Citations

ABSTRACT Objective Despite high stand‐alone performance, studies demonstrate that artificial intelligence (AI)‐supported endoscopic diagnostics often fall short in clinical applications due to human‐AI interaction factors. This video‐based trial on Barrett's esophagus aimed to investigate how examiner behavior, their levels of confidence, and system usability influence the diagnostic outcomes of AI‐assisted endoscopy. Methods The present analysis employed data from a multicenter randomized controlled tandem video trial involving 22 endoscopists with varying degrees of expertise. Participants were tasked with evaluating a set of 96 endoscopic videos of Barrett's esophagus in two distinct rounds, with and without AI assistance. Diagnostic confidence levels were recorded, and decision changes were categorized according to the AI prediction. Additional surveys assessed user experience and system usability ratings. Results AI assistance significantly increased examiner confidence levels ( p < 0.001) and accuracy. Withdrawing AI assistance decreased confidence ( p < 0.001), but not accuracy. Experts consistently reported higher confidence than non‐experts ( p < 0.001), regardless of performance. Despite improved confidence, correct AI guidance was disregarded in 16% of all cases, and 9% of initially correct diagnoses were changed to incorrect ones. Overreliance on AI, algorithm aversion, and uncertainty in AI predictions were identified as key factors influencing outcomes. The System Usability Scale questionnaire scores indicated good to excellent usability, with non‐experts scoring 73.5 and experts 85.6. Conclusions Our findings highlight the pivotal function of examiner behavior in AI‐assisted endoscopy. To fully realize the benefits of AI, implementing explainable AI, improving user interfaces, and providing targeted training are essential. Addressing these factors could enhance diagnostic accuracy and confidence in clinical practice.

AbstractThe Enhanced EHR-facilitated Cancer Symptom Control (E2C2) Trial is a pragmatic trial testing a collaborative care approach for managing common cancer symptoms. There were challenges in identifying cancer site and metastatic status.This study compares three different approaches to determine cancer site and six strategies for identifying the presence of metastasis using EHR and cancer registry data.The E2C2 cohort included 50,559 patients seen in the medical oncology clinics of a large health system. SPPADE symptoms were assessed with 0 to 10 numeric rating scales (NRS). A multistep process was used to develop three approaches for representing cancer site: the single most prevalent International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) code, the two most prevalent codes, and any diagnostic code. Six approaches for identifying metastatic disease were compared: ICD-10 codes, natural language processing (NLP), cancer registry, medications typically prescribed for incurable disease, treatment plan, and evaluation for phase 1 trials.The approach counting the two most prevalent ICD-10 cancer site diagnoses per patient detected a median of 92% of the cases identified by counting all cancer site diagnoses, whereas the approach counting only the single most prevalent cancer site diagnosis identified a median of 65%. However, agreement among the three approaches was very good (kappa > 0.80) for most cancer sites. ICD and NLP methods could be applied to the entire cohort and had the highest agreement (kappa = 0.53) for identifying metastasis. Cancer registry data was available for less than half of the patients.Identification of cancer site and metastatic disease using EHR data was feasible in this large and diverse cohort of patients with common cancer symptoms. The methods were pragmatic and may be acceptable for covariates, but likely require refinement for key dependent and independent variables.

Despite significant improvements in survival of patients with multiple myeloma (MM), outcomes remain heterogeneous, and a significant proportion of patients experience suboptimal outcomes. Importantly, traditional prognostic factors based on data from patients treated with older therapies no longer capture prognosis accurately in the contemporary era of novel triplet or quadruplet therapies. Therefore, risk stratification requires refinement in the context of available and investigational treatment options in routine practice and clinical trials, respectively. The current identification of high-risk MM (HRMM) in routine practice is based on the Revised International Staging System, which stratifies patients using a combination of widely available serum biomarkers and chromosomal abnormalities assessed via fluorescence in situ hybridization. In recent years, a substantial body of evidence concerning additional clinical, biological, and molecular/genomic prognostic factors has accumulated, along with new MM risk stratification tools and consensus reports. The International Myeloma Society, along with the International Myeloma Working Group, convened an Expert Panel with the primary aim of revisiting the definition of HRMM and formulating a practical and data-driven consensus definition, based on new evidence from molecular/genomic assays, updated clinical data, and contemporary risk stratification concepts. The Panel proposes the following Consensus Genomic Staging (CGS) of HRMM which relies upon the presence of at least one of these abnormalities: (1) del(17p), with a cutoff of >20% clonal fraction, and/or TP53 mutation; (2) an IgH translocation including t(4;14), t(14;16), or t(14;20) along with 1q+ and/or del(1p32); (3) monoallelic del(1p32) along with 1q+ or biallelic del(1p32); or (4) β2 microglobulin ≥5.5 mg/L with normal creatinine (<1.2 mg/dL).

ABSTRACT An increasing number of cancer patients are benefiting from long term cancer-directed therapy with an ever-expanding arsenal of novel agents from monoclonal antibodies to small molecules and cellular therapies in addition to the mainstay cytotoxic chemotherapy. All these therapies are accompanied by a unique array of adverse events, which include kidney toxicity. In this context, the need for onco-nephrology expertise continues to grow. Oncologists and hematologists collaborate closely with onco-nephrologists to determine: (i) treatment options for special populations based on accurate assessment of patients’ glomerular filtration rate, (ii) supportive therapies for those whose treatment course is complicated by acute kidney injury, and (iii) pharmacologic strategies to continue or restart cancer therapy during kidney dysfunction. Here we outline 10 tips for common clinical scenarios in an outpatient onco-nephrology clinic.

3553 Background: Vilastobart (XTX101) is tumor-activated, high affinity, Fc-enhanced aCTLA-4 designed to focus activity toward the tumor and minimize systemic adverse events. Fc-enhancement augments FcγR co-engagement on antigen presenting cells and has been linked with efficacy of aCTLA-4 combinations in patients (pts) with microsatellite stable (MSS) colorectal cancer (CRC) and other tumors [Fakih ASCO GI 2025]. Vilastobart was generally well-tolerated and demonstrated evidence of anti-tumor activity in pts with immunologically “cold” advanced solid tumors, both as a monotherapy and in combination with atezolizumab [Davar SITC 2024]. Methods: Phase 2 of the NCT04896697 study evaluated the initial recommended Phase 2 dose of vilastobart 100 mg Q6W in combination with atezolizumab 1200 mg Q3W in pts with MSS CRC who had at least 1 prior chemotherapy regimen in the metastatic setting, excluding pts with prior immune checkpoint inhibitors. Pts with (LM) and without liver metastasis (NLM) were eligible. Tumor biopsies were obtained before and during treatment for translational analyses. Results: As of January 13, 2025, 40 pts were dosed in Phase 2. Median age was 55 (25-82) and 70% of pts had 3 or more prior lines of therapy. Of the 24 enrolled NLM pts, 11 were response-evaluable with an available on-treatment scan as of the data cut. In these 11 pts, two confirmed and one unconfirmed partial response (PR) were reported, all accompanied by significant decreases in ctDNA and serum tumor marker CEA and with each pt ongoing on therapy, for a preliminary ORR of 27%. One additional pt (with peritoneal metastasis) had a 24% reduction in target lesions (first scan) and was ongoing on therapy. Of the 16 enrolled LM pts, 7 were response evaluable, with one stable disease and another pt reporting a mixed response with significant serum tumor marker reductions, both ongoing on therapy. Of note, one LM pt in Phase 1C dose escalation treated with vilastobart (150 mg Q6W) combination had a confirmed PR with LM resolution. Six pts (15%) reported G3+ treatment-related adverse events (TRAEs), with two G4 laboratory TRAEs and no G5 TRAEs. Only three pts discontinued therapy for TRAEs. TRAEs occurring in ≥10% (all grade) or ≥5% (G3) of pts are summarized in the Table. Conclusions: The combination of vilastobart, a novel tumor-activated, Fc-enhanced a-CTLA-4, and atezolizumab demonstrated initial evidence of anti-tumor activity in late line, metastatic MSS CRC where immune checkpoint blockade has historically been relatively ineffective. Vilastobart was observed to have a differentiated safety profile distinct from systemically active a-CTLA-4, consistent with tumor-selective activation. Clinical trial information: NCT04896697 . AE term All Graden (%) Grade 3n (%) Fatigue 12 (30%) 0 Diarrhea 8 (20%) 0 Infusion related reactions 5 (13%) 0 Pyrexia 4 (10%) 0 ALT increased 4 (10%) 0 AST increased 4 (10%) 1 (3%) Colitis 2 (5%) 2 (5%)

1543 Background: Treatment toxicity may limit the ability of cancer patients to receive all recommended cycles of therapy. Four to six cycles of docetaxel plus cyclophosphamide (TC) is a common adjuvant chemotherapy regimen for early-stage breast cancer. We assessed the impact of routine collection of patient-reported outcomes (PROs), coupled with a collaborative care model-based symptom management intervention, on the number of cycles of TC received by patients with breast cancer. Methods: The Enhanced, EHR-facilitated Cancer Symptom Control (E2C2) trial was a cluster-randomized, pragmatic clinical trial, conducted between March 2019 and January 2023 at Mayo Clinic Rochester and within the Mayo Clinic Health System in Minnesota and Wisconsin. Patients regularly reported the severity of 6 SPPADE symptoms (Sleep deficit, Pain, Physical function impairment, Anxiety, Depression, and Energy deficit/fatigue) on 11-point numerical rating scales. Each symptom score was interpreted as none to mild (0-3), moderate (4-6), or severe (7-10). The E2C2 intervention included symptom management education modules, clinician decision support aids, and the option to discuss severe symptoms with a nurse, physical therapist, or social worker. Patients with breast cancer who received at least one cycle of TC were included in this analysis. We compared the number of cycles of docetaxel and cyclophosphamide completed by patients in the control condition versus those in the intervention condition. Results: We identified 198 patients with breast cancer who received TC during the control condition and 128 who received TC during the intervention condition. Median age was 61.3 years in the control group and 60.3 years in the intervention group, and 94% and 95% were white, respectively. Those receiving treatment during the control condition, on average, completed 3.53 cycles of docetaxel, while those receiving TC during the intervention condition completed 3.77 cycles ( p = 0.014). Seventy-eight percent in the control group completed at least 4 cycles of docetaxel, compared to 84% in the intervention group. Those receiving TC during the control condition completed an average of 3.71 cycles of cyclophosphamide, compared to 3.78 cycles in the intervention condition (p = 0.231). Eighty-four percent of patients in the control group and 86% in the intervention group completed at least 4 cycles of cyclophosphamide. Conclusions: Routine PRO surveillance, coupled with guideline-based collaborative care interventions, was associated with completion of a greater number of cycles of docetaxel. These findings suggest that routine symptom surveillance and management may enhance the docetaxel tolerance profile and improve treatment adherence in patients with early-stage breast cancer, allowing for delivery of an optimized treatment course.

9029 Background: Addressing guideline concordant FP and TPP with cancer is challenging, especially for OT. SIM, a well-known education tool, has not been traditionally used to teach these topics. As part of the hematology/oncology fellowship curriculum, we designed a SIM applying LSP to bridge this educational gap. Methods: Applying LSP of contrasting cases, elaboration, feedback, generation and just-in-time telling, chief OTs and faculty designed a SIM with 3 clinical scenarios (CS) and 3 standardized pts (SP): FP for a female pt hospitalized for acute lymphoblastic leukemia tx, FP for a male pt in clinic for germ cell tumor tx, and breast cancer tx for a pregnant pt in 2 nd trimester. OT received written educational materials (WEM) and attended a lecture 1 week and 1 day pre SIM, respectively. OT were split into 2 groups; 1 started in the FP CS and 1 in the TPP CS, and then they switched. To reduce cognitive overload, each OT participated in 1 of 2 FP CS: half in the female FP CS and half in the male. Faculty and chiefs led debriefing sessions after all CS. The FP debrief ended with a special “learner as a teacher” part, where OT who attended the female CS taught those from the male CS about their topic and vice versa. The SIM concluded with "take-home" points. Pre (after having WEM) and post SIM anonymous surveys and verbal feedback assessed OT comfort, perceptions, and knowledge. Results: A 3 hour SIM with 14 OT and 5 faculty was completed successfully. SIM ended with all OT verbally sharing a learning point and positive feedback. All OT filled pre and post SIM surveys. Pre SIM, 17% reported having prior FP education, and 42% had prior TPP education. However, only 8% and 25% felt adequately prepared for FP and TPP discussions, respectively. Post SIM survey showed major improvement in comfort and knowledge about FP and TPP (see table 1). 75% of OT favored a combination of SIM and learning by teaching over other SIM formats. 83% supported having a TPP lecture before SIM. 100% reported satisfaction with FP and TPP SIM’s quality and wanted similar activities in the future. 92% were satisfied with the learner as a teacher tool. 100% felt education about FP and TPP with cancer should be part of fellowship training. Moral/cultural concerns were discussed as possible challenges in FP education. Conclusions: The application of LSP through SIM significantly increased OTs’ comfort discussing FP and TPP with cancer, and objectively enhanced their knowledge on these topics compared to after reading WEM. The SIM was largely perceived as a successful activity to educate about challenging topics and should become part of fellowship curriculum. Pre SIM % Post SIM % Comfortable addressing FP 17 92 Comfortable addressing TPP 8 100 Correct answers about female FP 63 83 Correct answers about male FP 67 94 Correct answers about TPP with cancer 68 98