The Malmö Diet and Cancer Study (MDCS) is a population-based prospective cohort study with the primary aim to examine whether a Western diet rich in fat and low in fruit and vegetables increases the risk of certain forms of cancer. Between 1991–1996, a total number of 30 446 individuals; 12 120 (39.8%) men and 18 326 (60.2%) women between 44–74 years where enrolled (from a background population of 74 138) 27 . Until 31 Dec 2008, 626 incident cases of CRC had been registered in the MDCS. Cases were identified from the Swedish Cancer Registry up until 31 Dec 2007, and from The Southern Swedish Regional Tumour Registry for the period of 1 Jan - 31 Dec 2008. All tumours with available slides or paraffin blocks were histopathologically re-evaluated on haematoxylin and eosin stained slides. Histopathological, clinical and treatment data were obtained from the clinical and/or pathology records. TNM staging was performed according to the American Joint Committee on Cancer (AJCC). Information on vital status and cause of death was obtained from the Swedish Cause of Death Registry up until 31 Dec 2009. Follow-up started at date of diagnosis and ended at death, emigration or 31 Dec 2009, whichever came first. None of the CRC cases registered until 31 Dec 2008 was lost due to emigration during follow-up. Median follow-up time was 3.35 years (range 0–17.69) for the full cohort (n = 626) and 6.05 years (range 1.03-17.69) for patients alive (n = 344). Patient and tumour characteristics of the cohort have been described in detail previously 26 28 29 . Ethical permission was obtained from the Ethics Committee at Lund University for the MDCS (Ref. 51/90), and the present study (Ref. 530/2008).

Cases with an insufficient amount of tumour material were excluded, whereby a total number of 557 (89.0%) tumours were suitable for tissue microarray (TMA) construction. Areas representative of cancer were marked on haematoxylin & eosin stained slides and TMAs were constructed as previously described 26 28 . In brief, two 1.0 mm cores were taken from each tumour and mounted in a new recipient block using a semi-automated arraying device (TMArrayer, Pathology Devices, Westminster, MD, USA). As demonstrated previously, there was no selection bias regarding the distribution of clinicopathological characteristics between the TMA cohort and the full cohort 28 .

For immunohistochemical analysis, 4 μm TMA-sections were transferred to glass slides (Menzel Super Frost Plus), dried at room temperature, and baked in a heated chamber for 2 hours at 60°C. Deparaffinization and antigen retrieval was performed using the PT Link system (DAKO, Glostrup, Denmark) in a high pH buffer. Staining was performed in an Autostainer Plus (DAKO) with monoclonal antibodies against p21 (clone SX118, DAKO) diluted 1:25, p27 (clone SX53G8, DAKO) diluted 1:100, and p53 (clone DO-7, DAKO) diluted 1:100, using a di-amino-benzidine (DAB) based visualization kit (K801021-2, Dako Denmark A/S). Counterstaining was performed using Mayer’s hematoxylin. Immunohistochemical staining and evaluation of cyclin D1 expression had been performed previously using a monoclonal anti-cyclin D1 antibody DSC-6 (DAKO) diluted 1:50 26 . Nuclear expression of p21, p27 and cyclin D1 was evaluated in a similar fashion, taking both the fraction of positive cells and staining intensity into account. Specifically, the nuclear staining intensity was denoted as 0 (negative), 1 (weak), 2 (moderate) or 3 (strong), and the proportion of positive tumour nuclei as 0 (0 to 1%), 1 (2 to 25%), 2 (26 to 50%), 3 (51 to 75%) and 4 (> 75%). p53 positivity was defined as > = 50% tumour cells with strong nuclear staining intensity in acordance with previous studies 30 .

For definition of MSI screening status, immunohistochemical staining for MLH1, PMS2, MSH2 and MSH6 was performed as previously described 29 . Immunohistochemical stainings were evaluated as negative when all tumour cells showed loss of nuclear staining. Surrounding stromal cells and tumour infiltrating lymphocytes served as internal controls for each biopsy core. A nuclear reaction of tumour cells was assessed as a positive staining. MSI screening status was evaluated using monoclonal antibodies against MLH1 (Clone ES05, DAKO, diluted 1 : 100), PMS2 (Clone A16–4, 556415, BD Pharmingen, San Diego, CA, USA, diluted 1 : 300), MSH2 (Clone FE11, NA27, Calbiochem, San Diego, CA, USA, diluted 1 : 100), and MSH6 (EPR3945, Epitomics, Burlingame, CA, USA, diluted 1 :100) as previously described 31 . Immunohistochemical stainings were evaluated as negative when all tumour cells showed loss of nuclear staining. Surrounding stromal cells and tumour infiltrating lymphocytes served as internal controls for each biopsy core. Any nuclear reaction of tumour cells was assessed as positive staining. Tumour samples lacking nuclear staining of MLH1, PMS2, MSH2 or MSH6 were considered to have a positive MSI screening status, and referred to as MSI, and tumours with negative MSI screening status are referred to as MSS.

Immunohistochemical staining of beta-catenin was performed with a monoclonal anti-beta-catenin antibody (# 610153 BD Pharmingen, San Diego, Ca, USA), diluted 1:5000. The staining was evaluated as previously described 32 33 , whereby membranous staining was denoted as 0 (present) or 1 (absent), cytoplasmic staining intensity as 0–2 and nuclear staining intensity as 0–2. The total score ranging from 0 (corresponding to membrane staining only, as in normal colonic mucosa) to 5 (tumours with strong nuclear and cytoplasmic staining) was then divided into three categories; 1 = score 0–1, 2 = score 2–3 and 3 = score 4–5).

All immunohistochemical stainings were evaluated by two independent observers (SW and KJ), who were blinded to clinical and outcome data. Scoring differences were discussed in order to reach consensus.

Chi Square and Spearman’s correlation (R) tests were used to explore the associations between clinicopathological and investigative parameters. Kaplan-Meier analysis and log rank test were used to illustrate differences in cancer specific survival (CSS) according to beta-catenin overexpression and MSI screening status. Cox regression proportional hazards models were used for estimation of hazard ratios (HRs) for death from CRC in both uni- and multivariable analysis, adjusted for age, gender, T-stage, N-stage, M-stage, differentiation grade and vascular invasion. A backward conditional selection method was used for variable selection by the model. All tests were two-sided. A p-value of 0.05 was considered significant. All statistical analyses were performed using IBM SPSS Statistics version 20.

Representative immunohistochemical images of the investigative markers in tumours from four different cases are shown in Figure 1A & B.

As seen in Table 1 there were significant associations between MSI and older age, female sex, proximal tumour location, non metastatic disease (N0,M0) and low differentiation grade. Further, beta-catenin expression was associated with distal tumour location, less advanced T-stage and intermediate or high tumour differentiation grade.

Columnar percentage in parentheses. χ2-test of association was used for 2x2 tables and χ2-test for linear trend for tables with more than two rows and/or columns. The categories marked as missing were not included in the analysis. *Significant at the 0.05 level. **Significant at the 0.01 level. N1 = 1-3 positive nodes, N2= > 4 positive nodes. MSS = Microsatellite stable, MSI = Microsatellite unstable. Proximal colon = appendix, cecum, ascending colon, hepatic flexure; descending colon = splenic flexure and descending colon.

Associations between beta-catenin overexpression, MSI screening status and investigative factors are shown in Table 2. MSI screening status was inversely associated with beta-catenin overexpression. MSI correlated significantly to cyclin D1 expression (fraction and intensity), p21 expression (fraction and intensity), and inversely with p53 expression and p27 expression (fraction and intensity). Beta-catenin overexpression correlated significantly to p53 expression, cyclin D1 expression (fraction), and p27 expression (fraction and intensity).

Columnar percentage in parentheses. χ2-test of association was used for 2x2 tables and χ2-test for linear trend for tables with more than two rows and/or columns. The categories marked as missing were not included in the analysis. *Significant at the 0.05 level. **Significant at the 0.01 level. MSS = Microsatellite stable, MSI = Microsatellite unstable.

Kaplan-Meier analysis showed significant correlations between MSI and prolonged CSS in the whole cohort (Figure 2A), and in stage III-IV disease (Figure 2E) but not in stage I-II disease (Figure 2C). Beta-catenin overexpression correlated significantly to a prolonged CSS in stage III-IV disease only (Figure 2F).

Uni- and multivariable (adjusted for age, T-stage, N-stage, M-stage, differentiation grade and vascular invasion) Cox regression analysis confirmed the significant correlation between MSI and an improved CSS, both in the full cohort and in stage III-IV disease (Table 3). Beta-catenin overexpression was also associated with an improved CSS, in the full cohort and patients with stage III-IV disease, although these associations were less significant than for MSI status (Table 3). Notably, while only reaching statistical significance in multivariable analysis, the HRs for beta-catenin overexpression should be considered as being similar in univariable and multivariable analysis (0.75 vs 0.70 in the full cohort and 0.73 vs 0.67 in stage III-IV disease).

*Significant at the 0.05 level. **Significant at the 0.01 level. P-value from multivariable analysis adjusted for age (>/<=75 years), gender, T-stage (I-II, III, IV), N-stage (0,1,2), M-stage (0, 1), differentiation grade (high-intermediate vs low) and vascular invasion (+/−/unknown). Beta-catenin low = beta-catenin score 0–2. Beta catenin high = beta-catenin score 3–5.

Next, we examined the potential treatment predictive value of MSI screening status and beta-catenin overexpression. Subgroup analysis of stage III disease patients in strata according to adjuvant treatment (no treatment, 5-fluorouracil + leucovorin or 5-fluorouracil + leucovorin + oxaliplatin) revealed no significant correlations of MSI or beta-catenin overexpression and CSS (Figure 3).

The prognostic and treatment predictive value of MSI and beta-catenin overexpression did not differ according to gender (data not shown). Similar trends for the prognostic and treatment impact of MSI status and beta-catenin expression were seen for 5-year overall survival (data not shown).