Browse Articles
Discover research articles across all indexed journals
ROCK-2: An ongoing open-label, randomized, multicenter phase 2 study of rocbrutinib versus investigator's choice of therapy in patients with R/R non-GCB DLBCL.
TPS7106 Background: Diffuse large B-cell lymphoma (DLBCL) is the most common type of lymphoma. Aberrant activation of the Bruton's tyrosine kinase (BTK)-related signaling pathway is present in the non-GCB subtype of DLBCL (Davis et al. Nature 2010). Multiple BTK inhibitors (BTKi) have demonstrated efficacy in relapsed or refractory (R/R) non-GCB DLBCL (Strati et al. Haematologica. 2021; Wilson et al. Nat Med. 2015; Yang et al. Blood Adv. 2022), however, there remains room for further improvement. Rocbrutinib is a novel, highly selective, the fourth generation covalent (irreversible) and non-covalent (reversible) BTKi. Phase 1 study showed that in 45 patients with R/R non-GCB DLBCL who had received ≥2 prior lines of systemic therapy, rocbrutinib monotherapy achieved an overall response rate (ORR) of 57.8% and a complete response (CR) rate of 31.3%( Song et al. CSCO. 2025). This article describes the design and progress of a phase 2 pivotal study aimed at comparing the efficacy and safety of rocbrutinib versus investigator's choice of therapy in patients with R/R non-GCB DLBCL. Methods: ROCK-2 (NCT07189065; CTR20253693) is an open-label, randomized, multicenter phase 2 study conducted in China. Eligible patients must have non-GCB DLBCL, not otherwise specified (NOS) per the 2017 WHO classification, received ≥2 prior lines of systemic therapy. Additionally, patients must have measurable disease, an ECOG performance status of 0-2, and adequate hematologic and organ function. Key exclusion criteria include central nervous system involvement, DLBCL transformed from indolent lymphoma, prior refractoriness to BTK-targeting agents, uncontrolled systemic diseases, active infection. Approximately 150 eligible subjects will be randomized 1:1 to receive either rocbrutinib or BR/R 2 , stratified by the number of prior lines of therapy (2 vs ≥3) and International Prognostic Index (IPI) score (0-2 vs 3-5). Subjects in the experimental group will receive rocbrutinib [200 mg once daily (QD)] until disease progression or unacceptable toxicity. Subjects in the control group will receive either BR or R 2 according to the investigator's choice. The BR regimen consists 6 cycles of rituximab (375 mg/m² on day 1) plus bendamustine (90 mg/m²/d on days 1-2). The R 2 regimen consists of rituximab (375 mg/m² on day 1 of cycles 1-6) plus lenalidomide (20 mg QD on days 1-21 of each 28d-cycle until disease progression or unacceptable toxicity). The primary endpoint is ORR assessed by Independent Review Committee (IRC) according to the 2014 Lugano criteria. Secondary endpoints include ORR assessed by investigator (INV), and CR rate, progression-free survival (PFS), duration of response (DOR), time to response (TTR), overall survival (OS) assessed by IRC or INV. The first subject was enrolled on Nov 27 th ,2025. Recruitment is ongoing. Clinical trial information: NCT07189065 .
Identifying patients with human epidermal growth factor receptor 2 (HER2)–low and –ultralow breast cancer (BC): Use of digital, artificial intelligence (AI)–based computational algorithms to assist HER2 scoring by pathologists.
1022 Background: Trastuzumab deruxtecan is approved in HER2-low (IHC 1+ or 2+/ISH negative) or -ultralow (IHC 0 with membrane staining in ≤10% of tumor cells) metastatic BC per DESTINY-Breast04 and -06 trials. Manual HER2 IHC scoring can be time-consuming and subjective. This retrospective, real-world study assessed manual scoring (ground truth) vs standalone AI-computational pathology-assisted (CPa) tools for scoring. Methods: Whole slide images (WSIs; N = 600, scanned with Aperio AT2) of BC samples stained with PATHWAY HER2 (4B5) assay, originally scored as HER2 IHC 0 (n = 400) or 1+ (n = 200), were rescored by 3 pathologists and using 4 CPa/AI tools in development as either IHC 0 absent membrane staining, 0 with membrane staining in ≤10% of cells, 1+, or 2+. Using 2023 ASCO/CAP guidelines, each pathologist performed 2 blinded readings per WSI; a reconciled score was used if the 2 readings differed. Manual consensus required ≥2/3 agreement. Concordance between pathologist manual consensus vs standalone CPa scoring was measured by positive and negative percentage agreement (PPA; NPA), overall percentage agreement (OPA), and Cohen κ, with review time recorded. Results: Of 600 WSIs, 586 (97.7%) had manual consensus. CPa tools were faster (Table) than manual scoring (manual median review time: 7.0 min; range, 3.0-11.5). PPA was high (≥92.5%) between manual consensus scoring and CPa tools; NPA across the tools was 79.1%, 68.9%, 67.4%, and 23.1%. Overall concordance varied across CPa tools; OPA ranged from 48.5% to 75.4% and Cohen κ from 0.26 to 0.61. Conclusions: Integrating CPa/AI tools as decision support aids for pathologists may reduce pathologist review time and augment pathologist inter-observer reproducibility, especially in HER2-ultralow identification. Refinement of CPa/AI algorithms in development may improve scoring to an even greater extent. DP tool (N = 586 a ) Review time, median (range), min PPA, b,c % (95% CI) NPA, b,d % (95% CI) OPA, b,e % (95% CI) Cohen κ e (95% CI) RV73X 2.7 (0.3-56.8) 94.9 (92.5-96.7) 79.1 (70.6-85.7) 75.4 (71.7-78.9) 0.61 (0.56-0.66) MQ52G 0.7 (0.1-8.5) 93.4 (90.9-95.4) 68.9 (60.5-76.2) 73.0 (69.3-76.6) 0.57 (0.52-0.63) KL84Q 3.1 (0.5-84.4) 92.5 (89.8-94.6) 67.4 (59.1-74.8) 69.5 (65.6-73.2) 0.54 (0.48-0.59) ZX19P 2.5 f (Not available) 99.1 (97.6-99.6) 23.1 (16.4-31.5) 48.5 (44.4-52.6) 0.26 (0.22-0.31) a Primary and metastatic samples from biopsies, effusions, fine needle aspirations, and surgical resection; due to the inbuilt pre-QC module, a few WSI outputs were not processed by RV73X and ZX19P. b Rounded to 1 decimal. c Agreement for consensus-positive cases (IHC 0 with membrane staining, 1+, or 2+). d Agreement for consensus-negative cases (IHC 0 absent membrane staining). e Agreement across all IHC scores. f Mean review time; median not available.
Aromatase independent estradiol synthesis in triple-negative breast cancer cells.
e13110 Background: Breast cancer is a leading cause of cancer related deaths in women in the United States. With advancements in universal screening technology, still one in eight women is likely to be diagnosed in their lifetime. Multiple factors including genetic predispositions (BRAC-1/2, TP53, STK11), environmental exposures (air pollutants, industrial chemicals), and individual risk factors (nulliparity, obesity) are presumed to contribute to the progression of breast cancer. Breast cancers are hormonally regulated signaling to Estrogen (Er + ), Progesterone (Pr + ), or Human Epidermal Growth Factors Receptor (HER 2+ ). Another variant, triple-negative breast cancer (TNBC), minimally expresses hormone receptors and lacks targeted therapy. TNBC typically follows an aggressive course leading to high morbidity and mortality rates. Steroidogenic hormone estradiol plays a key role in promoting breast cell growth and can in turn lead to tumorigenesis. Aromatase is the enzyme that converts androgens to estradiol. In TNBC, aromatase activity is often low or absent leading to TNBC being considered estrogen independent. Using tumorigenic and unaffected breast tissue, our lab has identified a binding partner with aromatase, aromatase interacting parter in breast (AIPB), whose over expression down regulates estradiol synthesis. Use of anti-helminth medications as potential adjunctive treatments for TNBC is an emerging field of investigation. We also observed that AIPB is not acutely regulated, but stimulated by estrogen agonist kinetically. Our current research suggests anti-helminth Pyrvinium pamoate influences expression of AIPB in TNBC cells leading to the down regulation of estradiol synthesis. Methods: Genetically engineered TNBC MDA-MB-231 cells capable of up-regulating expression of AIPB with an addition of tetracycline in a stable fashion were used in the experiment. We compared activity and expression of the stable cells in the presence and absence of tetracycline induced with 50,000-fold (1.0 ng to 50,000 ng) concentration of Pyrvinium pamoate for 48-hours. Estradiol synthesis was measured using radioimmunoassay and protein expression via immunoblotting by a specific antibody. Results: We observed a decrease in estradiol synthesis with increased expression of AIPB when cells were co-induced with tetracycline (Doxycycline). Cells induced with Pyrvinium pamoate showed an increased expression of AIPB by immunoblotting when the concentration of Pyrvinium pamoate was increased more than 20.0 μg. Conclusions: In the absence of aromatase, AIPB is a critical regulator of estradiol synthesis in triple-negative breast cancer appearing early in tumor development and thus serving as a potential biomarker for early prediction of breast cancer.
Reframing access to innovative oncology medicines: The impact of policy and governance reform in a Bismarck-type system in Bosnia and Herzegovina.
1645 Background: Access to innovative oncology medicines is often assumed to depend primarily on available financial resources. In health systems organized under the Bismarck model, however, access is frequently constrained by structural and policy-related factors, including fragmented governance, rigid reimbursement pathways, and infrequent updating of positive drug lists. These limitations may result in delayed access, restricted indications, and misalignment with contemporary clinical standards, even in the absence of absolute budget shortages. In Bosnia and Herzegovina, such constraints are compounded by complex administrative arrangements, leading to pronounced regional disparities in cancer care. In Sarajevo Canton prolonged waiting times and loss of treatment eligibility underscored that the key barrier was not lack of money, but lack of timely policy action. In October 2024, Sarajevo Canton launched a program shifting financing responsibility locally to improve timely access. Methods: We analyzed drug access before and after program initiation in 641 patients, assessing waiting times from indication to treatment approval and the proportion of patients losing indication prior to therapy availability. Results: Patients whose indication was established before the program were significantly more likely to remain without therapy compared to those after initiation (23.8% vs. 10.3%, p < 0.001). Among 547 patients with available waiting time data, mean waiting time decreased from 251.6 days (SD ≈ 190; range 21–910) to 40.4 days (SD ≈ 21; range 6–199), representing a >6-fold reduction. Variability also decreased, making access more predictable and equitable. Levene’s test indicated unequal variances (F = 517.3, p < 0.001), and Welch’s t-test confirmed the difference (t(157.5) = 13.91, p < 0.001; 95% CI 181–241 days), with a very large effect size (Cohen’s d ≈ 2.0). Conclusions: This policy intervention dramatically reduced waiting times and improved equity in therapy access. This demonstrates that substantial improvements in oncology drug access can be achieved without increasing overall healthcare spending. The observed gains were the result of a deliberate policy and governance shift—redefining responsibility, prioritization, and accountability—rather than additional financial input. This intervention highlights that, within Bismarck-type systems, access gaps are often policy failures rather than funding failures, and that targeted policy change can serve as a powerful lever to rapidly improve equity and quality of cancer care.
Radiotherapy combined with temozolomide and fluoxetine for newly diagnosed grade 4 glioma: Preliminary report from a single-arm, prospective phase II clinical trial.
e14073 Background: Grade 4 gliomas, including glioblastoma (GBM), are associated with poor prognosis and limited treatment options, It is common for patients with glioma to experience depression, and some studies have reported a significant association between depression and adverse survival outcomes in malignant glioma. Fluoxetine, initially developed and approved as a selective serotonin reuptake inhibitor (SSRI) for the treatment of depression, has exhibited anti-tumor effects in glioma cells by inducing lysosomal stress. A retrospective study indicated that concomitant chemoradiotherapy combined with fluoxetine significantly extends the survival of patients with GBM. Our study aims to investigate the efficacy and safety of the combination of fluoxetine and temozolomide with radiotherapy for the treatment of newly diagnosed Grade 4 glioma. Methods: This is a single-arm, single-center, prospective phase II clinical trial for patients with newly diagnosed Grade 4 glioma confirmed by surgical pathology or biopsy, was registered on May 20, 2024 in the Chinese Clinical Trial Registry(ChiCTR2400084456 ). Patients receive radiotherapy (60 Gy/30F) with concurrent daily temozolomide (75 mg/m²/day) and fluoxetine (starting at 20 mg/day, increased to 40 mg after 1 week). Following chemoradiation, patients undergo 6 cycles of adjuvant temozolomide (150–200 mg/m²/day for 5 days every 28 days) and continue fluoxetine (40 mg/day) for 6 months. The primary endpoint is 1-year progression-free survival(PFS) rate, and secondary endpoints include overall survival, safety, quality of life score, and cognitive function. Results: From May 20, 2024, to November 10, 2025, a total of 27 patients were successfully enrolled, including 25 cases of glioblastoma and 2 cases of grade 4 astrocytoma. Among the enrolled patients, there are 19 cases with wild-type IDH1/2, 6 cases that were not tested, and 2 cases with an IDH1 mutation. 13/27(48.1%) exhibited depressive tendencies (SDS score > 50) prior to chemoradiotherapy. As of December 31, 2025, a total of 17 patients who had completed one year of follow-up or experienced disease progression were included. The 1-year PFS rate was 52.9% (9/17). The most common adverse reactions included alopecia (17/17, 100%), nausea (7/17, 41.2%), insomnia (4/17, 23.5%), and seizures (3/17, 17.6%), all of which were grade 1-2. Only 1 case (5.9%) experienced a grade 3 adverse reaction, manifested as elevated alanine aminotransferase levels. Conclusions: These preliminary findings suggest that the combination of radiotherapy, temozolomide, and fluoxetine is feasible, has manageable toxicity, and demonstrates potential PFS benefit in patients with newly diagnosed Grade 4 glioma. We will continue to complete the follow-up of this study cohort, and further investigation in larger cohorts is warranted to confirm our results. Clinical trial information: ChiCTR2400084456.
Evaluation of autologous tumor-infiltrating lymphocytes (GT201) plus toripalimab in recurrent or metastatic head and neck cancer.
6049 Background: Immune checkpoint inhibitors (ICIs) are the standard treatment for recurrent and/or metastatic (R/M) squamous cell carcinoma of the head and neck (HNC) after failure of platinum-based chemotherapy; however, the clinical benefit remains limited, with reported overall response rate (ORR) of approximately 13% and median progression-free survival (PFS) of about 2 months in the second line setting. GT201 is an autologous TIL therapy engineered to express membrane-bound IL-15 (mbIL-15), which may enhance immune activation in the tumor microenvironment and promote durable response. We report preliminary safety and efficacy results from an open-label, single-arm study evaluating GT201 in combination with the PD-1 inhibitor toripalimab in patients with R/M HNC (NCT06190275). Methods: The primary endpoint was safety, including treatment-emergent adverse events (TEAEs) graded per CTCAE v5.0. Secondary endpoints included ORR, disease control rate (DCR), PFS, duration of response (DOR), and overall survival (OS), assessed per RECIST v1.1. Results: As of November 30, 2025, 6 patients with R/M HNC were treated (median age of 57 years; median 1 prior line of therapy). Histology indicated 5 squamous cell carcinoma and 1 lymphoepithelial carcinoma. All patients received 1–2 cycles of bridge therapy, followed by lymphodepletion (low-dose in 5 patients; intermediated-dose in 1 patient), GT201 infusion (5×10 9 -5×10 10 viable cells), and high-dose IL-2 (600,000 IU/Kg; 4-6 doses). Five patients subsequently received toripalimab; one patient progressed prior to PD-1 inhibitor treatment. Maximum follow-up was 15.5 months. Most of AEs were Grade 1-2. Grade ≥ 3 AEs were primarily related to lymphodepletion and IL-2, and included cytopenia, neutropenia, lymphocytopenia, monocytopenia, hypokalemia, rash, and increased bilirubin; all resolved or improved to Grade ≤ 2 within 14 days. The ORR was 66.7% (4/6), including 2 complete response (CR) and 2 partial response(PR); DCR was 83.3% (5/6). One patient with CR remains progress-free exceeding 12 months. Median PFS and OS have not yet been reached. GT201 cells expanded robustly and persisted in peripheral blood for at least 6 months post-infusion. Conclusions: GT201 combined with toripalimab demonstrated a manageable safety profile and encouraging antitumor activity in heavily pretreated R/M HNC, supporting further clinical development of this combination. Clinical trial information: NCT06190275 . Research Sponsor: Grit Biotechnology.
The genetic polymorphism underlying immune checkpoint inhibitor–induced endocrine toxicity.
12024 Background: Immune checkpoint inhibitors (ICIs) have transformed cancer therapy by enhancing T cell–mediated anti-tumor immunity; however, immune-related adverse events (irAEs)—particularly endocrine toxicities (e.g., thyroiditis, hypophysitis, adrenal insufficiency; incidence 5–15%)—pose significant clinical risks, including lifelong hormonal dependency. Although germline genetic factors, especially human leukocyte antigen (HLA) variants, are implicated in irAE susceptibility, prior studies lack granularity in toxicity subtyping and comprehensive genetic characterization across diverse clinical contexts. Methods: We integrated high-resolution HLA genotyping with precise phenotypic annotation of endocrine irAEs (subclassified by organ specificity, onset, severity per CTCAE v5.0) in multi-center retrospective and prospective cohorts. A hypothesis-free genome-wide approach identified novel HLA loci, followed by functional validation of candidate variants. Machine learning models were trained to predict toxicity risk by integrating HLA markers with clinical covariates (ICI type, cancer diagnosis, baseline characteristics). Results: Subtype-specific HLA associations were identified (e.g., distinct HLA-DRB1 alleles linked to thyroiditis versus hypophysitis). Collinearity analysis revealed significant positive correlations between non-endocrine irAE pairs (e.g., capillary hyperpermeability and gastrointestinal ulcers; P < 0.01), suggesting shared immunopathogenic pathways. The integrated risk model demonstrated robust stratification performance for endocrine toxicity susceptibility. Cohort demographics reflected real-world ICI utilization patterns (predominantly lung cancer), while clinical response data underscored challenges in achieving durable remission (notable progression rates). Conclusions: This study delineates the HLA-driven genetic architecture underlying endocrine irAEs and elucidates interdependencies among irAE subtypes. Integration of genetic risk profiling with clinical parameters enables personalized toxicity risk assessment, advocating for proactive, multidisciplinary monitoring protocols. These findings advance precision immunotherapy by informing biomarker-guided patient selection, optimizing safety, and refining clinical decision-making frameworks for ICI administration. Clinical trial information: MR-11-25-020518.
Postoperative re-profiling for identification of missed drivers and a high-risk fusion subgroup in non-pCR lung adenocarcinoma after neoadjuvant immunotherapy: A multicenter retrospective study.
8064 Background: For stage II-III driver-negative non–small cell lung cancer (NSCLC), neoadjuvant chemoimmunotherapy followed by surgery is standard. However, patients without pathological complete response (non-pCR) have suboptimal outcomes and heterogeneous benefit from adjuvant immunotherapy. Baseline driver-negative status may be confounded by false-negative PCR- or DNA-only testing and detecting missed drivers postoperatively could alter adjuvant strategies. This study characterized post-treatment driver alterations and its prognostic value in non-pCR NSCLC after neoadjuvant immunotherapy. Methods: The retrospective multicenter study enrolled 247 stage II-III lung adenocarcinoma (LUAD) patients initially tested as EGFR L858R/19del- and ALK-negative and treated with neoadjuvant immunotherapy (2018-2025). 76 lung squamous cell carcinoma (LUSC) cases were included for exploratory analysis. Postoperative samples underwent 35-gene synchronous DNA/RNA next-generation sequencing (DR-NGS) to identify SNV/indel and fusion events and assess prognostic associations. Results: Of 247 LUAD cases, 179 passed NGS quality control (QC). Driver alterations were identified in 104 (58.1%) patients, including 26 fusions: RETn = 9), MET ex14 skipping (n = 6), ALK(n = 4), ROS1(n = 3), NRG1(n = 2), MET fusion(n = 1), NTRK(n = 1) and 80 SNVs/indels :EGFR (n = 30), KRAS G12C/D (n = 16), HER2/3 (n = 16), BRAF (n = 2), KRAS non-G12C/D (n = 16). Notably, 10 classic EGFR mutations (19del/L858R) 、9 rare EGFR mutations and 4 ALK fusions were newly identified, indicating baseline omissions. Driver-positive tumors had significantly higher residual tumor burden(median: 51.8% vs. 35.8%, p = 0.003) and a higher proportion of females (43.3% vs 20.0%, p = 0.001). Median recurrence-free survival (mRFS) differed significantly among fusion-positive, mutation-positive, and driver-negative groups (20.9 vs 41.5 vs 60.3 months; p = 0.024). In patients receiving adjuvant immunotherapy, mRFS was 34.6 months in driver-positive and not reached in driver-negative patients. Driver detection was rare ( < 3%) in QC-failed samples (n = 68), which exhibited a higher major pathological response than QC-passed samples (58.1% vs 14.5%), indicating tumor cellularity as critical for detection. Exploratory analysis in 55 QC-passed LUSC samples showed a low driver detection rate (7.3%), suggesting limited utility in this cohort. Conclusions: DR-NGS reveals a high prevalence of drivers in non-pCR LUAD after neoadjuvant immunotherapy, with positivity associated with higher residual tumor burden. Fusion-positive patients have the poorest prognosis, identifying a subgroup that may benefit from tailored adjuvant strategies. These findings support routine postoperative molecular re-profiling in non-pCR patients to guide individualized treatment.
Real-world survival outcomes of radiotherapy-based treatment strategies for esophageal cancer: A population-based analysis from the Hong Kong CADRS.
e16122 Background: Radiotherapy is an essential component of treatment for esophageal cancer across definitive, perioperative, and salvage settings. However, survival outcomes associated with different radiotherapy-based strategies in real-world clinical practice remain incompletely understood. We conducted a population-based analysis to evaluate survival outcomes and prognostic factors among patients with esophageal cancer treated with radiotherapy in Hong Kong. Methods: Using the Hong Kong Clinical Data Analysis and Reporting System (CADRS), we identified patients with esophageal cancer who received radiotherapy between 2007 and 2022. Patients were categorized into three groups according to treatment strategy: definitive radiotherapy with or without chemotherapy, perioperative radiotherapy with or without chemotherapy combined with surgery, and salvage radiotherapy with or without additional surgery or chemotherapy. Overall survival (OS) was estimated using the Kaplan–Meier method and compared using the log-rank test. Prognostic factors were evaluated using univariate and multivariate Cox proportional hazards models. Results: A total of 1,926 patients were included, of whom 1,028 (53.4%) received definitive radiotherapy, 446 (23.1%) perioperative radiotherapy, and 452 (23.5%) salvage radiotherapy. The median age was 64 years, and 83.9% of patients were male. Median OS differed significantly among treatment strategies ( P < 0.001), with the longest survival observed in the perioperative radiotherapy group (20.0 months, 95% CI 16.3–24.1), followed by the definitive radiotherapy group (7.1 months, 95% CI 6.4–7.8), and the salvage radiotherapy group (3.0 months, 95% CI 2.5–3.6). In multivariate analysis, perioperative radiotherapy was independently associated with improved OS compared with definitive radiotherapy (HR 0.585, 95% CI 0.514–0.665, P < 0.001), whereas salvage radiotherapy was associated with worse OS (HR 1.568, 95% CI 1.390–1.769, P < 0.001). Concurrent chemotherapy remained independently associated with improved survival (HR 0.656, 95% CI 0.584–0.736, P < 0.001). Conclusions: In this large population-based real-world cohort, survival outcomes among patients with esophageal cancer treated with radiotherapy varied markedly according to treatment strategy. Perioperative radiotherapy was associated with the most favorable survival, while outcomes following salvage radiotherapy remained poor. These findings highlight the importance of treatment intent when interpreting real-world outcomes of radiotherapy in esophageal cancer.
Risk stratification and age-related disparities with abemaciclib responses in early ER-positive/HER2-negative breast cancer.
e12527 Background: Adjuvant abemaciclib improves outcomes in high-risk hormone receptor–positive, HER2-negative early breast cancer. However, the prognostic relevance of individual monarchE eligibility criteria and age-related differences in treatment decisions and outcomes remain poorly understood in real-world settings. Methods: We analyzed three complementary datasets: the SEER registry, the METABRIC genomic cohort, and a Japanese multi-institutional real-world cohort. Patients were classified using a four-tier system adapted from the monarchE trial (NCT03155997): Group 1, lymph node–negative; Group 2, 1–3 positive nodes with low grade and tumor size < 5 cm; Group 3, 1–3 positive nodes with high grade or tumor size ≥5 cm; and Group 4, ≥4 positive nodes. Molecular characteristics were evaluated in the METABRIC cohort using gene set variation analysis (GSVA). Patients were stratified by age into adolescent and young adult (AYA, 15–39 years), perimenopausal (40–54), menopausal (55–64), and older (≥65) groups. Disease-specific survival (DSS) was analyzed using Kaplan–Meier methods and Cox proportional hazards models, including interaction analyses between age and risk group. Results: The monarchE-based risk classification consistently stratified DSS in both SEER and METABRIC cohorts. Group3 patients exhibited significantly worse DSS compared with Group2 patients, underscoring than even within N1 disease, biological risk features strongly affect prognosis. In the METABRIC analysis, transcriptomic differences between groups reflected variations in proliferation-related and cell-cycle pathways, consistent with the pathological criteria of the monarchE trial. In the Japanese real-world cohort, abemaciclib initiation increased with nodal burden, 40% in Group 3 and 72% in Group 4, suggesting that treatment decisions were largely driven by nodal count rather than histologic or molecular features. Elderly patients were markedly less likely to receive abemaciclib, despite showing comparable or slightly inferior DSS relative to younger patients. Importantly, interaction analyses revealed no significant age-by-biology synergy, indicating that tumor aggressiveness is largely independent of age. The slightly inferior DSS observed in older patients appears not to be attributable to intrinsic tumor biology. AYA patients showed poorer outcomes compared to other age groups in Group 4. Conclusions: Our findings validate the prognostic relevance of the monarchE-based risk classification, demonstrating that high-risk N1 patients represent a clinically distinct subgroup. In addition, older patients were less likely to receive intensive therapy despite comparable disease aggressiveness, which may partly explain their inferior DSS. These results highlight the need for risk- and age-adapted treatment strategies in early-stage breast cancer.
Phase 2 data from ROSETTA Lung-02, a global randomized phase 2/3 trial of pumitamig (PD-L1 × VEGF-A bsAb) + chemotherapy in 1L NSCLC.
8513 Background: Pumitamig (BNT327/BMS986545) is an investigational PD-L1 × VEGF-A bsAb designed to restore effector T-cell function by binding PD-L1 and localizing VEGF-A neutralization within the tumor microenvironment. We present the prespecified interim analysis of the Phase 2 dose-optimization part of the global Phase 2/3 ROSETTA Lung-02 trial (NCT06712316) evaluating the recommended Phase 3 dose of pumitamig + chemotherapy in 1L NSCLC. Methods: The Phase 2 part of ROSETTA Lung-02 enrolled pts with treatment-naïve advanced NSCLC, no actionable genomic alterations, regardless of PD-L1 status, ECOG PS ≤1, and ≥1 measurable lesion per RECIST v1.1 into two substudies based on histology (nonsquamous [NSQ] and squamous [SQ]). Pts were randomized 1:1 to 1400 mg (dose level [DL] 1) or 2000 mg (DL2) pumitamig + histology-specific chemotherapy Q3W (NSQ: carboplatin + pemetrexed; SQ: carboplatin + paclitaxel). Primary endpoints were safety, overall response rate (ORR), and best percentage change in tumor size from baseline. Key secondary endpoints include duration of response (DOR) and disease control rate (DCR). Results: The Phase 2 part enrolled 44 pts (NSQ n=23; SQ n=21; data cutoff Nov 21, 2025). Median age was 66 y (range: 41–87), and 27 (61.4%) pts had ECOG PS 1. Among 40 response-evaluable pts, best overall response was CR in 2 pts, PR in 26, and SD in 12, for an ORR of 70.0% (28/40; confirmed ORR was 52.5% [21/40], 5 pending confirmation) and DCR of 100%. DOR data were not mature at the time of data cutoff. Median best change in tumor volume was -38.2% (NSQ -36.6%; SQ -39.7%). In NSQ NSCLC, ORR was 66.7% (14/21) and, by pumitamig dose, it was 72.7% for DL1 (8/11) and 60.0% for DL2 (6/10). In SQ NSCLC, ORR was 73.7% (14/19), 81.8% for DL1 (9/11) and 62.5% for DL2 (5/8). Central PD-L1 levels were available for 35 pts (PD-L1 <1% [n=20], 1–49% [n=9], ≥50% [n=6]), with activity across PD-L1 levels. Circulating tumor DNA dynamics were assessed. Median treatment duration was 4.5 mo (range: 0.1–8.8) with 30 (69.8%) pts still on treatment. In the safety set (N=43), 40 (93.0%) pts had a treatment-related adverse event (TRAE). Grade ≥3 TRAEs were reported in 19 (44.2%) pts and were considered pumitamig-related in 8 (18.6%). Pumitamig-related TRAEs led to treatment discontinuation in 2 (4.7%) pts. Immune-related AEs (irAEs) occurred in 6 (14.0%) pts and grade ≥3 irAEs in 1 (2.3%). Bleeding events were reported in 7 (16.3%) pts, with only 1 event being grade 3. Conclusions: In these first global data for a PD-(L)1 × VEGF-A bsAb in 1L NSCLC regardless of PD-L1 status, pumitamig + chemotherapy showed encouraging efficacy with a manageable safety profile. Efficacy of the lower dose + chemotherapy was particularly encouraging (ORR 72.7% in NSQ, 81.8% in SQ) and is being evaluated vs pembrolizumab + chemotherapy in the ongoing Phase 3 part of the trial. Clinical trial information: NCT06712316 .
Cancer recognition and assessment through non-invasive evaluation (CRANE): A prospective clinical validation study of a multi-biomarker class multi-cancer early detection (MCED) test in Japan.
TPS10629 Background: Japan has established screening programs for gastric, colorectal, lung, cervical, and breast cancers. Blood-based multi-cancer early detection (MCED) tests utilize the measurement of shared tumor-derived biomarkers (ctDNA, proteins) and machine learning to simultaneously detect several cancer types. By complementing current cancer screening, MCED testing has the potential to detect a wide range of cancers at earlier stages and expand existing screening efforts. CRANE is a prospectively collected, case-control study of Exact Sciences’ MCED test (Cancerguard) to assess the overall sensitivity and specificity of the MCED test in a Japanese population. The secondary objective is to evaluate test performance across cancer types and stages. Methods: The study will include Japanese participants aged 45-84 years. Participants must provide written informed consent prior to providing a peripheral blood sample. A cancer cohort of approximately 1,000 participants with newly diagnosed confirmed cancer will be included in the study. Cancer cases will be recruited to obtain a diverse representation of cancer types and stages while also considering prevalence rates by cancer type. Emphasis will be placed on the 16 cancers with the highest incidence in Japan: lung, gastric, prostate, colorectal, breast, liver, pancreatic, bladder/renal pelvis, non-Hodgkin lymphoma, esophageal, kidney, uterine, cervical, ovarian, biliary tract, and head and neck. Approximately 1,000 non-cancer control participants will be included in the study, corresponding to the age and gender distribution from current population estimates reported by the Statistics Bureau of Japan. These participants will have no known cancer at enrollment, with status confirmed based on up-to-date routine cancer screening consistent with Japanese national cancer screening programs and subsequently through a 12-month follow-up after enrollment. Participants are enrolled at clinical sites in Japan, including cancer screening centers affiliated with the Japan Cancer Society. Whole blood will be collected at study sites in LBgard tubes, shipped to a local laboratory, processed to isolate the plasma and buffy coat, and then frozen. The frozen plasma and buffy coat samples will be shipped to the Exact Sciences clinical laboratory (Madison, WI, USA) for analysis. Approximately 300 non-cancer control samples and ~50 cancer case samples will undergo interim testing to assess test logistics and preliminary performance. No protocol modifications or enrollment pauses will occur based on interim testing results. Final testing will occur once all participants have been enrolled. The study duration is expected to be approximately 24 months. This study is funded by Exact Sciences Corporation. Clinical trial information: UMIN000059407.
Arrhythmic outcomes following immune checkpoint inhibitor (ICI) therapy in patients with non–small cell lung cancer (NSCLC) without heart failure or myocarditis.
e24006 Background: Despite the established prognostic benefit, immune checkpoint inhibitors (ICIs) are associated with immune-related adverse events (irAEs), including cardiovascular complications such as new-onset heart disease and myocarditis. However, data on the occurrence of atrial and ventricular arrhythmias following ICI therapy remain conflicting. We therefore leveraged a large real-world database to evaluate the association between ICI therapy and incident arrhythmias in patients with non–small cell lung cancer (NSCLC). Methods: We conducted a retrospective cohort study using data from the U.S. Collaborative Network, spanning January 2020 through January 2025. Adult patients with NSCLC treated with ICI therapy were included. Patients with a prior history of cardiac arrest, implantable cardioverter-defibrillator implantation, myocarditis, or any atrial or ventricular arrhythmia before the index date were excluded. Additionally, patients who developed myocarditis during follow-up were excluded given its known association with arrhythmic risk. The index date was defined as the date of first ICI administration, and patients were followed for up to 36 months. The primary outcome was the incidence of overall arrhythmia, defined as new-onset atrial fibrillation (AF), atrial flutter, supraventricular tachycardia (SVT), or ventricular tachycardia (VT) after ICI initiation. Secondary outcomes included AF requiring cardioversion or antiarrhythmic therapy, AF or atrial flutter, and VT. Baseline characteristics, including demographics, comorbidities, laboratory values, and medication use, were balanced using propensity score matching to improve comparability between groups. Results: After propensity score matching, 72,498 patients were included in each group, with no significant differences in baseline characteristics or medication use. The mean age was 77 years; 55% were male, 68% were White, and 9.8% were Black. Over a 36-month follow-up period, patients receiving ICI therapy had a higher incidence of overall arrhythmia (HR 1.07, 95% CI 1.05–1.10), AF requiring cardioversion or antiarrhythmic therapy (HR 1.15, 95% CI 1.10–1.20), AF or atrial flutter (HR 1.04, 95% CI 1.01–1.07), and VT (HR 1.12, 95% CI 1.03–1.23). Conclusions: Among patients with NSCLC without prior heart failure, myocarditis, or arrhythmia, ICI therapy was associated with a significantly increased risk of atrial and ventricular arrhythmias. These findings suggest that ICI therapy in NSCLC is associated with clinically meaningful arrhythmic risk even in the absence of myocarditis or heart failure, supporting proactive cardiovascular surveillance during treatment.
Efficacy of neoadjuvant immunochemotherapy followed by surgery in locally advanced buccal squamous cell carcinoma: A single-center real-world study.
6081 Background: The 5-year overall survival (OS) rate for locally advanced (stage III–IV) buccal squamous cell carcinoma (LA BSCC) is under 50%. In stage IV disease, the 2-year OS rate is under 45%, and the 5-year OS rate is under 30% (29.3% for the anterior buccal and only 18.7% for the posterior buccal). Evidence is limited regarding whether betel nut chewing and tumor subsites influence the efficacy of neoadjuvant immunochemotherapy. Methods: Retrospective analysis of 115 patients with LA BSCC received neoadjuvant therapy at Hunan Cancer Hospital between August 2021 and October 2024. Treatment regimen: Neoadjuvant albumin-bound paclitaxel plus cisplatin or carboplatin combined with a PD-1 inhibitor, followed by definitive surgery. Postoperative adjuvant radiotherapy or chemoradiotherapy was administered based on pathological findings. Primary endpoint: Pathological complete response and major pathological response (pCR/MPR) rate; Secondary endpoints: Overall survival (OS) and disease free survival (DFS) at 1 and 2 years; Subgroup analyses: Outcomes were evaluated by betel nut chewing history, tumor subsite, and TNM stage. Results: Of 113 evaluable patients, 16% had stage III and 84% had stage IV disease. A history of betel nut chewing was present in 71.9%. The objective response rate to neoadjuvant therapy was 83.5%, including a 28.3% clinical complete response rate. 98 patients proceeded to surgery; the pCR/MPR rate was 27.6%, and the partial pathological response (pPR) rate was 52.0%. With a median follow-up of 29 months (15-53 months), 1-year OS and DFS rates were 80.5% and 74.3%, respectively; 2-year OS and DFS rates were 70.0% and 67.5%, respectively. 2-year OS did not differ significantly by betel nut chewing ( P =0.8578) or by tumor subsite (anterior vs posterior buccal; P =0.5568). More advanced TNM stage was associated with poorer survival ( P =0.0352). Patients achieving pCR or MPR had significantly better 2-year OS and DFS than those who did not ( P =0.0386 and P =0.0102, respectively). Grade≥3 adverse events occurred in 18% of patients and no treatment-related deaths were observed. Conclusions: Neoadjuvant immunochemotherapy increased the 2-year OS rate of LA BSCC to 70%, with the 27.6% pCR/MPR rate. Treatment efficacy did not differ by betel nut chewing or by tumor subsite, whereas TNM stage remained an independent prognostic factor. Deep pathological response (pCR/MPR) was associated with superior long-term outcomes. Number 2-yr OS rate(%) p value 2-yr DFS rate(%) p value Tumor stage III 18/113 88.89 0.1558 78.75 0.0352 IVa 64/113 71.12 66.54 IVb 31/113 54.17 39.13 Tumor subsite Anterior buccal 88/113 69.85 0.5568 60.30 0.9963 Posterior buccal 25/113 64.29 60.00 Pathological response rate pCR+MPR 27/98 78.95 0.0386 78.95 0.0102 pPR 51/98 83.43 72.66 SD+PD 21/98 54.55 47.37 Betel nut use + 82/113 69.22 0.8578 60.02 0.9238 - 31/113 67.88 60.87
Standard-dose ripretinib plus sunitinib versus ripretinib dose escalation in advanced GIST after progression on four prior therapies: Preliminary results of a multicenter cohort study.
11534 Background: Patients with advanced GIST who progress after four lines of therapy, including imatinib, sunitinib, regorafenib, and ripretinib, face limited treatment options. Ripretinib dose escalation is a potential option recommended by NCCN and CSCO guidelines, while combination therapy with TKIs targeting complementary pathways may offer an alternative for patients harboring multiple resistance mutations. This study aims to evaluate the efficacy of ripretinib dose escalation versus combination therapy following progression on standard-dose ripretinib to identify more effective treatment strategies. Methods: This study is a multicenter cohort study designed to evaluate the efficacy and safety of ripretinib dose escalation versus combination therapy with sunitinib in patients with advanced GIST who have progressed after four prior lines of therapy. Eligible patients receive ripretinib dose escalation group (150 mg twice daily) or the combination therapy group (ripretinib 150 mg once daily plus sunitinib 25 mg once daily). The primary endpoint was progression-free survival (PFS). Results: Between September 2023 and December 2025, 30 patients were enrolled (16 in the ripretinib dose escalation group and 14 in the combination group). Primary KIT11 mutation were identified in 17 patients, 62.5%(10/16) in the dose escalation group and 50% (7/14) in the combination group, while KIT 9 mutations were 2 patients both in the combination group . As of December 30, 2025, 5 patients in the ripretinib dose escalation group remained on treatment with stable disease, whereas all patients in the combination group had experienced disease progression; 3 of these showed slow progression and continued therapy. mPFS1 with standard-dose ripretinib was comparable between groups (6.0 months [95% CI: 4.7–7.3] vs. 7.0 months [95% CI: 5.2–8.8]). Following progression, ripretinib dose escalation achieved a mPFS2 of 8.3 months (95% CI: 4.7–12.0), significantly longer than combination therapy (3.1 months [95% CI: 2.0–4.3]) . OS data remain immature. Ripretinib dose escalation demonstrated better tolerability, with no grade ≥3 adverse events reported. In contrast, the combination group experienced higher toxicity, 35.7% discontinued treatment due to severe adverse reactions. Conclusions: Preliminary results indicate that ripretinib dose escalation demonstrated superior efficacy and a more favorable safety profile compared with combination therapy involving sunitinib. These findings support ripretinib dose escalation as the preferred treatment option for patients with advanced GIST who have progressed after four lines of standard therapy. However, the sample size in this study is still small, and further research with a larger sample is needed to validate this conclusion.
More than pain relief: Associations between provider compassion, anxiety, and burnout in ED oncology care.
12078 Background: Compassionate care is a cornerstone of high-quality oncology practice, especially in high-stress environments like the emergency department (ED). However, the relationship between perceived provider compassion and patient outcomes such as pain, psychological distress, and provider wellbeing is not well understood. Methods: We conducted a cross-sectional survey of oncology patients presenting to the ED. Participants completed validated assessments including the Brief Pain Inventory, the Schwartz Center Compassionate Care Scale (SCCCS), the Pain Catastrophizing Scale, the Pain Anxiety Symptom Scale, and the Self-Compassion Scale. Separately, ED providers were surveyed to assess their own burnout, rated on a 0–8 scale. Patients were grouped into High Compassion (SCCCS >100) and Low Compassion (≤100) categories based on their ratings of provider compassion. Group comparisons were performed using t-tests, and linear regressions explored associations between compassion scores and patient-reported and provider-reported outcomes. Results: Of 57 patients, 22 (39%) reported high provider compassion. Median final pain was lower in the High Compassion group (2.0 vs 4.0), though not statistically significant (p=0.18). Patients in the High Compassion group had significantly lower pain anxiety scores (29.2 vs 52.5, p=0.014) and were seen by providers with significantly lower burnout scores (3.1 vs 5.6, p=0.022). A trend toward lower pain catastrophizing was also observed (15.6 vs 25.7, p=0.069). No significant differences were found in pain reduction or self-compassion. Conclusions: In a cohort of ED oncology patients, higher perceived provider compassion was significantly associated with lower pain anxiety and with lower provider-reported burnout, but not with pain scores. These findings suggest that compassion may be more strongly linked to patients’ emotional experiences and the wellbeing of the providers delivering care. Integrating compassionate care practices into acute oncology settings may support both patient psychological health and provider sustainability. Key outcomes by compassion group (>100 vs. ≤100). Outcome High Compassion (Mean) Low Compassion (Mean) p-value Final Pain (0–10) 2.33 3.67 0.183 Pain Anxiety Total 29.2 52.5 0.014 Catastrophizing Total 15.6 25.7 0.069 Self-Compassion Total 23.7 23.7 0.996 Provider Burnout (0–8) 3.1 5.6 0.022
Rural-urban disparities in musculoskeletal functional limitation among cancer patients in the United States.
e13799 Background: As cancer survivorship increases, long-term physical function has emerged as a critical determinant of quality of life and independence. Musculoskeletal impairment related to cancer treatment, metastatic disease, and deconditioning may be exacerbated by disparities in access to rehabilitative and orthopaedic care. Rural populations face well-documented barriers to specialty care, yet rural-urban differences in physical function among cancer patients remain poorly characterized at a national level. Thus, we evaluated the association between rural residence and functional limitation among U.S. adults with cancer history. Methods: We analyzed pooled 2018–2024 data from the National Health Interview Survey, restricting the sample to adults with history of cancer. Rural populations were defined as living in a nonmetropolitan county. Functional limitation was defined as difficulty walking or climbing steps (“some difficulty”, “a lot of difficulty”, or “cannot do at all”). Covariates included age, sex, ethnicity, health insurance, private health insurance, and medical history (arthritis, asthma, coronary heart disease, prior myocardial infarction, chronic obstructive pulmonary disease). Sample weights were applied to account for complex survey design. A multivariable logistic regression was estimated. Results: 26,117 cancer patients were included (unweighted), 17.2% (95% CI: 16.7-17.8) of which resided in rural counties. Average age was 67.0 years old (66.5-67.6). 44.3% (43.6-45.0) of patients were male. 36.6% (35.9-37.3) of patients reported functional limitation overall; 42.4% (40.7-44.1) of rural patients reported limitation versus 35.8% (35.0-36.7) of urban patients. In multivariable logistic regression, cancer patients living in a rural county had 1.20 (1.10-1.31) times increased odds of reporting functional limitation (p < 0.001). Conclusions: Rural residence is independently associated with worse physical function among U.S. cancer patients, even after adjustment for demographic, socioeconomic, and medical factors. These findings suggest that geographic disparities in access to rehabilitative, orthopaedic, and survivorship care may contribute to long-term functional impairment. Targeted interventions to improve access to musculoskeletal and rehabilitation services in rural communities may reduce disability and improve survivorship outcomes among cancer patients.
Ischemic heart disease as a competing cause of death among patients with lung cancer in the United States: A multiple cause-of-death analysis, 1999–2023.
11163 Background: Despite improvement in lung cancer prevention, early diagnosis & management, Ischemic heart disease (IHD) has become a major concurrent cause of mortality in lung cancer. Population level evidence on long term deaths due to IHD & lung cancer especially across sex & racial groups are lacking. Identification of these trends is necessary for cardio-oncology risk stratification & promote equitable survivorship care. Methods: We used the CDC WONDER Online Database to obtain mortality data for IHD & Lung Cancer from 1999-2023 from patients aged 45-85 & above. We used current final multiple cause of death data from 1999-2020 & 2018-2023. We used ICD-10 codes I20-I25 for IHD & C33-34 for Lung cancer. Data were stratified by Race (Non-Hispanic [NH] American Indian or Alaska Native, NH Asian or Pacific Islander, NH Black or African American, NH White, Hispanic or Latino), sex (male & female) and age groups (45-85+). We calculated age-adjusted mortality rates (AAMR) per 100,000 population based on the 2000 US standard population. Temporal mortality trends were analyzed using Joinpoint regression. (P < 0.05). Results: From 1999-2023, AAMR from lung cancer & IHD decreased in most demographic groups. Males had higher AAMR than females, with mortality decreasing from 20.3 to 10.6 per 100,000, compared with a decline from 6.8 to 4.6 per 100,000 among females. Joinpoint regression identified substantial reduction among males from 1999-2010 (annual percent change [APC] −2.35%, p < 0.05) & 2010-2016 (APC −5.45%, p < 0.05), followed by stabilization. Among females, declines were observed from 2005-2012 (APC −2.76%, p < 0.05) & 2012-2015 (APC −7.35%, p < 0.05), with a nonsignificant plateau thereafter. NH Black exhibited higher AAMRs despite significant declines from 2005-2018 (APC −4.11%, p < 0.05). NH White showed persistent declines from 1999-2017, followed by a marginal increase during 2017-2021. Hispanic & Asian or Pacific Islander individuals maintained the lowest AAMRs, with decreasing trends & brief late-2010s increases in select subgroups, while American Indian or Alaska Native individuals exhibited greater year-to-year variability without statistically significant joinpoints. Conclusions: During 1999-2023, mortality from IHD & lung cancer decreased significantly in the US. Males & NH Black experienced a major disease burden. Recent improvement in death trends among the select patient population showed appropriate cardiovascular risk management in survivors with lung cancer. This analysis highlighted the importance of cardio-oncology care models & precision-guided, equity-driven interventions to decrease cardiovascular mortality among patients with lung cancer.
Maintenance therapy after induction chemotherapy with anti-EGFR–antibody and survival in metastatic colorectal cancer: A retrospective real-world study.
e15613 Background: Maintenance therapy (MT) after induction chemotherapy in metastatic colorectal cancer remains clinically uncertain, as its benefit has been demonstrated primarily for progression-free survival (PFS), whereas a consistent impact on overall survival (OS) has not been unequivocally established. Methods: We retrospectively analyzed patients with RAS-wild-type metastatic colorectal cancer treated at Moscow City Oncology Hospital No. 62 between 2009 and 2024 with anti-EGFR therapy. OS and PFS were evaluated according to receipt of MT. Survival outcomes were estimated using the Kaplan–Meier method and compared with the log-rank test. Multivariable Cox models were used with adjustment for sex, age, metastatic burden, anti-EGFR treatment category, ECOG performance status at first line, and year of treatment initiation. Results: We identified 781 patients who received anti-EGFR therapy; 355 (46%) were treated in the first line, 380 (50%) in the second or third line, and 30 (3.9%) in the fourth or later lines. Among patients receiving MT, regimens included fluoropyrimidine monotherapy (26.8%), anti-EGFR monotherapy (10.7%), fluoropyrimidines plus anti-EGFR (31.8%), fluoropyrimidines plus anti-VEGF (25.1%), and FOLFIRI plus bevacizumab (5.7%). Compared with observation, MT was independently associated with improved median OS (36.1 vs 30.5 months; unadjusted HR = 0.82, 95% CI 0.68–0.99, p = 0.042; adjusted HR = 0.69, 95% CI 0.56–0.85, p < 0.001). MT was also associated with improved PFS in line-specific analyses. In the first line, median PFS was 13.8 months (n = 295) versus 9.6 months (n = 441) (crude HR = 0.62, 95% CI 0.53–0.73; p < 0.001; adjusted HR = 0.58, 95% CI 0.49–0.69; p < 0.001). In the second line, median PFS was 13.1 months (n = 144) versus 6.1 months (n = 452) (crude HR = 0.53, 95% CI 0.43–0.65; p < 0.001; adjusted HR = 0.52, 95% CI 0.42–0.65; p < 0.001). In the third line, median PFS was 11.1 months (n = 68) versus 4.5 months (n = 323) (crude HR = 0.47, 95% CI 0.35–0.63; p < 0.001; adjusted HR = 0.52, 95% CI 0.38–0.70; p < 0.001). Conclusions: In this large real-world cohort of patients with RAS wild-type mCRC treated with anti-EGFR–based therapy, MT was independently associated with clinically meaningful improvements in OS and PFS across multiple lines of treatment. These findings should be interpreted with caution, as MT was preferentially administered to patients with objective response and preserved performance status, potentially introducing selection bias and confounding the observed OS benefit. As a non-randomized retrospective analysis, the study remains vulnerable to residual confounding from unmeasured or incompletely captured factors.
Sequencing antibody-based and CAR-T therapies in relapsed/refractory follicular lymphoma: An anchored indirect comparison and parallel meta-analysis.
e19053 Background: Therapeutic options for relapsed/refractory (R/R) follicular lymphoma (FL) now include dual-targeted CD19/CD20 antibody regimens, CD20×CD3 bispecific antibodies, and chimeric antigen receptor (CAR) T-cell therapies. However, the relative efficacy of antibody-based platforms and their positioning relative to CAR-T remain unclear due to the absence of head-to-head trials. We performed an anchored indirect comparison to evaluate progression-free survival (PFS) between antibody-based regimens and conducted a parallel meta-analysis to contextualize CAR-T outcomes. Methods: A systematic review identified phase III randomized trials in R/R FL sharing a lenalidomide–rituximab (R²) backbone. An anchored Bucher indirect comparison was performed between tafasitamab plus R² and epcoritamab plus R² using reported PFS hazard ratios (HRs). CAR-T therapies, which lack randomized comparators, were analyzed separately using a random-effects meta-analysis of proportions to estimate pooled overall response rate (ORR) and complete response (CR) rates. Safety outcomes were compared descriptively across therapeutic classes. Results: Compared with R², tafasitamab plus R² improved PFS (HR 0.43, 95% CI 0.32–0.58), as did epcoritamab plus R² (HR 0.21, 95% CI 0.14–0.31). In the anchored indirect comparison, epcoritamab plus R² demonstrated superior PFS relative to tafasitamab plus R² (HR 0.49, 95% CI 0.30–0.80; p=0.0047). In a parallel meta-analysis of CAR-T therapies (ZUMA-5 and ELARA; n=178), pooled ORR was 90.2% (95% CI 83.0–94.6) with moderate heterogeneity (I²=65%), and pooled CR rate was 73.6% (95% CI 66.5–79.7). CAR-T therapies achieved the highest depth of response but were associated with higher rates of acute toxicity, including cytokine release syndrome and neurotoxicity. Bispecific antibody therapy demonstrated strong PFS benefit with outpatient administration but higher rates of serious infections, while dual-targeted antibody therapy showed intermediate efficacy with a potentially favorable tolerability profile. Conclusions: In an anchored indirect comparison, CD20×CD3 bispecific antibody therapy was associated with significantly improved PFS compared with dual-targeted CD19/CD20 antibody therapy in R/R FL. CAR-T therapies achieved the highest response rates in parallel analyses but with increased acute toxicity. These findings support a risk-adapted, sequential treatment approach in R/R follicular lymphoma.