Table 3:

Risk estimates for early-onset versus late-onset CRC associated with a 95-SNP PRS among participants with and without Lynch Syndrome, in the Ohio cohort^a

PRS per 1 SD	N (cases)	N (controls)	OR (95% CI)	P value	P value for interactionb
Including Lynch and Non-Lynch Cases
All Subjects					0.0369
<50 Years-Old	574	979	1.73 (1.54, 1.95)	1.39E-19
≥50 Years-Old	2525	1463	1.47 (1.37, 1.58)	1.77E-28
Negative Family History					0.0106
<50 Years-Old	449	931	1.81 (1.59, 2.07)	9.64E-19
≥50 Years-Old	1885	1271	1.45 (1.34, 1.56)	1.16E-21
Positive Family History					0.1517
<50 Years-Old	106	48	1.28 (0.84, 1.97)	0.2530
≥50 Years-Old	565	192	1.55 (1.30, 1.84)	1.12E-06
Excluding Lynch Cases
All Subjects					0.0149
<50 Years-Old	537	979	1.82 (1.61, 2.06)	2.63E-21
≥50 Years-Old	2471	1463	1.49 (1.39, 1.60)	1.11E-29
Negative Family History					0.0107
<50 Years-Old	438	931	1.83 (1.60, 2.09)	7.50E-19
≥50 Years-Old	1856	1271	1.46 (1.35, 1.57)	4.30E-22
Positive Family History					0.5627
<50 Years-Old	80	48	1.53 (0.98, 2.41)	0.0635
≥50 Years-Old	540	192	1.61 (1.34, 1.92)	2.34E-07

^aThe logistic regression models include age, sex, principal components, and polygenic risk score.

^bP value produced from interaction term with continuous PRS (per SD) and age (<50 versus ≥50 years).