OBJECTIVEDetermine prevalence, progression rates, and associations of newly detectable macular atrophy (MA) in patients with choroidal neovascularization (CNV) secondary to neovascular age-related macular degeneration (nAMD) with/without ranibizumab treatment.DESIGNPost hoc analysis of MA in patients with occult/minimally classic nAMD who received monthly intravitreal ranibizumab (0.3 or 0.5 mg) or sham injections for 24 months (M) in MARINA, a phase III trial in treatment-naive patients (NCT00056836).PARTICIPANTSSeven hundred six patients with nAMD: ranibizumab 0.3 mg, n = 236; 0.5 mg, n = 237; sham, n = 233.METHODSMacular atrophy, assessed by color fundus photographs/fluorescein angiography, was classified as "within," "adjacent," or "nonadjacent" to the original CNV lesion. Factors associated with MA were assessed by multivariate logistic regression.MAIN OUTCOME MEASURESPrevalence/incidence of newly detectable MA over time, association with CNV area, MA progression rate, association of MA with visual acuity (VA), changes in CNV/leakage area, and factors predictive of new MA at 24M.RESULTSAt 24M, new MA was detected in 36.8%, 40.4%, and 21.0% of eyes for ranibizumab 0.3 mg, 0.5 mg, and sham, respectively, most frequently within the area of the baseline CNV lesion (93.2%, 85.0%, and 69.0%). Rate of MA progression was similar across arms (∼ 0.3 to 0.4 mm/year). There was strong association between absence of fibrosis and detectable MA (odds ratio, 2.7; 95% confidence interval [CI], 1.29-5.56), whereas an association was not identified between detectable MA and baseline VA, baseline fellow eye atrophy, ranibizumab treatment, or change in leakage/CNV area at 24M. Ranibizumab-treated eyes gained VA with (0.3 mg: 5.3 letters [95% CI, -3.3, 13.8]; 0.5 mg: 9.8 [4.7-15.0]) or without new MA (0.3 mg: 6.4 [4.1-8.6]; 0.5 mg: 8.0 [5.3-10.6]), whereas VA in sham-treated eyes deteriorated with/without new MA (-14.7 [-23.6, -5.8] and -14.0 [-16.9, -11.1], respectively).CONCLUSIONSNew MA was more frequently detected in ranibizumab-treated than sham-treated eyes. Macular atrophy progression was similar across arms. Multivariate analysis showed that absence of fibrosis was the only variable associated with increased MA. Regardless of MA presence/location at baseline or throughout the study, ranibizumab-treated eyes showed clinically significant improvements in VA, whereas VA in sham-treated eyes worsened.FINANCIAL DISCLOSURE(S)Proprietary or commercial disclosure may be found after the references.