Browse Articles

Minimally invasive vs open pancreaticoduodenectomy for periampullary tumors: A systematic review and updated meta-analysis.

e16474 Background: Laparoscopic pancreaticoduodenectomy (LPD) is increasingly being performed for periampullary tumors. However, its safety and efficacy compared to open pancreaticoduodenectomy (OPD) remain debated. LPD is a complex procedure that requiresadvanced laparoscopic skills. This meta-analysis aims to compare the outcomes of LPD and OPD. Methods: We systematically searched electronic databases, including PubMed, Embase, and the Cochrane Library, from inception through January 2026 to compare LPD with OPD and analyze the following outcomes: mortality, morbidity, major complications, delayed gastric emptying, postoperative pancreatic fistula, postoperative pancreatic hemorrhage, R0 resection rate, readmission, operative time, blood loss, hospital stay, and lymph nodes harvested. Forest plots were generated, and data were analyzed using a random-effects model to calculate pooled risk ratios (RRs) and mean differences (MDs) with 95% confidence intervals (CIs); heterogeneity was assessed using I² statistics. Results: A total of nine studies (n = 1,491) were included, comparing LPD (n = 752) with OPD (n = 739). LPD was associated with a significantly shorter hospital stay (MD 1.01, 95% CI 0.27–1.74, P = 0.0071, I² = 31.4%) and less blood loss (MD −97.23, 95% CI −149.15 to −45.31, P = 0.0002, I² = 87.7%) compared with OPD. However, LPD was associated with a longer operative time (MD −53.22, 95% CI −89.70 to −16.74, P = 0.0042, I² = 90.7%). There were no significant differences in mortality (RR 1.09, 95% CI 0.49–2.41, P = 0.8293, I² = 12.3%), morbidity (RR 1.01, 95% CI 0.89–1.14, P = 0.9144, I² = 0.0%), major complications (RR 1.13, 95% CI 0.83–1.55, P = 0.4366, I² = 50.1%), delayed gastric emptying (RR 1.09, 95% CI 0.79–1.49, P = 0.6028, I² = 47.2%), postoperative pancreatic fistula (RR 1.03, 95% CI 0.77–1.33, P = 0.9303, I² = 0.0%), postoperative pancreatic hemorrhage (RR 1.03, 95% CI 0.76–1.41, P = 0.8300, I² = 0.0%), R0 resection rate (RR 1.01, 95% CI 0.99–1.03, P = 0.5658, I² = 1.9%), readmission (RR 0.97, 95% CI 0.68–1.39, P = 0.8712, I² = 0.0%), or lymph nodes harvested (MD 0.17, 95% CI −9.97 to 10.31, P = 0.9741, I² = 0.0%). Conclusions: Laparoscopic pancreaticoduodenectomy is associated with a shorter hospital stay and less blood loss but a longer operative time compared with OPD, with similar postoperative outcomes, suggesting that LPD is a safe and feasible approach for periampullary tumors in selected patients.

A phase I study to assess the safety, tolerability, pharmacokinetics, and preliminary antitumor activity of KH617 for patients with advanced solid tumors (including diffuse gliomas of adulthood).

2046 Background: Outcomes in patients (pts) with recurrent glioblastoma (rGBM) are poor, with historical median overall survival (mOS) of 7-8 months. KH617 is a sesquiterpene compound that exhibits a wide spectrum of anti-cancer properties, has low toxicity, and is capable of traversing the blood-brain barrier. It can inhibit the growth, proliferation, and migration of tumors via multiple mechanisms, mitigate the malignant phenotype of tumor cells, and facilitate the transdifferentiation of tumor cells into normal cells. This is a phase I study of KH617 in pts with advanced solid tumors (including diffuse gliomas of adulthood). Here, we report the results of the dose escalation stage. Methods: Pts with histologically confirmed advanced solid tumors (including diffuse gliomas of adulthood) that were no longer responding to standard therapies were enrolled. The Bayesian Optimal Interval (BOIN) design was used for dose escalation, having explored in 7 successive cohorts (5-60mg/kg). Results: As of Dec, 30 2025, 32 pts were enrolled across 7 dose levels, including 23 pts with rGBM, 7 with non-glioblastoma, and 1 each with adenocarcinoma and spinal cord ependymoma. The median age was 51 years, and 19 (59.4%) pts were male. No dose-limiting toxicity (DLT) or treatment-related serious adverse events (TRSAEs) were observed, and the maximum tolerated dose (MTD) was not reached. The most frequently reported (≥10%) treatment-related serious adverse events (TRAEs) were white blood cell count decreased, neutrophil count decreased, lymphocyte count decreased, abnormal ECG T wave, and sinus bradycardia. Only 1(3.13%) pt reported grade ≥3 adverse event. Efficacy signals were observed especially in the 23 rGBM pts, who were administered KH617 across 7 doses (n = 1, 1, 3, 2, 8, 7, 1 in 5, 10, 17, 25, 33, 45, 60mg/kg cohorts, respectively), the mOS was 16.07 months. At the recommended phase 2 dose (RP2D) of 33mg/kg, 8 rGBM pts were included in the efficacy evaluation, objective response rate (ORR) was 12.5%, median progression-free survival (PFS) was 1.84 months, mOS exceeded 21 months (3 died and 5 were still alive). In the pharmacokinetic (PK) study, C max no longer increased with escalating doses from 25 mg/kg to 45 mg/kg, indicating that the peak plasma concentration had already reached the maximum plateau level under the current dosing. Combined the efficacy data with the PK data, 33 mg/kg was confirmed as the RP2D. Conclusions: KH617 has shown significantly superior anti-tumor activity compared to the historical mOS in pts with rGBM, while also being well-tolerated. Clinical trial information: CTR20223286.

Clinical features and outcomes of EBER-negative nasopharyngeal carcinoma: A large cohort study across China.

e18144 Background: The current TNM staging system and treatment guidelines are primarily based on EBER-positive nasopharyngeal carcinoma (NPC) patients, leaving a lack of high-level evidence to guide treatment decisions for EBER-negative patients. This study aims to clarify its clinical features and inform future therapeutic strategies. Methods: We conducted a multicenter study across seven centers in China. Key inclusion criterion required NPC patients to have undergone EBER in situ hybridization. The Kaplan-Meier curves were used for survival analysis, and inter-group differences were compared using the log-rank test. Multivariable Cox proportional hazards models evaluated the impact of clinical and pathological factors on survival outcomes. Propensity score matching (PSM) was performed at a 1:4 ratio between EBER-negative and EBER-positive cohorts, with matching based on clinical characteristics. A prognostic risk stratification model was performed. Results: This study analyzed 248 (20%) EBER-negative and 992 (80%) EBER-positive patients. Among these, 876 were male (71%). The EBER-negative NPC was significantly associated with the keratinizing subtype (26.2% vs. 0.7%, P < 0.001). Cox multivariable regression analysis of 1,240 patients identified EBER as an independent prognostic factor for NPC (HR = 0.49, 95% CI = 0.35-0.70, P < 0.0001). Compared to EBER-positive patients, EBER-negative NPC had significantly inferior 5-year overall survival (79.9% vs. 90.2%), progression-free survival (PFS, 67.6% vs. 77.3%), and locoregional recurrence-free survival (84.3% vs. 92.1%, all P < 0.001). Conversely, EBER-negative patients exhibited a lower risk of distant metastasis, with a lower 5-year distant metastasis-free survival rate (92.3% vs. 86.0%, P = 0.006). Multivariate analysis shows T stage (HR = 1.45, 95% CI = 1.05–1.99, P = 0.02) and hypertension status (HR = 1.68, 95% CI = 1.03–2.75, P = 0.04) as independent prognostic factors for PFS in the EBER-negative cohort. The combined model, which included T stage, hypertension status, and others, had a higher discriminative performance for PFS (AUC, 0.741) compared to hypertension status, T stage alone, or other combination models (0.54, 0.58 and,0.62, respectively). Based on these factors, patients were divided into three risk groups, with 5-year PFS rates of 80.1%, 64.3%, and 52.5%, respectively (p=0.0011). Conclusions: EBER-negative NPC is a distinct, aggressive subtype characterized by keratinizing histology and poorer overall survival, yet it is associated with a lower risk of distant metastasis compared to EBER-positive NPC. A practical risk-stratification model was built. Further studies are needed to confirm these findings and explore potential clinical applications.

ALFA score as a predictor of long-term benefit with frontline brigatinib in <i>ALK</i> + advanced NSCLC: A ctDNA-based composite biomarker from ALTA-1L.

8646 Background: Optimal sequencing of ALK-TKI in 1st-line ALK+ NSCLC remains controversial. Although 3 rd -generation (G) TKI are increasingly used upfront, a subset of patients (pts) may achieve prolonged outcomes with 2 nd G TKI. However, baseline (B) biomarkers to identify these pts are lacking. We evaluated whether a composite clinical-molecular score integrating (B) ctDNA and biological parameters could predict systemic and intracranial outcomes with upfront brigatinib. Methods: Post hoc analysis of pts with ALK+ NSCLC from the ALTA-1L trial with available (B) ctDNA NGS. ALK variants, co-occurring genomic alterations (co-GAs), and clinical/biological variables (PS, albumin, derived neutrophil-to-lymphocyte ratio [dNLR], and metastatic sites) were analyzed. Independent prognostic factors for progression-free survival (PFS) and intracranial PFS (IC-PFS) were identified by multivariable Cox models. A composite score (ALFA), based on the Cox models’ coefficients was developed. Results: Among 124 pts, ctDNA detected ALK variant in 53% (n = 66), mostly EML4::ALK (V1 n = 25, V3 n = 24, V2 n = 7, V5 n = 4). co-GAs were present in 46%, including suppressor gene alt. in 35%. High dNLR was observed in 34% and hypoalbuminemia in 21%. In multivariable analysis, high dNLR (HR 1.67, 95% CI 1.02–2.73; p = 0.04), low albumin (HR 0.34, CI 0.20-0.57; p &lt; 0.001) and suppressor co-GAs (HR 2.24, CI 1.32-3.79; p = 0.003) were independently associated with shorter PFS. Non-V1 variants showed a trend toward inferior PFS (HR 1.73; p = 0.05). For IC-PFS, high dNLR (HR 1.52; p = 0.03), brain metastases (HR 2.98; p &lt; 0.001), and suppressor co-GAs (HR 2.41; p = 0.002) remained independently prognostic. The ALFA score stratified pts into low- (59%), intermediate- (27%), and high-risk (14%) groups, with median PFS of 35.5, 11.1, and 5.5 months (mo.), respectively (p &lt; 0.0001; c-index 0.72). A similar separation was observed for intracranial PFS (IC-PFS; p &lt; 0.0001), with median IC-PFS not reached in the low-risk group, 21.1 mo in the intermediate-risk group, and 5.5 mo. in the high-risk group (c-index 0.72). Overall survival (OS) showed consistent and significant discrimination across ALFA risk groups (log-rank p &lt; 0.0001) with a median OS NR in low-risk, 44.3 mo. in intermediate-risk, and 19.3 mo. in high-risk pts. Conclusions: The ALFA score integrates baseline ctDNA-derived molecular features with routine blood-based parameters to robustly stratify clinical outcomes in ALK + pts treated with frontline brigatinib. This composite score identifies a subset of patients with durable systemic and intracranial benefit from second-generation ALK inhibitors, supporting its potential role as a practical risk stratification tool in routine clinical practice.

Open, laparoscopic, versus robotic surgery for Siewert type II/III adenocarcinoma of esophagogastric junction: A systematic review and Bayesian network meta-analysis.

e16093 Background: The optimal surgical approach for Siewert type II/III adenocarcinoma of the esophagogastric junction (AEG) remains debated, with open, laparoscopic, and robotic techniques each in use. Direct comparative evidence, particularly for robotic gastrectomy, is scarce. This study aimed to comprehensively evaluate the efficacy and safety of these three modalities. Methods: A systematic review and Bayesian network meta-analysis were conducted following PRISMA guidelines. Four databases were searched through November 2025 for studies comparing open, laparoscopic, and robotic surgery for Siewert II/III AEG. Pairwise and network meta-analyses were performed, with outcomes pooled as odds ratios (RR) or mean differences (MD) with 95% confidence (CI) or credible intervals (CrI). Treatment rankings were estimated using the Surface Under the Cumulative Ranking Curve (SUCRA). Results: Fifteen studies (1847 laparoscopic, 1399 open, 56 robotic cases) were included. Pairwise meta-analysis demonstrated laparoscopic surgery offered significant advantages over open surgery: reduced intraoperative blood loss (MD: -92.09 mL, 95% CI: -128.07 to -56.12), shorter hospital stay (MD: -1.98 days, 95% CI: -2.13 to -1.83), fewer overall complications (RR: 0.79, 95% CI: 0.66 to 0.94), and improved 5-year overall survival (RR: 1.33, 95% CI: 1.06 to 1.67). Open surgery maintained superiority in lymph node yield (MD: 1.38 nodes, 95% CI: 0.23 to 2.54). Compared to laparoscopy, robotic surgery was associated with higher lymph node retrieval (MD: 5.39 nodes, 95% CI: 0.36 to 10.41) and lower blood loss (MD: -32.44 mL, 95% CI: -55.16 to -9.72). The Bayesian network meta-analysis provided a comparative hierarchy: Robotic surgery ranked highest (best) for minimizing blood loss (SUCRA 82.5%), reducing overall complications (SUCRA 88.7%), achieving the lowest anastomotic leakage rate (SUCRA 99.7%), and enabling the shortest hospital stay (SUCRA 90.9%). Open surgery ranked highest for thorough lymphadenectomy (SUCRA 98.5%) and shortest operative time (SUCRA 81.6%). Laparoscopic surgery was ranked most favorable for reducing postoperative ileus risk (SUCRA 89.7%). No statistically significant differences in mortality were observed between the three approaches. Conclusions: While open surgery remains the benchmark for lymphadenectomy and operative efficiency, minimally invasive approaches offer distinct benefits. Laparoscopic surgery provides reliable recovery advantages. Robotic surgery emerges as promising for superior perioperative safety, particularly regarding anastomotic integrity, though these findings require validation in larger prospective studies. The surgical approach should be individualized based on tumor factors, expertise, and resources.

GZMK+ T cells in glioblastoma and an activated, non-exhausted phenotype.

e14071 Background: Glioblastoma (GBM) bears a survival estimate below 10% at 5 years, despite surgery and standard chemoradiation. Immunotherapeutic strategies have not improved disease outcomes, primarily due to the brain’s uniquely immunosuppressive tumor microenvironment (TME). Our group formerly described a GZMK + effector T cells subpopulation, enriched in GBM, which has been so far poorly characterized in its function and spatial localization. Methods: By using 27-parameters flow cytometry, we profiled 162.466 tumor infiltrating lymphocytes (TILs) from 8 newly diagnosed GBM patients (pts) surgically treated at our Institution and matched peripheral blood mononuclear cells (PBMC). Formalin-fixed paraffin embedded (FFPE) GBM slides were stained by hematoxylin-eosin (H&amp;E) and imaging mass cytometry (IMC). Tumor regions of interest (7-13 per case) were selected on H&amp;E and mapped to IMC slides (Hyperion Imaging System). A CCR5/CXCR3-based flow cytometry gating strategy was established to enrich GZMK+ CD8+ T cells from PBMC by excluding naïve (CCR7+CD95-), NK(CD56+) and MAIT(CD161 hi ) cells. CD8+ T cells from PBMC were stimulated in vitro with cytokines present in the GBM TME (IL-1β, IL-2, IL-6, IL-12, IL-15 and TNF). Cells were cultured for 7 days and phenotyped by flow cytometry; cell fate and division were tracked using CellTraceViolet labelling. Results: Flow cytometry analyses confirmed the high presence of GZMK+ and GZMK+/GZMB+ double positive CD8+ T cell populations within GBM tissue. In particular, we identified a GZMK+ subpopulation (~40% of CD8+ T cells) displaying a memory, activated, non-exhausted phenotype (CCR5+, CD69+, CD27+, PD1+, TIGIT-, CXCR3+, CCR4+, HLA-DR+). IMC revealed that CD8⁺ T cells were spatially localized in close proximity to activated myeloid cells (IBA1⁺, CD16⁺, CD68⁺, CD163⁺, CD14⁺, HLA-DR⁺), consistent with either microglia or macrophages. GZMK+ CD8+ T cells isolation from PBMCs based on CCR5/CXCR3 co-expression, led to an enrichment of up to 95% of GZMK+ CD8+ T cells. Compared to other CD8+ memory T cell populations, bona fide GZMK+ CD8+ T cells sorted from PBMC exhibited enhanced degranulation capacity (CD107a), higher INFg production and increased polyfunctionality upon stimulation with PMA-Ionomycin or aCD3/aCD28+IL-2. In vitro stimulation with IL-12+IL-15 induced GZMB expression in GZMK+ CD8+ T cells from PBMCs, recapitulating the GZMK+GZMB+ CD8+ TILs phenotype observed in GBM and establishing a robust in vitro model to study GBM-infiltrating T cells. Conclusions: GBM infiltrating GZMK+ CD8+ T cells display a memory, activated, non-exhausted phenotype. In vitro IL-12+IL-15 stimulation induces GZMB expression in GZMK⁺ CD8⁺ T cells, partially recapitulating the phenotype of GBM-infiltrating T cells. Collectively, these findings identify the predominant activated GZMK⁺ CD8⁺ T-cell population as a promising therapeutic target in GBM.

Pre-transplant MRD negativity and TP53 mutation status determine survival after CAR-T therapy bridged to allo-HSCT for refractory/relapse B-ALL.

6556 Background: The feasibility and outcomes of bridging relapsed/refractory (R/R) B-lineage acute lymphoblastic leukemia (B-ALL) patients to allogeneic hematopoietic stem cell transplantation (allo-HSCT) following CAR-T therapy require further investigation. This study evaluated long-term outcomes and prognostic factors in this setting. Methods: We conducted a retrospective analysis of 300 consecutive R/R B-lineage leukemia patients (295 B-ALL, 98.3%) who underwent allo-HSCT after CAR-T therapy. Somatic mutation screening via next-generation sequencing was performed on 179 patients. Univariate Cox regression analyzed factors associated with transplant outcomes. A contemporaneous cohort of 36 transplanted patients without prior CAR-T served as a comparator. Results: Median age at transplant was 14 years (range, 1-68). Prior to HSCT, 220 (73.3%), 75 (25.0%), and 5 (1.7%) patients received CD19-, CD22-, and Mix-CAR-T, respectively; 265 (88.3%) achieved complete remission (CR), and 39 (13.0%) were minimal residual disease (MRD)-positive. The 3-year disease-free survival (DFS), overall survival (OS), cumulative incidence of relapse (CIR), and non-relapse mortality (NRM) for the CAR-T bridging cohort were 64.0% (95%CI, 58.6-69.8%), 74.5% (95%CI, 69.5-79.9%), 28.4% (95%CI, 23.6-34.1%), and 7.6% (95%CI, 5.04-11.3%), respectively. Outcomes did not differ significantly from the non-CAR-T cohort (OS, p=0.54; DFS, p=0.24). Failure to achieve CR pre-HSCT (HR=3.31, p&lt;0.0001) and MRD-positivity (HR=3.69, p&lt;0.0001) were risk factors for relapse, while chronic graft-versus-host disease was protective (HR=0.35, p=0.0002). Pre-HSCT MRD-positivity, fungal infection, and viral cystitis increased NRM risk. TP53 (15.1%), KMT2D (14.5%), NRAS (14.0%), KRAS (13.4%), and ABL1 (9.5%) were the most frequently mutated genes. TP53 mutation was a significant risk factor for inferior DFS (HR=3.37, p&lt;0.0001) and OS (HR=3.56, p&lt;0.0001). GNAS and KDM6A mutations predicted higher CIR, while ETV6 mutation was associated with increased NRM. Conclusions: Bridging to allo-HSCT after CAR-T therapy is a feasible and effective strategy for R/R B-ALL, providing substantial long-term survival. Achieving pre-transplant CR and MRD-negative status is critical for outcome. Somatic mutations, particularly TP53, hold significant prognostic value, aiding in risk stratification and treatment optimization for these patients.

Patient experience and quality of life in a decentralized cancer care model for advanced prostate cancer.

e13611 Background: Decentralized cancer care models may reduce treatment burden for patients with advanced prostate cancer (CaP), particularly those facing mobility, transportation, of scheduling challenges. However, prospective patient-reported experience and quality-of-life data comparing home- versus clinic-based oncology care remain limited. Mayo Clinic Cancer CARE Beyond Walls (CCBW) is a patient-centered program designed to deliver selected standard-of-care cancer therapies in the home with virtual clinical oversight. This observational study evaluated patient-reported experience, feasibility, safety, and quality of life among men with advanced or metastatic CaP receiving care at home. Methods: Men with advanced/metastatic CaP receiving active standard-of-care therapy including hormone-based treatments, oral anticancer therapies, anti-bone resorptive agents, or intravenous chemotherapy who were participating in a CCBW randomized trial were enrolled into a companion observational study. After confirmation of treatment tolerance in clinic, patients received home-based care for 24 weeks (Arm A) or in-clinic care for 8 weeks followed by home-based care for 16 weeks (Arm B). Patient-reported outcomes were collected at baseline and Weeks 8, 16, and 24, using CAHPS-based measures and CCBW-specific surveys assessing overall cancer care experience, patient preference for site of care, comfort with home infusions, functional/symptom assessment and quality of life. Descriptive analyses were performed. Results: Eighteen participants were analyzed (Arm A n = 10; Arm B n = 8). Participants were medically stable, and predominantly insured by Medicare. Survey completion was 100% at baseline and 83% at all follow-up time points. At Weeks 16 and 24 when all respondents had received home-based care, 100% and 92% reported being “quite a bit” or “very much” comfortable receiving treatment at home. Communication with at-home and remote nursing teams was rated highly, and most respondents reported transition, scheduling, and equipment use easy or very easy. At Weeks 16 and 24, 45% and 40% of respondents reported improved quality of life from receiving home-based care, with no reports of worsening. No emergency department visits, hospitalizations, or urgent care visits occurred in either arm. Conclusions: In this prospective, observational study, home-based oncology care delivered through CCBW was feasible, safe, and associated with high patient satisfaction among men with advanced CaP receiving diverse systemic therapies. These findings support further evaluation and expansion of decentralized care integrating home-based treatment with virtual clinical oversight.

A phase 2 study of ipatasertib in combination with pembrolizumab for first-line treatment of recurrent or metastatic squamous cell cancer of the head and neck.

6006 Background: Single agent pembrolizumab in relapsed/metastatic head and neck squamous cell carcinoma (R/M HNSCC) has limited activity. The immunosuppressive tumor microenvironment (TME) includes regulatory T cells (Treg)s and myeloid-derived suppressor cells (MDSC)s, which may contribute to the low responses to anti-PD-1 therapy. Preclinical studies demonstrate that anti-PD-1 antibodies induce Treg activation through the AKT pathway. AKT blockade selectively inhibits the proliferation of human Tregs compared to conventional T cells. Furthermore, inhibiting the AKT pathway limits MDSC infiltration and differentiation while boosting effector T cell function within tumors. Ipatasertib is an oral highly selective small-molecule inhibitor of all three isoforms of AKT. This phase II trial compares the efficacy of combination ipatasertib plus pembrolizumab (I+P) versus pembrolizumab (P) monotherapy in R/M HNSCC. Methods: This is a prospective, two-arm, phase II, multicenter trial for 1 st line treatment of R/M HNSCC. Patients were randomized 1:1 to either Arm 1 - P 200mg on day 1 with I 400mg daily on days 1-14 of 21-day cycles, or Arm 2 – P monotherapy. PD-L1 CPS score ≥1 was required. The primary objective is to compare the PFS between the two arms. Secondary objectives included safety and ORR per RECIST 1.1. Results: As of 1/21/2026, 52 patients were randomized, with 27 enrolled in the I+P arm. The median age was 67 and 77% were male. The primary tumor sites were 46% oral cavity, 38% oropharynx, and 15% larynx. Among pts with oropharynx primary, 75% were p16 positive. PD-L1 CPS score was ≥20 in 60%. In the I+P arm the most common G1-3 treatment-related adverse events (TRAEs) occurring in ≥10% were diarrhea (70.4%), fatigue (44.4%), Nausea (40.7%), AST increase (18.5%), ALT increase (14.8%), and maculopapular rash (14.8%). Only 1 pt had grade 3 diarrhea. In P arm the most common G3 TRAEs were maculopapular rash (29.2%), AST increase (16.7%), and diarrhea (12.5%). There were no G4 or G5 TRAEs. Four patients required dose reduction of ipatasertib, primarily for diarrhea. Ten pts in the I+P arm and 6 pts in P arm required dose interruptions. One patient in each arm discontinued due to adverse events. At the data cutoff 4 pts remain on treatment in I+P arm and 2 remain on P arm. The ORR in I+P arm and P arm was 41% and 17% respectively. The CR rate was 15% in I+P arm and 4.2% in P arm. The DCR (CR+PR+SD) was 70% in I+P arm and 42% in P arm. With a median follow up of 7.0 months, the PFS in I+P arm is 8.1 months (95% CI: 4 – NA) and in P arm is 6.2 months (95% CI 1.9 – 13.5). Conclusions: I+P demonstrated an acceptable safety profile and shows promising clinical activity in R/M HNSCC. Clinical trial information: NCT05172258 .

Artificial intelligence-driven insights into silver-doped zinc ferrite (SI=25): Advancing biofilm control, drug delivery, and tissue engineering for cancer therapy

Stable Analog Weight Programming in Single‐Crystalline van der Waals Ferroelectric Transistors for Reliable Computing‐in‐Memory

ABSTRACT Emerging ferroelectric non‐volatile memories are revolutionizing von Neumann architectures by providing efficient hardware for both AI training and inference. However, as ferroelectric dimensions scale toward the nanoscale, reliable modulation is hindered by interfacial degradation and phase instability, leading to synaptic weight drift and computational inaccuracies. Here, a high‐performance ferroelectric‐van der Waals transistor (FeFET) for computing‐in‐memory by integrating a single‐crystalline Bi 2 O 2 Se (BOS) layer into a ferroelectric/MoS 2 heterostructure is demonstrated. The implementation of an asymmetrical capacitive stack ensures effective polarization‐charge compensation during fine‐state switching, achieving precise multi‐level weight programming with significantly suppressed carrier fluctuations. Fabricated through a low‐temperature process, the BOS‐based FeFET exhibits exceptional reliability, including 10‐year retention at 85°C, endurance exceeding 10 11 cycles, stable 32‐state analog switching with 0.9% retention variation over 10 000 s, and ultra‐low programming error. Atomically smooth heterointerfaces yield high spatial uniformity (7% variation) across the FeFET array, enabling a hardware neural network that achieves 98.5% accuracy in nonlinear classification. Furthermore, by incorporating intrinsic ferroelectric switching variance into the training phase, it is elucidated how device imperfections can be leveraged to reshape learning dynamics in pixel‐wise semantic segmentation. This work establishes a comprehensive co‐design methodology bridging advanced ferroelectric materials, device engineering, and algorithmic optimization for next‐generation neuromorphic computing.

A clinicopathological and molecular profiling study of <i>EGFR</i> mutations in a cohort of 215 lung cancer patients from Eastern India.

e20708 Background: Epidermal Growth Factor Receptor (EGFR) mutations are a critical biomarker in non-small cell lung cancer (NSCLC), guiding targeted therapy decisions. While global prevalence and patterns are well-documented, regional variations influenced by genetic, environmental, and lifestyle factors necessitate localized studies. This study aimed to characterize the prevalence, patterns, and clinicopathological correlations of EGFR mutations in a lung cancer cohort from Eastern India. Methods: This observational study analyzed 215 lung cancer patients. Formalin-fixed paraffin-embedded (FFPE) tissues or blood samples were collected. DNA was extracted, and EGFR mutation analysis was performed using real-time PCR (TRUPCR EGFR kit). Clinicopathological data, including age, sex, smoking history, histology, and survival, were collected. Statistical analyses were performed using MS Excel and Friedman’s ANOVA. Survival analysis was conducted using Kaplan-Meier plots. Results: EGFR mutations were identified in 38% of patients. Exon 19 deletions were the most common mutation (39% of mutated cases). Mutations were more frequent in patients aged &gt;60 years (62%), males (25% of total cohort), smokers, and those with adenocarcinoma. A significant proportion of mutated cases were found in smokers (52%) and passive smokers (24%). Systemic inflammatory markers (NLR, PLR) showed variations across exon mutation types. Survival analysis revealed a statistically significant improvement in survival for patients with any EGFR exon mutation compared to wild-type (p=0.019). No significant survival differences were observed based on sex, age, histology, or smoking status. Conclusions: This study confirms the high prevalence of EGFR mutations in an Eastern Indian cohort, aligning with global trends in demographic and histological associations. However, the notable frequency of mutations among smokers and the significant survival advantage associated with any exon mutation highlight potential region-specific characteristics. These findings underscore the importance of universal EGFR testing for all NSCLC patients in this region, irrespective of smoking history, to optimize access to targeted therapies.

CXCR4/CXCL12 antagonism to disrupt microenvironment-mediated resistance for autologous transplant conditioning in multiple myeloma: A systematic review and mechanistic synthesis.

e19523 Background: Despite advances in therapy, multiple myeloma remains incurable, with relapse driven by residual disease protected within the bone marrow microenvironment. The CXCR4/CXCL12 axis regulates myeloma cell homing and retention in stromal niches, promoting survival and limiting therapeutic efficacy. Although CXCR4 antagonists are routinely used for stem cell mobilization, pathway disruption also mobilizes malignant plasma cells and attenuates stromal resistance. Melphalan conditioning for autologous stem cell transplantation represents a key opportunity for MRD eradication, yet stromal protection remains a barrier. While CXCR4/CXCL12 antagonism has been explored with established anti-myeloma therapies, its role in chemosensitization and relevance to transplant conditioning have not been systematically synthesized. This review evaluates whether CXCR4/CXCL12 disruption enhances therapeutic sensitivity in myeloma and its rationale as an adjunct to transplant conditioning. Methods: We conducted a systematic review following PRISMA guidelines. MEDLINE, EMBASE, Scopus, and Web of Science were searched from inception through January 10, 2026 using terms related to multiple myeloma, CXCR4/CXCL12 antagonism, stromal interactions, and chemosensitization. After deduplication, records were independently screened. Eligible studies included preclinical, translational, and clinical investigations of CXCR4/CXCL12 blockade combined with anti-myeloma therapies; reviews and non-myeloma malignancies were excluded. Findings were narratively synthesized due to heterogeneity; formal risk-of-bias assessment was not performed. Results: Of 348 screened records, 22 studies met inclusion criteria. Preclinical studies showed that CXCR4/CXCL12 disruption significantly reduced stromal protection and enhanced sensitivity to melphalan, proteasome inhibitors, and immunomodulatory drugs. Early-phase clinical trials combining CXCR4 antagonists with standard regimens demonstrated significant biological activity and acceptable safety but were not powered for survival outcomes. Reported effects included increased circulating myeloma cells, response signals, and manageable hematologic toxicity. No randomized studies directly evaluated CXCR4/CXCL12 antagonism in autologous transplant conditioning. Conclusions: Across preclinical and early clinical studies, CXCR4/CXCL12 disruption increases myeloma cell vulnerability by limiting microenvironment-driven drug resistance. Although interpretation is constrained by heterogeneity and preclinical predominance, the convergence of chemosensitization and mobilization supports further investigation of CXCR4/CXCL12 blockade during melphalan conditioning to enhance cytoreduction in autologous transplantation.

Dexamethasone-associated hyperglycemia during chemotherapy in endometrial and ovarian cancer: A retrospective study.

e17583 Background: Dexamethasone is frequently administered during chemotherapy to prevent hypersensitivity reactions, nausea, and emesis. However, corticosteroid-induced glycemic disturbance remains underrecognized in gynecologic oncology. We aim to characterize how dexamethasone exposure during chemotherapy is associated with random glucose variability across the treatment cycles in ovarian and endometrial cancer patients. Methods: A retrospective study of ovarian and endometrial cancer patients treated with dexamethasone during their chemotherapy regimens at the McGill University Health Centre between November 2012 and November 2025 was performed. Random glucose measurements for each chemotherapy cycle were extracted from medical records, along with the total cumulative dexamethasone dose across the six cycles of chemotherapy. ROC analysis identified a BMI value predictive of hyperglycemic event (&gt;11.1mmol/L). A linear regression analysis was used to assess the association between BMI and the mean change in random glucose levels, adjusting for total dexamethasone dose. All analyses were stratified by cancer type to examine metabolic responses across patient subgroups. Results: Overall, 300 patients were included (150 ovarian, 150 endometrial); ovarian patients had a median BMI of 25.1 (IQR: 22.5-29.1) and a median age of 63.0 years (54.5-70.0), while endometrial patients had a median BMI of 28.1 (IQR: 24.1-35.4) and a median age of 65.0 years (IQR: 56.7-73.0). Median cumulative dexamethasone dose was 186mg (range: 16-500mg). ROC analysis identified BMI of 23.55 and 25.45 as a predictor of hyperglycemic events in ovarian and endometrial patients, respectively. Hyperglycemic episodes had a rate of 22.0% and 35.0% in ovarian and endometrial cancer patients, respectively (p=0.627). In the endometrial cancer cohort, a greater proportion of patients with BMI&gt;25.0 experienced ≥1 hyperglycemic event compared to those with BMI&lt;25.0 (20.2% vs. 0.0%; 16 vs. 0 events; p=.006), whereas no significant difference was observed in the ovarian cancer cohort (18.6% vs. 9.3%; 8 vs. 5 events; p=.180). In ovarian cancer patients, neither BMI (p=.171) nor dexamethasone dose (p=.250) was associated with random glucose level. In contrast, for endometrial cancer patients, both BMI (p=.045) and dexamethasone dose (p=.032) were independently associated with increased random glucose levels. Conclusions: Dexamethasone exposure during chemotherapy is associated with clinically meaningful increases in random glucose in select gynecological cancer subgroups, particularly in overweight patients. These findings highlight the importance of BMI-informed risk stratification and proactive glucose monitoring to mitigate dexamethasone-related hyperglycemia in gynecologic oncology care.

Toxicity assessment clinic as an efficacious tool to provide fast care to oncology patients on systemic therapy by minimizing emergency department visits and streamlining admissions directly from the cancer clinics.

12124 Background: The Toxicity Assessment Clinic (TAC) at the Hudson Regional Cancer Centre was established to improve access to care for cancer patients experiencing treatment-related toxicities, ensuring timely access to oncology-specific symptom management. The TAC is nurse led and utilizes an evidence-based oncology assessment tool to determine the appropriate care pathway, ensuring symptom assessment and treatment in a timely and safe manner. Methods: We reviewed the volume and characteristics of patients seen in the TAC for one year (2024), including visit modality, primary causes for TAC consultation, and appropriateness of TAC referrals by identifying related vs. unrelated complaints and determining the proportion of patients without a family physician who utilized TAC services for unrelated concerns. Results: There were 15,387 total chemotherapy clinic visits. Among them, there were 692 visits to TAC (4.5%). In total, 94.8% of patients had a family physician. Majority (93.9%) called TAC directly, 4.6% were sent by chemo suite nurses and 1.4% were referred from outside the cancer center. In addition, there were 997 registered nurse phone calls to follow up on health issues occurring within the clinic hours. Median age was 69 years [range 19-93]; 61% of patients were over 65. 61.9% of TAC patients had a solid tumor diagnosis, 24.5% had a hematological malignancy, and 6.8% had gynecological cancers. Gastrointestinal malignancies (26.9%), breast (17.5%) and lung (12.3%) were most prevalent. 47.9% had stage IV cancer, 27.2% had stage III. 93.5% of patients were on active systemic therapy, with 80.8% on first line. In-person TAC visits composed 89.1% of consults, and 10.9% had assessments over the phone. Skin lesions (7.8%), diarrhea (6.9%), disease progression (6.5%), urinary tract infection (5.8%) and respiratory infections (5.6%) were the most common issues. In 92.3% of cases, the presenting complaint was considered directly related to the cancer or therapy. Consult assessment durations were the following: 30 minutes or less (23.2%), 30-60 minutes (71.8%), and ≥60 minutes (4.9%). Of note, 76% of phone calls required ≤30 minutes compared to 39% patients who attended TAC in person. 52.6% of admitted patients from TAC required ≥ 60 min of assessment time. In 85.7% of cases, medical problems were resolved during the TAC visit; 1.4% were sent to ED and 12.9% were directly admitted to hospital. Conclusions: The TAC model is feasible and provides rapid assessment of patients on systemic therapy, particularly in the front-line setting. It is a cost-effective alternative to ED visits and streamlines admissions directly from cancer clinics. TAC improves patient flow for vulnerable populations such as elderly and immunocompromised patients.

Estimated clinical benefits of timely biomarker testing and matched therapy in HR+/HER2– metastatic breast cancer in the United States: A population model.

e13049 Background: Targeted therapies for actionable genomic changes ( PI3K pathway and ESR1 ) have transformed the treatment (tx) landscape and improved outcomes for HR+/HER2– metastatic breast cancer (mBC) patients (pts). However, their clinical benefits depend on timely genomic testing and appropriate tx selection, both of which remain suboptimal in real-world (RW) practice. This study modeled the population-level clinical impact of improving timely genomic testing and administering appropriately matched tx in U.S. pts with HR+/HER2– mBC, compared with current RW practice. Methods: A Markov-based population model simulated U.S. adults with HR+/HER2– mBC initiating first-line (1L) endocrine therapy (ET), estimating outcomes over 3 years across two tx lines. The base case reflected RW testing (34–62%) and tx uptake (6–29%) by genomic alteration, compared with an optimal scenario assuming 100% timely testing and matched tx. Matched tx was defined as administering ET plus mutation-specific targeted tx when available or ET alone or with non-selected agents for those without actionable mutations for each line. Timely testing was defined as genomic results available before starting a tx line with a targeted option available. Additional scenarios, varying testing and tx rates independently were evaluated. Outcomes included population-level overall survival (OS) across two lines and 1-year progression-free survival (PFS) by line. Model inputs, including PFS, OS, genomic alteration prevalence, and RW testing and tx rates, were derived from published literature. Sensitivity analyses assessed the impact of uncertainty in key inputs. Results: In a modeled population of 44,552 pts, 100% timely genomic testing and tx, vs current RW rates increased receipt of matched tx by 14% in 1L and 68% in 2L. Modeling 100% genomic testing and subsequent administration of appropriately guided tx, compared with RW rates, was associated with a 3.6% absolute improvement in 3-year OS, corresponding to 1,598 fewer OS events, including 578 among 1L patients with PIK3CA-mutated disease. Scenario analyses showed that all but the scenario with no test-to-treat was associated with an improvement in OS vs the base case. In year 1, the comparison of 100% test-to-treat versus base case showed 1,664 progressions or deaths avoided in 1L and 2,353 in 2L. Conclusions: Delays in genomic testing and suboptimal alignment of tx with genomic findings limit the clinical benefit of targeted tx for HR+/HER2– mBC. This model suggests that timely testing and matched therapy, particularly in 1L pts with PIK3CA mutations, may reduce disease progression and mortality. Interpretation should consider modeling limitations, including assumptions about tx adherence and uptake.

Radiation therapy utilization and survival associations in metastatic Merkel cell carcinoma in the immunotherapy era: A national database study.

e21561 Background: Immune checkpoint inhibitors (ICIs) have become the standard first-line systemic therapy for metastatic Merkel cell carcinoma (MCC). Radiotherapy (RT) remains commonly used for local control and palliation, and growing biologic interest suggests that RT may enhance antitumor immune responses when combined with ICIs. However, population-level evidence regarding contemporary patterns of RT use and its association with survival in immunotherapy-treated patients with metastatic MCC is limited. Methods: We conducted a retrospective population-based cohort study using the National Cancer Database (NCDB), including adults diagnosed with metastatic MCC between 2011 and 2022. Temporal trends in treatment utilization were described across the study period. Among patients diagnosed in the immunotherapy era (2017–2022), factors associated with receipt of RT were examined using multivariable logistic regression. To evaluate survival, analyses were restricted to immunotherapy-treated patients, and overall survival was compared between those who received RT and those who did not using propensity score-matched Kaplan-Meier and Cox proportional hazards analyses, with a focus on 2-year overall survival. Results: A total of 1,736 patients with metastatic MCC were included. Following 2017, immunotherapy use increased substantially, while chemotherapy use declined; RT utilization remained relatively stable over time. In multivariable analyses restricted to the immunotherapy era, higher primary tumor stage and lymph node dissection were independently associated with RT receipt. Among immunotherapy-treated patients in the propensity score-matched cohort, RT was not associated with improved 2-year OS (aHR,1.09; 95% CI, 0.84-1.42), and Kaplan-Meier analyses demonstrated no significant survival difference. Conclusions: In the contemporary immunotherapy era, RT continues to be selectively incorporated into the management of metastatic MCC, particularly in patients with greater local or regional disease burden. However, among immunotherapy-treated patients, RT was not associated with improved short-term overall survival at the population level. These findings support a continued role for RT in local disease management while underscoring the need for prospective studies to better define its optimal integration with immunotherapy.

Treatment beyond progression after early radiographic changes on first-line ipilimumab/nivolumab in metastatic clear cell renal cell carcinoma.

e16506 Background: Ipilimumab/nivolumab (I/N) is a standard first-line (1L) regimen for metastatic clear cell renal cell carcinoma (mccRCC) with durable benefit in patients achieving deep responses. However, management of patients with mixed responses or early radiographic progression remains undefined, with limited data supporting the common practice of treatment beyond progression (TBP). We evaluated treatment patterns and clinical outcomes associated with TBP following early radiographic changes in I/N-treated patients with mccRCC. Methods: We retrospectively analyzed patients with mccRCC who completed four doses of 1L I/N at Memorial Sloan Kettering Cancer Center. Patients with clinician-assessed mixed response (concurrent regression of some lesions with progression or new lesions) or radiographic progression (overall increase in tumor burden and/or new lesions) on 1st post-induction scan were classified as TBP or not, while those without progression were categorized as early disease control (DC). Kaplan–Meier methods estimated time from 1st post-induction scan to start of next therapy (TTNT) or death (OS), with prespecified landmark estimates at 24 months (TTNT) and 5 years (OS) used for between-group comparisons by log-rank tests. Time on treatment beyond progression (TTBP) was calculated for TBP patients. Results: Among 137 patients completing I/N induction, 54 achieved DC and 83 had mixed response (n=43) or progression (n=40) at 1st post-induction imaging; 47/83 (57%) continued TBP. Among TBP patients, 18 (38%) achieved DC on 2nd scan, (median interscan interval 10.1 weeks), 13 (28%) had further growth, and 6 lacked evaluable 2nd scans. Median TTBP was 12.3 months (CI, 3.8–18.0), and 16 TBP patients (34%) had not received subsequent therapy at last follow-up. At 24 months, TTNT was 12% for no-TBP, 38% for TBP, and 74% for DC (p&lt;0.001). At 5 years, OS was 23% (95% CI, 12–37) for no-TBP, 50% (95% CI, 32–66) for TBP, and 65% (95% CI, 42–80) for DC (p&lt;0.001). Conclusions: In this real-world cohort, TBP after I/N induction was common, with 38% of patients achieving DC and over half remaining on therapy for ≥1 year. However, outcomes with TBP were inferior to those with clear radiographic DC, and long-term benefits for this regimen were rarely observed in TBP patients, including those achieving DC. These findings highlight heterogeneity of early imaging outcomes and support prospective refinement of patient selection for TBP. Second-scan outcomes by first-scan response after I/N induction (n=137; 131 with evaluable second scan). 1st scan category Response on 2nd scan Stable disease on 2nd scan Mixed response on 2nd scan Progression on 2nd scan Total Response 17 (14%) 16 (39%) 4 (10%) 4 (10%) 41 Stable disease 2 (15%) 6 (46%) 3 (23%) 2 (15%) 13 Mixed response 16 (37%) 4 (9%) 16 (37%) 7 (16%) 43 Progression 12 (30%) 5 (12%) 5 (12%) 12 (30%) 34 Total 47 31 28 25 131

Emergency vs elective presentation as a care-delivery phenotype in presumed early-stage lung cancer: A National Inpatient analysis.

e23223 Background: Early-stage lung cancer is potentially curable with surgery, yet emergency or nonelective presentation represents a care-delivery phenotype reflecting failures in early diagnosis or access to care rather than tumor biology alone. The national impact of emergency presentation among surgically treated, presumed early-stage lung cancer patients is poorly defined. This study evaluates emergency versus elective presentation as a care-delivery phenotype and its association with perioperative morbidity, failure-to-rescue, and inpatient outcomes. Methods: A survey-weighted analysis of the National Inpatient Sample (2016–2023) was conducted. Adult lung cancer hospitalizations were identified using ICD-10-CM C34* in any diagnosis position. A presumed early-stage surgical cohort was defined by curative-intent lung resection without metastatic disease, malignant pleural effusion, or palliative care coding. The primary exposure was nonelective versus elective admission. Outcomes included in-hospital mortality; ICU-level care, defined by invasive mechanical ventilation, shock, or respiratory failure; major complications; failure-to-rescue; length of stay; and hospitalization cost and charges. Multivariable survey-weighted models adjusted for patient demographics, payer, neighborhood income, hospital characteristics, and year. Results: In the weighted cohort of 422,925 hospitalizations, nonelective admission occurred in 27.3%. After multivariable adjustment, nonelective admission was independently associated with higher odds of in-hospital mortality (adjusted odds ratio [aOR] 2.59, 95% CI 2.22–3.03), ICU-level care (aOR 4.53, 95% CI 4.32–4.76), and major complications (aOR 5.00, 95% CI 4.78–5.23), as well as longer length of stay (+2.42 days). Notably, failure-to-rescue among patients with major complications did not differ by admission type (aOR 0.90, 95% CI 0.77–1.05), indicating that excess mortality associated with nonelective presentation is driven by presentation severity and care pathway rather than differences in rescue quality. In unadjusted analyses, nonelective presentation was associated with higher mortality (2.21% vs 0.68%), greater ICU-level care use (33.8% vs 8.9%), more frequent major complications (45.0% vs 12.4%), and longer length of stay (7.84 vs 4.99 days) compared with elective admissions. Conclusions: In a procedure-defined, presumed early-stage lung cancer surgical cohort, nonelective admission identifies a common care-delivery failure phenotype associated with higher perioperative morbidity, ICU-level care, in-hospital mortality, and longer hospitalization. Emergency presentation represents a systems-level vulnerability in curative-intent lung cancer care and serves as a scalable and actionable marker for quality improvement and disparity-focused analyses.

Prognostic impact of cancer therapy–related cardiotoxicity on oncologic outcomes.

e24023 Background: Cancer therapy-related cardiotoxicity is a common complication of modern anticancer treatments and may adversely affect oncologic outcomes by limiting treatment delivery. While cardiotoxicity is well studied as a cardiovascular safety endpoint, its prognostic impact on cancer outcomes remains incompletely defined and appears to vary across cancer types and treatment settings. Methods: We conducted a systematic review (PROSPERO: CRD420261294684) of observational studies evaluating the association between cardiotoxicity and oncologic outcomes. Eligible studies included adults (≥18 years) with solid tumors or hematologic malignancies receiving systemic anticancer therapy, including anthracyclines, HER2-targeted therapies, immune checkpoint inhibitors, cellular therapies, or other agents. Studies were required to report oncologic outcomes stratified by cardiotoxicity status. Cardiotoxicity was defined using study-specific criteria and included left ventricular ejection fraction decline (typically ≥10% to &lt; 50%), clinical heart failure, myocarditis, clinically significant arrhythmias, biomarker elevation, or cardiotoxicity leading to treatment interruption or discontinuation. Primary outcomes were overall survival (OS) and progression-free or disease-free survival (PFS/DFS). Results were synthesized narratively due to heterogeneity. Results: Eight observational studies met inclusion criteria (3,620 patients). Study designs included retrospective cohorts and prospective registries, with sample sizes ranging from 60 to 1,399 patients and median follow-up of 6-72 months. Associations between cardiotoxicity and oncologic outcomes were heterogeneous. Cardiotoxicity was independently associated with worse OS in patients receiving CAR-T therapy (HR 1.85, 95% CI 1.17-2.92; p = 0.009) and in the cohorts experiencing severe cardiotoxicity (HR 10.2, 95% CI 5.5-19.2; p &lt; 0.001). In contrast, adjuvant trastuzumab-treated breast cancer cohorts showed no consistent association between cardiotoxicity and OS (HR 1.68, 95% CI 0.83–3.41; p = 0.148). For DFS or recurrence-free survival, cardiotoxicity leading to treatment interruption was associated with inferior outcomes (HR 1.6, 95% CI 1.1-2.3), while other cohorts showed non-significant trends. Across studies, cardiotoxicity was consistently linked to treatment interruption, dose reduction, or early discontinuation. Conclusions: Cardiotoxicity is associated with adverse oncologic outcomes primarily in patients with severe or clinically overt cardiac events and in intensive treatment settings. Mild or asymptomatic cardiotoxicity, particularly in adjuvant HER2-positive breast cancer, was not consistently associated with worse survival but frequently resulted in treatment modification. Prospective studies using standardized cardiotoxicity definitions are needed to better define prognostic risk.