The objective of this study was to assess the genomic prediction accuracy (GPA) of antibody response to PRRS in purebred sows and crossbred gilts. Data on 2,180 commercial crossbred gilts (CrossData), from 7 breeding companies, and 512 purebred multiplier Landrace sows (PureData), from one breeding company, were used to assess the ability to predict PRRS antibody response, measured as sample-to-positive (S/P) ratio, using 38,191 single nucleotide polymorphisms (SNPs). S/P ratio was measured in the CrossData at 40.7 ± 16 d after gilts entered commercial herds with a history of health problems, whereas in the PureData, S/P ratio was measured 46 d after a natural PRRS outbreak. Two prediction strategies were used: 1) CrossData was used for training and PureData for validation; 2) 7-fold cross-validation using the CrossData. Previous results showed that S/P ratio is mainly controlled by 2 regions on chromosome 7 (SSC7), at 24–31 Mb and 128–131 Mb. Therefore, different sets of SNPs were used for prediction: all SNPs (All_SNP), SNPs on SSC7 24–31 Mb (QTL1_SNP), SNPs on SSC7 128–131 Mb (QTL2_SNP), SNPs on both regions (SSC7_SNP), all SNPs except those on the SSC7 regions (Not7_SNP). GPA was measured as the correlation between genomic estimated breeding values and pre-adjusted phenotypes, divided by square root of heritability. Heritability estimates were 0.46 (PureData) and 0.31(CrossData). When training using the CrossData and validating on PureData, GPAs were 0.49 (All_SNP), 0.55 (QTL1_SNP), 0.30 (QTL2_SNP), 0.63 (SSC7_SNP), and 0.15 (Not7_SNP). GPAs were slightly lower using from cross-validation in the CrossData: when averaging across folds: 0.32 (All_SNP), 0.30 (QTL1_SNP), 0.27 (QTL2_SNP), 0.39 (SSC7_SNP), and 0.1 (Not7_SNP). The highest and lowest GPA from individual folds were −0.14 (Not7_SNP) and 0.60 (SSC7_All). These results show that S/P ratio can be accurately predicted using SNPs in pure and crossbred female pigs. In addition, greater accuracy can be obtained using the 2 QTL on SSC7 than the whole genome. Financial support from Genome Canada, the Canadian Swine Health Board, and PigGen Canada.