Browse Articles
Discover research articles across all indexed journals
FUEL-IT Lung: Fasting to unleash and enhance lung cancer immunotherapy—A VA Lung Precision Oncology Program trial.
TPS8666 Background: Single-agent PD-(L)1 inhibitors are standard first-line therapy for advanced non-small cell lung cancer (NSCLC) with PD-L1 expression ≥50%, yielding objective response rates of approximately 45% and median overall survival of 20-26 months. However, immune-related adverse events (irAEs) occur in up to 20% of patients. Fasting-mimicking diets (FMD) induce metabolic reprogramming by reducing insulin-like growth factor 1 (IGF-1) and glucose while increasing IGF-binding protein 1. Preclinical data demonstrate that FMD enhances anti-tumor immunity by increasing T-cell mediated tumor cytotoxicity, reducing myeloid-derived suppressor cells, and suppressing T-regulatory cell function. Recent murine models show that FMD combined with anti-PD-L1 therapy prevents or reverses immune-mediated myocardial infiltration, reduces systemic inflammation, and is more effective than anti-PD-L1 alone in delaying tumor growth while reshaping the tumor microenvironment. Our prior feasibility study (IUSCCC-0662) in 10 advanced lung cancer patients demonstrated 80% compliance with FMD, with minimal toxicities. These findings suggest FMD may enhance checkpoint inhibitor efficacy while potentially reducing irAEs. Methods: This is an open-label, randomized pilot study evaluating the feasibility and safety of FMD combined with pembrolizumab in Veterans with newly diagnosed stage IV NSCLC and PD-L1 expression ≥50%. Eligible patients must have ECOG performance status 0-2 and adequate organ function. Patients receive pembrolizumab 200 mg IV every 3 weeks. The study employs a partial crossover design: Arm 1 receives regular diet with pembrolizumab for cycles 1-3, then crosses over to FMD with pembrolizumab for cycles 4-6; Arm 2 receives FMD with pembrolizumab for cycles 1-3, followed by regular diet thereafter. FMD consists of a 4-day cycle with calorie restriction (fats:carbs:protein ratio of 50:40:10) administered starting on the day of pembrolizumab infusion, followed by transitional diet on day 5, for 3 consecutive cycles. Primary endpoints include feasibility (proportion completing 3 FMD cycles, target ≥70%) and safety. Secondary endpoints include immune-mediated toxicity rates, objective response rate, disease control rate, and 12-month progression-free survival. Exploratory objectives assess FMD impact on metabolic markers (glucose, ketones, IGF-1, IGFBP-1), immune cell populations, body composition via CT imaging, physical function (Short Physical Performance Battery), and quality of life (FACT-L, EORTC QLQ-C30). Enrollment of 33 patients per arm is planned for total of 66 patients. Accrual began in October 2025. Clinical trial information: NCT06671613 , Funded by VA ORD.
Optimizing post-prostatectomy management: A machine learning tool to identify low-risk patients and minimize unnecessary burden in localized prostate cancer.
e17121 Background: Biochemical recurrence (BCR) is a key early indicator of disease progression in localized prostate cancer (LPC). While uniform, intensive follow-up is standard, it leads to significant resource burden and patient anxiety, as patients particularly at minimal risk may not require it. It remains to be established whether machine learning models can provide efficient, data-driven decision-making solutions in this clinical context. Methods: This is a retrospective cohort study using structured electronic medical record data from a tertiary care center (West China Hospital). Eligible patients had clinical LPC and underwent RP between 2017-2024, with at least one year of follow-up. BCR was defined as PSA≥0.2 ng/mL with two consecutive increases post-RP. Predictors encompassed biopsy, pathological Gleason scores, prostate size, pT stage, surgical margin status, pre- and post-RP PSA levels, and inflammatory ratios. Random Forest (RF) and XGBoost models were developed with hyperparameter tuning via 10-fold cross-validation. Model performance was evaluated using 500 bootstrap resamples to obtain optimism-corrected Area Under Curve (AUC), Sensitivity, Specificity, PPV, NPV, and Calibration-In-The-Large (CITL). Clinical utility was assessed via Decision Curve Analysis (DCA) to guide follow-up decisions. Results: Among 763 male patients (mean age 67.3 years) in the cohort, 78 (10.2%) patients experienced BCR within 1 year (median 6 months). After bootstrap correction, the RF model achieved superior discrimination (optimism-corrected AUC: RF 0.90 vs. XGboost 0.82, P<0.05) and better calibration (CITL: RF -0.04 vs. XGboost 0.23, P<0.05). Both models demonstrated a consistently high NPV (0.97), supporting safe exclusion of low-risk patients from intensive surveillance. DCA revealed that using the RF model provided a superior net benefit across the risk thresholds of 0-54.2% compared to XGboost and default strategies of "follow-all" or "follow-none." A model-guided management strategy could reduce approximately 70% of health utilization cost per low-risk patient and reallocate about 19 clinician workdays to higher-value care annually. Conclusions: Using rigorous internal validation, we developed a highly discriminative and well-calibrated RF model capable of accurately identifying patients at minimal short-term BCR risk. The model's high NPV and demonstrated clinical utility suggest strong potential to safely de-escalate post-RP surveillance. Prospective multi-center validation is warranted to confirm its generalizability and its impact on clinical workflow and resource optimization. Optimism-corrected performance metrics. Algorithm AUC Sensitivity* Specificity* PPV* NPV* CITL RF 0.90 0.82 0.79 0.19 0.97 -0.04 XGboost 0.82 0.81 0.66 0.15 0.97 0.23 *With a Youden threshold.
Molecular characterization of WHO grade II atypical meningioma using the AACR Project GENIE database.
2088 Background: Atypical meningioma (AM) is a WHO grade II neoplasm arising from meningeal arachnoid cap cells and is distinguished from WHO grade I meningiomas by parenchymal invasion and rapid proliferation. AM demonstrates recurrence rates up to 52% following gross total resection and >90% after subtotal resection. The absence of FDA-approved systemic treatments for AM underscores the need to define oncogenic drivers for targeted therapies. This study leverages the American Association for Cancer Research (AACR) Project Genomics Evidence Neoplasia Information Exchange (GENIE) database to characterize the genomic landscape of AM and identify key genetic drivers for therapeutic targets. Methods: The AACR GENIE database was accessed from cBioPortal (v18.0-public) on December 12, 2025 to identify all AM patient samples. Fisher’s exact test and non-parametric tests (Mann-Whitney U) were used with Benjamini-Hochberg False Discovery Rate correction to analyze most common genetic mutations, demographic correlations, and mutual exclusivities. Unknown values were excluded from analysis. Results: The cohort identified 402 AM samples from 394 patients. The cohort was slightly female-predominant (52.8%) and largely adult (94.0%). Most patients were White (47.2%), followed by Asian (7.4%) and Black (3.6%). Most samples (n=307, 76.4%) were primary tumors and 48 (11.9%) were metastatic. NF2 was the most prevalent mutation (n=228; 56.7%), followed by TERT promoter mutations (n=52; 12.9%). Chromatin-modifying genes, including KMT2D (n=24; 6.0%), KMT2A (n=19; 4.7%), and KMT2C (n=16; 4.0%), were frequently altered. Additional recurrent mutations were observed in SPTA1 (n=21; 5.2%), MT-ND5 (n=20; 5.0%), FAT1 (n=20; 5.0%), PRKDC (n=18; 4.5%). Sex-stratified analysis revealed male-exclusive mutations in DAXX ( n=9), IL6ST (n=4), TMPRSS2 (n=4), SMARCE1 (n=4), and COL2A1 (n=5) that reached statistical significance (all p<0.05). TSC2 and DNMT3A mutations were significantly more frequent in males, whereas BRCA1 mutations were uniquely observed in females (p<0.05). NF2 mutations were enriched in Asian patients (p<0.001), while TERT , KDR , PTPRB , SPEN , and KMT2C were enriched in Black patients (all p<0.05). NF2 and TERT demonstrated significant mutual exclusivity (p<0.05). Co-occurrence was observed between ROS1 / ATM and KMT2D / TSC2 (p<0.05). Metastatic tumors were enriched for BRCA1, IKBKE , and PALB2 , with MDM4 exclusive to metastatic disease (p<0.001). Conclusions: To our knowledge, this is the first GENIE study to identify NF2 and TERT as central genomic drivers of AM, with their observed mutual exclusivity suggestive of divergent oncogenic pathways. Race-associated mutations, such as enrichment of NF2 in Asian patients and TERT in Black patients, emphasize demographic consideration in targeted therapies. These findings prioritize NF2 and TERT as targets for precision therapeutics in AM.
Final analysis of KOMET (NCT04924608), a phase 3 study of selumetinib in adults with NF1-PN.
3110 Background: In 2025 the EMA and FDA expanded the approval of selumetinib (SELU; ARRY-142886, AZD6244) to adults with neurofibromatosis type 1 (NF1) and symptomatic, inoperable plexiform neurofibromas (PN). We report the final exploratory analysis of SELU efficacy and safety from the Phase 3, randomized, double-blind, placebo (PBO)-controlled KOMET trial. Methods: Adults (≥18 yrs) with NF1-PN were randomized 1:1 to 28-day cycles of oral SELU 25 mg/m 2 BID or PBO with crossover to SELU at the end of Cycle (C) 12 or earlier if radiological progression was confirmed by independent central review (ICR). The single-arm objective response rate (ORR) by use of volumetric MRI analysis per ICR REiNS and safety were exploratory endpoints at final DCO (last participant (pt) completed C24; 17 Mar 2025)). Pharmacokinetics (PK) data were collected at steady state C1, Day (D) 8. A planned sample of 73 pts per arm with a 2-sided 5% alpha Fisher’s exact test had >99% power to detect the difference between a SELU ORR of 20% and PBO ORR of 0%. Results: Overall, 145 pts (SELU: 71; PBO: 74) were randomized; 66 pts in the PBO group crossed over to SELU treatment for the open-label period (SELU period). In the SELU period, 137 pts were treated; 96 pts (66.2%) (SELU: 45, PBO/SELU: 51) completed the study and were judged by the investigators at DCO to have clinical benefit at study completion and, consequently, continued receiving SELU during the post-trial access program. Median actual exposure to SELU during the SELU period was 554 days and maximum exposure to SELU treatment was 1156 days. The median relative dose intensity of SELU treatment was 99.5%. The ORR in the SELU group (N = 71) was 23.9% (95% CI; 14.6, 35.5). Of the 17 pts who achieved objective response, 10 (58.8%) remained in response for ≥12 months while 5 pts (29.4%) had not yet reached 12 months follow-up from onset of response. Median duration of response was not reached in the SELU group by final analysis. During the SELU period (N = 137), 51 pts (37.2%) had a maximum AE severity of ≥Grade 3 (most frequent were known SELU AEs); ≥1 AEs, possibly related to SELU, were experienced by 127 pts (92.7%); of these, 31 pts (22.6%) had ≥Grade 3 AEs. Serious AEs (SAEs) were experienced by 24 pts (17.5%); 6 pts (4.4%) had SAEs possibly related to SELU. AEs of special interest were experienced by 85 pts (62.0%); increased blood creatine phosphokinase (43.1%), increased ALT/AST (13.9% each), and peripheral edema (13.1%) were the most frequent (≥10% of pts). AEs were manageable; the AE-related discontinuation rate for the trial was 9.5%. At C1 D8, 64 pts who received SELU were included in the PK Analysis Set. Median t max was 1.5 h and geometric mean AUC (0-12) was 2986 h × ng/mL for SELU. Conclusions: In the first international, randomized, PBO-controlled trial in adults with NF1-PN, SELU achieved a sustained and durable, clinically meaningful reduction in PN volume per ICR REiNS. No new safety concerns were identified and AEs were manageable. Clinical trial information: NCT04924608 .
Distinction of FLT3-ITD transcriptomic signature from FLT3-TKD and the frequency of this signature in acute myeloid leukemia without <i>FLT3</i> mutation.
6538 Background: The FLT3 gene plays a major role in acute myeloid leukemia and frequently overexpressed irrespective of its mutation status. FLT3 internal tandem duplication (ITD) and tyrosine kinase domain (TKD) mutation are both associated with adverse prognosis, but the ITD is associated with more aggressive disease and higher relapses than the TKD. Therapeutically, FLT3-TKD has different resistance implications from FLT3-ITD. We used transcriptomic data to compare AML FLT3-ITD-positive (ITD+) with FLT3-TKD-positive (TKD+), then tested if AML FLT3-negative (FLT3-) may have RNA signature similar to those seen in ITD+ or TKD+ AML. Methods: Between 2023 and 2025, 906 AML samples were RNA profiled using next generation sequencing. A targeted NGS panel of approximately 1,600 genes was used. More than 100 million reads were required to accept RNA results. The required percentage of spliced RNA reads was above 20%. Of these samples, 171 had FLT3-ITD and 115 had FLT3-TKD mutations and 619 were FLT3-. Results: We first explored if expression profiling can define a subgroup of TKD+ cases that are similar to those with ITD+ AML. Using 22 genes in random forest modeling, expression profiling segregated ITD+ AML from TKD+ AML with AUC of 0.734 (95% CI: 0.638-0.831, Precision: 0.940). The top genes relevant for this segregation were GRB10, PTK7, WT1, TEC, CASP3, NR6A1, and CD47. We then build a random-forest-based model to distinguish ITD+ from FLT3- cases. We used the 171 ITD+ AML and a set of 220 FLT3- AML cases to distinguish between the two groups. Using 45 genes in this model we were able to segregate with the ITD+ cases based on expression with AUC of 0.889 (95% CI: 0.833-0.945, precision : 0.935). Top genes were LUC7L2, NFYC, PRPF40B, PRPF8, SF3A1, SMAD5, THRA, ACVR1, EVI2A, MCM3AP. This highly predictive model was then used to test independent set of 399 FLT3- AML cases. Of these cases 319 (80%) had a signature similar to ITD+ AML. Similarly, we used the 115 TKD+ with the 220 FLT3- AML cases and built a random forest-based model to distinguish between the two groups. This model required 70 genes and was less robust but segregated the TKD+ AML with AUC of 0.784 (95% CI: 0.701-0.868, precision: 0.842). Testing an independent set of 399 FLT3- cases using this model showed TKD signature in 236 (59%) of FLT- cases. All these positive TKD signature cases were also positive for the ITD signature. Conclusions: This data suggests that significant number of AML cases are biologically driven by changes similar to those seen in FLT3-mutant cases. While clinical trial is needed for confirmation, these cases with FLT3-ITD signature may respond efficiently if treated with therapy including FLT3 inhibitors in fashion similar to that seen ITD+ cases. Furthermore, this data suggests that ITD signature might be a better biomarker than TKD mutation for selecting patients to be treated with FLT3 inhibitors.
Correlative analysis of phase II trial of olaparib and durvalumab in patients with <i>IDH</i> -mutated cholangiocarcinoma.
4157 Background: IDH1/2 mutations occur in ~25% of cholangiocarcinomas (CCAs). Preclinical studies suggest synergy between PARP inhibition and PD-L1 blockade. We previously conducted a phase II trial of olaparib plus durvalumab in advanced IDH -mutant CCA. We report matched baseline tissue and serial cell-free DNA (cfDNA)-based correlative analysis of pts samples. The primary objective was to develop a cfDNA methylome signature as a non-invasive biomarker. Methods: This single-arm, open-label phase II study (NCT03991832) enrolled patients with unresectable/metastatic IDH -mutant CCA and ≤2 prior lines of systemic therapy. Patients received olaparib 300 mg orally twice daily and durvalumab 1500 mg IV every 4 weeks. Blood samples were collected at baseline and monthly while on treatment for planned correlatives to examine R2HG, S2HG, and artemin levels using HPLC (High-Performance Liquid Chromatography). Cell-free methylated DNA immunoprecipitation and high-throughput sequencing (cfMeDIP-seq) was performed on pts cfDNA samples and case-matched IDH -wildtype pts. Formalin-fixed, paraffin-embedded blocks of baseline tumor biopsies were used for multispectral fluorescent IHC (FL-IHC) using a 6-marker panel composed of pan-cytokeratin, CD4, CD8, CD163, PD-L1, and CD68. Results: Ten patients were enrolled (median age 63.5 y; 50% female; 90% received prior platinum). Median treatment duration was 1.95 mo (range 1.8–13.5). No complete or partial responses were observed. DCR was 30% (3 patients with stable disease); one patient remained on therapy for 13.5 mo before progression. Median PFS was 1.97 mo (95% CI 1.73–3.93). HPLC from baseline plasma showed lower baseline artemin levels correlated with stable disease (p=0.038), while R2HG/S2HG ratios showed no association with outcome. FL-IHC showed enriched immune cell populations, specifically CD4 + T cells (p=0.0032) and a trend for CD68 + macrophages (p=0.062), in the tumor stroma compared with tumor core. 54 plasma samples (from 8 patients) were profiled with cfMeDIP-seq along with 8 IDH -wildtype controls. Using top 100 differentially methylated regions, the circulating tumour DNA methylome exhibited a highly specific signature, accurately discriminating IDH -mutant and IDH -wildtype CCA regardless of treatment status. The specific differentially methylated gene pathways between groups were comprehensively examined in both on-treatment and baseline plasma samples. Conclusions: Advanced IDH -mutant CCA harbors an immune-excluded phenotype, which may underlie primary resistance to immunotherapy. Plasma cfDNA methylome signature can accurately discriminate CCA with IDH mutations. Clinical trial information: NCT03991832 .
Comparative outcomes of prophylactic cranial irradiation and brain surveillance in limited-stage small cell lung cancer: A systematic review and meta-analysis.
e20142 Background: Prophylactic cranial irradiation (PCI) improves survival outcomes versus no PCI in limited-stage small-cell lung cancer (LS-SCLC), but prior trials predated routine brain MRI; the benefit over modern MRI/CT surveillance is unclear as the phase 3 MAVERICK trial is ongoing. We performed a meta-analysis to compare PCI with routine imaging surveillance and inform practice in the modern imaging era. Methods: We systematically searched PubMed, Embase, and the Cochrane Library for studies published from 2023 to November 18, 2025. We included patients with LS-SCLC in complete or partial remission without brain metastases confirmed by brain MRI or CT. The intervention was PCI; the comparator was routine brain MRI or CT surveillance. The primary outcome was overall survival (OS); secondary outcomes were brain metastasis–free survival (BMFS) and progression-free survival (PFS). Outcomes were summarized as hazard ratios (HRs). When HRs were not reported, they were estimated from Kaplan–Meier curves. Random-effects models were used, and heterogeneity was summarized with I². Analyses were performed in R (version 4.4.2). For studies with propensity score matching (PSM), matched cohorts were used for meta-analysis. Results: A total of 714 patients receiving PCI and 709 patients managed without PCI were included across eight studies. The meta-analysis used propensity score–matched (PSM) cohorts when available (six studies) and full cohorts from two non-PSM studies. Across included cohorts, first-line therapy was radiotherapy with or without concurrent or sequential chemotherapy in 1,327 (92.2%) patients and thoracic surgery in 112 (7.8%) patients. None of the included studies reported whether immunotherapy was used as first-line therapy. The timing of surveillance brain imaging varied across studies, ranging from every 3–4 months during the first 1–2 years, every 3–6 months during years 2–3, and every 3–12 months thereafter. Additionally, the timing and regimens of salvage treatments differed among studies. For OS (8 studies), PCI was associated with better survival (pooled HR 0.74, 95% CI 0.57-0.96; I² = 56.5%, Q = 16.11). For PFS (3 studies), PCI was related to improved outcomes (pooled HR 0.77, 95% CI 0.65- 0.90; I² = 0.0%, Q = 0.44). For BMFS (4 studies), PCI was linked to superior BMFS (pooled HR 0.37, 95% CI 0.25- 0.54; I² = 0.0%, Q = 2.67). Conclusions: Compared with imaging surveillance, PCI was associated with improved OS, PFS, and BMFS in LS-SCLC patients. Heterogeneity among comparator follow-up strategies underscores the need for randomized trials reporting survival outcomes and long-term neurocognitive effects. These findings may help inform clinical decision-making while awaiting results from the ongoing MAVERICK randomized trial.
Timing of adjuvant nivolumab and survival outcomes in bladder cancer: A real-world analysis.
e16562 Background: Adjuvant nivolumab improves outcomes in high-risk muscle-invasive bladder cancer (MIBC) when initiated within 120 days of radical cystectomy, but optimal timing in real-world practice remains uncertain. Postoperative recovery, neoadjuvant chemotherapy (NAC), and logistical barriers often delay treatment. We evaluated whether earlier initiation of adjuvant nivolumab is associated with improved survival outside of clinical trials. Methods: We performed a retrospective cohort study using the TriNetX database (2017–2024), including adults with bladder cancer who underwent radical cystectomy followed by adjuvant nivolumab. Patients who underwent neoadjuvant chemotherapy were excluded. Patients were stratified by time from cystectomy to nivolumab initiation: <60 days vs 60–180 days. Cohorts were propensity score matched (1:1) on age, sex, race, comorbidities, and treatment history. Overall survival (OS) and time to next treatment (TTNT) were evaluated using Kaplan–Meier methods and log-rank testing. Results: In the primary analysis, both matched cohorts included 114 patients initiating nivolumab <60 days and 60–180 days after cystectomy. At 1 year, OS was significantly higher in the < 60-day group (88% vs 75%; p = 0.009). At 3 years, earlier initiation of nivolumab < 60 was still associated with a significantly higher OS (72% vs 53%; p = 0.016). No significant difference in TTNT was observed (56 vs 56 days; p = 0.808). Conclusions: Early initiation of adjuvant nivolumab within 60 days of radical cystectomy is associated with improved overall survival in real-world bladder cancer patients. These findings support timely postoperative initiation of immunotherapy when clinically feasible. These results should be interpreted cautiously due to the limited sample size. Further studies incorporating a larger sample size are warranted.
Bridging the gap in endometrial cancer: A multi-center initiative to standardize biomarker testing and community team-based care.
e23321 Background: The incidence of endometrial cancer (EC) has increased in the past decade. Integration of biomarker (BM) testing, chemo-immunotherapy, and multidisciplinary collaboration are essential to improve outcomes but adds complexity for oncologists. This initiative involved oncology teams from community clinics to assess current practices and develop and implement action plans to overcome barriers to optimal evidence-based treatment and patient (pt)-centered care. Methods: In 2024, healthcare providers (HCPs, N = 61) from four US community oncology clinics completed baseline surveys assessing attitudes, practice patterns, and challenges in EC management. HCPs participated in audit-feedback sessions (N = 65) to identify root causes of practice gaps and develop and implement action plans. Clinic champions (1-4 per clinic) completed follow-up surveys (N = 11) and participated in a workshop (N = 10) to assess sustainable practice changes and exchange solutions related to action plan implementation at their sites. Results: Baseline surveys identified variability in BM testing practices across community clinics. While most HCPs routinely assessed HER2 expression (72%), microsatellite instability or mismatch repair (67%), p53 abnormality (67%), and estrogen and progesterone receptors (66%), fewer assessed tumor mutational burden (41%), POLE mutations (20%), or NTRK fusions (15%). Reported use of chemo-immunotherapy, which is often BM-directed, for advanced or recurrent EC varied across HCPs (median: 70%, range 5%-100%). HCPs’ top-reported barriers to integrating BM-directed therapies were (1) keeping up with the latest guideline recommendations and clinical evidence for BM-directed therapies and (2) pt factors. Top barriers to multidisciplinary team-based care included low frequency of communication or communication barriers and misalignment on methods of communication. Thirty-eight percent of HCPs emphasized that enhanced collaboration across interprofessional teams would most improve care for EC pts. Prior to the sessions, only 40% of HCPs felt somewhat or very comfortable (4/5 Likert scale) interpreting and applying BM results to guide treatment for EC pts; following the sessions, this increased to 57%. Most clinic champions (55%) reported implementing BM testing workflow improvements after the sessions, including the adoption of shared communication platforms with pathologists to streamline requests and shorten turnaround times. Conclusions: This multi-center initiative revealed challenges in BM integration and team-based coordination in the care of EC pts in community oncology clinics. Audit-feedback sessions, action planning, and structured collaboration can support improvements in testing workflows and communication, and ultimately, may help community clinics accelerate adoption of evidence-based treatment for EC pts.
Machine learning–integrated routine biomarkers for breast cancer risk stratification: A scalable strategy for resource-limited settings.
e12557 Background: Early breast cancer diagnosis is essential for improving survival outcomes and optimizing healthcare resource utilization. However, in resource-limited settings and low- and middle-income countries, access to screening programs and advanced risk assessment tools remains limited, and scalable strategies leveraging routinely collected data are underexplored. We evaluated routine blood-based biomarkers, individually and in combination, using a machine learning (ML) approach to develop a low-cost, scalable breast cancer risk assessment tool. Methods: This cross-sectional study included 236 women aged 40–70 years evaluated at Barretos Cancer Hospital between January 2024 and January 2025. Participants comprised newly diagnosed breast cancer patients (n = 125) and women undergoing screening with benign imaging findings (BI-RADS 1–2; n = 111). Routine laboratory markers included complete blood count, metabolic and inflammatory markers, thyroid and sex hormones, and tumor markers. Group comparisons were performed using Student’s t -test. A ridge regression model was trained using biomarkers significantly associated with breast cancer (p < 0.05). Results: The median age was 52 years. The majority of participants were White (51.7%), followed by Brown (37.1%) and Black (10.3%); 56.8% were postmenopausal. Several biomarkers were significantly higher in women with breast cancer (p < 0.05), including CA 15-3 (p = 0.018), CEA (p = 0.014), FSH (p = 0.016), T3 (p < 0.001) and Mean Corpuscular Hemoglobin Concentration - MCHC (p = 0.014). No strong correlations were observed among selected variables (maximum correlation 0.19 between CEA and CA 15-3). Incorporation of biomarkers with p < 0.05 (CA 15-3, CEA, FSH, T3 and MCHC) into the ridge regression model yielded a mean AUC of 0.72 (95% CI, 0.68–0.75) based on 1,000 iterations of stratified random subsampling, with sensitivity of 0.63 ± 0.08, specificity of 0.67 ± 0.08, accuracy of 0.65 ± 0.05, balanced accuracy of 0.65 ± 0.05, negative predictive value of 0.63 ± 0.05, and positive predictive value of 0.67 ± 0.06. Conclusions: An ML model combining routine blood-based biomarkers demonstrated potential as a low-cost pre-screening tool for breast cancer risk stratification. This approach may help prioritize diagnostic resources, particularly in resource-limited settings where access to advanced imaging and invasive diagnostic procedures is constrained. External validation in independent and diverse populations is required to confirm generalizability and support future clinical application.
Diagnostic performance of methylene blue vital staining for early detection of oral epithelial dysplasia and carcinoma in high-risk populations: A meta-analysis.
e18008 Background: Early detection of oral precancer and cancer significantly improves prognosis, yet reliable and affordable screening tools remain limited in low-resource settings. Methylene blue staining is a simple, low-cost adjunctive technique proposed for identifying early mucosal lesions. This diagnostic test accuracy (DTA) meta-analysis evaluated its pooled sensitivity, specificity, and overall discriminative performance. Methods: A systematic search of databases was performed through November 2025. Studies assessing methylene blue staining for detecting oral precancer or cancer, using histopathology as the reference standard, and providing sufficient data to derive 2×2 contingency tables were included. Pooled sensitivity and specificity were calculated using a bivariate random-effects model. Statistical analyses were performed using Stata 17.0, R 4.5.2, and Meta-DiSc 2.0. Results: A total of nine studies were included. The pooled sensitivity of methylene blue staining was 88.5% (95% CI; 79.2%-94.0%). The pooled specificity was 56.1% (95% CI; 36.1%-74.3%), reflecting moderate discrimination and a notable false-positive rate. The SROC curve showed an AUC of 0.846, supporting good overall diagnostic performance. The pooled positive likelihood ratio (LR+) was 2.02, while the negative likelihood ratio ratio (LR-) was 0.20. Substantial heterogeneity was observed for both sensitivity (I² = 78.9%) and specificity (I² = 86.9%), likely due to differences in lesion stage, staining technique, and study design. Conclusions: Methylene blue is a highly sensitive and moderately accurate adjunctive tool for early detection of oral precancer and cancer, particularly valuable in low-resource settings. However, limited specificity suggests it should complement, not replace, definitive diagnostic evaluation. To improve its clinical application, more standardized, prospective research is required.
Dentist-led systematic oral education programs in the prevention of radiation-induced oral toxicities in patients with head and neck cancer (BRIGHT): A randomized clinical trial.
12144 Background: Radiotherapy-induced oral toxicity (RIOT) is a common complication in patients with head and neck cancer (HNC). International guidelines have recognized the role of oral-health education in preventing RIOT. However, most recommendations remained at the level of general principles and lacked a detailed, systematic and operationalized process, which hinders effective clinical delivery of oral-health education. Given that dentists are uniquely positioned to manage oral health, this study aims to evaluate the effectiveness of a dentist-led systematic oral-health education program (OHEP) on RIOT and provide a standard operating procedure (SOP) for its implementation. Methods: In this single-centre, assessor-blinded, randomised controlled trial, 150 patients with non-metastatic HNC scheduled for definitive radiotherapy (60–72 Gy) were enrolled at West China Hospital in China. Participants were randomly assigned (1:1) to receive either a dentist-led oral health education program (OHEP) or routine oral health care. The primary endpoint was the incidence of oral mucositis (OM), assessed by the WHO Oral Mucositis Grading Scale. Secondary endpoints included the incidence of severe oral mucositis (SOM, defined as WHO grade 3–4) as well as the time to onset and duration of OM (SOM). Outcome assessments were conducted by trained research assistants blinded to group allocation. Analyses were performed in both intention-to-treat and per-protocol populations, with two-sided P < 0.05 considered statistically significant. This study is registered with ClinicalTrials.gov, NCT06690346, and is now complete. Results: Of 150 randomised patients, 75 were allocated to each arm, and 142 (94.7%) completed the trial and were included in the intention-to-treat analysis. The incidence of OM was significantly lower in the OHEP arm than in the control arm (70.7% [53/75] vs. 88.0% [66/75] P = 0.016), with a significantly delayed onset (median 17 vs. 14 days; P = 0.006) and shorter duration (median 39 vs. 60 days; P < 0.001). Consistent results were observed in the per-protocol population. However, no significant difference was observed in the incidence, onset, or duration of SOM between arms. Conclusions: Dentist-led OHEP significantly reduced the incidence and delayed the onset of radiotherapy-induced OM. These results support dentist-led OHEP as a feasible, non-pharmacological approach to mitigate oral toxicity and improve supportive care in patients receiving radiotherapy for HNC. Clinical trial information: NCT06690346 .
Survival outcome of sandwich therapy with induction chemoimmunotherapy followed by concurrent chemoradiotherapy and sequential immune maintenance therapy in locally advanced unresectable esophageal squamous cell carcinoma (ESCC) patients: A multi-center prospective study.
4068 Background: PD-1 inhibitor, combined with chemotherapy, is recommended in first-line treatment for metastatic ESCC, but its role remains undefined in locally advanced ESCC. This trial aimed to investigate the survival outcome and safety of the ‘Sandwich Therapy’ consisting of induction chemoimmunotherapy followed by definitive chemoradiotherapy (dCRT) and sequential immune maintenance therapy. Methods: This is a multicenter, prospective study. Patients with histologically confirmed, clinical stage I–IVA and unresectable esophageal squamous cell carcinoma (ESCC) were included. Patients received 1-8 cycles of chemoimmunotherapy (intravenous paclitaxel (175 mg/m 2 ) plus platinum-based chemotherapy), then concurrent chemoradiotherapy followed by tislelizumab (200mg intravenously, every 3 weeks) until disease progression or the end of the study. The primary outcome was progression-free survival (PFS). The secondary outcomes included overall survival (OS) and the incidence of adverse events (AE). Results: Between August 2022 and Aug 2025, 77 patients were enrolled and all proceeded to the immune maintenance therapy phase. The median age was 61 years (IQR 55-69), and 78% had clinical stage III or IV disease. With median follow-up of 19 months (95CI), median PFS was 17.0 months (95% CI:12.0–36.0) , with 1-, 2- and 3-year PFS rate of 56.4%, 39.9% and 25.6%, respectively. The median OS was 45.0 months (95%CI:20.1-69.3), with 1-, 2- and 3-year OS rate of 74.9%, 63.2% and 50%, respectively. The objective response rate (ORR) was 55.8%, and the disease control rate was 92.2%. Grade ≥3 AE occurred in 27 (35.1%) patients during the treatment. The most common grade ≥3 AE was lymphopenia, which was reported in 22 (28.6%) patients. Conclusions: For locally advanced ESCC, induction chemoimmunotherapy followed by dCRT and subsequent maintenance of immunotherapy may improve PFS and OS with a manageable safety profile. Clinical trial information: NCT05515315 .
waveLINE-011: Zilovertamab vedotin plus R-CHP versus polatuzumab vedotin plus R-CHP as first-line therapy in germinal center B-cell–like (GCB) diffuse large B-cell lymphoma (DLBCL).
TPS7104 Background: DLBCL is a heterogeneous malignancy broadly categorized into 3 subtypes: activated B-cell–like (ABC) DLBCL, GCB DLBCL, and unclassified DLBCL. The standard-of-care first-line treatment option for DLBCL regardless of subtype is polatuzumab vedotin plus rituximab, cyclophosphamide, doxorubicin, and prednisone (R-CHP). However, while polatuzumab plus R-CHP prolongs PFS compared with R-CHOP in ABC DLBCL, it provides no benefit over R-CHOP in GCB DLBCL. Alternative first-line treatment options for GCB DLBCL are needed. Receptor tyrosine kinase–like orphan receptor 1 (ROR1) is an oncofetal transmembrane protein that is overexpressed in DLBCL, making it an attractive therapeutic target. Zilovertamab vedotin is an antibody-drug conjugate comprising an anti-ROR1 antibody with a proteolytically cleavable linker and the cytotoxic antimicrotubule agent MMAE. In the phase 1 waveLINE-001 study, zilovertamab vedotin was shown to have antitumor activity in heavily pretreated lymphoid cancers and promising efficacy and manageable safety in combination with R-CHP as first-line treatment for DLBCL. The randomized, open-label, phase 2 waveLINE-011 study (NCT06890884) has been designed to evaluate the efficacy and safety of zilovertamab vedotin plus R-CHP versus polatuzumab vedotin plus R-CHP in participants with previously untreated GCB DLBCL. Methods: Key eligibility criteria include age ≥18 years, previously untreated histologically confirmed GCB DLBCL (including but not limited to: DLBCL not otherwise specified GCB type and high-grade B-cell lymphoma [HGBL] GCB subtype), PET-positive disease at screening, an ECOG performance status score of 0 to 2, and an International Prognostic Index (IPI) score of 2 to 5. Participants with a history of transformation of indolent disease to DLBCL or with primary mediastinal B-cell lymphoma or gray zone lymphoma are excluded. All participants will be randomly assigned (1:1) to zilovertamab vedotin 1.75 mg/kg plus R-CHP or polatuzumab vedotin 1.8 mg/kg plus R-CHP on day 1 of every 3-week cycle for 6 cycles. Randomization will be stratified by geographic region (Western Europe vs US vs rest of world), IPI score (2 vs 3-5), bulky disease (<7.5 cm vs ≥7.5 cm), and histopathology (HGBL vs others). The primary end point is CR rate at the end of therapy per Lugano criteria by BICR. Secondary end points include PFS, EFS, duration of CR, OS, safety, and change in PROs. Response assessments (CT/FDG-PET) will be performed after cycle 4 day 1 but on or before cycle 5 day 1, and at 12 weeks after the cycle 4 scan. Follow-up efficacy assessments will be completed every 24 weeks for 2 years from the end of treatment assessment, then every year for 3 years. Adverse events will be graded per NCI CTCAE v5.0. Approximately 594 participants will be enrolled. Recruitment is ongoing. Clinical trial information: NCT06890884 .
Effect of dental prophylaxis on long-term systemic risks in therapy-induced oral mucositis: A real-world cohort analysis.
e24195 Background: Oral ulcerating mucositis (OUM) is a common complication of radiation and antineoplastic therapy, often leading to pain, infection risk, and long-term systemic effects. Dental prophylaxis (DP) may mitigate these risks by improving oral hygiene, but its long-term impact remains understudied. The objective of this study is to evaluate the association between DP and long-term outcomes, including pain, sepsis, malaise, fever, lymphadenopathy, vascular diseases, respiratory complications, pleural diseases, and ischemic heart disease (IHD), in adults with therapy-induced OUM. Methods: Retrospective cohort study using the TriNetX federated health research network (deprecated COVID-19 Research Network, 89 healthcare organizations). Cohort 1 included 41,958 patients aged ≥18 years with OUM due to radiation or antineoplastic therapy without DP. Cohort 2 included 387 similar patients with DP. Index event was the first OUM diagnosis; outcomes were assessed starting 1-day post-index with no end date (up to 20 years). Propensity score matching was not applied. Measures of association (risk difference, ratio, odds ratio), Kaplan-Meier survival, and number of instances for each outcome, excluding patients with prior outcomes where specified. Results: Patients without DP showed higher risks for several outcomes. For lymphadenopathy (including prior occurrences), risk was 0.213 vs 0.158 (risk difference 0.055 [95% CI, 0.018-0.091]; P = 0.008), with reduced survival probability (48.89% vs 69.45%; log-rank P < 0.001; hazard ratio [HR] 1.560 [95% CI, 1.213-2.007]). Respiratory diseases had higher risk (0.296 vs 0.241; P = 0.077) and worse survival (44.85% vs 48.93%; log-rank P = 0.006; HR 1.461 [95% CI, 1.113-1.919]). IHD risk was elevated (0.122 vs 0.087; P = 0.054), with poorer survival (67.89% vs 69.03%; log-rank P = 0.008; HR 1.644 [95% CI, 1.133-2.384]). No significant differences were observed for pain, sepsis, malaise, fever, arterial diseases, or pleural diseases. Conclusions: DP was associated with reduced long-term risks of lymphadenopathy, respiratory complications, and IHD in patients with therapy-induced OUM. These findings suggest potential benefits of routine dental interventions in oncology care, warranting prospective studies.
Midday immune checkpoint inhibitor (ICI) infusion influence on survival in patients with metastatic cancer: A real-world chrono-immunotherapy study.
11162 Background: Circadian rhythms regulate immune activity, and earlier ICI dosing has been linked to better survival. However, prior findings remain inconsistent, often relying on predefined time categories. To address this, we clustered actual infusion times to assess survival associations in a real-world cohort. Methods: This retrospective study included metastatic cancer patients treated with ICIs. Mean dosing times (up to 24 infusions/patient) were analyzed using circular statistics and clustered via k-means. Overall survival (OS) was defined from ICI initiation to progression, death, or last follow-up, and analyzed via Kaplan-Meier and multivariable Cox regression adjusted for cancer and ICI type. Results: A total of 640 patients (median age 61 years [range 23–87]; 78% male) were included. The most common tumor types were non-small cell lung cancer (52%), renal cell carcinoma (19%), and malignant melanoma (12%), followed by urothelial (5.5%) and small cell lung cancer (4.8%). The majority of patients received nivolumab (74%) or pembrolizumab (11%), while a smaller proportion were treated with other ICIs including atezolizumab (6.3%), avelumab (2%), or ipilimumab-based combinations (5.4%). ECOG performance status was 0–1 in 80% of patients. The most frequent metastatic sites were lymph nodes or soft tissue (78.6%), lungs (48.1%), bones (32.8%), liver (17.7%), and central nervous system (14.2%). Based on mean ICI infusion times, patients were grouped into three clusters: cluster 1 (2:15 PM, n = 76), cluster 2 (1:31 PM, n = 151), and cluster 3 (11:04 AM, n = 413). Median OS was 50.4, 28.2, and 15.7 months in Clusters 2, 1, and 3, respectively (log-rank p = 0.003). In multivariable analysis, Cluster 2 showed improved OS compared to Cluster 1 (Hazard Ratio [HR] 0.52, 95% Confidence Interval [CI] 0.31–0.86, p = 0.01), while no difference was seen between Clusters 3 and 1 (HR 1.13, 95% CI 0.75–1.70, p = 0.59). Conclusions: ICI administration around midday was associated with the longest survival, aligning with circadian peaks in T-cell infiltration, dendritic cell migration, and immune responsiveness. These findings support a personalized chrono-immunotherapy approach based on intrinsic biological rhythms.
Laparoscopic versus open distal gastrectomy for clinical T4a gastric cancer: Short-term outcomes and interim survival analysis of the UMC-UPPERGI-01 randomized clinical trial.
e16142 Background: Laparoscopic distal gastrectomy (LDG) is widely accepted for early and advanced stages gastric cancer (GC), but its oncological efficacy for T4a tumors remains controversial. We previously demonstrated the surgical safety of LDG with comparable short-term outcomes to open surgery in cT4a GC (JAMA Surgery, 2025). This interim analysis aims to evaluate the oncological outcomes of LDG versus open distal gastrectomy (ODG) in this high-risk population. Methods: In this phase III, single-center, non-inferiority RCT (NCT04384757), patients with cT4aN0-3M0 GC were randomized 1:1 to LDG or ODG with D2 lymphadenectomy. The primary endpoint is 3-year disease-free survival (DFS) with a non-inferiority margin of 1.45 for the Hazard Ratio (HR). Survival rates were estimated using the Kaplan-Meier method, and HRs were calculated using Cox proportional hazards models. All procedures were performed exclusively by qualified surgeons. Results: After exclusion, a total of 208 patients (LDG: 104; ODG: 104) were analyzed. Baseline, pathological characteristics and adjuvant chemotherapy rates were well-balanced . Regarding short-term outcomes, although LDG had longer operative time (220.0 vs 153.7 min; P < 0.001) and higher median blood loss (80 vs 50 mL; p = 0.003), there were no significant differences in 30-day morbidity (22.1% LDG vs 21.2% ODG; p = 0.87) or severe complications (2.9% vs 3.8%; p > 0.99). Recovery parameters including time to first flatus, oral tolerance, and length of hospital stay were also comparable. Moreover, the time from surgery to initiation of adjuvant chemotherapy did not differ between groups. At the interim survival analysis, the 3-year DFS rate was 67% (95% CI, 57%–79%) in the LDG group and 70% (95% CI, 62%–80%) in the ODG group. The estimated HR for DFS was 1.35 (95% CI, 0.78–2.32; p = 0.26). Although the HR point estimate was within the margin, non-inferiority could not be formally established at this stage as the upper limit of the 95% CI (2.32) exceeded the margin (Delta_0 = 1.45). The 3-year overall survival (OS) rate was 76% (95% CI, 66%–86%) in the LDG group and 74% (95% CI, 64%–86%) in the ODG group (HR 1.16; 95% CI, 0.64; 2.11; p = 0.63). Recurrence patterns did not differ significantly between groups. Subgroup analyses (age, tumor size, macroscopic type) showed no significant differences in DFS between the two approaches. Conclusions: This interim analysis of the UMC-UPPERGI-01 trial suggests that LDG provides comparable 3-year survival outcomes to ODG for cT4a gastric cancer. While the statistical non-inferiority is not yet established due to the current interval of follow-up and event accrual, the oncological trends and equivalent recurrence patterns support LDG as a viable surgical alternative in experienced surgeons. Final analysis will provide definitive evidence. Clinical trial information: NCT04384757 .
Oncologists’ knowledge, clinical practices, barriers and facilitators of cardio-oncology guideline-concordant care.
e24011 Background: Cancer and cardiovascular (CV) disease are leading global causes of morbidity and mortality, with rising CV complications across all cancer types.The 2022 European Society of Cardiology (ESC) Cardio-Oncology guidelines provide several (over 150 of 272) class 1 recommendations for cardio-oncology care. We evaluated knowledge and practice of the class 1 recommendations pertinent to oncology practice. Methods: A cross-sectional questionnaire, hosted on REDCap, was distributed to Canadian oncology providers from November 2025-January 2026. It tested knowledge of class 1 recommendations (4 general questions and additional questions by prescribed therapy), awareness of ESC guidelines, and perceived barriers to clinical uptake. Data were analysed descriptively, with associations between responses and respondent characteristics explored descriptively and using logistic regression. Results: Sixty-three respondents completed the survey. Most practiced in academic settings (75.8%) and devoted ≥50% of time to clinical work (88.7%). Almost half of respondents were within the first 10 years of practice (42.8%) and reported moderate to high exposure to cardio-oncology (44%). Forty percent (19/47) got 3 or 4 correct answers out of 4 general knowledge questions. Over half (63.3%) identified the correct tool to calculate baseline CV risk, while fewer correctly identified high baseline CV toxicity risk scenarios (38.8%) or appropriate cardio-oncology referral indications (39.6%). Less than 50% of anthracycline prescribers correctly identified the correct timing for post-treatment echocardiography. Over 70% of immune checkpoint inhibitor prescribers were unaware of recommended baseline CV assessment. In contrast, > 90% of HER2-therapy prescribers correctly identified management of symptomatic LVEF < 50% and nearly all CAR-T prescribers correctly identified the recommended baseline CV assessment. Male sex at birth was quantitatively but not statistically correlated with correctly answering knowledge questions (OR 2.6, 95%CI 0.73-9.26, p = 0.14) but greater exposure to cardio-oncology (regular involvement and/or an academic interest in cardio-oncology) and years in practice were not. Three quarters of respondents reported rarely or never consulting ESC guidelines. Key barriers included guideline length, time constraints, and limited access to cardio-oncology services. Simplified guidelines geared to oncologists, workflow prompts, and timely access to investigations and cardio-oncology specialists were perceived to facilitate guideline concordant care. Conclusions: Oncology providers are well positioned to identify individuals at CV risk at all treatment phases, yet have low knowledge and report barriers to implementation of oncology-specific ESC guidelines. Implementation strategies should be identified and evaluated to improve this gap.
Association of inpatient outcomes with carcinoid syndrome among malignant neuroendocrine tumor hospitalizations in the United States: NIS 2016–2022.
e16320 Background: Carcinoid syndrome (CS) may contribute to inpatient morbidity and resource utilization among patients with neuroendocrine tumors (NETs). However, contemporary national-level data describing the inpatient burden, complications, and outcomes associated with CS remain limited. We aimed to characterize national inpatient trends and assess outcomes associated with CS in malignant NET hospitalizations. Methods: Adult hospitalizations in the National Inpatient Sample (NIS) 2016–2022 with malignant NET diagnoses (ICD-10-CM C7A*/C7B*) were identified. CS was defined by any-diagnosis E34.0*. Survey weights and design-based variance estimates were applied. Outcomes included in-hospital mortality, complications, length of stay (LOS), and total charges (TOTCHG). Survey-weighted regression adjusted for demographics, socioeconomic/hospital factors, calendar year, payer, and key complications (sepsis, septic shock, acute kidney injury [AKI], acute respiratory failure, palliative care). Results: Among 33,318 NET hospitalizations (weighted n = 166,590), 3.24% had CS. Compared with non-CS hospitalizations, CS hospitalizations were older (66.1 vs 64.5 years), more often female (55.3% vs 51.2%), and more often Medicare-insured (62.1% vs 55.3%). AKI was more frequent with CS (30.6% vs 21.0%; P < 0.001), while sepsis and acute respiratory failure rates were similar. Unadjusted mortality was higher with CS (7.14% vs 5.55%), but CS was not independently associated with mortality after adjustment (aOR 1.19, 95% CI 0.90–1.58). CS was independently associated with longer LOS (+0.51 days) and higher charges (+$11,172). CS prevalence did not significantly vary by year. Conclusions: CS was present in ~3% of malignant NET hospitalizations and was associated with increased AKI and independently higher LOS and charges, without an independent association with in-hospital mortality after adjustment. Prospective studies are needed to clarify causal pathways and evaluate targeted inpatient interventions to prevent renal injury and reduce resource utilization.
Real-world evidence of immunotherapy-associated outcomes in patients with hepatocellular carcinoma: Insights from TriNetX database.
e16193 Background: Immunotherapy has become a cornerstone in the management of hepatocellular carcinoma (HCC), with immune checkpoint inhibitors (ICIs) and their combinations now established as preferred first-line systemic therapies. While ICIs have improved overall outcomes in HCC, only a subset of patients derive meaningful benefit. This underscores the critical need for understanding the clinical predictors to optimize treatment response in this population. Currently, large scale epidemiological studies are lacking. This investigation aims to explore the utilization of immune checkpoint inhibitors in treatment of HCC and associated treatment response and outcomes in real-world setting. Methods: For this cohort study, we utilized the US Collaborative Network within the TrinetX database, encompassing 111 healthcare organizations. We analyzed all participants aged 18 years and older diagnosed with hepatocellular carcinoma. Key clinical covariates including demographics, tumor characteristics, liver disease etiology, laboratory biomarkers, treatment, and outcomes were analyzed. We examined the lines of treatment and treatment pathways to assess the differences in survival probabilities using the Kaplan-Meier method. Cox proportional hazards regression modeling was utilized to identify clinical predictors and laboratory biomarkers for treatment response. P < 0.05 was considered significant. Results: We identified 206,159 patients diagnosed with HCC. The majority of the patients were male (69%) with a mean age of 63.9 years (standard deviation: 12.2 years). The cohort was 49% White, 9% Black, 17% Asian, and 8% Hispanic. Liver cirrhosis was present in 54%, while 20% had alcoholic liver disease, 33% had viral hepatitis, and 24% developed hepatic failure. Of these, 46.5% were treated with only Sorafenib, 12.1% were treated with Atezolimumab and Bevacizumab combination, 10.5% with Nivolumab, 6.5% with Pembrolizumab, and 4.6% with Durvalumab + Tremelimumb. After balancing for clinical covariates and year of HCC diagnosis, we identified different survival probabilities based on Kaplan-Meier curves (log rank P < 0.05). Further multivariable analyses are underway to account for time on treatment. Conclusions: We have identified a large cohort of patients with HCC, representative of the real-world setting. With the increased utilization of immune checkpoint inhibitors it is crucial to identify clinical characteristics and laboratory biomarkers associated with treatment response and outcomes. This study will allow us to fill this gap in knowledge.