Table 1.

Prespecified (unadjusted) and post hoc (adjusted) primary biomarker endpoint analyses: change in hs-CRP from baseline to follow-up

hs-CRP Median (IQR)	Baseline (mg/L)	14 + 3 Days (mg/L)	Median percent change from baseline	Median change from baseline (mg/L)	p valuea
Icosapent ethyl (n = 44)	3.2 (0.9, 11.6)	1.6 (0.6, 4.4)	−25.0 (−80.1, 26.7)	−0.5 (−6.9, 0.4)	0.011
Usual care (n = 47)	2.3 (0.7, 6.5)	2.1 (0.5, 5.8)	−5.6 (−57.1, 84.2)	−0.1 (−3.2, 1.7)	0.51

Within-group comparisons of hs-CRP levels for the IPE and usual care cohorts. A significant relative reduction of 25% was observed in the hs-CRP level (median change from baseline of −0.5, p = 0.011) in the IPE cohort while there were no significant changes in the usual care cohort (median change from baseline of −0.1, p = 0.51). The between-group difference was not significant for the unadjusted values (p = 0.082 for change from baseline hs-CRP), but after post hoc adjustment for age, sex, and baseline cardiovascular risk, the between-group p value was significant (p = 0.043 for change from baseline hs-CRP). Sex and age adjustments: men <45 versus ≥45 years and women <55 versus ≥55 years. Adjustments were warranted based on current literature and occurred post hoc (see STAR Methods section) (O’Driscoll et al., 2020). Baseline cardiovascular risk is described as the absence or presence of cardiovascular comorbidities. hs-CRP, high-sensitivity C-reactive protein; IQR, interquartile range.

^ap value (within-group median change from baseline); p value (between-group, unadjusted, prespecified) = 0.082; p value (between-group, adjusted, post hoc) = 0.043.