473 Background: The growth of intestinal epithelium is dependent on the Wnt/β-catenin signaling pathway. The R-Spondin (RSPO) family of proteins form complexes with Wnt ligand and RNF43 protein, which promote ubiquitination and degradation of Wnt receptors. Patients with gastrointestinal cancers harboring RSPO3 gene fusions or RNF43 gene mutations may benefit from Wnt pathway inhibitors. Methods: We retrospectively studied 51 patients at our institution diagnosed with gastrointestinal cancers with RSPO3 gene fusions or RNF43 gene mutations identified using Caris Life Sciences next-generation sequencing (NGS). Microsatellite-instability (MSI) was evaluated using NGS. Tumor mutational burden (TMB) was estimated from 592 genes by counting all nonsynonymous missense mutations not previously defined as germline mutations, expressed as mutations per megabase. First-line median progression-free survival (PFS) and median overall survival (mOS) were estimated using the Kaplan-Meier method. Results: Of the 51 patients in our study, 44 (86.3%) had RNF43 mutations and 7 (13.7%) had RSPO3 fusions, 6 of whom had colorectal cancer (CRC) and had PTPRK as the fusion partner, 1 case of small bowel cancer had a RSPO3: IFNGR1 fusion. Median age at diagnosis was 64.0 ± 12.0 years. The most common primary tumors were CRC in 23 patients (45.1%) and pancreas in 12 patients (23.5%). Co-occurring mutations in CRC patients with RSPO3 fusions included KRAS (50%, Q61H, G12C, and G12V), BRAF V600E (50%), and TP53 (100%), none were MSI-H, median TMB (mTMB) was 9.5 (range 7-11). Sixty-seven percent of RNF43-mutated CRC patients had the G659fs mutation. Co-occurring findings in RNF43-mutated CRC patients included 67% MSI-H, median TMB 27 (range 6-123), KRAS (28%), BRAF V600E (28%), and TP53 mutations (33%). Twenty-two patients (43.1%) had stage IV disease, 18 (35.3%) had stage III disease and 11 (21.6%) had stage II disease. Thirty RNF43-mutated patients (58.8%) underwent definitive surgical resection, with a mean time to recurrence of 22.2 months and mOS of 43.0 months (95% CI, 15.6-70.4). Among RNF43-mutated stage IV CRC (mCRC) patients, median PFS was 8.0 months (95% CI, 5.9-10.1). Stratified by stage, mOS was 111.0 months for stage II disease, 72.0 months for stage III disease, and 37.0 months for stage IV disease (95% CI, 14.7-59.3). Conclusions: RNF43 correlates with overall favorable mOS in patients with mCRC. mOS in RNF43-mutated mCRC is similar to those of RAS and BRAF wild-type subgroups as reported in the TRIBE study (Cremolini et al. 2015). Therefore, RNF43 mutational status appears to be useful in predicting clinical outcomes for patients with stage IV CRC and overlaps significantly with MSI-H. RSPO3 fusions are rare events and the clinical outcomes of this subset of patients with tumors with RSPO3 fusions remains largely unknown.