Browse Articles

Disitamab vedotin (RC48), tislelizumab, and S-1 as first-line therapy for HER2-overexpressing advanced gastric or gastroesophageal junction adenocarcinoma (GC/GEJC): Updated survival and exploratory biomarker results from the RCTS trial.

4065 Background: Anti-PD1 antibody has significantly improved survival in patients with HER2-overexpressing and PD-L1 combined positive score (CPS) ≥1 GC/GEJC when added to trastuzumab and chemotherapy. Recent trials revealed that HER2-targeted antibody-drug conjugates combined with anti-PD1 antibody, have also shown promising efficacy in this population. We report updated survival, safety, and exploratory biomarker results of RC48 combined with tislelizumab and the oral fluoropyrimidine S-1 as first-line therapy for patients with HER2-overexpressing GC/GEJC. Methods: This single-arm, multicenter clinical trial enrolled patients with unresectable or metastatic HER2-overexpressing (IHC 3+ or 2+, regardless of FISH status) GC/GEJC. Patients received RC48 (2.5 mg/kg), tislelizumab (200 mg), and S-1 (40–60 mg BID for 14 days) every 3 weeks until disease progression or intolerable toxicity. The primary endpoint was objective response rate (ORR). Secondary endpoints were progression-free survival (PFS), overall survival (OS) and safety. Exploratory analyses included dynamic peripheral T-cell receptor (TCR) sequencing and longitudinal circulating tumor DNA/cell-free DNA (cfDNA) profiling. Results: 57 patients were enrolled, 71.9% HER2 IHC 3+, 17.5% IHC 2+/FISH+, and 10.5% IHC 2+/FISH-; 45.6% had CPS ≥1. Until Dec 31, 2025, the median follow-up was 28.2 months. In the ITT population, the confirmed ORR was 89.5% (95% CI: 78.5–96.0%), including 8.8% complete responses; disease control rate was 98.2%. Median PFS was 13.8 months (95% CI: 10.3–24.0), and median OS was 31.9 months (95% CI: 22.1–not reached). In the CPS ≥1 and CPS < 1 subgroups, ORRs were 92.3% (95% CI: 74.9–99.1%) and 87.1% (95% CI: 70.2–96.4%), with median PFS 16.7 vs 10.0 months and median OS 31.9 vs 25.4 months, respectively. Grade ≥3 treatment-related adverse events occurred in 64.9%, mainly decreased white blood cell count (29.8%), decreased neutrophil count (12.3%), anemia (10.5%), and peripheral motor neuropathy (10.5%). In patients with paired TCR samples, higher treatment-induced clonal expansion score was associated with improved PFS (HR 0.34; P = 0.017) and OS (HR 0.33; P = 0.037), with particularly prognostic discrimination in the CPS < 1 subgroup. In patients with longitudinal cfDNA, cfDNA tumor fraction increases preceded imaging-defined progression in most patients who progressed. Conclusions: The combination of RC48, tislelizumab and S-1 as a first-line therapy shows encouraging response rates and updated survival with manageable safety in HER2-overexpressing GC/GEJC, including patients with CPS < 1. TCR clonal expansion may serve as a robust biomarker of antitumor immune activation, independent of PD-L1 expression levels. Clinical trial information: NCT05586061 .

Trajectories of circulating epithelial tumor cells for early prostate cancer detection.

e17118 Background: Prostate cancer is a major cause of cancer-related mortality among older men, highlighting the need for reliable and minimally invasive early detection strategies. Prostate-specific antigen (PSA)–based screening remains controversial due to limited mortality benefit, the risk of overdiagnosis, and subsequent overtreatment of indolent disease. In clinical practice, patients often retain limited information after consultations, and inconsistent risk communication by physicians can hinder informed shared decision-making. Serial assessment of circulating epithelial tumor cells (CETCs/CTCs) may represent a non-invasive approach to improve risk stratification in men at increased risk of prostate cancer. Methods: CETCs/CTCs were quantified in peripheral blood samples from men aged 50–85 years enrolled in a screening cohort (n = 35) and from patients undergoing PSMA-PET imaging for suspected prostate cancer (n = 49). PSA levels were measured at the time of PSMA-PET, and serial CETC/CTC trajectories were analyzed in relation to PSMA-PET findings. Diagnostic performance was evaluated using receiver operating characteristic (ROC) analysis and survival outcomes. Results: PSA levels did not significantly differ between patients with PSMA-PET–positive and PSMA-PET–negative findings (p = 0.94). In contrast, CETC/CTC counts were significantly higher in PSMA-PET–positive patients compared with individuals in the screening cohort (p < 0.001). ROC analysis identified an optimal cut-off value of 450 CETCs/CTCs per mL of blood, yielding a sensitivity of 0.6 and a specificity of 0.7. Longitudinal analysis demonstrated distinct CETC/CTC trajectories: increasing counts were observed in 11 of 13 patients (90%) who subsequently developed PSMA-PET–positive findings, whereas only 2 of 21 patients (8%) with decreasing CETC/CTC counts were PSMA-PET positive. Kaplan–Meier analysis revealed a significantly reduced PSMA-PET–free survival in patients with rising CETC/CTC levels (p < 0.001; hazard ratio 9.48). Conclusions: Serial monitoring of CETCs/CTCs provides dynamic and minimally invasive information on prostate cancer activity. Increasing CETC/CTC trajectories were strongly associated with PSMA-PET positivity, supporting their potential role as early biomarkers for prostate cancer detection, progression assessment, and treatment monitoring.

Deficiencies in oncology marketing applications leading to FDA rejections: An assessment of complete response letters.

e23048 Background: The U.S. Food and Drug Administration (FDA) issues Complete Response Letters (CRLs) to notify sponsors of the agency’s decision not to approve drug marketing applications. The CRL describes the deficiencies in drug development that justified FDA’s decision. Until recently, CRLs were not publicly available. In 2025, FDA published CRLs of marketing applications submitted between 2020 and 2024, including those that were approved later. We reviewed oncology-related CRLs to assess the deficiencies impacting approvals of anti-cancer drugs. Methods: We identified oncology-related CRLs for original and supplement New Drug Applications (NDAs) and Biologics License Applications (BLAs). From each CRL, we extracted data on drug indications; CRL date; and the deficiencies that justified FDA’s decisions, categorized as: Chemistry, Manufacturing, and Controls (CMC); manufacturing facilities; clinical inspections; non-clinical; clinical pharmacology; microbiology; trial design; clinical efficacy; and clinical safety. We consulted the Drugs@FDA website to identify applications that were approved later, with respective approval dates. For unapproved applications, we consulted the sponsors’ website to supplement information not available in the CRL. We descriptively report the proportion of CRLs citing each deficiency type. For applications that were later approved, we report the time intervals from CRL date to approval date. Results: Of 45 oncology-related CRLs (25 for NDAs and 20 for BLAs); 24 (53%) addressed single tumor types; and 18 (40%) addressed monotherapy indications. The number and proportion of CRLs with each deficiency type were: manufacturing facilities (33, 73%); CMC (28, 62%); clinical efficacy (8, 18%); clinical pharmacology (7, 16%); clinical safety (6, 13%); trial design and microbiology (5, 11% each); clinical inspections (4, 9%); and non-clinical (3, 7%). Of the 28 (62%) applications that were approved later, the median time from CRL to approval was 22.8 months (IQR: 30.6). The table reports the median times to approval based on deficiency type. Conclusions: Manufacturing deficiencies were the most common reasons for FDA’s rejections of oncology marketing applications. Deficiencies leading to CRLs are associated with substantial delays in time to market, with clinical efficacy and safety deficiencies leading to potentially longer delays. The findings highlight the value of risk mitigating strategies in specific areas of drug development. Time to approval from receipt of CRL (N=28). Deficiency Type Median (range), months Manufacturing Facility (n=20) 16.8 (8.6, 120.8) CMC (n=21) 18.2 (8.6, 120.8) Clinical Pharmacology (n=6) 21.6 (8.6, 120.8) Microbiology (n=3) 24.9 (15.4, 35.3) Trial Design (n=4) 30.3 (15.3, 120.8) Clinical Efficacy (n=7) 39.6 (9.1, 120.8) Clinical Safety (n=4) 56.3 (9.1, 120.8) Any Deficiency (n=28) 22.8 (8.6, 120.8)

Age-specific risks of leukemia in Li Fraumeni syndrome in children, adolescents, young adults, and adults.

e22653 Background: Leukemia is a well-recognized risk in Li Fraumeni Syndrome (LFS), but age-specific risks in males and females estimated using modern genetic epidemiology analyses within the world’s largest study of LFS families have not been reported. Methods: Pedigree data from 1,610 families with LFS from the LiFT UP study comprised of 47,393 family members were analyzed by Kaplan Meier methods and modified segregation analysis using a Bayesian framework. All families included had germline TP53 mutations. Age-specific and lifetime risks were calculated. Standardized incidence ratios were calculated to estimate relative risk using observed leukemia cases in pedigrees vs. SEER data for expected counts. Missing ages in family members were imputed. To minimize ascertainment bias, we excluded information from probands who were ascertained because of a leukemia diagnosis (conditioning on the proband). Results: Leukemia was reported in 165 females, 191 males (356 individuals) from a total of 289 families. Among these, leukemia was diagnosed in 30 female probands and 21 male probands, who were excluded as part of the ascertainment adjustment. The youngest age-at-diagnosis was <1 in females and 1 in males and the average age of diagnosis was 27.5 years in females and 27.4 years in males. Among individuals with recorded diagnosis age, 50 females and 46 males with leukemia were TP53 carriers, and 5 females and 10 males with leukemia were non-carriers, with the remaining 81 females and 93 males untested. The lifetime risk of leukemia is 9.9% in TP53+ females, compared to a SEER risk of 1.6% (Standardized Incidence Ratio (SIR)=11.7, confidence interval (CI)=8.8-15.4) and 16.4% in TP53+ males compared to a SEER risk of 2.9% (SIR = 12.7, CI=9.4-16.8). Age-specific SIRs are high especially in children, adolescents, and young adults, with the highest SIR in females aged 10-14 (SIR=23.2, CI=14.5-35.1) and highest SIR in males aged 15-19 (SIR=23.6 CI=14.4-36.4). Cumulative leukemia risks in children & adolescents (≤19 years) old are 1.5% in females and 2.1% in males. Conclusions: TP53 carriers with LFS have elevated risk of leukemia, especially in young carriers compared to population risks. Pediatric, adolescent, and young adults are at the highest relative risk of leukemia in LFS, although risks remain elevated throughout the lifetime of TP53 carriers in both females and males.

A three-dimensional multimode lumped-element resonator for collective spin manipulation and dispersive readout

We report a three-dimensional lumped-element multimode microwave resonator that enables homogeneous collective manipulation and dispersive readout of a macroscopic spin ensemble. By exploiting geometric symmetry, two anti-symmetric modes with strongly suppressed crosstalk are engineered to spatially overlap and couple to the same ensemble at distinct frequencies. Using negatively charged nitrogen-vacancy (NV−) centers in diamond at 28 mK, we observe a collective coupling strength of 5.0 MHz, which is in the vicinity of strong-coupling regime, and demonstrate nondestructive dispersive readout via a detuned mode. The compact design, tunable coupling, and high field homogeneity make this resonator a versatile device for hybrid spin–photon systems and multimode solid-state quantum technologies.

Deep eutectic solvents in biomedical applications: A mini review

Programmable Electrothermal Quad‐Functional Metamaterials for Decoupled Multi‐Field Control

ABSTRACT The extension of multifunctional metamaterials to coupled electro‐thermal systems holds the potential for transformative applications in adaptive camouflage, energy transport, and microelectronics. However, strong electron‐phonon interactions and the intrinsic coupling between voltage‐current and temperature‐heat flux fields limit existing designs to one or two functionalities. Here, we report a field‐line‐guided coordinate transformation strategy for constructing an electrothermal quad‐functional metamaterial (ETQFM) capable of simultaneously controlling current, voltage, heat flux, and temperature fields within a single architecture. By mapping prescribed field trajectories into spatially distributed material‐geometry configurations, the transformation decouples electrical and thermal responses, enabling independent realization of cloaking, concentrator, and rotation across both domains. As a proof of concept, six distinct prototypes are fabricated by 3D printing of laser‐sintered metal powders and experimentally validate, each exhibiting four independent functions with high fidelity and agreement with simulations. The strategy also allows functional switching in symmetric architectures by rotating boundary conditions and remains robust against low‐power thermal perturbations, establishing a generalizable framework for multi‐physics field manipulation.

The price of a breath: The economic and mortality burden of checkpoint inhibitor pneumonitis (2016–2020).

12142 Background: The landscape of oncologic toxicity has drastically changed because of the rapid introduction of Immune Checkpoint Inhibitor (ICI) indications, ranging from the treatment of melanoma to thoracic cancers. Immune-related adverse events (irAEs) are a novel range of toxicities brought about by the rapid growth of ICI. The overall national burden of severe inpatient toxicities is still unknown, despite the well-described outpatient care of irAEs. We measured the prevalence, mortality, and financial impact of severe irAEs necessitating hospitalization using the Nationwide Inpatient Sample (NIS). Methods: We evaluated adult hospitalizations secondary to major irAEs (pneumonitis, colitis, hepatitis, and endocrinopathies) using ICD-10-CM codes for a retrospective cohort analysis of the National Inpatient Sample database from 2016 to 2020. Only patient with cancer diagnosis associated complications were included to reduce confounders. Statistical evaluations and outcomes were generated using STATA version 19. The use of mechanical ventilation, total hospital charges, and in-hospital mortality were the major outcomes. Temporal trends and mortality factors were examined using survey-weighted regression models. Results: We identified a weighted national cohort of 109,225 irAE-related hospitalizations from 2016 to 2020. Epidemiology: Immune-mediated Colitis was the dominant driver of volume (54.6%) for inpatient admission, followed by Pneumonitis (22.2%) and Hepatitis (18.2%). The proportion of admissions attributable to Pneumonitis was increased from 21.1% in 2016 to 24.0% in 2020 (p < 0.001), coincident with expanded ICI use in lung cancer. Outcomes & Acuity: Pneumonitis was associated with the longest mean length of stay (LOS) of 6.9 days and the highest raw mortality (7.8%) compared to Colitis (1.6%). In multivariable analysis, Pneumonitis was the strongest independent predictor of in-hospital mortality (aOR 4.98; 95% CI 4.12–6.02; p < 0.001)relative to Colitis. Conversely, female sex was associated with a significant survival advantage (aOR 0.76; p = 0.001). Economic Burden: Pneumonitis was the most resource-intensive toxicity (Mean Charge: $74,298 vs. $43,831 for Colitis). The aggregate national hospital bill for irAEs rose from $1.14 Billion in 2016 to a peak of $1.75 Billion in 2020. Conclusions: Pneumonitis is the primary cause of death and economic expense, whereas colitis is the primary cause of irAE hospitalizations. Between 2016 and 2020, the total cost of hospitalization in the United States almost doubled, reaching approximately $1.75 billion per year. Health systems need to prepare for an increase in high-acuity, resource-intensive pulmonary toxicity as ICI indications spread into the adjuvant environment.

Temporal changes in the algebraic connectivity of collateral sensitivity maps.

e15007 Background: Drug resistance remains one of the greatest challenges in oncology, often limiting the long-term efficacy of cancer therapies. Collateral sensitivity, the phenomenon where resistance to one therapy alters susceptibility to others, offers a potential strategy to overcome treatment failure in cancer. Nevertheless, majority of the studies characterize this phenomenon at endpoints naively assuming complete sampling of the underlying evolutionary mechanisms, agnostic to the evolutionary trajectories. Thus, temporal dynamics of collateral sensitivity and/or how much observational variance due to under sampling exists remains largely unexplored. Methods: We developed a spatial cellular automata model to study temporal dynamics on a latent continuous space. Individual cells occupy a two-dimensional grid and undergo stochastic proliferation, death, and mutation. Each cell carries a continuous genotype vector that maps to multi-drug response phenotypes through a non-linear mapping determining the “fitness” of individual clones allowing selective pressure to shape the evolutionary trajectories. The phenotype values (collateral responses induced by the drug of interest) for each cell were tracked over time and the drug-drug mutual-information matrices were calculated across the population of cells at each timepoint. To characterize the change in drug-drug associations during resistance evolution, we examined the phenotype-covariance network and associated spectral properties of the graph Laplacians. Results: We performed the simulations under different scenarios: drug/no-drug, single/multi -peaked landscape. As expected, the diversity of the genotype space was higher under no-drug condition as opposed to treatment condition. Interestingly however, the fiedler value (algebraic connectivity) of the MI matrices showed a non-linear pattern, initially increasing and subsequently decreasing in the drug treatment setting. In contrast under no drug treatment the fiedler value continued to increase. Conclusions: This framework provides a flexible, quantitative approach to studying collateral sensitivity in continuous tumor phenotypes, moving beyond static or discrete models. By examining temporal changes in algebraic connectivity, we observe distinct patterns under selection and no selection conditions, highlighting that collateral sensitivity maps are not fixed properties but evolve over time. These results suggest that quantifying collateral sensitivity depends on both evolutionary context and sampling of the underlying phenotype space, including factors such as treatment pressure and landscape structure. Overall, this model establishes a foundation for future work aimed at understanding how evolving phenotypic organization impacts multi-drug response and adaptive therapy strategies.

Neoadjuvant chemotherapy with NALIRIFOX for locally advanced colorectal cancer: A multicenter, single-arm, phase II trial.

e15638 Background: Dual-agent regimens such as CAPOX and FOLFOX have demonstrated significant tumor downstaging efficacy in the neoadjuvant treatment of locally advanced colorectal cancer (LACC). In contrast, the triple-agent regimen FOLFOXIRI, while potentially superior in metastatic settings, has limited clinical utility owing to its poor tolerability profile.. Liposomal irinotecan, an optimized formulation of irinotecan, reduces adverse events through enhanced drug delivery, thereby improving the feasibility of triple-drug combination therapy. This trial aimed to evaluate the efficacy and safety of the NALIRIFOX regimen as neoadjuvant therapy for LACC. Methods: Patients with advanced colon cancer (cT4N1-2M0) or upper/middle rectal cancer (cT3-4N0-2M0, MRF+/EMVI+) who were candidates for R0 resection with anal preservation were enrolled. Eligible patients received NALIRIFOX (liposomal irinotecan 50 mg/m², oxaliplatin 60 mg/m², leucovorin 400 mg/m², fluorouracil 2400 mg/m²) every two weeks for 4–8 cycles, followed by surgical resection. The primary endpoint was the pathological complete response (pCR) rate. Key secondary endpoints included downstaging rate, R0 resection rate, tumor regression grade (TRG), perioperative complications, 2-year recurrence rate, disease-free survival (DFS), and safety. Results: From December 2024 to June 2025, 30 patients were enrolled. One patient with posterior uterine wall metastasis was inadvertently included and postoperatively confirmed to have malignant disease. The median age was 58 years (range 37–75), and 27 patients (90.0%) had stage III disease. Of these, 28 patients (93.3%) completed 4–8 cycles of chemotherapy, and 27 patients (90.0%) underwent surgery, achieving a 100% R0 resection rate. One 75-year-old patient discontinued treatment due to an adverse event (AE), while two declined surgery for personal reasons. No postoperative complications were reported. The pCR rate was 7.4% (2/27), and 77.8% (21/27) of patients achieved TRG 1–2. Tumor downstaging was observed in 85.2% (T stage), 74.1% (N stage), and 63.0% (TNM stage) of patients. Treatment-emergent adverse events (TEAEs) of any grade occurred in all 30 patients, with 4 (13.3%) experiencing grade ≥3 events; all four completed at least 4 treatment cycles. The most common grade ≥3 TEAEs were intestinal obstruction (6.7%), leukopenia (3.3%), and neutropenia (6.7%). Conclusions: The triple-agent neoadjuvant regimen containing liposomal irinotecan, oxaliplatin, and 5-fluorouracil/leucovorin demonstrated promising efficacy in patients with LACC, reflected in substantial tumor downstaging. The regimen also showed a favorable and manageable safety profile, supporting further investigation as a potential neoadjuvant treatment option for LACC. Clinical trial information: ChiCTR2400092630.

A four-subtype model of cervical cancer based on <i>PIK3CA</i> mutation, <i>STK11</i> mutation or deletion, and <i>YAP1</i> amplification: Multiple PI3K pathway inhibitors as suppressors of growth and cooperators with HPV-directed immunotherapy.

e17522 Background: Cervical cancer is driven by human papillomavirus (HPV) infection yet lacks molecularly defined subtypes and approved targeted therapies. We defined molecular subtypes of cervical cancer and identify actionable therapeutic vulnerabilities. Methods: Whole-exome sequencing on cervical tumors from Guatemala and Venezuela, with validation in TCGA, AACR Genie, and Caris cohorts. Functional studies were conducted in cervical cancer cell lines using targeted inhibitors, immune co-culture assays, and proliferation analyses. Results: We identified four somatic mutation subtypes of cervical cancer: (I) No mutation ( PIK3CA / STK11 wild-type (wt) without YAP1 amplification); (II) PIK3CA -mutant, wt for STK11 , YAP1 ; (III) YAP1 -amplified, wt for PIK3CA , STK11 ; and (IV) STK11 -mutated or deleted, wt for PIK3CA , YAP1 . STK11 alterations were significantly enriched in adenocarcinomas compared with squamous cell carcinomas (23%, X 2 =4.4; p = 0.0037) and were confirmed in the AACR Project GENIE (p = 0.035) and Caris cohorts (p = 2.5e-06). YAP1 amplification and STK11 alterations were associated with younger age at diagnosis and poorer overall survival compared to subtype I. STK11 alterations also correlated with inferior outcomes following immunotherapy compared to subtype I. PI3Kα-specific inhibitors (alpelisib and inavolisib) selectively suppressed proliferation in PIK3CA -mutant cervical cancer cell lines but not in PIK3CA wt cell lines. However, the pan-AKT inhibitor capivasertib showed variable activity, inhibiting some but not all PIK3CA -mutated cell lines and the SiHa (PIK3CA wt) cell line. Alpelisib reduced the expression of HPV16 E7, CD274/PD-L1 , YAP1 , and EGFR exclusively in PIK3CA -mutant cell lines. We tested for synergy between PI3Kα inhibition and immune-directed therapy. In HPV16-positive, HLA-A2, PIK3CA -mutant cells, alpelisib enhanced antigen-specific T cell–mediated cytotoxicity, with maximal effect observed following drug pretreatment and washout prior to T cell exposure. Conclusions: Our findings define clinically relevant molecular subtypes of cervical cancer and identify PIK3CA -mutant tumors as sensitive to PI3Kα inhibition. In published clinical studies, 7 patients with PIK3CA -mutant cervical cancer treated with alpelisib achieved partial response or stable disease. Combining PI3K inhibitors with immunotherapy represents a promising strategy for advanced cervical cancer. Distribution of cervical cancer subtypes by histology. Type I WT Type II PIK3CA Type III YAP1 Type IV STK11 SCC AD SCC AD SCC AD SCC AD Sum 1425 641 724 241 174 23 105 83 % Adeno in type 31% 25% 12% 44% Total (%) 2068 (60%) 965 (28%) 197 (5.7%) 188 (5.5%) Pooled data from Guatemala, TCGA, AACR Genie, and Caris datasets. SCC, squamous cell carcinoma; AD, adenocarcinoma.

Delivering complex pancreatic cancer care locally: Five-year FOLFIRINOX outcomes from Northern Ireland.

e16358 Background: Combination chemotherapy (FOLFIRINOX) improves survival in metastatic pancreatic cancer, with PRODIGE4/ACCORD11 reporting a median overall survival (OS) of 11.1 months. Real-world outcomes across all stages, particularly in regional centres, remain under-reported. This study evaluates five-year outcomes for patients receiving FOLFIRINOX in routine practice and compares them with historic trial benchmarks. Methods: A retrospective cohort analysis was conducted of 32 consecutive pancreatic cancer patients treated with FOLFIRINOX between 2020–2025 at a regional UK cancer centre. Demographic, clinical, and treatment variables were collected, including Eastern Cooperative Oncology Group (ECOG) performance status, stage, treatment intent, dose modifications, and survival. OS was calculated from diagnosis to death or last follow-up. Outcomes were descriptively compared with PRODIGE4/ACCORD11. Results: Median age was 67 years (range 49–78); 56% were female. ECOG performance status was favourable, with &gt; 85% ECOG 0–1. Stage distribution was: stage 1–2 (22%), stage 3 (41%), stage 4 (37%). Dose modifications were frequent across the cohort. All patients received FOLFIRINOX as first-line systemic therapy. Several early-stage and locally advanced patients also underwent surgery, Stereotactic Ablative Body Radiotherapy (SABR), or Concurrent Chemo-Radiotherapy(CCRT) as part of multimodality management. Across the entire cohort, median OS was approximately 13 months. In the stage 4 subgroup, median OS was ~10 months, closely aligning with the 11.1 months OS reported in PRODIGE4/ACCORD11 despite broader eligibility and real-world heterogeneity. Patients with stage 1–3 disease receiving multimodality therapy demonstrated prolonged survival, with several alive at last follow-up. Conclusions: FOLFIRINOX delivered real-world survival outcomes comparable to historic phase III data, even within a regional centre and across a broader patient population. Integration of surgery, SABR, and CCRT supported favourable outcomes in earlier-stage disease. These findings highlight the feasibility of delivering complex pancreatic cancer care outside tertiary centres and support ongoing pathway optimisation and trial integration. Key clinical characteristics and outcomes of pancreatic cancer patients. Variable Result Median Age (Range) 67 (49-78) Gender 56% F , 44% M ECOG Performance status &gt; 85% ECOG 0-1 Stage at diagnosis Stage 1-2: 22%Stage 3: 41%Stage 4: 37% Treatment intent Adjuvant: 22%Locally advanced: 41%Metastatic/Palliative: 37% Definitive local therapy Surgery: 12 patientsSABR: 7 patientsCCRT: 4 patientsNone: 9 patients Dose modifications 90% Median OS (all patients) ~ 13 months Median OS (stage 4) ~ 10 months Comparison with PRODIGE4/ACCORD11 Trial OS: 11.1months (metastatic)Real-World OS comparable despite broader eligibility

Adjuvant durvalumab or tislelizumab combined with S-1 for resected high-risk biliary tract cancers in China: A multicenter, two-cohort, single-arm, open-label, phase 2 trial preliminary analysis.

e16275 Background: Patients with biliary tract cancer (BTC) have a poor prognosis and high postoperative recurrence rates. Immunochemotherapy has been approved as a first-line treatment for advanced BTC. We aimed to explore the efficacy and safety of durvalumab (an anti-PD-L1 immunotherapy) or tislelizumab (an anti-PD-1 immunotherapy) combined with S-1 chemotherapy as a postoperative adjuvant treatment in patients at high risk of BTC recurrence. Methods: This was a multicentre, two-cohort, single-arm, phase 2 trial in BTC patients with R0-resected at high-risk recurrence. The study was being conducted at 2 tertiary hospitals in China. Patients received eight 21-day cycles of durvalumab or tislelizumab plus S-1. Following S-1 discontinuation, maintenance therapy with durvalumab or tislelizumab alone was administered in 28-day cycles for a maximum of 6 cycles. Propensity score matching (PSM) (1:2) analysis was used to balance potential bias. The primary endpoint was recurrence-free survival (RFS), and the secondary endpoints included overall survival (OS) and safety. Survival was assessed in the full analysis set (FAS) and safety was assessed in the safety set. This study is registered with ClinicalTrials.gov (NCT06490107) and is currently ongoing but no longer recruiting new patients. Results: Between Jun 4, 2024, and Jan 10, 2026, 54 patients with BTC were enrolled: 40 in the durvalumab plus S-1 cohort and 14 in the tislelizumab plus S-1 cohort. At the data cutoff of Jan 12, 2026, the median follow-up period was 12.75 months (95% confidence interval [CI], 8.25-17.24) and 14 (26%) patients had recurrence. The median RFS and OS of all patients were not reached. In FAS, 1-year RFS rate was 62.2% (95% CI, 48.0-80.7) and 1-year OS rate was 92.9% (95% CI, 83.5-100). 1-year RFS rate was 62.9% (95% CI, 47.3-83.5) and 1-year OS rate was 94.1% (95% CI, 83.6-100) with durvalumab plus S-1. 1-year RFS rate was 59.3% (95% CI, 32.2-100) and 1-year OS rate was 87.5% (95% CI, 67.3-100) with tislelizumab plus S-1. No significant differences in RFS and OS were observed between two cohorts, both before (p = 0.733) and after PSM (p = 0.939). The most common treatment-related adverse events (TRAEs) were neutropenia (41%), increased AST/ALT (33%), rash (30%), nausea (28%), pigmentation (28%). Grade 3 TRAEs occurred in 15 (28%) patients, including neutropenia (n = 5), rash (n = 3), pruritus (n = 2), increased AST/ALT (n = 2), elevated bilirubin concentrations (n = 2), diarrhoea (n = 1). No grade 4 TRAEs or treatment-related deaths occurred. Conclusions: Adjuvant therapy with S-1 combined with either durvalumab or tislelizumab demonstrates promising efficacy with an acceptable safety profile in patients with resected BTC at high risk recurrence. Further validation of this immunochemotherapy regimen is warranted in larger, multicentre trials. Clinical trial information: NCT06490107 .

Myxoid/round cell liposarcoma misdiagnosis incidence, etiology, and importance of NY-ESO-1 testing: Retrospective analysis of the Tempus database.

11542 Background: Liposarcoma is composed of multiple unique sarcoma subtypes, specifically myxoid/round cell liposarcoma (MRCLS), well differentiated and dedifferentiated liposarcoma (WD LPS and DD LPS), and pleomorphic liposarcoma. These are genetically distinct entities with different treatment paradigms. MRCLS is driven by a translocation between DDIT3 and FUS or EWSR1, detectable in 100% of MRCLS. WD and DD LPS are characterized by amplification of MDM2 and CKD4, and pleomorphic LPS is genetically complex. Recently, new treatments for liposarcoma are specific to the LPS type (CDK4/6 inhibitors; CTA directed engineered T Cells) which highlights the importance of an accurate liposarcoma subtype diagnosis. Methods: A retrospective analysis of the tempus RNA database for DDIT3 translocations and NY-ESO-1 (CTAG1B) expression in MRCLS was conducted. 101 patients (pts) with MRCLS were identified; 27 pts lacked Next Generation Sequence data for tumor samples leaving 74 pts that were selected for analysis. RNA was sequenced with either the Tempus RS or the RS.v2 assays. Patients' tumor samples with tumor purity &lt;30% were excluded from assessment of CTAG1B expression. Gene expression values in this analysis were normalized by transcripts-per-million (TPM) transformed to log2(TPM+1) units. The threshold of 2 (log2(TPM+1)) was set as the cut off of true positive for CTAG1B expression. Results: The DDIT3 fusion was identified in 65/74 (88%) patients. In 9/74 (12%) patients, there was no DDIT3 fusion detected. Out of the 65 patients that were positive for the MRCLS translocation 60/65 (92%) were DDIT3-FUS, and 5/65 (8%) were DDIT3-EWSR1. The MRCLS DDIT3 fusion-negative samples were 5/9 (56%) positive (pos) for MDM2 amplification and CDK4 amplification. The Tempus tumor of origin algorithm predicted all DDIT3 fusion-pos MRCLS samples to be soft tissue sarcoma (where predictions were made; 53 of 65 pts) and the DDIT3 fusion-neg samples to be fibrosarcoma in one case and liposarcoma in 7 cases. The last fusion-neg case did not have a tumor of origin prediction. The median expression of CTAG1B was 7.14 TPM in all patients, 7.2 TPM (range 3.1 to 8.7) in DDIT3-FUS samples, 7.5 TPM (range 5.7 to 7.7) in DDIT3-EWSR1 samples, and 0.04 (range 0.0 to 3.7) in DDIT3 fusion-neg samples. There was no difference in CTAG1B expression by biopsy site, metastatic status, stage, tumor purity, or center of diagnosis. Conclusions: It is of critical importance to accurately determine the liposarcoma subtype which influences the selection of new targeted treatments. This analysis highlights a 12% rate of misdiagnosis of MRCLS and suggests that WD/DD LPS can be misdiagnosed as MRCLS. Additionally, this study predicts that 100% of MRCLS with confirmed DDIT3 translocation will be NY-ESO-1 (CTAG1B) pos which highlights the utility of NY-ESO-1 as a diagnostic marker for NY-ESO-1 directed therapies.

The HER2 paradox in ovarian clear cell carcinoma: Expression and intrinsic resistance to trastuzumab deruxtecan (T-DXd).

5563 Background: Trastuzumab deruxtecan (T-DXd) shows potent activity in HER2-expressing ovarian cancer (OC). However, efficacy across histologic subtypes remains under-characterized. Specifically, data on T-DXd response in ovarian clear cell carcinoma (OCCC), a distinct chemo-resistant entity, are lacking. This study aimed to evaluate T-DXd efficacy in recurrent OC across different histologic subtypes. Methods: We retrospectively analyzed 59 patients with recurrent OC who received T-DXd treatment between 2020 and 2025 at Yonsei Cancer Center. HER2 status (IHC), objective response rate (ORR), progression-free survival (PFS), duration of response, and safety were assessed by histology. Results: Patients included HGSC (n=38), Mucinous (n=9), OCCC (n=10), and Endometrioid (n=2). Median age was 55; median prior lines of therapy was 3. Notably, 100% of OCCC, mucinous, and endometrioid tumors were HER2 2+/3+, vs. 89.5% in HGSC. Despite high HER2 expression, outcomes diverged. ORR was 55.3% (26/47) for HGSC/Mucinous vs. 0% (0/12) for OCCC/Endometrioid ( P &lt;0.001). Specifically, HGSC achieved 57.9% ORR (2 CRs) vs. 0% in OCCC. PFS was significantly shorter in the OCCC/Endometrioid group (median 1.2 vs 6.9 months; HR 18.3; P &lt;0.0001). OCCC patients showed rapid progression. Regarding safety, among 54 evaluable patients, 36 (66.7%) experienced Grade 3-4 adverse events, and 5 cases (9.3%) of ILD occurred; however, no new safety signals or treatment-related deaths were reported. Conclusions: T-DXd showed robust efficacy in recurrent HGSC and mucinous OC but no response in OCCC and endometrioid OC. Thus, HER2 expression alone does not guarantee T-DXd response in OCCC, possibly due to its chemo-refractory nature. Mechanisms of T-DXd resistance in OCCC warrant investigation. Tumor response. Histology HGSC (n=38) Mucinous (n=9) Endometrioid (n=2) Clear cell (n=10) Objective response 1 (n, %) 22 (58%) 4 (44%) 0 (0%) 0 (0%) CR 2 (5%) 0 (0%) 0 (0%) 0 (0%) PR 20 (53%) 4 (44%) 0 (0%) 0 (0%) SD 14 (37%) 4 (44%) 0 (0%) 0 (0%) PD 2 (5%) 1 (11%) 2 (100%) 10 (100%) 1 P =0.0006.

Health-related quality of life (HRQoL) with neoadjuvant and adjuvant (neoadj-adj) enfortumab vedotin (EV) plus pembrolizumab (pembro) in participants (pts) with muscle-invasive bladder cancer (MIBC) who are cisplatin ineligible: Phase 3 KEYNOTE-905 study.

4510 Background: In KEYNOTE-905/EV-303 (NCT03924895), neoadj-adj EV + pembro and radical cystectomy (RC) + pelvic lymph node dissection (PLND) showed significant and meaningful improvements in EFS, OS, and pathologic complete response vs RC + PLND in pts with MIBC who were ineligible for or declined cisplatin (cis) therapy. Prespecified exploratory HRQoL outcomes are reported. Methods: Pts with MIBC who were ineligible/declined cis were randomized 1:1 to neoadj-adj EV + pembro and RC + PLND or RC + PLND (control). HRQoL was analyzed in pts who completed ≥1 PRO assessment. EQ-5D-5L and Functional Assessment of Cancer Therapy (FACT)–Bladder-Cystectomy (Bl-Cys) with FACT-general (FACT-G) and Bladder Cancer Index (BCI) were assessed predose on D1 of neoadjuvant cycles 1 (baseline [BL]), 2, and 3; pre- and post-surgery; and predose on D1 of adjuvant cycles 1, 2, 4, 8, and 12 in the EV + pembro arm. In the control arm, assessments were at pre- (BL) and post-surgery. In both treatment (Tx) arms, assessments were also done at Tx discontinuation, Q12W for ≤2y, and Q24W thereafter. End points included mean (95% CI) change from BL to post-surgery wk 18 in FACT-G score; FACT–Bl-Cys scores (Bl-Cys subscale and Trial Outcome Index [TOI]); BCI urinary, bowel, and sexual scores; and EQ-5D-5L visual analog scale (VAS). Results: At post-surgery wk 18, completion and compliance rates for all assessments were ≥65% in the EV + pembro arm and &gt;86% in the control arm. Mean changes from BL to post-surgery wk 18 were similar between arms for FACT-G total and FACT–Bl-Cys subscale scores; FACT–Bl-Cys TOI score; BCI urinary, bowel, and sexual scores; and EQ-5D-5L VAS score (Table). BCI sexual and bowel score diminution was observed in both arms. For all PRO assessments, mean changes from BL over time were similar between Tx. Conclusions: The addition of neoadj-adj EV + pembro did not decrease HRQoL 18 wk post-surgery relative to RC + PLND alone; BCI bowel and sexual domains worsened in both arms, consistent with prior reports of RC impact. Given the superior efficacy and manageable safety, these results support the benefit/risk profile of EV + pembro and RC + PLND as Tx for pts with MIBC ineligible for or declining cis. Clinical trial information: NCT03924895 . Mean change from BL to post-surgery wk 18, (95% CI), n EV + pembro Control FACT-G total score −2.73 (−6.22 to 0.75)n = 102 −2.84 (−6.11 to 0.43)n = 75 FACT–Bl-Cys (subscale/symptom index) total score 1.31 (−0.70 to 3.32)n = 102 1.85 (−0.59 to 4.29)n = 75 FACT–Bl-Cys TOI −1.89 (−5.61 to 1.84)n = 102 −0.58 (−4.35 to 3.20)n = 75 BCI urinary −2.03 (−6.12 to 2.07)n = 101 −0.05 (−5.90 to 5.80)n = 70 BCI bowel −4.75 (−8.73 to −0.77)n = 101 −4.27 (−7.93 to −0.61)n = 70 BCI sexual −15.46 (−20.53 to −10.39)n = 94 −17.88 (−24.36 to −11.40)n = 61 EQ-5D-5L VAS −2.52 (−7.23 to −2.19)n = 102 −0.39 (−5.14 to 4.36)n = 75

The oncology data paradigm in practice: A feasibility study of structured data capture with minimal support (stage one of a multi-phase implementation).

e13660 Background: We previously showed success with collection of structured data elements within the electronic health record using the minimum common oncology data elements (henceforth, mCODE tool) supported by soft-stop close visit validation (CVV) and staff support interventions in a limited cohort. Since the additional interventions are resource-intensive and difficult to apply across an entire department, we therefore implemented a department-wide roll-out supported by the least resource-intensive interventions while monitoring uptake. Methods: The mCODE tool collects five data elements: stage, disease status, performance status, intent of treatment, and intent to change therapy. The Department of Oncology providers received education on the mCODE tool and were asked to implement use at a faculty meeting; soft-stop CVV was enabled at that time. The percentage of data captured for each data element was collected from the notes of providers who did not participate in prior pilots for the 4 months post-implementation. mCODE tool usage was defined as entry of at least one data element except stage (as stage has long been a structured data element in Epic, can be entered independently of the mCODE tool, and many providers were entering stage prior to mCODE tool introduction). Results: 86 MDs and 20 APPs for a total of 106 providers were included, resulting in 14,973 visits and 8,745 unique patients. 73.6% of providers (n =78) utilized the mCODE tool. Of those who utilized the tool, median overall data capture across the four data elements was 24.0%. 22.6% of providers (n =24) achieved &gt; 80% overall data capture and 32.1% (n =34) had over 50% overall data capture despite minimal intervention. 75.6% of MDs (n = 65) utilized the tool with a median data capture of 16.3% compared to 65% of APPs (n = 13) with a median data capture of 90.3%. Rates of data capture for individual data elements were similar. Conclusions: Use of soft-stop CVV appears to drive utilization of the mCODE tool but is insufficient to promote &gt; 80% data capture for most providers without more intervention. Supportive interventions such as individualized assistance with incorporation into existing templates and directed communication from leadership remain necessary to promote use of structured data elements and will be implemented in sequence to understand how we achieve the target data capture ( &gt; 80%) at scale.

Randomized, Placebo-Controlled Trial of B-Cell Depletion for Prevention of Corticosteroid-Requiring Chronic Graft-Versus-Host Disease

PURPOSE Chronic graft-versus-host disease (cGVHD) is a multisystem alloimmune disorder associated with abnormal B-cell biology and aberrant antibody responses. As B-cell–directed therapy can effectively treat established cGVHD, we tested whether prophylactic B-cell depletion could prevent the development of corticosteroid-requiring cGVHD following allogeneic transplantation. METHODS We performed a randomized, placebo-controlled, and blinded trial comparing four doses of the B-cell–depleting antibody obinutuzumab (1,000 mg once on days 90, 180, 270, and 365 after transplantation) with placebo in transplant recipients receiving tacrolimus-based GVHD prevention at higher risk of cGVHD. The primary end point was the 1-year incidence of corticosteroid-requiring cGVHD. We measured antibody responses against Y chromosome–encoded minor histocompatibility (H-Y) antigens and correlated their occurrence with corticosteroid-requiring cGVHD incidence. RESULTS One hundred seventy-eight participants were analyzed. The prophylactic administration of obinutuzumab resulted in profound B-cell depletion, a significant reduction in the incidence of steroid-requiring cGVHD at 1 year (13.3% v 35.2%; P = .0005), and an improvement in immunosuppression-free, relapse-free survival (48% v 34% at 2 years; P = .02). Neutropenia was more common in the obinutuzumab arm, but nonrelapse mortality was not different. In participants without preformed H-Y antibodies at the time of study intervention, obinutuzumab resulted in the most significant reduction in steroid-requiring cGVHD at 12 months (8.6%) compared with obinutuzumab participants with H-Y antibodies (40%) or placebo participants regardless of antibody status (41% with antibodies, 57% without antibodies). CONCLUSION In allogeneic transplant recipients at higher risk of cGVHD, early B-cell depletion results in a significant reduction in the incidence of corticosteroid-requiring cGVHD.

The electronic frailty index used to identify high-risk older adults receiving chimeric antigen receptor T-cell therapy.

11009 Background: Chimeric antigen receptor T-cell therapy (CAR-T) has transformed outcomes in relapsed/refractory hematologic malignancies and can be effective in older adults. Frailty, a syndrome of vulnerability common in older adults, is associated with poor outcomes but difficult to measure in clinical practice. The relationship between frailty and CAR-T outcomes is poorly understood. Methods: We conducted a retrospective analysis of 260 patients age 65+ years receiving commercial CAR-T at Massachusetts General Brigham (MGB) for multiple myeloma and B cell non-Hodgkin lymphoma. We abstracted patient characteristics and clinical outcomes from the electronic health record (EHR). Using EHR data from the Research Patient Data Registry, we calculated the MGB-eFI, a validated electronic frailty score based upon a deficit accumulation approach to frailty. The MGB-eFI assesses 31 age-related health deficits and yields a score ranging 0-1 (&lt; 0.1 is robust, 0.1-0.2 is pre-frail, 0.2-0.3 is frail, and &gt; 0.3 is very frail). We then evaluated associations of the eFI (as a categorical variable of robust/prefrail versus frail/very frail) with cytokine release syndrome (CRS), immune cell-associated neurotoxicity syndrome (ICANS), overall survival (OS), and event-free survival (EFS) using univariate and multivariable logistic and Cox regression models controlling for covariates. Results: Among 260 patients (median age 73 years [range 65-91], 45% female), the most common CAR-T products used were liso-cel (35%), tisa-cel (18%), ide-cel (18%), cilta-cel (12%), and axi-cel (10%). With the eFI, 10% of patients were robust, 25% prefrail, 28% frail, and 37% very frail. In univariate analyses, frail/very frail was associated with worse OS (hazard ratio [HR] 1.66, p = 0.013), worse EFS (HR 1.78, p = 0.001), higher rates of ICANS (odds ratio [OR] 1.99, p = 0.013), and higher rates of grade 3+ ICANS (OR 4.49, p = 0.003). Frailty was not associated with risk of CRS. In multivariable analyses controlling for age, sex, CAR-T product, lactate dehydrogenase, ferritin, number of prior therapies, and bridging therapy use, frail/very frail by eFI remained associated with worse EFS (HR 1.69, p = 0.005) but not with OS (HR 1.37, p = 0.150). Frail/very frail was numerically associated with increased ICANS, but this did not meet statistical significance (OR = 1.71, p = 0.090). In a multivariable analysis controlling for age, CAR-T product, and ferritin (due to event sample size), frail/very frail was associated with higher rates of grade 3+ ICANS (OR 3.75, p = 0.017). Conclusions: Frailty measured using an automated eFI is associated with worse EFS and higher rates of high-grade ICANS, despite controlling for other clinical and patient factors. The eFI holds promise as a tool that if integrated into the EHR can help clinicians with risk stratification and CAR-T decision-making for older adults with hematologic malignancies.

Impact of body composition on prognosis, treatment toxicity, and surgical complications in locally advanced gastric cancer.

e16128 Background: Weight loss and skeletal muscle wasting are frequent in cancer and may influence treatment tolerance and outcomes. Computed tomography (CT) based body composition analysis at the L3 vertebra is an accurate method to quantify skeletal muscle in routine oncology care. Methods: We performed a multicenter retrospective cohort study including 202 adults with locally advanced gastric cancer (LA-GC), stage IB – III, treated in four Portuguese hospitals, between January 2020 and December 2022. Skeletal muscle area was assessed on baseline CT at the L3 vertebra level, using Data Analysis Facilitation Suite (DAFS) software, and skeletal muscle index (SMI) was calculated. Sarcopenia was defined as an SMI below sex specific mean. The primary objective was to evaluate the association of sarcopenia with FLOT Chemotherapy Dose Limiting Toxicities (DLTs). Secondary objectives were to evaluate the association between sarcopenia and Postoperative Complications after gastrectomy, Relapse Free Survival (RFS) and Overall Survival (OS). Results: Mean age was 69 years, 65% had ECOG PS 0, 53% received FLOT chemotherapy protocol. Mean SMI was 49,6 cm²/m² in males and 40,9 cm²/m² in females. SMI correlated positively, though moderately, with BMI (p &lt; 0.01; r = 0.424). Sarcopenia was associated with higher risk of DLTs (p = 0.021; OR 2.56, 95% CI 1.15-5.73) and postoperative complications (p = 0.024; OR 2.16, 95% CI 1.11-4.21). Although sarcopenia was not significantly associated with RFS (p = 0.186) or OS (p = 0.168) at 30 months follow up, a numerical difference was observed (64% vs 56% of patients did not relapse and 74% vs 63% were alive, for non sarcopenic vs sarcopenic patients). Conclusions: Sarcopenia significantly increased the risk of chemotherapy toxicity and postoperative complications in LA-GC. The effect of sarcopenia on OS and RFS was not statistically significant in this locally advanced disease cohort, possibly because surgical intervention modifies the disease trajectory. The lack of standardized CT based sarcopenia cut-offs remains a major barrier to clinical implementation. Nevertheless, early detection of sarcopenia using CT imaging may represent a valuable and precise tool to identify patients requiring closer monitoring, chemotherapy dose optimization, and early referral for multimodal interventions, including pharmacologic therapy, structured exercise programs, and personalized nutritional support.