Browse Articles

Beyond survival: Longitudinal prognostication in glioblastoma using conditional survival—A population-based study.

2045 Background: Glioblastoma multiforme (GBM) carries a poor prognosis, and the initial survival projections provided at diagnosis do not accurately depict the evolving risk that exists for all GBM survivors. Conditional survival (CS), defined as the probability of surviving an additional period after having lived to a specific time point, provides a much more accurate way of defining prognosis for these patients. We evaluated conditional survival patterns in a large population-based cohort of patients with glioblastoma. Methods: Adult patients who were diagnosed with primary glioblastoma between 2000 and 2022 were identified from the SEER 17 Registries. The estimated overall survival (OS) was determined by using the Kaplan-Meier method. We also calculated one year and two-year conditional survival (CS) probabilities at multiple landmark times (0-24 months). In addition, we stratified the conditional survival by the status of surgical resection and whether or not the patient received radiation therapy. Results: There were 52,125 patients in our sample. As expected, the median OS was very poor, dropping from 37.3% at 12 months to 16.0% at 24 months. Nevertheless, the conditional survival was significantly better than overall survival. One-year CS increased from 40.4% at diagnosis to 57.2% among patients surviving 24 months, while two-year CS improved from 17.3% to 39.4% over the same landmarks. Patients who underwent surgical resection had higher early CS compared with biopsy or no surgery, though differences diminished at later landmarks. Similar trends were found in CS rates in patients who received radiation therapy, indicating survivorship selection over time as opposed to a decrease in the effectiveness of treatments. Conclusions: Conditional survival in glioblastoma improves markedly with increased time of survival. Therefore, it is essential to use CS in place of traditional survival statistics to provide more accurate prognoses for glioblastoma patients. Landmark Time (Months) Overall Survival 1-Year Conditional Survival 2-Year Conditional Survival 1-Year CS (Resection) 1-Year CS (No resection/Biopsy) 1-Year CS (RT Yes) 1-Year CS (RT No/Unknown) 0 92.4% 40.4% 17.3% 46.7% 17.1% 48.7% 14.1% 6 56.3% 42.0% 20.9% 43.2% 32.3% 42.5% 37.9% 12 37.3% 42.9% 24.5% 43.1% 40.7% 42.4% 47.9% 18 23.7% 49.8% 31.7% 49.9% 48.7% 49.2% 56.0% 24 16.0% 57.2% 39.4% 57.1% 57.2% 56.5% 63.2% RT: Radiotherapy.

Clinicopathological analysis and therapeutic management of radiation-induced sarcomas of the head and neck: A single-center retrospective study.

e23537 Background: Radiation-induced sarcoma of the head and neck (RISHN) is a rare but devastating late toxicity of curative radiotherapy (RT). We conducted a single-center cohort study to analyze the clinicopathological characteristics, management, survival, and hemorrhagic risk of RISHN and to identify associated prognostic factors and effective clinical treatments. Methods: We conducted a retrospective analysis of 25 pathologically confirmed RISHN patients treated at Fudan University Shanghai Cancer Center (January 2019 to December 2025), constituting the largest single-institution case series to date. The clinical data include diagnostic age, RT dose, latency, RISHN site, histological subtype, stage, first symptoms and treatment. The survival analysis was estimated by the Kaplan–Meier method. Results: The median age at RISHN diagnosis was 47 years and the median latency was 10.7 months. The most common primary tumor was nasopharyngeal carcinoma (NPC, 21 cases, 84.0%), receiving RT doses ≥66 Gy. RISHN arose chiefly in paranasal sinuses and nasopharynx. Bleeding and nasal congestion were most common presenting symptoms. The Median follow-up time was 24.2 months. The major histological subtypes of RISHN included post-radiation sarcoma (11 cases), undifferentiated sarcoma (4 cases) and osteosarcoma (4 cases). Surgery was the cornerstone of initial management and gross total resection (GTR) was achieved in 11 patients. However, the median EFS was only 10.4 months with the 1-year EFS rate of 41.1%. GTR conferred a significant EFS benefit over nonGTR (incomplete resection or systemic therapy alone), with median EFS times of 12.9 versus 6.9 months (p=0.03; HR=0.34). The median OS was 28.3 months and a numerical advantage in median OS was observed for GTR group (36.9 months) over nonGTR group (20.0 months) (p=0.17, HR=0.40). In 18 patients receiving first-line systemic therapy, the disease control rate (DCR) was 83.4% and the median PFS was 7.7 months. Exploratorily, patients who received immunotherapy had numerically longer PFS (12.7 vs 6.0 months, p=0.53, HR=0.66) and OS (48.9 vs 28.3 months, p=0.99, HR=1.01) than those who did not. Hemorrhage was a notable clinical manifestation especially in anlotinib-treated RISHN patients, whereas none of six patients undergoing prophylactic carotid embolization experienced significant bleeding subsequently. Conclusions: RISHN had a poor prognosis. GTR was an independent prognosis predictor of RISHN, while the EFS was unsatisfactory. Systemic therapy offers an important treatment option. First-line treatment with immunotherapy may improve survival outcomes in RISHN patients. Hemorrhage mitigation, particularly careful use of anti-angiogenic therapy and selective prophylactic endovascular intervention, should be integrated into multidisciplinary care.

Clinical and genetic landscape of individuals evaluated in a CHIP cardiology clinic: A single-center experience, 2020 to 2025.

6593 Background: Despite growing observational data linking clonal hematopoiesis of indeterminate potential (CHIP) to adverse cardiovascular outcomes, the cardiovascular evaluation and management of individuals with clonal hematopoiesis remain undefined. We report the experience of a dedicated CHIP Cardiology Clinic. Methods: We conducted a retrospective review of patients evaluated in the CHIP Cardiology Clinic at the Brigham and Women’s Hospital. Baseline demographics, comorbidities, medication use, and treatment interventions were extracted from the electronic medical record. Results: Fifty-three patients were evaluated between 2020 and 2025, of whom 39 (73.6%) had clonal hematopoiesis (CH), out of which 12 (30.8%) met criteria for CHIP. Among individuals with CH, the most frequently identified mutations were D N M T 3 A (30, 76.9%), T E T 2 (23, 59.0%), and A S X L 1 (6, 15.4%). The mean age was 65.9 years (range 40-81), and 23 (58.9%) were male. Hypertension and dyslipidemia were each present in 23 (59.0%) patients. Existing cardiovascular diseases included coronary artery disease (7, 18.0%) and arrhythmias (3, 7.7%). Following evaluation in the CHIP Clinic, and shared decision-making discussion in view of the dearth of evidence regarding outcome improvement in those with CHIP, lipid-lowering therapies were initiated or increased in 28 (71.8%) patients, with statins initiated or dose escalated in 21 (53.8%) and ezetimibe in 11 (28.2%). Blood pressure medications were initiated or intensified in 5 (12.8%) patients. Ischemic evaluation was performed in 3 (7.7%) patients due to suggestive symptoms. We developed a “patient page” to acquaint the patients and families with CHIP. (https://doi.org/10.1001/jamacardio.2024.3773). Conclusions: In this first longitudinal report from a dedicated CHIP Cardiology Clinic, a substantial proportion of patients harbored somatic mutations associated with elevated cardiovascular risk. Institution of preventive cardiovascular medications, in particular lipid-lowering therapies, was a frequent intervention. These findings support the feasibility of CHIP-focused cardiovascular care but underscore the need for prospective trials to evaluate the impact of targeted preventive strategies in this population.

Partial versus radical nephrectomy for renal cell carcinoma: A systematic review and meta-analysis.

e16551 Background: Renal cell carcinoma (RCC) is a common urological challenge, for which radical nephrectomy (RN) has served as the traditional surgical treatment for a long time. However, partial nephrectomy (PN) has gained acceptance as a nephron-sparing alternative and has the benefits of preserving renal function and reducing long-term metabolic morbidity. Although PN is increasingly favoured for localized tumors, its oncological safety relative to RN remains a subject of ongoing debate. This systematic review and Meta-analysis aims to compare the clinical efficacy and safety of PN versus RN in patients with RCC. Methods: Electronic databases were searched for studies comparing RN and PN for RCC from inception till January 2026. Inclusion criteria included randomized and non-randomized controlled trials, retrospective and observational studies/cohorts. Meta-analysis was conducted using R software (V.4.5.2) with the "META" package. Binary outcomes were pooled using MH Random Effects model with Odds Ratio (OR) and continuous outcomes using mean differences (MD) with the inverse variance random-effects model. Heterogeneity was assessed with τ² and I² statistics, sensitivity analysis was performed using a leave-one-out approach, and publication bias was evaluated using funnel plots and Egger’s regression test. Results: This systematic review and meta-analysis included 45 studies with a population size of 79,417 (PN = 28,192; RN = 51,225). PN was associated with a statistically significant decrease in all-cause mortality (OR = 0.63; 95% CI: [0.41, 0.96]; p = < 0.0001), and cancer-specific mortality (OR = 0.45; 95% CI: [0.25, 0.81]; p = < 0.0001), and demonstrated improved overall survival (OR = 1.46; 95% CI: [1.14, 1.88]; p = < 0.0001). However, RN was associated with improvement in postoperative complications (OR = 1.29; 95% CI: [1.08, 1.54]; p = < 0.0001), incidence of urinary fistula (OR = 7.02; 95% CI: [3.20, 15.39]; p = 0.538) and reduced operation time (MD = 14.15; 95% CI: [4.50, 23.81]; p = < 0.0001). Outcomes including eGFR, blood loss, length of stay, and cardiovascular complications, were comparable between the two surgical approaches. Conclusions: PN demonstrates superior overall and cancer-specific survival compared to RN, significantly reducing mortality risk for RCC patients. However, these benefits come at the cost of increased operative times and a higher risk of postoperative complications, specifically urinary fistulas. Surgeons should balance the long-term survival benefits of PN against the reduced perioperative risks associated with RN.

Nonlinear relationship between circulating homocysteine and parathyroid hormone among patients with osteoporotic fractures: A cross-sectional study based on a hospital cohort

Osteoporotic fractures are frequent among older adults and are often accompanied by disturbances in endocrine and metabolic regulation. Parathyroid hormone (PTH) serves as a key regulator of bone remodeling, while elevated plasma homocysteine—an established marker of vascular injury and skeletal fragility—has been implicated in osteoporosis. Yet, the interaction between these two biochemical factors remains insufficiently characterized. This study investigated the link between circulating homocysteine and serum PTH concentrations among 2,190 hospitalized patients with osteoporotic fractures, emphasizing nonlinear associations and modifying influences. Fasting blood samples obtained at admission were used to measure plasma homocysteine and intact PTH levels. Multiple regression and spline-based modeling techniques were employed to estimate independent relationships and potential threshold effects, with subgroup analyses considering hypertension, renal function, and magnesium status. Homocysteine demonstrated an independent, positive correlation with PTH (P < 0.001). A nonlinear dose–response relationship was observed, wherein PTH increased progressively with homocysteine up to about 18 μmol/L, followed by a sharper rise beyond this point, suggesting a statistically significant threshold. The relationship was more evident among patients with hypertension, impaired renal function, or low magnesium levels, indicating that vascular strain, renal stress, and mineral imbalance may heighten PTH activity. Elevated homocysteine levels may be associated with higher PTH concentrations in individuals with osteoporosis. However, the clinical relevance of homocysteine in metabolic assessment and risk stratification requires further validation.

Leucaena leucocephala leaves extract mediated green synthesis of TiO2 nanoparticles, characterization and photocatalytic application

Beyond Earth: Resilience of Quasi‐2D Perovskite Solar Cells in Space (Adv. Mater. 31/2026)

High‐Efficiency Targeted Mitochondrial Transfer and AUTAC4‐Enhanced Dual Renewal Strategy for Rheumatoid Arthritis

ABSTRACT Rheumatoid arthritis (RA) is a chronic autoimmune disease with limited therapeutic effectiveness of conventional biomaterials, which often lack targeted accuracy, delivery efficiency, and biocompatibility. Here, we present a biomimetically engineered carrier material using mitochondria as “living materials” to restore cell homeostasis in RA. The dual action carrier consists of a folic acid‐modified macrophage membrane targeting activated M1 macrophages in RA joints, and it enables in situ mitochondrial transfer with more than twofold increase of delivery efficiency, which is a critical limitation of current approaches. By facilitating precise intracellular transfer of healthy mitochondria, and incorporating autophagy targeting chimera 4 (AUTAC4) in order to selectively destroy dysfunctional mitochondria, this design achieves complete mitochondrial renewal, increasing energy metabolism and homeostasis. In an RA model, the Dual‐Action Mitochondrial Renewal Therapy (DAMRT) showed significant therapeutic potential. It could be used as a novel platform for treatment for RA and other mitochondrial dysfunction.

Micromechanical characteristics of compression densification of spent mushroom residue biomass based on DEM

A direct fluorescence assay for quantitative analysis of ornithine decarboxylase activity and inhibitor screening

Clinical impact of baseline and peri-treatment concomitant medication exposure and polypharmacy on outcomes of immune checkpoint inhibitor therapy: A real-world pan-tumor analysis.

11011 Background: Immune checkpoint inhibitors (ICIs) have transformed cancer therapy, yet real-world patients frequently receive multiple concomitant medications that may influence treatment outcomes. These medications may alter the gut microbiome and modulate host immune function, impacting ICI efficacy and survival. The clinical effects of baseline and peri-treatment medication exposure on ICI outcomes remain incompletely characterized. Methods: We conducted a retrospective study of ICI-treated solid tumors at a tertiary academic center. Concomitant medications were captured at ICI initiation (baseline chronic exposure) and during a peri-treatment window (acute exposure, 30 days before or after ICI initiation). Drug classes included antibiotics, proton pump inhibitors, corticosteroids, NSAIDs, opioids, anticoagulants, antiplatelets, statins, cardio-metabolic and psychotropic agents, and supplements. Polypharmacy was defined as ≥5 baseline medications. Outcomes included overall survival (OS), progression-free survival (PFS), immune-related adverse events (irAEs), and objective response using RECIST 1.1. Multivariable Cox proportional hazards models were used to adjust for confounding. Results: The cohort included 772 patients (60.1% male) with a median BMI of 25.8 kg/m² (IQR 23.0–29.1), and 66.2% had stage IV disease. On multivariable analysis for objective response (RECIST), baseline polypharmacy (≥5 medications) (OR 0.65, 95% CI 0.43–0.96; p=0.033), acute antibiotic exposure within 30 days of ICI initiation (OR 0.63, 95% CI 0.39–0.99; p=0.049), and metastatic disease (stage IV) (OR 0.51, 95% CI 0.26–0.98; p=0.045) independently predicted inferior response, while a higher number of immunotherapy cycles was associated with improved response (OR 1.06 per cycle, 95% CI 1.05–1.08; p<0.001). Acute antibiotic exposure within 30 days before ICI initiation was also independently associated with worse progression-free survival (HR 1.70, 95% CI 1.05–2.76; p=0.031). On multivariable Cox regression for overall survival, acute opioid exposure (HR 1.51, 95% CI 1.06–2.15; p=0.023), baseline polypharmacy (HR 1.78, 95% CI 1.19–2.65; p=0.005), and advanced disease (HR 2.78, 95% CI 1.73–4.47; p<0.001) were independently associated with worse survival, whereas a higher number of immunotherapy cycles (HR 0.96 per cycle, 95% CI 0.94–0.98; p<0.001) and prior surgery of the primary tumor (HR 0.51, 95% CI 0.33–0.79; p=0.002) were associated with improved survival. No baseline polypharmacy or specific medication exposure independently predicted the occurrence of irAEs. Conclusions: Concomitant medication exposure significantly impacts ICI outcomes, potentially through modulation of host immunity and the gut microbiome, supporting routine medication review at ICI initiation.

Survival outcomes of high and intermediate-risk hormone receptor (HR)-positive human epidermal growth receptor-2 (HER2)-negative early breast cancer patients (EBC) who did and did not receive (neo)adjuvant chemotherapy.

e12530 Background: Patients with HR-positive HER2-negative EBC have protracted risk of disease relapse despite 5 years of adjuvant endocrine therapy. Combination of endocrine therapy with CDK4/6 inhibitors have shown to improve clinical outcomes in these settings according to MonarchE and NATALEE studies. Prior to the use of adjuvant CDK4/6 inhibitors, (neo)adjuvant chemotherapy in addition to adjuvant endocrine were considered for these patients, and such decision was mainly based on cancer stage. The objective of this study was to determine the survival outcomes of HR-positive HER2-negative EBC cancer patients who did and did not receive (neo)adjuvant chemotherapy. Methods: Patient data from the Hong Kong Breast Cancer Registry was retrieved for this retrospective study. Ethics approval was obtained from regional committee of 11 public and 4 private hospitals/clinics in Hong Kong. Patients who underwent definitive surgery for EBC between January 2006 and December 2011 were categorized into 3 groups: Group 1 consisted of patients who satisfied both MonarchE and NATALEE criteria; Group 2 were those eligible for NATALEE criteria but not MonarchE and Group 3 were remaining patients who mainly had stage 1 cancers. Patients background characteristics were collected. DDFS, iDFS and OS were evaluated using the Kaplan–Meier method. Results: A total of 3078 HR-positive HER2-negative EBC were included: Group 1, n = 912; Group 2, n = 600; Group 3, n = 1566. The number (%) of patients who received (neo)adjuvant chemotherapy were 480 (30.7%) in Group 1, 539 (90.0%) in Group 2 and 856 (94.0%) in Group 3. For patients within Group 2 and Group 3, 10-year iDFS, DDFS and OS were significantly better for those who had received (neo)adjuvant chemotherapy. Specifically, for Group 3, the 10-year iDFS, DDFS and OS among those who did and did not have (neo)adjuvant chemotherapy were 70.5% vs 54.0% (p = 0.028), 75.4% vs 69.5% (p = 0.007) and 80.8% vs 62.5% (p = 0.001). Survival outcomes among Group 1 patients did not differ between those who had or had not received (neo)adjuvant chemotherapy. Conclusions: Our findings confirmed that the role of (neo)adjuvant chemotherapy is significantly associated with improvement in patients’ outcomes in intermediate and high-risk HR-positive HER2-negative EBC patients. High-risk patients who did not receive (neo)adjuvant chemotherapy had dismal outcomes. Acknowledgement: we thank Drs Joanne Chiu and Miranda Chan for their support.

Effects of SFRT on quality of life in patients with bulky advanced cancer: A retrospective single-center cohort analysis.

e24058 Background: Bulky tumors remain a major therapeutic challenge in modern oncology. Spatially Fractionated Radiation Therapy (SFRT) is an emerging technique shows great potential. We therefore conducted a retrospective review of two-year data from our center, aimed to evaluate the impact of SFRT on quality of life and survival outcomes in patients with bulky tumors. Methods: Patients who received SFRT between February 2023 and May 2025 were analyzed in this retrospective study. Collected data comprised demographic characteristics, as well as follow-up information on symptomatic response, best overall response, survival duration, and treatment-related adverse events. Results: A total of 60 patients were included, with a median age of 66 years (26–88). Among them, 38 (63.3%) were male and 22 (36.7%) were female. Patients with 14 primary tumor types were included, and irradiation targeted 10 distinct anatomical sites. The most common clinical symptoms were pain (41, 68.3%), dyspnea (13, 21.7%), tenesmus and hematochezia (2 each, 3.3%), intestinal obstruction (1, 1.7%), and portal hypertension (1, 1.7%). The median pretreatment tumor diameter was 83.55 mm (36.43–2240.50), and median tumor volume was 288.20 cm³ (33.04–2209.73). 18 patients were treated with lattice radiotherapy (LRT) and 42 with Stereotactic Centralized Ablative Radiation Therapy (SCART) technique. The median BED was 88.8 Gy ( 40-195). All patients underwent SFRT with daily cone-beam CT (CBCT) guidance. Within 2–4 weeks after radiotherapy, 91.7% (55/60) of patients showed symptomatic relief, including pain improvement in 92.7% (38/41), resolution of dyspnea in 100% (13/13), cessation of bleeding in 100% (2/2), relief of tenesmus in 100% (2/2), and no improvement in intestinal obstruction or portal hypertension (0/1 each). Follow-up imaging was available for 51 patients, showing tumor regression in 45 cases (6 showed increased volume after treatment). According to RECIST 1.1 criteria, 17 patients achieved partial response (PR), 33 had stable disease (SD), and 1 experienced progressive disease (PD). The overall response rate (ORR) was 33.3%, and the disease control rate (DCR) was 98.0%. A BED ≥ 88 Gy was identified as an independent predictor of treatment response in both univariate and multivariate analyses (P = 0.024). With a median follow-up of 15.8 months, 36 of 60 patients had died by September 17, 2025. The median overall survival (OS) was 7.6 months (95% CI: 4.3-10.9). All adverse events were mild to moderate in severity. No grade 3 or higher toxicities were observed. Conclusions: SFRT allows for localized high-dose irradiation in large-volume tumors, demonstrating favorable tolerability, promising symptom relief (correlating with quality of life improvement), and encouraging local control. A strategy of safe BED escalation could enhance palliative efficacy.

Three-decade trends in bladder cancer burden in the Middle East and North Africa: Divergent patterns between regional and Saudi Arabian populations.

e16616 Background: Urinary bladder cancer (UBC) represents a significant public health challenge in the Middle East and North Africa (MENA) region, with age-standardized incidence rates substantially exceeding global averages. Despite this burden, regional epidemiological data remain limited, and the evolving trends in Saudi Arabia have not been systematically compared to broader MENA patterns. This study leverages Global Burden of Disease (GBD) data to characterize longitudinal trends in UBC mortality, incidence, and prevalence across the MENA region and Saudi Arabia from 1990 to 2023. Methods: Age-standardized rates for UBC mortality, incidence, and prevalence were extracted from the GBD 2023 database for the MENA region and Saudi Arabia, stratified by sex. Trends were analyzed using joinpoint regression to calculate annual percentage changes (APC) and average annual percentage changes (AAPC). Pairwise comparisons between MENA and Saudi Arabia trends were performed to assess statistical significance. Results: In MENA, mortality declined from 4.01 to 3.55 per 100,000 (AAPC: -0.40%), incidence increased from 6.85 to 7.44 (AAPC: 0.16%), and prevalence rose from 29.71 to 37.70 (AAPC: 0.71%). In Saudi Arabia, mortality decreased from 4.96 to 3.73 (AAPC: -0.86%), with divergent sex trends: male mortality declined (AAPC: -1.20%) while female mortality increased (AAPC: 0.50%). Saudi incidence remained stable (AAPC: -0.12%) while prevalence increased (AAPC: 0.48%). Males had 3-4 times higher rates than females. MENA mortality declined until 2010 then reversed; incidence increased after 2009. Trends differed significantly between regions (p < 0.001). By 2050, MENA mortality is projected to decline 5.0% while Saudi Arabia mortality increases 7.65%. Conclusions: While UBC mortality has declined modestly in both MENA and Saudi Arabia over the past three decades, rising incidence and prevalence rates underscore a growing disease burden. The marked divergence in temporal trends between Saudi Arabia and the broader MENA region, along with notable sex-specific disparities—particularly the increasing female mortality and incidence in Saudi Arabia—highlight the need for tailored public health interventions. These findings emphasize the importance of strengthening regional cancer surveillance, enhancing early detection programs, and implementing targeted tobacco control and metabolic health initiatives to address the evolving epidemiological landscape of bladder cancer in the region.

Sequential bTAE-HAIC combined with lenvatinib and sintilimab for infiltrative hepatocellular carcinoma: A single-center perspective.

4131 Background: Infiltrative hepatocellular carcinoma (HCC) is defined as a tumor with ill-defined border, no evidence of convex margination, and no typical enhancement pattern by imaging. This subtype of HCC is usually associated with poor prognosis and is frequently diagnosed at advanced stage. Blank-microsphere transcatheter arterial embolization (bTAE) and hepatic arterial infusion chemotherapy (HAIC) of oxaliplatin, 5-fluorouracil and leucovorin are effective and safe for HCC. This prospective study is to evaluate the safety and efficacy of bTAE-HAIC combined with Lenvatinib and Sintilimab for intermediate-advanced infiltrative HCC. Methods: This prospective phase II trial is a single-center, single-arm study designed to assess the efficacy and safety of TAE and HAIC, combined with sequential Lenvatinib and Sintilimab on infiltrative HCC. Patients with primary infiltrative HCC treated at our center between October 2023 and December 2024. The primary endpoint was objective response rate (ORR) according to mRECIST criteria. Secondary endpoints included overall survival (OS), progression-free survival (PFS), and treatment-related adverse events (AEs). Results: A total of 30 patients who met the criteria were enrolled in this study and received at least 2 courses of TAE-HAIC combined with Lenvatinib and Sintilimab treatment. The median follow-up time was 7.1 months, and 10 patients died during the period. 5 patients (16.7%) achieved complete response (CR) after sequent ablation or radiotherapy, 20 patients (66.7) were evaluated as PR. The overall ORR was 83.3%. The median OS was 16.0 months (95%CI: 5.6-26.5 months). The 3-month, 6-month and 12-month rates of OS were 93.3%, 83.0% and 60.2%. The median PFS was 9.5 months (95%CI: 4.9-14.1 months), and the PFS rates at 3-month, 6-months and 12-month were 95.7%, 68.1% and 20.4%, respectively. The median liver-specific PFS was 11.7 months (95%CI: 7.7-15.7 months). The overall incidence of treatment related AEs was 76.7% according to the CTCAE 5.0 criteria. The incidence of Grade 1-2 AEs was 73.3%, and only one patient experienced Grade 3 AE. 10 patients (33.3%) received dose reduction, 9 patients (30.0%) discontinued drug therapy. Conclusions: The preliminary results of current study indicates that the combination regimen is safe and effective, and the AEs and complications are controllable and tolerable. This study provides a reference for further phase III clinical research, and a novel treatment option for the infiltrative HCC. Clinical trial information: NCT06070636 . Baseline characteristics. Variable All Patients (N = 30) Male 26 (86.7%) BCLC Stage C 28 (93.3%) Extrahepatic Metastasis 13 (43.3%) Tumor Number- Multiple（≥3） 16 (53.3%) Largest Diameter (cm) >10 cm 24 (80.0%) Largest Diameter (cm) >15 cm 8 (26.7%) Macrovascular Invasion 25 (83.3%) Portal Hypertension 12 (40.0%)

Real-world effectiveness of pembrolizumab among patients with TMB-H advanced solid tumors.

e23360 Background: Pembrolizumab has been conditionally approved for previously treated solid tumor patients with high tumor mutational burden (TMB-H ≥10 mut/Mb, as determined by an FDA-approved test) who lack satisfactory alternative treatment options. Real-world evidence of pembrolizumab effectiveness in the TMB-H population is limited. Methods: We conducted a retrospective study to evaluate the effectiveness of pembrolizumab in TMB-H patients using the US-based deidentified Flatiron Health-Foundation Medicine Pan-Tumor Clinico-Genomic Database. All adult patients with advanced/metastatic solid tumors (excluding melanoma and NSCLC), who were TMB-H (based on FoundationOne CDx), non-microsatellite instability high and received pembrolizumab monotherapy in 2L/2L+ were included. Real-world response (rwR) assessment was abstracted through manual medical chart review, based on clinician documentation following radiographic imaging. The prespecified primary analysis included patients with ≥1 rwR assessment. Real-world response rate (rwRR) was the proportion of patients with a complete response (rwCR) or partial response (rwPR). Results: The primary population included 306 patients across 25 tumor types, most commonly bladder, breast, and gastric. Median age was 68 years; 51.8% female; 80.3% non- Hispanic/Latino; 105 patients had a response, yielding a rwRR 34.3%. The rwRR was 25.2% (95% CI:17.6–34.2) for TMB 10-13 mut/Mb (n=29) and 39.8% (95% CI: 32.8–47.1) for TMB ≥13 mut/Mb (n=76). Median duration of response was 14.5 months; median time to response was 2.8 months. Sensitivity analysis with all patients (counting those without rwR assessment as non-responders) is shown in the table. Conclusions: This observational study reflects real-world practice, including non-standardized visit schedules and chart-abstracted responses without imaging re-read, which may limit comparability of rwRR to the ORR by RECIST in clinical trials. Nevertheless, these findings were consistent with the trial that supported the approval and provided complementary evidence on pembrolizumab effectiveness in TMB-H cancers treated in routine care. Real-world responses among the study populations. Primary population (n=306) All population (n=407) n (%) 95% CI n (%) 95% CI Response (CR or PR) 105 (34.3) (29.0, 39.9) 105 (25.8) (21.6, 30.3) Complete Response (CR) 30 (9.8) (6.7, 13.7) 30 (7.4) (5.0, 10.4) Partial Response (PR) 75 (24.5) (19.8, 29.7) 75 (18.4) (14.8, 22.5) Non-Response 201 (65.7) (60.1, 71.0) 302 (74.2) (69.7, 78.4) Stable Disease (SD) 65 (21.2) (16.8, 26.3) 65 (16.0) (12.5, 19.9) Progressive Disease (PD) 126 (41.2) (35.6, 46.9) 126 (31.0) (26.5, 35.7) Pseudo-Progression 7 (2.3) (0.9, 4.7) 7 (1.7) (0.7, 3.5) Indeterminate response 1 (0.3) (0.0, 1.8) 1 (0.2) (0.0, 1.4) Not documented 2 (0.7) (0.1, 2.3) 2 (0.5) (0.1, 1.8) No response assessment - - 101 (24.8) (20.7, 29.3)

Admission acuity and inpatient outcomes across major gastrointestinal malignancies in the United States, 2018–2022.

11097 Background: Hospitalization for gastrointestinal (GI) malignancies frequently occurs in the setting of acute decompensation or unplanned admission. National data quantifying the association between admission acuity and inpatient outcomes across major GI cancer subtypes remain limited. Methods: A serial cross-sectional analysis was conducted using the 2018–2022 Healthcare Cost and Utilization Project National Inpatient Sample. Adult hospitalizations with a principal diagnosis of esophageal (C15), gastric (C16), colorectal (C18–C20), liver or intrahepatic bile duct (C22), or pancreatic cancer (C25) were identified. Admission acuity was classified as elective or non-elective. Outcomes included in-hospital mortality (primary), length of stay (LOS), and hospitalization cost estimated using cost-to-charge ratios. National estimates accounted for survey weighting, clustering, and stratification. Survey-weighted multivariable logistic regression evaluated factors independently associated with in-hospital mortality, adjusting for admission type, cancer subtype, demographics, payer, neighborhood income quartile, hospital characteristics, and calendar year. Results: From 2018–2022, 214,740 unweighted GI cancer hospitalizations represented an estimated 1.07 million hospitalizations nationally. Colorectal cancer accounted for 56.7% of admissions, followed by pancreatic (17.6%), liver or intrahepatic bile duct (10.5%), gastric (8.6%), and esophageal cancer (6.6%). Overall, 52.9% of hospitalizations were non-elective. In-hospital mortality was higher for non-elective than elective admissions (5.92% vs 1.89%), with longer LOS (7.61 vs 6.15 days) and lower mean hospitalization costs ($25,644 vs $32,166). In adjusted analyses, elective admission was associated with substantially lower odds of in-hospital mortality compared with non-elective admission (adjusted odds ratio [aOR] 0.395; 95% CI 0.328–0.476). Relative to esophageal cancer, adjusted mortality odds were lower for colorectal, gastric, and pancreatic cancers, while no significant difference was observed for liver or intrahepatic bile duct cancers. Conclusions: Across major gastrointestinal malignancies, non-elective admission was associated with substantially higher inpatient mortality and longer length of stay independent of cancer subtype and sociodemographic factors. These nationally representative findings provide benchmarking data for admission acuity as a marker of inpatient risk in gastrointestinal oncology.

Enabling federated discovery leveraging agentic AI systems: The Cancer AI Alliance and DataVoyager.

e20516 Background: Artificial intelligence (AI) has the potential to transform oncology, but robust AI models require diverse, multi-institutional data that preserve patient privacy. Federating data across institutions is a promising solution, yet historically difficult to implement at scale. The Cancer AI Alliance (CAIA) was created to enable secure, large-scale federation of patient data across leading cancer centers. Moreover, to address analytic barriers faced by investigators, we developed a federated implementation of DataVoyager (DV), an interactive AI agent for data analysis by Ai2. Here we present early results leveraging agentic AI to evaluate real-world outcomes in non-small cell lung cancer (NSCLC). Methods: A researcher entered a natural language query into DV: “Describe the overall survival (OS) for patients who receive osimertinib or alectinib for lung cancer.” A large language model (LLM; a secure, open-weight gpt-oss-120B model) translated the query into a statistical analysis plan, and a second LLM autonomously generated Python code for implementation. Through the federated framework, code was securely distributed to four edge-nodes where it was executed locally on de identified EHR data. CAIA site edge-nodes returned model summaries to DV, which synthesized aggregated outputs and produced summary statistics, metadata, and visualizations. All plans, code, assumptions, and discussions were archived for independent validation and reproducibility. Results: CAIA enabled federated access to data from >1 million patients across four institutions. DV harmonized cohort identification, estimated demographics, and conducted federated survival analyses in CAIA’s combined stage I–IV NSCLC cohort (Table). DV produced an analysis plan and outputs (descriptive tables, Kaplan-Meier plots, plain-language summaries, code, and logs) in 1 min 24 s, within privacy safeguards. The results appeared broadly consistent with published clinical trials and real-world observations, supporting the validity of the federated workflow. The generated code was verified as correct by human expert review. Conclusions: Combining CAIA’s federated infrastructure with DV’s agentic AI represents a potential new paradigm in oncology discovery. CAIA brings together leading cancer centers to enable secure analytics without patient-level data exchange. In parallel, DV provides a novel capability to interrogate large, distributed datasets without coding expertise or costly infrastructure, delivering complete analytic workflows in minutes. Together, these advances enable privacy-preserving, reproducible, and scalable real-world evidence generation, as demonstrated in NSCLC. Characteristic Alectinib (N = 603) Osimertinib (N = 2823) Age <65 years 68% 46% Sex Male 43% 30% Female 57% 70% Race White 73% 64% Non-white 20% 30% Missing 7% 6% OS 2-year 78% 63% 5-year 69% 52%

Negative hyperselection as a predictive biomarker of benefit from first-line doublets + anti-EGFRs/bevacizumab (bev) in microsatellite stable (pMMR/MSS) <i>RAS</i> and <i>BRAF</i> wild-type (wt) and HER2− metastatic colorectal cancer (mCRC): An individual patient data pooled analysis of seven clinical trials (CTs).

3550 Background: Primary tumor sidedness is nowadays a major driver for the choice of biologics in the first line of pMMR/MSS RAS and BRAF wt mCRC patients (pts). It has been suggested that it might be replaced by a broader molecular profiling, but the evidence provided by individual trials is uncertain for the adoption of different panels and underrepresentation of gene-altered cases. We recently showed that HER2+ status does not predict resistance to first-line doublets + anti-EGFRs. Methods: We collected individual patient data from 7 CTs in first-line mCRC: TRIBE2, TRIPLETE, VALENTINO, CAPRI-2, FIRE-3, PARADIGM and CALGB/SWOG80405. pMMR/MSS RAS and BRAF wt and HER2− cases treated with doublets + anti-EGFRs/bev with available NGS data were included. Hyperselected tumors were those without MET+ status, ALK/ROS1/NTRKs/RET fusions, and HER2/PIK3CAexon20/PTEN/AKT1 mutations. Propensity score adjustment was used to assess clinical outcomes (aHR and aOR) according to first-line doublets + anti-EGFRs/bev. Results: Out of 1198 included pts, 990 (83%) had left-sided and 208 (17%) right-sided tumors. Anti-EGFR based regimens were associated with numerically longer OS (aHR: 0.86) in left-sided tumors, but not in right-sided ones (aHR: 0.98), with an insignificant p for interaction [int] = 0.60). Hyperselected and gene-altered cases were 1077 (91%) and 121 (9%), respectively. The anti-EGFR benefit in OS was magnified in hyperselected pts (aHR: 0.83), with an aHR of 1.16 in gene-altered ones (p int =0.11, adjusted also for tumor sidedness). As reported in the table, there was no interaction between the effect of biologics and primary tumor location either in the hyperselected (p int =0.86) or in the gene-altered group (p int =0.41). Conclusions: Hyperselection outperforms tumor sidedness as a predictive marker of benefit from doublets + anti-EGFR/bev in pMMR/MSS RAS/BRAF wt and HER2− mCRC. The OS benefit from anti-EGFR-based regimens is restricted to pts with hyperselected tumors, independently of primary tumor location. Therefore, in hyperselected mCRC, doublets + anti-EGFRs remain the standard for left-sided pts and may deserve consideration in right-sided ones. Hyperselected Gene-altered Left Right P int Left Right P int Anti-EGFRs N=537 Bev N=365 Anti-EGFRs N=87 Bev N=88 Anti-EGFRs N=57 Bev N=31 Anti-EGFRs N=19 Bev N=14 ORR (%) 81 67 72 65 74 52 58 57 aOR [95% CI] 2.29 [1.67 – 3.14] 1.44 [0.75 – 2.77] 0.24 3.48 [1.29 – 9.97] 1.02 [0.25 – 4.20] 0.23 mPFS* 13.2 12.9 11.1 10.2 11.2 11.3 8.0 8.8 aHR [95% CI] 1.02 [0.88 – 1.19] 1.01 [0.73 – 1.39] 0.73 1.06 [0.63 – 1.77] 1.23 [0.57 – 2.66] 0.75 mOS* 38.8 35.5 34.6 29.2 30.2 27.4 20.2 32.1 aHR [95% CI] 0.84 [0.71 – 0.98] 0.85 [0.61 – 1.17] 0.86 1.05 [0.63 – 1.75] 1.46 [0.66 – 3.22] 0.41 *Months.

Real-world outcomes of prolonged versus shorter duration pembrolizumab following chemoradiation in non–small cell lung cancer: A propensity-matched analysis.

e20636 Background: The optimal duration of pembrolizumab following chemoradiation in non–small cell lung cancer (NSCLC) remains uncertain in real-world practice. Current guidelines do not provide clear recommendations regarding treatment duration beyond clinical trial protocols, resulting in substantial variability in real-world practice. While extended immunotherapy exposure is commonly pursued, its association with improved survival and healthcare utilization outside clinical trials is incompletely characterized. We compared outcomes between patients receiving prolonged versus shorter durations of pembrolizumab using a large real-world database. Methods: We conducted a retrospective cohort study using the TriNetX U.S. large de-identified EMR database from 72 U.S. healthcare organizations, including adult patients with NSCLC who received chemoradiation followed by pembrolizumab. Patients were categorized into: Full-duration pembrolizumab (continued ≥12 months), and Short-duration pembrolizumab (discontinued earlier). Propensity score matching (1:1) was performed for age, sex, race, smoking history, and comorbidities. Primary outcome was overall mortality. Secondary outcomes included hospitalizations, critical care utilization, emergency department visits, palliative care use, and prolonged services. Kaplan–Meier and Cox proportional hazards models were applied. Results: After matching, 450 patients were included in each cohort with well-balanced baseline characteristics. Mortality: Full-duration: 37.1% Short-duration: 46.4% Risk difference: −9.4% (p = 0.005) Hazard ratio (HR): 0.67 (95% CI, 0.55–0.83) Median overall survival: Full-duration: 1505 days Short-duration: 791 days Log-rank p &lt; 0.001 Hospitalizations, critical care, emergency visits, and palliative care were similar between groups. Prolonged services occurred less frequently in the full-duration group (HR 0.61, 95% CI 0.38–0.97). Conclusions: In this large real-world cohort, the longer duration of pembrolizumab following chemoradiation was associated with significantly improved overall survival without increased healthcare utilization. These findings support sustained immunotherapy exposure when clinically feasible and highlight the value of real-world evidence in informing treatment duration decisions. Clinical implications of this study: supports continuation of pembrolizumab beyond early discontinuation when tolerated, demonstration survival benefit without excess healthcare burden.