688 Background: Pancreatic cancer is one of the leading causes of cancer-related death and is associated with poor overall survival. Glucagon-like peptide-1 receptor agonists (GLP-1RAs), widely prescribed for diabetes and obesity, have been investigated for potential impact in patients with pancreatic cancer, although their safety and clinical effects remain uncertain. This study evaluates the association between GLP-1RA use and clinical outcomes in a real-world cohort of patients with pancreatic cancer. Methods: Using the TriNetX Network database (72 healthcare organizations), we identified patients with pancreatic cancer who were prescribed GLP-1 receptor agonists (GLP-1RAs; n = 1,972) from 2010 to 2022, and matched controls without exposure to GLP-1RAs (n = 57,645). Propensity score matching balanced demographics and comorbidities, yielding 1,821 patients in each group. The primary outcomes were all-cause mortality and overall survival, assessed with Kaplan–Meier survival curves. Secondary outcomes included hospitalization, postoperative infection/sepsis, venous thromboembolism (VTE), acute kidney injury (AKI), and chronic kidney disease (CKD). Risk ratios (RRs) with 95% confidence intervals were estimated over a three-year follow-up period. Results: After propensity matching, the mean age was 72.0 years in the GLP-1RA group and 71.8 years in the non-GLP-1RA group. GLP-1RA users had lower all-cause mortality compared with non-users (40.6% vs 44.3%; risk ratio [RR] 0.92, 95% CI 0.85–0.99; p = 0.025) and longer median survival (1,065 vs 824 days; hazard ratio [HR] 0.90, 95% CI 0.81–0.997; p = 0.043). Hospitalization was less frequent in GLP-1RA users (40.2% vs 46.1%; RR 0.87; p = 0.040). Postoperative infection/sepsis (18.1% vs 21.3%; RR 0.85; p = 0.023) was also significantly lower in the GLP-1RA group. No significant differences were observed for VTE (15.1% vs 16.4%; p = 0.335), AKI (18.7% vs 20.2%; p = 0.317), or CKD (10.4% vs 8.4%; p = 0.073). Conclusions: GLP-1RA use was associated with lower mortality, longer survival, and reduced risks of hospitalization and postoperative infection/sepsis in patients with pancreatic cancer, without increased risk of VTE or renal complications. Prospective studies are warranted to confirm these findings and further clarify the role of GLP-1RAs in pancreatic cancer care.

824 Background: Immune checkpoint inhibitors (ICIs) have improved survival in patients with cancer but can also cause immune-related adverse events (irAEs). The clinical benefits of adherence to the irAE guidelines are unclear. This study aims to evaluate whether adhering to these guidelines improves irAE outcomes. Methods: We retrospectively reviewed patients with gastrointestinal cancer who developed irAEs after receiving ICIs between June 2017 and August 2024. The use of corticosteroids and second-line immunosuppressants was evaluated based on adherence to contemporary international guidelines (NCCN, ESMO, and ASCO). IrAE recovery was defined as the complete discontinuation of corticosteroids together with restoration of the affected organ function to baseline levels. A multivariable logistic regression analysis was performed to evaluate factors associated with achieving irAE recovery within 70 days. The cutoff of 70 days was chosen because the median time from the initiation of corticosteroid therapy to irAE recovery in this cohort was 70 days. Results: Of the 737 patients treated with ICIs, 127 (17.2%) developed irAEs, including 29 (22.8%) who experienced multiple irAEs. Excluding 53 endocrine irAEs, 100 of 116 (86.2%) were treated with corticosteroids. Among the 100 irAEs, there were two groups; 74 irAEs managed according to guidelines (Group A) and 26 managed in a non–guideline-adherent manner (Group B). There were significant differences in second-line immunosuppressant use between the two groups ( p = 0.0327). The most common irAEs were pneumonitis (n = 34, 20.1%), followed by colitis (n = 21, 12.4%), hepatitis (n = 15, 8.9%), and dermatitis (n = 15, 8.9%). Fifty-four irAEs (54.0%) were grade 3 or 4. There was no statistically significant difference in the irAE recovery rate between the two groups, with rates of 87.8% (65/74) and 76.9% (20/26), respectively. However, Group A required a significantly shorter duration of corticosteroid treatment than Group B (median: 63.0 days vs 146.5 days, p < 0.001). This finding was consistent across irAE grades and subgroups requiring second-line therapies. Among 23 patients who received second-line therapies, initiating treatment within two weeks after the start of corticosteroid therapy was associated with fewer relapses (0/14, 0% vs 5/9, 55.6%; p = 0.0407) and shorter corticosteroid treatment duration (median: 49.0 days vs 132.5 days, p < 0.001). Multivariable analysis revealed that guideline adherence was the only factor significantly associated with achieving irAE recovery within 70 days (odds ratio = 4.57, p = 0.0129). Conclusions: Adequate initial corticosteroid doses and the appropriate use of second-line immunosuppressants resulted in a significantly shorter corticosteroid treatment duration.

411 Background: Immunotherapy combined with chemotherapy has become the first-line treatment for advanced esophageal squamous cell carcinoma (ESCC) , but myelosuppression remains a common Adverse Event. However, existing preventive measures have limitations, and there is a lack of multi-lineage prevention and treatment strategies. Trilaciclib is a highly effective, selective, and transiently reversible cyclin-dependent kinase 4/6 inhibitor. Administered intravenously prior to chemotherapy, it protects hematopoietic stem and progenitor cells, thereby reducing chemotherapy-induced myelosuppression. We conducted this study to evaluate the efficacy and safety of trilaciclib in patients with advanced ESCC receiving immunotherapy combined with chemotherapy. Methods: Patients with unresectable metastatic/recurrent ESCC received trilaciclib (240 mg/m², IV, Q3W) combined with immunotherapy (Q3W), paclitaxel (175 mg/m², IV, Q3W), and carboplatin (AUC = 4-6, IV, Q3W). Chemotherapy was administered for ≤ 6 cycles. Trilaciclib was administered for the same number of cycles as chemotherapy. Each treatment cycle was 21 days, and treatment continued until disease progression, intolerable toxicity, completion of the protocol-specified treatment duration, or patient withdrawal. The primary endpoint was the incidence of grade ≥ 3 neutropenia during chemotherapy. Secondary endpoints included the incidence of grade ≥ 3 thrombocytopenia during chemotherapy, incidence of grade ≥ 3 anemia during chemotherapy, disease control rate (DCR), progression-free survival (PFS), and safety (adverse events [AEs]). Results: As of April 23, 2025, 10 patients were enrolled, with a median age of 65.5 years. For 90% of pts, an ECOG performance status (PS) of 0 was documented at baseline assessment (10% ECOG PS 1). During chemotherapy, 10 patients (100%) developed neutropenia, 1 patient (10%) developed anemia, and no patient developed thrombocytopenia. Among these, only 1 patient (10%) had grade ≥ 3 neutropenia, and no grade ≥ 3 anemia or thrombocytopenia was observed. Other AEs were mainly grade 1-2, including decreased appetite (50%), fatigue (40%), headache (40%), and alopecia (10%). Nine patients (90%) had stable disease (SD), resulting in a DCR of 90%. The median PFS did not reach. Conclusions: The combination of trilaciclib reduced the incidence of grade ≥ 3 chemotherapy-induced myelosuppression in patients with advanced ESCC receiving immunotherapy combined with chemotherapy. Notably, no thrombocytopenia of any grade was observed in this study to date, and no adverse impact on antitumor efficacy was noted. Preliminary results suggest that trilaciclib can improve chemotherapy tolerance in advanced ESCC patients, and further study data are highly anticipated.

741 Background: IP strategies directly target the peritoneal compartment in PDAC, where systemic chemotherapy alone demonstrates poor efficacy. Approaches include HIPEC, normothermic intraperitoneal paclitaxel (NIPEC/IP-PTX) given with systemic therapy, and PIPAC. This systematic review synthesizes evidence on feasibility, perioperative safety, and clinical outcomes of IP therapy in PDAC. Methods: We conducted a systematic search in PubMed, Embase, Scopus, and Cochrane databases for studies published through August 2025. Eligible studies included prospective or retrospective studies evaluating HIPEC, NIPEC/IP-PTX, or PIPAC with extractable safety or oncologic outcomes. Outcomes included feasibility, 30-day mortality, perioperative complications, locoregional recurrence (LRR) in the adjuvant setting, overall survival (OS) from peritoneal metastasis (PM) diagnosis or first IP therapy, conversion-to-resection, and histologic/radiologic response. Results: Sixteen studies met criteria, including 7 on HIPEC, 5 on NIPEC/IP-PTX, and 4 on PIPAC. In the adjuvant setting, 2 HIPEC studies demonstrated high feasibility (>85%), 0–5% 30-day mortality, and complication rates comparable to pancreatectomy. A randomized trial reported significantly lower LRR with adjuvant gemcitabine HIPEC. Median OS in treated arms was ~17–18 months, with 1-year OS ~63%. In the cytoreductive surgery (CRS) plus HIPEC setting for isolated peritoneal metastases, patients achieved CC-0/1 cytoreduction in most cases, with median OS ~24–41 months and 1-, 2-, and 3-year OS ranging 60–91%, 33–66%, and 39–59%, respectively. Thirty-day mortality was 0–4.3%, and major morbidity 20–43%. For NIPEC/IP-PTX, multicenter and single-institution cohorts consistently reported median OS ~14–16 months, 1-year OS ~60%, and conversion-to-resection rates of 17–24%. Ascites control and feasibility were reproducible, and device-related adverse events were generally manageable. PIPAC programs established oxaliplatin at 120 mg/m² as the recommended dose, with ~50–60% histologic regression by PRGS and disease stabilization in 50–60% after 1–2 cycles. Median OS in PDAC subsets was ~9–13 months, with low major toxicity, though early discontinuation was frequent in progressive disease. Conclusions: Intraperitoneal therapies for PDAC are feasible and show promising biologic and clinical signals. Adjuvant HIPEC reduces locoregional recurrence without excess mortality; CRS plus HIPEC can achieve multi-year survival in selected patients with isolated PM metastases; NIPEC/IP-PTX offers reproducible survival and conversion-to-resection; and PIPAC provides histologic regression with acceptable safety.

26 Background: Haystack MRD is a highly accurate, tumor-informed ctDNA assay designed for minimal residual disease (MRD) detection. To evaluate clinical performance and inter-laboratory reproducibility, retrospective testing of samples from early-stage colorectal cancer (CRC) patients enrolled in the randomized DYNAMIC and DYNAMIC-III trials was performed. Methods: Residual specimens from 59 patients (51 stage II, 8 stage III CRC) with annotated clinical outcomes were analyzed using the CLIA-validated Haystack MRD assay, which targets up to 50 tumor-informed, patient-specific variants derived from whole-exome sequencing of tumor and matched normal DNA. Testing was performed on 60 cell-free DNA (cfDNA) samples from 59 unique patients, containing a median of 3,800 genomic equivalents from plasma collected post-operatively (n=40) and post-adjuvant chemotherapy (n=20). Concordance of MRD results with clinical recurrence was assessed. Orthogonal validation was also conducted to evaluate reproducibility across two independent CLIA-certified laboratories: Quest Diagnostics Oncology Center of Excellence (Lewisville, TX) and Haystack Oncology (Baltimore, MD). Results: All 60 samples tested with Haystack MRD in parallel at the two sites yielded concordant MRD results (100% agreement). MRD positivity was observed in 10 of 12 samples from patients who experienced clinical recurrence, while 48 samples from patients who remained recurrence-free tested MRD-negative (sensitivity: 83%; specificity: 100%; positive predictive value [PPV]: 100%; negative predictive value [NPV]: 96%). Quantitative ctDNA measurements were between 0.32 (min) and 316.21 (max) mean ctDNA molecules/ml and were highly correlated across sites (Pearson correlation coefficient = 1.000; p<0.0001), supporting strong inter-laboratory reproducibility. Additional testing is ongoing. Conclusions: This study provides both clinical and orthogonal validation of the Haystack MRD assay using well-characterized specimens from the DYNAMIC trials. Results demonstrate high concordance with clinical recurrence outcomes and reproducibility across independent laboratories. These findings support the use of Haystack MRD as a robust approach for MRD detection following curative-intent treatment in stage II/III CRC, with potential to guide adjuvant therapy decisions and enable proactive, ctDNA-informed monitoring.

416 Background: Variable anatomy, prompted by bowel filling and respiratory motion, necessitates larger margins for planning target volumes (PTV) in radiotherapy for gastric lymphoma patients, leading to higher doses to organs at risk (OAR). The deep inspiratory breath-holding (DIBH) technique during radiotherapy has emerged as a protective method that improves OAR sparing. Therefore, this systematic review aims to systematically evaluate the benefits of DIBH compared to free breathing (FB) techniques during radiotherapy for gastric lymphomas, along with the physiological mechanisms involved. Methods: This systematic review included three comparative studies following a comprehensive search across PubMed, Google Scholar, and the Cochrane Library, involving a total of 33 patients with gastric lymphomas, in whom radiotherapy was planned under both DIBH and FB. The radiotherapeutic techniques assessed were intensity-modulated radiotherapy (IMRT), 3D-conformal radiotherapy (3D-CRT), step-and-shoot IMRT (SIMRT), volumetric-modulated arc therapy (VMAT), and tomotherapy. The dosimetric parameters included Gy (Gray), mean dose (Dmean), maximum dose (Dmax), etc. Results: Median heart dose (MHD) showed a consistent and significant reduction with DIBH across all techniques, including IMRT: 4.9→2.6 Gy; 3D: 8.54→4.46 Gy; SIMRT: 6.97→3.63 Gy; VMAT: 6.83→3.42 Gy; and tomotherapy: 7.06→3.80 Gy (all p<0.001). An additional study reported a decline from 7.1 Gy (FB 1.5 cm PTV) to 3.2 Gy (DIBH 1 cm PTV) (p<0.001). Kidney doses showed modest changes, with Dmean reduced from 4.3 Gy with FB 1 cm to 3.8 Gy with DIBH. For the left kidney, only one study demonstrated a significant reduction (p = 0.01). Spinal cord doses were also lower with DIBH; Dmax decreased from 19.79 to 17.56 Gy (VMAT), and Dmean from 8.7 Gy (FB 1.0 cm) and 9.7 Gy (FB 1.5 cm) to 7.9 Gy (DIBH) (p=0.02 and p=0.001). Findings in the lung were inconsistent: one study reported lower Dmean with DIBH, while others noted slight increases, e.g., left lung 3.0 vs. 2.6 Gy for FB 1 cm; 3.3 vs. 3.0 Gy for FB 1.5 cm. Liver outcomes were non-significant primarily, with reported reductions from 9.4 to 8.3 Gy and 13.91 to 12.64 Gy, although one study showed a higher Dmean with DIBH compared to FB 1 cm (p = 0.04). Conclusions: In conclusion, based on the current literature, DIBH significantly improves organ at risk (OAR) sparing, especially for the heart, right kidney, and spinal cord, during radiotherapy for gastric lymphomas. While results for the liver and lungs are mixed, no significant differences have yet been reported for the small bowel. These findings support the routine use of DIBH to minimize cardiac and critical organ radiation during radiotherapy, thereby facilitating safer clinical and patient practices.

91 Background: Previous evidence have evaluated the impact of GLP-1 receptor agonists (GLP-1 RA) on colon cancer risk in patients with obesity. We aim to provide updated population-based data, analyzing the potential effect of GLP-1RA on the risk of colon cancer in patients with or without obesity. Methods: This retrospective observational study leveraged the TriNetX database, forming two propensity-score–matched cohorts of type 2 diabetes (T2DM) patients without prior colon cancer. One cohort consisted of insulin-treated individuals not using GLP-1 agonists, and the other included new users of GLP-1 agonists. Both groups excluded patients with inflammatory bowel disease, personal or family history of colon cancer or polyps, or hereditary colon cancer syndromes. Matching accounted for age, sex, ethnicity, and race, resulting in 965,333 comparable patients. The main outcome was the incidence of new colon cancer diagnoses identified via ICD-10 codes. Analyses included odds ratios (ORs), risk ratios (RRs), Kaplan–Meier survival curves with hazard ratios (HRs), and statistical significance was set at p < 0.05. Results: Patients not treated with GLP-1 agonists had significantly higher colon cancer risk across all metrics: RR = 3.06 (95% CI: 2.921–3.207), OR = 3.08 (95% CI: 2.935–3.223), and HR = 2.702 (95% CI: 2.578–2.833), all with p < 0.0001. The absolute risk difference was 0.498% (95% CI: 0.478–0.517%). Additionally, the GLP-1 agonist cohort demonstrated superior overall survival regarding colon cancer. Conclusions: GLP-1 RA use was associated with significantly lower colon cancer risk and improved survival compared with insulin therapy. These findings align with previously published data showing the potential protective role of GLP-1 RA in reducing malignancy risk, supporting the need for further studies to evaluate their long-term beneficial profile.

807 Background: Geographic access to clinical trials is critical for equitable cancer care. Gastrointestinal (GI) cancers—including colorectal, pancreatic, liver, gastric, esophageal, and biliary—account for over a quarter of U.S. cancer deaths. Despite this burden, the geographic distribution of active GI cancer trials has not been systematically evaluated, particularly by cancer subtype and population factors such as rurality, socioeconomic status, and race/ethnicity. Methods: All active, interventional Phase I–III GI cancer trials were identified from ClinicalTrials.gov (accessed September 5, 2025), restricted to U.S.-based therapeutic studies. Trial sites were geocoded by ZIP code and mapped to 2020 Census ZIP Code Tabulation Areas (ZCTAs). Geographic access was defined as the proportion of the U.S. population residing within 30 or 60 miles of a site using Haversine distance. Analyses were stratified by cancer subtype, rurality (RUCA codes), income quartile, and race/ethnicity (ACS 2023). Results: We identified 958 active GI cancer trials (287 colorectal, 260 pancreatic, 136 gastric, 86 esophageal, 108 hepatocellular carcinoma [HCC], 81 cholangiocarcinoma) across &gt;4,400 U.S. ZIPs. At 30 miles, population coverage exceeded 80% for colorectal and pancreatic cancers but was lower for HCC (61%) and cholangiocarcinoma (64%). At 60 miles, coverage was &gt;90% for colorectal, pancreatic, and gastric cancers but remained &lt;80% for HCC and cholangiocarcinoma. Subgroup analysis showed 30-mile coverage for HCC was 71.0% among urban vs 9.6% among rural residents, and 42% vs 86% for lowest- vs highest-income quartiles. Hispanic and Black populations had 30-mile coverage of 67% and 70%, respectively. Conclusions: Geographic coverage of GI cancer trials in the U.S. varied by cancer type, with lower access for HCC and cholangiocarcinoma compared with colorectal and pancreatic cancers. Coverage was also reduced in rural, lower-income, and Hispanic and Black populations. These findings highlight geographic and demographic gaps in trial availability and can inform efforts to improve equitable trial access. Population coverage within 30 and 60 miles of GI cancer trial sites, by cancer type and subgroups. Cancer Type No. Trials (NCT IDs) Trial Phase Distribution (I/II/III) No. Sites (ZIPs) Pop Cov (%) (30/60m) Urban Pop Cov (%)(30/60m) Rural Pop Cov (%)(30/60m) Low Income Quartile Cov(%)(30/60m) High Income Quartile Cov(%)(30/60m) Hispanic Pop Cov(%)(30/60m) Black Pop Cov(%)(30/60m) Colorectal 287 123/149/15 1211 83/93 91/96 44/80 65/84 98/99 84/91 85/94 Pancreatic 260 109/132/19 1079 81/92 89/95 42/79 62/83 97/99 81/88 83/92 Gastric 136 59/68/9 966 79/90 87/93 41/75 60/78 96/99 81/87 81/90 Esophageal 86 36/46/4 546 71/84 80/90 24/61 49/69 93/97 74/82 75/85 Liver (HCC) 108 34/67/7 271 61/76 71/83 10/41 42/61 86/93 67/79 70/80 Cholangio 81 24/52/5 398 64/80 74/86 17/50 45/63</

TPS797 Background: Resectable pancreatic ductal adenocarcinoma (R-PDA) continues to be associated with high recurrence and poor long-term outcomes. While neoadjuvant chemotherapy is increasingly utilized, its superiority over upfront resection with adjuvant therapy remains uncertain. Tumor treating fields (TTFields) exert anti-mitotic effects by disrupting microtubule assembly and impairing cytokinesis through alternating electric fields. Preclinical models and clinical data suggest additive activity with chemotherapy. In the phase II PANOVA trial (n=40) in advanced PDA, TTFields combined with gemcitabine-based therapy demonstrated promising activity with acceptable safety, the main device-related adverse event being grade 3 dermatitis. Methods: This multicenter, single-arm phase II trial will assess perioperative TTFields in combination with gemcitabine plus nab-paclitaxel (GNP) in patients with R-PDA. Key eligibility criteria include: visible pancreatic mass, measurable disease, ≤180° venous involvement without arterial encasement, patent portal–splenic confluence, and no distant metastasis or extra-regional lymphadenopathy. Patients will initiate treatment with TTFields (compliance goal &gt;80%) and three cycles of GNP, followed by restaging. Those proceeding to resection will receive an additional three cycles of adjuvant GNP with TTFields. The Bayesian Optimal Phase II (BOP2) design specifies dual primary endpoints: 2-year overall survival (OS) and resection rate. The null hypotheses are OS of 40% (vs. 60% target) and resection rate of 60% (vs. 75% target). A total of 30 evaluable patients are required, with an interim analysis after 15 patients; early termination occurs if ≤8 resections are achieved at that point (power 83%, type I error 10%). Secondary objectives include safety/tolerability, overall response rate, TTFields compliance, relative dose intensity, and long-term survival outcomes. The study opened to accrual with a planned enrollment of 38 patients. At the time of submission, two patients had been enrolled, and one patient had resection (ClinicalTrials.gov identifier: NCT05624918). Clinical trial information: NCT05624918 .

32 Background: The clinical significance of CRC risk imparted by prevalent recessive variants MUTYH-G396D and Y179C (Caucasian carrier rate 1/50) and the low-penetrance APC-I1307K variant (Ashkenazi Jewish carrier rate ~1/15) is debated. Evidence supporting differential mutational spectrum, immune biology, and clinical outcomes in tumors that arise in carriers could lend support to their relevance. Here, we compare CRC tumor characteristics and survival of MUTYH-G396D/Y179C and APC-I1307K carriers to pts w/MUTYH or APC wild-type (WT) tumors. Methods: Retrospective review of CRC pt samples (n = 38,622) submitted to a CLIA-certified laboratory (Caris Life Sciences) for NGS of DNA (592-gene or whole exome) and RNA (whole transcriptome). Deficient mismatch repair/microsatellite instability (dMMR/MSI-H) was assessed by IHC/NGS. PD-L1 expression was tested by IHC (SP142; positive (+): ³2+, ³%5). Presumed monoallelic (mMut) and biallelic mutations (bMut) included those with or without, respectively, a second-hit mutation or variant allele frequency ³60%. Statistical significance determined by Fisher’s-Exact/Mann Whitney/X 2 tests. Real-world overall survival was obtained from insurance claims data, w/Kaplan-Meier estimates used for comparison. Results: MUTYH-G396D/Y179C (296 mMut, 88 bMut) and APC-I1307K (23 mMut, 133 bMut) alterations were identified. MUTYH bMut was more frequently TMB-High (³ 10 mutations/Mb: 29.1% vs mLOF 11.6%, p&lt;0.001; WT 10.3%, p&lt;0.001) and less frequently dMMR/MSI-H (1.3% vs mLOF 9.7%, p=0.018; WT 7.8%, p=0.036). CMS2 (34.1%) and CMS4 (31.7%) were the most common subtypes in MUTYH bMut, w/similar distributions in mMut (p=0.52) and WT (p=0.84). COSMIC signatures SBS18 (damage by reactive oxygen species) and SBS36 (defective base excision repair) were highly enriched in MUTYH bMut (21% and 17%, respectively, vs 6% and 1% in mMut, and 2% and &lt;1% in WT). APC bMut tumors were less frequently TMB-H (10.1% vs mLOF 13.0%, p=0.67; WT 18.0%, p=0.019), dMMR/MSI-H (6.4% vs mLOF 10.5%, p=0.50; WT 18.7%, p=0.002), and PD-L1+ (4.1% vs mLOF 13.6%, p=0.072; WT 9.6%, p=0.041). APC bMut were enriched with the canonical CMS2 subtype (36.3% vs 18.8% mMut, p=0.17; WT 9.8%, p&lt;0.001) and less frequently CMS1 subtype (9.9% vs 31.3%, p=0.19; WT 37.1%, p&lt;0.001). No SBS signatures were enriched in APC bMut. In pMMR/MSS CRC, MUTYH bMut was associated with similar prognosis as mMut and WT (median OS: 29.3 vs 32.0 and 28.2 months, p=0.57), while the prognosis of APC bMut was superior to mMut WT tumors (median OS: 40.5 vs 34.2 and 21.1 months, p&lt;0.001). Conclusions: The APC-I13017K variant may positively impact prognosis in pts with pMMR/MSS tumors. Tumor differences derived from distinct molecular signatures in MUTYH bMut and APC bMut warrant further investigation as CRC treatment targets.