Browse Articles

Risk factors for second primary malignancies (SPM) in patients with chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL): A real-world study.

e19022 Background: Patients with CLL/SLL are at increased risk of developing SPM. Identification of key clinical risk factors associated with SPM is critical to inform long-term surveillance strategies and optimize disease management. Methods: Adult patients diagnosed with CLL/SLL from Jan 1, 2019 to May 31, 2023 were identified from the Symphony open claims database. Patients were categorized into three cohorts: (1) treated with first-line covalent BTK inhibitor (cBTKi), (2) treated with first-line chemoimmunotherapy (CIT), and (3) those who are managed with active observation. The 36-month incidence of SPM (excl. nonmelanoma skin cancer [NMSC] and hematologic malignancies [HM]) was estimated. Multivariable logistic regressions were used to assess associations between baseline characteristics (demographics; region; obesity; 3-factor risk estimate scale [TRES], a validated comorbidity risk score; and select comorbidities) and SPM occurrence. Results: In total 86,654 patients with CLL/SLL were identified (cBTKi treated, n=11,352; CIT treated, n=12,589; and on active observation, n=62,713). Over 36-month follow-up, the incidence of SPM (excl. NMSC and HM) was 13.1% with cBTKi, 16.6% with CIT, and 11.8% managed with active observation ( P <.0001). In the overall population, age ≥70 yr (odds ratio [OR], 1.43; 95% CI, 1.36-1.50), male sex (OR, 1.32; 95% CI, 1.26-1.38), higher TRES score (OR, 1.11; 95% CI, 1.05-1.18), chronic pulmonary disease (OR, 1.26; 95% CI, 1.19-1.34) and liver disease (OR, 1.30; 95% CI, 1.19-1.41) were independently associated with increased risk of SPM. Similar risk factors were identified when cBTKi, CIT, and patients on active observation, were assessed individually. Conclusions: Older age, male sex, higher disease burden, and chronic comorbidities were associated with increased SPM risk in patients with CLL/SLL. Treatment with CIT was linked to a higher SPM risk compared with cBTKi therapy or observation. Patients with these risk factors should be monitored for the development of SPM during CLL/SLL management. Risk factors, OR (95% CI) Overall CLL/SLL (N=86,654) cBTKi (n=11,352) CIT (n=12,589) Observation (n=62,713) Age (yr) ≥70 vs <70 1.43 (1.36-1.50) 1.35 (1.19-1.54) 1.22 (1.09-1.36) 1.52 (1.43-1.61) Sex Male vs female 1.32 (1.26-1.38) 1.33 (1.17-1.51) 1.19 (1.07-1.33) 1.34 (1.27-1.42) Race/ethnicity Non-Hispanic Black vs Non-Hispanic White 1.07 (0.99-1.16) 1.07 (0.87-1.30) 0.92 (0.75-1.12) 1.11 (1.01-1.23) Hispanic vs Non-Hispanic White 1.05 (0.95-1.16) 0.83 (0.61-1.13) 1.21 (0.98-1.49) 1.04 (0.92-1.17) TRES Scores 2-3 vs scores 0-1 1.11 (1.05-1.18) 1.25 (1.07-1.47) 1.24 (1.09-1.41) 1.05 (0.98-1.13) Select comorbidities Chronic pulmonary disease 1.26 (1.19-1.34) 1.13 (0.96-1.34) 1.22 (1.06-1.39) 1.29 (1.21-1.39) Liver disease 1.30 (1.19-1.41) 1.08 (0.82-1.42) 1.20 (1.01-1.43) 1.32 (1.18-1.47)

AFP response-based model as a prognostic monitor in unresectable hepatocellular carcinoma management.

e16211 Background: Increasing evidence showed atezolizumab plus bevacizumab (A+B) with transarterial therapies (ABT therapy) was a highly effective regimen for unresectable hepatocellular carcinoma (uHCC). While, it is still unclear whether dynamic changes of alpha-fetoprotein (AFP) can early predict tumor response, residual tumor activity, and prognosis of HCC patients in the era of combined therapy. This study aimed to explore the AFP trajectory patterns and its impact on clinical outcomes following ABT therapy. Methods: This retrospective study enrolled uHCC patients with baseline AFP levels ＞25 ng/mL who achieved normalization following ABT treatment. Analyses included time to first AFP-complete response (AFP-CR, AFP normalization measured from treatment initiation) and AFP-progression-free survival (AFP-PFS, defined as the time from the first normalization of AFP to the first documented recurrence of AFP). Results: Among the 276 patients who received ABT as first-line therapy, 91 patients had elevated baseline AFP level that normalized after treatment. The median time to first AFP-CR was 2.4 months. At the time of first AFP-CR, tumor response assessed by mRECIST criteria showed CR in 13 patients (14.3%), partial response (PR) in 58 patients (63.7%), and stable disease (SD) in 20 patients (22.0%), whereas assessment by RECIST 1.1 criteria recorded PR in 37 patients (40.7%) and SD in 54 patients (59.3%). Ultimately, disease progression occurred in 27 (29.7%) patients. Among the 44 patients who achieved a mRECIST-based radiological CR (rCR), it occurred later than AFP-CR in 29 patients (65.9%), with a median lag time of 3.0 months (IQR: 2.0-5.1). Of the remaining 15 patients with mRECIST-rCR, it was concurrent with AFP-CR in 10 patients (22.7%) and occurred earlier in 5 patients (11.4%). The median AFP-PFS was 24.5 months, closely aligning with median mRECIST-PFS of 27.4 months. A total of 73 patients maintained sustained normal AFP level until the last follow-up. Conclusions: AFP trajectories provide valuable information complementary to radiological assessment in the combined modality era. Further prospective study is warranted to solidify the utility of AFP trajectory patterns in optimizing management for uHCC.

Interrogation of fibrolamellar cancer to provide clinically relevant insights into therapeutic vulnerabilities.

10041 Background: Fibrolamellar carcinoma (FLC) is a rare disease that primarily afflicts young adults. Systemic therapy with surgery is the only curative intervention for advanced disease with 5-year survivals of 45%. We examined 3-dimensional tissue explants from FLC patients to measure the efficacy of cytotoxics, targeted agents and novel combinations. Methods: Of 41 FLC specimens; 17 female, 24 male, median age 26 (14-60), 36 (88%) were previously treated, 18/36 (50%) Oxaliplatin & Gemcitabine-Lenvatinib (GEMOX-LEN), 11 (31%) 5FU & Interferon and 18 (50%) Nivolumab. Samples were examined by Ex Vivo Analysis of Programmed Cell Death (EVA/PCD) as previously described. Following disaggregation and 72-hour drug exposures, 5-point dose-response curves were interpolated to provide lethal concentration 50% (LC50) for comparison with 10,000 prior human tumor explant analyses. Drug synergy was evaluated by the method of Chou & Talalay (Chou TC. Cancer Res (2010)15;70(2):440-6). Of 24 drugs and combinations selected for testing, FLC yields were sufficient for 18/24 (75%) to be fully analyzed. To explore mechanisms of response/resistance, single agent LC50’s compared drug activity against all solid tumors by Z-score. Results: By rank order the highest activity for single-agents identified Vorinostat (HDAC), followed by Phenformin (Mitochondrial Complex I) and 6-Diazo-5-Oxo-L-norleucine (Glutaminolysis), then Cobimetinib (MEK) Regorafenib (TKI), KAT (Hexokinase) Lenvatinib (TKI) and Irinotecan (TOPO I). The lowest activity was found for Navitoclax (BH3) followed by Celecoxib (NSAID) and Quercetin (Natural product). Z scores for drug combinations favored GEMOX and 5FU & Interferon compared with other solid tumors. Previously treated tumors were more drug resistant. GEMOX combined with LEN proved synergistic in 40% of samples. Despite low single agent activity, Quercetin showed synergy with Lenvatinib. Discussion: We have previously described the ex vivo study of small molecules as probes of human tumor biology. As FLC uniformly expresses the DNAjb1-PRKACA gene-fusion, we used drugs with known mechanisms of action to interrogate operative pathways and vulnerabilities of the FLC phenotype associated with the DNAjB1-PRKACA genotype. We have previously shown that Vorinostat and Phenformin activities correlate with the MYC-directed BRD-inhibitor JQ1 while Protein Kinase A is known to regulate MYC and glutaminolysis is an established hallmark of MYC activation. Conclusions: The results suggest that the DNAjB1-PRKACA reprograms FLC toward a metabolic phenotype that is vulnerable to epigenetic modulation, mitochondrial targeting and glutaminolysis inhibition. The study met its objective to establish a database for the future management of FLC patients. This will allow each patient’s tissue to be used to craft novel therapies predicated upon new insights into the biology of FLC.

Evaluating value in the age of accelerated approval: A pembrolizumab case study.

11084 Background: The FDA accelerated approval (AA) program was established to expedite approval of promising drugs that meet unfulfilled clinical needs. AA is granted on the basis of surrogate endpoints, decreases time to market, and increases access to therapies. Pursuing multiple AA and label expansions can also be a significant component of a drug’s lifecycle strategy. Among 322 anticancer drug approvals (2009-2020), 59% were label expansions, with only 16% granted to new drugs with a novel mechanism of action. Our goal was to understand changes in value across multiple label-expanding AAs for high-cost anticancer medications. We studied pembrolizumab, a highly successful immune checkpoint inhibitor, evaluating value from its initial approval through multiple label expansions. Methods: We calculated value scores for different indications of pembrolizumab using the ASCO Value Framework Net Health Benefit Worksheet (2020) and ESMO-MCBS v2.0. The value calculations require mature survival outcomes, so we could not calculate scores for AA indications awaiting data maturity (2) or those eventually withdrawn from the market (2). We utilized the FDA drug database and primary literature to gather safety and efficacy outcomes for each indication. Analysis generated descriptive statistics about how the value of pembrolizumab evolved from 2014 to 2025. Results: We calculated value scores for 28 pembrolizumab indications, 12 with AA. Value scores varied widely across indications, with ASCO Net Health Benefit Score (NHBS) ranging from 17 (biliary tract cancer) to 59 (MSI/dMMR colorectal cancer) and ESMO MCBS scores ranging from 1 to 5. NHBS was not perfectly collinear with OS gains. The rolling average of NHBS across all indications over time remained fairly stable with a mean score of 37.67 (range: 30- 41). We also compared the NHBS for each AA indication to the cumulative NHBS average at the time each AA was granted. The NHBS for HCC, HNSCC, and cervical cancer were lower than respective pre-AA NHBS averages. The other 9 AAs yielded a final NHBS higher than each’s pre-AA average NHBS. Conclusions: The NHBS formula generated varying values for different pembrolizumab indications, distinct from simple measures such as OS-rank order. Although the average NHBS remained fairly constant over time, it was lower than the maximum NHBS (59) and less than the NHBS of other breakthrough approvals (osimertinib for NSCLC: 110, nivolumab for melanoma: 53). If the average value of multiple indications in a drug's indication portfolio is significantly lower than a few blockbuster successes, clinicians could be vulnerable to an availability heuristic and overestimate the value of new AAs, based on anecdotal understanding of prior successes. In an environment of fast-paced AAs and label expansions, clinicians, patients, and payers should develop and deploy rapid value assessment tools in order to sustainably maintain access to high-yield cancer therapies.

Real-world outcomes comparing the use of GLP-1 receptor agonists in patients with endometrial cancer and comorbid obesity: A propensity score matched analysis from a global federated health research network.

5543 Background: Glucagon-like peptide-1 (GLP-1) receptor agonists are widely used for the treatment of type 2 diabetes and obesity. Emerging evidence suggests that these agents may have potential anti-tumor activity as GLP-1 signaling may influence pathways involved in cell proliferation and inflammation. Endometrial cancer is the most common gynecologic malignancy, and obesity represents a well-established, modifiable risk factor that may adversely affect oncologic outcomes through chronic inflammation and metabolic dysregulation. Our study aims to evaluate the impact of GLP-1 receptor agonists on outcomes among patients with endometrial cancer and comorbid obesity. Methods: A retrospective cohort study was conducted using the TriNetX Global Collaborative Network, covering January 2000 to December 2025, encompassing data from 168 global healthcare organizations. Patients aged 18 and above with endometrial cancer and comorbid obesity were identified and then stratified into two groups based on treatment with GLP1 agonists or not. The two groups were then propensity-matched based on age, ethnicity, medical comorbidities, and stage of cancer. Outcomes were assessed for five years from treatment initiation and included overall mortality, myocardial infarction (MI), ischemic stroke; development of peritoneal, lymph node, and visceral metastases; sepsis, pancreatitis, and ED visits. Time-to-event analyses were performed using Kaplan-Meier methods and Cox proportional hazards models. Results: After propensity matching, 6352 patients were included in each cohort with balanced baseline characteristics. Our analysis found that patients with endometrial cancer and comorbid obesity who received GLP-1 agonists had a significantly lower risk of overall mortality (Hazard Ratio [HR]: 0.45, 95% CI: 0.39-0.51, p-value <0.001), development of peritoneal metastases (Risk Difference [RD]: -2.16%, 95% CI: -2.74% to -1.58%, p-value <0.001), nodal metastases (RD: -0.41%, 95% CI: -0.72% to -0.09%, p-value =0.01), visceral metastases (RD: -3.76%, 95% CI: -4.60% to -2.93%, p-value =0.01), and sepsis with and without shock (RD: -1.42%, 95% CI: -2.14% to -0.70%, p-value <0.001) compared to patients who did not receive GLP-1 agonists. There was no statistically significant difference in the rates of MI, ischemic stroke, ED visits, and pancreatitis between the two cohorts. Conclusions: In this large real-world analysis, patients with endometrial cancer and comorbid obesity who received GLP-1 agonists had lower rates of overall mortality; peritoneal, lymph node, and visceral metastases, and sepsis with or without shock. Additional longitudinal cohort studies are imperative to explore the associations between these agents and inform clinical practice guidelines in this patient population.

Inequities in breast cancer mortality with contributing cardiometabolic disorders among U.S. women aged ≥45 years, 1999–2024.

e12704 Background: While breast cancer survival has improved, cardiometabolic disorders including obesity, diabetes, cardiovascular diseases (e.g., hypertension, thrombosis, cardiomyopathy, stroke, heart failure), liver disorders (e.g., non-alcoholic steatohepatitis, hepatic steatosis), and certain gynecological conditions are increasingly recognized as competing risks for mortality. Trends in deaths with breast cancer as the underlying cause and cardiometabolic disorders as contributing causes remain poorly characterized, particularly across demographic and geographic subgroups. This analysis focuses on women ≥45 due to low mortality counts (n = 2,988) in younger women. Methods: We analyzed U.S. mortality data (1999–2024) for women aged ≥45 with breast cancer as the underlying cause and cardiometabolic disorders as contributing causes. Age-adjusted mortality rates (AAMR) per 100,000 were calculated. Joinpoint regression estimated Annual Percent Change (APC) and average annual percent change (AAPC) to evaluate trends stratified by race, U.S. census region, and urbanization. Results: A total of 167,620 deaths occurred among women aged ≥45 years during 1999–2024 (overall AAMR 8.40 per 100,000). Mortality declined modestly from 1999–2011 (APC −0.39%, p = 0.01), reaching an AAMR of 7.51 in 2015, before reversing. The sharpest increase occurred during 2018–2021 (APC +6.88%, p < 0.01), with AAMRs rising from 7.87 in 2018 to 9.58 in 2021 and remaining elevated through 2024 (AAMR 9.90). Substantial racial disparities were observed; Black women consistently exhibited a mortality burden approximately twofold higher than White women. From 2015–2024, mortality among Black women rose significantly (APC +3.56%, p < 0.001), reaching an AAMR of 15.72 in 2024, compared with 9.54 among White women. Regionally, the South showed a sustained increase since 2015 (APC +4.91%, p < 0.001) with the highest 2024 AAMR (11.36). The West also saw a post-2016 rise (APC +5.38%, p < 0.001), while the Northeast and Midwest showed no significant net increase. Nonmetropolitan areas faced the most concerning trends: micropolitan and noncore areas saw increases beginning in 2017 (APC +6.53% and +13.35%, respectively), with 2020 AAMRs of 10.68. Medium metropolitan counties also rose post-2016 (APC +5.47%, p < 0.001). Conversely, large central and fringe metropolitan areas showed long-term declines (AAPC −0.15% and −0.14%, respectively) with no significant net increase. Conclusions: Mortality for breast cancer with cardiometabolic contributions is rising among women ≥45, disproportionately affecting Black women, the South, and non-metropolitan areas, highlighting the need for targeted interventions in cardio-oncology, survivorship, and lifestyle strategies including structured exercise.

Prion-like dynamics of galectin-3 as a driver of colorectal cancer progression: A digital pathology and computational simulation study of supramolecular aggregation.

e15092 Background: Modern oncology is identifying "onco-prions"—proteins that regulate malignancy through biophysical behaviors analogous to neurodegenerative prions. Galectin-3 (Gal-3) is a chimeric lectin characterized by an N-terminal intrinsically disordered domain (IDR) that drives Liquid-Liquid Phase Separation (LLPS), forming dynamic biomolecular condensates and extracellular "lattices" that stabilize oncogenic signaling. While Gal-3 expression is a known prognostic factor, the clinical significance of its physical state—transitioning from soluble monomers to supramolecular aggregates—remains poorly defined in colorectal cancer (CRC). Methods: We integrated computational modeling with clinical validation in a retrospective cohort of 94 CRC tissue samples (Stages I–IV), 10 adenomas and 8 normal colon tissue samples. To characterize the "prion-like" potential of Gal-3, we employed the PLAAC (Prion-Like Amino Acid Composition) algorithm to identify prion-like domains (PrLDs) and utilized the Neurosnap platform for deep learning-based structural simulations of phase separation dynamics. Findings were validated using QuPath-based digital image analysis, applying custom segmentation scripts to quantify Gal-3 aggregation density (nuclear "speckles," cytoplasmic clusters, and extracellular lattices) across CCR progression, including normal, adenomas and tumor stages. Results: Bioinformatic analysis with PLAAC confirmed high-probability PrLDs within the Gal-3 N-terminal domain, which Neurosnap simulations identified as the primary driver of multivalent self-association. In clinical samples, QuPath-assisted quantification revealed a significant correlation between Gal-3 aggregation density and advanced tumor stage (p < 0.05). Conclusions: Galectin-3 operates as a functional prion-like agent in CRC, where its supramolecular aggregation state dictates its pathogenic potential. The integration of PLAAC, Neurosnap, and QuPath provides a robust workflow for characterizing these "onco-prion" dynamics. Our results suggest that the physical aggregation of Gal-3 is a superior biomarker for progression compared to simple protein expression levels. These findings support a shift toward "phase engineering" therapies, such as dominant-negative variants (Gal-3C) or allosteric inhibitors, to dismantle the prion-like scaffolds sustaining the malignant phenotype.

Utility of Baveno VII criteria for endoscopy risk stratification prior to systemic therapy in unresectable hepatocellular carcinoma: A real-world Veterans Health Administration experience.

e23388 Background: Patients with unresectable hepatocellular carcinoma (HCC) are at risk for gastroesophageal varices and bleeding. First-line atezolizumab–bevacizumab therapy requires baseline esophagogastroduodenoscopy (EGD) to exclude high-risk varices, which may delay treatment and increase procedural burden, particularly in resource-limited settings. The Baveno VII criteria noninvasively identify cirrhotic patients at low risk for clinically significant portal hypertension (CSPH) using platelet count and liver stiffness. We hypothesized that Baveno VII criteria could risk-stratify unresectable HCC patients for baseline EGD urgency. Methods: We conducted a retrospective cohort study of adults with unresectable HCC treated at a single Veterans Health Administration center from 2020–2025. Baseline characteristics and events were sourced from the electronic health record. Cirrhosis severity was stratified using Baveno VII criteria as low-risk CSPH, probable CSPH, or definite CSPH. The primary endpoint was variceal bleeding requiring intervention. Secondary endpoints included overall bleeding events and CTCAE-graded bleeding severity. Associations were analyzed using chi-square or Fisher’s exact tests. Results: Ninety-three patients were included. The primary endpoint occurred in three patients (3.2%), all with advanced cirrhosis, including probable CSPH (n = 1) or decompensated cirrhosis (n = 2). No variceal bleeding occurred among patients without cirrhosis or those classified as low risk for CSPH. Cirrhosis severity was not significantly associated with the primary endpoint (p = 0.34). Of the thirty-three patients (35%) who did not undergo baseline EGD, seven experienced a bleeding event with no significant association (p = 0.89). Most of the total nineteen bleeding events were non-variceal and low-grade. Higher-grade bleeding occurred in patients with probable CSPH or decompensated cirrhosis. Race was significantly associated with the primary endpoint (p < 0.001). Conclusions: In this cohort developed from real-world data, variceal bleeding requiring intervention during systemic therapy was uncommon and limited to patients with advanced cirrhosis. No bleeding occurred in patients without cirrhosis or those classified as low risk by Baveno VII criteria. These findings suggest Baveno VII criteria may support a more individualized, resource-conscious approach to baseline EGD before systemic therapy. Prospective multicenter validation is needed. Association between baseline characteristics and bleeding events. Variceal Bleed (p-value) Any Bleed Bleed Severity Age 0.70 0.24 0.60 Race <0.001 0.33 0.19 Child-Pugh 0.23 0.03 0.40 Baveno VII/Cirrhosis Severity 0.34 0.88 0.09 Beta Blocker Use 0.90 0.80 0.67 First Line Therapy Receipt 0.22 0.57 0.24 EGD Receipt 0.93 0.89 0.20

Correction: Evaluation of an acrylic acid hydrogel dosimeter for 3-D dose verification in radiotherapy using MRI

Avalanche leakage current mechanism in vertical GaN devices on foreign substrates

Avalanche breakdown, the most desirable breakdown mode in power devices, is typically characterized by a positive temperature coefficient (PTC) of breakdown voltage (BV) extracted from leakage current–voltage (I–V) characteristics. Here, we report a suppression and reversal of this PTC in avalanche-capable gallium nitride (GaN) p–n diodes with high dislocation densities grown on foreign substrates and elucidate the underlying mechanism. Vertical GaN p–n diodes with optimized edge termination are fabricated on GaN, patterned sapphire, and sapphire substrates, providing a controlled increase in dislocation density. Although all devices exhibit robust avalanche behavior in circuit-based tests, their breakdown I–V characteristics show distinct temperature dependences, evolving from a clear PTC to a negative temperature coefficient (NTC) with increasing dislocation density. This behavior arises from the interplay between avalanche carrier generation and dislocation-assisted carrier transport. Avalanche-generated carriers traverse the depleted drift region via dislocation-assisted variable-range hopping (VRH), which intrinsically exhibits an NTC. The superposition of avalanche generation and VRH transport flattens the breakdown I–V characteristics and yields a weakened PTC or NTC of the extracted BV. These results indicate that an NTC of BV in high-dislocation-density power devices does not preclude avalanche capability, suggesting the limitation of leakage I–V-based avalanche identification.

Carbon dots probes: Structural & functional insights towards emerging food and agricultural applications

Multi-modal data fusion and deep reinforcement learning for dynamic resource scheduling in intelligent manufacturing systems under variable market demand

RPAP3: Structural evolution, chaperone networks, and disease implications of a transcriptional Co-chaperone

Efficacy and safety of cosibelimab 800 mg every 2 weeks for locally advanced cutaneous squamous cell carcinoma: Updated follow-up from a pivotal study.

9585 Background: Cosibelimab, a high-affinity programmed death ligand-1 (PD-L1)–blocking antibody, is approved by the US Food and Drug Administration (FDA) at a recommended dose of 1200 mg every 3 weeks (Q3W) for the treatment of adults with locally advanced or metastatic cutaneous squamous cell carcinoma (laCSCC or mCSCC) who are not candidates for curative radiation or surgery. The CSCC expansion cohort of the pivotal open-label Phase 1 study (NCT03212404) evaluated cosibelimab 800 mg every 2 weeks (Q2W) in patients with laCSCC or mCSCC and 1200 mg Q3W in patients with mCSCC. Results through March 2023 demonstrated robust and sustained responses with cosibelimab 800 mg Q2W. The 800-mg Q2W and 1200-mg Q3W regimens are considered similar based on pharmacokinetic criteria outlined by the FDA for alternative PD-L1–blocking antibody dosing regimens. We present new follow-up data for the 800-mg Q2W laCSCC cohort. Methods: Eligible patients were ≥18 years old and had an Eastern Cooperative Oncology Group performance status of 0 or 1 and unresectable laCSCC not amenable to local therapy. Cosibelimab 800 mg was administered intravenously on Day 1 of each 14-day cycle until complete response, worsening disease progression, toxicity, or clinical deterioration. The primary endpoint was objective response rate (ORR; best overall response: complete or partial) by independent central review per RECIST V1.1; duration of response (DOR) was the key secondary endpoint. Follow-up data presented are from investigator assessment. Other secondary endpoints included safety. Results: By the January 31, 2025, data cutoff, 64 patients with laCSCC received ≥1 dose of cosibelimab 800 mg Q2W; their median (range) age was 77 (51–95) years, and 66% were male. Patients received a median of 29 (2–89) doses over a median duration of 60 (4–184) weeks. The ORR (95% confidence interval) was 50% (37%, 63%), including 17 complete and 15 partial responses; over a median follow-up duration of 31 (1–59) months the median DOR was not reached. Treatment-emergent adverse events (TEAEs) were reported in 61 (95%) patients and considered treatment-related in 50 (78%) patients; none were fatal. The most common AEs were anemia and diarrhea, recorded in 17 (27%) patients each. Grade ≥3 TEAEs were reported in 26 (41%) patients and considered treatment-related in 7 (11%). Hypertransaminasemia, in 2 (3%) patients, was the only serious treatment-related TEAE reported. Immune-related AEs (irAEs) were recorded in 22 (34%) patients; Grade ≥3 irAEs occurred in 1 (2%) patient and were dermatologic in nature (maculo-papular and pruritic rashes) and considered treatment-related. Conclusions: Responses to treatment with cosibelimab 800 mg Q2W remained robust and durable with stable ORR after continued follow-up in a larger cohort of patients with laCSCC, with no new safety signals identified. Clinical trial information: NCT03212404 .

From risk to reward: Redefining the role of GLP-1 receptor agonist in pancreatic cancer primary prevention.

e16449 Background: Pancreatic cancer (PC) carries a dismal 5-year survival of < 12%, and no FDA-approved agents exist for primary prevention. PC risk is strongly linked to obesity, diabetes, and chronic inflammation. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) directly target obesity-driven inflammatory pathways, reduce metabolic stress, and modulate signaling pathways (PI3K/AKT/mTOR) that drive cancer apoptosis. Based on this biological rationale, we conducted the first large-scale real-world study to evaluate the safety and efficacy of GLP-1RA for primary prevention of PC in high-risk patients. Methods: Leveraging a large real-world database (TriNetX), we identified patients aged 18-90 years with ≥1 PC risk factor (tobacco use, obesity, type 2 diabetes, alcohol use disorder, chronic pancreatitis, first-degree relative with PC, genetic susceptibility to PC) based on ICD-10 codes. Patients were stratified by GLP-1RA exposure into two cohorts: Cohort A (GLP-1RA users with ≥ 2 months of prescriptions) and Cohort B (controls). The primary endpoint was incident PC, and the secondary endpoints were adverse events related to GLP-1RAs. Patients with outcomes before the study window were excluded. Propensity score matching adjusted for confounders. Multivariate logistic regression was used to estimate odds ratios for categorical outcomes, and the Cox proportional hazards model was used to estimate hazard ratios for time-to-event outcomes. Subgroup and sensitivity analyses were employed to validate the strength of our results. Results: Median follow-up was 3,397 days for GLP-1RA users and 2,531 days for controls. Demographics were balanced between cohorts (mean age 42 years; 74% White, 15% Black; 72% female). Among 572,330 high-risk patients (286,165 per cohort), PC incidence was significantly lower among GLP-1RA users (0.05%; 134/286,165) than among non-users (0.13%; 362/286,097), with a 2.6-fold reduction in the absolute incidence of PC among high-risk patients (NNT of 1,255). GLP-1RA use was associated with a 71% reduction in PC incidence (HR 0.289; 95% CI 0.237–0.352). GLP-1RA use was associated with an increased risk of gastrointestinal adverse events, including nausea/vomiting (HR 1.108; 95% CI 1.096–1.121), abdominal pain (HR 1.081; 95% CI 1.066–1.097), diarrhea (HR 1.199; 95% CI 1.182–1.217), and constipation (HR 1.159; 95% CI 1.143–1.176). Conclusions: GLP-1RA use significantly reduced PC incidence by 71% in high-risk patients, the strongest real-world signal yet for a pharmacologic prevention of this lethal malignancy. Although the absolute incidence remained low, the magnitude of the risk reduction underscores the transformative potential of GLP-1RA in preventive oncology. These findings address a critical unmet need in pancreatic oncology. Prospective mechanistic and randomized studies are warranted to validate these associations.

Multi-omics analysis from the IMMUcan consortium to identify predictors of chemo-immunotherapy response in early-stage triple-negative breast cancer.

2588 Background: Combining PD-1/PD-L1 immune checkpoint blockade (IO) with neoadjuvant chemotherapy (CT) is now a standard treatment for early-stage triple-negative breast cancer (eTNBC). However, no validated biomarker is currently used to guide patient selection for CT-IO benefit. We aimed to define baseline tumor–immune–stroma ecosystem states that explain differential responses and could inform future biomarker-driven trial designs. Methods: The IMMUcan consortium prospectively enrolled 422 patients with eTNBC treated with CT alone (n=221) or CT-IO (n=201). Pretreatment FFPE biopsies underwent whole-exome sequencing (n=397), RNA sequencing (n=360), multiplex immunofluorescence (2 panels, n=353/368), and imaging mass cytometry (n=341). Response was assessed as pathological complete response (pCR) versus residual disease (n=400). Clinical, genomic, transcriptomic, and spatial features were analyzed for associations with pCR and treatment interaction. Multi-omics integration was performed using Multi-Omics Factor Analysis (MOFA) to derive latent factors and ecosystem states. Results: Baseline clinicopathological characteristics were balanced between groups, while the pCR rate was higher with CT-IO than CT alone (73.1% vs 54.8%; Δ=20.4%; p<0.01). High tumor grade and tumor-infiltrating lymphocytes predicted pCR in both groups without IO-specific predictive value. Among transcriptomic TNBC subtypes, immunomodulatory, mesenchymal (M), and luminal androgen-receptor (LAR) derived the greatest benefit from CT-IO. Spatial profiling identified as key CT-IO responsive feature the enrichment of activated PD-1⁺GZMB⁺CD8⁺ T cells in proximity to tumor cells. MOFA identified biological axes with predictive value, including organized adaptive immunity, effector functions, immune-memory, stromal/angiogenic barrier, and luminal-metabolic lineage. Organized adaptive immunity predicts CT-IO benefit in M and LAR subtypes (OR: 5.42; 95% CI 0.87-33.6; p interaction = 0.048; and OR: 2.97; 95% CI 0.98-8.9; p interaction = 0.036, respectively). Conversely, LAR tumors characterized by luminal–metabolic lineage showed no benefit from CT-IO. Conclusions: Integration of multi-omics data revealed distinct baseline biological ecosystem states capable of identifying patients most likely to benefit from, or exhibit resistance to, CT-IO. These findings provide a comprehensive framework for designing biomarker-driven therapeutic strategies in next-generation neoadjuvant trials for early-stage TNBC.

An LNP-mRNA–based in vivo cell engager as a producer of T macrophage tumor triads in situ for targeted anti-solid tumor therapies.

2547 Background: The solid tumor microenvironment promotes infiltration of myeloid cells, with tumor-associated macrophages as the most abundant innate immune population. A recent study showed tumor-reactive T cells enriched in functional clusters with tumor or antigen-presenting cells, and T cells expanded from these clusters ex vivo exhibit enhanced tumor-killing activity. Bispecific and trispecific antibodies have emerged to link T cells with tumor and myeloid cells, improving antitumor immunity. However, their clinical use is limited by structural complexity, manufacturing challenges, and immunogenicity risks. Here, we developed a “live-cell engager” by harnessing and engineering macrophages in vivo to form T-macrophage-tumor triads, activating both T cells and macrophages near tumor cells to boost antitumor responses. Methods: We engineered macrophages to express a single-chain variable fragment (scFv) targeting CD3 (αCD3), a tumor-specific chimeric antigen receptor (CAR) and CD80 on their surface. Each was encoded by separate mRNAs encapsulated within one lipid nanoparticle formulation (LNP-C-3-80). Mouse bone marrow-derived macrophages (BMDMs) were used ex vivo to evaluate T cell activation and antitumor function. The efficacy of LNP-C-3-80 was further tested in immunocompetent syngeneic models including ID8 ovarian cancer and Hepa1-6 hepatocellular carcinoma. Results: In vitro, BMDMs engineered with LNP-3-80 effectively engaged T cells with macrophages, significantly boosting T cell activation compared to LNP-3 alone, as shown by increased CD69 expression and elevated secretion of IFN-γ, IL-2, and Granzyme B. Furthermore, BMDMs engineered with LNP-C-3-80 were able to engage both T cells and tumor cells, and when co-cultured with T cells, showed tsuperior antitumor activity over control groups (LNP-3-80, LNP-3, LNP-GFP). The LNP-delivered αCD3 and CD80 mRNAs were mainly expressed on macrophages but not on T cells, facilitating macrophage-T cell-tumor cell cluster formation. This proximity enabled mutual activation and synergistic tumor cell killing. In vivo, LNP-C-3-80 achieved nearly complete tumor regression and strong prevention of recurrence in ovarian and liver cancer models after intraperitoneal or intravenous administration. Toxicity studies in healthy C57BL/6J mice showed no organ damage or adverse serum biochemical changes, indicating a favorable safety profile. Conclusions: In summary, in vivo macrophage engineering promotes the formation of T cell–macrophage–tumor cell clusters, reducing immune cell–cancer cell distance and concurrently activating T cells. LNP-C-3-80 shows significantly enhanced antitumor efficacy in syngeneic solid tumor models with minimal toxicity. These results highlight LNP-C-3-80’s strong potential as a promising therapeutic strategy against diverse solid tumors.

GP regimen chronomodulated induction chemotherapy combined with concurrent chemoradiotherapy for locally advanced nasopharyngeal carcinoma: A phase III, multicenter, randomized controlled study.

TPS6137 Background: Radiotherapy is the primary treatment for nasopharyngeal carcinoma (NPC). Currently, induction chemotherapy combined with concurrent chemoradiotherapy has become a Category I recommendation for NPC treatment. Studies show that the GP regimen is one of the most effective induction chemotherapy protocols for locally advanced NPC. However, the incidence of grade 3-4 acute toxicities with the GP regimen is as high as 75.7%, severely affecting patient treatment compliance. To reduce toxicity while maintaining efficacy, this study focuses on a "chronomodulated chemotherapy" strategy, which adjusts drug administration timing based on biological rhythms to minimize damage to normal tissues. Methods: Clinical Data: 1. Sample Size Calculation: This study is a randomized controlled trial with a non-inferiority design. Based on a 3-year recurrence-free survival rate of 85.3% in the conventional group, a non-inferiority margin (δ) of 10%, a one-sided α of 0.025, a power (1-β) of 0.8, equal group sizes, and accounting for a 10% dropout rate, the calculated total sample size is 434 subjects. 2. Inclusion Criteria: Treatment-naïve patients with Stage III or IVa non-keratinizing nasopharyngeal carcinoma (UICC/AJCC 8th edition). 3. Exclusion Criteria: History of prior malignancy, prior radiotherapy, or presence of other severe diseases. 4. Withdrawal Criteria: Occurrence of severe adverse events or intolerable toxicities during the study. Treatment Protocol: 1. Induction Chemotherapy Phase: Experimental Group: Gemcitabine 1000 mg/m², IV infusion over 30 min on days 1 & 8, administered between 08:00-09:30; Cisplatin 80 mg/m², continuous IV infusion (civ) on day 1, delivered evenly over 12 h from 10:00 to 22:00. Repeated every 3 weeks (Q3W) for 3 cycles. Control Group: Gemcitabine 1000 mg/m², IV infusion over 30 min on days 1 & 8; Cisplatin 80 mg/m², IV infusion over 2-3 h on day 1. Q3W for 3 cycles. 2. Concurrent Chemoradiotherapy Phase: Concurrent Cisplatin Regimen: Experimental Group: Cisplatin 100 mg/m², civ on day 1, delivered evenly over 12 h from 10:00 to 22:00. Q3W for 3 courses. Control Group: Cisplatin 100 mg/m², IV infusion over 2-3 h on day 1. Q3W for 3 courses. 3. Radiotherapy: Intensity-Modulated Radiation Therapy (IMRT) was used, with a total dose of 69.96-72.6 Gy delivered in 33 fractions. Innovation: Therapeutic efficacy for nasopharyngeal carcinoma has reached a plateau, making “toxicity reduction” a key research focus. The toxicity-reducing approach of chronomodulated chemotherapy holds promising application prospects for locally advanced nasopharyngeal carcinoma. Current Status: To date, 122 patients have been enrolled. Clinical trial information: ChiCTR2400086032.

Investigating mechanisms of resistance to immune checkpoint inhibitors (ICIs) in renal cell carcinoma (RCC).

4545 Background: Though RCC often exhibits substantial CD8 T-cell infiltration, CD8 abundance does not predict clinical benefit from ICIs. We sought to elucidate tumor microenvironment features that uncouple CD8 presence from ICI response and to evaluate strategies to overcome resistance mechanisms. Methods: We performed single-cell RNA sequencing on 70 RCC tumors (59 clear-cell; 11 non–clear-cell), including samples obtained during systemic therapy (ICI-based therapy, n = 46; others, n = 24). Best response to ICI-based therapy was assessed according to RECIST: 18 responders (R; CR/PR) and 11 non-responders (NR; PD). Tumor-reactive TCRs were inferred using VDJdive. Non-negative matrix factorization identified gene programs associated with NR. Bulk RNA-seq data from landmark clinical trials (HCRN, JAVELIN Renal 101, CheckMate 9ER, COSMIC-313) were analyzed using Scissor (Sun et al. Nat Biotechnol 2022) to map PFS-associated bulk signals to single-cell states in our reference dataset, with summarized enrichment as an observed-to-expected (O/E) ratio. Ligand–receptor (NicheNet) and pathway analyses implicated upstream drivers. Functional validation was performed using primary human T cells cultured under normoxia or hypoxia, with or without TGFβ1. A TGFβ activation assay was used to evaluate rescue of T-cell function by the clinical-stage latent TGFβ1-selective inhibitor SRK-181. Results: We profiled 443,337 high-quality cells including 86,865 CD8 T cells. In NR tumors, a subset with high tumor-specific T-cell abundance (5/11, 45.5%) was enriched for an exhausted CD8 state with a tissue-resident memory program (Tex-rm; p < 0.01). Across multiple ICI-treated clinical trial cohorts, we identified Tex-rm cells as most associated with shorter PFS (O/E 1.34–1.82). Ligand–receptor and pathway analyses implicated enhanced TGFβ signaling in Tex-rm compared with other Tex. In vitro, hypoxia plus TGFβ1 increased a Tex-rm–like phenotype (CD69⁺CD103⁺PD-1⁺TIM-3⁺) relative to controls (p < 0.05 for each). TGFβ inhibited T-cell IFNγ production, while treatment with the TGFβ inhibitor SRK-181 restored it (p < 0.05). Conclusions: We identified a Tex-rm program associated with ICI intrinsic resistance in RCC despite high abundance of tumor-specific T cells. Functional data support a role for TGFβ signaling in promoting this dysfunctional state, suggesting that inhibiting TGFβ1 may help restore T-cell function and overcome ICI resistance in some non-responders.

Benchmarking histology foundation models to identify Prosigna high-risk breast cancer patients.

e12572 Background: Currently, a variety of commercial molecular tests, including MammaPrint, Oncotype DX, and Prosigna, are used to guide adjuvant therapy decisions for early-stage breast cancer patients with hormone receptor-positive/HER2-negative phenotypes. Despite their effectiveness, these tests present challenges in terms of complexity, turnaround time, and cost, motivating the development of AI-based digital pathology tools as alternative prognostic methods. Methods: We benchmarked two self-supervised histology foundation models (UNI-v1 and Prov-GigaPath) for prediction of Prosigna-derived recurrence risk (ROR-PT) from paired H&E whole-slide images. Training and testing were performed using more than 7 million image tiles (224 × 224 pixels) derived from 534 breast cancer patients collected at the University Hospital of North Norway. Both models were fine-tuned using end-to-end task adaptation with slide-level supervision to classify patients into high-risk versus intermediate/low-risk groups. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC) and standard classification metrics. Results: In the overall cohort, molecular subtypes were distributed as follows: luminal A 65% (n = 347), luminal B 32% (n = 173), basal 2% (n = 8), and HER2-enriched 1% (n = 6). The Prosigna ROR-PT scores were dichotomized into low-risk and high-risk groups, comprising 427 (80%) and 107 (20%) patients, respectively. In the hold-out test set (n = 107, 20%), the UNI and Prov-GigaPath models achieved sensitivities of 0.82 and 0.73, specificities of 0.70 and 0.78, and AUCs of 0.79 and 0.81, respectively. While Prov-GigaPath demonstrated slightly higher precision (PPV), the UNI model showed higher sensitivity, suggesting a broader identification of high-risk patients (Table). In addition, model ensembling did not show performance gains over the best unimodal model. Furthermore, a strong correlation (Pearson’s r = 0.65, p < 0.001) was observed between the predictions of both foundation models and the continuous Prosigna ROR scores. Conclusions: The findings of this initial study demonstrate that AI-driven analysis of routine H&E images represents a promising, cost-effective, and rapid surrogate to Prosigna for prediction of recurrence risk in breast cancer. These results will be further validated in independent external cohorts. Model AUC (CI, 95%) Sensitivity Specificity PPV NPV UNI 0.79 (0.65-0.89) 0.82 0.70 0.42 0.94 Prov-GigaPath 0.81 (0.76-0.96) 0.73 0.78 0.46 0.92