Abstract Highly penetrant mutations responsible for Mendelian cancer syndromes have not been systematically identified in large populations. Historically, highly penetrant cancer mutations have been identified in large families with multiple cancer cases using linkage analysis. Next-generation sequencing (NGS) allows for efficient genomic screening of large numbers of familial cancer cases. Highly penetrant mutations such as those found in the APC or mismatch repair (MMR) genes for colorectal cancer (CRC) or BRCA1/2 genes for breast cancer, are rare and often population-specific. Therefore, to identify these rare, highly penetrant cancer-predisposing mutations using NGS requires a large sample size that can make this approach prohibitively expensive. Here, we present a new, robust NGS pooling strategy for identification of rare, highly penetrant mutations. Our strategy is at least five times less expensive than traditional germline NGS of cancer patients. We screened 1,046 CRC cases and 1,006 unaffected controls from Colon Cancer Family Registry (Colon CFR), to identify highly penetrant CRC mutations in 40 known or candidate CRC genes. Cases were patients with familial CRC and controls were spouses or unrelated healthy individuals. DNA from 1,046 cases and 1,006 controls were combined into 480 pools (3 cases/pool and 8 controls/pool) for targeted sequencing of the 40 selected genes with mean depth of >1000X in cases and >700X in controls using Qiagen GeneRead Customized Targeted Panel. Following quality control, alignment of reads, and calculation of allele frequencies, we selected likely functional variants (nonsynonymous and nonsense) in cases that were not found in controls and absent or very rare (minor allele frequency <0. 001) in other populations. We found 264 rare variants in 273 cases: six stopgain and 258 nonsynonymous variants. While, >40% (n = 106) of the variants have never been reported, almost all the variants identified in the MMR genes (MLH1, MSH2, MSH6, and PMS2) were reported in dbSNP. However, only 11 of 39 MMR variants have sufficient evidence to be considered pathogenic in ClinVar. In addition to known mutations, we found novel, deleterious variants in ABCC1, APC (including stopgain), AXIN2, BMPR1A, CHEK2, CTNNB1, ENG, MUTYH, POLD1, POLE, PTCH1, SH2B3, TGFBR2, and TP53 among other genes. Previously observed variants of unknown significance (VUS) were found in BMPR1A, EPCAM, and SMAD4. Deleterious variants were selected for Sanger sequencing to identify individual genotypes and validate NGS results. Our study identified a large proportion of previously unreported, likely deleterious variants in known or candidate CRC susceptibility genes. The targeted NGS pooling strategy used in this study proved to be a robust and cost-effective alternative for screening a large cohort of familial CRC cases and controls for mutations in CRC susceptibility genes, making it applicable for screening other cancer types and complex diseases with dominant mode of inheritance. Citation Format: Leon Raskin, Yan Guo, Stephen B. Gruber, Daniel D. Buchanan. Robust targeted sequencing strategy identifies highly penetrant mutations in hereditary colorectal cancer cases from Colon Cancer Family Registry. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr LB-372.