Background: We conducted this prospective, single-arm, phase II trial to improve conversion efficacy by PD-1 antibody camrelizumab (Cam) combined with Nab-POF regimen in patients with initially unresectable locally advanced or limited metastatic gastric or gastroesophageal (G/GEJ) adenocarcinoma. Objective: Patients with initially unresectable locally advanced or limited metastatic gastric or gastroesophageal junction adenocarcinoma were enrolled. Design: Eligible patients received Cam 200mg, nab-paclitaxel 125 mg/m2, oxaliplatin 85 mg/m2 or with trastuzumab if Her2 positive, followed by fluorouracil 2400 mg/m2 as a 48-hour continuous infusion every 2 weeks. Tumor response and surgical feasibility were assessed every 3 cycles. If confirmed feasible for surgical resection after 6 cycles, patients would undergo surgery. The primary endpoint was R0 resection rate. Results: A total of 53 patients was enrolled and 52 were evaluable. The R0 resection rate was 75.0% (39/52), meeting the primary endpoint. ORR was 88.5% (46/52). The total CR rate was 23.1% (12/52). The median PFS and OS were not reached. The 3-year PFS and OS rates were 56.9% (95% CI = 45.7%-86.3%) and 62.8% (95% CI = 41.6%-78.0%), respectively. Patients tolerated treatment well. Grade 3/4 AEs were 42.3% (22/52) and neutrophil decrease was the most common at 36.5%. [68Ga]Ga-NOTA-GSI PET/CT scan demonstrated that patients with elevated SUV value of metastatic lesion had significantly longer PFS (P = 0.019) and a tendency of longer OS (P = 0.111). Conclusion: Cam combined with Nab-POF regimen chemotherapy provides a new conversion strategy for patients with initially unresectable locally advanced or limited metastatic G/GEJ cancer.

4063 Background: The MGC-FLORA study is a prospective biomarker-exploratory trial primarily designed to identify predictive biomarkers (including gut microbiota and blood-based biomarkers) for first-line immunochemotherapy in advanced gastric/gastroesophageal junction (G/GEJ) adenocarcinoma. A key exploratory question is whether adding anti-angiogenic therapy (fruquintinib) modifies the efficacy landscape and its associated biomarkers. This interim analysis compares clinical outcomes between chemotherapy plus immunotherapy (CHEM cohorts) and chemotherapy plus immunotherapy plus fruquintinib (FRUQ cohorts). The FRUQ cohort is concurrently registered as an independent phase II clinical study (NCT06158919). Methods: This analysis included patients enrolled before September 30, 2025, with follow-up until January 15, 2026. Patients with previously untreated, unresectable locally advanced or metastatic G/GEJ adenocarcinoma were enrolled into two cohorts in this non-randomized study: CHEM (PD-1 inhibitor plus XELOX/SOX) and FRUQ (PD-1 inhibitor plus XELOX/SOX plus fruquintinib 4mg QD, days 1-14). After 6 induction cycles, oxaliplatin was discontinued in both cohorts, and maintenance therapy consisted of S-1/capecitabine plus PD-1 inhibitor, with or without continued fruquintinib. Blood and stool samples were collected at baseline (C1D1), after two treatment cycles, and at progression for future biomarker analysis (gut microbiota, cfDNA/cfRNA); the present report focuses on clinical outcomes. Results: A total of 126 patients were included (CHEM: n = 83; FRUQ: n = 43), with a median follow-up of 10.97 months. In the overall population, median PFS was 8.38 months and median OS was 18.53 months. Among evaluable patients, ORR was 47.1% (16/34) in the CHEM cohort versus 76.7% (33/43) in the FRUQ cohort, indicating a higher response rate with the addition of fruquintinib. Median PFS was 8.38 months for CHEM and 10.35 months for FRUQ (HR = 0.63, 95% CI 0.37-1.08; p = 0.09). After propensity score matching for sex, age, liver metastasis status, number of target lesions, and Claudin18.2 expression (n = 29 per cohort), the PFS benefit associated with fruquintinib became more pronounced (HR = 0.45, 95% CI 0.20-1.02; p = 0.05). Conclusions: This initial analysis from the biomarker-oriented MGC-FLORA study suggests that the addition of fruquintinib to first-line immunochemotherapy may improve clinical efficacy in advanced G/GEJ adenocarcinoma. Consistent trends toward prolonged PFS were observed in both the overall and propensity score–matched populations, supporting a potential therapeutic benefit of fruquintinib. These findings warrant further validation in randomized trials and integrated biomarker analyses.

This single-center, open-label, exploratory trial aimed to investigate the efficacy and safety of envafolimab combined with lenvatinib and chemotherapy in previous treated advanced G/GEJ adenocarcinoma. Eligible patients with HER2-negative, microsatellite stable (MSS) advanced G/GEJ adenocarcinoma who had progressed after first-line treatment were enrolled in this phase II trial. Patients without previously receiving PD-1/PD-L1 inhibitors were enrolled in Group A, and those who progressed on the first-line PD-1 inhibitors were included in Group B. Patients in both groups received envafolimab (200 mg, subcutaneous injection (sc), days 1 and 15, Q4W) combined with lenvatinib and albumin-bound paclitaxel (100 mg/m2, IV, days 1, 8 and 15, Q4W, up to 6 cycles) until disease progression, unacceptable toxicity, or refusal of continuation. The primary endpoint was objective response rate (ORR). A total of 30 patients were included for safety and efficacy analysis. As of data cutoff (Sep 14, 2024), the median follow-up was 17.0 months (IQR: 8.0–18.8) in Group A. The ORR was 60.0

Background: Patients with refractory metastatic colorectal cancer (mCRC) face limited treatment options after failure of standard therapies. This single-arm, phase II study aimed to evaluate the efficacy and safety of rechallenge strategies using previously effective regimens in late-line mCRC. Methods: Patients who progressed after ≥2 lines of prior chemotherapy, with a prior progression-free survival (PFS) ≥4 months and a ≥4-month treatment-free interval on that regimen were enrolled. Patients received rechallenge chemotherapy (oxaliplatin-, irinotecan-, or raltitrexed-based) with or without targeted agents (bevacizumab or cetuximab). Primary endpoint was investigator-assessed PFS. Secondary endpoints included objective response rate (ORR), disease control rate (DCR), overall survival (OS), and safety. Results: Forty-three patients were enrolled (31 received chemotherapy plus targeted agents; 12 received chemotherapy alone). One patient discontinued treatment, leaving 42 patients evaluable for tumor response. The median PFS and OS were 3.97 months (95% CI: 2.46-5.48) and 13.03 months (95% CI: 9.68-16.38), respectively, while the ORR and DCR were 2.4% and 61.9%. Subgroup analysis showed that patients receiving chemotherapy plus targeted agents had higher DCRs (80.0% for bevacizumab-based regimens and 72.7% for cetuximab-based regimens vs. 18.2% for chemotherapy alone;) and longer median PFS (4.17 and 4.50 months vs. 1.57 months, respectively). Grade 3 or 4 adverse events were reported in 39.5% of patients, with no severe adverse events or treatment-related deaths observed. Conclusion: Chemotherapy rechallenge strategies, particularly when combined with targeted agents, demonstrated promising clinical activity and acceptable safety in selected heavily pretreated patients with mCRC who previously achieved sustained disease control.

157 Background: For the third-line treatment of metastatic colorectal cancer (mCRC), the standard regimens of regorafenib or trifluridine/tipiracil (TAS-102) monotherapy had demonstrated limited objective response rates (ORR) and progression-free survival (PFS). A phase Ib study reported the promising efficacy with the combination . However, no randomized trials have directly compared the combination against regorafenib monotherapy. This multicenter, randomized, open-label, phase II trial aims to evaluate the efficacy and safety of regorafenib plus TAS-102 versus regorafenib alone in patients with mCRC who have progressed after at least two prior lines of therapy. Methods: Patients progressing after ≥2 prior lines of treatment (including fluoropyrimidines, oxaliplatin, irinotecan) were randomized 1:1 to the combination group and regorafenib group, stratified by RAS status and primary tumor location. The combination group received regorafenib (80mg, qd, q2w) + TAS-102 (35mg/m² bid, d1-5, q2w). The regoragenib group received regorafenib monotherapy (120mg daily, d1-21, q4w). Treatment in each group was continued until disease progression or intolerable toxicity. The primary endpoint was PFS. Secondary endpoints included ORR, disease control rate (DCR) , overall survival (OS) and safety. Results: From July 1 th 2023 to January 31 th 2025, a total of 102 patients were enrolled and randomly assigned in this trial, of whom 51 were assigned to the combination group and 51 to the regorafenib group. 65 (63.7%) were male and 37(36.3%) were female. In the intention to treatment (ITT) analysis, after a median follow-up of 11.7 months, the median PFS (Kaplan–Meier estimate) was 6.10 months (95% CI 4.30-8.17 months) among the participants in the combination group and 2.77months (95% CI 2.10 - 3.47 months) among those in the regorafenib group (HR 0.34; 95% CI 0.21-0.54; p < 0.001). The ORR was 11.0% (6/51) vs. 0 (0/51), and DCR was 66.7% (34/51) vs. 29.4% (15/51) in the combination group and the regorafenib group, separately, in the ITT analysis. Safety profile indicated that 21.6% (11/51) and 19.6% (10/51) participants experienced adverse events with a maximum grade of 3/4 in the the combination group and the regorafenib, respectively. Conclusions: The combination regimen of regorafenib and TAS-102 demonstrated clinically meaningful PFS improvement with manageable toxicity in patients with refractory mCRC, supporting further investigation in phase 3 trials. Clinical trial information: NCT05970705 .

148 Background: After failure of first-line oxaliplatin (LOHP)/5-FU for metastatic colorectal cancer (mCRC), irinotecan (CPT-11) is the standard second-line chemotherapy. Cetuximab (C225) plus CPT-11 can improve efficacy in RAS/RAF wild-type patients, even after CPT-11 failure. This study was designed to evaluate whether a sequential strategy of second-line CPT-11 monotherapy followed by third-line C225 plus CPT-11 (s-IRI-CetuIRI) could provide more treatment lines and longer cumulative progression-free survival (PFS), compared to using combined C225 and CPT-11 (c-CetuIRI) directly in the second-line setting. Methods: RAS wild-type, cetuximab-naive mCRC patients after failure of first-line LOHP/5-FU were randomized to s-IRI-CetuIRI arm and c-CetuIRI arm, according to the stratification factors of primary tumor location and numbers of metastatic lesions. The primary endpoint was PFS (defined as the sum of second and third-line PFS in s-IRI-CetuIRI arm, and second-line PFS for c-CetuIRI arm). Results: Between November 2019 and November 2024, 42 eligible patients were enrolled in the study. The main primary tumor side was on the left side, with 68.4% (13/19) in the s-IRI-CetuIRI group and 78.3% (18/23) in the c-CetuIRI group. 21% (4/15) and 30.4% (7/23) of patients received bevacizumab in their previous treatment in both groups. At a median follow-up of 30.5 months (range 22.1-38.8 months), the median PFS (12.1 months vs 4.97 months, P=0.036) and OS (41.23 months vs 21.97 months, P=0.001) were significantly longer in the s-IRI-CetuIRI group than in the c-CetuIRI group. Subgroup analysis showed that patients with 1-2 metastases had better PFS and OS when treated with s-IRI-CetuIRI (P=0.043 and 0.001, respectively). Patients with left-sided primary tumors also had a longer PFS trend in the s-IRI-CetuIRI group (P=0.057) and better survival (P=0.043). Conclusions: The results of the study suggest that the sequential strategy of third-line C225 plus CPT-11 after second-line CPT-11 monotherapy has better survival than direct combination therapy in the second-line setting, especially in patients with low metastatic lesion burden and left-sided primary tumor site. Clinical trial information: NCT04833036 .

4057 Background: This prospective phase II study investigates a novel therapeutic strategy combining PD-1 inhibitor sintilimab (Sin) with short-course intensified chemotherapy regimen Nab-POF as first-line treatment for metastatic gastric cancer (MGC), based on the hypothesis that short-duration, high-intensity chemotherapy may potentiate antitumor immunity while reducing the cumulative immunosuppressive burden of prolonged chemotherapy. Methods: Eligible patients (pts) received Sin (200mg IV), nab-paclitaxel (125mg/m²), oxaliplatin (85mg/m²), and fluorouracil (2400mg/m² as a 48-hour infusion) (plus trastuzumab for HER2+ pts) Q2W. After 6 cycles, pts without disease progression received maintenance treatment (capecitabine + Sin). Upon disease progression, HER2- pts underwent reinduction with the Nab-POF regimen ± Sin, for up to 6 cycles, followed by maintenance therapy (S-1 +/- Sin). Tumor response was assessed every 3 cycles. PFS1 was defined as the time from informed consent to first progression. PFS2 was defined as the time from consent to second progression. If reinduction therapy was not performed, PFS2 equals to PFS1. The primary endpoint was PFS2 in HER2- pts. Results: In total, 123 pts were enrolled up to 09/2025. 118 pts were evaluable. This report presents the results for the HER2- population. Among 90 HER2- pts, the median age was 60 (23-83), with 72.2% male. The common metastatic sites included the liver (36.7%), peritoneum (26.7%), bone (13.3%), and lungs (6.7%). ORR was 82.2% (74/90) and DCR was 100%. Nine pts received complete response and CR rate was 10.0%. 2-year DFS and OS rate were both 100% among pts with CR. 60 pts progressed and 47 patients died. 25 pts experienced reinduction with the Nab-POF regimen ± Sin. Median PFS1 was 13.6mo (95%CI 9.8-16.1) and median PFS2 was 16.2mo (95%CI 11.0-23.2). Median OS was 21.3mo (95%CI 11.3-NR). 1-year OS rate was 72.2% (95%CI 63.5-82.1) and 2-year OS rate was 44.2% (95%CI 33.9-57.6). Subgroup analysis showed whether signet ring cells are present or not, whether liver metastasis occurs or not, and the level of CPS were not related to the therapeutic effect. All pts experienced treatment-related adverse events (AEs). Grade 3/4 AEs were 55.6% (50/90) and neutrophil decrease was the most common at 36.7% (33/90). Immune-related AEs occurred in 22(24.4%) pts, including 6 (6.7%) grade 3/4 events, which were effectively controlled. Conclusions: Sin plus short-course intensified chemotherapy regimen Nab-POF demonstrates high response and CR rate, and remarkable long-term effects. This approach may represent a feasible first-line treatment strategy, warranting further evaluation in randomized trials. Clinical trial information: NCT05982301 .

4045 Background: Tecotabart vedotin (TV; LM-302) is a novel and potent MMAE-based ADC targeting CLDN18.2. Previous data (ASCO 2025, abstract#4039) demonstrated promising efficacy and manageable safety profile of TV plus PD-1 inhibitor in patients (pts) with systemic therapy-naive advanced gastric or gastroesophageal junction (G/GEJ) cancer. Updated phase II results of TV plus toripalimab (T) with or without chemotherapy in this first-line setting are presented. Methods: Eligible pts had histologically confirmed, no prior systemic therapy, HER2-negative G/GEJ adenocarcinoma with CLDN18.2 positivity (IHC 2+/3+ in ≥1% of tumor cells). Pts received TV (1.8 mg/kg Q2W) plus T (3 mg/kg Q2W), with or without capecitabine or S-1. The primary endpoint was PFS per RECIST v1.1. Secondary endpoints were efficacy, safety, PK, and biomarker analysis. Data cutoff: November 25, 2025. Results: A total of 71 pts with G/GEJ adenocarcinoma were enrolled: 39 in the TV+T cohort and 32 in the TV+T+Chemo cohort. Baseline CLDN18.2 high expression (≥25%) was presented in 82.1% of pts in the TV+T cohort and 84.4% in the TV+T+Chemo cohort. PD-L1 CPS ≥1 was observed in 61.5% and 81.3% of pts, respectively. Median follow-up was 16.16 months in the TV+T cohort and 14.82 months in the TV+T+Chem cohort. The median PFS (95% CI) was 10.68 months (6.28, NE) and 12.55 months (6.80, NE), respectively. Among pts with CLDN18.2 ≥25%, the median PFS (95% CI) was 12.22 months (6.87, NE) and NE (8.44, NE), the median DOR (95% CI) was 13.86 months (9.36, NE) and NE (7.10, NE), ORR was 64.5% and 69.2%, and DCR was 96.8% and 96.2%, respectively. Median OS was not reached in either cohort. In pts with CLDN18.2 <25%, TV+T (N=7) and TV+T+Chemo (N=5) showed a median PFS (95% CI) of 5.72 (1.15, NE) and 6.80 (5.55, NE) months, ORR of 14.3% and 40.0%, DCR of 71.4% and 100%, and OS (95% CI) of 10.78 (2.56, NE) months and NE (8.25, NE), respectively. TRAEs occurred in all pts. Grade ≥3 TRAEs occurred in 66.7% of pts in the TV+T cohort and 81.3% of pts in the TV+T+Chemo cohort. The most common Grade ≥3 TRAEs (incidence ≥20%) were decreased neutrophil count in the TV+T cohort, and decreased neutrophil count and decreased white blood cell count in the TV+T+Chemo cohort. TEAEs leading to dose reduction occurred in 17.9% (Grade ≥3: 12.8%) and 43.8% (Grade ≥3: 18.8%) of pts, respectively. TEAEs led to treatment discontinuation occurred in 15.4% (Grade ≥3: 7.7%) and 31.3% (Grade ≥3: 28.1%), respectively. No treatment-related deaths were reported. Conclusions: TV+T, with or without chemotherapy, demonstrated encouraging anti-tumor activity and manageable safety as first-line treatment for CLDN18.2-positive advanced G/GEJ cancer. The doublet shows comparable efficacy to the triplet with enhanced tolerability, supporting further large-scale clinical investigation. Clinical trial information: NCT05934331 .

Background East Asian Helicobacter pylori (H. pylori) strains are commonly classified as a single hspEAsia lineage characterized by elevated virulence. However, gastric cancer incidence varies markedly across China, suggesting that clinically relevant bacterial heterogeneity may exist within this framework. A systematic assessment of fine-scale population structure and its functional correlates in Chinese H. pylori remained limited.Methods We analyzed whole-genome sequencing data from 1,243 H. pylori isolates collected from 20 provinces and regions across China, including 50 newly sequenced clinical strains from Shanghai. Fine-scale population structure was resolved using coancestry-based clustering and chromosome painting. Subpopulations were further characterized by pangenome composition, virulence factor repertoires, genome-wide fixation index (Fst), and predicted antibiotic resistance-associated mutations. E-test minimum inhibitory concentration (MIC) assays were performed to compare phenotypic susceptibility with mutation-based resistance prediction.Results Six geographically structured subpopulations were identified within Chinese hspEAsia. SubtypeCentral represented a widely distributed mainland lineage, whereas subpopulations from Inner Mongolia and Taiwan showed the greatest genetic divergence. Chromosome painting revealed strong within-lineage ancestry cohesion in SubtypeTaiwan, contrasted by extensive admixture in Inner Mongolia and Yunnan. Recurrent high-Fst loci across subpopulations, including glnA, frpB4, and HP1501, highlighted genomic regions contributing disproportionately to population differentiation. Marked heterogeneity in virulence profiles was observed. SubtypeInnerMongolia showed a higher prevalence of cagA-negative or Western-type cagA variants and a reduced overall repertoire of virulence genes. Predicted antibiotic resistance patterns were also strongly subtype dependent. Notably, SubtypeTaiwan exhibited an exceptionally high rifampicin resistance rate driven almost exclusively by a single rpoB A2414V mutation. E-test validation in the newly collected isolates provided supportive phenotypic evidence for the mutation-based resistance strategy.Conclusions Chinese H. pylori hspEAsia strains comprise multiple regionally structured subpopulations with distinct evolutionary histories, gene content, virulence profiles, and predicted resistance determinants. This fine-scale genomic classification provides a biological basis for understanding regional disparities in gastric cancer risk and genotypic resistance, and supports the need for subtype-aware surveillance and region-specific clinical management strategies in China.

4039 Background: Claudin 18.2 (CLDN18.2), a tight junction protein highly expressed in gastric and gastroesophageal junction (GEJ) cancers, has emerged as a promising therapeutic target. LM-302 (tecotabart vedotin), a novel and potent MMAE-based ADC targeting CLDN18.2, has shown promising efficacy and safety as monotherapy in heavily pretreated advanced gastric cancer. This pooled analysis evaluates the efficacy and safety of LM-302 in combination with the anti-PD-1 antibody toripalimab as a first-line treatment option for patients with gastric, GEJ, or esophageal adenocarcinoma (EAC). Methods: Eligible patients with histologically confirmed, previously untreated, unresectable, HER2-negative gastric, GEJ, or EAC were included. Patients received LM-302 (1.6 mg/kg Q3W, 2.0 mg/kg Q3W or 1.8 mg/kg Q2W) and toripalimab (240 mg Q3W or 3 mg/kg Q2W). Endpoints included safety, objective response rate (ORR), disease control rate (DCR), progression-free survival (PFS), OS, and biomarker analysis. Data cutoff: January 7, 2025. Results: A total of 43 gastric, GEJ, or EAC patients (median age: 62.3 years; 65.1% male) from Australia and China were treated. No dose-limiting toxicities (DLT) were observed across all these dose levels. Treatment-related adverse events (TRAEs) related to LM-302 were reported in 39 patients (90.7%), with common events (≥20%) including anemia, decreased white blood cell count, decreased neutrophil count, increased aspartate transaminase (AST), vomiting, loss of appetite, and nausea. Grade ≥3 TRAEs related to LM-302 occurred in 16 patients (37.2%), the most common events (≥5%) were decreased neutrophil count (14.0%), increased alanine transaminase (11.6%), increased AST (9.3%), and anemia (7.0%). Among 41 efficacy-evaluable patients (median follow-up: 6.01 months), ORR was 65.9% (95% CI: 49.4-79.9%), and DCR was 85.4% (95% CI: 70.8- 94.4%). In 32 GC patients with CLDN18.2 expression in ≥25% of tumor cells (IHC 2+/3+), ORR was 71.9% (95% CI: 53.3-86.3%) and DCR was 96.9% (95% CI: 83.8-99.9%). Among these patients, ORR was 63.3% (95% CI: 35.1-87.2%) for patients with PD-L1 CPS < 1 and 77.8% (95% CI: 52.4-93.6%) for patients with PD-L1 CPS ≥1. Median PFS and OS were not reached; one patient with PD-L1 CPS < 1 achieved PR and remained on treatment for 14.70 months. Conclusions: LM-302 combined with toripalimab demonstrated encouraging anti-tumor activity and manageable safety as first-line treatment for patients with CLDN18.2-positive gastric, GEJ, and EAC, including those with low-to-moderate CLDN18.2 expression. These findings support further large-scale clinical trials to confirm efficacy, safety and clinical utility. Clinical trial information: NCT05188664 ; NCT05934331 .

Programmed cell death protein-1 (PD-1) inhibitors plus chemotherapy have been the standard of care in the first-line treatment of advanced gastric or gastroesophageal junction (G/GEJ) adenocarcinoma; however, the survival benefits are modest in patients with low programmed death ligand 1 (PD-L1) expression. Here we investigated the efficacy and safety of cadonilimab (PD-1/cytotoxic T lymphocyte antigen-4 (CTLA-4) bispecific antibody) plus chemotherapy as first-line treatment in G/GEJ adenocarcinoma. The prespecified interim analysis is reported here. This was a randomized, double-blind, placebo-controlled phase 3 study. Eligible patients were adults with untreated, unresectable, locally advanced or metastatic G/GEJ adenocarcinoma. Patients were randomized 1:1 to receive cadonilimab (10 mg kg−1 every 3 weeks) or placebo plus chemotherapy (every 3 weeks). The primary endpoint was overall survival (OS) in the intention-to-treat population (one-sided significance level, P = 0.025). Secondary endpoints included OS in patients with a PD-L1 combined positive score ≥5, progression-free survival, objective response rate, duration of response and safety. As of 18 August 2023, 610 patients from 75 study centers were randomized to cadonilimab (n = 305) or placebo (n = 305). With a median follow-up of 18.7 months, the cadonilimab group had a significantly longer median OS (14.1 versus 11.1 months; hazard ratio (HR) 0.66; 95 NCT05008783 . In a prespecified interim analysis of the randomized, double-blind phase 3 COMPASSION-15 trial, patients with advanced HER2-negative gastric/GEJ cancer treated with the anti-PD-L1/CTLA-4 bispecific Ab cadonilimab plus chemotherapy showed significantly improved overall survival compared with patients treated with placebo plus chemotherapy as first-line treatment.

OBJECTIVE:The aim of the study was to evaluate the feasibility of functional MR in predicting the clinical response to chemotherapy in patients with colorectal liver metastases (CLM). METHODS:A total of 196 eligible patients were enrolled in the study between August 2016 and January 2023. Functional MR was performed at baseline and after one cycle of chemotherapy. The diffusion kurtosis radiomic texture features were extracted and a signature model was built using the R package. The initial 100 cases were designated as the training set, the following 48 cases were designated as the validation set, and the final 48 cases were designated as the intervention validation set. RESULTS:Good performance for the response prediction (AUC = 0.818 in the training set and 0.755 in the validation set) was demonstrated. The objective response rates (ORRs) in the high-risk subgroup were significantly lower than the low-risk subgroup in the training and validation sets. Worse progression-free survival and overall survival rates were noted in the high-risk population. In the intervention set 22.9% (11/48) of the chemotherapy regimens for patients were changed in response to the model-predicted results and the ORR reached 77.1% (37/48), which was significantly higher than the training and validation sets [47.97% (71/148); P = 0.000]. CONCLUSIONS:A functional MR signature effectively predicted the chemotherapy response and long-term survival. The adjustment of the regimen guided by the model significantly improved the ORR.

ABSTRACT Background Cytotoxic agents remain the mainstay treatment for advanced gastrointestinal cancer. However, the number of cytotoxic agents is limited, and the treatment effect is not satisfactory. Therefore, new agent and combination strategies are to be explored. Methods The antitumor efficacy of low‐dose epigenetic modifiers (LD‐EMs) of 5‐azacytidine and entinostat, cytotoxic agents of paclitaxel, cisplatin, oxaliplatin, 5‐fluorouracil, and a novel cytotoxic agent TIC10, and the combination of LD‐EMs and cytotoxic agents was investigated in vivo. Flow cytometry and immunohistochemistry were conducted to analyze the immune phenotype in the tumor microenvironment. The proliferation and apoptosis analyses were performed in vitro. Results LD‐EM therapy demonstrated superior tumor inhibition compared with commonly used chemotherapy in gastrointestinal cancer. A novel cytotoxic agent TIC10 resulted in weak to moderate tumor growth inhibition (TGI). LD‐EMs exhibited a more pronounced antitumor effect than TIC10 alone (CT26: TGI of 74.5% vs. 46.2%, respectively; p < 0.05; HNM007: TGI of 52.0% vs. 21.4%, respectively; p < 0.05; AKR: TGI of 53.8% vs. 10.1%, respectively; p < 0.05). The combination of TIC10 and LD‐EMs led to a more pronounced tumor reduction with tolerable toxicity. Analysis of the immune profiles showed increased percentages of CD45+ lymphocytes, CD3+ and CD8+ T cells, M1 macrophages, and dendritic cells, whereas decreased percentages of M2 macrophages and myeloid‐derived suppressor cells under treatment with LD‐EMs and combination therapy. Mechanistic studies revealed that LD‐EMs activated the RIG‐I–MAVS pathway, stimulated type I interferon responses, and subsequently promoted chemokine secretion. In contrast, TIC10 suppressed cell viability and induced cell apoptosis. Conclusions LD‐EMs remodeled the tumor microenvironment to an immune‐promoting environment. Although TIC10 could suppress cell viability and induce cell apoptosis. A combination of LD‐EMs and TIC10 indicated a rational strategy through complementary mechanisms.

334 Background: For initially unresectable locally advanced or metastatic gastric or gastroesophageal junction adenocarcinoma, the prognosis is poor and chemotherapy is the main treatment option. The AIO-FLOT3 Trial suggested that conversion therapy might improve outcome in pts with limited metastatic gastric cancer (GC). We conducted this prospective, single-arm, phase II trial to improve conversion efficacy of PD-1 inhibitor Cam combined with Nab-POF regimen (nab-paclitaxel , oxaliplatin, fluorouracil) in pts with initially unresectable locally advanced or limited metastatic gastric or gastroesophageal junction adenocarcinoma. Methods: Eligible pts received Cam 200mg, nab-PTX 125 mg/m 2 , oxaliplatin 85 mg/m 2 or with trastuzumab if Her2 positive, each was an IV followed by fluorouracil 2400 mg/m 2 as a 48-h continuous IV on day 1, every 2 weeks. Patients were assessed for tumor response and surgical feasibility every 3 cycles, and if confirmed feasible for surgical resection by both surgical and medical experts after 6 cycles, they would undergo surgery within 3-6 weeks. The primary endpoint was R0 resection rate. Results: In total, 53 pts were enrolled up to 01/2024. 52 pts were evaluable. 50% are younger than 65 years of age. Male was 73.1%. The metastatic sites included liver 46.2% (24), retroperitoneal lymph nodes 40.4% (21), Krukenberg 5.8% (3), and locally advanced unresectable tumors 21.2% (11). 4 pts (7.7%) were dMMR/MSI-H and 10 (19.2%) were Her2 positive. ORR was 88.5% (46/52) and DCR was 98.1% (51/52). 45(86.5%) pts were assessed as resectable. 3 pts (6.8%) refused surgery and 3 (6.8%) were cCR under observation. The R0 resection rate was 75.0% (39/52). CR rate was 23.1% (cCR, 5.8%; pCR,17.3%). The mPFS was not reached (95% CI = 17.0-NE), and the mOS was not reached (95% CI = 25.4-NE), either. The 3-year PFS and OS rate were 56.9% (95% CI = 45.7%-86.3%) and 62.8% (95% CI = 41.6%-78.0%), respectively. The 3 cCR pts have survived utill the last follow-up date of 07/2025 without tumors for 17, 29, and 34 months, respectively. Pts tolerated treatment well, although all pts experienced treatment-related adverse events (AE). Grade 3/4 AEs were 42.3% (22/52) and neutrophil decrease was the most common at 36.5%. 10 patients (19.2%) experienced immune related AEs (irAEs), and 2 of them experienced grade 3 irAEs. Conclusions: The PD-1 inhibitor Cam combined with the Nab-POF regimen safely induced a high conversion rate with very high R0 resection and high 3-year PFS and OS rates, preliminarily demonstrating promising effects, providing a new conversion drug treatment option for pts with initially unresectable locally advanced or limited metastatic advanced GC. These findings warrant further prospective, randomized controlled investigations. Clinical trial information: NCT04510064 .

4046 Background: For unresectable locally advanced or metastatic HER2-negative gastric cancer (GC), the combination of chemotherapy with immunotherapy, specifically PD-1 inhibitors, has emerged as a new standard of care in first-line treatment of advanced GC, but the patients’ outcomes remain poor. Fruquintinib (Fru) is a highly selective inhibitor of VEGFR1/2/3. We sought to explore the efficacy and safety of fruquintinib combined with PD-1 inhibitors and chemotherapy in the first-line treatment of HER2-negative advanced gastric or gastroesophageal junction adenocarcinoma. Here, we report the updated results. Methods: Eligible pts with HER2 negative locally advanced unresectable or metastatic gastric or gastrooesophageal junction adenocarcinoma, without any systemic anticancer treatment for advanced disease were included in the study. pts received 6 cycles of combined first line treatment with Fru (4mg p.o. qd, d1-14, q3w, ) combined with PD-1 inhibitor (investigator's choice of sintilimab 200mg or nivolumab 360mg intravenously q3w) and chemotherapy (investigator's choice of XELOX or SOX) regimen. The following maintenance treatment was Fru combined with PD-1 inhibitor and S-1/capecitabine until disease progression or intolerable toxicity. The primary endpoint was progression-free survival (PFS). Secondary endpoints included objective response rate (ORR), overall survival (OS), disease control rate (DCR) and safety. Results: As of Jan 24, 2025, a total of 33 pts (22 male, 11 female) median age 62 (range: 35-77) were enrolled and received at least one dose of treatment. 30 patients had at least one tumor assessment post treatment, with an ORR of 80.0% (24/30; 95%CI 61.4-92.3) and DCR of 100%(30/30; 95%CI 88.4-100.0). In the intention to treatment (ITT) population, the median PFS was 9.43 months (95% CI 5.29-13.24) and the 9-month PFS rates was 57% (95%CI 40.0-83.0). The most common AEs of all grades were neutrophil count decreased (33.3%), palmar-plantar erythrodysesthesia syndrome (33.3%), and fatigue (30.3%). Conclusions: The combination of fruquintinib and chemotherapy with PD-1 blockade in the first-line treatment for unresectable locally advanced or metastatic HER2-negative GC had shown promising efficacy and acceptable safety profile. The results warrant further investigations in a large cohort. Clinical trial information: NCT06158919 .

442 Background: Paclitaxel injection (paclitaxel IV) is category 1 preferred regimen in the second-line therapy of gastric cancer, while it has shortcomings including vehicle-led safety risk, long time injection, premedication or frequent hospital visit. Paclitaxel oral solution (Liporaxel), the world's first successfully developed oral formulation, is an alternative. This study aimed to establish non-inferiority in efficacy and comparable safety profile of paclitaxel oral solution versus paclitaxel IV, as a second-line monotherapy in gastric cancer. Methods: This is a randomized, open-label, non-inferiority phase 3 trial conducted at 53 centers in China. Patients with unresectable or recurrent or metastatic gastric cancer progressed after fluoropyrimidine- or fluoropyrimidine plus platinum-based first-line therapy were randomly assigned at 1:1 ratio (stratified by gastrectomy, ECOG PS and prior chemotherapy) to receive paclitaxel oral solution (200mg/m 2 twice daily on days 1, 8, 15 of a 28-day cycle) or paclitaxel IV (175mg/m 2 on day 1 of a 21-day cycle). Co-primary endpoints were progression-free survival (PFS) assessed by blind independent review committee (BIRC) and overall survival (OS), with non-inferiority margin of hazard ratio (HR) of 1.18 for PFS and 1.16 for OS. Results: From April 22, 2019, to January 31, 2022 (data cut-off), 536 patients were randomized to receive either paclitaxel oral solution (n=268) or paclitaxel IV (n=268), with a median follow-up of 13.4 vs. 12.6 months. PFS met non-inferiority criteria, with BIRC-assessed median PFS of 3.02 months (95% confidence interval [CI]: 2.69, 3.71) in the oral group vs. 2.89 months (95% CI: 2.53, 3.48) in the IV group (HR=0.894, 95% CI: 0.719, 1.112, p =0.311). OS (cut-off on February 15, 2023, for OS primary analysis) demonstrated superiority for paclitaxel oral solution, with median OS of 9.13 months (95% CI: 7.72, 10.97) vs. 6.54 months (95% CI: 5.75, 7.26), showing 2.59-month improvement (HR=0.770, 95.5% CI: 0.635, 0.934, p =0.006). For the treatment-related adverse events (TRAEs), the oral group showed lower incidences of neuropathy (22.3% vs. 38.7% all grade), alopecia, fatigue, musculoskeletal and connective tissue disorders, and no hypersensitivity occurred without premedication. The most common ≥Grade 3 TRAEs were neutrophil count decreased (47.9% in oral vs. 54.5% in IV), white blood cell count decreased (41.5% vs. 35.3%) and anemia (16.6% vs. 10.9%). Grade 5 TRAEs were rare with comparable rates across two groups (four [1.5%] vs. three [1.1%]). Conclusions: Paclitaxel oral solution demonstrated non-inferiority in PFS and superiority in OS compared to paclitaxel IV, with clinically manageable and favorable safety profile, supporting paclitaxel oral solution as a second-line treatment option for gastric cancer. Clinical trial information: CTR20190050.

Background:Although patients with gastric cancer liver metastases (GCLM) may achieve survival benefits after radical surgery, there is controversy regarding the surgical indications and choice of surgical methods. This study aims to investigate the impact of hepatic resection (HR) on the prognosis of patients with GCLM who have undergone radical resection of the primary tumor. Methods:This study conducted a retrospective analysis of 120 patients with GCLM who have undergone resection of the primary tumor. The patients were divided into an HR group and a non-hepatic resection (NHR) group. Propensity score matching (PSM) was analyzed and patients' prognoses were followed up and compared. Results:The PSM analysis included a total of 88 patients. The HR group had a median overall survival (OS) time of 35.0 months [95% confidence interval (CI): 30.7-39.3], with 1-, 2-, and 3-year survival rates of 88.0%, 81.5%, and 46.8% respectively. The NHR group had a median OS time of 16.0 months (95% CI: 10.5-21.5), with 1-, 2-, and 3-year survival rates of 56.8%, 30.8%, and 22.4% respectively. The median OS time was statistically different between the two groups. Extrahepatic metastasis (hazard ratio =2.777; 95% CI: 1.598-5.223; P=0.002) and HR (hazard ratio =0.154; 95% CI: 0.040-0.594; P=0.007) were significant factors for OS. In the HR group, laparoscopic surgery (P=0.004) and extrahepatic recurrence (P=0.008) were significant factors for intrahepatic recurrence-free survival (IHRFS). Conclusions:HR can significantly improve the prognosis of GCLM with resected primary tumors. Laparoscopic surgery is preferred as the surgical approach. Patients with extrahepatic recurrence have a shorter IHRFS.

e15525 Background: Irinotecan-based regimens are commonly used as second-line therapy for metastatic colorectal cancer (mCRC). Irinotecan liposomes can improve the distribution and metabolism through enhanced permeability and retention effect. This study explored the efficacy and safety of irinotecan liposome (HR070803) in combination with 5-fluoruracil/leucovorin (5-FU/LV) and bevacizumab as a second-line treatment for patients with mCRC. Methods: In this phase 2 study (ChiCTR2300075966), patients with histologically confirmed mCRC who had failed prior first-line chemotherapy (fluoropyrimidine plus oxaliplatin, with either bevacizumab or an anti-EGFR antibody for RAS wild-type tumors) were eligible for enrollment. Exclusion criteria included patients with pathologically confirmed pMMR/MSI-H, BRAF V600E mutation, or prior irinotecan-based therapy. Bevacizumab (5.0 mg/kg), HR070803 (60 mg/m²), and 5-FU/LV (2400/400 mg/m²) were administered every two weeks until disease progression. The primary endpoint was the objective response rate (ORR) according to RECIST version 1.1. Secondary endpoints included disease control rate (DCR), duration of response (DoR), progression-free survival (PFS), overall survival (OS), quality of life, and tolerability. Results: From January 15, 2024, to December 27, 2024, 34 patients were enrolled. The median age was 58 years (range, 26–73), with 76.5% of patients being male and 73.5% having left-sided colon or rectal cancer. KRAS mutations were present in 41.2% of patients, and 2.9% had ERBB2 amplification. Hepatic metastases were observed in 76.5% of patients, and 58.8% had metastatic involvement of more than one organ. Of the 27 patients with evaluable treatment efficacy, the confirmed ORR was 29.6% (8/27 patients; 95% CI, 12.2-48.0), and the DCR was 81.5% (22/27 patients; 95% CI, 65.8-97.1). Median DoR, PFS, and OS were not yet reached. Treatment-related adverse events (TRAEs) of grade 3 or 4 occurred in 47.1% of patients (16/34); the most common TRAEs were hematologic and gastrointestinal. TRAEs leading to treatment discontinuation occurred in 2 patient (5.9%). Conclusions: Irinotecan liposome (HR070803) in combination with 5-FU/LV and bevacizumab demonstrated promising efficacy and durability in this patient population. These results support further investigation of irinotecan liposome (HR070803) as a treatment option for patients with mCRC. Clinical trial information: ChiCTR2300075966 .

375 Background: The combination of first-line chemotherapy and PD-1 has emerged as a standard treatment, enhancing the efficacy in advanced gastric/gastroesophageal junction (GEJ) cancer. The efficacy of the combination of second-line chemotherapy and PD-1/PD-L1 remains unclear. This study aims to observe the efficacy of Envafolimab (PD-L1 inhibitor ) plus tyrosine kinase inhibitors (TKIs) and second-line chemotherapy in the treatment of patients (pts) with advanced GEJ adenocarcinoma. Methods: Eligible pts with HER2-negative, microsatellite stable (MSS), advanced GEJ adenocarcinoma were enrolled in this study. Pts who never received PD-1 or PD-L1 inhibitors before would assign into Group A. Pts in Group B have failed first-line PD-1 antibody combined chemotherapy, with optimal efficacy as or above stable disease (SD). Pts in both groups received Envafolimab (200mg, hypodermic injection, days 1, 15, Q4W) combined with Lenvatinib (8mg/12mg, po, once a day) and albumin paclitaxel (100mg/m2, IV, days 1, 8 and 15, Q4W, up to 6 cycles) until disease progression, unacceptable toxicity, or pts refusal to continue treatment. The primary endpoint was objective response rate(ORR) per iRECISTv1.1, and the secondary endpoints included disease control rate (DCR), progression free survival (PFS), overall survival (OS) and safety. Results: From Oct 2022 to Apr 2024, 32 pts were enrolled. Both groups enrolled 16 pts equally, and 15 pts of each group were evaluable for efficacy analysis. The average ages were 60.07 ± 11.73 y and 50.13 ± 12.18 y in group A and B respectively. In group A, 7 pts had been tested for PD-L1 status, of which 2 (13.3%) was positive and 5 (33.3%) were negative. PD-L1 status was negative in 5(33.3%) pts, positive in 9 (60.0%), and untested in 1(6.7%) in group B. As of Sep 14, 2024, The ORR was 60.0% (9/15) in group A and 46.7% (7/15) in group B. The DCR was 100.0% (15/15) in both groups. After a median follow-up of 17.0 mo (IQR: 8.0-18.8) in group A, the mPFS was 8.2 mo (95% CI = 6.1-10.4), and the mOS was 14.8 mo(95% CI = 7.4-22.2). With median follow-up of 9.0 mo (IQR: 6.1-16.2) in group B, the mPFS was 5.9 mo (95% CI= 3.8-8.1) and the mOS was 11.5 mo (95% CI = 3.1-19.9). Overall incidences of adverse events (AEs) of any grade was 100% (30/30), and 80.0% (24/30) pts had ≥ grade 3 AEs during treatment. Most treatment-related adverse events (TRAEs) of grade ≥3 included decreased neutrophil count (73.3%), decreased leukocyte count (63.33%), Anemia (10%), febrile neutropenia (6.67%). Conclusions: In pts of advanced gastric cancer who have previously received PD-1 inhibitors or not, this second-line combination therapy demonstrates promising preliminary effects. Pts who have not previously received ICIs might benefit more from this combination therapy. Clinical trial information: NCT06030934 .

Cancer-associated fibroblasts (CAFs) play a crucial role in the progression of pancreatic ductal adenocarcinoma (PDAC). Here, integrated single-cell RNA sequencing analysis is utilized to comprehensively map CAFs in the human PDAC tumor microenvironment (TME). Normal fibroblasts (NFs) and nine distinct CAF subtypes are identified including newly identified CAF subtypes, CDCP1+FTL+ CAFs, transitional CAFs (tCAFs), interferon simulated genes (ISG)+ myofibroblastic CAFs (myCAFs), and proliferative CAFs (pCAFs). CDCP1+FTL+ CAFs, pCAFs, and ISG+ myCAFs are associated with unfavorable clinical outcomes. CDCP1+FTL+ CAFs exhibit enhanced glycolysis and iron metabolism, resisting ferroptosis. The antigen-presenting CAFs (apCAFs) show high heterogeneity, consisting of multiple subtypes expressing distinct immune cell signatures. The CAF subtypes display differentiation plasticity, transitioning from early normal-like CAFs (nCAFs) to inflammatory CAFs (iCAFs) and myCAFs, ultimately leading to more invasive pCAFs. AP-1 family members FOS and JUN regulate the malignant phenotype conversion of NFs to nCAFs, while transforming growth factor-β (TGFβ) and interferon-γ (IFNγ) signals trigger the interconversion between classic myCAFs and iCAFs, respectively. A close interaction between CAFs and myeloid cells (especially neutrophils) is further observed in PDAC-TME, mainly mediated by CXCR4-CXCL12 chemotaxis. This work depicts a detailed CAF map and its dynamic interconvertible shift, providing important insights for combined targeted CAFs therapy.