Browse Articles

Impact of corticosteroid exposure on long-term immune checkpoint inhibitor responses in solid tumors.

12160 Background: Multiple analyses offer conflicting evidence regarding the implications of immune related adverse events (irAE) on survival of patients with solid tumors treated with immune checkpoint inhibitor(s) (ICIs). Furthermore, the treatments used to manage irAEs also have an unclear impact on progression free survival (PFS) and overall survival (OS) on patients with solid tumors treated with ICIs. We present the PFS, OS and tumor specific survival (TSS) for pts with metastatic solid tumors treated with ICIs at an academic cancer center. Methods: 848 pts with metastatic solid tumors diagnosed between 2014 through January 2026 were identified via retrospective chart review. PFS was defined as the time from starting ICIs to progression or death, OS was defined as the time from starting ICIs to death and TSS was defined as the time from starting ICIs to death from malignancy (when available). The corticosteroid (CS) exposure was defined as low if prednisone (or equivalent) < 1mg/kg/day and/or length of exposure < / = 21 days. The CS exposure was defined as high if prednisone (or equivalent) was > / = 1mg/kg/day and/or length of exposure > / = 22 days. The CS exposure was defined as very high if prednisone (or equivalent) > 20mg/day and length of exposure was > 42 days. Steroid sparing agents were considered low impact if given once and within 14 days of starting CS. Steroid sparing agents were considered high impact if given more than once. Steroid sparing agents were considered very high impact if given more than once and/or given after 14 days of starting CS. Acute irAE was defined if symptom and immunosuppression resolved within 42 days. Chronic irAE was defined if symptom and/or immunosuppression continued beyond 42 days. The effects of CS, steroid sparing agent exposure on PFS, OS, TSS were evaluated using multivariable Cox proportional hazards models. The effect of acute irAE versus chronic irAE on PFS, OS, TSS were evaluated using multivariable Cox proportional hazards models. Results: 64.8% of patients experienced an irAE and 44.7% of patients developed G3-4 irAE requiring corticosteroids +/- steroid sparing agent. The low impact corticosteroid exposure did not affect PFS, OS or tumor specific survival. The very high impact CS exposure was associated with decrease in PFS (p = 0.02) but not with OS or TSS (p = 0.3). The high impact steroid sparing agent was associated with decrease in PFS (not significant) but the very high impact steroid sparing agent was associated with decrease in PFS. The chronic irAE was associated with slight improvement in PFS (not significant). Conclusions: The presence of an irAE was not significantly associated with PFS, OS or tumor specific survival treated with ICIs. There was a negative impact on the length of corticosteroid exposure and steroid sparing agents on PFS, TSS. The delay or interruption in treatment due to irAE did not affect survival.

Updated results from inMMyCAR, the ongoing first-in-human phase 1 study of KLN-1010 in patients (pts) with relapsed and refractory multiple myeloma (RRMM).

7509 Background: KLN-1010 is a modified lentiviral vector generating a novel, fully human anti-BCMA CAR-T cell in vivo when administered intravenously without preparative chemotherapy to pts with RRMM. Here, we report updated findings from inMMyCAR, the first sponsored, multicenter study of an in vivo CAR-T therapy for pts with RRMM. Methods: Eligibility required RRMM with measurable disease, adequate end-organ and bone marrow (BM) function, and ≥3 prior lines of therapy, including a proteosome inhibitor, an immunomodulatory drug, and an anti-CD38 monoclonal antibody. Results: To date, 6 pts were infused across two dose levels (6×10 6 and 2×10 7 IU/kg). Pts ranged in age from 61-72 and had received 3-5 prior lines of therapy. 5 pts had high-risk cytogenetics. BM plasma cell involvement ranged from <5% to 80%. 1 pt had extramedullary disease (EMD). Median time from consent to infusion was 13.5 days. All pts experienced treatment-emergent adverse events. Infusion-related reactions (IRR) occurring in 3 pts (2 at 2×10⁷ IU/kg, 1 at 6×10⁶ IU/kg) were manageable and resolved within 6-48 hrs. Cytokine release syndrome (CRS) was observed in 4 pts, with median onset on day 11 and median duration of 3.5 days; all events were Gr 2 and were managed with tocilizumab and corticosteroids. No immune effector cell–associated neurotoxicity syndrome (ICANS) or delayed neurotoxicity were observed. Peak absolute lymphocyte counts occurred between days 13-22, with median counts of 6.9×10 9 /L (range 2.3-43.1×10 9 /L) and no associated clinical sequelae. CAR-T cells were detectable in peripheral blood through 4 months follow-up and were predominantly memory-phenotype by month 3 (M3). All 6 pts achieved minimal residual disease (MRD) negativity (5 at 10 -6 and 1 at 10 -5 sensitivity) at M1 in the BM. MRD negativity was maintained through M6 (10 -6 sensitivity) in the pt with the longest follow-up. All pts achieved a response by IMWG criteria, and responses deepened over time. The pt with EMD showed complete resolution by M1. Conclusions: Results from inMMyCAR continue to demonstrate compelling early clinical activity. All pts achieved early MRD-negative responses and deepening of IMWG response over time. The pt with the longest follow-up remains MRD-negative with a stringent CR at M6. Activity was also observed in EMD, with radiologic resolution within 1 month. KLN-1010 was generally well-tolerated, with manageable toxicities related to IRR and CRS, supporting potential feasibility for outpatient administration. No ICANS or delayed neurotoxicity were observed. The study remains ongoing; updated results will be presented. Clinical trial information: NCT07075185 . Pt Dose, ×10 7 IU/kg Timepoints of MRD-negative results a IMWG response Duration of follow-up, months 1 2 M1, M3, M6 sCR 6+ 2 2 M1, M3 VGPR 5+ 3 2 M1, M3 VGPR 4+ 4 0.6 M1, M3 VGPR 3+ 5 0.6 M1 sCR 1+ 6 0.6 M1 PR 1+ a By next-gen flow cytometry or sequencing with sensitivity of 10 -5 or 10 -6 .

A randomized phase III study of gemcitabine plus cisplatin with S-1 versus gemcitabine plus cisplatin with immune checkpoint inhibitor in unresectable biliary tract cancer.

TPS4252 Background: Gemcitabine plus cisplatin (GC) has been established as the standard first-line treatment for unresectable biliary tract cancer (BTC). Recent phase III trials have demonstrated the clinical benefit of adding immune checkpoint inhibitors (ICIs) to GC. Meanwhile, GC combined with S-1 has also shown promising efficacy in East Asian populations. However, no randomized phase III trial has directly compared GC plus ICI with GC plus S-1. This study aims to compare the efficacy and safety of GC plus ICI with GC plus S-1 in patients with unresectable BTC. Methods: This is an ongoing, multicenter, open-label, randomized phase III trial (UMIN000051689). Eligible patients have histologically confirmed unresectable biliary tract cancer, including intrahepatic cholangiocarcinoma, extrahepatic cholangiocarcinoma, and gallbladder cancer, with no prior systemic therapy. Patients are randomized in a 1:1 ratio to receive either Arm A: gemcitabine plus cisplatin plus S-1 or Arm B: gemcitabine plus cisplatin combined with an immune checkpoint inhibitor. Randomization is stratified by primary tumor site, disease status (recurrence vs primary unresectable), and planned immune checkpoint inhibitor regimen for patients assigned to Arm B. The primary endpoint is overall survival. Secondary endpoints include progression-free survival, objective response rate, disease control rate, safety, and health-related quality of life. The planned sample size is 460 patients (230 per arm), providing 80% power to detect a statistically significant difference in overall survival with a two-sided alpha of 0.05. Current Status: Patient enrollment began in August 2023. As of January 2026, approximately 60% of the planned patients have been enrolled. Enrollment is ongoing, and follow-up continues. Conclusions: This randomized phase III trial will clarify the optimal first-line systemic treatment for unresectable biliary tract cancer by directly comparing GC plus immune checkpoint inhibition with GC plus S-1. Clinical trial information: UMIN000051689.

Bridging the fertility divide: Disparities across counseling, referral, and utilization of fertility preservation in young breast cancer patients.

e13565 Background: As survival improves for reproductive-age women with breast cancer (BC), fertility preservation (FP) is a key component of high-quality oncologic care. ASCO and ASRM guidelines recommend early counseling and timely referral prior to gonadotoxic therapy; however, real-world FP delivery remains inconsistent. Prior studies often evaluate FP as a single outcome, limiting insight into where attrition occurs. We conducted a systematic review to map disparities across the FP continuum of counseling, referral, and utilization. Methods: We systematically reviewed observational studies evaluating FP among reproductive-age women with BC. Eligible studies reported rates of fertility counseling, referral to reproductive specialists, and/or FP utilization. Data were extracted without imputation or pooling. Disparities by race/ethnicity, socioeconomic status, insurance type, and geography were summarized. Given heterogeneity in outcome definitions, quantitative synthesis was restricted to studies with extractable denominators or adjusted estimates. Results: Four observational studies met inclusion criteria. Fertility counseling was reported in one institutional cohort, with 40 of 248 patients (≈16%) receiving counseling. Referral to fertility specialists was evaluated in one population-based study, with 178 of 4,452 patients referred (≈4.0%), demonstrating regional variation. FP utilization was reported in three studies and remained uniformly low, ranging from 0.6% in a large population-based cohort (206/36,468) to 3–4% in institutional cohorts (10/248 and 20/627). Across studies, non-Hispanic Black and Hispanic patients consistently demonstrated lower FP utilization than non-Hispanic White patients (adjusted prevalence ratio 0.31, 95% CI 0.21–0.46). Lower socioeconomic status, public insurance, and non-urban residence were also associated with reduced utilization. In contrast, associations between race or insurance status and counseling were weaker and inconsistent. Conclusions: Despite guideline endorsement, FP delivery in young women with BC demonstrated substantial attrition across the care continuum, with the greatest inequities occurring at utilization rather than counseling. These findings indicate that informational interventions alone are insufficient. Policy driven solutions including standardized referral mandates, insurance reform, and integrated oncology-reproductive care models are urgently needed to ensure equitable access to FP as a standard component of cancer care. Fertility preservation outcomes among young women with breast cancer. Outcome Study Events / N Observed Proportion Counseling Swain 2023 40 / 248 ~16% Referral Korkidakis 2019 178 / 4,452 ~4.0% Utilization Swain 2023 10 / 248 ~4.0% Utilization Mannion 2024 20 / 627 ~3.2% Utilization Meernik 2023 206 / 36,468 ~0.6%

Real-world impact of comprehensive genomic profiling in colon cancer: Evidence from community oncology centers in western India.

e15587 Background: NGS-based mutation profiling is increasingly used to guide oncology treatment. This study compares comprehensive genomic profiling (CGP) with limited panels & evaluates its role in precision medicine in a developing-country setting. Methods: This retrospective study included consecutive colon cancer patients from community oncology centers in Western India diagnosed between March 2018 & May 2025. Clinical, molecular & survival data were extracted from EMR & analyzed using the Kaplan–Meier method. Results: A total of 1100 colon cancer patients were evaluated who underwent NGS testing (10.54%; 116/1100) & the mean age was 63 years (IQR 54.3-72.0); with M:F ratio 1.9:1 & stage distribution being 13.8%, 38.8%, & 47.4% for Stage I/II, III & IV respectively. CGP was performed in 87.9% (102/116) to evaluate SNVs, CNVs & indels spanning across over 15000 loci for 1080 tumor while short panel testing was done in 12.0% (14/116) to assess mutation detection rates for key genes (BRAF, ERBB2, KRAS, NRAS, PIK3CA, TP53, NTRK) wherein 79.4% were tissue-based & 20.7% were liquid biopsy-based. Majority (59.5%) of the patients tested were therapy naive while 40.5% were pre-treated. The most common subtype in the cohort was adenocarcinoma (96.6%); commonly involved primary sites were sigmoid colon (n = 46) & ascending colon (n = 41) with a predominance of left-sided over right-sided disease (56.0% vs 38.8%). The common mutations reported were: TP53 (n = 59), KRAS (n = 41), APC (n = 25), PIK3CA (n = 14), BRAF(n = 9), HER2 (n = 3) NRAS (n = 1); comutation commonly observed were KRAS + TP53 ( n = 18) & BRAF + TP53 ( n = 6). MSI high & TMB high ( > 10 muts) were reported in 11.3%(10/88) & 26.6% (4/15) respectively. PD-L1 testing was done in 53.4% (62/116) with positivity (CPC/TPS > 1) in 51.6% (32/62). First line therapy mainly constituted chemotherapy (n = 94) & surgery (n = 81); common chemotherapy regimens were FOLFOX (n = 43) & CAPOX (n = 31). Targeted therapy was administered in first line to 19.8% (n = 22/111) mostly: Bevacizumab (n = 18), Cetuximab (n = 5) & regorafenib (n = 2) with 77.3% ORR. Immunotherapy was administered to 6% (7/116) with distribution: Pembrolizumab (n = 3), Nivolumab(n = 2) Dostarlimab (n = 1), Atezolizumab (n = 1) & 50% CBR. Among first line recipients, best responses were complete response in 30.6%, partial response in 33.3% & stable disease in 3.6%. Median overall survival (mOS) was 21.4 mo in BRAF mutant cohort vs 49.5 mo in other mutations (p = 0.035). mOS for NRAS/KRAS mutant cohort was 28.8 mo vs not reached in others (p = 0.032). TP53 mutant cases had mOS of 28.8 mo vs 49.5 mo in non-mutants (p = 0.191). mOS was not reached in immunotherapy marker mutant patients vs 49.5 mo in non-mutants (p = 0.091). Conclusions: This study supports CGP as a clinically impactful tool in precision oncology emphasizing its importance & improving its integration and accessibility in resource-limited healthcare settings.

Development and clinical utility of an H&E-based tumor origin prediction model.

3021 Background: Tumor of unknown primary origin remains a critical unmet need in oncology, as histologic and anatomic diagnoses are essential for ensuring effective and timely therapies. We previously developed the Tempus Tumor Origin (TO) Laboratory Developed Test (LDT), an RNA sequencing-based machine learning algorithm that discriminates between clinically relevant cancer subtypes. However, RNA-sequencing data is not always available, with a failure rate that can be > 20% in the literature. In this study, we develop and assess the accuracy of a hematoxylin and eosin (H&E)-based machine learning algorithm to predict tumor origin. We also explore combining predictions from H&E and RNA models to improve accuracy further. Methods: We curated a cohort of 93,770 H&E whole slide images (WSIs) of tumor biopsies and surgical resections across 67 cancer subtypes from 73,634 patients. The cohort was split into 60/20/20 for training, validation, and testing, stratified by cancer subtypes and clinical covariates. We extracted tile embeddings from WSIs using a foundation model, H-optimus-0, and trained an attention-based multiple instance learning model to predict cancer subtype labels derived from clinically abstracted documents. Using a subset of 54,878 samples that also have predictions from the RNA-based Tempus TO algorithm, we trained and evaluated a multilayer perceptron multimodal model that takes the prediction scores from the H&E model and the RNA model as input. Results: On 7,092 test set samples that had both modalities available, the H&E, RNA, and multimodal models achieved top-1 accuracies of 0.85, 0.91, and 0.92 and top-3 accuracies of 0.94, 0.97, and 0.97, respectively. Among subtypes with at least 30 samples in the test set, the multimodal model statistically significantly outperformed the RNA model on five subtypes and underperformed on two. In addition, the H&E model had a top-1 accuracy of 0.77 and a top-3 accuracy of 0.90 on 3,028 samples in the test set that failed RNA sequencing due to quantity insufficient (QNS). Table 1 presents H&E model performance on the RNA QNS samples by diagnosis subtypes. Conclusions: We developed a high-performing H&E-based algorithm to predict tumor origin. With shorter turnaround time and wider availability, the H&E model can potentially deliver accurate tumor origin predictions faster and, critically, in cases where RNA-seq is unsuccessful or unavailable. Furthermore, leveraging both H&E and RNA in a multimodal model may improve the performance of future tumor origin algorithms. H&E model performance on RNA QNS test set samples in the five most prevalent diagnosis subtypes and other subtypes combined. Diagnosis subtype N Top-1 accuracy (%) Top-3 accuracy (%) Lung adenocarcinoma 497 84 94 Prostatic adenocarcinoma 387 92 95 Breast carcinoma 348 83 95 Colorectal adenocarcinoma 327 85 93 Pancreatic adenocarcinoma 272 78 94 Other 1,197 64 83

Gradient Valence Engineering Synchronizes Charge‐Carrier and Catalytic Dynamics for Efficient Solar Water Oxidation

ABSTRACT The efficiency of photoelectrochemical water splitting is constrained by the kinetic mismatch between ultrafast charge separation and slow catalytic turnover. Inspired by the spatiotemporal precision of photosystem II, we designed a redox‐engineered BiVO 4 /Fe‐HOTP (BVO/R‐Fe‐HOTP) photoanode, with ‘R‐’ denoting the sample subjected to sequential NaBH 4 reduction and O 2 oxidation treatment (where HOTP refers to the 2,3,6,7,10,11‐hexaoxidotriphenylene multidentate ligand). This architecture establishes a programmable valence gradient that bridges charge separation and catalytic water oxidation. Through controlled redox engineering, we grew an amorphous Fe‐HOTP layer on BVO, establishing a continuous transition from electron‐rich Fe δ+ ( δ < 2), at the interface, to highly oxidized Fe 3+ , at the outer surface. Under light illumination, surface Fe 3+ is further oxidized to Fe 4+ , generating active redox sites that enable a turnover frequency (TOF) of 82 s −1 . This architecture reduces interfacial band offsets for ultrafast hole injection and establishes a built‐in potential gradient that extends carrier lifetime to 0.03 s. Thus, the BVO/R‐Fe‐HOTP photoanode delivers a photocurrent density of 6.1 mA cm −2 at 1.23 V RHE and, when coupled with a Si solar cell, achieves unbiased solar water splitting with a solar‐to‐hydrogen efficiency of 4.58%. These results establish gradient valence engineering as an effective strategy for synchronizing charge‐carrier and catalytic dynamics.

Nanoparticle shape effects on MHD radiative hybrid nanofluid flow over a rotating cone with quadratic chemical reaction

Dual-vector gene therapy for hearing loss secures first FDA approval

Translating ferroptosis into oncology: challenges, opportunities and future directions

Tailoring of Layered Bismuth‐Based Materials: Advanced Functionalities for Environment, Energy, Photonics, Electronics, and Biomedicine

ABSTRACT Layered bismuth‐based materials exhibit exceptional potential for a wide range of advanced functional applications, enabled by their versatile structural chemistry and tailorable physicochemical properties. This review begins by summarizing their fundamental structural classifications, including layered bismuth oxides, layered bismuth chalcogenides, and other layered bismuth compounds, along with their intrinsic physicochemical properties. It then discusses strategies for precise property modulation, such as atomic‐scale structural engineering, bandgap tailoring, polarization control, and heterointerface engineering. Focusing on their diverse and rapidly expanding utility, the article highlights applications in environmental remediation, energy conversion and storage, biomedical systems, and next‐generation electron devices, including post‐Moore core electronics, topological and quantum devices, chip‐enabling technologies, as well as the construction of high‐performance sensing, micro‐energy, neuromorphic computing, and heterogeneously integrated systems. Key challenges and future directions toward practical implementation are also analyzed. In summary, this review provides a comprehensive guide for the rational, application‐oriented design of these materials across interdisciplinary fields.

Effect of intratumoral alpha radiation on tumor growth delay and tumor microenvironment in an orthotopic colorectal liver metastasis murine model

Potassium suppresses allosteric activation of ZAP-70-dependent T cell receptor signaling

Updated analysis of adjuvant chemotherapy after radical resection of metachronous colorectal cancer metastases.

e15618 Background: The role of adjuvant systemic therapy following complete (R0) resection of metachronous colorectal cancer (CRC) metastases or local recurrence remains a subject of debate. This non-randomized study was designed to assess outcomes associated with adjuvant chemotherapy compared with surgery alone after radical resection of metachronous CRC metastases. Methods: Patients aged ≥18 years with histologically confirmed CRC and resectable metachronous metastases (disease-free interval ≥6 months) at any site were assigned to either surgery alone (active surveillance) or adjuvant mFOLFOX6 for 6 months. The primary endpoints were 2-year disease-free survival (DFS) and second disease-free survival (DFS2). Non-inferiority was defined as an absolute difference in 24-month DFS between the surgery-alone and adjuvant chemotherapy groups not exceeding 10%, with the upper bound of the one-sided 95% confidence interval below this margin. Results: Between June 2008 and December 2022, 145 patients were enrolled and assigned to the surgery alone or ACT groups; baseline demographic and clinicopathologic characteristics were well balanced between groups. Median follow-up was 39.8 months (95% CI, 34.3–52.5) in the surgery alone and 45.3 months (95% CI, 39.1–52.5) in the ACT group. Median DFS was 24.4 months (95% CI, 14.6–41.2) and 17.9 months (95% CI, 14.4–26.7), respectively. Two-year DFS rates were 51.2% (95% CI, 39.1–67.0) and 43.2% (95% CI, 33.7–55.3) (p = 0.30). Non-inferiority of surgery alone compared with ACT for 2-year DFS was demonstrated according to the prespecified margin (non-inferiority p = 0.022). In post-progression analysis, DFS2 was significantly longer in the surgery alone group (median 70.4 months [95% CI, 32.7–NR] vs 35.5 months [29.3–50.8]; HR 0.52, 95% CI 0.31–0.89, p = 0.014). After progression, repeat surgery without chemotherapy was more common in the non-adjuvant group (33% vs 19%), whereas surgery followed by chemotherapy was more frequent in the ACT group (22% vs 38%). The majority of patients in both groups subsequently received palliative systemic therapy (44% vs 43%). Conclusions: In patients with completely resected metachronous CRC metastases, omission of adjuvant chemotherapy did not compromise DFS and was associated with longer DFS2. These findings support a more individualized approach to adjuvant treatment selection. Prospective randomized studies are warranted to confirm these observations.

A first-in-human study of ATX-295, an oral inhibitor of KIF18A, in patients with advanced or metastatic solid tumors, including ovarian cancer.

TPS3167 Background: ATX-295 is an oral inhibitor of KIF18A, an adenosine triphosphate (ATP)-dependent, plus end-directed mitotic kinesin. KIF18A facilitates chromosomal alignment and spindle microtubule dynamics during mitosis in certain advanced solid tumors with chromosomal instability (CIN). In preclinical studies, ATX-295 demonstrated robust antiproliferative activity in chromosomally instable solid tumor models, including in high grade serous ovarian cancer. In vivo, oral administration of ATX-295 induced dose-dependent tumor growth inhibition in CIN platinum-resistant ovarian and squamous non-small cell lung cancer cell-line derived xenograft (CDX) and patient derived xenograft (PDX) models. Methods: NCT06799065 is a multi-center, first-in-human, Phase 1/2, open-label, single-arm, dose-escalation and expansion study to evaluate the safety profile of ATX-295 and determine the recommended phase 2 dose (RP2D) in subjects with locally advanced or metastatic solid tumors. The study in progress will be conducted in two parts: dose escalation, followed by dose expansion. Participant enrollment and continuous safety assessment will be guided by a mTPI-2 design (Guo, 2017) to identify an optimal dose. To assess evidence of preliminary antitumor activity, a Simon 2-stage design (Simon, 1989) will be used during dose expansion. Key eligibility criteria: patients with histologically confirmed solid tumors who have locally recurrent or metastatic disease, including high grade serous ovarian cancer (HGSOC) and squamous non-small cell lung cancer (sqNSCLC); refractory to or relapsed after all standard therapies with proven clinical benefit. For the expansion cohorts, participants must have histological confirmation of HGSOC and be determined to be platinum-resistant, platinum-refractory, or platinum-intolerant; have measurable disease; have an Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 1. Primary endpoints will evaluate safety and tolerability of ATX-295 at pharmacologically active dose(s) and/or schedule(s) and determine the recommended phase 2 dose (RP2D). Secondary endpoints will evaluate pharmacokinetics (PK) of ATX-295 in plasma and evaluate pharmacodynamic (PD) effects of ATX-295 to assess preliminary evidence of anti-tumor activity. The study is actively recruiting. Clinical trial information: NCT06799065 .

Avelumab + sacituzumab govitecan (Ave + SG) vs avelumab monotherapy (Ave mono) as first-line (1L) maintenance treatment for advanced urothelial carcinoma (aUC): Updated subgroup analyses of JAVELIN Bladder Medley based on metastatic sites.

4568 Background: Ave 1L maintenance is a recommended treatment option for patients (pts) with aUC without progression after 1L platinum-based chemotherapy (PBC). In the JAVELIN Bladder Medley phase 2 trial (NCT05327530), 1L maintenance with Ave + SG (Trop-2–directed antibody-drug conjugate) improved progression-free survival (PFS) vs Ave mono (primary endpoint). In pts with aUC, presence of visceral metastases (including liver or lung) is associated with poorer prognosis. We report updated subgroup analyses from the primary analysis of JAVELIN Bladder Medley based on metastatic site. Methods: Pts with unresectable locally advanced or metastatic UC without progression after 4-6 cycles of 1L PBC were randomized 2:1 to receive Ave + SG or Ave mono, stratified by presence of visceral metastases at start of 1L PBC. Primary endpoints were investigator-assessed PFS (measured from randomization) and safety; overall survival (OS) was a secondary endpoint. For visceral and nonvisceral subgroups, PFS and OS data in the Ave mono arm were extended per protocol using propensity score–weighted data from the JAVELIN Bladder 100 phase 3 trial; extended data were not available for other subgroups. Results: At the start of 1L PBC, of 74 and 37 pts in the Ave + SG and Ave mono arms, respectively, 37 (50.0%) and 19 (51.4%) had visceral metastases, 20 (27.0%) and 11 (29.7%) had lung metastases, 17 (23.0%) and 7 (18.9%) had liver metastases, 17 (23.0%) and 11 (29.7%) had bone metastases, and 26 (35.1%) and 11 (29.7%) had lymph node–only disease. At data cutoff (Apr 28, 2025), in the Ave + SG and Ave mono arms, respectively, median follow-up for PFS was 15.7 and 25.1 mo. Across all subgroups, PFS was prolonged with Ave + SG vs Ave mono (Table); OS analyses were immature. In the Ave + SG and Ave mono arms, respectively, grade ≥3 treatment-related adverse events occurred in 72.2% and 5.6% of pts with visceral metastases, 57.9% and 0% of pts with lung metastases, 82.4% and 0% of pts with liver metastases, 82.4% and 0% of pts with bone metastases, 70.3% and 5.6% of pts with nonvisceral metastases, and 76.9% and 9.1% of pts with lymph node–only metastases. Conclusions: In the primary analysis of JAVELIN Bladder Medley, Ave + SG as 1L maintenance improved PFS vs Ave mono, irrespective of metastatic site. Clinical trial information: NCT05327530 . PFS, median (95% CI), mo Ave + SG Ave mono HR (95% CI) Site of metastases at start of 1L PBC Visceral* 9.03 (5.62-13.80) 2.20 (1.91-3.71) 0.49 (0.28-0.84) Nonvisceral 14.69 (7.43-NE) 7.33 (4.21-11.10) 0.61 (0.33-1.13) Lung 8.77 (4.17-9.46) 1.81 (0.07-9.26) 0.43 (0.17-1.09) Liver 8.77 (5.55-9.36) 2.69 (1.91-NE) 0.48 (0.17-1.32) Bone 9.26 (5.49-17.64) 2.33 (0.07-5.45) 0.40 (0.16-0.99) Lymph node only 14.00 (7.43-NE) 7.59 (1.87-NE) 0.65 (0.26-1.66) *Pts could have ≥1 site of visceral metastases.HR, hazard ratio; NE, not estimable.

Impact of earlier time of first cycle immunotherapy administration on progression-free survival in real-world metastatic non–small cell lung cancer.

e20579 Background: Immune checkpoint inhibitors (ICIs) have revolutionized the treatment of advanced non-small cell lung cancer (NSCLC). Recent evidence suggests that earlier time of day (ToD) ICI administration is associated with improved clinical outcomes across multiple cancer types, implicating circadian regulation of immune function in immunotherapy efficacy. Data in NSCLC, particularly from real-world and racially and ethnically diverse populations, remain limited. We investigated the association between ToD of immunotherapy administration and survival outcomes in a racially and ethnically diverse population with metastatic NSCLC. Methods: We conducted a retrospective cohort study of patients with Stage IV NSCLC treated with ICIs between 2014 and 2025. Time of day of the first ICI infusion was extracted from infusion start times. Patients were stratified into earlier and later ToD groups using median, tertile, and quartile cutoff analyses. Progression-free survival (PFS) and overall survival (OS) were estimated using Kaplan-Meier methods and Cox proportional hazards models adjusted for age, sex, race/ethnicity, ECOG performance status, histology, PD-L1 expression, and socioeconomic status. Results: A total of 215 patients were included. 42% were Black, 30% Hispanic, and 28% Caucasian/Asian. Late ToD group was defined as ≥11:30 am for the median analysis, ≥2:30 pm for the tertile analysis, and ≥3:02 pm for the quartile analysis. PFS and OS did not differ significantly by race or ethnicity. Later time of ICI administration was associated with inferior PFS across median (adjusted HR 1.46, 95% CI 1.04-2.04, p=0.028), quartile (HR 1.87, 95% CI 1.17-2.98, p=0.0009), and tertile (HR 1.95, 95% CI 1.3-2.92, p=0.001) analytic groups when compared with earlier administration. A non-significant trend toward improved OS with earlier ToD administration was observed. ECOG performance status ≥2 was independently associated with poorer PFS and OS across all groups. Higher socioeconomic distress, measured by patient zip codes and the Distressed Communities Index, was independently associated with poorer OS in both univariate and multivariate analysis. Conclusions: In this real-world study of metastatic NSCLC, earlier first-cycle ICI administration was associated with significantly improved PFS and a consistent trend toward improved OS, regardless of race or ethnicity. Performance status and socioeconomic distress were also independently associated with survival outcomes. These findings suggest that the circadian timing of ICI administration may impact immunotherapy efficacy, warranting further investigation of chronotherapy-informed ICI administration.

Pulmonary embolism-related mortality in pancreatic cancer patients aged ≥45 years in the United States, 1999-2023: A CDC WONDER analysis.

e16411 Background: Pancreatic cancer is among the most prothrombotic malignancies; however, long-term national trends in pulmonary embolism (PE)–related mortality within this population remain poorly defined. We evaluated temporal patterns and demographic disparities in PE-associated mortality among individuals with pancreatic cancer. Methods: We analyzed U.S. mortality data from the CDC WONDER database (1999–2023), identifying decedents with pancreatic cancer (ICD-10 C25) and PE (ICD-10 I26) listed as underlying or contributing causes of death. Age-adjusted mortality rates (AAMRs) were calculated using the 2000 U.S. standard population. Joinpoint regression was used to estimate annual percent change (APC) and average annual percent change (AAPC), stratified by age, sex, race, census region, and urbanization. Results: Overall PE-related mortality among individuals with pancreatic cancer increased significantly over the study period (AAPC 4.4%, 95% CI 3.7–5.1; P < 0.001). Rising trends were consistent across all census regions (AAPC range 4.0%–4.5%) and in both metropolitan (4.0%) and non-metropolitan (3.9%) areas. Age-stratified analyses demonstrated steady growth among adults aged 45–64 years (AAPC 4.3%), while those aged ≥65 years experienced a marked inflection in 2018, followed by rapid increases of 9.8% annually (95% CI 6.2–13.5). Significant long-term increases were observed among White (AAPC 4.5%) and Black or African American individuals (AAPC 4.6%). Sex-specific trends showed initial increases through 2019, followed by abrupt annual declines in males (−18.0%) and females (−18.5%), resulting in non-significant overall 24-year trends by sex. Most deaths occurred in medical facilities (60.1%, n = 6,135). Texas (n = 146) and Ohio (n = 105) reported the highest absolute numbers of deaths. Conclusions: PE-related mortality among individuals with pancreatic cancer has increased substantially over the past two decades, driven largely by a sharp rise in adults aged ≥65 years since 2018. The predominance of inpatient deaths suggests persistent vulnerability despite hospital-based thromboprophylaxis strategies. The recent decline in sex-specific rates after 2019 warrants further investigation to determine whether it reflects evolving anticoagulation practices, changes in clinical management, or shifts in cause-of-death reporting.

Predictors of persistent opioid use following curative-intent radiation for head and neck cancer.

e18099 Background: Head and neck cancer (HNC) patients often experience pain from intense radiation treatment. Opioids work well to minimize this side effect. However, persistent use during and after treatment can cause significant health problems. There is little data in the literature that can help predict a predisposition for prolonged opioid use in HNC patients who receive radiation. Analysis of such data could aid HNC providers and patients in making appropriate and informed pain management choices. Methods: This study reviewed a retrospective single-institution cohort of HNC patients treated with definitive radiation, chemoradiation, or induction chemotherapy followed by chemoradiation between October 2011 and December 2023. Multivariable logistic regression was used to identify factors associated with continued opioid use at three and six months following radiation. Fine-Gray competing-risk regression evaluated time to first opioid use among opioid-naïve patients. Results: Among 789 patients (81% male; median age 61 years), 95.1% received opioids within 6 months of diagnosis. 40% of these patients continued opioid use three months after treatment and 15% continued use six months following treatment. Patients who continued opioid user at three months post treatment did so independently of other factors that were evaluated such as former smoking, radiation treatment alone, prior opioid use, and palliative medicine involvement. Continued use after six months was associated with opioid use at three months and palliative medicine involvement. Thin weight status and palliative medicine involvement was associated with earlier opioid initiation, while treatments of induction chemotherapy and radiation alone were associated with delayed opioid initiation. Interestingly, patients with advanced disease when care began, had lower long-term opioid risk. Conclusions: When HNC patients get curative radiation treatment, long-term opioid use is common. Early identification of high-risk patients may facilitate targeted opioid treatment strategies that help with pain management while also reducing long-term opioid dependence.

Anal cancer incidence among Medicaid enrollees with HIV in the United States.

e16479 Background: Anal cancer incidence remains elevated among people with HIV (PWH); however, incidence among PWH enrolled in Medicaid, particularly across U.S. geographic regions, remains poorly characterized. Approximately 40% of U.S. adults with HIV under age 65 are enrolled in Medicaid, and anal cancer screening was formally recommended for PWH by the U.S. Department of Health and Human Services in 2024. Characterizing anal cancer incidence in this population is therefore a critical public health and cancer prevention priority. Methods: We used the National Medicaid Database to estimate anal cancer incidence among PWH from 2016–2020. Anal cancer (ICD-10 C21) and HIV diagnoses (B20–B24, Z21, R75, B97.35) were identified using claims data. Individuals with < 12 months of Medicaid enrollment, dual Medicare–Medicaid eligibility, or anal cancer diagnosed prior to HIV diagnosis were excluded. Absolute incidence rates (per 100,000 person-years) were estimated nationally and by region, state, and county, restricting geographic estimates to areas with ≥10 cases. Sex-stratified incidence rates were estimated at national and regional levels. Results: Between 2016–2020, 1,047 anal cancer cases (246 females, 801 males) were diagnosed among 611,305 PWH. Overall anal cancer incidence was 52.7 per 100,000 person-years, with substantially higher incidence among males (68.7 per 100,000) than females (30.0 per 100,000). Marked regional variation was observed among males, with the highest incidence (per 100,000) in the Midwest (97.2) and South (94.7) and the lowest in the Northeast (54.6) and West (55.4). Among females, incidence rates were comparable in the Midwest (26.9), Northeast (27.6), and West (24.1), but were relatively higher in the South (40.2). At the state level, Missouri had the highest incidence (122.0), followed by Georgia (112.2) and Ohio (94.7), whereas New York (37.2) and New Jersey (42.3) had the lowest incidence in the nation. At the county level, the lowest incidence was observed in New York State counties—Queens (28.0), Kings (35.6), Bronx (46.9), and New York County (49.6)—as well as Los Angeles County (47.5). Conclusions: Anal cancer screening has been recommended in New York State since the mid-2000s under guidelines issued by the New York State Department of Health AIDS Institute. The consistently lower anal cancer incidence among PWH in New York and New Jersey, particularly within New York counties, may suggest a potential population-level benefit of early and sustained screening efforts. This study also demonstrates substantial geographic disparities, with the highest incidence rates observed in the Midwest and South, underscoring the importance of targeted prevention efforts in these regions.