Skip to main content
. Author manuscript; available in PMC: 2012 Jul 18.
Published in final edited form as: Gastroenterology. 2010 Jun 9;139(3):797–805.e1. doi: 10.1053/j.gastro.2010.06.005

Table 3.

Treatment efficacy in preventing metachronous colorectal adenoma, stratified by biomarker level at baseline

Biomarker level at baseline*
Low
High
P
PGE2
Placebo, events/total (%) 19/42 (45.2) 21/53 (39.6)
DFMO/sulindac, events/total (%) 12/53 (22.6) 3/44 (6.82)
RR (95% CI) 0.50 (0.28 – 0.91) 0.17 (0.05 – 0.54) 0.132
Spd:Spm
Placebo, events/total (%) 31/68 (45.6) 24/61 (39.3)
DFMO/sulindac, events/total (%) 5/64 (7.81) 12/72 (16.7)
RR (95% CI) 0.17 (0.07 – 0.41) 0.42 (0.23 – 0.77) 0.087
Putrescine
Placebo, events/total (%) 24/62 (38.7) 31/67 (46.3)
DFMO/sulindac, events/total (%) 7/70 (10.0) 10/66 (15.2)
RR (95% CI) 0.26 (0.12 – 0.56) 0.33 (0.18 – 0.61) 0.796

Abbreviations: DFMO, difluoromethylornithine; PGE2, prostaglandin E2; RR, relative risk; CI, confidence interval; Spd:Spm, spermidine-to-spermine ratio

*

Low versus high determined by the median biomarker level: low is below the median; high is above the median

P value for the likelihood ratio test of interaction between treatment group and low versus high biomarker level at baseline