal Mucinous neoplasms: a clinicopathologic analysis of 107 cases. Misdraji et al., (2003), Am J Surg Path 27:1089.
2. Disseminated peritoneal adenomucinosis and peritoneal mucinous carcinomatosis. A clinicopathologic analysis of 109 cases with emphasis on distinguishing pathologic features, site of origin, prognosis, and relationship to "pseudomyxoma peritonei". Ronnett et al., (1995), Am J Surg Path 19:1390.
3. Adenocarcinoma of the appendix.
4. Serosal involvement in gastrointestinal cancer: its assessment and significance. Ludeman & Shepherd. (2005), Histopath 47:123.