Browse Articles

Using digital screening to increase early identification of hereditary breast and ovarian cancer syndrome: A multicenter cluster-randomized trial.

10526 Background: Early identification of hereditary breast and ovarian cancer syndrome (HBOCS) in primary gynecological care is essential for timely referral to specialized counseling and risk-adapted prevention. Yet, identification mostly occurs after disease onset, thereby thwarting preventive efforts. To facilitate early identification of healthy HBOCS carriers along the HBOCS care pathway in Germany, we developed a modular cross-sectoral digital care platform (dVP_FAM). This study compared the impact of dVP_FAM with standard care on increasing the proportion of healthy individuals with HBOCS. This project was funded by the Innovation Fund of the German Federal Joint Committee. Methods: We conducted a two-arm, cluster-randomized trial including 42 primary gynecological centers (20 intervention, 22 control) in Germany. In the intervention centers, patients filled out a digital HBOCS screening questionnaire based on the inclusion criteria of the German Consortium HBOC. Results were reviewed and validated by primary care gynecologists, who then decided whether to initiate referral to a specialized HBOC center via the digital platform. Control centers followed usual care. A mixed-effects logistic regression model with random intercepts was used to account for clustering at the study center level. Odds ratios (ORs) and 95% confidence intervals were estimated. Results: In the intervention centers, 5.534 individuals completed digital screening; 1.116 (20.2%) met inclusion criteria (IC+), and 507 (9.2%) were uncertain about family history. Of these 1.623 individuals, 664 requested an appointment at an HBOC center. A total of 467 individuals (intervention group (IG): n = 329, mean age = 45.6 (SD 12.3), 99.4% female; control group (CG): n = 138, mean age = 46.2 (SD 13.5), 99.3% female) met the primary endpoint. Among IG, 282 participants (85.7%; [83.3%, 88.6%]) were healthy at the initial counseling at the HBOC center compared with 110 participants (79.7%; [73.5%, 84.7%]) in the CG. After adjusting for center effects, the observed odds of healthy counselees were lower in the CG than in the IG (OR = 0.67; 95% CI, 0.28–1.59; p = 0.368). Conclusions: The IG showed a higher proportion of healthy individuals at risk for HBOCS. Although not significant, this trend may suggest that supporting primary care physicians with a digital screening platform facilitates early identification of HBOCS within cross-sectoral care pathways. Clinical trial information: DRKS00030371.

Real-world characteristics, treatment patterns, and outcomes of newly diagnosed diffuse large B-cell lymphoma (DLBCL) patients treated in first-line (1L) in the United States (US) community oncology setting.

e19069 Background: Most US DLBCL patients receive their care in community practices. In that setting, there is greater variability in patient characteristics, care delivery, and resources that may affect management and outcomes. While 1L therapy with R-CHOP predominates, the approval of Pola-R-CHP expanded options. However, there is limited data characterizing the drivers of treatment patterns and outcomes in the community setting. The objective of this study is to describe real-world clinical profiles and time-to-next-treatment or death (TTNTD) among patients treated with 1L therapy in the US community oncology practice setting. Methods: Patients newly diagnosed with DLBCL (not otherwise specified) who initiated a 1L regimen within the US Oncology Network from January 1, 2020 to January 31, 2025 were included. Structured electronic health record data was analyzed until death or last contact date by September 2025. TTNTD was estimated using Kaplan-Meier methods. Patient characteristics and outcomes were summarized descriptively by 1L regimen (R-CHOP, Pola-R-CHP, or Other). Results: Among the 4,291 eligible patients, the 1L therapies included 3,138 (73%) R-CHOP, 411 (10%) Pola-R-CHP, and 742 (17% overall; 83% R-mini-CHOP and 17% other chemotherapy) Other. Baseline characteristics differed with higher stage in Pola-R-CHP and older age at diagnosis in Other, which was primarily R-mini-CHOP. Adoption of Pola-R-CHP increased from 2% to (2022) to 28% (2024), while R-CHOP declined but remained most used across years (86%-54%). Two-year TTNTD was 70.6% for R-CHOP, 73.6% for Pola-R-CHP, and 47.3% for Other. Median TTNTD was 68.5 months for R-CHOP, not reached for Pola-R-CHP, and 17.9 months for Other. Conclusions: Pola-R-CHP utilization is increasing, but R-CHOP was the most used 1L regimen in the US community setting. While TTNTD is favorable for Pola-R-CHP, high rates of relapse remain, highlighting ongoing need for optimization of frontline therapy and supportive care pathways to improve real-world effectiveness. Patient characteristics and outcomes by 1L treatment. Variable R-CHOP (n=3,138) Pola-R-CHP (n=411) Other (n=742) Median Age (Q1, Q3) 68 (59, 75) 69 (61, 75) 82 (78, 85) Female, n (%) 1,368 (43.6%) 175 (42.6%) 357 (48.1%) Stage (I-II), n (%) 1,029 (32.8%) 42 (10.2%) 215 (29.0%) Stage (III-IV), n (%) 1,482 (47.2%) 294 (71.5%) 388 (52.3%) <2 ECOG, n (%) 1,617 (51.5%) 232 (56.4%) 312 (42.0%) ≥2 ECOG, n (%) 274 (8.7%) 36 (8.8%) 151 (20.4%) TTNTD Events, n (%) 933 (29.7%) 95 (23.1%) 375 (50.5%) Median TTNTD (Months; 95% CI) 68.5 (63.5, NR) NR (34.2, NR) 17.9 (13.5, 26.0) 2-Year TTNTD (95% CI) 70.6% (68.8, 72.3)% 73.6% (68.4, 78.1)% 47.3% (43.2, 51.3)% CI = Confidence interval, ECOG = Eastern Cooperative Oncology Group, NR = Not reached.

Socioeconomic and access disparities in ovarian cancer treatment and outcomes: A retrospective cohort study.

e17569 Background: Ovarian cancer ranks eighth in common causes of cancer/cancer-related death worldwide. Its high mortality can be attributed to 70% of cases being diagnosed at advanced stage disease, leading to a high relapse rate despite multimodal therapy, such as chemotherapy, surgery, and newer targeted immunotherapy. Despite efforts to improve screening among women for health disparities, barriers to healthcare often remain multifaceted and difficult to identify. There are gaps in the literature regarding how variables such as insurance type and distance relative to the treatment facility affect disease outcomes. This study aims to identify key socioeconomic factors (SES) and health disparities, and to assess whether these variables influence receipt of any treatment within one month of initial diagnosis and progression-free survival (PFS). Methods: This single-system retrospective cohort study analyzed 555 adult patients with ovarian cancer cases diagnosed between January 1, 2016, and February 28, 2025, as identified in the Henry Ford Health cancer registry. PFS was defined as the time from systemic therapy initiation to the first documented disease progression (recurrence) or death from any cause, with patients without an event censored at the earlier of the last disease assessment or the data extraction date (Dec 10, 2025). A multivariable logistic regression model and a Cox proportional hazards model were used to assess the associations between SES and (1) receiving treatment within one month and (2) PFS, respectively. Results: Older age groups had lower odds of receiving treatment within one month compared with patients younger than 50 years. Patients living 25–50 miles from their initial diagnosis facility had 20.9% higher odds of receiving treatment within one month compared to those within 25 miles. Stage II and III patients had 19.1% and 23.5% lower odds of receiving treatment within one month, respectively, than stage I patients. Patients aged 50-59 had nearly a 3-fold higher hazard ratio (HR) of death or disease progression (DDP) compared to 18-49, while 60-69 had a 2.2-fold higher HR. Patients aged 70-79 had a 4.2-fold higher HR; those aged 80+ had a 4.7-fold higher HR. Compared to White patients, Black patients had a 1.7-fold higher HR of DDP. Patients with dual eligible payer had a 2.4-fold higher HR of DDP compared to private payers. Patients who lived 25–50 miles from their initial diagnosis facility (IDF) had 72.1% higher HR of DDP compared to those within 25 miles, and those who lived 50+ miles from their IDF had more than 3-fold higher HR. Conclusions: Older age groups and patients with more advanced cancer stages had significantly lower odds of receiving any treatment within one month. Among patients undergoing systemic therapy, older age, black race, higher cancer stage, dual eligible payer status, and living farther from the diagnosis facility were significant risk factors for DDP.

Neoadjuvant and adjuvant pembrolizumab plus chemotherapy for locally advanced gastric or gastroesophageal junction adenocarcinoma: Outcomes from the microsatellite instability-high population of the phase 3 KEYNOTE-585 study.

4074 Background: In the randomized phase 3 KEYNOTE-585 study (NCT03221426), perioperative pembrolizumab (pembro) plus chemotherapy (chemo) significantly improved pathologic complete response (pCR) vs placebo (pbo) plus chemo (diff, 10.9%; 95% CI, 7.5-14.8; P < 0.00001), with no new safety signals in participants (pts) with locally advanced gastric or gastroesophageal junction (G/GEJ) adenocarcinoma. We present outcomes for the microsatellite instability-high (MSI-H) population. Methods: Eligible pts had untreated resectable locally advanced G/GEJ adenocarcinoma (including Siewert type 2 or 3 tumors); MSI was centrally assessed. Pts enrolled in the main cohort (n = 804) received neoadjuvant pembro 200 mg intravenously (IV) every 3 weeks (Q3W) or pbo plus chemo (cisplatin + capecitabine or 5-FU) for 3 cycles; after surgery, pts received adjuvant pembro or pbo plus chemo Q3W for 3 cycles, then adjuvant pembro or pbo Q3W for 11 cycles. Pts in the fluorouracil, docetaxel, and oxaliplatin (FLOT) cohort (n = 203) received neoadjuvant pembro 200 mg IV Q3W or pbo Q3W for 3 cycles plus FLOT Q2W for 4 cycles; after surgery, pts received adjuvant pembro or pbo Q3W for 3 cycles plus FLOT Q2W for 4 cycles, then adjuvant pembro or pbo Q3W for 11 cycles. End points were pCR (blinded independent central review [BICR]), event-free survival (EFS; RECIST v1.1 by BICR), overall survival (OS), and safety. We report outcomes in the main and FLOT cohorts combined. The data cutoff date was February 16, 2024 (final analysis). Results: Data from the 81 pts with MSI-H status were analyzed (n = 43, pembro + chemo; n = 38, pbo + chemo). The pCR rate was 39.5% (95% CI, 25.0-55.6) in the pembro group and 0% (95% CI, 0.0-9.3) in the pbo group (diff, 39.5%; 95% CI, 26.3-54.5). The median EFS was not reached (NR; 95% CI, 49.9-NR) and 60.1 months (95% CI, 18.6-NR), respectively (HR, 0.58; 95% CI, 0.26-1.31); 24/48-month EFS rates were 78%/73% in the pembro group and 68%/68% in the pbo group. The median OS was NR (95% CI, NR-NR) in both treatment groups (HR, 0.50; 95% CI, 0.19-1.30); 24/48-month OS rates were 91%/86% in the pembro group and 76%/74% in the pbo group. Treatment-related adverse events (AEs) occurred in all 43 pts (100%) in the pembro group and 37 (97%) in the pbo group. Grade 3-5 treatment-related AEs occurred in 32 (74%) and 25 (66%) pts, respectively. Discontinuation of any drugs due to treatment-related AEs occurred in 15 pts (35%) in the pembro group and 12 (32%) in the pbo group. No deaths due to treatment-related AEs were reported. Conclusions: In this post hoc analysis, efficacy outcomes for pts with MSI-H G/GEJ adenocarcinoma suggested a consistent trend with numerically more pronounced difference between the treatment groups, with a manageable safety profile. Further studies in this population are warranted. Clinical trial information: NCT03221426 .

Safety and preliminary efficacy of M9466 (HRS-1167) in combination with abiraterone acetate and prednisone/prednisolone (AA-P) in patients (pts) with metastatic prostate cancer in the phase 1 DDRiver 501 study.

5038 Background: M9466 (HRS-1167) is a highly potent, selective next-generation PARP1 inhibitor being investigated in pts with locally advanced/metastatic solid tumors. Here we report safety and preliminary efficacy data from Module 3 of the DDRiver 501 study, a Phase 1, open-label, global multicenter study, evaluating M9466 with AA-P in pts with metastatic prostate cancer (NCT06421935). Methods: DDRiver 501 Module 3 enrolled eligible pts with metastatic castration-resistant prostate cancer (mCRPC) or metastatic hormone-sensitive prostate cancer (mHSPC), irrespective of prior anticancer therapy received. Pts received M9466 (either 50 or 100 mg once-daily) with AA-P until disease progression. Primary objective was safety. Additional objectives included efficacy (objective response rate [ORR], PSA50 [confirmed decrease in prostate-specific antigen levels of >50% from baseline]), pharmacokinetics (PK), molecular responses (e.g. >50% reduction from baseline of methylation-based tumor fraction in circulating tumor DNA [ctDNA]) and tumor genetic profiling. Results: As of Nov 21, 2025, Module 3 enrolment was complete (N=21; M9466 50 mg n=10, 100 mg n=11; mHSPC n=1, mCRPC n=20; prior treatment included: PARP inhibitors n=0, androgen receptor pathway inhibitors [ARPI] n=21). Thirteen (62%) pts carried HRR gene mutations, including four (19.0%) with BRCA1/2 mutations, by ctDNA analysis. Pts received M9466 for a median (range) of 18.1 (6.0–30.1) and 26.9 (6.0–42.0) weeks (50 and 100 mg arms respectively). Safety outcomes are presented (table). No dose-limiting toxicities or serious treatment-related adverse events to any study intervention were observed. Across pts with baseline RECIST measurable disease, confirmed ORR was 18.8% (3/16, 95% CI: 4.0, 45.6). Two of the three responders had RAD50 loss of function mutations. In pts with baseline PSA ≥2 ng/mL, 3/17 (17.6%) achieved a PSA50 response. Molecular response and PK data will also be presented. Conclusions: Preliminary data from DDRiver 501 Module 3 demonstrate M9466 with AA-P is generally well tolerated and demonstrates antitumor activity in ARPI-pretreated pts with mHSPC/mCRPC. Clinical trial information: NCT06421935 . Safety overview. n (%) M9466 50 mg (n=10) M9466 100 mg (n=11) Any TEAE 9 (90.0) 11 (100) TEAEs (>25% of total pts) Anemia Fatigue Platelet count decreased Neutrophil count decreased 7 (70.0)6 (60.0)2 (20.0)3 (30.0) 9 (81.8)2 (18.2)5 (45.5)3 (27.3) Any ≥Grade 3 TEAE 4 (40.0) 5 (45.5) ≥Grade 3 TEAEs (≥n=2 of total pts) Anemia Platelet count decreased Neutrophil count decreased 2 (20.0)1 (10.0)1 (10.0) 5 (45.5)1 (9.1)1 (9.1) M9466-related TEAEs leading to: Discontinuation Reduction Interruption 02 (20.0)4 (40.0) 1 (9.1)*4 (36.4)4 (36.4) *Due to anemia. TEAE, treatment-emergent adverse events.

Influence of <i>TONSL</i> on tumor suppressor function of <i>RAD51</i> and resistance to CDK4/6 inhibitors in ER+ breast cancer.

1097 Background: Homologous Recombination (HR) and Replication Stress (RS) resolution mechanisms help to maintain genomic stability. TONSL, MMS22L, RAD51, BRCA1/2 are the few proteins that cooperatively coordinate HR and RS resolution. Therefore, genes in these pathways are expected to function as tumor suppressors. However, only BRCA 1/2 function as tumor suppressors as their germline mutations increase cancer susceptibility. TONSL and RAD51 are frequently overexpressed in tumors. Here, we investigated the relationship between TONSL-RAD51 axis in breast cancer (BC) outcome and response to CDK4/6 inhibitors (CDK4/6i) and endocrine therapy (ET). Methods: 10,980 ER+ BC samples tested by NGS (592, NextSeq; WES/WTS, NovaSeq, Caris Life Sciences). Dual expression group TONSL/RAD51 -high (H), TONSL -H/ RAD51 -low (L), TONSL -L/ RAD51 -H, and TONSL / RAD51 -L were classified by RNA expression above or below median. HRD score and TONSL /HRD-H/L survival was analyzed using METABRIC (ER+ BC, n = 1904) dataset. TONSL-RAD51 role in sensitivity to CDK4/6i was determined using TONSL -amplified ER+ MCF-7 cells. Real-world median overall survival (mOS) of TONSL/RAD51 groups was derived from insurance claims and calculated from biopsy, start of ET or CDK4/6i to last contact using Kaplan-Meier. Statistical significance was assessed using chi-square and Mann-Whitney U with multiple comparison adjustments (q&lt;0.05). Results: TONSL -H had higher HRD score (33.7 vs 21.1, p&lt;0.01) compared to TONSL -L (n=952 each). TONSL /HRD-H (n=573) had worse mOS (109 m vs 138 m, p&lt;0.05) compared to TONSL /HRD-L (n=630). TONSL -H had higher frequency of LOH (31.6% vs 19.8%), BRCA2 (5.1% vs 2.9%) and BRCA1 (1.5% vs 0.8%) mutation, all q&lt;0.05. TONSL/RAD51 -H had worse mOS compared to TONSL -H/ RAD51 -L, TONSL -L/ RAD51 -H or TONSL / RAD51 -L (Table). TONSL -H/ RAD51 -L had worse survival with ET compared to other groups. With ET+CDK4/6i, TONSL/RAD51 -H and TONSL- H /RAD51 -L had similar but worse mOS compared to TONSL -L/ RAD51 -H and TONSL / RAD51 -L (Table). TONSL knockdown in CDK4/6i resistant MCF-7 cells created TONSL -L/ RAD51 -H phenotype and cells regained the sensitivity to CDK4/6i. Conclusions: These findings suggest that TONSL -H independent of RAD51 expression is associated with poor response to ET or ET+CDK4/6i. Therapeutic targeting of TONSL to create TONSL-L/RAD51-H status may improve response to ET+CDK4/6i. Moreover, as the activity of TONSL is regulated by various protein complexes, destabilizers of TONSL-protein complexes could potentially help in sensitizing ER+ BC to ET or CDK4/6i. TONSL/RAD51 -HOS in months TONSL -H/ RAD51 -LOS in months TONSL-L/RAD51 -HOS in months TONSL/RAD51 -LOS in months p-value Overall 35 m (n=4176) 41.4 m (n=1473) 37.9 m (n=1476) 44.3 m (n=4173) &lt;0.01 ET 63.4 m (n=1361) 58 m (n=461) 65.2 m (n=508) 76.1 m (n=1430) &lt;0.01 ET+CDK4/6i 73.8 m (n=2261) 73.8 m (n=866) 84.6 m (n=799) 80.7 m (n=2291) &lt;0.01

Phase II study of olaparib plus ATR inhibitor ceralasertib in patients with metastatic breast cancer and germline BRCA1/2 pathogenic variants who progressed on prior PARP inhibitor therapy.

1142 Background: PARP inhibitors (PARPi) olparib and talazoparib improve objective response rate (ORR) and disease-free survival (DFS) compared with chemotherapy in patients with metastatic breast cancer and germline BRCA 1/2 pathogenic variants (gBRCA 1/2 PVs). Although the ORR is approximately 60%, about 40% of patients do not respond, and nearly all responders eventually develop progressive disease. Therefore, new strategies to overcome PARPi resistance are needed. Preclinical studies suggest that ATR inhibition combined with olaparib may restore PARPi sensitivity. Phase I studies have shown that olaparib (Ola) plus the ATR inhibitor ceralasertib (Cerala) is safe and have established the recommended phase II dose. This prospective phase II study evaluated the ORR of Ola plus Cerala in patients with metastatic breast cancer with gBRCA 1/2 PVs who previously progressed on PARPi therapy. Methods: Eligible patients had metastatic breast cancer (HER2- negative), prior PARPi exposure, and no more than two prior lines of chemotherapy. After informed consent, all patients underwent the baseline research biopsy of a metastatic lesion and provided blood samples for exploratory analysis aimed at identifying mechanisms of PARPi resistance. Treatment consisted of olaparib 150 mg orally twice daily on days 1-28 plus ceralasertib 80 mg orally twice daily on days 1-14 of each 28-day cycle. The primary endpoint was ORR per RECIST 1.1. Secondary endpoints included safety, DFS, and duration of response. Exploratory endpoints focused on biomarker analysis of PARPi resistance. This abstract reports the primary end point. Results: Fifteen patients were enrolled. The median age was 39 years. Eight patients had hormone receptor-positive disease; 6 of these had received endocrine therapy before first-line PARPi. Two patients achieved a partial response (PR) and two had stable disease (SD), resulting and an ORR of 13.3% and a clinical benefit rate of 26.7%. No grade 4 toxicities were observed. Grade 3 adverse events included anemia (N=5), neutropenia (N=1), decreased white blood cell count (N=1), lymphopenia (N=1), and thrombocytopenia (N=1). Conclusions: The combination of olaparib plus ceralasertib is well tolerated and demonstrates clinical activity in patients with metastatic breast cancer with gBRCA 1/2 PVs who have progressed on prior PARPi therapy. A study of this combination in the first-line setting may help determine whether ATR inhibition can overcome or delay PARPi resistance. Biomarker analysis to identify mechanisms of resistance are ongoing and will be reported separately. Clinical trial information: NCT04090567 .

Enhancing grid stability using dynamic reserve power point tracking techniques

Grid stability is of prime importance for grid-tied solar power systems as they are prone to power quality issues caused by the varying intensity of sun radiation and grid disturbances. To maintain grid stability, various PV power tracking algorithms have been developed. However, classical power tracking models often fail to maintain grid stability and sustain required power reserves under real-time variations in grid conditions and solar generation. To address this limitation, a Dynamic Reserve Power Point Tracking (DRPPT) control algorithm is proposed to ensure grid stability by dynamically adjusting reserve power. By continuously monitoring the PV array and grid conditions, the proposed controller determines the dynamic solar reserve power and accordingly selects the suitable operating mode. The operating point of PV array is then regulated by Flexible Power Point Tracking (FPPT) technique, which performs fine-tuning of the reference voltage followed by grid injection. By combining FPPT and Maximum Power Point Tracking (MPPT) functionalities, the proposed DRPPT controller maintains optimal reserve levels, ensuring the grid can rapidly respond to sudden changes in power supply or demand while meeting customer requirements. The proposed model is tested on hardware and simulated on software, and both results show the ability of DRPPT algorithm to handle real-time grid frequency changes and adapt the RPPT operation accordingly to meet grid stability standards. The proposed model has achieved superior THD mitigation, thereby improving grid stability, with 53.75%, 50%, and 7.5% lower THD compared to the conventional RPPT, Global FPPT (GFPPT), and Genetic Algorithm (GA)-based FPPT techniques, respectively.

Bioprospecting of seaweed: Au nanocluster synthesis, characterization and theoretical studies

Cluster‐Free Intrinsic Assembly for Efficient and Stable Perovskite Light Emitting Diodes

ABSTRACT Metal halide perovskites have emerged as transformative candidates for next‐generation optoelectronic materials, yet their performance remains constrained by an inherently rapid and uncontrolled crystallization process driven by pre‐aggregated clusters in precursors. While cluster‐related challenges have been partially explored in perovskite photovoltaics, systematic investigations into their mechanisms and practical solutions for perovskite light‐emitting diodes (PeLEDs) remain scarce. Herein, we introduce a cluster‐free intrinsic assembly strategy to fundamentally reshape the crystallization dynamics of perovskites. By exploiting the diuretic furosemide (FRSM) as an ionic binder, we achieve simultaneous coordination of all ionic components within the perovskite precursor, effectively suppressing cluster formation and redirecting crystallization toward a cluster‐free intrinsic assembly pathway. The resulting perovskite films exhibit homogeneous high‐quality perovskite nanocrystal structure, with exceptional optoelectronic properties and remarkable ambient stability. These advancements enable PeLEDs with a record external quantum efficiency (EQE) of 31.0% alongside unprecedented operational stability (equivalent T 50 &gt;310 000 h at 100 cd m −2 , T 90 &gt; 1000 h at 1000 cd m −2 ), establishing new performance benchmarks for PeLEDs. Our work establishes a paradigm linking precursor‐state engineering to film‐quality determinism, demonstrating that eliminating conventional cluster‐dominated aggregation pathways can revolutionize the assembly process and unlock the potentials of perovskite optoelectronics.

Texture-controlled anisotropic strength and ductility in hot-rolled Mg-2Zn-0.1Ca alloy

KDM5A methylation modulates its genomic demethylase and transcriptional actions

Novel adaptive versus continuous dosing of encorafenib and binimetinib in combination with nivolumab in advanced <i>BRAF V600E</i> mutant melanoma.

9538 Background: In BRAF V600 mutant melanoma, continuous treatment with BRAF/MEK inhibitors invariably begets acquired resistance. Preclinical success of intermittent dosing to overcome resistance did not translate to clinical benefit in the randomized SWOG S1320 trial, possibly due to fixed schedule of intermittent dosing (5 wk on/3 wk off cycles). Leveraging intra-tumoral competition dynamics to slow expansion of resistant tumor subpopulations via patient-specific adaptive MAPK therapy may improve outcomes. We hypothesized that adaptive drug scheduling of BRAF/MEKi combined with anti-PD-1 would lead to non-inferior or improved outcomes with less toxicity compared to a continuous triplet regimen. Methods: In a pilot prospective trial, pts with advanced BRAF V600E mutant melanoma, naïve to systemic tx in this setting, with detectable plasma circulating tumor DNA (ctDNA) were eligible (using BRAFV600E ctDNA ddPCR). Ten pts were randomized 1:1 to continuous (Arm A) or patient-specific adaptively dosed (Arm B) ENCO plus BINI, with continuous biweekly NIVO in both arms. A mathematical model was calibrated for Arm B using pts’ ctDNA values (baseline and biweekly sampling) and standard CT response imaging every 12 weeks. This model was a coupled system of differential equations to guide treatment decisions biweekly by simulating the future response on or off drugs (ENCO+BINI) based on that pt’s known ctDNA levels and response data. Therefore, pts in Arm B all had differing lengths of time that they received ENCO+BINI. Primary endpoint was feasibility of the adaptive approach, by adherence to protocol from baseline to week 12. Results: Ten pts, median age 59, 70% male, 80% with stage IV melanoma, were randomized. In Arm B, NIVO alone was often effective enough to not trigger resuming ENCO+BINI; in Arm A, pts were on E+B for a median of 61 weeks (89% of time on tx, as drugs could be held for toxicity), while in adaptive Arm B, patients were on E+B for median 11 weeks (24% of time on tx). ORR was 100% in Arm A and 60% in Arm B, all partial responses, with 2 stable disease in Arm B. At median follow-up of 45.5 months, median PFS was 26.9 months (95% CI 5.5-NA) with 3 progressors in Arm A. Median PFS was not reached (17.8-NA) in Arm B and 2 progressed (log-rank p= 0.5). Median OS was 32 months (8.8-NA) in Arm A (2-year OS 60%, 3 pts died). Median OS was not reached (17.8-NA) in Arm B (2-year OS 75%,1 pt died), log-rank p=0.18. In Arm A, there were 16 treatment-related grade 3 adverse events, with only 6 grade 3 AEs in Arm B. Conclusions: In this innovative study, we demonstrated the feasibility of patient specific mathematical modeling incorporating ctDNA and imaging, to inform adaptive dosing of ENCO+BINI with concurrent NIVO. The adaptive arm did not compromise efficacy and had lower toxicity. Successful confirmation of this approach in larger trials may favorably impact patient QOL and reduce cost of therapy. Clinical trial information: NCT03543969 .

Impact of pembrolizumab infusion timing on outcomes in early-stage triple-negative breast cancer: A real-world exploratory analysis (Neo-Real).

610 Background: Neoadjuvant pembrolizumab combined with chemotherapy is the standard of care for early-stage triple-negative breast cancer (TNBC) based on KEYNOTE-522, improving pathologic complete response (pCR) and event-free survival (EFS). Emerging data suggest that immunotherapy timing may influence oncologic outcomes; however, evidence in early-stage TNBC and in combination regimens is limited. We conducted an exploratory real-world analysis to evaluate the association between infusion timing and clinical outcomes. Methods: Neo-Real/GBECAM-0123 is a multicenter real-world cohort including patients with early-stage TNBC treated with neoadjuvant pembrolizumab-based chemotherapy in Brazil and Argentina. In this analysis, patients treated between 2019 and 2024 with available infusion timing data were included. Infusion timing was classified as Early (≤20% of pembrolizumab infusions initiated at or after 14:00) or Late (&gt;20% initiated at or after 14:00). The primary endpoint was pCR (ypT0/Tis ypN0). Secondary endpoints included EFS, defined according to KEYNOTE-522 criteria. Sensitivity analyses using timing definitions were performed. Results: Among 419 patients with available infusion timing data (Early: n=265; Late: n=154); 385 underwent surgery and were evaluable for pCR. Baseline clinicopathologic characteristics were largely balanced between groups, although stage III disease was numerically more frequent in the Early group (30.9% vs 23.6%, p=0.134). Treatment features, including use of dose-dense chemotherapy and number of neoadjuvant pembrolizumab cycles (median 8 in both groups), were similar. pCR rates did not differ according to infusion timing (Early: 62.5% vs Late: 64.1%; p=0.827). Late infusion timing was associated with improved EFS, with a 2-year EFS of 90.7% versus 82.7% (p=0.013). In multivariable Cox regression adjusting for age, clinical stage, pathologic response, tumor grade, histology, and number of neoadjuvant pembrolizumab cycles, Late infusion timing remained associated with improved EFS (HR 0.34; 95% CI 0.15–0.80; p=0.013). Sensitivity and subgroup analyses, including evaluation of the timing of the first pembrolizumab infusion using an earlier cut-off (11:00), as well as analyses by histology, clinical stage, and pathologic response, yielded consistent results. Conclusions: In this exploratory real-world analysis of patients with early-stage TNBC treated with neoadjuvant pembrolizumab-based chemotherapy, infusion timing was not associated with pCR. However, later infusion timing was associated with improved EFS. These findings contrast with observations reported in other tumor types and underscore the need for further studies to clarify the impact of immune checkpoint inhibitor infusion timing on clinical outcomes in TNBC.

Liver transplantation as a therapeutic strategy for far-advanced hepatic malignancies: Survival outcomes from the LT-FAM study.

e16238 Background: As liver transplantation (LT) indications expand, it is increasingly explored for select patients with hepatic malignancies far beyond conventional LT criteria. These far-advanced tumors are characterized by very large or innumerable lesions, extreme tumor marker elevation, and previously treated extrahepatic metastatic disease. We evaluated oncologic and survival outcomes following LT across five malignancy types. Methods: We retrospectively reviewed patients undergoing LT from 2015–2025 for far-advanced hepatocellular carcinoma (HCC), hilar cholangiocarcinoma (hCCA), intrahepatic cholangiocarcinoma (iCCA), nonresectable colorectal liver metastases (nCRLM), and metastatic neuroendocrine tumors (NET). Far-advanced disease was defined by excessive tumor size or burden, extreme tumor marker elevation, and/or treated extrahepatic metastases prior to LT. Tumor characteristics and survival outcomes/Kaplan–Meier estimation were analyzed. Results: Fifty-six patients underwent LT, 93% via living donors. All malignancy groups demonstrated extensive tumor burden with high rates of multifocal or innumerable hepatic disease and biomarker elevation. Treated extrahepatic metastases were present in 55% of HCC and 29% of nCRLM recipients. Median overall survival was 2.7 years for HCC, 1.9 years for hCCA, 2.6 years for iCCA, 2.4 years for nCRLM, and 6.8 years for NET. Recurrence ranged from 18–64%, and mortality from 9–50%, with NET demonstrating the most favorable outcomes. Conclusions: In carefully selected patients with hepatic malignancies far exceeding traditional LT criteria, LT, predominantly via living donor pathways, may achieve meaningful survival despite high-risk tumor features. These findings support the feasibility of LT within an expanding transplant oncology paradigm and emphasize selection strategies guided by tumor biology rather than conventional criteria alone. HCC N=11 hCCA N=14 iCCA N=11 nCRLM N=14 NET N=6 Age at Transplant, years (median) 31 46 59 53 58 Sex (n, %Male) 4 (36%) 10 (71%) 7 (64%) 11 (79%) 6 (100%) Living Donor Liver Transplant (n, %) 11 (100%) 13 (93%) 11 (100%) 14 (100%) 3 (50%) Number of hepatic lesions (n, %) 1-3 4-7 8-10 Innumerable 3 (27%)5 (45%)2 (18%)1 (9%) 14 (100%) 9 (82%)1 (9%)1 (9%) 5 (36%)3 (21%)2 (14%)4 (29%) 1 (16%)5 (83%) Max Tumor Diameter (mean, cm) 10 3.6 7.0 6.6 4.8 Extrahepatic Metastatic disease (n, %) 6 (55%) 0 0 4 (29%) 0 Pre-LT Oncologic Treatments (n, %) Liver Resection 7 (63%) 3 (23%) 0 5 (36%) 2 (33%) Local-Regional Therapy 11 (100%) 0 6 (55%) 0 0 Hepatic Artery Infusion Pump -- -- -- 7 (50%) -- Systemic Chemotherapy 9 (82%) 14 (100%) 8 (73%) 14 (100%) 5 (83%) Radiation Therapy 2 (18%) 11 (79%) 0 2 (14%) 0 Immunotherapy 7 (64%) 0 2 (18%) 0 0 Recurrence Post Transplant (n, %) 7 (64%) 5 (36%) 2 (18%) 7 (50%) 3 (50%) Deaths Post Transplant (n, %) 3 (27%) 4 (29%) 1 (9%) 4 (29%) 3 (50%) Overall Survival (median, years) 2.7 1.9 2.6 2.4 6.8

Pan-cancer discovery of clinical and genomic determinants of immune-related adverse events.

11132 Background: Immune-related adverse events (irAEs) from immune checkpoint inhibitors (ICIs) can be lifelong, fatal, or require treatment discontinuation, substantially limiting the clinical benefit of immunotherapy. Despite their impact, predictors of irAE susceptibility remain poorly defined, largely due to the lack of scalable approaches for toxicity phenotyping. Here, we leverage large language models and integrated clinico-genomic data to identify determinants of irAEs. Methods: We developed a custom retrieval augmented generation and large language model (RAG-LLM) pipeline to automatically annotate 6 key adverse events (adrenal insufficiency, hepatotoxicity, hyperthyroidism, hypothyroidism, colitis, and pneumonitis) using free text from clinical notes across Memorial Sloan Kettering Cancer Center. We first validated RAG-LLM predictions using a gold standard prospectively collected adverse event dataset from 8,119 patients across 1,057 individual clinical trials. RAG-LLM imputations were then scaled to 55,406 (12,291 ICI-treated) patients. All patients had associated somatic and germline MSK-IMPACT panel sequencing data. Single nucleotide polymorphism (SNP) imputation was performed using GLIMPSE and time-to-event genome-wide association studies (GWAS) were performed using SPACox. HLA class I genotypes were imputed using HLA-HD. Random Survival Forest (RSF) models were trained to predict irAE occurrence. Results: The custom RAG-LLM pipeline had strong performance across all 6 irAEs (area under the ROC curve of 0.77-1.00). In the RAG-LLM imputed data, 799 (1.4%) patients had adrenal insufficiency, 2,803 colitis (5.1%), 7,130 hypothyroidism (12.9%), 448 hyperthyroidism (0.8%), 15,627 hepatotoxicity (28.2%), and 3,730 pneumonitis (6.7%). Among ICI-treated patients, pneumonitis (HR 1.4; 95% CI 1.1-1.8; p &lt; 0.01) and hepatotoxicity (HR 1.4; 95% CI 1.2-1.6; p &lt; 0.01) were associated with worse overall survival. The GWAS found two genome-wide significant hits that predicted adrenal insufficiency (rs115003145, HLA region) and hypothyroidism (rs7864322, FOXE1 gene enhancer region) with p &lt; 5x10 -8 . Fine mapping of the HLA SNP showed that HLA-C*06:02 was specifically associated with increased risk of adrenal insufficiency in ICI-treated patients only (HR 1.6; 95% CI 1.2-2.2; p &lt; 0.01). RSF model performance for irAE occurrence had F1 scores of 0.67-0.85. For pneumonitis, patients with the highest risk quartile by RSF model had 7.1% risk of pneumonitis at 1 year compared to 0.4% for the lowest risk quartile (HR 14.9; 95% CI 10.4-21.3; p &lt; 0.01). Conclusions: We developed and validated a novel custom RAG-LLM pipeline that allows automatic annotation of irAEs using clinical notes. Using this pipeline, we identified novel biomarkers of ICI-related adrenal insufficiency and hypothyroidism. Finally, we developed predictive models for irAE prediction that can be used at the point of care.

Linking T-cell receptor dynamics to outcomes in neoadjuvant-treated melanoma.

9573 Background: Clinical Stage III (cSIII) melanoma patients are at high risk of recurrence following therapeutic lymph node dissection (TLND) and adjuvant therapy. Neoadjuvant immunotherapy based regimens have shown promising results in this setting, with pathological response being the best predictor of recurrence free survival (RFS) and distant metastases free survival (DMFS). T-cell Receptor (TCR) abundance and clonality are being explored as biomarkers. Methods: In the phase II NeoACTIVATE (NCT03554083) trial, patients were treated based on BRAF status with 12 weeks of neoadjuvant atezolizumab, vemurafenib and cobimetinib (Arm A, BRAF-mutated) or atezolizumab and cobimetinib (Arm B, BRAF wildtype), followed by TLND and 6 months adjuvant atezolizumab. TCR sequencing (n=23) was performed on the pretreatment involved lymph node (iLN) and on the same node, along with an adjacent uninvolved node (uiLN) after TLND. The fraction productive of T cells (fpTC) was defined as the percentage of in-frame TCR sequences (capable of encoding a functional expressed receptor protein) of all TCR sequences identified. Simpson Clonality Index (SCI) was used to measure relative abundance of specific clones, with a higher score indicating a more monoclonal repertoire. Comparisons were made using the Wilcoxon Rank Sum test. Results: 30 patients were enrolled, 15 in each arm. 2 patients were not operated on, and one did not receive adjuvant treatment, leaving 27 patients for outcome analysis. 3-year RFS was 55.6% (95% CI 9.5%-77.8%), median of 62.2 months (95% CI 19-NR); DMFS was 64.3% (95% CI 48%-86.3%), median 62.2 months (95% CI 33.9-NR). The fpTC was higher in the pretreatment iLN for those without an RFS event (p=0.04) with a trend in the uiLN as well (p=0.06). Patients without a DMFS event had a higher fpTC in the uiLN (p=0.03) but not in the iLN. The SCI was significantly higher only in the post treatment iLN for patients without an RFS event (p=0.03) and was not significantly different when compared based on DMFS status. When performing pairwise comparisons of nodes based on recurrence status, we found that patients who did not suffer an RFS event had a significantly higher SCI in the post treatment iLN compared to the uiLN (p=0.02). Conclusions: Neoadjuvant immunotherapy and targeted therapy for cSIII melanoma can lead to durable RFS. Higher pretreatment fpTC as well as higher post-treatment iLN TCR clonality were both associated with improved RFS. These data indicate that TCR repertoire characteristics within iLN and uiLN may reflect differential immune responses to neoadjuvant therapy and could inform future studies evaluating biomarkers to stratify risk of recurrence and allow personalization of therapy. Clinical trial information: NCT03554083 . fpTC Simpson Clonality Index RFS event DMFS event RFS event DMFS event Pretreatment iLN 0.04 0.12 0.43 0.11 Posttreatment iLN 0.09 0.09 0.03 0.12 uiLN 0.06 0.03 0.79 0.48 Δ iLN (post-pre) 0.79 0.37 0.07 0.41 Δ posttreatment iLN – uiLN 0.21 0.21 0.02 0.08

Phase II pilot study of pembrolizumab (PEM) and neoadjuvant radiation therapy (nRT) in high-risk soft tissue sarcomas (STS).

11584 Background: Pathologic necrosis (PN) after neoadjuvant therapy in STS has been associated with improved outcomes. We investigated clinicopathologic outcomes of concurrent PEM/nRT in STS. Methods: Adults with localized STS [FNCLCC grade 3 ( &gt; 1.5 cm) or grade 2 ( &gt; 3 cm)] received 3 doses of PEM (200 mg IV q21 d) concurrently with nRT, followed by surgery. Primary endpoint was the rate of post-treatment complete pathologic necrosis (CPN; ≥90% PN). 26 subjects provided 94% power to detect an increase in CPN rate from 15% to 40%; ≥7 cases with CPN would exclude a 15% CPN rate. Secondary endpoints included toxicity (CTCAE 5.0), response (RECIST 1.1), and event-free survival (EFS). Results: 27 STS patients were enrolled including UPS (10; 37%), DD liposarcoma (8; 30%), myxofibrosarcoma (2; 7%), solitary fibrous tumor (2; 7%), and other STS (5; 19%). 26 patients were evaluable for CPN and 25 for EFS. All except 2 (7%) received 3 planned PEM doses. All completed planned RT (50-50.4 Gy). 2 (7%) with UPS had partial responses prior to surgery. Median tumor size was 10 cm (range 2.1-26 cm). R0 resection was achieved in 92%. 14 (54%) patients had grade 3 tumors; 12 (46%) had grade ≥2. Median post-treatment PN was 55% (range 0-99%) with 5/26 (19%) having CPN. Median follow-up time was 33.4 months (m) from PEM initiation (range 1.6-60 m). 12/25 patients had an EFS event. Among all patients (n = 27), 12 developed only distant metastatic disease, and 1 developed local and distant recurrence. AEs were consistent with anticipated effects of PEM and RT. Two grade 5 events occurred (ischemic colitis, prostate cancer), both judged unrelated to study therapy. 3 patients experienced wound dehiscence, and one had wound infection. EFS at 12 m was 68%; at 57.5 m, it was 52%. Post-treatment PN as a continuous variable was inversely associated with EFS (HR = 22.47, p = 0.004). Increased tumor size was also associated with inferior EFS (HR = 1.09/cm, p = 0.04). In multivariable analyses, both necrosis and tumor size remained independently associated with EFS (HR = 26.52 and 1.09, p = 0.004 and 0.041, respectively). Sex, age, histology, tumor grade/location, and radiation modality were not associated with EFS. Conclusions: PEM did not increase the proportion of patients with post-treatment CPN after RT. Unexpectedly, increased tumoral necrosis after treatment was associated with inferior EFS. It is possible that increased post-treatment tumoral necrosis may identify a subset of STS patients less likely to benefit from PEM/nRT or who have inherently more aggressive tumor biology. These findings warrant additional studies for validation and exploration of the mechanisms and clinical applications of these findings. PN after PEM/nRT might provide a readily available biomarker that could be used to adjust therapy in those unlikely to benefit from PEM/nRT. Correlative analysis using pre- and post-treatment tumor are on-going and will be presented. Clinical trial information: NCT03338959 .

Real-world (RW) effectiveness and safety of lurbinectedin (lurbi) for previously treated extensive-stage small cell lung cancer (ES-SCLC): Final primary and subgroup analysis results of Jazz EMERGE 402.

8079 Background: Lurbi received accelerated approval in 2020 for ES-SCLC that progressed on or after platinum-based treatment (Tx) based on a phase 2 basket trial and full approval in 2025 combined with atezolizumab as 1st-line maintenance Tx for ES-SCLC based on the phase 3 IMforte trial. Methods: The phase 4, prospective, observational Jazz EMERGE 402 trial (NCT04894591) evaluated lurbi for previously treated ES-SCLC in RW practice in North America (final data cut: July 17, 2025). The primary endpoint was overall response rate (ORR). Key secondary endpoints were progression-free survival (PFS), overall survival (OS), and safety. Effectiveness was assessed in all patients (pts) and in prespecified subgroups. Results: At the final data cut, 267 pts had received ≥1 cycle of lurbi. At baseline (BL), median (min–max) age was 67 (29–89) years, 52 (19%) pts had an ECOG PS ≥2, 67 (25%) had brain metastases, 84 (31%) had liver metastases, 88 (33%) had a chemotherapy-free interval (CTFI) &lt;90 days, and 196 (73%) had ES-SCLC as the initial diagnosis. Pts received lurbi as 2nd-line (169 [63%]), 3rd-line (78 [29%]), or later (20 [7%]) Tx. Median (Q1–Q3) number of lurbi Tx cycles and Tx duration were 4 (2–7) and 91 (56–170) days. Granulocyte colony-stimulating factor was used in 99 (37%) pts (71 [27%] as primary prophylaxis). Six (2%) pts were receiving Tx at study completion; 261 (98%) discontinued Tx, with disease progression (183 [70%]) the primary reason. While lurbi was effective among all pts and poor-prognosis subgroups, better outcomes tended to occur in pts with CTFI ≥90 days and ECOG PS &lt;2 (Table). Eighty-eight (33%) pts had Tx-related adverse events (TRAE); anemia (21 [8%]) and neutropenia (19 [7%]) were most common. Rates of serious neutropenic infection (5 [2%]), anemia (4 [1%]), and neutropenia (3 [1%]) were low. Conclusions: Lurbi was associated with clinically meaningful effectiveness and predictable/manageable safety in previously treated ES-SCLC in a RW population that included poor-prognosis subgroups. Clinical trial information: NCT04894591 . Effectiveness among all pts and by subgroup. AllN = 267 Age &lt;65 Yearsn = 101 Age ≥65 Yearsn = 166 CTFI &lt;90 Daysn = 88 a CTFI ≥90 Daysn = 137 a Initial LSn = 69 b Initial ESn = 196 b ECOG PS &lt;2n = 178 c ECOG PS ≥2 n = 52 c ORR, d % (95% CI e ) 29 (23, 36) 30 (20, 43) 28 (20, 37) 24 (14, 37) 28 (19, 38) 33 (21, 47) 27 (20, 36) 26 (19, 35) 41 (24, 59) PFS, d months, median (95% CI) 3.3 (2.6, 4.1) 4.1 (2.8, 4.5) 2.9 (2.2, 3.8) 2.9 (2.0, 4.1) 3.3 (2.4, 4.2) 4.0 (2.4, 5.8) 3.3 (2.5, 4.1) 3.3 (2.6, 4.2) 2.6 (1.7, 5.2) OS, months, median (95% CI) 7.6 (6.4, 8.7) 8.2 (7.0, 10.6) 6.6 (5.8, 8.7) 5.8(4.6, 6.6) 9.3(7.1, 10.9) 8.7 (6.2, 10.6) 6.8 (6.0, 8.7) 8.1(6.6, 9.6) 5.2 (2.4, 7.9) a CTFI missing for 42 pts. b Stage missing for 2 pts. c ECOG PS missing for 37 pts. d Per RECIST v1.1 in pts with BL measurable disease. e Estimated using Clopper-Pearson exact method. LS, limited stage.

A deep learning approach to quantitatively capture cerebellar overgrowth in pediatric NF1 patients.

e14001 Background: Macrocephaly is one of the most common clinical manifestations of NF1 patients. As compared to normals, NF1-associated brains show enlarged gray and whiter matter volumes with varying degrees of regional specificity. Despite this knowledge, little is known about NF1-specific cerebellar volume changes in childhood, particularly relative to their age-matched and gender-matched controls. We sought to quantitatively capture cerebellar enlargement in pediatric NF1 patients against a large cohort of age and gender-matched, control patients. Methods: Clinical data and 3D T1 MPRAGE sequences (either pre or post contrast) were acquired from 61 pediatric NF1 patients, between the ages of 1 and 21, seen at the University of Minnesota Masonic Children’s Hospital (UMMCH) and satellite clinics. Imaging and clinical data from 956 control patients was obtained from a variety of public sources. SynthSeg, an open-source, whole brain segmentation model was used to segment all intracranial structures including the cerebellum from all available imaging. Results: Males and females with NF1 had, on average, 17.1% and 12.9% larger cerebellar volumes, respectively, compared to age-matched peers. Median percentile for cerebellar volume in males and females with NF1 was 96.4 and 94.6, respectively. Cerebellum-to-intracranial volume ratio was overall maintained in males with NF1 (median: 52nd percentile), but not in females with NF1 (24th percentile). Conclusions: Pediatric NF1 cerebellar volumes can be captured by automated means. Pediatric NF1 patients appear to have significantly enlarged cerebellar volumes as compared to their peers. Despite overall cerebellar enlargement in both genders, females had a relatively larger overgrowth of total intracranial volume compared to their cerebellums.