Browse Articles

Sociodemographic disparities in sepsis-associated mortality among patients with gastrointestinal malignancies in the United States, 1999-2020.

e23077 Background: Patients with gastrointestinal (GI) malignancies are increasingly treated with multimodal systemic and procedural therapies, including cytotoxic chemotherapy, immune checkpoint inhibitors, and corticosteroids for treatment-related toxicities, which can compromise immune function and increase the risk of severe infections. However, population-level trends and sociodemographic disparities in sepsis-associated mortality among patients with GI malignancies in the contemporary treatment era remain incompletely characterized. This study evaluates long-term mortality trends for sepsis among patients with gastrointestinal malignancies. Methods: We analyzed U.S. death certificate data from the CDC WONDER database (1999–2020) for adults aged ≥25 years with gastrointestinal malignancies (ICD-10 C15–C26) listed as the underlying cause of death and sepsis (A40.0–A41.9) listed as a contributing cause. Age-adjusted mortality rates (AAMRs) per 100,000 population were calculated. Temporal trends were assessed using Joinpoint regression to estimate annual percent change (APC) and average annual percent change (AAPC) with 95% confidence intervals (CI), stratified by sex, race/ethnicity, U.S. census region, and urban–rural residence. Results: From 1999 to 2020, 140,650 deaths were identified among patients with GI malignancies and sepsis. Sepsis-associated mortality increased over time, with AAMRs rising from 2.98 in 1999 to 3.35 in 2020 (AAPC 0.83%, 95% CI, 0.44–1.22). Mortality rates were significantly higher in males than females (3.82 vs 2.22), with the most significant rise in male mortality occurring between 2012 and 2020 (APC 2.62, 95% CI, 1.84–3.40). Racial disparities were evident with consistently higher mortality rates observed among African American individuals throughout the study period (AAMR 4.77) as compared to Hispanics (3.41), Asians (3.36), American Indians (2.91), or Whites (2.71). Regionally, the South exhibited the greatest rise in mortality (AAPC 1.38%, 95% CI 0.88–1.88). Although large metropolitan areas had higher absolute mortality, rural populations experienced the sharpest increase in mortality over time (AAPC 1.81%, 95% CI 1.25–2.38). Conclusions: Sepsis-associated mortality in patients with GI malignancies has risen steadily in the United States over the past two decades, with a disproportionate burden among males, African Americans, and rural populations despite advances in cancer therapy. These findings underscore gaps in treatment-associated supportive oncology care, and structural factors influencing access to infection surveillance and timely access to high acuity care. This highlights the need for targeted interventions to diagnose and treat early sepsis, especially in vulnerable populations, to reduce preventable mortality in the modern era of GI cancer management.

Genomic and transcriptomic correlates of deep PSA response in patients with metastatic hormone-sensitive prostate cancer (mHSPC).

5089 Background: Despite advances in mHSPC treatment with combination therapies, patient outcomes remain heterogeneous. Achieving an undetectable PSA level is a strong prognostic marker for improved overall survival (OS). Baseline molecular predictors of deep PSA response are lacking. We aimed to identify genomic and transcriptomic determinants associated with undetectable PSA using a large clinical-genomic cohort. Methods: The Tempus Lens Platform was used to query the de-identified Tempus multimodal database and analyze a cohort of patients with mHSPC who underwent DNA (Tempus xT) and RNA (Tempus xR) clinical sequencing. Patients were restricted to those with stage IV disease within 90 days of sample collection. Samples collected within 12 months prior to or within 3 months of androgen deprivation therapy (ADT)-based therapy initiation were considered hormone-sensitive (n = 525). Patients were classified as PSA-low ( < 0.1 ng/mL, n = 285) or PSA-high (≥0.1 ng/mL, n = 240) based on the PSA level at 6 months (±45 days) post-treatment initiation. Genomic alterations, gene expression, and transcriptomic signatures were compared between PSA groups. Real-world OS was defined as the time from ADT start to death or loss to follow-up. Hazard ratios (HR) were estimated using Cox regression and adjusted for age. Significance was calculated using a Wald test at p < 0.05. Results: Median age at diagnosis was 67.5 years; 69.7% were White and 18.5% were Black/African American. The majority (91%) had de novo metastatic disease. First-line treatments included ADT+androgen receptor pathway inhibitor (ARPI) (65% vs 64%), ADT alone (15% vs 18%), ADT+ARPI+docetaxel (19% vs 15%), and ADT+docetaxel (1.3% vs. 3.5%) for PSA-low vs PSA-high, respectively. Baseline PSA was lower in patients with PSA-low vs PSA-high status (median 24 vs 36 ng/mL, p = 0.01). Somatic alterations in SPOP (17% vs 11%, p = 0.04) and ZFHX3 (2.5% vs 6%, p = 0.05) differed between patients with PSA-low and PSA-high, respectively. No significant differences in expression of actionable novel targets including PSMA, TROP2, B7-H3, or STEAP1 were observed between cohorts. Median OS was not reached for patients with PSA-low and was 47 months for patients with PSA-high, with 36-month landmark OS of 83% vs 66%, respectively. On multivariable analysis, PSA-low was independently associated with improved OS (HR 0.51, 95% CI 0.31-0.85, p = 0.01). Conclusions: In a real-world cohort, deep PSA response at 6 months portends significantly improved OS in patients with mHSPC and is associated with more SPOP and fewer ZFHX3 alterations. No transcriptomic differences in emerging therapeutic targets were identified. Validation in independent cohorts is warranted.

Lorlatinib as first-line therapy for ALK-positive advanced non-small cell lung cancer: A retrospective, multicenter, real-world study from China.

e20740 Background: The evolution of ALK tyrosine kinase inhibitors (TKIs) has led to significantly prolonged survival in patients with ALK-positive advanced non-small cell lung cancer (NSCLC). Notably, in the phase III CROWN study, the median progression-free survival (PFS) for the third-generation ALK TKI lorlatinib remained unreached after 5 years of follow-up. Currently, published real-world evidence on lorlatinib primarily focuses on its efficacy and safety in later-line therapy, and a significant gap exists in real-world data for lorlatinib as a first-line therapy. Therefore, we conducted this retrospective, multicenter, single-arm, real-world study in Zhejiang Province, China. Methods: This retrospective, multicenter, single-arm study enrolled adult patients with histologically or cytologically confirmed ALK-positive advanced NSCLC receiving first-line lorlatinib therapy from 2022 to 2025. Patient data were collected from institutional Electronic Medical Record (EMR) systems by investigators. The primary endpoint was PFS. Secondary endpoints included disease control rate (DCR), objective response rate (ORR), intracranial ORR, and safety. Results: As of the cutoff date, a total of 82 eligible patients with ALK-positive advanced NSCLC were included in this analysis. The median age was 52 (range: 26–86), 48.8% were female, 78.0% were never-smokers, 96.3% had adenocarcinoma histology, and 20.7% had baseline brain metastasis. The median follow-up of 15.07 months (95% CI: 10.77–19.36). The median PFS for first-line lorlatinib in patients with ALK-positive advanced NSCLC has not yet been reached (95% CI: NR–NR). The 12-month PFS rate was 96.1% (with baseline brain metastasis: 73.2%; without baseline brain metastasis: 100%). The ORR was 70.8% among evaluable patients (n = 72), with a 100% DCR and 92.9% intracranial ORR. The most common treatment-emergent adverse events (TEAEs) were hypertriglyceridemia (80.5%) and hypercholesterolemia (80.5%). The incidence of grade 3–4 TEAEs was 31.7%, which was predominantly comprised of hypertriglyceridemia (22.0%) and hypercholesterolemia (7.3%). No treatment-related deaths were reported. Conclusions: Lorlatinib demonstrated potent efficacy in the first-line treatment of Chinese patients with ALK-positive advanced NSCLC, achieving a 12-month PFS rate of 96.1% regardless of baseline brain metastasis status, with a safety profile consistent with the CROWN study. The most common grade 3–4 adverse events were hypertriglyceridemia and hypercholesterolemia. The study is ongoing, with plans to further expand the sample size and conduct long-term follow-up.

Predictors of in-hospital mortality among critically ill patients in ICU with cancer: A retrospective MIMIC-IV cohort analysis.

e23111 Background: Critically ill patients with cancer admitted to the intensive care unit (ICU) remain at high risk for adverse outcomes. Identifying early predictors of mortality could greatly improve risk stratification and clinical strategies. The goal of this study was to pinpoint correlations between various factors and in-hospital mortality among adult cancer patients admitted to the ICU. Methods: The research utilized the MIMIC-IV (v3.1) database in a retrospective cohort study design. The study population encompassed adult patients with a malignancy diagnosis who were admitted to the ICU between 2008 and 2019. Malignancy diagnoses were determined using ICD-9 and ICD-10 codes, focusing exclusively on each patient's initial ICU admission. Researchers extracted demographic data, comorbidity information, initial vital signs, laboratory results, SOFA scores, and Charlson Comorbidity Index (CCI) values. The primary outcome measured was in-hospital mortality. Multivariable logistic regression models, adjusted for 13 clinical and demographic factors, were developed to identify independent predictors, with model performance assessed using AUROC. Results: A total of 10,321 patients were included (mean age 66.4 ± 13.4 years; 58.3% male). Overall, in-hospital mortality was 20.2%. Non-survivors were older and had higher SOFA scores (7.5 vs 3.8, p < 0.001) and CCI (8.7 vs 7.7, p < 0.001). In the multivariable model, higher SOFA score (OR 2.68, 95% CI 2.50-2.88), Charlson Index (OR 1.46, 95% CI 1.36-1.57), and heart rate (OR 1.733, 95% CI 1.63-1.83) were independently associated with increased mortality, while higher MAP and Spo2 were protective. The model demonstrated good discrimination (AUROC 0.82). Conclusions: For critically ill cancer patients, the severity of their illness and the burden of comorbidities are the most significant predictors of in-hospital mortality. Early physiological parameters might provide a valuable tool for timely prognostication and aid in clinical decision-making. Clinical predictors of in-hospital mortality in critically Ill cancer patients. Predictor Variable Survivors (N = 8,240) Non-Survivors (N = 2,081) Adjusted Odds Ratio (95% CI) P-value Age, mean ± SD (years) 66.0 ± 13.5 68.2 ± 13.1 1.04 (1.00–1.07) <0.001 SOFA score, mean ± SD 3.8 ± 2.9 7.5 ± 4.2 2.68 (2.50–2.88) <0.001 Charlson Comorbidity Index 7.7 ± 3.0 8.7 ± 2.8 1.46 (1.36–1.57) <0.001 Heart rate (per unit increase) — — 1.73 (1.63–1.83) <0.001 Mean arterial pressure (MAP) — — 0.83 (0.78–0.89) <0.001 SpO₂ (per unit increase) — — 0.65 (0.61–0.69) <0.001 Platelet count — — 1.13 (1.07–1.20) <0.001 White blood cell count — — 1.09 (1.04–1.14) <0.001 Creatinine — — 0.91 (0.86–0.97) 0.003 Abbreviations: CI: Confidence Interval, MAP: Mean Arterial Pressure, OR: Odds Ratio, SOFA: Sequential Organ Failure Assessment, SpO₂: Peripheral Oxygen Saturation.

Real-world outcomes of first-line treatments in patients with metastatic non-small cell lung cancer: Final analysis of the Registurk-Lung registry.

e20633 Background: Real-world data are critical to understanding treatment patterns and outcomes in metastatic non–small cell lung cancer (mNSCLC) outside clinical trials. The REGISTURK-LUNG registry was designed to evaluate demographic, clinicopathological characteristics, treatment strategies, and survival outcomes of patients with mNSCLC treated in routine clinical practice in Turkiye. Methods: REGISTURK-LUNG is a multicenter, observational registry conducted across 41 centers. This final analysis includes 5,293 patients diagnosed with stage IV NSCLC between January 2022 and March 2024. Demographic features, histopathology, biomarker testing, first-line treatment patterns, progression-free survival (PFS), overall survival (OS), and response rates were analyzed. Survival outcomes were estimated using the Kaplan–Meier method and compared using log-rank tests. Hazard ratios (HRs) were calculated where applicable. Results: The median age was 68 years (range 29–94), and 84.1% of patients were male. Adenocarcinoma (52.8%) and squamous cell carcinoma (32.9%) were the most common histologies. PD-L1 testing was performed in 22.1% of patients. First-line treatment consisted of chemotherapy alone in 88.2%, immunotherapy-containing regimens in 11.8% (single-agent immunotherapy 2.4%; chemo-immunotherapy 9.4%). Median PFS was significantly longer in patients receiving immunotherapy-containing regimens compared with chemotherapy alone (7.5 vs 5.0 months; HR 0.73, 95% CI 0.59–0.91; p = 0.004). Median OS was also improved with immunotherapy-containing treatments (16.0 vs 13.6 months; HR 0.64, 95% CI 0.53–0.88; p < 0.001). Worse PFS and OS were observed in patients with multiple metastatic sites, brain or liver metastases, and poorer ECOG performance status. Higher PD-L1 expression was associated with improved OS (p = 0.001). Conclusions: In this large real-world registry of patients with metastatic NSCLC, first-line immunotherapy-containing regimens were associated with significantly improved PFS and OS compared with chemotherapy alone. These findings support the effectiveness of immunotherapy in routine clinical practice and highlight the prognostic impact of metastatic burden, performance status, and biomarker status in mNSCLC.

Gene–treatment interaction modeling to identify temozolomide-sensitive <i>IDH</i> -mutant diffuse gliomas.

2085 Background: IDH-mutant diffuse gliomas are biologically heterogeneous. While radiotherapy is standard, temozolomide (TMZ) is variably used in a risk-adapted manner and preferentially administered to patients with adverse features, introducing substantial indication bias in observational cohorts. Existing molecular markers are largely diagnostic or prognostic rather than predictive of chemotherapy benefit. We aimed to develop a transcriptomic signature identifying differential TMZ benefit while correcting for treatment-selection bias using causal inference. Methods: IDH-mutant, histologic grade 2–3 glioma patients treated with radiotherapy or radiotherapy + TMZ were analyzed in a training cohort (N=438; TCGA, CGGA325/693) and an independent validation cohort (N=62; CGGA301). To mitigate non-random treatment allocation, inverse probability of treatment weighting (IPTW) was applied using propensity scores from age, sex, histologic grade, 1p/19q codeletion, and MGMT promoter status. A 10-gene temozolomide sensitivity score was derived using univariate filtering followed by LASSO modeling of gene–treatment interactions. Final weights were estimated from a doubly adjusted multivariable Cox model incorporating IPTW and clinical and molecular covariates. Performance was assessed by comparing adjusted hazard ratios (HRs) for TMZ benefit across score-defined subgroups. Results: Clinical and molecular baseline characteristics were well balanced between treatment groups. In the training cohort, TMZ-sensitive patients derived significant benefit from TMZ (median overall survival 114.0 vs 85.7 months; adjusted HR 0.58, 95% CI 0.35–0.96; P=0.035). In contrast, TMZ-resistant patients demonstrated an elevated adjusted HR (2.98, 95% CI 1.63–5.45; P&lt;0.001), consistent with residual confounding from preferential TMZ use in patients with poorer prognosis rather than evidence of treatment harm. A significant difference in treatment effect was observed between subgroups (P&lt;0.001). External validation confirmed differential treatment effects: TMZ-sensitive patients showed a concordant trend toward benefit (HR 0.51, 95% CI 0.13–1.94; P=0.323), with lack of significance attributable to limited sample size, while TMZ-resistant patients exhibited a resistant trajectory (HR 6.04, 95% CI 1.56–23.38; P=0.009), with the difference in treatment effect remaining significant (P=0.011). Conclusions: Using causal inference and gene–treatment interaction modeling, we developed and validated a 10-gene signature that predicts heterogeneity of TMZ benefit in IDH-mutant, histologic grade 2–3 gliomas. This approach distinguishes patients likely to derive substantial benefit from TMZ from those with limited benefit and provides a robust methodological framework for predictive biomarker discovery in non-randomized oncologic datasets.

Sources matter: Learning to evaluate large language models for oncology treatment plans and changes observed by anchoring models to an oncology-specific database.

1617 Background: Large language models (LLMs) are increasingly evaluated for oncology decision support, yet guideline concordance and hallucinations remain variable across cancer types and tasks. Retrieval-augmented generation (RAG) may improve safety by constraining the evidence base used during generation. Prior oncology LLM studies report discordance with guideline-based care, but few describe a reproducible health-system “vetting mechanism” that couples guideline-only retrieval with clinician-scored safety rubrics across multiple oncology domains. Methods: We assembled 316 de-identified oncology cases in a tumor-board template: medical oncology (breast, GI, GU, leukemia; n=216), gynecologic oncology (n=50), and neuro-oncology/CNS metastatic cancer (n=50). Each case was run on three systems: baseline GPT-4 (O1), NCCN-guideline anchored RAG (O2; guideline-constrained retrieval), and a literature-anchored system (OpenEvidence; O3). Oncologists independently scored outputs using a modified Generative Performance Score (mGPS; −1 to +1) combining guideline concordance with hallucination penalties; readability/rationality was scored on a 1–5 Likert scale. Inter-rater reliability was assessed (κ =.63). This workflow was designed as a pre-deployment model credentialing step for health-system governance (source policy → case bank → clinician-scored acceptance criteria). Results: Across all 316 cases (case-averaged), O2 achieved the highest mean total mGPS (0.75) compared with O1 (0.43) and O3 (0.59). In the medical oncology cohort (n=216), O2 outperformed O1 and O3 (mean mGPS 0.72 vs 0.35 and 0.54), with higher guideline concordance (0.85 vs 0.68 and 0.75) and lower hallucination penalties (−0.09 vs −0.25 and −0.19); readability was also highest for O2 (4.17 vs 3.22 and 3.86). In gynecologic oncology (n=50), O2 again performed best (mGPS 0.827) relative to O3 (0.697) and O1 (0.647). In the CNS cohort (n=50), mean composite mGPS favored O2 (0.82) over O3 (0.68) and O1 (0.55), while readability remained stable across models (3.55–3.73). Severe hallucination/unsafe recommendation rate: O1 [26%], O2 [5%], O3 [17%] (pre-specified definition; e.g., major fabricated citation or guideline-discordant harmful recommendation). Conclusions: In a 316-case, multi-domain clinician-scored evaluation, limiting retrieval to oncology guideline sources improved safety-aligned performance relative to unconstrained generation or literature-only anchoring. A guideline-constrained RAG configuration paired with clinician-scored rubrics and tail-risk reporting can serve as a practical health-system vetting mechanism for governance, re-validation, and change control of LLM decision support, including in resource-limited settings where reliable guideline-consistent recommendations are essential.

Serum magnesium level and survival outcomes in extensive-stage small cell lung cancer treated with first-line chemoimmunotherapy: A retrospective cohort study.

e20030 Background: Recent studies suggested associations between higher serum Magnesium (Mg) level and better survival outcome in patients receiving immunotherapy, especially in small cell lung cancer (SCLC). However, the prognostic significance of Mg in extensive-stage (ES) SCLC treated with first-line chemoimmunotherapy remains unclear. We evaluated the association between baseline serum Mg and survival outcomes in ES-SCLC receiving platinum–etoposide plus PD-(L)1 blockade. Methods: We conducted a retrospective cohort study of adults with ES-SCLC treated with first-line platinum–etoposide plus atezolizumab or durvalumab at Mayo Clinic via Mayo Clinic Platform in 2018–2025. Baseline average serum Mg level was defined as the mean of all Mg labs from day −30 to day 0 relative to chemoimmunotherapy initiation and categorized as low ( &lt; 1.7 mg/dL), normal (1.7-2.0 mg/dL), or high ( &gt; 2.0 mg/dL) using clinical cutoff. Low and normal groups were combined a priori due to small low-Mg counts. The primary endpoint was overall survival (OS), estimated by KM and compared by log-rank test. OS was measured from chemoimmunotherapy initiation to death and censored at last follow-up. Multivariable Cox proportional hazards models adjusted for age, gender, ECOG, liver/brain metastases status, and smoking status. Results: The cohort included 214 patients, 124 (58%) were female; median age was 68 years; 49% had liver metastases and 30% had brain metastases. Median OS was 10.4 months (95% CI 7.6-15.9) among 99 patients with high serum Mg level, compared with 7.5 months (95% CI 6.7-10.7) in 115 patients with low or normal serum Mg level (Log-rank p = 0.126). High Mg was associated with a trend toward longer OS in unadjusted analyses (Hazard ratio [HR] 0.80, 95% CI 0.59-1.07, p = 0.129); however, the association was not statistically significant after multivariable adjustment (aHR 0.81, 95% CI 0.60-1.10, p = 0.180). Conclusions: Among patients with ES-SCLC treated with first-line platinum–etoposide plus PD-(L)1 blockade, higher baseline serum Mg was associated with numerically longer OS but was not independently associated with survival after multivariable adjustment. These findings suggest baseline serum Mg level may be associated with prognosis and warrant validation in larger cohorts. Baseline characteristics by baseline serum magnesium category in ES-SCLC treated with first-line chemoimmunotherapy. Variable Overall (N=214) High Mg &gt;2.0 (n=99) Low/Normal Mg ≤2.0 (n=115) p value* Age, median (IQR) 68 (62, 74) 68 (61, 74) 69 (62, 75) 0.201 Female, n (%) 124 (58%) 57 (58%) 67 (58%) &gt;0.999 ECOG ≥2, n (%) 58 (27.1%) 28 (28%) 30 (26%) 0.719 Liver metastases, n (%) 105 (49%) 49 (49%) 56 (49%) 0.648 Brain metastases, n (%) 65 (30%) 33 (33%) 32 (28%) 0.469 Ever smoker, n (%) 197 (92.1%) 90 (90.9%) 107 (93.0%) 0.974 *p values: Wilcoxon (age); χ²/Fisher (categorical).

Tumor evolution following treatment with sacituzumab govitecan for metastatic breast cancer: A proteomic analysis.

e13006 Background: Sacituzumab Govitecan (SG) is a Trop2-targeting antibody drug conjugate (ADC) approved for the treatment of HER2-non amplified metastatic breast cancer (MBC). While changes to receptor target density can be associated with response and resistance in HER2-targeted ADCs, the relationship between dynamic changes in Trop2 (or other ADC targets) and SG resistance is unknown. Methods: Institutional databases were used to identify patients treated with SG for MBC who had paired pre and post treatment biopsies. Paired tissue samples were sectioned and submitted to both mass spectrometry (MS, mPROBE) and multiplex immunofluorescence (mIF, Zeiss). Proteomic quantification of key proteins was compared using the Wilcoxon signed rank test. mIF was described qualitatively. Results: Fourteen patients had tissue pairs from both pre and post-SG that passed quality control and underwent quantitative analysis. Seven patients (50%) had ER+ disease. Median progression free-survival (PFS) on SG was 4.7 months and median overall survival was 15.8 months. Baseline Trop2 expression ranged from 296-6,453amol/ug (median 1,560 amol/ug). At the individual patient level, temporal changes in Trop2 expression varied between -96% to 400% (median -18%) after SG treatment. Similar findings were apparent with TOPO1, where 6/14 (43%) of patients had &gt; 25% loss in TOPO1 expression, and the remaining had stable or increased expression. In aggregate, after adjusting for multiple hypothesis testing, no unidirectional quantitative changes in mean protein expression were observed for Trop2 (-393 amol/ug, p = 0.6), TOPO1 (-200 amol/ug, p = 0.12), TOPO2A (-40 amol/ug, p = 0.2), SLFN11 (66 amol/ug, p &gt; 0.9), HER2 (-53 amol/ug, p &gt; 0.9), HER3 (-16 amol/ug, p = 0.8), or Nectin4 (-127 amol/ug, p = 0.042, q = 0.6). mIF results helped visualize protein changes in Trop2, HER2, HER3, and Nectin4 and corroborated observed changes seen from mass spectrometry. Conclusions: Dynamic patient-level changes in Trop2 and several other putative ADC biomarkers were observed, suggesting heterogeneity in ADC resistance. Further mechanistic investigation will be important to understanding heterogeneity in resistance, and quantitative proteomic methods such as MS and mIF may have utility in characterizing ADC resistance and response.

IBI363 (TAK-928) plus chemotherapy as first-line (1L) treatment for advanced non–small cell lung cancer (NSCLC).

8586 Background: IBI363 is a first-in-class, PD-1/IL-2 α-bias bispecific antibody fusion protein with great potential for immune-cold and immunotherapy (IO)-resistant tumors. Previously, IBI363 monotherapy demonstrated encouraging and durable efficacy in patients (pts) with IO-treated advanced NSCLC (2025 ASCO #8509). Here, we report a phase 1 study evaluating IBI363 plus platinum-based doublet-chemotherapy (PDC, pemetrexed/paclitaxel plus platinum) in 1L NSCLC. Methods: Previously untreated pts with advanced NSCLC, without sensitizing EGFR, ALK, or ROS1 alterations were enrolled. The safety lead-in evaluated IBI363 at 1.5 or 3 mg/kg Q3W plus PDC. In dose optimization (PD-L1 TPS &lt; 50% required), pts were randomized 1:1:1 to 3 cohorts: 3-1.5 mg/kg (IBI363 3 mg/kg plus PDC in cycle 1, then 1.5 mg/kg Q3W plus PDC in subsequent cycles), 3 mg/kg (IBI363 3 mg/kg Q3W plus PDC) and 1.5 mg/kg (IBI363 1.5 mg/kg Q3W plus PDC). Stratification factors were squamous vs non-squamous, and PD-L1 TPS &lt; 1% vs 1-49%. Primary endpoints were safety and efficacy assessed by investigators per RECIST v1.1. Secondary endpoints included pharmacokinetics (PK) and immunogenicity. Results: As of December 22, 2025, 80 pts were enrolled in safety lead-in (N = 11) and dose optimization (N = 69) with median follow-up (mFU) of 5.8 months (range: 0.9-9.5). Baseline characteristics (median age: 64 years, male: 88.8%, ECOG PS 1: 81.3%, stage IV: 72.5%, sqNSCLC: 66.3%) were balanced among 3-1.5 mg/kg (N = 23), 1.5 mg/kg (N = 28) and 3 mg/kg (N = 29) cohorts. Median treatment duration of IBI363 was 25.0 weeks (range: 4.0-41.1) while 65.0% pts with ongoing treatment, and 88.9% pts having completed ≥4 cycles of PDC. Grade≥3 (G3+) treatment-emergent adverse events (TEAEs) occurred in 81.3% pts including 65.2% for 3-1.5 mg/kg, 82.1% for 1.5 mg/kg and 93.1% for 3 mg/kg cohorts. Common TEAEs were anemia (any grades 78.8%, G3+ 18.8%), neutrophil count decrease (75.0%, G3+ 42.5%), white blood cell count decrease (63.8%, G3+ 20.0%), arthralgia (51.3%, G3+ 2.5%), and platelet count decrease (45.0%, G3+ 17.5%). IBI363-related AEs led to corresponding treatment discontinuation and death in 6.3% and 1.3% pts. In dose optimization, 65.2% pts had PD-L1 TPS &lt; 1 and 34.8% had TPS 1-49%. Efficacy was evaluable in 62 pts with at least 1 tumor assessment. For 3-1.5 mg/kg cohort (N = 22), ORR was 86.4% (95% CI: 65.1-97.1, confirmed ORR [cORR]: 81.8%) and DCR was 100% (95% CI: 84.6-100). ORR was 85.7% for sqNSCLC (N = 14) and 87.5% for non-sqNSCLC (N = 8). For the 1.5 mg/kg (N = 19) and 3 mg/kg cohorts (N = 21), ORR was 57.9% (cORR: 42.1%) and 66.7% (cORR: 57.1%). PFS was immature (events 28.8%). Clinical PK data also supported an optimal benefit-risk profile based on efficacy and safety observed at 3-1.5 mg/kg. Conclusions: IBI363 plus PDC was well tolerated and showed encouraging efficacy in advanced NSCLC. Safety, efficacy, and PK data supported 3-1.5 mg/kg as the recommended dose. Clinical trial information: NCT06468098 .

CDH6-high gastric cancers as exhibitors of a reproducible vascular–stromal, immune-excluded tumor microenvironment.

e15179 Background: Cadherin-6 (CDH6) is a developmental adhesion molecule aberrantly expressed in subsets of gastric cancer and associated with invasive behavior. While CDH6-high tumors have been hypothesized to exhibit immunosuppressive tumor microenvironment (TME) features, the extent to which CDH6 expression reflects coordinated vascular, stromal, and immune organization across cohorts has not been systematically evaluated. Methods: We conducted a multi-cohort analysis of gastric cancer RNA-sequencing datasets. Within each cohort, tumors were stratified into CDH6-high and CDH6-low groups using cohort-specific median transcripts per million (TPM). Immune and stromal cell-type enrichment was inferred using the xCell deconvolution algorithm. Associations with angiogenic signaling were explored through transcriptomic correlation with vascular endothelial growth factor (VEGF) family members. Results: Across ten independent cohorts, comprising 1,811 clinically annotated samples, CDH6-high tumors demonstrated a reproducible vascular–stromal–enriched tumor microenvironment. Endothelial cells were increased in 10/10 cohorts (p &lt; 0.05), accompanied by enrichment of hematopoietic stem cells (9/10), pericytes (5/10), and elevated stromal scores (5/10), with concomitant epithelial depletion (5/10). Immune profiling showed consistent directional immune exclusion, including reduced γδ T cells (9/10), Th1 (6/10), Th2 (7/10), NK cells (5/10), CD8⁺ T-cell subsets (2–4/10), and reduced immune scores (0/10), with greater consistency in vascular and stromal features than immune effect sizes. Angiogenic signaling patterns were concordant with this phenotype, with VEGFB- and VEGFC-associated tumors showing endothelial expansion and increased stromal and microenvironment scores (9–10/10 cohorts), in contrast to VEGFA-associated stromal depletion and epithelial predominance. Conclusions: CDH6-high gastric cancers are associated with a reproducible, structurally organized vascular–stromal tumor microenvironment characterized by immune exclusion rather than immune absence. These findings support CDH6 expression as a marker of tumor microenvironmental architecture and highlight the relevance of angiogenic context for therapeutic stratification.

Temporal trends and disparities in gallbladder cancer mortality: A U.S. population-based study.

e16175 Background: Gallbladder cancer (GBC) is a relatively uncommon malignancy with one of the most unfavorable prognoses among gastrointestinal cancers and is characterized by marked geographic, racial, and sex-based disparities. Despite declining incidence in many regions, comprehensive national evaluations of long-term GBC mortality trends in the United States remain limited. Methods: Death certificate records from CDC WONDER were analyzed for decedents aged 25 years and above with gallbladder cancer (ICD-10 code C23) listed as the cause of death during 1999–2020. Age-adjusted mortality rates (AAMRs) per 100,000 and annual percent changes (APC) were derived, with results stratified by sex, age group, racial group, U.S. Census region, and urban–rural status. Statistical significance was determined at a p-value of &lt; 0.05. Results: From 1999–2020, GBC accounted for 47,504 deaths in the United States. The AAMR declined significantly from 2.32 in 1999 to 1.62 in 2020 (APC −1.44; 95% CI −1.65 to −1.24), with an overall AAMR of 1.83 (95% CI 1.81 to 1.85). Overall AAMR remained higher in females (2.21) compared to males (1.35). Female AAMR decreased from 2.78 to 1.96 (APC −1.45, 95% CI: -1.68 to -1.22), while male AAMR declined from 1.60 to 1.25 (APC −1.20, 95%CI -1.47 to -0.92). Age-stratified analysis showed substantially higher mortality in older adults. AAMR declined from 5.01 to 3.38 (APC −1.60) in the older age group, compared with a smaller decline in the middle-aged group (0.79 to 0.61; APC −0.91). Mortality among young adults was too sparse for trend analysis. By race/ethnicity, the highest overall AAMR was observed among Hispanic/Latino individuals (2.83), followed by Black/African American (2.60), Asian/Pacific Islander (2.09), and White populations (1.64). Significant declines were observed among Whites, Hispanics/Latinos, and Asian/Pacific Islanders, while trends among Black/African American individuals remained stable. Geographically, states in the highest AAMR decile included New York, New Mexico, and the District of Columbia, whereas Montana had the lowest overall AAMR. Mortality was higher in metropolitan compared with non-metropolitan areas throughout the study period. Regionally, the Northeast had the highest overall AAMR, followed by the Midwest, West, and South, despite significant declines across all regions. Conclusions: Although GBC mortality has declined nationwide over the past two decades, disparities persist by sex, race/ethnicity, geography, and urbanization. NH Black and Hispanic populations, as well as residents of high-burden states, continue to face disproportionate mortality. Targeted prevention, early detection, and equitable treatment access are urgently needed to reduce these inequities.

Timing of biomarker testing and associated clinical outcomes in ovarian cancer patients: A retrospective study.

e17574 Background: In ovarian cancer (OC) patients, biomarker testing can inform treatment decisions and thereby guide optimal management. The timing of testing for biomarkers such as BRCA , homologous recombination deficiency (HRD), and folate receptor α (FRα) is not currently standardized in clinical practice. We aimed to characterize biomarker testing rates and the impact of timing of testing on clinical outcomes in OC patients. Methods: This retrospective study utilized the US-based electronic health record-derived deidentified Flatiron Health Research Database to identify adult patients diagnosed with OC from Nov 2022 to Sep 2025 who had biomarker records available at the time of or post OC diagnosis. FRα immunohistochemistry (IHC), BRCA , and HRD testing rates were examined. Among patients who received ≥1 biomarker test and ≥1 line of therapy (LOT), we evaluated the impact of timing of biomarker testing post-diagnosis (0-3, 3-6, ≥6 mo) on first-line progression free survival (PFS) and overall survival (OS) using Cox proportional hazard models with inverse probability weights adjusted for patient demographic and clinical characteristics. A sensitivity analysis was conducted in patients with ≥2 LOT. Results: Of 1685 patients identified, testing rates were 92.3% for BRCA , 59.9% for HRD, and 43.8% for FRα IHC; the FRα IHC retesting rate was 7.9%. FRα testing rates increased in 2023–2025. Among 1029 patients (0-3 mo: n=782; 3-6 mo: n=151; ≥6 mo: n=96) with ≥1 biomarker test and ≥1 LOT, the adjusted hazard ratios (aHRs) for OS (95% CI) were 1.60 (1.08–2.39) and 1.46 (0.90–2.35) in patients tested at 3-6 and ≥6 mo, respectively, compared to patients tested 0-3 mo post-diagnosis, representing a 32-38% risk reduction of death in the 0-3 mo group. The PFS (95% CI) aHRs were 1.15 (0.85–1.55) and 1.45 (1.06–1.97) in patients tested 3-6 and ≥6 mo, respectively, vs patients tested 0-3 mo post-diagnosis. Among 504 patients (0-3 mo: n=233; 3-6 mo: n=113; ≥6 mo: n=158) with ≥1 FRα IHC test and ≥1 LOT, the OS (95% CI) aHRs were 2.14 (1.20–3.80) and 2.03 (1.20–3.41), representing a 51-53% lower risk of death in the 0-3 mo group; plus, PFS (95% CI) aHRs were 1.76 (1.20–2.58) and 2.59 (1.88–3.55) in patients tested at 3-6 and ≥6 mo, respectively, vs patients tested 0-3 mo post-diagnosis. Sensitivity analyses conducted in patients with ≥2 LOT showed consistent results. Conclusions: Testing OC patients early for BRCA , HRD, or FRα was associated with a significant reduction in risk of death and disease progression. The findings suggest that improvement in early biomarker testing has the potential to improve survival outcomes in OC patients.

Recurrence patterns and site-level survival in the previous phase III study of IL13-PE38QQR: Implications for drug distribution variability.

2017 Background: Interleukin-13–Pseudomonas Exotoxin A (IL13-PE38QQR) is a targeted cytotoxic fusion protein directed against the IL13α2 receptor expressed on glioblastoma stem cells and absent in normal brain tissue. The drug is administered by convection-enhanced delivery (CED) for recurrent glioblastoma (GBM). Early-phase studies demonstrated durable tumor control and long-term survival; however, a pivotal Phase III trial failed to meet its primary endpoint. We hypothesized that variability in clinical outcomes was driven primarily by limitations in drug delivery rather than tumor biology. Methods: We conducted a post hoc analysis of clinical, imaging, and survival data from the randomized Phase III PRECISE trial of IL13-PE38QQR versus Gliadel wafers. Treatment-site experience was defined as ≥3 efficacy-evaluable IL13-PE38QQR patients. A subset of patients underwent radiolabeled drug infusion to assess intraparenchymal distribution. Recurrence patterns were evaluated using serial MRI. Survival outcomes were compared by treatment arm and site experience. Results: Marked heterogeneity in drug distribution was observed in patients receiving radiolabeled IL13-PE38QQR, ranging from broad target coverage to minimal or absent delivery. In the IL13-PE38QQR arm, treatment at experienced centers correlated with significantly improved overall survival (median OS 48.4 vs 32.1 weeks; p=0.038), whereas site experience had no significant impact in the control arm. A 50% increase in 1-year survivorship was observed in the treatment arm compared with control. Long-term survivors most commonly exhibited distant or multifocal recurrence with relative preservation of the treated region, suggesting durable local tumor control where therapeutic exposure was achieved. Conclusions: Clinical outcomes following IL13-PE38QQR therapy were strongly associated with delivery efficiency and treatment-team experience. Radiolabeled distribution studies, experience-dependent survival effects, and recurrence patterns in long-term survivors indicate that outcome variability was driven primarily by technical limitations in drug delivery rather than tumor biology. This study represented the first large-scale evaluation of CED in the absence of purpose-built delivery technology, relying on complex, nonstandardized surgical workflows. These findings highlight the importance of standardized procedures, specialized training, and purpose-built delivery platforms to enable reproducible therapeutic exposure and optimize the clinical potential of IL13-PE38QQR. Clinical trial information: NCT00076986 .

Neoadjuvant chemotherapy with/without PD-1i in locally advanced laryngeal/hypopharyngeal cancer: A multicentered retrospective study.

6085 Background: The efficacy of adding programmed death-1 (PD-1) inhibitors to neoadjuvant chemotherapy (NACT) for improving survival in locally advanced laryngeal/hypopharyngeal carcinoma remains unclear. This study aimed to compare progression-free survival (PFS) between patients receiving NACT with or without a PD-1 inhibitor (NACT+PD-1i). Methods: This multicenter retrospective study enrolled 267 patients with T2-4N0-3M0 laryngeal/hypopharyngeal squamous cell carcinoma who received NACT with (n=165) or without (n=102) a PD-1 inhibitor, followed by definitive radiotherapy, between 2019 and 2024, from Fudan University Shanghai Cancer Center, Fujian Cancer Hospital and Sun Yat-sen University Cancer Center. The primary endpoint was PFS. Inverse probability of treatment weighting (IPTW) was used to adjust for baseline imbalances. Results: The NACT+PD-1i group demonstrated significantly improved PFS versus NACT alone (median PFS: not reached vs. 33.0 months; hazard ratio [HR]=0.61, 95% confidence interval [CI]: 0.41–0.90; p=0.012). The 2-year PFS rates were 70.2% and 56.5%, respectively. This PFS benefit persisted after IPTW (p=0.012). PD-1 inhibition was an independent favorable factor for PFS in multivariable analysis (unadjusted, HR=0.54, p=0.005; IPTW-adjusted, HR= 0.63, p=0.040). Notably, the NACT+PD-1i group showed superior regional recurrence-free survival (RRFS) (2-year RRFS: 92.1% vs. 81.0%; HR=0.31, p=0.001) and higher objective response rate (ORR) (93.9% vs. 85.9%, p=0.027), particularly for cervical lymph nodes (94.2% vs. 81.6%, p=0.002). Achieving an ORR after neoadjuvant therapy was a strong predictor for improved outcome, either in unadjusted PFS (HR=0.40, p=0.0005) or in IPTW-adjusted PFS (HR=0.49, p=0.0005). Conclusions: In this real-world analysis, adding a PD-1 inhibitor to NACT significantly improved ORR and PFS in locally advanced laryngeal/hypopharyngeal carcinoma, supporting its further evaluation in neoadjuvant strategies.

Biomarker-driven assessment of immunochemotherapy with or without fruquintinib as first-line treatment for advanced gastric/GEJ adenocarcinoma: Initial clinical results and subgroup analysis from the MGC-FLORA study.

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.

Neoadjuvant systemic chemotherapy combined with camrelizumab (CAM) and apatinib (APA) for BCLC stage A/B hepatocellular carcinoma (HCC) beyond Milan criteria: An interim analysis of a phase 2 trial.

e16283 Background: Patients(Pts) in BCLC stage A/B HCC beyond Milan criteria have a high recurrence rate after radical surgery. However, the standardized perioperative treatment strategy has not been established.This study aimed to investigate to efficacy and safety of systemic chemotherapy with mFOLFOX7 regimen combined with camrelizumab (CAM, immune checkpoint inhibitors) plus apatinib (APA, anti-angiogenesis) as conversion therapy for potentially resectable HCC. Methods: In this a prospective, single arm, phase II study, treatment-naive pts with BCLC stage A/B HCC beyond Milan criteria (assessed by MDT as unsuitable for radical resection surgery) were included.The enrolled pts received CAM (200 mg, q3w), APA (250 mg, qd), and mFOLFOX7 (oxaliplatin 85 mg/m 2 , leucovorin 200 mg/m 2 , 5-fluorouracil bolus 400 mg/m 2 on day 1, and 5-fluorouracil infusion 2400 mg/m 2 for 46 hours; q3w; up to 4 cycles). The surgical feasibility of pts was evaluated every 2 cycles . Sequential therapy began 2 weeks after surgery, continuation of CAM and APA was administered for up to 12 months, approximating 17 cycles.The primary endpoint was major pathological reactions (MPR), defined as viable tumor cells in less than 50% of tumor bed. Secondary endpoints were complete pathological reactions (pCR), objective response rate (ORR, by RECIST v1.1 and mRECIST), disease-free survival (DFS), overall survival (OS), and safety. Exploratory endpoint is presence of microvascular invasion (MVI), defined as postoperative histologically confirmed MVI. Results: At the data cut-off (Jan. 20, 2026), a total 21 pts were screened. Among these, 19 pts were enrolled (3 woman and 16 men; median age, 61years; range, 38 - 73 years) and received neoadjuvant therapy. The median tumor size was 71 mm (range, 27 - 147). 17 pts had HBV infection and 9 pts had more than 1 tumors. Out of 19 pts, 13 pts were in follow-up/ finished treatment(4 pts recurrence after surgery), 4 pts withdraw from treatment (inoperability or refusal of surgery), while 2 pts were still receiving CAM plus APA. After a median of 3 (2-4) treatment cycles, 15 pts have undergone hepatic surgery, R0 resection rate was100%, 9 (60.0%) pts achieved MPR. Among pts with MVI, M0 was found in 86.7% (13/15) pts, M1 was found in 13.3% (2/15) pts. 15 pts Hepatic Reserve Function (Indocyanine Green-15min) was normal before hepatic surgery. The most common TEAE included Alanine aminotransferase increased (52.6%), Platelet count decreased (42.1%) and Hypoalbuminemia (42.1%); and Grade 3-4 TRAE occurred in 7 of 19 pts. None of them had Serious Adverse Drug Reactions (SADRs). Conclusions: Neoadjuvant mFOLFOX7 combined with CAM and APA is effective and tolerable in pts with BCLC stage A/B HCC beyond Milan criteria. Preoperative neoadjuvant therapy has reduced the rate of MVI. More data would be further analyzed and reported. Clinical trial information: NCT06607107 .

Identification of drug-related cardiovascular risks: A comprehensive analysis using the FAERS database (2004–2024)

Objective: This study aimed to analyze drug adverse events, identify high-risk medications associated with cardiovascular disease (CVD), and provide evidence for clinical medication safety. Methods: We conducted a retrospective analysis of the FAERS database (2004Q1–2024Q4) using nine CVD-related terms from MedDRA 27.0. Four signal detection methods—PRR, ROR, BCPNN, and MGPS—were employed to identify significant drug—event associations. Results: A total of 1,560,242 CVD-related reports were included. Rofecoxib, phenylpropanolamine and rosiglitazone were strongly associated with CVD, with ROR values of 73.4 (95% CI: 71.16–75.7), 66.8 (95% CI: 23.25–191.94), and 58.69 (95% CI: 57.68–59.76), respectively. These drugs have since been withdrawn or restricted. Subgroup analysis revealed gender and age differences in CVD risk: testosterone related drugs may be associated with cardiovascular disease risk in elderly and male patients, while alendronic acid may be associated with increased cardiovascular risk in female patients. These preliminary associations may stem from mixed indications rather than direct effects of the drugs. Validate and quantify in the real world doxorubicin and other anticancer agents are correlated with cardiotoxicity in children known risks. Additionally, potential associations with previously unreported CVD risks were suggested for paricalcitol, busulfan, gemtuzumab ozogamicin, clofarabine, and clofazimine, which warrant further attention. Conclusion: This study showed that some drugs were significantly associated with CVD, and some of them were newly discovered signals. These findings provide a basis for the generation of new hypotheses about the cardiovascular risk of drugs, but the causal relationship needs to be confirmed by further research.

A Trade‐Off Between Thermodynamics and Kinetics of O ₂ Binding for Highly Active and Selective Electrocatalytic Oxygen Reduction Reaction

ABSTRACT O 2 binding to metal ions is an essential step in the oxygen reduction reaction (ORR). Tailoring O 2 binding is anticipated to realize highly active and selective four‐electron ORR. Herein, we report on fine‐tuning the molecular pocket of Co porphyrins to get a trade‐off between thermodynamics and kinetics of O 2 binding for ORR with high activity and selectivity. Three Co tetra (2‐amidophenyl)porphyrins with an αααα structure but different steric hindrance at the pocket bottleneck are synthesized. The pocket thermodynamically favors O 2 binding by stabilizing O 2 adducts and thus improves the 4e selectivity, but the sterically hindered pocket entrance may obstruct the access of O 2 to Co. By tuning the pocket entrance, we tailored the rate constant and equilibrium constant of O 2 binding, and consequently, we achieved efficient ORR with a half‐wave potential of 0.83 V versus RHE and a transferred electron number of 3.78. This performance is remarkable among reported mononuclear Co porphyrins.

Synthesis of Sb2S3 nanostructures and carbon-based composites: Enhancing electrochemical catalysis and DSSC performance