Browse Articles

Allostatic load and myosteatosis in women with metastatic breast cancer at diagnosis.

e13018 Background: llostatic load (AL), a measure of cumulative physiological stress, and myosteatosis, fatty infiltration of skeletal muscle, have each independently been associated with adverse cancer outcomes. AL reflects chronic activation of stress-related physiological pathways that adversely affect multiple organ systems; therefore, it may represent an upstream contributor to alterations in body composition, including myosteatosis. However, few studies have assessed if AL correlates with myosteatosis. This study examined the association between high AL score and myosteatosis in women with newly diagnosed metastatic breast cancer (MBC). Methods: This cross-sectional study used data from a retrospective cohort study of women receiving care at an urban academic medical center (January 1, 2000-2024). Women aged ≥18 years with computed tomography (CT)-confirmed MBC and an L3 CT image at diagnosis were included. AL score was calculated using a definition from previous publications: a nine-biomarker composite score (range 0-9), with biomarker values outside established clinical thresholds assigned 1 point. The median AL score of the population was used to classify subjects as having high (≥2) or low AL ( < 2). Myosteatosis was defined using BMI-specific CT skeletal muscle radiodensity cutoffs ( < 41 HU for BMI < 24.9 kg/m2, < 33 HU for BMI ≥25 kg/m2). Spearman’s correlation and Fisher's Exact test were used to assess the association between high AL and myosteatosis in this MBC population. Results: In the analytic sample (n = 114), the mean (SD) age was 59.5 (13.2), the Caucasian population was most prevalent (46.5%, n = 53/114), the median (IQR) BMI was 29.4 (9.5), and the majority (62.3%) had high functional status at diagnosis (ECOG 0-1). The median (IQR) AL score was 2 (2) and 41.2% of women had high AL. The prevalence of myosteatosis was 46.5% (n = 53/114). No association was observed between AL score and skeletal muscle radiodensity (r = -0.036, p = 0.76). Chi-square analyses similarly demonstrated no association between AL and myosteatosis (p = 0.154). Conclusions: Among women with MBC, myosteatosis was moderately prevalent. However, no association was identified between AL score and myosteatosis at diagnosis. Further research with standardized definitions and longitudinal data is needed to clarify their independent and combined roles in prognosis of cancer outcomes.

Early versus delayed initiation of first-line immune checkpoint inhibitors in metastatic non–small cell lung cancer: A real-world analysis of treatment timing and care context.

e20660 Background: Immune checkpoint inhibitors (ICIs) are standard first-line therapy for metastatic non–small cell lung cancer (NSCLC) without targetable mutations. In real-world practice, the timing of ICI initiation varies and may reflect differences in clinical urgency and care pathways rather than treatment efficacy alone. We evaluated whether early versus delayed initiation of first-line ICIs is associated with survival and early healthcare utilization in metastatic NSCLC. Methods: We conducted a retrospective cohort study using the TriNetX Global Collaborative Network, including electronic health records from approximately 170 healthcare organizations. Adults (≥18 years) with stage IV NSCLC who received first-line immunotherapy (pembrolizumab, nivolumab, atezolizumab, durvalumab, or cemiplimab) were included. Patients were stratified by time from metastatic diagnosis to ICI initiation: early (≤30 days) versus delayed (30–90 days). Propensity score matching was performed for demographics and comorbidities. To minimize immortal time bias, overall survival (OS) was indexed to the date of stage IV diagnosis, while 90-day outcomes, including mortality, emergency department (ED) visits, hospitalizations, and intensive care unit (ICU) admissions, were indexed to the date of ICI initiation. Kaplan–Meier analyses and hazard ratios were used to compare outcomes. Results: After propensity score matching, 1,424 patients were included in each cohort. When overall survival was indexed to the date of metastatic diagnosis, there was no significant difference between patients initiating first-line ICIs within 30 days versus 30–90 days (median OS 732 vs 819 days; HR 1.05, 95% CI 0.95–1.16; log-rank p=0.351). Ninety-day mortality following ICI initiation was similar between groups (14.5% vs 13.0%; p=0.217), as were rates of ED or observation visits (30.2% vs 30.8%; p=0.714). However, earlier ICI initiation was associated with lower rates of acute hospitalization (40.2% vs 44.7%; p=0.015) but higher rates of ICU admission within 90 days of treatment initiation (41.1% vs 35.3%; p=0.002). Conclusions: In this real-world analysis of metastatic NSCLC restricted to first-line immunotherapy, initiation of ICIs within 30 days versus 30–90 days of diagnosis was not associated with differences in overall survival when survival was indexed to metastatic diagnosis. Earlier initiation was, however, associated with increased ICU utilization shortly after treatment initiation, suggesting greater clinical instability at the time therapy is started. In practice, awaiting sequencing results before starting immunotherapy is essential and thus the effect of time to initiation of immunotherapy needs further assessment via larger retrospective studies in an attempt to eliminate confounding variables.

A machine learning algorithm for optimizing treatment selection for patients with up to three hepatocellular carcinomas measuring ≤ 3 cm.

4019 Background: Liver resection (LR) and radiofrequency ablation (RFA) are recommended for patients with early-stage hepatocellular carcinoma (HCC) with three nodules measuring ≤3 cm and preserved liver function. This study aimed to develop a predictive model to guide treatment decisions based on survival outcomes. Methods: This study included 18,958 patients with up to three HCCs measuring ≤3 cm from the nationwide survey of Japan. The Recurrent Deep Survival Machines (RDSM) model was employed for deep survival analysis. We employed 10-fold cross-validation, the concordance index (C-index), and overall survival (OS) to assess model performance. Survival curves were compared using the log-rank test. To identify potential confounding factors, 1:1 propensity score matching (PSM) was performed. Results: Patients undergoing LR demonstrated significantly longer OS than those receiving RFA (5-year survival rate 81.4% vs. 73.1%; P < 0.005). The trained RDSM model achieved a C-index of 0.68. In the deep learning (DL) model, patients undergoing recommended treatment demonstrated significantly longer survival than those who did not (5-year survival rate 81.2% vs. 73.9%; P < 0.005; PSM, 81.9% vs. 76.7%; P < 0.005). The DL modeling recommended LR in 6,966 (84.5%) patients undergoing RFA, especially those showing typical imaging patterns (early enhancement and washout in the computed tomography images [85.9% vs. 61.7% and 84.1 vs.74.3%, respectively]). Conclusions: DL modeling effectively helped treatment allocation for patients with up to three HCCs measuring ≤3 cm. Our study indicates the potential utilization of DL modeling in the treatment allocation of patients with up to three HCCs measuring ≤3 cm.

Phase II study evaluating ivonescimab in combination with chemotherapy as first- and second-line treatment of advanced or metastatic gastric and gastroesophageal adenocarcinoma patients (UCGI 52 – GRACIE).

TPS4239 Background: The prognosis of advanced/metastatic gastroesophageal adenocarcinomas (GEA) has recently improved with new targeted therapies, including immune checkpoint inhibitors, mostly developed in the first-line setting. However, for patients (pts) without actionable biomarkers or for those who become refractory to first-line therapy, a significant unmet need remains for effective therapeutic options. Ivonescimab, a bispecific antibody targeting both VEGF-A and PD-1, appears to be a promising strategy in this setting. Methods: This phase 2 multicenter two-cohort, non-randomized study (NCT06846346) is designed to evaluate the efficacy and safety of ivonescimab + chemotherapy in pts with advanced or metastatic GEA, with and without actionable biomarker. Main eligibility criteria are age ≥ 18 years, histologically proven GEA, stage 4 disease, no prior treatment for pts without actionable biomarker (cohort 1) or only one prior line for pts with actionable biomarker (cohort 2), ECOG-PS 0-1, and adequate organs functions. The tumor has to be measurable (RECIST 1.1). Treatment regimens are ivonescimab 20mg/kg IV on D1 Q2W combined with: cohort 1: FOLFOX (oxaliplatin 85mg/m² IV, folinic acid 400 mg/m² IV [or L-folinic acid 200 mg/m²], 5-FU bolus 400 mg/m² then 5-FU 2400 mg/m² as a 46-hours continuous IV infusion on D1 Q2W; oxaliplatin will be stopped after 8 cycles); cohort 2: at investigator discretion, paclitaxel (80 mg/m2 IV at D1, D8 and D15, Q4W for at least 4 cycles) or irinotecan (180 mg/ m2 IV on D1 Q2W). Treatment will be administered until disease progression or unacceptable toxicity. The primary endpoint is ORR by central review. Secondary endpoints include ORR assessed by investigator, duration of response, PFS, OS, safety, time to PS deterioration >2, and quality of life. According to a single stage phase II A’Hern’s design, a sample of 33 evaluable pts in cohort 1 and 47 evaluable pts in cohort 2 is required to distinguish between a maximum futility proportion of 0.5 (cohort 1)|0.2 (cohort 2) and a minimum efficacy proportion of 0.75 (cohort 1)|0.4 (cohort 2) with a one-sided significance level of 0.04 and 90% power. 88 patients have to be enrolled to observe a minimum of 80 evaluable patients. Planned accrual duration is 1 year, 12 patients have already been enrolled. Clinical trial information: NCT06846346 .

Impact of metformin on outcomes with immune checkpoint inhibitors in renal cell carcinoma: A large-scale retrospective analysis.

e16558 Background: Immune checkpoint inhibitors (ICIs) are a mainstay in the treatment of renal cell carcinoma (RCC). Metformin, a commonly used biguanide for type 2 diabetes, has demonstrated anti-tumor and immunomodulatory effects that may enhance ICI efficacy. However, the clinical impact of concurrent metformin use on outcomes in patients with RCC treated with ICIs remains unclear. Methods: We conducted a retrospective cohort study using the TriNetX Global Collaborative Network, a database of electronic medical records from 158 healthcare organizations. We identified patients with RCC using ICD-10-CM C64 who were initiated on ICIs ( nivolumab, pembrolizumab and avelumab) and were divided into two cohorts: those receiving concurrent metformin and those who were not. Propensity score matching (PSM) was performed in 1:1 ratio to balance the cohorts for age, sex, race, comorbidities, and concomitant medications. The primary outcome was overall survival (OS) over a 3-year follow-up period. Secondary outcomes included Major Adverse Cardiovascular Events (MACE) and immune-related adverse events (irAEs) including pneumonitis, colitis, thyroiditis, and skin eruptions. Results: A total of 12,271 patients with RCC treated with ICIs were identified: 1,321 in the Metformin cohort and 10,950 in non-metformin cohort. After 1:1 PSM, two well-balanced cohorts of 1,132 patients each were analyzed. Baseline characteristics were statistically similar between both cohorts (p > 0.05; Table 1). Metformin use in combination with ICI was associated with a significantly higher survival probability compared with ICI alone (61.40% vs 51.79%; HR 0.75, 95% CI 0.65–0.87). There was no statistically significant difference in the incidence of serious immune-mediated toxicities, including pneumonitis, thyroiditis, skin eruptions, and colitis and MACE. Conclusions: In this large-scale, real-world analysis, concurrent metformin use was associated with significantly improved OS in patients with RCC treated with immune checkpoint inhibitors. These findings suggest that metformin may serve as a beneficial adjuvant to therapy. However, additional prospective clinical trials should be performed to validate these findings and further elucidate the potential driving signaling mechanisms. Baseline characteristics after PSM. Characteristic RCC with Metformin (N=1,132) RCC without Metformin (N=1,132) P value Age at index [mean +/- SD (years)] 64.6 +/- 9.8 64.5 +/- 11.1 0.829 Sex Female, n (%) 298 (26.3) 293 (25.9) 0.811 Male, n (%) 834 (73.7) 839 (74.1) 0.010 Race White, n (%) 841 (74.3) 842 (74.4) 0.962 Hispanic, n (%) 119 (10.5) 126 (11.1) 0.636 Black or African American, n (%) 69 (6.1) 59 (5.2) 0.363

A validation study of a novel biomarker-based scoring system (EAST score) for limited-stage small-cell lung cancer: A secondary analysis of JCOG0202 and JCOG1011.

8081 Background: Concurrent chemoradiotherapy (CCRT) followed by prophylactic cranial irradiation (PCI) is potentially curative for limited-stage small-cell lung cancer (LS-SCLC), yet reliable markers for identifying long-term survivors remain undefined. The Enhanced Assessment for SCLC Treatment (EAST) score was originally derived from a retrospective, single-center training cohort for prognostic stratification of LS-SCLC patients receiving CCRT. This score integrates N3 status, serum lactate dehydrogenase (LDH), pro-gastrin-releasing peptide (ProGRP), and cytokeratin 19 fragment (CYFRA 21-1) levels measured at diagnosis. This study (JCOG2401A) validated its prognostic utility using data from two randomized controlled trials (RCTs) of LS-SCLC. Methods: The validation cohort comprised patients enrolled in JCOG0202 and JCOG1011: randomized phase 3 and 2 studies, respectively, that compared multiple consolidation chemotherapy regimens following CCRT. External validation of the dichotomized EAST score (low: score 0–1, high: 2–5) was performed by assessing discrimination and calibration for progression-free survival (PFS) and overall survival (OS). Exploratory subgroup analyses were conducted in an integrated cohort of the training cohort (N = 224; median follow-up, 64.5 months) and the validation cohort. Hazard ratios (HRs) for PCI were estimated using propensity score-weighted Cox models. Results: Out of 309 patients in these trials, 205 with complete data for EAST score components were included in the validation cohort. Median age was 62 years. Low- vs. high-risk groups showed stage (I-II/III) distributions of 28/72% vs. 11/89%, and tumor response (CR-PR/SD) distributions of 89/3% vs. 91/2%, respectively. The low-risk group (N = 114) showed better outcomes than the high-risk group (N = 91) for both PFS and OS (Table). Integrated cohort analysis revealed a numerical survival benefit from PCI in low-risk patients (N = 214), whereas the benefit was highly limited in the high-risk group (N = 202) (Table). Conclusions: The prognostic value of the EAST score was validated even in external RCT datasets. High-risk patients, characterized by early recurrence and inferior OS, derive minimal benefit from PCI. These findings provide a rationale for a planned risk-adapted RCT utilizing the EAST score. Median PFS, months PFS, HR (95% CI) Median OS, months OS, HR (95% CI) Low- vs. high-risk Training cohort (N=107/117) 20.6/9.4 0.48 (0.34-0.67) 53.0/34.2 0.67 (0.46-0.98) Validation cohort (N=114/91) 25.0/10.7 0.55 (0.40-0.77) 63.2/29.6 0.51 (0.36-0.73) PCI vs. no-PCI All (N=287/129) 14.5/12.9 0.91 (0.65-1.28) 50.0/35.5 0.78 (0.56-1.10) Low risk (N=152/62) 31.3/14.6 0.86 (0.49-1.48) 67.9/40.1 0.75 (0.43-1.31) High risk (N=135/67) 10.2/10.6 0.99 (0.71-1.39) 34.2/30.9 0.86 (0.59-1.28)

SPARTO: A phase I study of PD-1 inhibitor spartalizumab (PDR001) in combination with low dose of pazopanib in pediatric relapsed/refractory tumors.

10001 Background: The initial results of immunotherapy in children with refractory/relapsed tumors have been disappointing. Modulation of angiogenesis with anti-VEGF may enhance immunotherapy penetration and promote lymphocytes T infiltration. In this context, we propose combining immunotherapy with low doses of pazopanib to enhance therapeutic efficacy in pediatric patients. Methods: Pediatric part of the SPARTO trial (NCT05210413) was a multicenter, open-label, dose-finding phase I clinical trial conducted in France. Pediatric patients with recurrent/refractory solid tumors were eligible to receive pazopanib at a fixed low dose of 225 mg/m²/d combined to anti-PD1 spartalizumab at three candidate doses (2, 3, 4 mg/kg/q4w). Dose-finding was guided by Bayesian Optimal Interval design based on dose-limiting toxicities (DLTs) in cycle 1, with a target toxicity level of 25%. Recommended phase 2 dose (RP2D) was determined on the basis of DLTs and blood exposure of pazopanib and spartalizumab. Toxicities were evaluated using CTCAE v5.0 and responses were evaluated by clinical and radiologic assessment (RECIST1.1 or RANO for brain tumors). Results: From May 2022 to June 2025, 41 patients were enrolled at 8 sites (median age: 13.9y; range: 7-25). Among 39 evaluable patients, 4 DLTs occurred in 1/9 patients at 2 mg/kg of spartalizumab (DL0), 1/6 patients at 3 mg/kg (DL1) and 2/24 patients at 4 mg/kg (DL2). Three patients experienced macrophagic activation syndrome (grade 3 at DL0 and DL2, grade 4 at DL1) and one patient experienced liver enzymes increase grade > 3 at DL2. A total of 58 grade 3–4 adverse events were reported in 26 patients, predominantly hepatic and hematologic toxicities. Median area under the serum concentration-time of spartalizumab from 0 to 28 days post-infusion at cycle 1, were 12.9 g.h/L (IQR 10.7-15.8), 17.2 g.h/L (IQR 15.5-20.0) and 19.5 g.h/L (IQR 16.9-23.6), respectively for DL0, DL1 and DL2. Median trough concentration of pazopanib was 23.1 mg/L (IQR 15.8-32.1). DL2 was identified as the RP2D. Of 40 evaluable patients, one patient with osteosarcoma presented complete response, 4 had partial responses (2 osteosarcoma, 1 chordoma, 1 carcinoma) and 8 had stable disease (including 4 for more than 4 months). The median duration of treatment was 68 days (IQR 57—106); one patient completed 24 cycles and two were still on treatment at data cut-off (at cycle 16 and cycle 23). Conclusions: Spartalizumab combined with low dose pazopanib was well tolerated leading to a RP2D of 4 mg/kg/q4w. The regimen shows evidence of anti-tumor activity in pediatric sarcoma, especially osteosarcoma. Further investigations are ongoing to select the sub-population who could benefit of this combination. Funded by INCa and ARC Foundation (call for project CLIP2 “Novartis Innovative Molecules, INCa-ARC_14820”). Clinical trial information: NCT05210413 .

Timing of palliative care referral and overall survival in patients with melanoma brain metastases treated with ipilimumab and nivolumab.

e21607 Background: The association between timing of palliative care referral and survival outcomes in patients with melanoma brain metastases (MBM) treated with immune checkpoint inhibitors remains poorly characterized. Methods: We conducted a retrospective cohort study of patients with MBM treated with combination ipilimumab and nivolumab at Mayo 3 sites from 2016 to 2024. Palliative care referral timing relative to MBM diagnosis was categorized as early (≤56 days), late ( > 56 days), or none. Patients without referral documentation were excluded from referral-based analyses. Overall survival (OS) was estimated using Kaplan–Meier methods and compared with log-rank tests. Multivariable Cox proportional hazards models were used to identify factors independently associated with OS, with stepwise selection (entry p < 0.25; stay p < 0.15). Results: Among 147 patients, 139 had referral information: 33 (23.7%) early, 56 (40.3%) late, and 50 (36.0%) no referral. Median OS differed significantly by referral timing (log-rank p < 0.0001): 2.8 months (early), 7.9 months (late), and 47.8 months (no referral). Among patients with ECOG 2–3, a higher proportion had early or late referral (each 42.9%) compared with no referral (14.3%), whereas patients with ECOG 0–1 were more evenly distributed across referral groups (early 21.6%, late 40.0%, none 38.4%). In multivariable analysis of 88 patients with complete covariate data, both early (HR 4.54; p = 0.0016) and late referral (HR 2.86; p = 0.011) were independently associated with worse OS compared with no referral. Progressive disease versus objective response (HR 4.45; p = 0.0002), hemorrhagic metastases (HR 2.79; p = 0.0015), male sex (HR 2.07; p = 0.044), and increasing age (HR 1.02 per year; p = 0.032) were also associated with inferior OS. Treatment with ipilimumab 1 mg/kg plus nivolumab 3 mg/kg showed a trend toward improved OS compared with ipilimumab 3 mg/kg plus nivolumab 1 mg/kg (HR 0.48; p = 0.055). Conclusions: Timing of palliative care referral was strongly associated with overall survival in patients with MBM treated with ipilimumab and nivolumab. Although patients with ECOG 2–3 were more frequently referred early or late, ECOG category overall did not clearly differentiate referral timing groups, suggesting referral may reflect unmeasured clinical factors beyond baseline performance status. Prospective studies incorporating more granular functional and symptom measures are needed to better define the role of palliative care integration in this population. Multivariable cox model for overall survival. Variable Comparison HR P value Referral timing Early vs None 4.54 0.0016 Late vs None 2.86 0.011 Best overall response PD vs ORR 4.45 0.0002 SD vs ORR 2.16 0.098 Hemorrhagic metastases Yes vs No 2.79 0.0015 Sex Male vs Female 2.07 0.044 Age Per year increase 1.02 0.032 Regimen Ipi1+Nivo3 vs Ipi3+Nivo1 0.48 0.055

Plasma ISG15 as predictor of outcome from antiangiogenic therapy in solitary fibrous tumor: Analysis from the GEIS-69 trial.

e23520 Background: Solitary fibrous tumor (SFT) is a rare sarcoma subtype characterized by limited therapeutic options in advanced disease. Tyrosine kinase inhibitors (TKIs) with antiangiogenic properties show clinical activity, but predictive biomarkers of response are lacking. ISG15, an interferon-stimulated gene involved in immune and stress responses, has been implicated in resistance mechanisms to antiangiogenic TKIs, including pazopanib, in previous studies. Methods: Plasma ISG15 levels were quantified by ELISA in paired blood samples from 13 SFT patients within GEIS-69, treated with sunitinib followed by combination with nivolumab. Samples were collected at baseline and after 3 weeks of sunitinib monotherapy. Overall survival (OS) and progression-free survival (PFS) were estimated using Kaplan–Meier methodology. Survival differences were assessed using the log-rank test and univariate Cox proportional hazards regression. Optimal cut-offs for baseline ISG15 were defined using maxstat. Results: Baseline plasma ISG15 levels ranged from 162 to 3358 pg/mL (median 731 pg/mL). Patients with high baseline ISG15 levels showed significantly shorter PFS compared with those with low levels (median 2.8 vs 5.1 months; p = 0.049), while no statistically significant differences in OS were observed. In univariate Cox regression analysis, high baseline plasma ISG15 levels were associated with an increased risk of progression (HR 3.52, 95% CI 0.93–13.32; p = 0.064). Following sunitinib treatment, ISG15 levels increased in most patients (10/12 evaluable cases), with dynamic changes ranging up to 19.2-fold. However, neither OS nor PFS differed significantly according to the magnitude of ISG15 induction during treatment. Conclusions: High baseline plasma ISG15 levels are associated with shorter PFS in SFT patients treated with sunitinib, supporting its potential role as a predictive biomarker in the antiangiogenic setting. Antiangiogenic therapy represents the current standard first-line treatment for unresectable or advanced disease. In contrast, treatment-induced ISG15 upregulation does not appear to correlate with clinical outcome. These findings warrant validation in larger cohorts.

Efficacy and safety of cadonilimab in patients with MSI-H/dMMR locally advanced or metastatic gastrointestinal malignancies: A real-world retrospective study.

2623 Background: Immunotherapies targeting PD-1 and CTLA-4 have emerged as promising therapeutic strategies for gastrointestinal malignancies (GIM), particularly those dMMR or MSI-H phenotypes. Cadonilimab, a novel bispecific antibody co-targeting PD-1 and CTLA-4, has showed potent anti-tumor effect across multiple cancer types. This real-world retrospective study systematically evaluated the efficacy and safety of cadonilimab in patients with MSI-H/dMMR locally advanced or metastatic GIM. Methods: This study retrospectively included patients with MSI-H/dMMR unresectable locally advanced or metastatic GIM who administered cadonilimab(6 mg/kg Q2W) since November 2022 in The Affiliated Hospital of Qingdao University. The primary endpoint was ORR as assessed by investigators per RECIST v1.1. Secondary endpoints included DCR, PFS, DoR, safety profile, and exploratory biomarker analysis (interleukins [IL-2, IL-6, IL-8, IL-10]) was performed on blood samples collected at baseline and before each treatment cycle. Results: As of November 7, 2025, a total of 21 patients were enrolled. The median follow-up duration was 19.4 months. The ORR was 76.19% and DCR was 100%. Median PFS and median DoR were not reached. The 12-month PFS rate was 74.9% (95% CI, 49.6%-88.8%) and 18-month PFS rate was 68.1% (95% CI, 41.6%-84.5%). Among 14 gastric cancer (GC) patients, ORR was 78.57% (5 CR, 6 PR), median PFS was not reached, 12-month PFS rate was 77.9% (95% CI, 58.6%-100%), and 18-month PFS rate was 68.2% (95% CI, 46.3%-100%). The ORR of GC treated in 1 st line was 80%, 4 achieved CR (40%). Among 7 colorectal cancer (CRC) patients, ORR was 71.43% (2 CR, 3 PR), median PFS was not reached, 12-month PFS rate was 71.4% (95% CI, 49.6%-88.8%), and 18-month PFS rate was 71.4% (95% CI, 29.4%-91.9%); the ORR of 1 st line treatment was 83.33%. By metastatic site: ORR was 85.71% in patients with peritoneal metastasis and 100% in those with pelvic metastasis. Notably, one patient achieved pCR after cadonilimab conversion therapy, and another patient receiving cadonilimab as fourth-line monotherapy achieved PFS of 25.4 months. As of data cutoff, 5 patients remained on treatment. Regarding safety, 33.33% of patients experienced grade 1-2 immune-related adverse events (irAEs), and 4.76% had grade ≥3 irAEs. No treatment discontinuation due to adverse events. Exploratory biomarker analysis showed that ORR was 90.91% in patients with high IL-2 expression, 91.67%,92.86% and 81.82% in patients with normal IL-6, IL-8 and IL-10 levels, respectively. Conclusions: Cadonilimab demonstrates potent and durable anti-tumor activity with a favorable safety profile in MSI-H/dMMR locally advanced or metastatic GIM. Additionally, interleukins may serve as potential predictive biomarkers for cadonilimab efficacy, providing insights for precision medicine in this patient population.

Risk of developing brain/central nervous system (CNS) metastases across multiple <i>ERBB2</i> -altered cancer types: Genotype as shaper of phenotype.

2022 Background: ERBB2 (HER2) alterations are associated with increased risk of brain metastases in breast cancer and non-small cell lung cancer (NSCLC). However, the relationship between ERBB2 alterations and brain metastasis across other tumor types remains unclear. We hypothesized that ERBB2 -altered tumors demonstrate increased propensity for brain metastases across multiple cancer types. Methods: Using the US-based deidentified Flatiron Health-Foundation Medicine Clinico-Genomic Database (FH-FMI CGDB), we analyzed patients with advanced solid tumors who received ≥1 line of therapy. ERBB2 alterations included pathogenic/likely pathogenic variants and amplifications. Brain/CNS metastases were identified via ICD-9/10 codes and validated against manual abstraction in NSCLC patients. Focusing on patients without brain/CNS metastases at diagnosis, we performed Cox proportional hazards analyses to evaluate time from initial diagnosis to first brain/CNS metastasis, modeling death as a competing risk. Multivariable models were adjusted for tumor types. Results: The analysis included 47,653 patients across 16 tumor types. Excluding breast/NSCLC, there were 29,272 patients: 2,149 ERBB2 -altered (7.3%) and 27,123 ERBB2 -wild type. 653 patients (30% of ERBB2 -altered patients) had mutations and 1644 patients (77%) had amplifications (some had both). In a multivariable analysis adjusting for tumor type and excluding breast/NSCLC, ERBB2 alterations independently predicted accelerated brain/CNS metastasis development (HR 1.55, 95% CI 1.29-1.86, p&lt;0.001). In separate disease-specific Cox models, the strongest associations were observed in gastric (HR 2.33, 95% CI 1.68-3.22, p&lt;0.001) and ovarian cancers (HR 2.52, 95% CI 1.45-4.38, p&lt;0.001) as well as breast (HR 1.70, 95% CI 1.48-1.96, p&lt;0.001) and NSCLC (HR 1.33, 95% CI 1.11-1.60, p=0.002). Conclusions: ERBB2 alterations predict a significantly increased risk of accelerated development of brain/CNS metastases independent of tumor types, with the strongest effects observed in gastric and ovarian cancers in addition to breast and lung cancers.

High-quality 4D anatomical and functional MRI for abdominal tumor motion management.

e16013 Background: 4D-MRI offers superior soft-tissue contrast for abdominal motion management but is often limited by the trade-off between image quality and scan duration. Moreover, existing techniques are largely restricted to anatomical imaging, lacking the integration of functional data. This study proposes a deep learning framework to generate high-quality (HQ) 4D anatomical and functional MRI from low-quality (LQ) inputs via accurate motion estimation and reconstruction, aiming to significantly enhance tumor tracking reliability and precision in Image-Guided Radiotherapy (IGRT). Methods: A total of 169 abdominal MRI datasets with complete liver coverage were partitioned into training and internal validation cohorts (8:2). Additionally, an independent external test cohort (n = 12) was acquired. The model employed a 3D U-Net-based architecture, utilizing HQ static 3D-MRI and LQ 4D-MRI inputs to synthesize HQ 4D anatomical MRI, HQ 4D functional MRI, and corresponding deformation vector fields (DVFs). To ensure spatial consistency, DWI was aligned via a hierarchical cross-contrast registration (HCR) pipeline. Image quality was quantified using Full Width at Half Maximum (FWHM) and CNR, while motion consistency was assessed via liver centroid trajectories. Results: In the validation and testing cohorts, quantitative motion analysis demonstrated consistent sub-voxel accuracy. For every individual patient and across all motion directions, the mean 3D trajectory errors were consistently &lt; 1 mm. Furthermore, the maximum error for each case remained below the original voxel dimensions (Training:1.56×1.56×3.0mm³; Testing: 2.68×2.68×2.7mm³). For a representative test case, Figure 1 presents the motion tracking curves and quality metrics (CNR, FWHM). Correspondingly, Figure 2 displays the generated HQ images (T1, T2, and DWI). As shown in Figure 2, while the tumor was indistinguishable on anatomical sequences, the generated HQ-DWI delineated the lesion, validating the model's capability to recover functional information for target definition. Conclusions: The proposed framework successfully reconstructs HQ 4D anatomical and functional MRI from LQ inputs while maintaining precise motion information. This personalized, multi-parametric 4D-MRI approach demonstrates feasibility for fast and reliable motion management, potentially enabling high-precision IGRT for abdominal cancers.

Association between clonal plasma cell S-phase after autologous stem cell transplantation and survival outcomes in multiple myeloma.

e19564 Background: Risk stratification in multiple myeloma (MM) relies on baseline staging systems and cytogenetic abnormalities, yet early relapse occurs in a subset of patients despite favorable risk features. Prior studies showed that clonal plasma cell S-phase assessed by flow cytometry using the plasma cell proliferation (PCPRO) assay at diagnosis and at autologous stem cell transplantation (ASCT) identifies biologically high-risk disease. The prognostic significance of clonal plasma cell S-phase following ASCT has not been well defined. Methods: We retrospectively analyzed MM patients who underwent ASCT within one year of diagnosis between January 1, 2013, and August 31, 2024. Patients without available post-ASCT clonal plasma cell S-phase assessment were excluded, yielding 150 evaluable patients. The proportion of clonal plasma cells in S-phase was determined by DNA content between G0/G1 and G2/M peaks and reported as a percentage. S-phase was assessed at the first post-ASCT marrow evaluation at day +60 (D45–75) and/or day +100 (D76–130), using the earlier assessment when both were available. Patients were classified as low (&lt;2%) or high (≥2%) post-ASCT S-phase. Landmark progression-free survival (PFS) analyses were performed from the post-transplant assessment. Multivariable Cox models adjusted for cytogenetic risk, maintenance therapy, disease response at transplantation, International Staging System (ISS), and age. Results: Among 150 evaluable patients, 36 (24%) had high post-ASCT S-phase (≥2%) and 114 (76%) had low post-ASCT S-phase (&lt;2%). Median age at ASCT was higher in the high S-phase group (64.5 vs 61.0 years; p=0.063), and advanced ISS stage (II/III) was more frequent (78% vs 57%; p=0.02). Cytogenetic risk, induction regimen, depth of response at transplantation, and maintenance therapy were similar between groups. Patients with high post-ASCT S-phase had significantly inferior PFS compared with those with low S-phase (median 17.0 vs 39.7 months; p=0.0003). On multivariable analysis, post-ASCT S-phase ≥2% remained independently associated with inferior PFS (HR 1.76, 95% CI 1.00–3.09; p=0.049). In an exploratory analysis among patients with paired S-phase assessments at the time of ASCT and post-ASCT, dynamic S-phase trajectories further stratified outcomes. Median PFS was longest in patients with persistently low S-phase (low→low; 39.7 months), intermediate in those converting from high to low S-phase (high→low; 27.2 months), and shortest in patients with newly emergent or persistently high post-ASCT S-phase (low→high: 15.0 months; high→high: 11.0 months) p=0.0019. Conclusions: Elevated proliferative activity of residual clonal plasma cells following ASCT identifies a biologically high-risk subset of MM patients with early relapse. Dynamic changes in post-ASCT S-phase further refine post-transplant risk stratification.

Global variation in thromboprophylaxis models of care and implementation supports for hospitalized and ambulatory cancer care.

e13627 Background: Cancer-associated thrombosis (CAT) is a leading cause of morbidity and mortality. As systemic therapies increasingly shift to ambulatory delivery—where most venous thromboembolism (VTE) events occur—effective models of care for thromboprophylaxis beyond the inpatient setting are critical. An expanding evidence base and evolving guidelines support risk-directed thromboprophylaxis across care settings, highlighting the importance of understanding real-world implementation. However, how thromboprophylaxis is operationalized globally—particularly in ambulatory cancer care—remains poorly described. Methods: A global, cross-sectional assessment of institutional thromboprophylaxis models of care for hospitalized and ambulatory adults with cancer was conducted in Q2 2025, with one response per service from oncology or antithrombotic pharmacists. Institution-level data on guideline availability, VTE risk assessment and prescribing workflows, preferred antithrombotics, and implementation barriers were collected and analyzed descriptively. Denominators reflect eligible responses per item. Results: Models of care were reported for 96 health services across Asia (27.1%), North America (27.1%), Oceania (20.8%), Europe (14.6%), and other regions (10.4%). Institutional CAT guidelines were available in 56.8% of hospitalized and 38.6% of ambulatory settings, with highest availability in Europe (85.7% hospitalized; 66.7% ambulatory). Risk assessment and prescribing occurred across multiple systems in 34.4% of institutions, most frequently in Latin America/Middle East/Africa (90%). For hospitalized patients, thromboprophylaxis strategies were evenly split between opt-out (default unless contraindicated; 48.8%) and opt-in (criteria-based; 51.2%), with regional variation (Oceania 83.3% opt-out; Europe/Asia 66.7% opt-in; North America 50/50). Ambulatory practices were inconsistent: VTE risk assessment was recommended for select patients in 30.4% of centers and not performed in 39.1%. System-level decision support was reported less often for ambulatory than hospitalized care (25.3% vs 62.2%). Oral anticoagulants were guideline-recommended more often in ambulatory than hospitalized care (68.8% vs 33.3%); apixaban more than rivaroxaban (71.9% vs 46.9%). Quality monitoring for ambulatory CAT was uncommon, with ≤15% of services routinely tracking risk assessment, prophylaxis use, or VTE rates. Barriers included limited pharmacist involvement (absent in 19.5% of ambulatory settings), fragmented systems, and variable anticoagulant availability. Conclusions: Global models of care for thromboprophylaxis vary substantially, with implementation less mature and less supported in ambulatory than hospitalized cancer care, underscoring the need for system-level implementation strategies in outpatient oncology.

Burden and temporal trends of pediatric central nervous system tumors in Kenya: A 15-year analysis from a national referral hospital.

e14045 Background: Pediatric central nervous system (CNS) tumors are a leading cause of childhood cancer mortality globally, yet epidemiologic data from low- and middle-income countries (LMICs) remain scarce. Understanding long-term tumor patterns is essential to inform equitable childhood cancer control. We describe the histologic spectrum and temporal trends of pediatric CNS tumors over 15 years at Kenya’s largest referral hospital. Methods: We conducted a retrospective review of children ≤18 years with histologically confirmed CNS tumors treated surgically at Kenyatta National Hospital between 2010 and 2024. Demographic, histologic, tumor grade, location, and year of diagnosis data were extracted from pathology records. Tumors were classified according to WHO CNS5 criteria; cases requiring molecular confirmation were designated as not otherwise specified due to absence of molecular testing. Temporal trends were analyzed by year and 5-year periods. Results: A total of 404 pediatric CNS tumors were identified. Gliomas were the most common tumors (41.1%), followed by embryonal tumors (32.2%), with medulloblastoma representing nearly one-quarter of all cases. Low-grade tumors accounted for 58.4% overall and increased over time (55.0% in 2010–2014 vs 61.7% in 2020–2024; p&lt;0.001). Annual surgical case volumes rose steadily, peaking in 2021, likely reflecting backlog after COVID-19–related service disruptions. The proportion of tumors diagnosed in children aged 0–4 years increased markedly from 22.6% to 35.1% across the study period. Supratentorial and infratentorial tumors were nearly equally represented. Conclusions: This 15-year analysis provides the most comprehensive characterization of pediatric CNS tumors in Kenya to date and highlights evolving tumor patterns in an LMIC setting. Rising detection of low-grade tumors and increasing diagnoses in very young children suggest improving access to care, while persistent reliance on histology alone underscores critical diagnostic gaps. These findings emphasize the need for expanded molecular pathology, multidisciplinary neuro-oncology services, and population-based registries to advance global childhood cancer equity.

How long is long enough? Outcomes with finite-duration bispecific antibody therapy in multiple myeloma.

7538 Background: Bispecific antibodies (BsAbs) have shown unprecedented efficacy in relapsed/refractory multiple myeloma (RRMM), and finite-duration therapy is an increasingly common yet poorly studied strategy. We aimed to analyze outcomes in MM patients (pts) treated with finite-duration teclistamab (tec) or talquetamab (tal). Methods: This single-institution retrospective cohort study included pts treated with tec or tal monotherapy at UCSF from 7/2021 to 1/2025 for RRMM, excluding pts with prior BsAb exposure and pts receiving BsAbs as a bridge to CAR-T. Patient-, disease-, and treatment-related characteristics were collected and summarized descriptively. To assess the impact of treatment duration while mitigating immortal time bias, we conducted 6- and 9-month landmark analyses of PFS, comparing pts who discontinued treatment before the landmark with pts still on treatment. Patients with PD or death before the landmark were excluded. Multivariate analyses (MVA) of survival outcomes used Cox regression with the following covariates: age, IMS-IMWG risk, ISS stage, extramedullary disease (EMD), ECOG score, and prior BCMA exposure. Results: In total, 120 consecutive pts were included with median follow-up 13.7 mos. Forty stopped therapy early for reasons other than PD (finite) and 80 did not (non-finite). Among finite pts, median treatment duration was 4.7 mos (range 0.2-32.4), 24 stopped due to toxicity, and 16 due to sustained remission. Clinical characteristics including age, ISS, IMS-IMWG risk, and ECOG were balanced between groups, but finite pts had less EMD (7.5% vs 25.0%, p=0.03) and triple-class refractory status (52.5% vs 73.8%, p=0.03) than non-finite pts. ORR was 97.5% (60% CR+) and 73.4% (29.1% CR+) in finite and non-finite pts, respectively. For finite, non-finite, and all pts, respectively, median PFS was not reached (NR) with 24-mo PFS 63%, 7.9 mos, and 16.53 mos, while median OS was NR, 24.8 mos, and NR. For landmark analysis of patients alive and progression-free at 6 mos, 22 were off therapy and 55 pts remained on therapy at 6 mos and median PFS was NR and 29.5 mos, respectively (univariate regression HR 0.71 [0.32-1.57], p=0.4). On MVA, the hazard ratio also favored the on-therapy group (HR 0.51 [0.21-1.22], p=0.13). At the 9-month landmark, 20 pts were off therapy and 42 pts were on, with univariate regression HR 1.71 (0.58-5.11, p=0.3) and MVA HR 1.1 (0.32-3.76, p=0.87). Conclusions: Finite-duration BsAb therapy was associated with durable responses in selected patients, although this group had more favorable baseline disease characteristics. In adjusted landmark analyses, early discontinuation before 6 months was associated with numerically shorter PFS without statistical significance, while PFS was comparable at the 9-month landmark. Randomized-controlled trials are required to define optimal treatment duration.

Determinants of survival in 356 patients failing frontline HMA-Ven for newly-diagnosed AML.

e18524 Background: Venetoclax (Ven) in combination with hypomethylating agents (HMA) is FDA-approved front-line therapy for elderly or unfit patients with newly-diagnosed acute myeloid leukemia (ND-AML); however, majority relapse or are refractory ( Pratz AJH 2024 ). We examined survival outcomes in the setting of frontline Ven-HMA failure. Methods: ND-AML patients with failure to achieve complete remission (CR/CRi) or loss of CR/CRi after front-line Ven+HMA, excluding post-transplant relapse were retrospectively studied. Response was evaluated per European Leukemia Net 2022 criteria ( Dohner Blood 2022 ). Results: 356 patients with ND-AML (median age 75 years; 65% males; 55% secondary) receiving frontline Ven+HMA (median; 3 cycles (1-26), had refractory [n=219 (62%)], or relapsed disease [137 (38%)]. At diagnosis, karyotype was complex in 142/347 (41%). Mutations involved TP53 (29)%, ASXL1 (19%), RUNX1 (17%), SRSF2 (15%), KRAS/NRAS (13%), IDH2 (7%), FLT3-ITD (7%), NPM1 (6%) and IDH1 (5%). At treatment failure, complex karyotype was present in 72/180 (40%), including 55 from baseline. In 137 patients with paired NGS, mutations persisted in TP53 (88%), IDH1 (86%), IDH2 (78%), NPM1 (75%), K/NRAS (73%), RUNX1 (70%), FLT3-ITD (65%) mutations. Mutations were acquired in a minority [ IDH1 / IDH2 (2% each) TP53 (3%), NPM1 (4%), FLT3-ITD (5%) K/NRAS (6%)]. Clearance of mutations were infrequent for TP53 (12%) and IDH1 (14%). At median follow-up of 4 months (mo) (0-80) from the time of relapse/refractory disease, 320 (90%) patients have died, with median survival of 4 mo (1-2-3 yr survival 19%/8%/5%). On multivariate analysis, peripheral blasts &gt;20%, complex karyotype, and wild-type IDH1 were independent predictors of inferior survival, while allogeneic stem cell transplant (ASCT) was associated with improved survival. A 3-point prediction model based on peripheral blasts ≥20%, complex karyotype, and wild-type IDH1 stratified patients into low-, intermediate-, and high-risk groups, with median survival of 7, 3, 2 months, respectively (p&lt;0.01). 13 patients underwent ASCT (median survival; 22.5 mo; 3-yr survival 33%), from low (n=10) or intermediate-risk groups (n=3). Salvage therapy (n=170, 48%) yielded CR/CRi of 31% for Ven+HMA (n=36), 54% intensive chemotherapy (n=26), 50% FLT3 inhibitors (i) (n=22), 37% IDH1/2i (n=19), 25% other regimens (n=69). Median survival was similar: Ven+HMA 8 mo, intensive chemotherapy 10 mo, FLT3i 6.5 mo, IDH1/2i 14 mo, other regimens 6 mo, but inferior with supportive care (2 mo; p&lt;0.01). Conclusions: The current study identifies peripheral blasts ≥20%, complex karyotype, and wild-type IDH1 as predictors of inferior survival in ND-AML relapsed/refractory to front-line Ven+HMA and underlines that while no specific salvage therapy conferred superior outcomes, ASCT was indispensable for long-term survival.

First-line HLX07 vs placebo combined with serplulimab and chemotherapy for nasopharyngeal carcinoma: A randomized, double-blind, multicenter phase 2 study.

6042 Background: Programmed cell death protein 1 (PD-1) blockade with chemotherapy confers significant survival benefit compared to chemotherapy alone and is the standard first-line therapy for recurrent or metastatic nasopharyngeal carcinoma (R/M-NPC). Epidermal growth factor receptor (EGFR) is overexpressed in approximately 85% of all NPCs and associated with poor outcomes, suggesting a potential target for improved efficacies. This study explores the efficacy of HLX07 (a novel humanized anti-EGFR antibody) versus placebo, in combination with serplulimab (PD-1 inhibitor) and chemotherapy as first-line treatment for R/M-NPC. Methods: This is a randomized, double-blind, multicenter phase 2 study. Patients with histopathologically confirmed, unresectable, R/M NPC that is not amenable to local or radical treatment and had no prior systemic therapy were randomized 2:1 to receive either HLX07 at 1000 mg (HLX07 group) or placebo (placebo group), along with serplulimab (300 mg) and chemotherapy (gemcitabine and cisplatin) Q3W intravenously. Primary endpoint was blinded independent central review (BICR)-assessed objective response rate (ORR) per RECIST v1.1. Secondary endpoints included other efficacy endpoints, safety, pharmacokinetics and biomarker explorations. Results: As of December 24, 2025, 75 patients were randomized to the HLX07 group (n=50) or placebo group (n=25). Efficacy results are reported for the per-protocol set (n = 72), which excluded two patients who violated the enrolment criteria and one with no post-baseline tumor assessment in the HLX07 group. With 26.0 months of follow-up, BICR-assessed confirmed ORR was 74.5% vs. 72.0% for the two groups. Overall, a trend of an improved median progression-free survival (PFS) was observed with HLX07 (17.3 months vs. 9.4 months, stratified hazard ratio [HR] 0.79, 95% CI 0.40–1.56). Median overall survival was not reached vs. 27.9 months (stratified HR 0.40, 95% CI 0.16–0.99) for the respective groups. Subgroup analysis revealed a trend of improved PFS in the HLX07 group compared to the placebo group for patients with PD-L1 CPS &lt;10 (median PFS, 8.1 vs. 6.8 months, HR 0.54, 95% CI 0.19–1.56) as well as for patients with EGFR H-score ≥200 (median PFS, not reached vs. 7.8 months, HR 0.30, 95% CI 0.08–1.15). 74 (98.7%) patients experienced treatment-emergent adverse events (TEAEs), with grade ≥3 TEAEs reported in 60 (80.0%) patients. TEAEs led to treatment discontinuation occurred in 14 (18.7%) patients. Deaths due to TEAEs were reported in 5 (6.7%) patients, with 1 (1.3%) in the HLX07 group that was treatment related. Conclusions: The addition of HLX07 to serplulimab and chemotherapy showed encouraging efficacy along with a manageable safety profile in patients with treatment-naïve R/M-NPC. Further investigation of this treatment regimen is warranted. Clinical trial information: NCT05513573 .

Polyfluoroalkyl‐Tagged Cell‐Penetrating Peptide‐Additives Enhance Intracellular Protein Delivery via Sustained Monomeric Lipid Interaction

ABSTRACT Recent advances in cell‐penetrating peptide (CPP)‐mediated intracellular protein delivery emphasized the critical role of sustained membrane association in enhancing delivery efficiency. Here, we report cell‐surface‐reactive, polyfluoroalkyl‐tagged polyarginine peptides with varying fluorine content as CPP‐additives that significantly enhance protein delivery in living cells. At low micromolar concentrations (2.5 µM), CPP‐additives containing 11–13 fluorine atoms enhanced intracellular protein delivery over 2‐fold relative to a tagless control without observable cytotoxicity. Live‐cell time‐lapse fluorescence imaging revealed that a CPP‐additive with 13 fluorine atoms showed prolonged membrane association (&gt;5 min) relative to a tagless control and facilitated rapid protein internalization within 10 min. Remarkably, surface‐enhanced infrared absorption spectroscopy (SEIRAS) with POPC membranes showed that fluorous CPP‐additives initially interacted with the lipid bilayer predominantly as aggregates but subsequently inserted into the membrane interior as monomers without fluorous tag‐tag association. Complementary molecular dynamics simulations of the initial membrane‐association step provided atomistic insight, showing partial lipid insertion of a monomeric CPP‐additive with 13 fluorine atoms while no insertion was observed for a tagless control within the same time scale. Collectively, our findings establish polyfluoroalkyl‐tagged CPP‐additives as potent, non‐cytotoxic vectors for intracellular protein delivery and provide mechanistic detail regarding the molecular basis of their lipid bilayer interactions.

Coherence control in graphene for photonic networking and quantum information storage