Table 5.
18q LOH and Patient Mortality in Non–MSI-High Colorectal Cancer in Strata of Tumor Location
| Tumor Location | Total No. | HR for Colorectal Cancer–Specific Mortality in 18q LOH-Positive Patients* | HR for Overall Mortality in 18q LOH-Positive Patients* |
||||||
|---|---|---|---|---|---|---|---|---|---|
| Stage-Matched HR | 95% CI | Multivariate HR | 95% CI | Stage-Matched HR | 95% CI | Multivariate HR | 95% CI | ||
| Rectum | 138 | 1.34 | 0.66 to 2.75 | 1.46 | 0.70 to 3.07 | 1.13 | 0.64 to 1.98 | 1.10 | 0.62 to 1.96 |
| Distal colon, splenic flexure to sigmoid | 174 | 1.30 | 0.59 to 2.88 | 1.78 | 0.78 to 4.05 | 0.88 | 0.53 to 1.47 | 1.01 | 0.60 to 1.71 |
| Proximal colon, cecum to transverse colon | 217 | 0.87 | 0.54 to 1.39 | 0.95 | 0.57 to 1.58 | 0.92 | 0.62 to 1.38 | 1.01 | 0.66 to 1.56 |
| P for interaction (18q LOH and location)† | .26 | .17 | .81 | .91 | |||||
NOTE. The multivariate, stage-matched (stratified) Cox regression model included the 18q LOH variable stratified by location, age, year of diagnosis, sex, family history of colorectal cancer, tumor grade, KRAS, BRAF, PIK3CA,p53, β-catenin, JC virus T antigen, LINE-1 methylation, MSI (low v microsatellite stability), and CpG island methylator phenotype.
Abbreviations: LOH, loss of heterozygosity; MSI, microsatellite instability; HR, hazard ratio.
*
Versus 18q LOH-negative patients as a referent.
†
For assessing a potential interaction, tumor location was dichotomized as proximal v distal (including rectum).