Browse Articles
Discover research articles across all indexed journals
Intraoperative precision margin assessment in breast-conserving surgery using handheld fluorescence microscopy.
580 Background: Breast-conserving surgery (BCS) is the standard treatment for early-stage breast cancer and is increasingly preferred by younger patients. The key challenge in BCS is ensuring complete tumor removal while preserving healthy breast tissue. Surgical margin assessment plays a crucial role in reducing the risk of local recurrence. Traditional intraoperative frozen sections (FS) are time-consuming, limit sampling, and may cause tissue damage. This study evaluates the performance of EndoScell, a handheld fluorescence microscopy system, as a rapid and efficient alternative for intraoperative margin assessment. Methods: EndoScell provides real-time, high-resolution imaging of tissue margins during surgery. This system rapidly stains and scans tissue, allowing for micron-level visualization of cellular features. In this study, EndoScell was used to scan marked surfaces of breast tissue during surgery. The surgeon performed an initial assessment, while images/videos were recorded. Specimens were then sent for intraoperative FS analysis for comparison. If FS identified positive margins, the area was re-excised by 3–5 mm, and the process was repeated. Routine paraffin pathology served as the gold standard, with blinded side-by-side reading of ES images to evaluate consistency. Results: As of December 31, 2025, 300 patients undergoing BCS were included. EndoScell demonstrated high diagnostic accuracy with a sensitivity of 96.4%, specificity of 95.1%, and AUC of 0.953, showing strong agreement with paraffin pathology. Notably, the time required for EndoScell imaging was significantly shorter than FS, with an average of 3.14 ± 1.48 minutes compared to 37.50 ± 13.34 minutes for FS (P<0.001), leading to substantial improvements in surgical efficiency. Conclusions: This study confirms the feasibility and non-inferiority of EndoScell for intraoperative margin assessment in breast-conserving surgery. The system enables real-time decision-making, reducing secondary surgeries, improving outcomes, and enhancing surgical safety. Its significant clinical and social value supports further development and implementation. Key metrics related to diagnostic accuracy and time efficiency between EndoScell and traditional frozen section methods. Metric EndoScell (ES) Frozen Section (FS) Positive Margin 54 56 Negative Margin 246 244 Sensitivity 0.964 1.000 Specificity 0.951 1.000 AUC 0.953 (95% CI: 0.912–1.003) 1.000 Imaging Time (min) 3.14 ± 1.48 37.50 ± 13.34 p-value <0.001 N/A Cohen's d -3.64 N/A
A phase 2 study evaluating SSGJ-707 (PF-08634404), a PD-1/VEGF bispecific antibody, + chemotherapy (chemo) in patients (pts) with first-line (1L) advanced/recurrent endometrial cancer (EC).
5627 Background: SSGJ-707 is a fully human immunoglobulin G4 bispecific antibody targeting programmed death 1 (PD-1) and vascular endothelial growth factor (VEGF). SSGJ-707 has demonstrated promising efficacy and manageable safety alone and in combination with chemo in phase 2 studies in solid tumors. We report results from the phase 2 SSGJ-707-ST-II-02 study (NCT06522828) of 1L SSGJ-707 + chemo in advanced/recurrent EC. Methods: Pts with newly diagnosed stage lIl/IV or recurrent EC with low potential for cure by radiation therapy (tx) or surgery and systemic tx-naive were enrolled. Pts received SSGJ-707 5 mg/kg or 10 mg/kg Q3W + chemo (carboplatin AUC 5 + paclitaxel 175 mg/m 2 ) Q3W for 6 cycles, followed by SSGJ-707 maintenance tx until loss of clinical benefit or intolerable toxicity for up to 2 years. Primary endpoints were safety and ORR (RECIST 1.1). Secondary endpoints include duration of response (DOR), PFS, OS, PK, and biomarkers. Results: At data cutoff (Nov 28, 2025), 32 pts with EC (26 mismatch repair proficient [pMMR], 6 MMR deficient [dMMR]) received 5 mg/kg (n=16) or 10 mg/kg (n=16) SSGJ-707 + chemo. Median SSGJ-707 tx duration was 5.6 mo (range, 0.2-13.6); 8 pts (25.0%) discontinued tx. In the evaluable population, confirmed ORR was 85.7% (12/14 pts) for the 5 mg/kg dose and 80.0% (12/15) for the 10 mg/kg dose. Results for pMMR and dMMR groups are in the Table. Median PFS, OS, and DOR were not reached for either dose. Any-grade tx-related adverse events (TRAEs) were reported in 30/32 pts (93.8%) and grade ≥3 TRAEs in 22/32 pts (68.8%). The most common TRAEs (≥40%) included neutrophil count decreased (59.4%), white blood cell count decreased (59.4%), anemia (56.3%), and platelet count decreased (56.3%). No TRAEs led to death. TRAEs led to discontinuation of SSGJ-707 in 2 pts (6.3%). Immune-related AEs occurred in 6 pts (18.8%): hypothyroidism (n=4), hyperthyroidism (n=3), and rash (n=2). VEGF-related AEs occurred in 13 pts (40.6%): blood pressure elevation (n=6), proteinuria (n=6), hemorrhage (urinary occult blood positive; n=2), perforation and fistula (n=2), and venous thrombosis (n=1). Conclusions: SSGJ-707 + chemo demonstrated promising efficacy with a manageable safety profile in pts with tx-naive advanced/recurrent EC, supporting further investigation of SSGJ-707 in these pts. Clinical trial information: NCT06522828 . 5 mg/kg Q3W (n=14) 10 mg/kg Q3W (n=15) pMMR (n=12) dMMR (n=2) pMMR (n=11) dMMR (n=4) Confirmed ORR, n (%) [95% CI] 10 (83.3) [51.6, 97.9] 2 (100)[15.8, 100.0] 9 (81.8)[48.2, 97.7] 3 (75.0)[19.4, 99.4] Complete response 0 0 1 (9.1) 0 Partial response 10 (83.3) 2 (100) 8 (72.7) 3 (75.0) Stable disease 2 (16.7) 0 2 (18.2) 1 (25.0) Progressive disease 0 0 0 0 CR/PR pending – – 1 a 1 Unconfirmed ORR, n (%) [95% CI] 10 (83.3) [51.6, 97.9] 2 (100)[15.8, 100.0] 10 (90.9)[58.7, 99.8] 4 (100)[39.8, 100.0] a One additional pt in the 10 mg/kg group confirmed in Dec.
Uveal melanoma: A retrospective cohort study of long-term outcomes and impact of surveillance imaging intervals on metastatic burden to guide risk-adapted surveillance strategies.
e21578 Background: Uveal melanoma (UM) is associated with a high lifetime risk of metastatic disease, characterized by prolonged dormancy periods (sometimes decades) and strong hepatic predilection. Molecular risk stratification using gene expression profiling (GEP) and Preferentially Expressed Antigen in Melanoma (PRAME) expression is commonly used for prognostication and surveillance discussions. However, there is a general lack of consensus on optimal surveillance imaging modalities and intervals with varied practices across experts and institutions. More frequent imaging may not necessarily lead to clinically actionable interventions and can generate considerable anxiety for patients. Methods: We conducted a single institution, retrospective cohort study of UM patients seen between 2014-2025. Of 619 patients total, 440 met the inclusion criteria of stage I–III UM at diagnosis with at least one year follow-up. We captured metastatic events and correlated with clinical stage, GEP class and PRAME status. Surveillance imaging intervals, calculated from the last normal imaging study to the first scan concerning for metastasis, were categorized as 6, 12, or >18 months. Metastatic tumor burden, including size (e.g. LDLHM = largest diameter of the largest hepatic metastasis) and number of metastases, was compared across surveillance imaging intervals for different molecular risk categories. Results: With a median follow-up of 4.4 years (inter-quartile range [IQR] 2.5-7.0), 83 of 440 patients (19%) developed distant metastases (DM). DM occurred across all GEP risk categories (class 1A - 10/109 [9%]; class 1B - 6/66 [9%] and class 2 - 49/121 [40%]) and most commonly involved the liver (86%) and/or lungs (17%). Among these 83 patients, median time-to-metastasis was 26 months (IQR 19-46) overall, 35 months (IQR 23-49) in GEP class 1A/1B and 23 months (IQR 17-33) in class 2. GEP class 2 was also associated with greater hepatic tumor burden at time of first detection, including larger tumors (median LDLHM 19 mm [IQR 10 – 32] for class 2 vs 10 mm [IQR 7 – 15] for class 1A/B; p = 0.14) and more tumors (median number of hepatic metastases 5 [IQR 2-10+] in class 2 vs 3 [IQR 2-3] in class 1A/B; p = 0.02). Median LDLHM for class 2 tumors was 13 mm [IQR 8-22] and 31 mm [IQR 23-52] with 6- and 12-months surveillance intervals, respectively (p = 0.009). Median LDLHM was 10.5 mm [IQR 8.25-17.5] for class 1A/1B with 12-months surveillance intervals. Conclusions: GEP risk categories influence biologic behavior of UM metastases, with metastases developing more commonly and sooner and with greater tumor burden for GEP class 2 vs class 1A/1B, warranting a risk-adapted surveillance approach. Our data support 6-months surveillance imaging interval for GEP class 2 tumors and 12 -months interval for GEP classes 1A and 1B.
Perioperative cadonilimab (AK104) in locally advanced MSI-H/dMMR gastric or gastroesophageal junction adenocarcinoma and colorectal adenocarcinoma: A basket study.
4102 Background: Microsatellite instability-high (MSI-H) / deficient mismatch repair (dMMR) tumors generate abundant neoantigens and are sensitive to PD-1 blockade. However, evidence supporting the use of PD-1/CTLA-4 blockade in the neoadjuvant setting for MSI-H/dMMR gastric and colorectal cancers remains scarce. Methods: In this multicenter, single-arm, dual-cohort basket study (NCT04556253), we used a Simon’s two-stage design to evaluate perioperative cadonilimab in patients with locally advanced MSI-H/dMMR gastric/gastroesophageal junction (GEJ) or colorectal adenocarcinoma. 29 patients were enrolled in cohort 1 (gastric/GEJ adenocarcinoma) and 18 patients were enrolled in cohort 2 (colorectal adenocarcinoma). The main inclusion criteria include newly diagnosed, histologically confirmed cT3–T4aNxM0 MSI-H/dMMR gastric/GEJ adenocarcinoma or colorectal adenocarcinoma, with an ECOG score of 0-1. Both cohorts received 3 cycles of neoadjuvant cadonilimab (10 mg/kg q3w) followed by radical resection, then 6 cycles of adjuvant cadonilimab (6 mg/kg q3w). The primary endpoint was pathologic complete response (pCR). Secondary endpoints included overall survival (OS), event-free survival (EFS), disease-free survival (DFS), major pathologic response (MPR) rate and incidence of adverse events (AEs). Results: As of the data cutoff date of January 24, 2026, the median follow-up was 14.6 months for cohort 1 and 12.6 months for cohort 2. In cohort 1, 2 patients were pathologically confirmed as pMMR prior to treatment. Among the 26 patients (mITT population), pCR rate was 69.2% (95% CI: 48.2%-85.7%), and MPR rate was 80.8% (95% CI: 60.6%-93.4%). The 12-month OS, EFS, and DFS rates were each 100% in mITT population. All 29 patients (ITT population) received neoadjuvant therapy. AEs during neoadjuvant treatment occurred in 24 patients (82.8%), with 3 patients (10.3%) experiencing grade 3-4 AEs. Common AEs (incidence ≥10%) included rash, anemia, constipation, and insomnia, which were primarily grade 1-2. Among the 18 patients (mITT population) in cohort 2, the postoperative pCR rate was 66.7% (95% CI: 41.0%-86.7%), and the MPR rate was 72.2% (95% CI: 46.5%-90.3%). The 12-month OS, EFS, and DFS rates were each 100% in mITT population. All 18 patients (ITT population) received neoadjuvant therapy. AEs during neoadjuvant treatment occurred in 17 patients (94.4%), with 2 patients (11.1%) experiencing grade 3-4 AEs. Common AEs (incidence ≥10%) included rash, infusion-related reaction, anemia, and cough, which were primarily grade 1. No grade 5 AEs were reported in this study. Conclusions: Perioperative cadonilimab demonstrates excellent antitumor activity and a manageable safety profile in patients with locally advanced MSI-H/dMMR gastric/GEJ adenocarcinoma and colorectal adenocarcinoma. Clinical trial information: NCT04556253 .
A population-based modular multi-objective optimization framework with decoupled exploration and exploitation for lung adenocarcinoma histopathological subtyping in H&E whole-slide images (WSI).
e20019 Background: Current state-of-the-art artificial neural network optimizers, such as Adam, are single-objective, single-solution and produce a single weight configuration optimized for a broad range of tasks such as histopathological subtyping of lung adenocarcinoma due to complex morphologic patterns. One-size-fits-all optimization limits clinical adaptability. Existing multi-objective approaches prioritize loss reduction without monitoring clinically meaningful metrics, leading to unpredictable generalization and underscoring the need for domain-calibrated multi-objective optimization frameworks. Methods: We proposed a universal, component-independent multi-objective optimization framework that decouples exploitation (search-space navigation) from exploration (elite models selection) to provide a Pareto front of high-performing trade-off solutions. Therefore, the performance metrics of the final Pareto-front solutions are no longer unexpected, since the calibration during training consistently promotes model parameters that achieve strong performance according to the selected evaluation metrics, which will contain more clinical control on the deep neural networks. The framework is modular: exploitation can use any loss-driven optimizer, and exploration can use any performance-metric–driven evolutionary selection method. In this study, for multi-label classification, we used Adam-based updates for exploitation and max–max non-dominated sorting for exploration. The multi-objective optimizer minimized clinically motivated losses (1−precision, 1−recall) while maximizing evaluation metrics (specificity, sensitivity) to identify high-performing trade-off models. Results: In multi-objective, multi-label classification experiments, we compared MAdam and enhanced MAdam using hypervolume (HV) as the primary performance indicator. Experiments used the WSSS4LUAD dataset, a weakly supervised benchmark for histologic subtyping of lung adenocarcinoma from H&E whole-slide images. Table 1 reports higher HV values for the proposed performance metric–driven non-dominated sorting (Scheme 2) compared with loss-driven non-dominated sorting (Scheme 1). Conclusions: We present a multi-objective optimization framework that decouples search-space navigation from solution selection to generate a Pareto-front of clinically meaningful trade-off models. This approach provides pathologists with multiple well-calibrated model options. It also improves convergence and achieves broader coverage of high-performing model parameters compared with conventional loss-driven non-dominated sorting strategies. Hypervolume (HV) comparison between schemes. Dataset Scheme 1 Scheme 2 Train 0.92 0.97 Validation 0.89 0.96 Test 0.89 0.96
A phase II study of bTAE-HAIC combined with lenvatinib and camrelizumab for patients with high tumor burden hepatocellular carcinoma: The TALEM-H trial.
4123 Background: Patients with high tumor burden hepatocellular carcinoma (HCC) face a dismal prognosis. TALEM-H is a single-arm, prospective trial, aiming to evaluate the safety and preliminary effectiveness of transcatheter arterial embolization with blank microspheres (bTAE) plus hepatic artery infusion chemotherapy (HAIC) combined with lenvatinib and camrelizumab as first-line therapy for patients with high tumor burden HCC. Methods: Forty patients with huge HCC (defined as a single tumor >10 cm or multiple tumors >15 cm in diameter) received bTAE and FOLFOX-HAIC (Oxaliplatin 85mg/m2 2h + Calcium Levofolinate 400mg/m2 2h + Fluorouracil 2400mg/m2 46h) and lenvatinib (4-8mg) once daily plus camrelizumab (200mg) Q3W. The primary endpoint was objective response rate (ORR) according to mRECIST criteria. Secondary endpoints included progression-free survival (PFS), overall survival (OS) and treatment-related adverse events (TRAEs). Results: As of January 10, 2026, the median follow-up time was 10.7 months. The median maximum tumor diameter was 124mm (range: 77.5–262 mm) and the median sum of the diameters of the three largest tumors was 182 mm (range: 110–328.5 mm). The intrahepatic ORR was 95%, with 7 cases (17.9%) of intrahepatic complete response (CR) and 30 cases (76.9%) of intrahepatic partial response (PR). The 1-year and 2-year PFS rates were 48.7% and 37.6%, respectively (median PFS: 10.4 months; 95% CI: 8.6–15.1). Median OS has not yet been reached. Further, TRAEs of any grade occurred in 35 patients (87.5%) and no treatment-related deaths occurred. Grade 3 or higher TRAEs occurred in 12.5% of patients, commonly hypoalbuminemia (5%), immune-mediated pneumonitis (2.5%), and anemia (2.5%). Conclusions: bTAE-HAIC combined with lenvatinib plus camrelizumab represents a promising first-line treatment, offering an exceptional ORR and encouraging survival outcomes for HCC patients with a high tumor burden. Clinical trial information: NCT06061276 . Baseline characteristics. Variable All Patients (N = 40) BCLC Stage C 32 (80%) Extrahepatic Metastasis 18 (45%) Tumor Number- Multiple（≥3） 28 (70%) Largest Diameter (cm) >10 cm 35 (87.5%) Sum of the diameters of the three largest tumors >20 cm 14 (35%) Macrovascular Invasion 27 (67.5%)
The impact of introducing cemiplimab as adjuvant treatment in high-risk cutaneous squamous cell carcinoma on NHS services and facilities.
e18031 Background: The recently published C-POST phase 3 trial has shown improved disease-free survival and reduced locoregional recurrence with adjuvant cemiplimab in high-risk locally advanced cutaneous squamous cell carcinoma (CSCC). Incorporating adjuvant cemiplimab into current standard-of-care regimes will represent a significant shift in practice. Our aim was to assess the impact of this change by calculating the number of potentially eligible patients treated in a busy tertiary head and neck cancer (HNC) centre and estimating the effect on resources for future service planning. Methods: With institutional approval (The Clatterbridge Cancer Centre NHS Foundation Trust), we identified HNC patients with cutaneous squamous cell carcinoma referred for radiotherapy after ablative surgery between October 1, 2024, and September 30, 2025. All radiotherapy referrals were captured. A retrospective case-note review identified patients meeting nodal and non-nodal high-risk criteria from the C-POST trial. Additional variables included age, gender, primary diagnosis, ethnicity, demographics, TNM staging, histology, radiotherapy site and dose, and reasons for not receiving radiotherapy. Results: 164 patients were referred, of whom 84.14% (138) were referred for adjuvant radiotherapy, 6.7% (11) for radical radiotherapy, 6.1% (10) for palliative radiotherapy, and 3.04% (5) were inappropriate referrals. Mean age was 80.5 years (SD 9.68) and M:F was 4:1. Overall, 90.85% (149) were eligible for adjuvant or radical radiotherapy. Among eligible patients, 65.8% (n=98) received radiotherapy, 20.1% (n=30) opted for surveillance, 8.7% (n=13) for surgical excision, 2.7% (n=4) declined treatment, and 2.7% (n=4) were unsuitable due to poor PS. In total, 54.87% (n=90) had high-risk nodal or non-nodal disease and were eligible for adjuvant cemiplimab. Within this group, 74.2% (n=66) had perineural invasion, 15.7% (n=14) recurrent CSCC with ≥N2b, ≥T3 or poorly differentiated histology, 5.7% (n=6) nodal disease with ECE, 2.2% (n=2) T4 lesions, and 1.1% (n=1) in-transit metastases. Based on C-POST, the cohort requires ~2849 hospital visits for adjuvant cemiplimab. Each patient needs an initial radiological assessment, 10 pretreatment assessment/blood visits, 10 treatment visits, 4 visits for surveillance scans, and 6 visits for consultant review. Immune-related toxicity management adds ~59 visits, assuming 20% experience one toxicity episode requiring ≥3 attendances plus ~5 additional visits for imaging or blood tests. Conclusions: Introducing adjuvant cemiplimab in high-risk cutaneous squamous cell carcinoma at a busy NHS tertiary cancer centre is feasible but will lead to major changes in treatment protocols and service provision. This audit provides a benchmark to aid service development and quantify the need for future departmental expansion.
Real-time quantification of laryngeal motion in early glottic cancer using MR-Linac cine imaging: Implications for adaptive PTV margins.
e18109 Background: To quantify intra-fraction laryngeal motion in patients with early-stage glottic cancer treated on a 1.5T MR-Linac and evaluate the potential for planning target volume (PTV) margin reduction. MR-guided radiotherapy (MRgRT) offers real-time visualization that could in turn facilitate personalized adaptive treatment strategies, while standard protocols used in conventional treatment options often have to rely on generous margins to account for uncertainty that in turn increases toxicity. Methods: A prospective observational study was conducted involving 6 patients diagnosed with T1–T2 early glottic carcinoma undergoing MRgRT on the Elekta Unity system. Patients were immobilized using an MR-safe head and neck base plate with a headrest and standard 5-point thermoplastic mask. Intra-fraction motion of the thyroid cartilage (surrogate for the glottis) was continuously monitored using the Comprehensive Motion Management (CMM) system via real-time cine MRI. The target volume with OARs and the tracking structures were contoured by an expert head and neck radiation oncologist. Motion data was analysed across Superior-Inferior (SI), Left-Right (LR), and Anterior-Posterior (AP) axes. Non-physiological outliers (>2 cm) representing swallowing or coughing events were excluded. Descriptive statistics (mean, SD, 95th percentile) were calculated to inform margin efficacy. Results: Analysis of continuous motion revealed that laryngeal stability is predominantly maintained during normal respiration, with transient excursions due to swallowing. After exclusion of outliers, the mean displacement was 0.81 ± 1.05 mm in the LR direction, 1.13 ± 1.14 mm in the AP direction, and 2.37 ± 2.06 mm in the SI direction. The 95th percentile of motion — representing the boundary required to cover the target 95% of the time — was 2.5 mm (LR), 3.3 mm (AP), and 6.5 mm (SI). Conclusions: Laryngeal motion during MRgRT is anisotropic, with the greatest displacement observed in the SI direction. The data suggests that conventional PTV margins (often uniform 3-5mm) may be safely reduced in the axial plane (LR/AP) to 3 mm, while the SI margin requires careful management or gating to account for swallowing. CMM allows for safe margin personalization, potentially reducing toxicity to the pharyngeal constrictors and larynx minus gross tumor volume (GTV). The result of our study provides novel insight into the dynamics of laryngeal motion using MR-Linac cine imaging and underscores the clinical potential for motion-based margin reduction in adaptive radiotherapy for early-stage glottis carcinoma.
SHR-1701 combined with fuzuloparib and chemotherapy as first-line therapy for advanced lung squamous cell carcinoma: Efficacy and safety results from a phase II study.
8569 Background: Although immune checkpoint inhibitors are established in the treatment of advanced lung squamous cell carcinoma (LUSC), many patients derive limited or no durable benefit, underscoring the need for novel therapeutic strategies. However, poly (ADP-ribose) polymerase inhibitors (PARPi) can upregulate PD-L1 expression and promote immune-mediated response, which may increase the efficacy of anti-PD-(L)1 based therapy. Additionally, blocking TGF-b signaling has the potential to facilitate the recovery from chemo-induced myelosuppression. This phase II study evaluates the efficacy and safety for first-line approach combining SHR-1701, a bifunctional anti–PD-L1/TGF-β trap, with the PARP inhibitor fuzuloparib, combined with standard chemotherapy, which may potentiate anti-tumor immunity. To our knowledge, this is the first clinical investigation of this multi-pathway targeting strategy in advanced LUSC. Methods: Treatment-naïve patients with advanced LUSC received induction therapy with SHR-1701 (30 mg/kg IV q3w) plus platinum-based chemotherapy (investigator’s choice) for 4 cycles. Patients with disease control proceeded to maintenance therapy with SHR-1701 (same dose) plus oral fuzuloparib (100 mg twice daily) until progression or unacceptable toxicity. The primary endpoint was progression-free survival (PFS) by Blinded Independent Review Committee (BIRC) per RECIST v1.1. Secondary endpoints included objective response rate (ORR), disease control rate (DCR), overall survival (OS), and safety. Results: Between February 2023 and June 2025, 71 treatment-naïve advanced LUSC patients were enrolled (median age 65 years; 96% male). As of data cutoff (December 1, 2025; median follow-up 17.4 months), 57 patients (80.3%) had started maintenance therapy (median 12 cycles). Disease progression or death had occurred in 27 of 71 patients (38.0%) by the cutoff date. Median PFS was 11.0 months (95% CI, 8.1 to not estimable). Among 69 response-evaluable patients, best overall response included 1 complete response (1.4%), 56 partial responses (81.2%, including 46 confirmed), and 11 stable disease (15.9%), yielding an unconfirmed ORR of 82.6% (57/69), a confirmed ORR of 68.1% (47/69), and a DCR of 98.6% (68/69). OS data were immature; the 12-month OS rate was 83.0% (95% CI 71.1–90.2). Treatment-related adverse events occurred in 94.4% (67/71) of patients, most commonly anemia (42.3%), increased blood creatinine (19.7%), and proteinuria (18.3%). No new safety signals were observed. Conclusions: SHR-1701 plus fuzuloparib and chemotherapy as first-line therapy demonstrated promising anti-tumor activity and a manageable safety profile in patients with advanced LUSC. Clinical trial information: NCT04937972 .
Real-world treatment patterns and outcomes of HER2-altered NSCLC in China: A retrospective multi-center study (CHAPTER Study).
e20502 Background: The HER2-targeted therapy trastuzumab deruxtecan (T-DXd) was approved for use in China in 2024, providing a novel option for patients with HER2-mutant non-small cell lung cancer (NSCLC). Despite its approval, data on treatment patterns and clinical outcomes in real-world settings remain scarce. This study aimed to analyze real-world treatment patterns and outcomes in HER2-mutant NSCLC, providing a foundation for optimizing HER2-targeted therapies in clinical practice. Methods: This retrospective analysis included clinical data from HER2-mutant NSCLC patients treated across four medical centers (Peking University Cancer Hospital & Institute, Peking University Cancer Hospital (Inner Mongolia Campus)/Affiliated Cancer Hospital of Inner Mongolia Medical University, The second affiliated hospital of Dalian medical university and The Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital) between December 2015 and April 30, 2025. Statistical analyses were performed using SPSS software and included log-rank tests, Kaplan-Meier survival analyses, univariate and multivariate models, and Clopper-Pearson interval estimation. Results: 404 patients with HER2-altered NSCLC were included, of whom 49.75% (201/404) had HER2 exon 20 insertion mutations. Among patients harboring HER2 mutations (with or without amplification) receiving first-line therapy, trastuzumab deruxtecan (T-DXd) monotherapy achieved a median real-world progression-free survival (rwPFS) of 19.40 months (N = 11), which was significantly longer than chemotherapy (N = 34, 5.90 months), immunotherapy-based regimens (N = 106, 8.20 months), non-T-DXd HER2-targeted therapy (N = 25, 6.07 months), and anti-angiogenic based therapy (N = 44, 8.47 months) (P = 0.04). For second-line therapy, the median rwPFS of T-DXd monotherapy (N = 25, 10.13 months) was numerically longer than chemotherapy (N = 11, 3.90 months) and IO-based therapy (N = 28, 6.03 months) (P = 0.25). Conclusions: This is the largest real-world cohort study of HER2-mutant NSCLC to date. Real-world treatment strategies for HER2-mutant NSCLC patients are highly variable. T-DXd demonstrated the most favorable outcomes in both the first-line and second-line settings, showing superior effectiveness compared to other available treatments. These findings underscore the potential of T-DXd as a key therapeutic option for HER2-mutant NSCLC patients. Future studies are warranted to further validate these real-world outcomes and explore optimal therapeutic strategies.
Preeclampsia gene signatures and PD-L1 in endometrial cancer.
e22589 Background: Aggressive endometrial cancers (EC), particularly p53-abnormal (p53abn), have poor outcomes despite immune-active features. Preeclampsia (PE), a pregnancy-specific hypertensive disorder, involves interferon-driven inflammation similar to anti-tumor immunity, yet whether PE-associated signatures identify immunotherapy-responsive EC phenotypes remains unexplored. We hypothesized that PE gene signatures might distinguish aggressive EC with distinct immune profiles relevant to immunotherapy selection. Methods: Transcriptomic and PD-L1 (22C3 IHC) data from 730 EC tumors (subtypes: p53abn, POLE, MSI-H, NSMP) were analyzed. Differential gene expressions between tumor subtypes were assessed using linear modeling with limma-voom. Gene set enrichment (GSEA) and over-representation analyses were conducted using Hallmark, GO Biological Process, KEGG_Medicus, and Reactome databases. Single-sample GSEA was performed to generate per-tumor PE scores from four published PE gene sets and a combined signature. Bipartite network clustering of PE scores defined PE-high/low groups among samples with detectable PE enrichment. Associations with PD-L1 were assessed via multivariable negative binomial regression adjusting for age, ancestry, stage, and molecular subtype. Results: Among 730 tumors analyzed, patient demographics included 45% aged 50–69 years; ancestry distribution of 41% European, 14% African, 16% other, and 29% missing; and molecular subtypes of p53abn (39%, aggressive), POLE (16%), MSI-H (13%), and NSMP (32%). Aggressive tumors showed immune activation with enrichment of interferon signaling (normalized enrichment score [NES] 1.42–1.56), antigen presentation (NES 1.45–1.50), and T-cell/leukocyte pathways (NES 1.40–1.51, GSEA FDR < 0.25). Over-representation analysis identified 290 immune pathways (FDR < 0.05), with T-cell activation, leukocyte adhesion, interferon response, and antigen processing among the most significant. PE signatures were enriched in aggressive tumors (combined PE OR 1.92, p=0.002), with 81/125 overlapping upregulated genes. PE enrichment correlated with PD-L1 expression (ρ=0.26, p=2.7×10⁻¹²) and independently predicted PD-L1 positivity (IRR 1.63 per SD, 95% CI 1.36–1.95, p=6.8×10⁻⁷), superseding molecular subtype (all p>0.21). Bipartite clustering identified PE-high (n=350) and PE-low (n=380) groups. PE-high tumors showed higher PD-L1 positivity (44% vs 28%, p=1.6×10⁻⁵), though expression intensity among positives was comparable (p=0.36). Conclusions: PE-associated gene signatures define an immune-active subset of aggressive EC characterized by interferon signaling and antigen presentation. These signatures correlate with PD-L1 expression and outperform molecular subtype in predicting PD-L1 positivity, suggesting potential utility for refining immunotherapy selection and warranting further clinical validation.
Epigenetic state-control and immune visibility as prognostic modules in breast cancer.
e12602 Background: Epigenetic reprogramming is proposed as an enabling dimension of cancer, but it remains unclear how expression programs of epigenetic regulators map onto downstream hallmark capabilities and clinical outcomes. We tested whether an expression-based “epigenetic state-control” module has a definable cancer hallmark footprint and subtype-specific prognostic relevance in breast cancer. Methods: We assembled a literature-derived 81-gene epigenetic state-control signature encompassing canonical chromatin regulatory systems (Polycomb, H3K9 repression, DNA methylation, SWI/SNF and NuRD remodeling, histone acetylation/deacetylation, arginine methylation, and 3D genome organization). The association with cancer hallmarks was evaluated using enrichment analysis on cancerhallmarks.com, with a multiple-testing correction (FDR). In parallel, we defined an 18-gene immune-visibility module that captures MHC-I antigen processing and presentation and IFN signaling. Prognostic performance for relapse-free survival (RFS) was assessed in KMplotter across PAM50 subtypes using the best cutoff, with multiple-testing adjustment (FDR < 10%). Results: The epigenetic regulator module showed broad hallmark associations, with strong enrichment for replicative immortality and genome instability, and additional enrichment for resisting cell death, metabolic reprogramming, and invasion-metastasis. In contrast, the epigenetic signature was not enriched for “evasion of immune destruction”. However, it was robustly captured by the immune-visibility module, supporting that immune evasion is best assessed using dedicated antigen presentation programs rather than inferred indirectly from chromatin regulators. In KMplotter, a higher epigenetic module score was associated with improved RFS in luminal A (HR 0.52, p = 0.00029) and luminal B (HR 0.61, p = 5×10⁻4), but not in basal-like or HER2-positive disease. Conversely, a higher immune-visibility score stratified RFS in basal-like (HR 0.32, p = 8.2×10⁻14), luminal B (HR 0.51, p = 3.5×10⁻5), and HER2-enriched tumors (HR 0.53, p = 1.4×10⁻4), but not luminal A. Conclusions: Epigenetic regulators exhibit a distributed hallmark footprint, dominated by immortality and genome instability. The divergent, subtype-specific prognostic patterns argue that epigenetic state-control and immune visibility are distinct axes, prioritizing tumors with high epigenetic state-control but low immune visibility as candidates for epi-drug combinations.
Efficacy and safety of donafenib combined with hepatic artery infusion chemotherapy and sintilimab as neoadjuvant therapy for hepatocellular carcinoma: A prospective, single-arm phase II trial (Neo-DoHAICs study).
e16176 Background: Recurrence is a major cause of treatment failure in hepatocellular carcinoma (HCC), especially for tumors with high-risk features such as multiplicity or vascular invasion. Yet, a standardized neoadjuvant regimen for resectable HCC is lacking. This study evaluates the efficacy and safety of a neoadjuvant combination of donafenib, sintilimab, and hepatic artery infusion chemotherapy (HAIC) in patients with resectable HCC at high risk of recurrence. Methods: This is a prospective, single-arm phase II study (NCT06512467). Patients with technically resectable CNLC IIb or IIIa HCC were included, provided they also had high-risk features such as an anticipated surgical margin < 1 cm or the presence of ipsilateral portal or hepatic vein tumor thrombus. Patients received 2-4 cycles of neoadjuvant therapy with donafenib (200 mg, bid), sintilimab (200 mg, q3w) and HAIC (oxaliplatin 85 mg/m 2 2h, leucovorin 400 mg/m 2 2h, fluorouracil bolus 400 mg/m 2 in the first 10 minutes, and fluorouracil infusion 1200 mg/m 2 for 23 hours, q3w) prior to elective surgical resection. The primary endpoint was the major pathological response (MPR) rate. Results: A total of 17 patients were enrolled from August 2024 to July 2025. 94.1% of patients were CNLC IIIa. The median maximum tumor diameter was 99.0 mm. 76.4% presented with Vp3/Vp4 portal vein tumor thrombus. 29.4% had hepatic vein tumor thrombus. To date, 16 patients have undergone surgical resection. Pathological assessment revealed the following rates: 31.3% for pathological complete response (pCR), 43.8% for MPR, 100.0% for R0 resection, and 25.0% for microvascular invasion. The objective response rate was 88.2% (2 complete responses, 13 partial responses [PRs]) per mRECIST and 52.9% (9 PRs) per RECIST 1.1, the corresponding median times to response were 1.6 months and 2.0 months, respectively. The disease control rate was 100.0% for both criteria. A association was found between radiographic and pathological response: 85.7% of radiographic responders achieved pCR or MPR, compared to none among non-responders ( P = 0.05). The 1-year event-free survival rate was 84.4%, while the 1-year recurrence-free survival rate was 73.9%. Both AFP and PIVKA-II levels decreased significantly from baseline to surgery. The albumin-bilirubin score did not show a significant change before surgery compared to baseline (-2.82 vs -2.86, P = 0.75). Treatment-emergent adverse events (TEAEs) occurred in 94.1% of patients. Grade 3-4 TEAEs were reported in two cases. The most common AEs were decreased platelet count and elevated ALT. Conclusions: Preliminary analysis indicates that the neoadjuvant combination of donafenib, HAIC, and sintilimab demonstrates favorable efficacy and safety in patients with HCC. The enrollment and follow-up are continuing. Clinical trial information: NCT06512467 .
NRG-GI008: Colon adjuvant chemotherapy based on evaluation of residual disease (CIRCULATE-NORTH AMERICA).
TPS3686 Background: Currently, there are no biomarkers validated prospectively in randomized studies for resected colon cancer (CC) to determine need for adjuvant chemotherapy (AC). However, circulating tumor DNA (ctDNA) represents a highly specific and sensitive approach (especially with serial monitoring) for identifying minimal/molecular residual disease (MRD) post-surgery in CC patients (pts), and may outperform traditional clinical and pathological features in prognosticating risk for recurrence. CC pts who do not have detectable ctDNA (ctDNA–) are at a much lower risk of recurrence and may be spared the toxicities associated with AC. Furthermore, for CC pts with detectable ctDNA (ctDNA+) who are at a very high risk of recurrence, the optimal AC regimen has not been established. We hypothesize that for pts whose CC has been resected, ctDNA status may be used to risk-stratify for making decisions about AC. Methods: In this prospective phase II/III trial, up to 1,912 pts with resected stage IIB, IIC, and III CC will be enrolled. Based on the post-operative ctDNA status using personalized and tumor-informed assay (Signatera, bespoke assay), those who are ctDNA– (Cohort A) will be randomized to immediate AC with fluoropyrimidine (FP)+oxaliplatin (Ox) for 3-6 mos per established guidelines v serial ctDNA monitoring. Patients who are ctDNA+ post-operatively, or with serial monitoring (Cohort B), will be randomized to FP+Ox v more intensive AC with addition of irinotecan (I) for 6 mos. One cycle of chemotherapy is allowed while awaiting ctDNA testing results for cohort assignment. The primary endpoints for Cohort A are time to ctDNA+ status (phase II) and disease-free survival (DFS) (phase III) in the immediate v delayed AC arms. The primary endpoint for Cohort B is DFS in the FP+Ox v FP+Ox+I arms for both phase II and phase III portions of the trial. Secondary endpoints include prevalence of detectable ctDNA post-operatively, time-to-event outcomes (overall survival and time to recurrence) by ctDNA status, and the assessment of compliance to adjuvant therapy. Biospecimens including archival tumor tissue, as well as post-operative plus serial matched/normal blood samples, will be collected for exploratory correlative research. Active enrollment across the NCTN started in June 2022 with CCTG sites joining in August 2023. Accrual is 1,123/1,912 (as of 1/20/2026). *Drs. Dasari and Lieu contributed equally to the development of this trial. Clinical trial information: NCT05174169 .
Phase 1b/2 trial of melphalan-percutaneous hepatic perfusion (PHP) therapy and nivolumab/relatlimab in patients with metastatic melanoma and liver metastasis.
TPS9600 Background: Despite advances with immune checkpoint inhibitors (ICIs), patients with metastatic melanoma and liver metastases (LM) experience poorer outcomes. In the CheckMate-067 subanalysis, patients with LM treated with nivolumab/ipilimumab had shorter median PFS (4.4 vs 18.1 months) and OS (28.2 months vs not reached) than those without LM, confirming its negative prognostic role. Nivolumab/ipilimumab and nivolumab/relatlimab show comparable efficacy; however, the latter is increasingly used for its improved safety profile (grade ≥3 adverse events: 59% vs 21%). Melphalan-PHP (Hepzato), a liver-directed therapy delivering high-dose melphalan through a closed filtration circuit, concentrates treatment in the liver while minimizing systemic exposure. Research by Yu et al. demonstrated that LM reduces peripheral and tumoral CD8+ T cells, creating an immune desert, while combining liver-directed therapy with ICIs can reactivate systemic immune responses. Based on these insights and the CHOPIN trial’s safety data, the proposed Phase 1b/2 study will assess the safety, tolerability, and preliminary efficacy of melphalan-PHP plus nivolumab/relatlimab for first line treatment of metastatic non-uveal melanoma patients with LM. Methods: Phase 1b/2, single-center, single-arm study will use a Simon two-stage (5+10) design (n=15). Patients will receive nivolumab/relatlimab on Day 1 of each 28-day cycle for up to 2 years and receive melphalan-PHP on Day 15 of Cycle 1 and Cycle 3. Eligible participants must have histologically or cytologically confirmed metastatic melanoma involving the liver and be systemic treatment-naïve in the metastatic setting. Prior adjuvant ICI is permitted provided it was completed >6 months prior to study enrollment. Key exclusions include metastatic uveal melanoma, prior treatment with melphalan-PHP or ICI combinations, uncontrolled or symptomatic brain metastases, prior grade 3+ adverse events from ICI requiring treatment discontinuation, ongoing immunosuppressive therapy, contraindications to anesthesia, and significant cardiac, hepatic (Child-Pugh B/C), or pulmonary disease. The co-primary objectives/endpoints in stage 1 are to assess the safety and tolerability as measured by 2 or less DLTs (any treatment-related grade 4+ non-hematologic event lasting >3 days); and preliminary efficacy as measured by 2 or more objective responses in hepatic and non-hepatic lesions. Secondary objectives include: disease control rate (DCR), progression-free survival (PFS), overall survival (OS), duration of response (DOR), and any tumor reduction. Open for enrollment in January 2026. Clinical trial registry number: NCT07281924. Clinical trial information: NCT07281924 .
The SERPENTINE trial: A phase II clinical study decoding immune responses by scRNA-seq and TCR-seq analysis in patients with MSS metastatic colorectal cancer (mCRC) with and without liver metastases treated with durvalumab and tremelimumab.
3607 Background: Immune checkpoint inhibitor (ICI) therapy aims to instruct immune cells to efficiently recognize and eliminate cancer cells. It has shown efficacy in MSI colorectal (CRC), and, although there is no approval yet, a subset of patients with MSS CRC included in clinical trials respond to ICI, indicating a significant knowledge gap in our current understanding of immunological mechanisms in MSS CRCs. The relationship between the cellular and molecular responses to ICI in the tumor microenvironment (TME) and the clinical outcome remains poorly understood. We sought to address this gap in understanding through a series of high-throughput functional genomics analyses. Methods: Patients (pts) with refractory MSS CRC were treated with durvalumab (Dur) 1500mg plus tremelimumab (Trem) 300mg at Week 0, followed by Dur 1500 mg Q4W. We characterized the cellular makeup of the metastatic TME using single-cell RNA-seq, and cataloged T-cells clonotypes within metastases and in the blood using single-cell and deep TCR-seq. We collected data for each patient before and upon treatment initiation to explore the dynamic changes of the TME and blood in response to ICI. To facilitate direct comparisons between metastatic and primary TMEs, we integrated our dataset with an external single-cell RNA-seq dataset from primary CRC tumors. Results: 24 pts were included in the analysis (18 with no liver metastases (NLM), 6 with liver metastases (LM)). Around 200,000 single-cell profiles were collected from biopsies taken from 6 pts with LM and 10 pts with lung metastases. Disease Control Rate (DCR) was 39% for MSS CRC without LM (2 PR + 5 SD) and 0% for those with LM (p=0.13). Robust gene expression changes within the immune compartment of the TME were observed upon treatment, irrespective of overall treatment outcome, yet specific biological programs were differentially altered across patient response groups. Differences in the composition of TMEs in the lung and liver metastases were found with an enrichment of T regulatory and IFN-responder CD4 populations in the lung vs more CD8+ T cell subtypes in the liver. Tumor TCR analysis revealed that, following ICI treatment, liver metastases displayed a high diverse TCR repertoire, whereas lung metastases were less diverse and showed marked clonal expansion, suggesting greater T cell activity within the lung lesions (p=0.035). Conclusions: Prospective deep immune repertoire profiling detected profound differences in TME according to dissemination patterns, which might explain differences in treatment outcomes. Despite low overall efficacy with Dur and Trem in MSS patients, a trend to better response rate and DCR was observed in pts without LM. We propose immune cell tracking as powerful tool to quantify the efficacy of ICI in real time. Clinical trial information: NCT06680739 .
Neoadjuvant treatment of recurrent high-grade gliomas with camrelizumab in combination with apatinib: A prospective phase II clinical study.
2061 Background: High-grade gliomas exhibit extremely recurrent and resistant to treatments, and require urgent therapeutic innovation. This phase II trial presents preliminary results of neoadjuvant camrelizumab (PD-1 inhibitor) combined with apatinib (VEGFR2 TKI) for recurrent high-grade gliomas. Methods: This single-center, single-arm, phase II trial (NCT04588987) enrolled patients (age >18, KPS ≥60) with recurrent high-grade gliomas suitable for re-resection after prior STUPP therapy. Patients received i.v. camrelizumab 200mg (D1) and oral apatinib 250mg (D1-D7). Surgery was performed 1 week post-drug discontinuation. Adjuvant treatment with i.v. camrelizumab 200 mg every two weeks (from 2 weeks ± 5 days post-surgery) in combination with continuous oral apatinib 250 mg/d (from 4 weeks post-surgery) was administered until death/unacceptable toxicity. Primary endpoint: overall survival (OS). Secondary endpoints: 1-year OS rate, 6-/12-month PFS rates, time to progression (TTP), KPS, and adverse events (AEs). Results: From October 2020 to December 2024, 28 patients were enrolled (median age 42 years [range 23-62]; 22 glioblastoma, 4 anaplastic astrocytoma, 1 anaplastic oligodendroglioma, 1 gliosarcoma). The 27 patients completed adjuvant therapy with follow-up, and 1 patients continue on adjuvant therapy. As of January 14, 2026, median OS for all patients was 11.100 months (95% CI: 8.818 -13.382), with 1-year OS rate 46.4%. Median PFS was 4.870 months (95% CI: 4.351 – 5.389), with 6-month PFS rate 39.3%. The 22 patients with recurrent glioblastoma (rGBM) showed the median OS of 11.670 months (95% CI: 8.303 -15.037), and the 1-year OS rate was 50.0%. The median PFS was 5.070 months (95% CI: 4.381 -5.759), with the 6-month PFS rate of 40.9% in the rGBM subgroup. Historical neoadjuvant anti-PD1 monotherapy for rGBM yielded OS 13.9 months and PFS 3.3 months (Nat Med. 2019). Conclusions: Camrelizumab in combination with apatinib neoadjuvant therapy showed excellent efficacy in recurrent high-grade gliomas, and was superior to PD1 inhibitor monotherapy neoadjuvant treatment in rGBM patients. Clinical trial information: NCT04588987 .
Record‐High Performance 2:17‐type SmCo Magnets via Fe‐Driven HRE Segregation
ABSTRACT Development of high‐performance SmCo magnets, simultaneously possessing high magnetic energy product ( BH ) max and low remanence temperature coefficient | α |, is critical for applications of wide‐temperature precision instruments. Conventional heavy rare‐earth (HRE) substitution improves temperature stability via antiferromagnetic coupling but inevitably sacrifices ( BH ) max , resulting in a persistent trade‐off between ( BH ) max and | α |. Herein, we propose a Fe‐HRE synergistic compositional‐design strategy that integrates Fe enrichment and HRE segregation to break this bottleneck. First‐principles calculations reveal that increasing Fe concentration provides a thermodynamic driving force for HREs segregation from the 1:5H cell boundary into the 2:17R matrix. Furthermore, molecular field simulations quantitatively demonstrate that HRE enrichment in the 2:17R phase enhances its temperature compensation effect and effectively overcomes this trade‐off. Guided by these insights, a series of Sm 0.4 Gd 0.6 (Co bal Fe x Cu 0.08 Zr 0.025 ) 7.2 ( x = 0.20–0.24) magnets are prepared. Magnetic and microstructural characterizations confirm that moderate Fe enrichment ( x = 0.22) not only improves ( BH ) max but also facilitates Gd segregation into 2:17R phase without microstructural degradations. These synergistic effects yield a record‐high ( BH ) max of 18.8 MGOe and α 20°C–300°C = −0.012%/°C. This work establishes a unified design framework integrating magnetic moment engineering with thermodynamic element distribution regulation, paving a viable path for high‐temperature‐stable SmCo magnets for aerospace precision instruments.
Fluoroacetate‐Mediated Dual‐Interface Ionic Stabilization in Perovskite Solar Cells
ABSTRACT Crystallization kinetics and ionic dynamics jointly govern the efficiency and stability of perovskite solar cells (PSCs). Here, we report a fluoroacetate‐mediated molecular strategy that regulates perovskite crystallization and ionic migration. Ethylammonium trifluoroacetate (EATFA) coordinates with lead and formamidinium ions, accelerating nucleation and moderating grain growth during the vacuum quenching process. Upon annealing, EATFA localizes at both top and bottom interfaces, where dual‐sided enrichment passivates deep‐level traps, enhances charge extraction, and suppresses light‐induced halide accumulation. Deep‐level transient spectroscopy (DLTS) and transient ion drift (TID) reveal that EATFA prevents the temperature‐induced transition of iodide ions into an interstitial‐mediated migration pathway observed in conventional films (activation energy decreased from 0.47 to 0.15 eV). Applied to 1.66‐eV wide‐bandgap PSCs, critical for silicon/perovskite tandems yet prone to ion‐migration degradation, this strategy enhances both power‐conversion efficiency (PCE) and operational stability under thermal, ultraviolet, and continuous stress, achieving 22.06% PCE in 1‐square‐centimeter blade‐coated devices and retaining 95% of initial efficiency after 2000 h of maximum‐power point tracking and 91% after 1000 h at 65°C. Similar improvements in 1.55‐eV perovskites confirm the bandgap‐independent nature of this approach, providing a unified route toward efficient and durable PSCs.
ENVELOPE trial (NCCH2412): Enfortumab vedotin in patients with locally advanced or metastatic small bowel adenocarcinoma refractory or intolerant to platinum-based combination therapy—A multicenter, phase II trial (trial in progress).
TPS4269 Background: Small bowel adenocarcinoma (SBA) is a rare malignancy with limited systemic options beyond first-line platinum-based combination therapy (FOLFOX/CapeOX) and no established second-line treatment. We recently reported that Nectin-4 is frequently highly expressed in SBA, providing a biological rationale for enfortumab vedotin, a Nectin-4–directed antibody-drug conjugate (ADC). We designed a multicenter phase II trial to evaluate the antitumor activity and safety of enfortumab vedotin in unresectable locally advanced or metastatic SBA refractory or intolerant to platinum-based combination therapy. To our knowledge, this is the first trial to evaluate the efficacy of ADC in this rare cancer. Methods: Key eligibility criteria include age ≥18 years, histologically/cytologically confirmed SBA, ECOG PS 0–1, prior FOLFOX or CapeOX with progression or intolerance, ≥1 measurable lesion (RECIST v1.1) and adequate organ function. If prior testing has identified alterations qualifying for approved tumor-agnostic therapies, patients must be refractory or intolerant to such therapies. Nectin-4 positivity is not required for enrollment. Exclusion criteria include a history of interstitial lung disease and poorly controlled diabetes. Enfortumab vedotin is administered at 1.25 mg/kg intravenously on days 1, 8, and 15 of each 28-day cycle until progression or unacceptable toxicity. Tumor assessments are performed by enhanced CT every 8 weeks up to week 24 and every 12 weeks thereafter. The primary endpoint is the objective response rate (ORR), as determined by blinded independent central review. Representative secondary endpoints include disease control rate, progression-free survival, overall survival, and safety outcomes, as assessed by the Common Terminology Criteria for Adverse Events v5.0. The null ORR is 5%, and the expected ORR is 25%. An exact binomial design (one-sided α = 0.05; 90% power) requires 25 evaluable patients, with the planned sample size of 27 to account for non-evaluable cases. In a prespecified translational study, tumor Nectin-4 expression will be assessed by immunohistochemistry, and its association with the efficacy of enfortumab vedotin will be explored. Biological specimens will be collected at baseline, during treatment, and at progression for multi-omics analysis to investigate mechanisms of response and resistance. The ENVELOPE trial has been approved by the Institutional Review Board of the National Cancer Center Japan. Enrollment opened on October 2025 and will continue through October 2027; follow-up will extend 12 months after the last patient is enrolled. As of December 31, 2025, 5 patients had been enrolled. Clinical trial information: NCT07347314 /jRCT2031250424.