K E Y P O I N T S l Assessing pretreatment genomic aberrations improves risk stratification of FL independent of clinical/ mutation markers. l Gain or cnLOH of 2p is correlated with 2-year progression; CDKN2A/B deletion with worse PFS; CREBBP and TP53 deletions predict worse OS.Although recent advances in molecular genetics have enabled improved risk classification of follicular lymphoma (FL) using, for example, the m7-FLIPI score, the impact on treatment has been limited. We aimed to assess the prognostic significance of copy-number aberrations (CNAs) and copy-neutral loss of heterozygosity (cnLOH) identified by chromosome genomic-array testing (CGAT) at FL diagnosis using prospectively collected clinical trial specimens from 255 patients enrolled in the SWOG study S0016. The impact of genomic aberrations was assessed for early progression (progressed or died within 2 years after registration), progression-free survival (PFS), and overall survival (OS). We showed that increased genomic complexity (ie, the total number of aberration calls) was associated with poor outcome in FL. Certain chromosome arms were critical for clinical outcome. Prognostic CNAs/cnLOH were identified: whereas early progression was correlated with 2p gain (P 5 .007; odds ratio [OR] 5 2.55 [1.29, 5.03]) and 2p cnLOH (P 5 .005; OR 5 10.9 [2.08, 57.2]), 2p gain specifically encompassing VRK2 and FANCL predicted PFS (P 5 .01; hazard ratio 5 1.80 [1.14, 2.68]) as well as OS (P 5 .005; 2.40 [1.30, 4.40]); CDKN2A/B (9p) deletion correlated with worse PFS (P 5 .004, 3.50 [1.51, 8.28]); whereas CREBBP (16p) (P < .001; 6.70 [2.52, 17.58]) and TP53 (17p) (P < .001; 3.90 [1.85, 8.31]) deletion predicted worse OS. An independent cohort from the m7-FLIPI study was explored, and the prognostic significance of aberration count, and TP53 and CDKN2A/B deletion were further validated. In conclusion, assessing genomic aberrations at FL diagnosis with CGAT improves risk stratification independent of known clinical parameters, and provides a framework for development of future rational targeted therapies. (Blood. 2019;133 (1):81-93) Follow-up for PFS, median (min, max), mo 89 (1, 173) 24 (11, 151) 86 (7, 173) 95 (1, 157) Follow-up for OS, median (min, max), mo 114 (2, 177) 135 (47, 177) 116 (12, 175) 111 (2, 175)*CHOP-R, 6 cycles of CHOP chemotherapy at 3-week intervals with 6 doses of rituximab; CHOP-RIT, 6 cycles of CHOP followed by consolidation with iodine 131 I tositumomab radioimmunotherapy. †LDH is not available from 5 patients, 4 on CHOP-R, and 1 on CHOP-RIT; FLIPI risk is not available from 11 patients, all on CHOP alone. ‡Early progression is defined as progression or death within 2 years after registration.