e15800 Background: Small bowel adenocarcinoma (SBA) is a rare malignancy of the digestive tract with limited knowledge about its genetic alterations. Methods: Eighteen tumor and normal matched samples were sequenced using the whole-exome-Illumina platform. Various agnostic analysis methods were used to identify relevant somatic mutations and the results were compared with the available TCGA datasets. To test potential targetable mutations we developed eight patient-derived xenografts (PDX) and three cell lines from freshly-collected primary or metastatic SBA tissues. Two kinase-activating ErbB2 mutations (V842I and Y803H) and one wild-type (wt) ErbB2 (n = 30) PDX models were used to measure the effect of Dacomitinib (Daco) on tumor growth. ErbB2-mutant tumor-derived cell lines were tested for Daco and Lapatinib (Lapa) response in-vitro. Reverse-phase-protein-arrays (RPPA) were used to identify molecular changes. Results: Several unexpected oncogenic mechanisms were suggested by the pattern of somatic mutations across the dataset, including mutations in the Notch and Hippo pathways. In addition, we found 6/18 APC truncating mutations exclusive to mutations in ZNRF3 or RNF43 genes, suggesting that non-APC wnt-activating mechanisms are important in SBA, while considered a minority in CRC. Importantly, 6/18 samples displayed ErbB2 mutations, of which 4 resided in the kinase domain, D769Y, V777L, Y803H and V842I. Nanomolar doses of both Lapa and Daco significantly inhibited ErbB2-mutants’ cell proliferation in-vitro. Target inhibition was confirmed by the RPPA results: EGFR-Y1068, ErbB2-Y1248 and PKCα-S657 dephosphorylation, PCNA reduction and PARP cleavage increase. In-vivo-administered Daco resulted in significant tumor reduction in ErbB2-V841I (39%, p = 0.03) and ErbB2-Y803H (59%, p = 0.03) tumors, and had no anti-tumor effect on wt-ErbB2 tumors. Conclusions: The generation of in-vitro and in-vivo model systems from rare cancers is possible and provides a valuable resource into understanding potentially relevant targetable mutations. Our findings suggest that SBA patients with ErbB2-activating mutations should be considered for clinical trials targeting this alteration.