Abstract Background: Advances in multiplexed imaging now profile cell phenotypes from 2D to 3D, creating new opportunities to analyze spatial organization in complex tissue. A 3D view of tissue architecture enables a re-examination of tumor-immune interactions in the tumor microenvironment, revealing spatially organized niches with direct biological and clinical relevance that may be obscured in 2D. Methods: We developed SpatialTopic1, a fast, scalable, unsupervised niche-detection method that identifies recurrent spatial patterns (“topics”) across multiplexed tissue images. SpatialTopic analyzes datasets with millions of cells within minutes using modest memory. With cell-type annotations as input, it applies to both spatial transcriptomics and proteomics (e.g., Xenium, CosMx, IMC and CODEX). Here, we extend SpatialTopic to 3D multiplexed images and demonstrate its applicability on a 3D CyCIF dataset from the melanoma invasive margin2. We also introduce a supervised framework of SpatialTopic that predicts spatial topic distributions using fixed, predefined topic compositions (“known topics”), enabling more scalable inference across multiple samples and facilitating links between spatial niches and clinical outcomes. Results: SpatialTopic delineates four main topics along the vasculature-to-tumor-core axis at the melanoma invasive margin: (1) Vascular topic: endothelial cells with CD4 T cells and macrophages; (2) Immune topic: mainly including CD4 T cells, dendritic cells, as well as Regulatory T cells (Tregs), and B cells; (3) Tumor-immune boundary topic: tumor cells mixed with dendritic cells and CD4 T cells; and (4) Tumor core topic. In this dataset, relative to the prior publication2, SpatialTopic more cleanly resolves vascular structures and reveals a graded shift in immune composition from vasculature toward the tumor boundary: decreasing macrophages and increasing CD4 T cells and dendritic cells at the invasive front of melanoma. 1.Peng, X. et al. Scalable topic modelling decodes spatial tissue architecture for large-scale multiplexed imaging analysis. Nat. Commun. 16, 6619 (2025). 2.Yapp, C. et al. Highly multiplexed 3D profiling of cell states and immune niches in human tumors. Nat. Methods 22, 2180-2193 (2025). Citation Format: Xiyu Peng, James Smithy, Mohammad Yosofvand, Caroline Kostrzewa, Fiona Ehrich, MaryLena Bleile, Jasme Lee, Michael A. Postow, Margaret K. Callahan, Katherine Panageas, Ronglai Shen. SpatialTopic exploring tumor ecosystem in 3D multiplexed imaging of melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 5499.

1515 Background: Single agent pembrolizumab is standardly administered every 3 weeks (q3wk). Although extended interval dosing every 6 weeks (q6wk) was approved during the pandemic, most oncologists (65%) prefer q3-week dosing to monitor and mitigate immune-related adverse events (irAEs). We hypothesized that telehealth (TH)-enabled safety monitoring may facilitate broader adoption of q6 week dosing with reduced in-person health care visits while maintaining safety and patient experience. Methods: MAking Telehealth Delivery of Cancer Care at Home Effective and Safe for Immunotherapy (MATCHES-IO) is a single-arm pragmatic trial among patients with solid tumors receiving single-agent pembrolizumab. The intervention consisted of q6-week in-person pembrolizumab infusions with interim (eg. weeks 3,9,15) telehealth toxicity assessments during the first 6 months of therapy. Intervention components included a TH clinician visit, home phlebotomy, and biometric monitoring. The primary outcome was the number of days with an in-person health care facility visit, ascertained from the electronic health record. Participants were compared to a contemporaneous matched cohort receiving standard q3-week pembrolizumab, matched 1:1 on cancer type, stage (I–III vs IV), and age (15-year bands). Safety was evaluated by comparing irAEs requiring steroid use. Patient experience was assessed using a survey that included likelihood to recommend the intervention. Results: Between May 2023 and February 2025, 60 patients were enrolled (median age 69.5 [range: 25-85], 38% female, 78% white). Cancer diagnoses included thoracic (52%), genitourinary (30%), and melanoma (18%) in cases and controls. In matched analyses, there was a 3.5-day difference in total healthcare facility contact days between cases and controls (median 8.5 vs. 12.0, p = 0.14) that did not achieve statistical significance. Rates of irAEs requiring steroids were similar between groups (25.0% vs. 27.1%, p = 0.84). Fifty-two patients (87%) in the intervention group completed a patient experience survey. The median likelihood to recommend score was 10 (Q1,Q3: 8.0, 10.0). Most respondents (96%) perceived a benefit from the telehealth platform, including saved time (85%), increased convenience for patients (75%) and caregivers (44%), and cost savings (52%). Conclusions: Extended interval pembrolizumab supported by TH-enabled toxicity monitoring with remote phlebotomy and biometric assessments was associated with reduced in-person health care facility days but did not achieve the significance threshold. The intervention was associated with high patient satisfaction and no increase in irAEs. These results may address clinicians’ concerns about the safety of q6 week immunotherapy dosing and inform future care-delivery strategies integrating telehealth with immunotherapy treatment.

Purpose: Pain management remains challenging for adolescent and young adult (AYA) cancer patients. Acupuncture and massage have been recommended for cancer-related pain management, but no prospective trials have been conducted in AYA patients. We explored the effects of acupuncture and massage on pain in AYAs. Methods: This subgroup analysis focused on AYA patients (aged 18-39 years) enrolled in a randomized controlled trial comparing acupuncture and massage for pain management in advanced cancer. Interventions were delivered weekly for 10 weeks, followed by monthly booster treatments through week 26. The primary outcome was the worst pain score from the Brief Pain Inventory, analyzed using a linear mixed-effects model. Results: Thirty participants met eligibility criteria (13 acupuncture; 17 massage), with a mean age of 31.1 years (standard deviation, 5.8); 57% were female; 67% were White; and 53% had solid tumors. Both groups experienced reduced pain over time. Relative to baseline, patients receiving acupuncture had a mean change of -1.26 points (95% confidence interval [CI], -2.54 to 0.01) at 10 weeks and a mean change of -1.46 points (95% CI, -2.78 to -0.14) at 26 weeks. Patients receiving massage experienced a mean change of -2.81 points (95% CI, -3.92 to -1.70) at week 10 and a mean change of -3.79 points (95% CI, -4.85 to -2.73) at week 26. Conclusion: AYA patients with advanced cancer who received either acupuncture or massage experienced clinically meaningful and sustained reductions in pain. These findings provide a promising foundation for future trials aimed at evaluating integrative pain management strategies in AYAs.

TPS1680 Background: Telehealth (TH) use for cancer care has rapidly expanded, both during and following the COVID-19 pandemic. However, evidence regarding safety, feasibility, effectiveness, equity, stakeholder satisfaction, and implementation of telehealth-enabled oncology care remains limited. We seek to assess the impact of TH on these outcomes in breast and prostate cancer patients. Methods: The MATCH-UP pragmatic cluster randomized trial is being conducted at Memorial Sloan Kettering Cancer Center to compare alternate models of TH use in oncology practice. Sixty-two physician practice clusters in breast and prostate medical oncology are randomized 1:1 to enhanced telehealth (ET) or usual practice (UP), stratified by disease type and clinic volume. ET emphasizes TH with expanded home services designed to deliver as many components of care as desired in patients’ homes. Usual care involves routine practice with TH visits at the discretion of the patient and MD or APP. Eligible patients have >=3 prior medical oncology visits and are not enrolled on a therapeutic trial. Automated enrollment is triggered by a routine oncology visit with waiver of informed consent. The ET intervention includes EHR-enabled telehealth scheduling defaults for routine follow-up visits, optional home phlebotomy, structured support for home administration of select injectable medications, and digital support for patients with TH access barriers. The primary endpoint is the proportion of routine medical oncology visits conducted in person among all routine medical oncology visits over 1 year. Secondary outcomes include healthcare utilization, no-show and late cancellation rates, and overall survival. Patient-reported outcomes include: quality of life, healthcare costs, experience of care, and preferences for use of TH at future visits. Clinician-reported outcomes include: experience of care and preferences for TH use at future visits. For patients in the ET arm only, telehealth accessibility, intervention uptake and efficiency, and implementation outcomes, including acceptability, appropriateness, and feasibility, are also being collected. The primary endpoint will be analyzed using generalized linear mixed models with a logistic link, accounting for repeated visits within patients and clustering within randomized practice units and adjusting for stratification factors. Enrollment will continue through March 2026 with follow-up for 12 months. To date, 7256 patients have been enrolled, with 3641 patients assigned to ET, (n=2437 [67%] breast cancer; n=1204 [33%] prostate cancer and 3,615 assigned to UP (n=2252 [62%] breast cancer; n=1363 [38%] prostate cancer). MATCH-UP will generate pragmatic evidence on the effectiveness, patient-centeredness, implementation, and equity-relevant barriers of a scalable, EHR-embedded enhanced telehealth model for breast and prostate oncology care. Clinical trial information: NCT06954337 .

Prostate cancer (PC) shows marked heterogeneity in clinical outcomes, from indolent to highly aggressive disease. Alterations in TP53, PTEN, RB1, and DNA damage repair genes are linked to poor patient outcomes, yet how the interaction and co-occurrence of these alterations impact prognosis remains unclear. We leveraged the AACR Project GENIE prostate cancer cohort to assess the prognostic impact of individual and concomitant gene alterations, focusing on key PC drivers (BRCA2, TP53, PTEN, RB1). the study was conducted in two stages: (i) identification of candidate alterations at individual gene level through a systematic literature review, followed by analyzing the prevalence and co-occurrence of genes in the AACR Project GENIE Public Release (PR) dataset. (ii) Prognostic analysis of the pre-selected biomarkers in the AACR Project GENIE Biopharma Collaborative (BPC) cohort. Analyses were performed in three clinical settings (entire BPC cohort, n=1114; subgroup of patients with localized disease at diagnosis, n=837; and subgroup of patients with metastatic PC with a biopsy collected before or at the time of mPC, n=455) using two statistical approaches: univariate analysis (UA) and inverse probability weighting using propensity scores (PS). Cox models were applied to estimate overall survival hazard ratios (HRs). after integrating the findings from the systematic review with the analysis of the GENIE PR (n=5684), 107 individual genes and 122 co-occurring alterations were selected for study. In the GENIE BPC (n=1114 patients) 75% were M0 at diagnosis, and 67% presented metastatic PC at any time. Alteration prevalence was: TP53, 25%; PTEN, 15%; BRCA2, 4%; RB1, 4%; co-alterations included, PTEN-TP53 (6.4%), RB1-TP53 (2.1%), PTEN-RB1 (1.3%), and BRCA2-TP53 (1.2%). All four key genes showed a statistically significant prognostic impact in the entire cohort with UA (BRCA2, HR:1.9; TP53, HR:2.6; PTEN, HR:2.4; RB1, HR:4.1, all p<0.001). Results were consistent in the subgroup analysis and when using PS. Additional genes such as FANCA, PIK3CA, and MYC were also associated with poor outcomes. Among co-alterations, BRCA2-TP53 (HR:2.0, p=0.05), RB1-TP53 (HR:3.6, p<0.001), PTEN-TP53 (HR:2.4, p<0.001); PTEN-RB1 (HR:5.5, p<0.001), and PTEN-TP53-RB1 (HR:9.0, p<0.001) were the strongest predictors of poor survival, with consistent effects across subgroups and analytical methods. Other combinations including MYC-TP53 (HR 2.2, p<0.001) and PIK3CA-TP53 (HR 3.2, p<0.001), also showed independent prognostic value. Genomic alterations in TP53, PTEN, RB1, and BRCA2, alone and in combination, independently associate with poor prognosis. Co-alteration patterns such as PTEN-TP53-RB1 define aggressive molecular subsets. This evidence supports integrating genomic interaction patterns for enhancing clinical risk models and guiding precision patient management. Furthermore, this study highlights the potential of AACR Project GENIE to enable data-driven development of biomarker-informed patient care strategies. Pablo Cresta Morgado, Victor Navarro, Guillermo Vilacampa, Haitham Alatoom, Manuel Ramos del Rio, Irbaz Riaz, Shawn Sweeney, Nikolaus Schultz, Ken Kehl, Gregory Riely, Wasim Abida, Katherine Panageas, Deborah Schrag, Philippe Bedard, Christine Micheel, Xindi Guo, Chelsea Nayan, Rodrigo Dientsmann, Joaquin Mateo. Genomic prognostic biomarkers in prostate cancer from the AACR GENIE cohort [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Innovations in Prostate Cancer Research and Treatment; 2026 Jan 20-22; Philadelphia PA. Philadelphia (PA): AACR; Cancer Res 2026;86(2_Suppl):Abstract nr A013.

TPS1683 Background: Tarlatamab is a bispecific T-cell engager (BiTE) therapy approved for the treatment of extensive-stage small cell lung cancer (ES-SCLC) and is the first BiTE therapy approved for a solid tumor. Due to the risk of cytokine release syndrome, patients are routinely hospitalized for ≥24 hours following each of the first two doses. These hospitalizations can pose substantial logistical, financial, and psychosocial challenges, particularly for patients in rural communities or with caregiver or employment constraints. HaH is an emerging care delivery model that provides hospital-level care in patients’ homes and has been associated with reduced readmission rates, improved patient experience, and lower cost in other populations. Whether HaH can safely and efficiently replace inpatient hospitalization for post-tarlatamab monitoring is unknown. The MAking Telehealth Delivery of Cancer Care at Home Effective and Safe (MATCHES)-Novel trial evaluates a HaH model compared with standard inpatient hospitalization following tarlatamab administration. Methods: MATCHES-Novel is an active, single-center, prospective randomized controlled trial comparing HaH with standard inpatient hospitalization for post-administration monitoring of tarlatamab in patients with ES-SCLC. Eligible patients have an ECOG performance status of 0-2, are initiating tarlatamab for an FDA-approved indication, reside within a 60-minute response catchment area, and have an available in-home caregiver. Participants are randomized 1:1 to HaH or standard inpatient monitoring following drug administration, with monitoring duration and discharge criteria aligned with institutional standards. The HaH intervention includes scheduled in-home community paramedic assessments on the day of and the day after infusion, home phlebotomy, telehealth physician oversight, patient and caregiver education, and 24/7 care escalation pathways. The primary objective is to compare efficiency, measured by the number of inpatient hospital days during the two 7-day periods after each infusion. Assuming a mean number of 4 inpatient hospital days in the control arm, enrollment of 70 patients provides ≥80% power to detect a statistically significant difference if the mean inpatient days per patient in the HaH arm is ≤2.8. Secondary objectives include safety outcomes (transfers from home to inpatient care, clinician adjudicated toxicity events) and patient, caregiver and physician reported experiences and preferences, assessed through prespecified surveys and interviews. These findings may inform future care-delivery models for administering innovative cancer therapies outside the inpatient setting, potentially improving access. The trial opened to accrual in April 2025, and 16 of the planned 70 patients have been enrolled. Clinical trial information: NCI-2025-03406 .

BACKGROUND:No standardized method exists for seizure assessment in glioma clinical trials. We describe the development and evaluation of RANO-TREAT (Tumor Related Epilepsy Assessment Tool) for seizure assessment and its association with changes on brain MRI. METHODS:Patients with glioma/glioneuronal tumors and ≥ 1 prior seizure along with clinicians completed RANO-TREAT in conjunction with brain MRIs, yielding multiple RANO-TREAT scores at clinic visits over time. Unweighted (primary) and weighted (post-hoc) scores were correlated with disease progression via MRI in all patients and patients with IDHmt tumors, separately. Cohorts were randomly split by patient into cohort-specific training and validation sets. Weights for RANO-TREAT items were defined by multivariable generalized estimating equation models in cohort-specific training sets and validated in cohort-specific validation sets. A nomogram was developed using overall cohort training and validation sets. RESULTS:Four hundred and ninety patients (310 IDHmt tumors) had ≥ 1 visits and 285 patients (168 IDHmt tumors) had ≥ 2 visits. Unweighted RANO-TREAT scores (OR:1.01; 95%CI:0.998-1.02; P = .13) and score changes (OR:1.00; 95%CI:0.99-1.02; P = .63) were not associated with progressive disease on MRI. Post-hoc analysis using training and validation sets demonstrated weighted RANO-TREAT scores were correlated with progressive disease in both overall cohort validation set (OR:2.51; 95%CI:1.80-3.52; P < .0001) and IDHmt cohort validation set (OR:4.53; 95%CI:2.11-9.75; P = .0001). Weighted analyses for patients with ≥ 2 visits showed similar associations in validation sets. CONCLUSIONS:This prospective study suggests an association of seizure control evaluated by a new standardized tool with disease progression in glioma. This tool requires further systematic evaluation in glioma clinical trials alongside more traditional endpoints.

9519 Background: IO102-IO103 is an investigational cancer vaccine that targets both tumor and immune-suppressive cells in the tumor microenvironment. IO102-IO103 has demonstrated clinical activity in combination with anti-PD-1 monotherapy in advanced melanoma, with concordant clonal T cell expansion observed in the tumor and peripheral blood in patients with radiographic response. IO102-IO103 has not yet been assessed in combination with nivolumab-relatlimab (nivo-rela). Methods: In this multicenter, phase 2 trial (NCT05912244) patients with unresectable, previously untreated non-uveal melanoma were treated with subcutaneous IO102-IO103 and intravenous nivo-rela for up to two years. IO102-IO103 was administered every two weeks for eight weeks, and every four weeks thereafter. PD-L1 protein expression was assessed using the 28-8 pharmDx assay on pre-treatment tissue. The primary endpoint was best overall response rate (BORR) by RECIST v1.1, with a plan to reject the null hypothesis based on the proportion of patients with ≥1% membranous PD-L1 staining in the tumor compartment. Secondary endpoints included progression-free survival (PFS), safety assessed by Common Terminology Criteria for Adverse Events v 5.0, and duration of response (DOR). PFS and DOR were summarized using Kaplan-Meier methods. Bulk T cell receptor (TCR) sequencing was performed on peripheral blood mononuclear cells using an RNA-based assay at baseline, week 4 and week 8. Results: Among 43 evaluable patients, the BORR was 60% (95% confidence interval [CI]: 44-75%), including 19 patients with partial response and 7 patients with complete response. At database lock (Dec 1 2025), median follow up was 10.1 months (interquartile range 6.1-19.3). The median duration of response was not reached (95% CI: 14, NR) and median PFS was 8.2 months (95% CI 6.8, not reached [NR]). Among patients with PD-L1 negative tumors (n = 25), the BORR was 52% (95% CI 31-72%) and median PFS was 8.2 months (4.2, NR). Grade 3-4 treatment-related adverse events occurred in 9 patients (21%), including adrenal insufficiency (2 patients), acute kidney injury (2), aseptic meningitis (1), arthritis (1), maculopapular rash (1), myositis (1), neutropenia (1), and colitis (1). There were no treatment related deaths. T cell clonal expansion was observed at on-treatment time points compared to baseline in patients with and without radiographic response. Conclusions: IO102-IO103 in combination with nivo-rela was associated with a higher objective response rate compared to historical data for nivo-rela alone, meeting the trial’s primary endpoint. No unexpected safety signals were observed. These findings support the further clinical investigation of IO102-IO103 with nivo-rela as an initial treatment for unresectable melanoma. Clinical trial information: NCT05912244 .

PURPOSE:Anxiety is prevalent, disruptive, and undertreated among survivors of cancer. Cognitive behavioral therapy (CBT) is the first-line treatment, but not all individuals have access, respond to treatment, or prefer this option because of stigma. Music therapy is effective for short-term anxiety reduction, but it is unknown whether it is noninferior to first-line CBT for long-term anxiety reduction. METHODS:This comparative effectiveness trial randomly assigned English- or Spanish-speaking survivors of cancer to seven weekly telehealth sessions of music therapy or CBT. The coprimary end points were changes in the Hospital Anxiety and Depression Scale (HADS) anxiety score at weeks 8 and 26. The noninferiority margin was 0.35 standard deviations, informed by a minimal clinically important difference (MCID) of 1.7 points. Secondary outcomes included fatigue, depression, insomnia, pain, cognitive dysfunction, and health-related quality of life. RESULTS:Among N = 300 patients, 74.7% was female, 76.5% was White, and 19.0% was Hispanic. At week 8, the mean change in HADS anxiety score was -3.12 (95% CI, -3.59 to -2.65) in music therapy and -2.97 (95% CI, -3.45 to -2.50) in CBT; the between-group difference was -0.15 (95% CI, -0.78 to 0.49), within the noninferiority margin of 1.20 (P < .001). At week 26, the mean change was -3.31 (95% CI, -3.78 to -2.85) in music therapy and -3.00 (95% CI, -3.47 to -2.53) in CBT; the between-group difference was -0.31 (95% CI, -0.95 to 0.32), within the noninferiority margin of 1.28 (P < .001). Both groups produced anxiety reductions exceeding the MCID and showed similar improvements in secondary outcomes. CONCLUSION:Music therapy is noninferior to CBT for anxiety in survivors of cancer. Both telehealth interventions produced clinically meaningful, durable improvements in anxiety.

1581 Background: Among the growing population of premenopausal patients with breast cancer, years-long endocrine therapy often requires repeated clinic-administered injections, creating substantial time burden and adherence challenges. This pilot evaluated the feasibility of a telemedicine-supported home injections program designed to improve access, convenience, and patient experience. Methods: Patients were recruited from 4 outpatient medical oncology practices at a comprehensive cancer center between October 2024 and January 2025. Eligibility criteria included a breast cancer diagnosis, an active treatment plan including leuprolide and/or denosumab, and use of the patient portal. Patients proceeded to pilot participation after insurance approval for outpatient administration of medications and (if applicable) acceptability of copay cost. Patients and/or caregivers received in-clinic training by nursing staff on injection preparation and administration, were provided both written and video-based educational material and were followed for up to 2 home injections over a 6-month period, with telemedicine support. The primary outcome was feasibility, assessed by home injection completion rates. Patient and clinician satisfaction were assessed using the Net Promotor Score (NPS) and rates of continued at home injection administration post pilot completion. Patients were also invited to participate in 60-minute semi-structured exit interviews. Results: Of 105 eligible patients, 54 agreed to participate in the pilot and 24 obtained insurance approval for the medication with an acceptable copay. All 24 patients were trained in injection administration of intramuscular leuprolide, of whom 50% had no prior injection experience. Overall, 96% successfully completed one home injection, 79% of patients completed two home injections, and 75% continued home administration after follow-up. Notably, only 1 patient discontinued home injection administration due to telemedicine scheduling related issues. Patients reported high satisfaction, citing time savings and convenience as key benefits. Both patients and providers strongly endorsed the model with highly compelling net promoter scores of 69 and 61, respectively. Conclusions: This pilot demonstrated the feasibility of a telemedicine supported home injection care delivery model for breast cancer patients in the oncology setting, evidenced by high completion rates and patient preference to continue home administration. Educational materials and optional telemedicine visits for initial injections were leveraged to support adherence without increased healthcare utilization. Further evaluation across broader geographic, demographic, and payor mix is warranted to inform scale up. An ongoing pragmatic trial (NCT06954337) is testing this approach as part of an innovative model of care: Enhanced Telehealth.

9557 Background: While inpatient toxicity associated with tumor infiltrating lymphocyte (TIL) therapy and IL-2 administration is well characterized, risks facing patients after hospital discharge are less understood. Better characterization of outpatient adverse effects (AEs) could guide the optimal frequency and duration of follow up for this growing patient population. Methods: We conducted a retrospective analysis of all patients with advanced melanoma discharged from Memorial Sloan Kettering Cancer Center after receiving investigational or commercial lifileucel from October 2020 to October 2024. We reviewed incidence and timing of new Grade 3+ treatment-related AEs (TRAEs), blood product administration, and readmission among all patients from the time of hospital discharge to the time of the start of a subsequent systemic therapy or death. Results: Fifty-three patients successfully discharged after lifileucel administration were identified; patient demographics are included in Table 1 . The median follow up time from discharge in survivors was 5 months (interquartile range (IQR): 3, 18). Two patients (4%) developed new Grade 3+ TRAEs following hospital discharge, including new Grade 3 neutropenia 73 days after discharge and new Grade 3 hypoxia 104 days from discharge. Hypoxia was secondary to pleural effusions that developed in the setting of renal thrombotic microangiopathy. Four patients (7.5%) were readmitted for TRAEs including cytopenias, dyspnea, and syncope while 7 patients (13%) were readmitted for melanoma progression. Readmissions occurred a median of 85 (IQR: 39, 128) days after lifileucel infusion and 69 days (IQR: 19, 98) after initial discharge. Twelve patients (23%) received at least one outpatient blood product transfusion, including packed red blood cells (PRBCs; 10 patients) and platelets (6 patients). Within 30 days of initial discharge, six patients (11%) received at least one PRBC transfusion and 3 patients (5.7%) received at least one platelet transfusion. The median number of transfused PRBC units was 2 (IQR: 1,4) and the median number of platelet transfusions was 4 (IQR: 2, 4). Conclusions: Rates of new severe toxicity and treatment-related readmissions were low among patients discharged post-lifileucel. About one in four patients required blood products after discharge. Identification of risk factors for the development of outpatient TRAEs may inform personalized care following lifileucel administration. Patient demographics. Characteristic N = 53 1 Age at TIL infusion 61 (42, 66) Sex Male 29 (55%) Female 24 (45%) Melanoma subtype Cutaneous 19 (36%) Uveal 9 (17%) Acral 8 (15%) Unknown primary 8 (15%) Mucosal 5 (9.4%) Other 4 (7.5%) BRAF status Mutated 13/51 (25%) Wild type 38/51 (75%) Treatment setting Investigational 45 (85%) Standard of care 8 (15%) 1. N (%); Median (interquartile range).

9555 Background: Mucosal melanoma (MM) is an aggressive subtype of melanoma with distinct biology, and outcomes in advanced disease are inferior compared with cutaneous melanoma. Frontline Chinese studies in MM have shown efficacy of combined VEGF/R and PD-1 blockade, but studies in more diverse populations and options in PD-1 resistance are lacking. Methods: We conducted a phase 2/1b single-center trial in patients (pts) with untreated, unresectable or advanced MM. Pts received standard nivolumab (nivo) plus axitinib (axi) 5mg PO twice daily. Primary endpoint of the phase 2 doublet arm was objective response rate (ORR) by RECIST 1.1 (H0=23%, Ha=48%). Clinical benefit rate (CBR) was defined as ORR or stable disease (SD) >6 months (mos). Upon progression with good tolerance, the phase 1b triplet arm pts received the addition of either stereotactic body radiotherapy (SBRT, 30Gy/5 fractions) or ipilimumab (ipi, 1mg/kg < 4 doses) to ongoing nivo + axi. The primary endpoint of the triplet was safety by CTCAE v5.0 and adverse events (AEs) of special interest (AESIs). Kaplan-Meier methods estimated time to event outcomes; ORR and AEs were reported as proportions with exact 95% confidence intervals. Results: N=21 pts were enrolled; N=20 were evaluable for efficacy. See Table for baseline population characteristics. Median follow up was 15 mos (IQR 6, 21), 45% of pts (95% CI: 23, 68) had an objective response; 3 complete and 6 partial responses. Median duration of response was 13 mos (8.6, not reached (NR)). SD persisted ≥ 6 mos in 2 of 7 pts; CBR was 55% (95% CI: 32,77). Median progression free survival (PFS) was 6.3 mos (3.5, NR), and 12-mos estimated PFS and overall survival was 37% (20,67) and 71% (52, 96), respectively. Rate of grade ≥3 treatment related AEs (TRAE) in the doublet arm (n=21) was 67% (95% CI: 43,85), most commonly hypertension & hepatitis, with two pt deaths; 1 nivo-related myasthenia gravis / myositis, and 1 nivo-related pancreatitis with steroid-related PJP pneumonia. 14 pts (70%) progressed on doublet therapy, of which 7 enrolled on the triplet arm. N=5 received ipi and N=2 SBRT (both to anorectal primaries and adjacent lymph nodes). There were 2 grade ≥3 TRAEs in the ipi triplet arm (hepatitis), 0 in the SBRT arm, no grade 5 events, and no AESIs. Zero of 4 evaluable pts in ipi triplet and 1 of 2 in the SBRT triplet responded (4+ mos, ongoing). Conclusions: The frontline combination of nivolumab and axitinib was effective in patients with unresectable or advanced outside of China, and a prospective global study randomized against immune checkpoint blockade is warranted. Adding either ipi or SBRT to nivo-axi appears safe in select pts with progressive disease and further studies are needed for pts with PD-1 resistant MM. Clinical trial information: NCT05384496 . Characteristic (n=21) N (%) Age (median) 73 years (IQR: 67, 82) Sex Female Male 13 (62%) 8 (38%) Race Caucasian Asian Black 16 (76%) 3 (14%) 2 (10%) Primary Site Anorectal Sinonasal Vulvovaginal 10 (48%) 8 (38%) 3 (14%) Stage Locoregionally advanced Metastatic 14 (67%) 7 (33%) LDH (median) 195 (IQR: 171,201)

This cohort study examines the response and survival rates associated with ipilimumab-nivolumab therapy in patients with progressive melanoma brain metastases after anti–programmed cell death 1 (anti–PD-1) therapy.

Anxiety and depression are common in many cancers but have not been systematically studied in patients with histiocytic neoplasms (HN). We sought to estimate rates of anxiety and depression and identify clinical features and patient-reported outcomes (PROs) associated with anxiety and depression in patients with HN. A registry-based cohort of patients with HN completing PROs including the Hospital Anxiety and Depression Scale (HADS) from 2018-2023 was identified. Moderate or severe anxiety or depression were respectively defined as a score of 11+ on the HADS anxiety or depression subscales. Associations of variables, including other validated PROs, with moderate or severe anxiety or depression were modeled with logistic regression to estimate odds ratios (OR) and 95% confidence intervals (CI). In 215 patients, approximately 1 in 3 met the criteria for anxiety or depression and 1 in 7 met the criteria for moderate or severe anxiety or depression. These estimates remained stable over a twelve-month trajectory. Rates of depression, but not anxiety, significantly differed across HN types with patients with Erdheim-Chester Disease experiencing the highest rate. In addition, neurologic involvement, unemployment, and longer undiagnosed illness interval were significantly associated with increased risk of depression. Financial burden, financial worry, and severe disease-related symptoms were correlated with increased risk of both anxiety and depression. Conversely, increased general and cognitive health-related quality of life (HRQoL) were correlated with decreased risk of both anxiety and depression. In patients with HN, anxiety and depression are prevalent, stable over time, and correlated with financial burden, symptom severity, and HRQoL.

Methotrexate (MTX) is first-line treatment for central nervous system lymphoma (CNSL) but poses significant toxicity risks, especially in patients with renal dysfunction. Glucarpidase 50 u/kg is approved for treatment of MTX toxicity once it develops. Here, we aim to explore empiric glucarpidase administration for patients at high-risk of MTX toxicity. This is a prospective study of planned low-dose glucarpidase (1000u) following MTX in patients with CNSL. Eligible patients had baseline CrCl <60 ml/min, or history of delayed MTX clearance or grade 2 or higher MTX-related toxicity. Each enrollment corresponded to one MTX and glucarpidase dose; re-enrollment was permitted. Glucarpidase was given ≥ 24 hours post-MTX. The primary endpoint was MTX clearance (≤ 100 nmol/L) 24 hours following glucarpidase. This study is registered on clinicaltrials.gov (NCT03684980). Twenty glucarpidase treatments were administered to 8 patients (6 men, 2 women) with a median age 72.5 (range, 53-83). The baseline Karnofsky Performance Status was 80 (range, 70-100). Enrollment indications included renal dysfunction (n=1), history of delayed MTX clearance (n=2), and MTX-related acute kidney injury (n=5). Patients enrolled one (n=5), four (n=2), or seven times (n=1). MTX levels were reduced to < 100 nmol/L 24 hours after glucarpidase following 17 of 20 treatments (85%). In the remaining 3 cases—occurring in the same patient — MTX levels were reduced by >97%. There were no events attributed to glucarpidase. There were five grade 3 events at least possibly attributed to MTX that included anemia (n=2), creatinine increase (n=1), and absolute lymphocyte count (ALC) decrease (n=2). Other MTX-attributed adverse events were grade 1 or 2. Anti-glucarpidase antibody data are pending. Planned low-dose glucarpidase 1000u results in rapid MTX clearance in patients at high risk of MTX toxicity. Further research is needed to determine whether it prevents MTX dose-reduction, reduces toxicity, or shortens hospital admission.

Primary central nervous system lymphoma (PCNSL) is an aggressive malignancy with poor prognosis for relapsed/refractory (r/r) disease. Inhibitors of the Bruton’s tyrosine kinase (BTK) signaling node yield radiographic response but early resistance is common. Here, we combine the BTK inhibitor ibrutinib with pan PI3K inhibitor, copanlisib. Eligible patients had r/r PCNSL, age≥18, ECOG≤2. Patients were dosed with concurrent daily ibrutinib and weekly intravenous copanlisib (days 1, 8, 15 of 28-day cycle) or sequentially treated with 1 cycle daily ibrutinib then 2 cycles copanlisib days 1, 8, 15. Strong CYP3A4/5 inducers/inhibitors were prohibited. Eighteen patients were enrolled; 14 to concurrent therapy, 4 sequential. Median age was 63 (range, 40-80), ECOG 1 (0-2), 10 women, 4 men. Patients had a median of 2 prior treatments (range, 1-10) and 11 (61%) were chemo-refractory. Prior treatments included methotrexate (MTX) (n=18), ibrutinib (n=6), radiation (n=2), and transplant (n=5). Partial or complete response was observed in patients treated concurrently (n=8, 57%) and sequentially (n=2, 50%). Median progression-free survival was 3 months. Pharmacokinetic data demonstrated low plasma (Cmax: 17.9 ng/mL) and nearly undetectable CSF (Cmax: 0.01 ng/mL) levels of ibrutinib during concurrent treatment suggesting direct drug-drug interaction. During sequential therapy, plasma levels of ibrutinib were in the expected range (serum Cmax: 212 ng/mL). There was 1 grade 5 lung infection (PJP pneumonia). Grade 3 and 4 events included rash (n=6), lymphocyte count decrease (n=4), 2 instances each of hypertension and elevation of aspartate and alanine aminotransferase, 1 instance of neutropenia, thrombocytopenia, hyperglycemia, and aspergillosis. The most common grade 1/2 events were hyperglycemia (n=11) and nausea (n=6). Combination copanlisib and ibrutinib yielded radiographic responses though this likely reflected effects of copanlisib rather than combination therapy. Sequential dosing is required for copanlisib and ibrutinib to reach cytotoxic levels. This study reiterates the importance of real-time pharmacokinetic monitoring.

HD-MTX is standard treatment for central nervous system lymphoma (CNSL). Administration typically requires hospitalization until drug clearance, due to toxicity risks. We previously demonstrated that glucarpidase, a bacterial recombinant enzyme that rapidly degrades MTX, facilitated outpatient MTX treatments during the COVID-19 pandemic. Here, we further assess feasibility of outpatient MTX with planned glucarpidase use. This prospective, non-randomized study included two cohorts. Cohort A received eight 14-day cycles of MTX 3.5 g/m2 and glucarpidase 2000u 24 hours later. Cohort B included rituximab 500 mg/m2, and 1000u glucarpidase in cycles 3-8 only. In both cohorts, cycles 1-2 were administered inpatient to ensure MTX tolerability, while cycles 3-8 were outpatient. Concurrent chemo/targeted therapies were permitted at physician discretion. The primary endpoint was serum MTX reduction 1 hour after glucarpidase. Patients not receiving at least 6 cycles of glucarpidase were replaced. Eighteen patients consented (Cohort A: 4; Cohort B: 14). Twelve transitioned to outpatient MTX with glucarpidase; six disenrolled before outpatient treatment due to preference or elevated creatinine. Median age was 73.5 (range, 44-82), median Karnofsky Performance Status (KPS) 80 (range, 60-90). One Cohort B patient withdrew after the first outpatient cycle due to preference. Remaining Cohort B patients completed all planned cycles. No hospital admissions or MTX dose-reductions occurred. Median MTX reduction 1 hour post-glucarpidase was 99.6% (range, 62.2-100%). Reduced clearance was noted in one Cohort A and three Cohort B patients; antibody analysis is ongoing. Treatment was well tolerated with no grade 3/4 adverse events attributed to glucarpidase. One Cohort A patient experienced a grade 2 allergic reaction. Grade 1 events included dizziness, fatigue (n=2), flushing, headache, infusion reaction, nausea, paresthesia, pruritus, and rash. Planned-use glucarpidase enables safe and feasible outpatient MTX administration in healthy patients who have previously tolerated MTX.

Multiplex immunofluorescence (mIF) is a promising tool for immunotherapy biomarker discovery in melanoma and other solid tumors. mIF captures detailed phenotypic information of immune cells in the tumor microenvironment, as well as spatial data that can reveal biologically relevant interactions among cell types. Given the complexity of mIF data, the development of automated analysis pipelines is crucial for advancing biomarker discovery. In pre-treatment melanoma samples from 50 patients treated with immune checkpoint inhibitors (ICIs), a higher stromal B cell percentage is associated with the clinical benefit of ICI therapy. The automatic detection of B cell aggregates with DBSCAN, a novel application of a computer-aided machine learning algorithm, demonstrates the potential for enhanced accuracy compared to pathologist assessment of lymphoid aggregates. TCF1+ and LAG3- T cell subpopulations are enriched near stromal B cells, suggesting potential functional interactions. These analyses provide a roadmap for the further development of spatial immunotherapy biomarkers in melanoma and other diseases.