Table 2. Multivariate logistic regression analysis to assess association between infiltrative tumour growth pattern and loss of CDH1 expression.
| Predictors for infiltrative tumour growth pattern (vs expansile-intermediate growth pattern) | Multivariate OR (95% CI) | P |
|---|---|---|
| Loss of CDH1 expression (vs intact) | 2.02 (1.23–3.34) | 0.006 |
|
Covariatesa | ||
| MSI-high (vs MSI-low/MSS) | 0.11 (0.04–0.31) | <0.0001 |
| BRAF mutation (vs wild type) | 4.44 (2.28–8.66) | <0.0001 |
| Loss of CTNNB1 membrane expression (vs intact) | 1.62 (0.98–2.68) | 0.06 |
Abbreviations: CI=confidence interval; CIMP=CpG island methylator phenotype; LINE-1=long interspersed nucleotide element 1; MSI=microsatellite instability; MSS=microsatellite stable; OR=odds ratio.
Covariates included in the initial models were as follow: sex, age (continuous, increase by 10 years), year of diagnosis of colorectal cancer (continuous, increase by 1 year), family history of colorectal cancer in first degree relatives (absent vs present), tumour location (caecum, ascending to transverse colon, splenic flexure to sigmoid colon, rectum), MSI (low/MSS vs high), CIMP (negative/low vs high), LINE-1 methylation (10% decrease, continuous), BRAF, KRAS, PIK3CA mutations and CTNNB1 membrane expression (intact vs lost). A backward stepwise elimination with a threshold of P=0.10 was used to select variables in the final models.