Browse Articles

Electronic health record–based gut disruption and perfluoroalkyl and polyfluoroalkyl substances drinking-water proxies in postmenopausal hormone receptor–positive breast cancer: A nested case-control study.

e24180 Background: Environmental exposures and gut microbiome disruption may influence hormone receptor-positive (HR+) breast cancer (BC), but large-scale studies are limited by lack of exposure biomarkers and stool profiling. To address this gap, we leveraged extensive electronic health record data in the NIH All of Us research program dataset to find proxies for these potentially related biomarkers. Methods: We conducted a matched nested case-control study in the NIH All of Us Controlled Tier using conditional logistic regression. We matched cases and controls on age, race, sex, and observation time. We defined a proxy of postmenopausal status as age ≥ 55 years at the index date. We defined incident HR+ BC as breast cancer with evidence of endocrine therapy. We derived a Perfluoroalkyl and Polyfluoroalkyl Substances (PFAS) drinking-water proxy, cumulative environmental load (CEL), by linking ZIP3 to U.S. EPA UCMR5 PFAS occurrence data published for 2023-2025. We created CEL by detect-frequency-weighting PFAS concentrations, winsorizing, applying log1p transformation, and SD-standardizing. We also evaluated a rank-based CEL in sensitivity analysis. We defined a gut disruption proxy (GDP) as ≥ 30 cumulative antibiotic days in the prior 5 years and/or prior GI dysfunction diagnoses (IBS, IBD, or C. difficile). We adjusted for area-level deprivation and insurance-related covariates as available, and tested a CEL and GDP interaction. Results: We excluded 24 non-informative matched sets and analyzed 13,696 observations across 4,582 strata (4,582 cases; 9,114 controls). GDP was associated with higher odds of HR+ BC (adjusted OR 4.37; 95% CI 3.82-4.99; p < 0.001). CEL was not associated with HR+ BC under the primary specification (OR 1.00 per SD; 95% CI 0.96-1.04; p = 0.92), and GDP did not modify the CEL association (interaction OR 0.97; 95% CI 0.82-1.15; p = 0.75). Rank-based CEL showed a modest association (OR 1.12; 95% CI 1.07-1.17; p < 0.001) without interaction (p = 0.25), suggesting potential sensitivity of PFAS–HR+ BC associations to exposure scaling and distributional assumptions. GDP components were each associated with HR+ BC, including antibiotic burden (OR 1.73 per SD; 95% CI 1.64-1.82; p < 0.001) and GI dysfunction history (OR 4.84; 95% CI 3.82-6.13; p < 0.001). Conclusions: An EHR-derived gut disruption proxy was strongly associated with incident postmenopausal HR+ BC, though this could reflect healthcare utilization bias within the dataset in addition to biological effects. Broadness of the ZIP3-linked UCMR5 PFAS drinking-water proxy likely contributed to no association under the primary log1p SD-standardized specification and no evidence of synergy with gut disruption, although rank-based scaling yielded a modest association.

Cytisinicline for smoking cessation among patients with a history of cancer: A post-hoc analysis from two phase 3 clinical trials.

e22676 Background: Smoking cessation may reduce recurrent and secondary cancers, and extend survival. Cytisinicline, a plant-based alkaloid that acts as a partial agonist of α4β2 nicotinic acetylcholine receptors, significantly increased smoking cessation vs placebo in two phase 3 trials (ORCA-2 [NCT04576949]; ORCA-3 [NCT05206370]). We present a post-hoc analysis evaluating the outcomes from participants who reported a history of cancer at baseline. Methods: The two RCTs enrolled a similar population of adults who currently smoked ≥10 cigarettes per day and who had at least 10ppm exhaled carbon monoxide values at baseline. Volunteers were randomly assigned to behavioral support with either cytisinicline 3 mg three times daily for 6 or 12 weeks, or placebo. This post-hoc analysis evaluated outcomes in 120 of the 1602 participants who reported a baseline cancer history (placebo, n = 38; cytisinicline 6 wks, n = 45; cytisinicline 12 wks, n = 37). The primary efficacy outcome was biochemically confirmed continuous smoking abstinence during the last 4 wks of treatment. Treatment-emergent adverse events (TEAEs) and associated treatment discontinuation were evaluated. Results: Among those included in this analysis, 77 patients (64.2%) were female; 108 patients (90.0%) were white; mean age 60.6 (SD 9.4) years; mean duration of smoking 44.4 (SD 10.9) years. Most common malignancies were basal cell carcinoma (n = 31), breast cancer (n = 21), and squamous cell carcinoma (e.g. lung, skin; n = 19). Cytisinicline increased continuous abstinence for the 6-wk treatment group for 11/45 (24.4%) vs 4/38 (10.5%) placebo at wks 3–6 (odds ratio [OR] 2.75 [95% CI 0.81–10.74]); and in the 12-wk cytisinicline treatment group for 12/37 (32.4%) vs 6/38 (15.8%) placebo at wks 9–12 (OR 2.56 [95% CI 0.84–8.24]). In the entire pooled cohort, cytisinicline was well tolerated (most common TEAEs observed in ≥5% of patients and with a frequency ≥2% higher in the cytisinicline vs placebo group were insomnia and abnormal dreams), with no serious treatment-related TEAEs, and treatment discontinuation due to adverse events of < 3% across treatment groups. Conclusions: Cytisinicline was well tolerated and showed numerically higher smoking cessation rates compared with placebo in a small cohort of patients with a history of malignancy. Smoking cessation remains an underutilized but highly modifiable determinant of cancer outcomes, including treatment efficacy, recurrence, and survival. While exploratory in nature, these findings suggest that cytisinicline offers a promising pharmacologic option to support smoking cessation in cancer survivors. Clinical trial information: NCT04576949 ; NCT05206370 .

Complex permittivity of gallium arsenide and sapphire to 220 GHz

Despite the industrial relevance of sapphire and gallium arsenide (GaAs) for electronic applications, there are few reports on the value of complex permittivity and dielectric loss at frequencies above 110 GHz using integrated circuit topologies such as coplanar waveguides. Available materials characterization reports at microwave frequencies use methods that quantify bulk material properties and may miss the effective material properties present in the substrate structure and most relevant for integrated circuits at the frequencies of interest. Using a wafer-level method to extract the complex permittivity of sapphire and GaAs to 220 GHz, we found that the complex permittivity of intrinsic GaAs (εon−wafer≈13) and the convoluted in-plane and out-of-plane complex permittivity of c-plane sapphire (εon−wafer≈10.5) did not show dispersion up to 100 GHz. This result shows the importance of wafer-level tests for sapphire, where on-wafer permittivity diverges from nominal single-axis values. Above about 100 GHz, different data analysis approaches yielded increasingly different permittivities. This result provides useful data to electronics designers, materials scientists, and physicists while also providing insight into mechanisms that could produce unexpected frequency dependence.

Graphene-based flexible SERS substrates with Ag nanohole structures for biochemical sensing

Synchronizing Tunable Luminescence and Shape Morphing in a Metal Nanocluster–Enabled Hydrogel Platform

ABSTRACT Soft hydrogels capable of simultaneous shape morphing and optical modulation under a single stimulus hold promise for next‐generation intelligent materials. However, most systems rely on complex architectures or multi‐component triggers, limiting integration and responsiveness. Here, we develop a structurally homogeneous nanocluster–gel platform by embedding water‐soluble gold nanoclusters (AuNCs) into a polyacrylamide matrix, enabling synchronized mechanical deformation and tunable luminescence in response to solvent polarity. Upon exposure to organic solvents, the gel rapidly contracts (<5 s), while spatial confinement of AuNCs within the polymer boosts quantum yield from 6.05% to 22.83%. The emission profile is programmable by varying cluster size, highlighting AuNCs as structurally tunable emission modulators. Kinetic analysis reveals non‐Fickian swelling/deswelling—governed by solvent diffusion and polymer relaxation—that synergistically modulates emission. This enables real‐time, quantitative optical detection of solvent gradients (R 2 = 0.996) with >97% reversibility over 10 cycles. The integrated responsiveness is demonstrated in a biomimetic lotus‐shaped gel that folds and dims upon water uptake, and is further validated in a leakage scenario where the hydrogel serves as a pipe encapsulation film, achieving rapid sealing and visual monitoring of solvent leakage. This work establishes a nanocluster–gel platform that unifies photophysics and actuation for adaptive, self‐indicating soft systems.

Efficacy and safety of mesutoclax (ICP-248) in combination with orelabrutinib in patients with B-cell malignancies: A pooled analysis.

7073 Background: Mesutoclax (ICP-248) is a next-generation BCL2 inhibitor, and orelabrutinib is a marketed BTK inhibitor for CLL/SLL, MCL, and MZL. However, the clinical activity of their combination in these malignancies remains undefined. This analysis evaluated the combination of mesutoclax and orelabrutinib across B-cell malignancies. Methods: Patients with relapsed and refractory (R/R) MCL, MZL were enrolled in a phase 1 study (NCT05728658), and treatment-naive (TN) CLL/SLL were enrolled in a phase 2 study (NCT06378138). R/R MCL and MZL patients received continuous daily mesutoclax (125 mg) and orelabrutinib (150mg) from cycle 1 day 1 continuously until disease progression or unacceptable toxicity. For CLL/SLL patients, induction therapy with orelabrutinib (150 mg QD, Cycles 1-17) was administered first, followed by mesutoclax (100 mg or 125 mg QD, Cycles 3-14). The orelabrutinib treatment continued beyond cycle 17 if the uMRD (≤10-4) was not achieved. Mesutoclax was implemented with a ramp-up schedule in all patients to mitigate the risk of TLS. Results: As of 05 Jan 2026, 60 patients were enrolled and treated in the studies: 8 R/R MCL, 10 R/R MZL, and 42 TN CLL/SLL (mesutoclax 100 mg, n=21; 125 mg, n=21). In R/R patients, the median number of prior lines of therapy was 1 (1-4). 6 (33.3%) were refractory to the last line of therapy. For TN CLL/SLL, 76.2% (32/42) of patients had moderate or high TLS risk, and 14.3% (6/42) had TP53 mutation or del (17p). Among 5 MCL and 8 MZL patients who had at least one disease evaluation, the overall response rate (ORR) was 100%, with CRR of 100% and 50%, respectively. Five patients (38.5%) achieved peripheral blood (PB) uMRD. In the 21 CLL/SLL patients receiving mesutoclax 125 mg, the ORR was 100% and the CRR was 38.1%, and the peripheral blood uMRD rate at 36-week was 65%. The median time to CR was 3.7 months in R/R group and 7.1 months in TN group. The 12-month PFS rate was 100% in CLL/SLL, while data for MCL and MZL are immature due to short follow-up. As the safety data cutoff (31 Dec 2025), the combination of mesutoclax and orelabrutinib was well tolerated with a favorable safety profile, and no new safety signals were identified compared to either agent as monotherapy. Most TEAEs were grade 1-2, with no TEAEs leading to drug discontinuation or death reported. The most common grade ≥3 TEAEs include neutrophil count decreased (35%), platelet count decreased (11.7%). Notably, no grade ≥3 anemia was reported. No clinical or laboratory TLS occurred. Conclusions: Mesutoclax in combination with orelabrutinib demonstrated a tolerable safety profile across B cell malignancy subtypes (MCL, MZL, CLL/SLL). Significant 100% ORR and deep response were observed in patients receiving mesutoclax 125mg combined with orelabrutinib. This all oral, chemo-free regimen has the potential to establish a novel therapeutic option for B-NHLs. Clinical trial information: NCT05728658 .

Association of social environment with clinically meaningful declines in health-related quality of life domain bother and function in individuals after prostate cancer treatment.

e17025 Background: Clinically meaningful (CM) changes in function and bother have a significant impact on health-related quality of life (HRQoL) after prostate cancer (PCa) treatment. The role of social environment in these changes is unclear. This study aims to examine the relationship between social environment and CM changes in HRQoL-related function and bother after PCa treatment. Methods: Deidentified patient data from CaPSURE registry were geocoded and linked with CDC Social Vulnerability Index (SVI) data, which ranks communities based on vulnerability to public health crises (high SVI: ≥75th percentile nationally). High SVI was defined as ≥75th percentile nationally. HRQoL (0-100 (best)) was assessed using the UCLA Prostate Cancer Index for urinary (UF and UB), sexual (SF and SB), and bowel (BF and BB) function and bother, at baseline and two years post-treatment. CM changes at two years were those > ½ standard deviation from mean baseline scores; changes in function or bother were deemed discordant. Multivariable logistic regressions assessed associations between SVI and discordant CM changes in HRQoL, adjusting for sociodemographics and treatment. Odds ratios (OR) and 95% confidence intervals (CI) were presented, p < 0.05 was significant. Results: Among 8,904 men, mean age at diagnosis was 65.36 years. The cohort was 86.5% White, 9.8% Black, and 3.7% Asian/Latino/Other. 22% had > 3 comorbidities [3–4 (18.3%), and ≥5 (3.9%)]. Discordant CM changes at two-years post treatment noted in 12.5% for urinary, 17.7% for sexual, and 9.5% for bowel domains. High-SVI patients had lower baseline HRQoL across all domains, for example, UF (89.03 vs 92.63), BF (85.08 vs 88.43), and SF (49.09 vs 53.81). Compared with radical prostatectomy, other treatment modalities were associated with higher odds of urinary and sexual function decline and increased bowel bother. High SVI was linked with lower odds of discordant CM change in sexual domain (OR 0.35, 95% CI 0.16-0.73), but not in urinary or bowel domains. Conclusions: High SVI is associated with worse baseline HRQoL scores and lower odds of discordant, clinically meaningful declines in sexual HRQoL, meaning greater risk of concurrent declines in both sexual function and bother. Linking patients from high SVI communities undergoing PCa treatment with more resources and support may improve HRQoL and survivorship.

State-level trends in colorectal cancer incidence and stage before and during the COVID-19 pandemic.

e15709 Background: Colorectal cancer (CRC) incidence differs across U.S. states, in part reflecting socioeconomic and rural differences as to where patients are diagnosed. During the COVID-19 pandemic, disruptions to routine care raised concern for delayed CRC detection and shifts toward more advanced disease at diagnosis. Methods: We analyzed CRC incidence using NAACCR Cancer in North America (CiNA) public-use data (1995–2022) for the 50 U.S. states and the District of Columbia. Age-adjusted incidence rates (per 100,000; standardized to the 2000 U.S. population) were examined for a pre-pandemic baseline (2010–2019), the pandemic year (2020), and early recovery (2021–2022). Socioeconomic context was defined using census-tract poverty and rurality (metro vs non-metro). We assessed baseline state-level incidence patterns, pandemic-related changes, and variation by socioeconomic and rural context. Stage at diagnosis was examined by race and ethnicity, stage and grade by socioeconomic context, and 2020 incidence changes were compared between screening-eligible adults (45–75 years) and younger adults (15–44 years). Results: Baseline CRC incidence varied substantially across states and was higher in states with a greater proportion of cases from high-poverty, non-metro areas (increase of 0.38 cases per 100,000 per 1% increase in this case mix; p < 0.001). In 2020, CRC incidence declined in all states (median −13.8%; interquartile range [IQR] −16.5% to −11.1%), with partial recovery in 2021–2022 (median −6.6%; IQR −9.4% to −3.8%). The magnitude of the 2020 decline did not differ meaningfully by socioeconomic or rural composition; however, states with more disadvantaged rural case mixes often experienced smaller-than-expected declines compared with national patterns. Across all racial and ethnic groups, the proportion of CRC diagnosed at regional or distant stage increased by 2.0 to 7.4 percentage points in 2020 and remained elevated in 2021–2022. Differences in stage and grade by socioeconomic status and rurality were modest. In age-stratified analyses, CRC incidence declined among screening-eligible adults aged 45–75 (median −13.9%; IQR −15.5% to −10.1%), whereas incidence among adults aged 15–44 did not show a comparable decline (median +7.9%; IQR +1.1% to +16.2%), yielding a median between-group difference of −19.0 percentage points. Conclusions: The COVID-19 pandemic was associated with a nationwide reduction in CRC diagnoses, incomplete recovery through 2022, and a shift toward later-stage disease. Socioeconomic and rural contexts influenced baseline CRC burden but explained only part of pandemic-related changes. Younger adults did not experience a comparable decline, consistent with disruption of screening-dependent detection pathways. These findings highlight the need to re-engage CRC screening and ensure timely diagnostic evaluation across diverse populations.

Phase 1, first-in-human, open-label study evaluating the safety, tolerability, pharmacokinetics, and preliminary efficacy of TGW101 in patients with advanced solid tumors.

TPS3163 Background: Tumor-associated glycoprotein 72 (TAG-72) is a non-internalizing glycoepitope that is aberrantly expressed in a broad range of epithelial cancers, predominantly adenocarcinomas. TGW101 is an antibody-drug conjugate (ADC) therapy that utilizes a novel, bio-orthogonal click chemistry-based linker technology that enables the targeting of non-internalizing cancer antigens and controlled payload release. The first component of TGW101 is a chemically cleavable ADC (TGW101-ADC) that consists of a diabody targeting TAG-72 and a highly stable trans -cyclooctene (TCO) linker conjugated to a monomethyl auristatin E (MMAE) payload. Once the TGW101-ADC binds to intratumoral TAG-72 and is largely cleared from the systemic circulation, the second drug component, a tetrazine trigger (TRG001), is administered, leading to rapid MMAE release from TGW101-ADC via a selective cleavage reaction with the TCO linker. Upon release of MMAE into the tumor microenvironment, it readily diffuses into surrounding tumor cells, binds to microtubules, inhibits cell cycle progression, and activates local immune cells. In preclinical studies, TGW101 treatment led to tumor regressions in the OVCAR3 (ovarian) cell line and GXF214 (gastric) patient-derived xenograft model, and improved survival in LS174T (colon) tumor bearing mice. Methods: This first-in-human, open-label Phase 1 study evaluates the safety and tolerability of TGW101 and defines a maximum tolerated dose (MTD). A Bayesian Optimal Interval (BOIN) design is used to determine the recommended dosing regimen(s) for expansion. Secondary objectives include assessing the pharmacokinetic (PK) profile, intratumoral pharmacodynamics, preliminary anti-tumor activity, and immunogenicity of TGW101. Exploratory analysis will investigate the relationship between TAG-72 expression and tumor response, as well as the effect of TGW101 exposure on tumor biomarkers. Up to 50 participants with advanced solid tumors, including breast, NSCLC, prostate, cervical, endometrial, ovarian, gastroesophageal, and non-squamous cell carcinoma of the head and neck are planned for enrollment. During dose escalation, increasing doses of TGW101-ADC, administered intravenously on Day 1 followed 7 days later by a fixed-dose infusion of TRG001, will be explored in cohorts of 3-4 participants treated in 3-week cycles. Once pharmacologically active dose levels are identified, specific cohorts may enroll additional participants to further explore safety, PK, tumor biomarker changes, and efficacy. Adverse events are assessed according to CTCAE v5 and tumor response is determined by RECIST 1.1 or PCWG3 criteria every 6 weeks for the first 2 cycles, then every 9 weeks. The trial began in May 2025 and 7 sites, all within the United States, are open for enrollment. Clinical trial information: NCT06959706 .

Longitudinal assessment of treatment information needs in an advanced prostate cancer online support group, 2016-2025.

e17056 Background: We see a growing trend in patients and caregivers using peer-led online communities for treatment advice. This study measured topic trends and tested automated topic tagging in these forums. Advanced prostate cancer treatment options changed quickly from 2016 to 2025. New developments in imaging, radioligand therapy, and PARP inhibitors have changed how advanced prostate cancer is managed. Patients and caregivers needed new information at the same pace. Methods: We analyzed 374,514 posts in 21,572 discussion threads from Malecare’s advanced-stage prostate cancer forum (2016-2019, 2021-2025). Threads were tagged for eight treatment domains using prespecified keyword rules. For validation, we used a stratified random sample of 405 threads, with 45 sampled per year. Two independent coders, blinded to automated tags, performed binary coding for each topic at the thread level. Percent agreement and Cohen's kappa were estimated with bootstrap 95% confidence intervals. Automated tags were compared with adjudicated gold-standard labels per topic, and sensitivity, specificity, false-positive rate, and false-negative rate were reported. Yearly prevalence estimates for radiopharmaceuticals and radioligand therapy, as well as for PARP and DNA repair, were corrected using the Rogan-Gladen adjustment. Results: Automated tagging identified an increase in discussions of radiopharmaceuticals and radioligand therapy from 14.8% of threads in 2016 to 26.1% in 2025. After correction, prevalence increased from 17.0% to 31.0%, showing a greater rise in interest. Similarly, discussions about PARP and DNA repair topics increased from 4.9% in 2016 to 12.7% in 2025. After correction, prevalence rose from 2.2% to 10.2%. Table 1 presents interrater agreement and automated tag error for all eight topics. Conclusions: Validated tagging enables tracking of patient information needs over time, enabling real-world telemetry to improve patient education. Online discussions mirrored important changes in advanced prostate cancer treatment and should inform the development of health education content. Clinicians and patient advocates can improve health education by adding detailed frequently asked questions about new treatments, including side effects and management, to better answer patient questions. Validation metrics by topic (n=405). Topic Agree (%) Kappa Sensitivity Specificity FPR FNR Androgen axis therapy 99.5 0.988 0.990 0.960 0.040 0.010 Imaging and staging 94.8 0.886 0.890 0.980 0.020 0.110 Chemotherapy 87.9 0.741 1.000 0.990 0.010 0.000 Radiopharmaceuticals and radioligand therapy 95.8 0.875 0.820 0.990 0.010 0.180 PARP inhibitors and DNA repair 95.1 0.718 1.000 0.970 0.030 0.000 Immunotherapy 96.3 0.853 0.850 1.000 0.000 0.150 Clinical trials 80.7 0.622 0.710 0.990 0.010 0.290 Treatment sequencing and resistance 71.6 0.474 0.600 0.730 0.270 0.400

MPOWER: A pilot trial among men with prostate cancer—Optimizing wellness by enhanced relief from hot flashes with acupuncture.

TPS12170 Background: Prostate cancer is the most commonly diagnosed cancer among men, and advanced disease often requires androgen deprivation therapy (ADT). ADT is associated with distressing side-effects, including hot flashes, sleep disturbance, fatigue, and cognitive symptoms. Acupuncture is a nonpharmacologic intervention shown to reduce hot flashes in other hormone-treated cancers, such as breast cancer; however, standardized acupuncture protocols for men receiving ADT are lacking. The primary objective of the MPOWER trial (NCT07335224) is to evaluate the feasibility and acceptability of study procedures, recruitment strategies, and delivery of a standardized acupuncture protocol to reduce hot flashes in men with prostate cancer. Feasibility will be assessed through enrollment and intervention engagement, and acceptability through satisfaction ratings and post-intervention qualitative feedback. Secondary objectives include patient-reported hot flash frequency and severity, hot flash interference, endocrine symptoms, sleep disturbance, and health-related quality of life. Methods: This randomized study compares Immediate Start Acupuncture (IA) with an Active Comparator: Delayed-Start Acupuncture (DA), allowing all participants eventual access to the intervention, among 24 men treated with ADT and/or androgen receptor pathway inhibitors. The intervention consists of a standardized 10-week protocol with weekly 30-minute manual acupuncture sessions delivered by a certified acupuncturist with oncology expertise. All participants receive lifestyle education from a nurse navigator as part of usual care. Participants randomized to the IA arm receive the acupuncture protocol during the initial 10-week intervention period, followed by 12 weeks of follow-up without acupuncture. Participants randomized to the activate comparator DA arm will receive lifestyle education per usual care without acupuncture during the initial 10-week comparison period, with bi-weekly check-ins serving as an attention control, and then initiate the acupuncture protocol after the delay period. Study assessments integrate real-time patient-reported hot flash diaries collected via a secure mobile application for remote symptom monitoring (Locus Health), validated patient-reported outcome measures, and continuous Fitbit monitoring to capture physiologic and sleep metrics, including heart rate variability, sleep duration, and nighttime awakenings. Feasibility outcomes include enrollment rates and intervention engagement. Acceptability is assessed quantitatively at multiple time points and qualitatively through semi-structured interviews. The primary patient outcome is the Hot Flash Severity Score, calculated as weekly mean hot flash frequency multiplied by severity. Enrollment began in January 2026, with accrual expected within 12 months. Clinical trial information: NCT07335224 .

Phase I study of intraperitoneal fast-manufactured IL-9-secreting CEACAM5-targeted CAR-T cells in advanced colorectal cancer with peritoneal metastases.

3514 Background: Colorectal cancer peritoneal metastasis (CRPM) remains an unmet clinical need with poor prognosis and limited benefit from systemic therapies. Our prior study on intraperitoneal (I.P.) CEA-targeting CAR-T cells (Nature Cancer) achieved a 25% ORR, yet limited persistence and exhaustion hampered efficacy. Consequently, we developed a novel Th9/Tc9-polarized fast- manufacturing process that exhibited superior T-cell fitness and antitumor function preclinically. We therefore conducted a Phase I study to evaluate the safety and preliminary efficacy of I.P. administration of these optimized CAR-T cells in CRPM population. Methods: This open-label, single-arm Phase I study enrolled patients with histologically confirmed, heavily pretreated CEA-positive CRPM. A standard 3+3 dose- escalation design evaluated three dose levels (DLs): DL1 (2.0 × 10 5 cells/kg), DL2 (3.0 × 10 5 cells/kg), and DL3 (4.0 × 10 5 cells/kg), followed by dose expansion. Primary objectives were safety and tolerability; secondary objectives included preliminary antitumor activity, cellular kinetics, and pharmacodynamics. Results: Between July 2024 and October 2025, 15 patients were enrolled; 14 were evaluable for efficacy. One dose-limiting toxicity (DLT; Grade 4 pneumonia) occurred at DL2. The most common grade 3 non-hematologic toxicity was diarrhea (53.3%). Cytokine release syndrome (CRS) occurred in all patients and was Grade 1/2 in 93.3%. No treatment-related deaths were reported. Among 14 efficacy-evaluable patients (median 3 prior lines of therapy), ORR was 57.1% and DCR was 100%, with maximum tumor shrinkage reaching 85.6%. Median PFS was 4.7 months (95% CI, 3.7–NE). Peripheral blood CAR- T cells expanded to a mean peak of 1.74×10 4 copies/μg, and serum CEA levels decreased in 100% of patients (median reduction: 67.5%). Conclusions: Intraperitoneal administration of fast-manufactured Th9/Tc9 like CEACAM5-targeted CAR-T cells demonstrated a manageable safety profile and encouraging antitumor activity in heavily pretreated CRPM patients, at approximately one-tenth of conventionally reported CAR-T cell doses. These findings support further development of fitness-enhanced, regionally delivered CAR-T strategies for peritoneal malignancies. Clinical trial information: NCT05396300 . Variable DL1 (n=3) DL2 (n=5) DL3 (n=6) Total (n=14) PR, n (%) 1 (33.3) 4 (80.0) 3 (50.0) 8 (57.1) SD, n (%) 2 (66.7) 1 (20.0) 3 (50.0) 6 (42.9) ORR, n (%) 1 (33.3) 4 (80.0) 3 (50.0) 8 (57.1) DCR, n (%) 3 (100.0) 5 (100.0) 6 (100.0) 14 (100.0) Median PFS, months 5.6 [3.8–NE] 3.4 [3.2–NE] NA 4.7 [3.4–NE]

Radiotherapy use among locally advanced cervical cancer patients (LACC) in the United States community oncology setting.

5523 Background: Brachytherapy with concurrent chemoradiation +/- pembrolizumab is the current standard of care for LACC. American Society for Radiation Oncology (ASTRO) recommends external beam radiation therapy (EBRT)+brachytherapy, and since 2020 has recommended intensity modulated radiation therapy (IMRT) over other forms of EBRT due to reduced toxicity risk. However, real-world evidence is needed to characterize patient outcomes based on the type of radiation received. Methods: Electronic medical record (EMR) data from The US Oncology Network identified adult LACC (FIGO 2018 Stage IB3-IVA) patients (pts) diagnosed between July 2018 – July 2022 and followed through June 2024. Pts were stratified by receipt of IMRT or Volumetric Modulated Arc Therapy (VMAT) and brachytherapy during the observation period. Patient characteristics and provider-documented progression were assessed descriptively. Results: This study included 250 LACC pts; 73 (29%) pts received IMRT or VMAT, 177 (71%) pts did not receive IMRT/VMAT; and 189 (76%) pts received brachytherapy. Overall, 61 pts (24%) received both IMRT or VMAT and brachytherapy. Among the 148 pts who were diagnosed in 2020 or later, 28% received IMRT or VMAT. Disease progression occurred among 33% of IMRT or VMAT pts and 26% of non IMRT or VMAT patients; progression was similar across brachytherapy and non-brachytherapy patients. Conclusions: Advanced radiation techniques such as IMRT or VMAT and brachytherapy are associated with improved outcomes for patients with LACC, yet their combined use only occurs among 1 in 4 LACC patients, which is notably less than clinical trials. These findings present an opportunity to broaden adoption of evidence-based radiation strategies, supporting optimal treatment delivery and improved patient outcomes. IMRT or VMAT non IMRT or VMAT Brachytherapy Non-brachytherapy Patient, N 73 177 189 61 Age (yrs, median) 51 54 51 59 Race (White), N(%) 50 (69) 101 (57) 114 (60) 37 (60) Race (non-White), N(%) 9 (12) 36 (20) 35 (19) 10 (16) Race (not documented), N(%) 14 (19) 40 (23) 40 (21) 14 (23) Stage I-II, N(%) 31 (42) 82 (46) 94 (50) 19 (31) Stage III, N(%) 23 (32) 65 (37) 64 (34) 24 (39) Stage IVA, N(%) 6 (8) 10 (6) 11 (6) 5 (8) Progression, N(%) 24 (33) 46 (26) 55 (29) 15 (25)

Targeting MAN1A1 to address immune checkpoint inhibitors resistance: A novel mechanism of vascular-endothelial reprogramming in NSCLC.

2566 Background: Despite the efficacy of immune checkpoint inhibitors (ICIs) in NSCLC, resistance remains a major hurdle. The role of non-immune components, particularly vascular endothelial cells (VECs), is poorly understood. We aimed to identify key drivers of ICI resistance by focusing on VEC-mediated remodeling of the tumor microenvironment (TME). Methods: We integrated transcriptomic data from public cohorts of ICI-treated and treatment-naïve NSCLC patients (GSE225620), proteomic profiling from in-house whole blood samples, and peripheral blood lymphocyte subset analysis from NSCLC patients and healthy controls. Differential expression analysis and machine learning (LASSO, SVM-RFE) were employed to identify key resistance-associated genes, including MAN1A1. Functional validation was performed using shRNA-mediated MAN1A1 knockdown in A549 and H1299 NSCLC cell lines, with apoptosis measured by flow cytometry. Single-cell RNA-seq data (GSE207422) from pre- and post-treatment NSCLC tumors were analyzed using Seurat for cell annotation. Cell-cell communication (CellChat) and pathway activity (AUCell) analyses were used to investigate VEC-immune cell interactions and their association with treatment response. Statistical analyses included Kaplan-Meier survival, Cox regression, and ROC curves to assess the prognostic and predictive value of MAN1A1. Results: Multi-omics analysis identified MAN1A1 as a key candidate associated with ICI resistance. Elevated MAN1A1 expression predicted inferior clinical outcomes, including shorter mPFS (3.2 vs. 8.5 months; P=0.019) in our cohort and reduced OS (47.4 vs. 65.1 months; HR=0.95, P=0.017) in the TCGA dataset. MAN1A1 expression effectively distinguished ICI non-responders (AUC=0.784). MAN1A1 knockdown significantly induced apoptosis in NSCLC cell lines (A549, P=0.00022; H1299, P=0.0004). Increased peripheral B-cell counts post-chemotherapy correlated with treatment response, and MAN1A1 protein levels inversely associated with B-cell levels (r=-0.38, P<0.05). Single-cell RNA-seq analysis revealed VECs as the primary expressors and communication hubs of MAN1A1. In non-responders, MAN1A1-high VECs were reprogrammed to drive a pro-inflammatory feedback loop with neutrophils via ANXA1-FPR1/NAMPT-integrin signaling axes. In contrast, VECs from responders exhibited a shift towards T-cell-recruiting LGALS9-CD45 signaling. Conclusions: Our study identifies MAN1A1 as a novel driver of ICI resistance in NSCLC, mediating vascular-endothelial reprogramming that fosters an immunosuppressive TME characterized by pro-tumor neutrophil recruitment and impaired T- and B-cell immunity via the ANXA1-NAMPT axes. Targeting MAN1A1 represents a promising therapeutic strategy to overcome ICI resistance, providing a strong rationale for future translational development.

Frontline intensified vs standard immunochemotherapy in high-risk diffuse large B-cell lymphoma: A systematic review and meta-analysis.

7065 Background: While the phase III CALGB 50303 trial comparing frontline REPOCH vs standard RCHOP for diffuse large B-cell lymphoma (DLBCL) showed comparable outcomes overall, it suggested a potential benefit of intensified therapy in high-risk patients, including high-grade (HG) and double/triple-expressor (DE/TE) disease. Subsequent studies evaluating REPOCH in HG and DE/TE DLBCL have been small, retrospective and nonrandomized, with conflicting results. Despite limited evidence and greater toxicity, REPOCH is commonly used in these cohorts. We conducted a systematic review and meta-analysis to assess outcomes after frontline REPOCH vs RCHOP in HG and DE/TE DLBCL, comprising the largest high-risk cohort to date. Methods: Original research studies evaluating outcomes after frontline REPOCH or RCHOP in HG or DE/TE DLBCL from 2009 to 2025 were included. HG was defined by MYC translocation, including double/triple-hit (DH/TH; MYC and BCL2 and/or BCL6 translocation). DE/TE was defined by MYC and BCL2 and/or BCL6 by IHC. Studies were excluded if they had <10 subjects or used agents in addition to the RCHOP or REPOCH backbones. RCHOP-like regimens with comparative substitutions were included (e.g., RCOMP). We used random effects models to calculate pooled response rates and survival outcomes. Results: 69 studies were included with 3979 total patients (1818 HG; 2161 DE/TE). 2698 patients (68%) were treated with RCHOP, and 1281 (32%) with REPOCH. Limited reporting of patient characteristics in HG or DE/TE cohorts precluded multivariable analysis of prognostic factors. While overall response rate (ORR) did not differ between treatment groups (relative risk [RR] 1.04, CI 0.92-1.19), patients treated with REPOCH were more likely to achieve complete response (CRR; RR 1.12, CI 1.01-1.23). This difference was especially notable in the HG subgroup (RR 1.4, CI 1.21-1.61), but there was no difference in the DE/TE subgroup (RR 1.04, CI 0.93-1.17). Across all patients, REPOCH was associated with significantly improved progression-free survival (PFS; hazard ratio [HR] 0.5, CI 0.39-0.65) and overall survival (OS; HR 0.59, CI 0.43-0.82). Landmark 2-year PFS (65% vs 52%, p = 0.013) and OS (75% vs 50%, p < 0.0001) were significantly higher for REPOCH. Subgroup analysis showed significantly improved 2-year OS with REPOCH in both HG (72% vs 47%, p < 0.0001) and DE/TE (67% vs 51%, p = 0.032) disease. Conclusions: In this large cohort, REPOCH was associated with significantly improved CRR, PFS and OS relative to RCHOP in HG and DE/TE DLBCL. Reported patient characteristics were limited, and the REPOCH arm may have preferentially included younger patients with better performance status. Nevertheless, our analysis provides evidence in an area lacking randomized trial data. While not definitive, our findings support the common clinical practice of treating select high-risk patients with REPOCH.

Diagnostic performance, safety, and biodistribution of [ ⁶⁸ Ga]Ga-OncoCAIX for PET/CT imaging: A first-in-human phase I clinical trial in clear cell renal cell carcinoma.

3075 Background: Conventional imaging techniques are routinely used in clear cell renal cell carcinoma (ccRCC). However, their ability to accurately characterise tumour biology and stage the disease remains limited. An attractive target for theranostic applications is represented by carbonic anhydrase IX (CAIX), which is overexpressed in ccRCC. [⁶⁸Ga]Ga-OncoCAIX is a novel small-molecule radiopharmaceutical for positron emission tomography (PET) imaging. The compound has been discovered by DNA-Encoded Chemical Libraries and designed to selectively bind to CAIX, allowing for tumour-specific uptake and favourable biodistribution in preclinical models. This multicentre, prospective, phase I study provides the first-in-human assessment of [⁶⁸Ga]Ga-OncoCAIX PET/CT in patients with kidney lesions. Methods: Twenty patients with suspected primary or recurrent ccRCC based on conventional imaging were planned for enrolment in the study. Each patient received a single intravenous injection of [⁶⁸Ga]Ga-OncoCAIX. PET/CT acquisitions were performed at 0, 10, 60 and 120 minutes after the injection with multiple blood and urine samples collections for pharmacokinetic and dosimetric analysis. Adverse events were recorded and graded according to CTCAE v5.0. Where available, PET findings were correlated with histopathology. Results: Twenty patients were included. Seventeen patients were evaluated for incidentally detected renal masses, and three for suspected recurrence identified on CT or MRI. [⁶⁸Ga]Ga-OncoCAIX was well tolerated, with no drug-related AEs reported. The tracer demonstrated mixed renal and hepatic clearance, favourable biodistribution, and low background activity in healthy tissues, including the kidneys. Physiological uptake was mainly observed in the gastrointestinal tract, particularly the stomach. PET-positive lesions showed rapid and selective tumour uptake shortly after injection, with increasing signal intensity at 60 and 120 minutes. Eight patients had positive PET findings, eleven scans were negative and one patient presented a lesion with mild tracer uptake. Available histopathological confirmation showed concordance with PET results, including three PET-positive ccRCCs, three PET-negative oncocytomas, and one PET-negative chromophobe RCC; the one lesion with mild tracer uptake corresponded to a papillary RCC with mild CAIX expression on histology. Dosimetric analyses are ongoing. Conclusions: [⁶⁸Ga]Ga-OncoCAIX shows an excellent safety profile and favourable biodistribution, as well as promising tumour-targeting properties in ccRCC lesions. These preliminary findings support its potential role for the detection and characterisation of ccRCC. Complete data on histopathology will confirm the diagnostic accuracy and potential clinical utility in patients with suspected ccRCC. Clinical trial information: NCT06840548 .

Inferring optimal detection of homozygous loss (homozygous deletion) in head and neck cancer: Associations with HPV status, primary disease site, and clinical outcomes.

6071 Background: Validated detection of homozygous loss is becoming increasingly important in clinical practice. Examples include targeting the AKT pathway in breast cancer with PTEN loss and ongoing multi-tumor trials of PRMT5 and MAT2A inhibitors, with MTAP loss serving as a key biomarker. Homozygous losses are challenging to detect and require intentional next-generation sequencing assay design and validation. We evaluated the prevalence of the most common homozygous losses in advanced head and neck squamous cell carcinoma (HNSCC) and their associations with HPV status, primary site, and clinical outcomes on standard first-line (1L) therapies. Methods: This study used the nationwide (US-based) de-identified Flatiron Health-Foundation Medicine head and neck cancer clinico-genomic database (FH-FMI CGDB), originating from approximately 280 US cancer clinics (~800 sites of care). Patients with advanced HNSCC who underwent tissue-based genomic profiling with FoundationOneCDx were included. First-line therapy included chemotherapy alone or in combination with cetuximab or immune checkpoint inhibitors (ICIs). Logistic regression assessed the associations of prior treatment, HPV status, and primary disease site with homozygous losses. Clinical outcomes were evaluated with Cox proportional hazards models. Results: Among 969 HNSCC specimens, homozygous losses were most frequent in CDKN2A (22.1%), CDKN2B (17.8%), MTAP (9.4%), and PTEN (5.3%). CDKN2A, CDKN2B and MTAP losses were enriched in tumors arising from the larynx and hypopharynx compared with the oral cavity and oropharynx and were largely mutually exclusive with HPV positivity. PTEN loss was most common in the oropharynx and was enriched in HPV-positive tumors and in metastatic liver biopsies. In univariable analysis, homozygous losses were not associated with outcomes after 1L therapy. In multivariable analysis, HPV negativity and higher ECOG scoring, but not PD-L1 status, were independently associated with worse overall survival. Conclusions: Using an assay that is FDA-approved to detect and report copy-number (CN) losses, homozygous losses of CDKN2A , CDKN2B , MTAP , and PTEN were frequently observed in HNSCC. These alterations were not directly associated with outcomes following 1L therapy. Similar to CDKN2A/B loss , MTAP loss was enriched in HPV-negative tumors and defines a clinically relevant subset of HNSCC (~10%) potentially eligible for emerging MTAP-targeted therapies.

Understanding psychological distress during chemotherapy: Assessment of anxiety and depression in Middle Eastern cancer patients.

e24117 Background: Anxiety and depression are common but often under-recognized in patients receiving chemotherapy and may adversely affect adherence, symptom reporting, and quality of life. Data from Middle Eastern oncology populations are limited, and use of brief screening tools is not widely implemented in regional practice. Methods: In this cross-sectional study at Burjeel Hospital, Oman, adults aged 18–75 years with solid tumors on active chemotherapy completed PHQ-9 and GAD-2 questionnaires during routine visits. Demographic and clinical data (age, sex, cancer type, stage, therapy line, ECOG performance status) were collected from records. Moderate-to-severe depression was defined as PHQ-9 ≥10 and clinically significant anxiety as GAD-2 ≥3. Descriptive statistics and multivariable logistic regression identified factors associated with higher symptom burden. Results: Of 112 eligible patients, 108 (96%) completed both instruments (61% female; mean age 54 ± 11 years). Tumor types reflected regional patterns: breast (28%), colorectal (19%), lung (14%), ovarian (6%), endometrial (4%), head and neck (9%), lymphoma (8%), others (12%). Overall, 38% met criteria for moderate-to-severe depression and 31% had significant anxiety. Depression was more frequent in females (46% vs 27%, p = 0.04), advanced disease (OR 2.3, 95% CI 1.1–4.7), and second-line or later chemotherapy (OR 1.9, 95% CI 1.0–3.8, p = 0.02). Anxiety correlated with depression (r = 0.68, p < 0.001) but not age, tumor site, or performance status. Conclusions: A high burden of anxiety and depression was observed among chemotherapy patients in Oman, underscoring the need for routine psychosocial screening. Integrating brief tools such as PHQ-9 and GAD-2 with referral pathways to psycho-oncology, counseling, and culturally adapted supportive programs may help reduce distress and improve adherence and outcomes. Depression and anxiety prevalence across clinical characteristics. Characteristic Subgroup Depression PHQ-9>10 n/N(%) Anxiety GAD-2 n/N (%) Depression (p) Anxiety (p) Gender Female 30/66 (46%) 23/66 (35%) 0.04 0.12 Male 11/42 (2 7%) 11/42 (26%) Disease Stage Early (I-II) 8/35 (22%) 8/35 (23%) <0.05 0.08 Advanced (III-IV) 33/73 (45%) 26/73 (3 6%) OR (I. 1-4.7) Treatment Line First-Line 15/54 (2 8%) 12/54 (22%) 0.02 0.03 Second-Line 16/3 8 (42%) 13/3 8 (3 4%) Third-Line + 10/16 (5 5%) 9/16 ( 56%) OR 1.9 (1.0-3.8) ECOG PS 0-1 30/85 (3 5%) 25/85 (29%) 0.45 >/= 2 11/23 (48%) 9/23 (39%)

EIK1003, a PARP1-selective inhibitor, in combination with paclitaxel (PTX): Initial combination and updated monotherapy results from a phase 1/2 study EIK1003-001 in advanced solid tumors.

3102 Background: Non-selective PARP inhibitors (PARPi) demonstrate efficacy in HRR-mutant tumors but cause myelosuppression, potentially mediated by PARP2 inhibition. Overlapping hematologic toxicities have limited PARPi combinations with chemotherapy, restricting their use largely to the maintenance setting. EIK1003 is a potent and selective PARP1 inhibitor that may enable chemotherapy combinations previously not feasible. Here we report the first clinical data for EIK1003 + PTX and updated results for EIK1003 monotherapy. Methods: Cohort 1C evaluated EIK1003 QD (dose level [DL] 1 to 4) + PTX 80 mg/m² QW (Days 1, 8, 15 of 21-day cycles) in patients (pts) with platinum-resistant recurrent ovarian cancer (OC) and HER2(-) breast cancer (BC). Cohort 1A evaluated EIK1003 monotherapy (10 to 160 mg QD). BOIN dose escalation design was used. Primary endpoints were safety and tolerability; secondary endpoints were PK and antitumor activity. Results: As of 21-Nov-2025 safety data cutoff, 51 patients were treated in Cohort 1C; 30 remain on treatment. Median age was 59 years (range 25-81), ECOG PS 0/1 (54.9%/43.1%), and 62% pts received ≥ 4 prior lines of therapy. One DLT of febrile neutropenia was reported at DL4. Gr ≥ 3 TEAEs occurred in 62.7% pts and were mainly hematologic (Table 1). All pts with Gr ≥ 3 anemia had baseline anemia. TEAEs led to dose reduction in 10 pts and discontinuation in 8 pts. No deaths occurred due to TEAEs. The ORR was 23.1% (1 CR, 8 PR), with responses in OC (n=5) and BC (n=4). EIK1003 PK in combination was consistent with monotherapy, with target concentrations achieved at all dose levels. In Cohort 1A, 65 pts were treated. Gr ≥ 3 TEAEs were 43.1%, with anemia and neutropenia being most common (Table 1). 3/6 pts with Gr ≥ 3 anemia had baseline anemia. ORR was 14.3% overall, and 31.3% (5/16) in PARP-naïve pts. Median DOR in confirmed responders was > 6 mos in both cohorts (Table 1). Safety, PK/PD, and preliminary efficacy supported advancement of two doses into Part 2 dose optimization. Conclusions: EIK1003 + PTX demonstrated a tolerable safety profile in a heavily pre-treated population, with hematologic toxicities comparable to historical PTX use (eg, KN-B96, ESMO 2025). These findings represent the first clinical evidence that PARP1-selective inhibitors can be safely combined with chemotherapy. Extended follow-up of EIK1003 monotherapy confirms durable tolerability with antitumor activity. Clinical trial information: NCT06253130 . Safety and efficacy summary. Cohort 1C (EIK1003 + PTX) Cohort 1A (EIK1003 monotherapy) Safety N = 51 N = 65 Gr ≥ 3 TEAEs, % (n) 62.7 (32) 43.1 (28) Neutropenia 43.1 (22) 7.7 (5) Anemia 15.7 (8) 9.2 (6) Efficacy* N = 39 N = 49 ORR per RECIST 1.1, % (n) 23.1 (9) 14.3 (7) DCR (CR+PR+SD), % (n) 74.4 (29) 38.8 (19) DOR, median (range), mos 6.4+ (4.4+, 7.7) 6.2+ (4.2+, 10.4+) * Efficacy evaluable population as of 17-Dec-2025.

Cadonilimab in combination with chemotherapy versus chemotherapy alone in patients with immunotherapy-refractory recurrent or metastatic nasopharyngeal carcinoma: A multicenter, randomized, phase III trial the CONQUEST trial).

6033 Background: Chemotherapy combined with immune checkpoint inhibitors (ICIs) is the standard first-line treatment for recurrent or metastatic nasopharyngeal carcinoma (R/M NPC); however, resistance is common. In the post-ICI setting, prospective trials have demonstrated limited efficacy, with objective response rates (ORRs) ranging from 4.2% to 34.3%. A prior phase II study of nab-paclitaxel, cisplatin, and capecitabine (TPC) plus cadonilimab demonstrated a promising ORR of 68% in ICI-refractory patients, providing the rationale for this phase III trial comparing TPC plus cadonilimab versus TPC alone in this population. Methods: In this randomized, multicenter, phase III trial, patients were assigned (1:1) to receive TPC plus cadonilimab or TPC alone. TPC consisted of nab-paclitaxel 200 mg/m² d1, cisplatin 60 mg/m² d1, and capecitabine 1000 mg/m² BID d1–14 (Q3W). In the experimental arm, cadonilimab (10 mg/kg d1) was added. Maintenance with capecitabine plus cadonilimab (experimental) or capecitabine alone (control) continued until disease progression, unacceptable toxicity, or 2 years. The primary endpoint was progression-free survival (PFS), defined as the time from randomization to disease progression or death from any cause. Secondary endpoints included overall survival (OS), ORR (defined as the best overall response), and safety. Results: A total of 84 patients were randomized (n=42/arm) at three hospitals. Baseline prior therapies were balanced ( P =0.533): 35 (83.3%) patients in the experimental arm and 37 (88.1%) in the control arm had received 1-3 prior lines of therapy, while 7 (16.7%) and 5 (11.9%) had received >3 prior lines, respectively. After a median follow-up of 10.2 months, PFS was significantly prolonged in the experimental arm versus the control arm (8.9 months [95% CI 6.3-11.4] vs. 5.1 months [95% CI 3.2-6.9]; HR 0.47, 95% CI 0.28-0.80; P =0.004). The ORR was 61.9% in the experimental arm versus 52.4% in the control arm, showing no statistically significant difference ( P =0.378). Median OS was not reached. All patients experienced adverse events (AEs). Grade 3-4 AEs occurred in 19 (45.2%) patients in the experimental arm versus 23 (54.8%) in the control arm ( P =0.383). The most common Grade 3-4 AEs were anemia (31.0% vs. 21.4%), neutropenia (21.4% vs. 21.4%), and leukopenia (16.7% vs. 14.3%). Immune-related AEs (irAEs) were more frequent in the experimental arm (34 [82.9%] vs. 25 [58.1%]; P =0.013); however, no Grade 3-4 irAEs were observed in either arm. Conclusions: TPC chemotherapy combined with cadonilimab demonstrated a statistically significant improvement in PFS compared with TPC alone in heavily pretreated, ICI-refractory R/M NPC patients, with a manageable safety profile (NCT06664983). Clinical trial information: NCT06664983 .