48uep6bbphidcol2|ID

48uep6bbphidvals|1751

48uep6bbph|2000F98CTab_Articles|Fulltext

Periampullary carcinoma (PC) is a heterogeneous group of cancers that includes ampullary, pancreatic head, distal bile duct and duodenal cancers. The incidence of these cancers in resected pancreato-duodenectomies varies in different parts of the world with pancreatic head cancer being the most common in the west (5% to 25%).^1-5 In the Indian population, the ampullary carcinoma forms the bulk.^6-7 

Expression of mismatch repair (MMR) proteins is the first line screening test for detecting microsatellite instability (MSI). MSI is one of the important pathogenetic mechanisms in Lynch syndrome. Studies investigating the prognostic value of microsatellite instability (MSI) in PC that have shown that tumors with MSI show significantly longer 5 year survival and better prognosis than those with low MSI (MSI-L) or microsatellite stable (MSS) tumors.^8-9 MSI is responsible for 15%-20% of colorectal cancer and a similar percentage of periampullary cancers have been shown to have MSI.¹⁰ The incidence of MSI in periampullary carcinoma reported in  literature varies widely from 0% to 22% in different studies using different methodologies. Intestinal type PC have shown to be associated more frequently with MSI than pancreatobiliary type PCs.⁸

The present study was planned to evaluate MMR protein expression as a marker for MSI in periampullary carcinoma and to correlate the loss of expression with its anatomical and histological subtypes and clinicopathological factors.

 

One hundred and eight consecutive cases of PC were retrieved retrospectively within a period of three years. All the cases were histologically classified into intestinal, pancreatobiliary, mixed or other types using a combination of morphological criteria given by Albores-Saavedra et al¹¹ and panel of IHC markers. The markers used for pancreatobiliary differentiation were MUC1 (polyclonal, dilution 1:50, Thermoscientific), CK17 (clone E3, dilution 1:40, Thermoscientific) and CK7 (clone OV-TL 12/30, dilution 1:80, Dako); and for intestinal differentiation were MUC2 (clone Ccp58, dilution 1:100, Novocastra), CDX2 (clone MU392A.UC, dilution 1:50, Biogenex) and CK20 (clone Ks20.8, dilution 1:8, Dako). IHC for MMR proteins included MLH1 (clone ES05, Novocastra, dilution 1:25), MSH2 (clone 25D12, Novocastra, 1:50), MSH6 (clone PU29, Novocastra, dilution 1:50) and PMS2 (clone EP51, DAKO, dilution 1:40). Antigen retrieval was done by microwave treatment using the EZ Antigen retriever (Biogenex). This was done in citrate buffer at pH 6.0 for MUC1, MUC2, MLH1, MSH2, MSH6 and CK17 and in EDTA buffer at pH 8.0 for CDX2, CK7, CK20 and PMS2 for 30 minutes. Sections were incubated with primary antibody at room temperature for one hour followed by secondary antibody (Dako LSAB Envision Universal kit) for 30 minutes at room temperature. Diaminobenzidine (DAB) was used as the chromogen. The slides were washed and counterstained with Mayer’s hematoxylin. MMR expression was considered lost only when there was a complete absence of nuclear staining in the presence of a positive control. 

 

Correlation of expression loss and intact expression of MMR proteins with different histological parameters were analyzed using chi-square test and survival analysis between these two groups was done using Kaplan Meier curves using log-rank test. Probability (p) values of < 0.05 obtained by the two tailed significance were regarded as statistically significant. SPSS software program, version 15 was used for all statistical analyses.

 

Ampullary cancers accounted for 91 (84.2%) cases of the 108 periampullary carcinomas. Carcinoma head of pancreas constituted another 9 (8.3%), distal CBD cancers 5 (4.6%) and duodenal cancers 3 (2.7%) cases. Using morphological criteria and IHC markers, 41 (38%) cases were classified as intestinal type, 54 (50%) cases as pancreatobiliary type, 6 (5.6%) cases as mixed type and 7 (6.5%) cases were put in the category of others type  (Table 1). Overall expression loss of MMR proteins was seen in 9 (8.3%) cases (Figure 1 a-e). The pattern of MMR protein loss is shown in Table 2. Among the histological types, loss of MMR protein was equally distributed between intestinal and pancreatobiliary types with 4 cases each and one case was of signet ring cell carcinoma. 6 of the patients with loss of MMR expression were male. There was no significant correlation between MMR expression loss and clinicopathological parameters in periampullary carcinoma (Table 3). The MMR expression loss was then correlated with histological differentiation. The intestinal type tumors with loss of MMR protein had tendency to have low grade, less lymphovascular invasion, low T-stage and low overall TNM stage (Table 4), however the difference was not found to be statistically significant. The pancreatobiliary tumors with loss of MMR protein expression on the other hand had a tendency for higher grade, more perineural invasion, earlier lymph node metastasis, higher T-stage and higher overall TNM stage (Table 4), however again the differences were not statistically significant.

 

Different studies have found variable rates of MSI or loss of MMR expression in periampullary cancers, ranging from 0% to 22%.^8,12-16 This may be attributed to the difference in methodologies used for MSI analysis, although the sensitivity of 92% and specificity of 99% for IHC to detect MSI in colorectal carcinoma has been shown in different studies.^18-20 It can also be explained by the different anatomic locations studied in different series. The role of MSI in periampullary carcinoma also varies with some studies reporting to have no relation with clinicopathological parameters whereas others show association with a better prognosis. Achille et al found MSI of about 20% in a series of 25 cases of ampullary carcinoma and reported a significant association between MSI and longer patient survival.¹² Imai et al reported a high rate of MSI (22.2%) and markedly frequent alteration of the TGF-b-receptor-II gene (77.8%) in ampullary carcinomas.¹³ Using PCR, Scarpa et al found 15.1% and Suto et al found 12.5% of ampullary carcinoma showed MSI.^14-15 Ruemmele et al found a sensitivity of 79% and a specificity of 100% for predicting MSI in ampullary carcinoma using MLH1, MSH2 and MSH6 immunostaining. They noticed 10% (14 cases) MSI-H in 144 ampullary adenocarcinoma and 7% in adenoma which was accompanied by loss of MMR protein in 79% and 50% cases respectively and concluded that MSI occurs as an early event at the stage of adenoma. They found that ampullary carcinoma with MSI had significantly longer survival and better prognosis than those with MSI-L or MSS cancers.⁸ Eleven of 15 cases were intestinal type, 3 were pancreatobiliary type and one case was of invasive papillary carcinoma. They found tumor grade, histologic subtype and survival to be significant correlated with MSI status on univariate analysis and only tumor grade and survival on multivariate analysis. Sessa et al showed MSI-H in 5 (9.5%) of 53 cases of ampullary carcinoma of whom all were of intestinal type and half were associated with mucinous or medullary component similar to colon cancer.⁹ In their study MSI status significantly affected survival on univariate analysis although it was not significant in multivariate analysis. Park et al and Rashid et al did not find MSI in any of their cases of ampullary carcinoma.^16-17 Suto et al reported MSI-H in 12.5% cases of ampullary carcinoma and did not find any correlation of clinicopathological parameters with MSI status.¹⁵

In the present study rate of loss of MMR protein in periampullary carcinoma (8.3%) was similar to those reported in the literature. The MMR expression loss in PC did not show any correlation with clinicopathological parameters. However, when it was correlated with histological differentiation, intestinal type with MMR loss featured better than pancreatobiliary with MMR loss. Since histological differentiation of intestinal and pancreatobiliary in periampullary carcinoma remains the key determinant for predicting prognosis, this finding holds little value. This is reiterated in the fact that the survival as well where median survival in periampullary carcinoma was marginally higher in cases with MMR but this difference disappeared when histological types were considered. 

 

Microsatellite instability is one of the important pathogenetic mechanisms in carcinogenesis. However, the present study did not find any evidence to suggest that it may impact the prognosis in any way. The difference in prognostic parameters in periampullary carcinoma were influenced by histological differentiation of intestinal and pancreatobiliary type rather than expression loss of MMR proteins.