Browse Articles
Discover research articles across all indexed journals
Stratifying survival benefit of encorafenib/cetuximab (EC) in BRAF V600E–mutant mCRC using clinical and inflammatory biomarkers: Real-world experience from greater Manchester.
e15563 Background: BRAF V600E-mutant metastatic colorectal cancer (mCRC) is an aggressive subtype with historically poor outcomes. Although encorafenib/cetuximab (EC) improves survival, real-world data and the prognostic utility of clinical and inflammatory biomarkers remain limited. This study presents real-world data from a large UK cancer centre, incorporating expanded biomarker and prognostic group analyses. Methods: This retrospective cohort study included 100 adults with BRAF V600E-mutant mCRC receiving encorafenib/cetuximab (EC) between March 2020 and August 2025. Treatment continued until progression or unacceptable toxicity. Overall (OS) and progression-free survival (PFS) were estimated via Kaplan-Meier method. Prognostic groups (metastatic burden, liver involvement, time from metastatic diagnosis to EC) and baseline inflammatory indices (NLR, PLR, MLR, SIRI, PNI) were analysed using median cut-offs. Univariable and multivariable Cox models identified independent prognostic factors. Results: Among the 100 patients included (n = 100), 49% were females and the median age was 64 years. ECOG performance status was 1 in 78% of patients, and 68% presented with right-sided primary tumours. Liver metastases were observed in 59% of patients, and 45% had undergone prior primary tumour resection. Over a median follow-up of 10 months, the median PFS was 5.85 months (95% CI 5.27–6.42), and median OS was 10.25 months (95% CI 7.39–13.11). Patients with favourable prognostic characteristics (1 or 2 metastatic sites, no liver metastases, and time to initiation of EC from metastasis diagnosis more than 6 months) achieved significantly longer PFS (7.26 vs 5.42 months, p = 0.014) compared to those with poor characteristics. Low PLR was associated with improved PFS (6.24 vs 5.26 months, p = 0.05), while higher PNI independently predicted inferior PFS on multivariable analysis (HR 1.71, 95% CI 1.10–2.66, p = 0.02). For OS, a lower baseline NLR was linked to improved survival (13.8 vs 9.1 months, p = 0.03). No clinical biomarker was significant in multivariable analysis. Conclusions: In this large real-world cohort, encorafenib/cetuximab outcomes align with pivotal trial data. Composite clinical prognostic groups and simple inflammatory biomarkers (NLR, PNI) offer meaningful prognostic information, supporting improved risk stratification and personalized treatment in patients with BRAF V600E–mutant mCRC.
Temporal trends and sociodemographic differences in influenza vaccination among U.S. cancer survivors, 2005-2024.
1674 Background: Given the increased risk of complications and adverse outcomes associated with influenza infection among individuals with cancer, even in the post-treatment setting, clinical oncology guidelines support annual vaccination in cancer survivors. Few studies have investigated long-term trends and potential disparities in the prevalence of influenza vaccination among individuals with a history of cancer. Therefore, we examined temporal trends in and factors associated with influenza vaccination in a nationally representative sample of cancer survivors in the United States. Methods: Using data from 20 survey years (2005-2024) of the National Health Interview Survey, we included adults aged ≥18 years with a self-reported history of cancer and complete data on influenza vaccination. The primary outcome was self-reported influenza vaccination in the past 12 months. Crude annual prevalence estimates were calculated overall and by select sociodemographic characteristics using survey-weighted logistic regression models. Temporal trends were quantified using annual percent change (APC) estimates. Results: Among 43,932 adults with a history of cancer (median age: 65.7 years), representing approximately 15.0 million U.S. cancer survivors annually, crude influenza vaccination prevalence increased from 45.5% (95% CI: 43.0-48.1) in 2005 to 63.3% (95% CI: 61.2-65.4) in 2024, corresponding to an absolute change of 17.8 percentage points (pp; 95% CI: 14.5-21.1) and an APC of +0.79 pp/year (95% CI: 0.52-1.06). Younger survivors aged 18-44 years experienced a particularly large absolute change over the 20-year study period (+32.4 pp; APC: +1.32 pp/year), yet fewer than half were vaccinated in 2024 (46.4%), compared to 77.4% of those aged ≥75 years. Cancer survivors without insurance (change: +24.7 pp; APC: +0.83 pp/year) had a 2024 vaccination prevalence of 32.0%, compared to insured survivors (63.8%). Survivors without a usual source of care (change: +31.0 pp; APC: +1.63 pp/year) had a 2024 vaccination prevalence of 39.9%, compared to those with a usual source of care (63.8%). Hispanic survivors (change: +25.3 pp; APC: +1.30 pp/year) had a 2024 vaccination prevalence of 51.9%, compared to non-Hispanic White survivors (64.9%). Conclusions: Influenza vaccination prevalence among U.S. cancer survivors increased substantially over the past two decades; however, more than one-third were unvaccinated in 2024. Despite overall gains, vaccination coverage remained particularly low among younger survivors, uninsured individuals, and those without a usual source of care, with persistent differences across racial and ethnic groups. Targeted clinical and public health strategies are needed to address these gaps and improve guideline-endorsed uptake of seasonal influenza vaccination among cancer survivors.
Safety of immune checkpoint inhibitors in solid tumor patients with hepatitis C: A real-world analysis.
11158 Background: Patients with hepatitis C virus (HCV) infection are commonly excluded from immune checkpoint inhibitor (ICI) trials. We evaluated hepatotoxicity and outcomes after ICI therapy in patients with versus without HCV. Methods: Retrospective cohort study using the TriNetX Research Network (2014-2024). Adults with solid tumors initiating ICIs were classified as HCV if they had detectable HCV RNA and/or ICD-10 B18.2 (chronic viral hepatitis C) within 1 year prior to ICI start. Cohorts were 1:1 propensity-score matched on demographics, comorbidities, cancer type, metastatic site, and baseline liver tests. Outcomes from day 1–90 after ICI started were ALT ≥150 U/L, total bilirubin ≥3 mg/dL, diagnosis-coded hepatotoxicity, and diagnosis-coded hepatic failure. Overall survival (OS) was assessed for up to 5 years. A sensitivity analysis restricted HCV definition to lab-confirmed HCV RNA (no ICD-10 inclusion). Results: In the primary analysis, 1,157 HCV and 109,309 non-HCV ICI-treated patients were identified; 1,137 patients remained per group after matching. Patients with Hep C had a higher risk of ALT ≥150 U/L, although this did not reach statistical significance (7% vs 5%; p = 0.066). There was no difference in the risk of bilirubin ≥3 mg/dL (5.0% vs 5.0%; p = 1.0). Diagnosis-coded hepatotoxicity was similar (1.8% vs 1.8%; p = 0.875), while diagnosis-coded hepatic failure was higher in the HCV cohort (4.7% vs 2.5%; p = 0.005). Median OS was 20.1 vs 17.2 months (p = 0.0595). In the RNA-only sensitivity analysis (308 matched patients), there was no significant difference in overall survival between patients with and without HCV prior to treatment (27% vs 32%; p = 0.7489). However, the risk of ALT ≥150 U/L was significantly higher in the HCV group (10% vs 5%; p = 0.0058). There was no significant difference in mean bilirubin values after treatment (p = 0.0636). Conclusions: Pre-existing HCV was associated with an increased risk of transaminitis in patients receiving immunotherapy, but there was no statistically significant difference in biliary toxicity or median overall survival. There is also an approximate two-fold increased risk of hepatic failure. These real-world data highlight the importance of closely monitoring for hepatotoxicity in patients receiving immunotherapy and support ICI use in patients with HCV.
Artificial intelligence–assisted treatment decision-making for cancer multidisciplinary team meetings: A proof of concept study.
e13673 Background: Multi-Disciplinary Team Meetings (MDTMs) are “gold standard” in the UK Cancer Care continuum. Multiple, fragmented systems complicate MDTM workflows due to lack of data integration and system coordination. A proof-of-concept exploratory, comparative analysis of a multiple Large Language Model (LLM) benchmarked Oncology Intelligence Platform was conducted prospectively for treatment decision making support in the Breast Cancer MDTM. The aim was to identify the most sustainable and scalable, real-world LLM, i.e., with maximal accuracy, minimal hallucination rate, and manageable GPU (Graphics Processing Unit) usage. Methods: A retrospective validation dataset of 225 matched National Health Service England (NHSE) radiology and histopathology text reports from randomised, heterogenous, confirmed and/or suspicious breast cancer patients with a wide range of disease stage (I to IV/early to advanced) and objective disease characteristic findings was curated from 2018 to 2024. This was a mixture of primarily unstructured, structured, unimodal, multi-source data. 125/225 were radiology consisting of Ultrasound, Mammogram, MRI Breast, CT and Bone Scans whereas the rest 100 were histopathology and/or supplementary reports. OncoflowTM, an Artificial Intelligence (AI) powered Cancer MDTM Coordinator CoPilot tool, was implemented. This software platform deployed multiple LLMs synchronously for treatment matching, i.e., mapping the objective output of data extracted from reports to recommendations from clinician preferred knowledge sources.These involved clinical practice guidelines namely European Society of Medical Oncology (ESMO), National Institute for Health and Care Excellence (NICE), Breast Systemic Anti-Cancer Therapy (SACT) Protocols from the Clatterbridge NHS Cancer Centre. Results: 6 LLMs were benchmark tested in a sandbox environment. LLM 1 and 4 were Base/Foundational models, Domain Specific (Medical) and Proprietary whereas the rest were Instruction tuned, Open Source with general purpose and multilinguality. LLM 4 was Autoencoding (Encoder only) whereas the rest were Autoregressive (Decoder only). All these LLMs were fine-tuned and a proprietary data processing strategy was used to enhance the models towards task suitability. LLM 2 showed Guideline (ESMO/NICE) Accuracy rate of 92%, SACT Protocol Accuracy Rate of 100%, hallucination rate of 3.4% and GPU usage of 18GB, outperforming others as the most preferred. Conclusions: The rising cancer incidence and rapidly evolving evidence-based treatment decision making needs have resulted in unsurmountable MDTM pressures. The 10 year health plan envisions to “make the NHS the most AI-enabled health system” “with AI seamlessly integrated into clinical pathways”. This will in turn invite incorporation into real world clinical workflows via electronic health record integration.
Differential immune gene signatures to inform clinical course of patients with metastatic colorectal cancer.
e15592 Background: Patients with proficient mismatch repair (pMMR) colorectal cancer liver metastases (CRLM) constitute a distinct subgroup with poor survival and attenuated responses to immunotherapy. We hypothesized immune gene expression signatures may predict clinical outcomes. Using data from our single institution experience and from a national database we sought to define relationships between overall survival (OS) and gene expression profiles among patients with metastatic colorectal cancer (CRC). Methods: Using our institutional cohort of CRLM patients, we compared gene expression between patients in the lowest OS quartile (n = 9) and those alive at follow-up (n = 21). 15,605 metastatic pMMR CRC tumors underwent Next-Gen sequencing of DNA and RNA at Caris Life Sciences. CRLM (n = 10,600) were stratified into high- and low-risk subgroups based on OS quartiles to validate findings in the Emory data. Additional metastatic sites were lung (n = 2995) and peritoneal (n = 2010). Deconvolution of bulk tumor RNA expression was used to profile the tumor microenvironment (TME) using quanTIseq. Significance was calculated using X 2 /Fisher’s exact, or Mann-Whitney U with p-values adjusted for multiple comparisons. OS from time of tissue collection to last contact was estimated from insurance claims data using Cox proportional hazards models to calculate hazard ratio (HR) and log-rank tests to calculate p-values. Results: Among our institutional cohort, tumors from low-risk patients demonstrated signatures associated with increased immune enrichment and myeloid predominance. Among pMMR CRLM within the Caris cohort, 3,113 (29%) were considered low-risk and 4973 (47%) were considered high-risk. Of the top 40 differentially expressed genes identified between Emory high- and low-risk cohorts, 21 (52.5%) genes were validated by the Caris cohort. Among pMMR CRLM patients within the low-risk cohort, there were signatures denoting increased immune infiltration with a similar myeloid predominance. Patients with pMMR CRLM tumors had worse OS compared to lung mets (21 vs 35 m; HR 1.7, 95% CI: 1.6-1.8 p < .001) but better OS compared to peritoneal mets (21 vs 18 m; HR 0.8, 95% CI: 0.8-0.9, p < 0.001). Compared to CRLM TMEs, peritoneal mets had increased Tregs, B cells, M1-like, and M2-like macrophages (FC: 1.3-2.3, p < .001) but decreased neutrophils and NK cells (FC: 0.65-0.89, p < .001), while lung mets had increased Tregs, B cells and M2-like macrophages (FC: 1.2-2.5, p < .001). Conclusions: Results from a single institution cohort and a national dataset identify key differences in immune signatures between CRLM that may influence outcomes. These trends persisted based on CRC metastatic site of disease: CRLM are devoid of T cells while peritoneal mets harbor tumors with suppressive TME features. These distinct immune profiles may contribute to worse OS observed among peritoneal and CRLM as compared to lung mets.
A phase 1 dose-escalation and -expansion study of AFN0328, a novel mRNA-based immunotherapy, in patients with HPV16/18-associated high-grade squamous intraepithelial lesions.
5544 Background: Persistent infection with high-risk human papillomavirus (HPV), particularly HPV16 and HPV18, drives the development of high-grade squamous intraepithelial lesions (HSIL), an obligate precancerous condition. The standard care of cervical HSIL requires either surgical excision (LEEP or cold-knife conization), or in some circumstance, watchful waiting. AFN0328 is a novel mRNA-based immunotherapy encoding HPV16/18-related antigens, designed to elicit antigen-specific immune responses and interrupt viral antigen-driven disease progression. Methods: Safety, tolerability and preliminary efficacy of AFN0328 were assessed in patients with HPV16- and/or HPV18-associated HSIL (CIN2/3) in an ongoing, multicenter, open-label Phase 1 trial. The study comprised two sequential parts: Part 1 employed a classic 3+3 dose-escalation design to assess safety and determine the maximum tolerated dose (MTD); Part 2 is a dose-expansion cohort at selected doses (150 μg and 300 μg) to further characterize safety, tolerability, and preliminary efficacy. AFN0328 was administered intramuscularly at 0, 4, and 12 weeks. The primary endpoints included safety and tolerability (dose-limiting toxicities within 28 days post-first dose, adverse events graded per CTCAE v5.0). For Part 2, the efficacy endpoints included viral clearance rate at Week 24 and Week 36, and histopathological regression at Week 36. Secondary endpoints included pharmacokinetic parameters, HPV16/18-specific IgG antibody titers, and pharmacodynamic markers (cellular immune responses and serum cytokines). Results: As of December 29, 2025, 23 patients had received at least one dose of AFN0328. No DLTs, serious adverse events, or grade ≥3 treatment-related adverse events (TRAE) were observed. PK analyses of first and last doses showed that circulating mRNA encoding HPV-related antigens peaked within 48 hours and declined gradually, with detectable levels persisting for up to at least 4 weeks across dose levels. All evaluable patients developed specific IgG responses and viral clearance was observed in evaluable patients during follow-up. Conclusions: AFN0328 demonstrated favorable safety and tolerability with robust immune responses in HPV16/18-associated HSIL, supporting further clinical development as an immunotherapy for HPV-driven precancerous disease. Clinical trial information: CTR20244013. Summary of safety and immunogenicity outcomes. Outcome Result Patients treated 23 Dose-limiting toxicities 0 Grade ≥3 treatment-related AEs 0 Most common TRAEs Injection-site reactions, fever Specific IgG response rate 100%* Observed viral clearance Yes* *Observed in evaluable patients during interim follow-up.
Targeting metal ion transport in pancreatic cancer: A prognostic signature and GraphBAN-predicted therapeutics framework.
e16000 Background: Pancreatic ductal adenocarcinoma (PDAC) remains a lethal malignancy with limited therapeutic options. Dysregulation of metal ion homeostasis is implicated in tumor progression and therapy resistance. This study aims to systematically identify metal ion transport-related prognostic genes, construct a reliable risk model, and discover novel therapeutic agents using artificial intelligence. Methods: Transcriptomic data from GEO databases (GSE183795 as training set, GSE28735 as validation set) and single-cell RNA-seq data (GSE155698) were analyzed. Differential expression analysis, functional enrichment, and protein-protein interaction networks were constructed. A prognostic risk signature was developed using LASSO-Cox regression. The tumor immune microenvironment was characterized using CIBERSORT. An innovative GraphBAN model integrated with CNN, ESM, GCN, and ChemBERTa was employed for drug prediction, followed by molecular docking. Single-cell analyses including trajectory inference and cell-cell communication were performed. Results: We identified and validated a five-gene prognostic signature (SLC20A1, SLC39A10, SLC5A3, SLC11A1, SLC4A4) significantly associated with overall survival in PDAC. The risk model demonstrated robust predictive power in both training (3-year AUC = 0.73) and external validation cohorts (5-year AUC = 0.97). High-risk patients exhibited an immunosuppressive microenvironment characterized by M2 macrophage infiltration. GraphBAN prediction and molecular docking identified Elephantin and Sinularin as high-affinity binders to SLC39A10 and SLC4A4, respectively. Single-cell analysis revealed the specific expression dynamics of these genes in macrophage and neutrophil subpopulations and delineated their evolving roles along pseudotemporal trajectories. Conclusions: We established a novel metal ion transport-related gene signature as an independent prognostic indicator for PDAC. Our integrative AI-driven framework successfully predicted candidate drugs targeting this pathway, providing a promising strategy for personalized therapy and highlighting the therapeutic potential of modulating metal ion homeostasis in PDAC.
Closing the gap in HER2 testing: CEP17-low FISH results and NGS discordance.
e15192 Background: Discordance between HER2 amplification detected by fluorescence in situ hybridization (FISH) and next-generation sequencing (NGS) has been observed across tumor types, often in the context of chromosome 17 alterations. Low CEP17 control signal, due to monosomy 17, partial centromeric loss, truncation artifact or probe failure, can artifactually inflate the HER2/CEP17 ratio, resulting in FISH positivity in the absence of true ERBB2 gene amplification. The biologic underpinnings of FISH and NGS differ: FISH evaluates single cell nuclei, while NGS measures bulk tumor DNA and may be influenced by tumor purity, policy and heterogeneity. As HER2-targeted therapies expand across malignancies, accurate genomic classification is critical to avoid unnecessary toxicity in patients unlikely to benefit. Methods: We retrospectively identified patients with HER2 immunohistochemistry (IHC) 2+ tumors who underwent reflex HER2 FISH testing between 2020 and 2025 at a single institution. Cases with HER2 FISH-positive results were selected for analysis. NGS results were collected when available. Absolute HER2 copy number and CEP17 signal were analyzed. Anti-HER2 therapy use, disease progression, and cardiotoxicity were summarized using descriptive statistics. Results: Among 33 HER2 FISH-positive tumors, the majority were breast primaries (84.8%), with 15.1% gastroesophageal origin. NGS was available in 81.8% of cases (27/33), confirming HER2 amplification in only 11.1% (3/27), yielding a discordance rate of 88.9%. A low CEP17 signal was observed in 42.4% (14/33). Notably, absolute HER2 copy number was ≥6.0 in 42.4% of tumors, while 57.6% (19/33) met FISH positivity criteria despite HER2 copy number < 6.0, supporting CEP17-driven ratio inflation as a major contributor to discordance. Anti-HER2 therapy was administered in 75.8% (25/33) of patients; among treated patients, 36.0% experienced progression and 24.0% developed cardiotoxicity. Conclusions: This single-institution cohort reveals a high rate of discordance between HER2 FISH positivity and NGS-confirmed amplification in IHC 2+ tumors, often driven by artificially low CEP17 signals. A modest HER2 copy number represents a clinically relevant pitfall that may lead to HER2-directed treatment in patients without clear genomic amplification. These findings highlight the limitations of ratio-based FISH interpretation alone and support multimodal HER2 adjudication incorporating absolute HER2 copy number, low CEP17 signal, and NGS copy-number context. Prospective validation is warranted to refine HER2 testing algorithms and to guide HER2 targeted therapy decisions in discordant cases.
Response-focused analyses of the impact of circadian rhythm on efficacy of first-line immunotherapy in melanoma.
9520 Background: Several retrospective studies have associated better clinical outcomes with morning infusions of immune-checkpoint inhibitors (ICI), attributing the same to circadian differences in T-cell activity. However, these studies have largely focused on survival endpoints, which are susceptible to confounding by skewed cohort distributions, treatment heterogeneity or other unknown variables. In this study, we explore the impact of timing of ICI infusions by primarily focusing on objective response as a more direct and immediate measure of immunologic activity, and possibly less susceptible to confounding variables. Methods: We conducted a single center, retrospective study of patients with advanced melanoma with response-evaluable disease and treated with an ICI regimen in the first-line setting. We recorded baseline patient and tumor characteristics and the start time of the first ICI infusion. Based on the median start time (~1:00 PM) of all ICI infusions in our clinic, we classified patients into “Early” (before 1:00 PM) and “Late” cohorts (1:00 PM & after). We analyzed objective response rate (ORR) per RECIST v1.1, time-to-response (TTR), duration of response (DOR), progression-free survival (PFS), and overall survival (OS), comparing endpoints between cohorts using logistic or Cox regression, with adjustments for variables including age, sex, ECOG score, and disease stage. Results: We identified 216 eligible patients treated between 2014 - 2025, including 78 in the Early cohort and 138 in the Late cohort. Median age was 64 years (range 19 - 95) and 69% were male. Median follow-up duration was 27 months (range 6 days - 12 years). There were no significant differences in baseline patient and tumor characteristics between the two cohorts. ORR for the Early cohort was 64.1% (50/78; complete response [CR] - 34.6%, partial response [PR] - 29.5%) and for the Late cohort was 61.6% (85/138; CR - 28.3%, PR - 33.3%) [p = 0.71]. There were no significant differences in TTR and DOR between the cohorts. The distribution of first ICI infusion start-time in the responders mirrored that of all ICI infusions in all patients with melanoma. Adjusted PFS and OS were also similar between the cohorts (see Table). Conclusions: Our response-focused analyses do not suggest an association of early timing of ICI infusion to better efficacy outcomes, in contrast to findings from other studies. Our findings are consistent with the known long terminal half-life of ICI agents, which should provide steady availability of ICI antibodies to immune cells throughout the day. Our results do not support implementing time-based ICI infusion strategies in the oncology clinics. Endpoint Early cohort Late cohort Hazard ratio (95% CI) P-value PFS, Median months (95% CI) 12 (9, NR) 16 (9, 28) 1.01 (0.68, 1.50) 0.97 OS, Median months (95% CI) 82 (66, NR) 88 (62, NR) 1.07 (0.61, 1.89) 0.82 CI = Confidence interval; NR = Not reached.
Metastatic breast cancer-related mortality with thoracic involvement among U.S. adults aged ≥ 25 years: A retrospective CDC WONDER analysis from 1999 to 2024.
e13094 Background: Breast cancer is one of the most commonly diagnosed malignancies and a leading cause of cancer-related mortality in the United States, with metastatic disease responsible for most deaths. The disease commonly spreads via hematogenous and lymphatic pathways to the thoracic compartment, particularly the lungs and pleura. Thoracic involvement is associated with severe, often fatal complications including malignant pleural effusion, lymphangitic carcinomatosis, and respiratory failure. However, long-term national mortality trends related to thoracic involvement in metastatic breast cancer remain poorly defined. This study evaluates long-term national mortality trends associated with thoracic involvement in metastatic breast cancer among U.S. adults aged ≥25 years from 1999 to 2024. Methods: We conducted a retrospective analysis using the CDC WONDER multiple cause-of-death database from 1999 to 2024. Deaths among adults aged ≥25 years with breast carcinoma (ICD-10: C50) and concurrent lung or pleural metastases (C78.0, C78.2) were identified. Age-adjusted mortality rates (AAMRs) per 100,000 population were calculated and stratified by age, sex, race/ethnicity, U.S. Census region, and urbanization status. Temporal trends were assessed using Joinpoint regression to estimate annual percent change (APC) and average annual percent change (AAPC). Results: From 1999 to 2024, a total of 79,749 deaths were attributed to metastatic breast cancer with thoracic involvement. Overall mortality showed a non-significant decline trend (AAPC: −0.39) with temporal variability. Females had markedly higher mortality than males (AAMR: 2.39 vs 0.03), with a significant decline from 1999 to 2008 (APC: −4.23). Among age groups, adults aged ≥65 years bore the highest mortality burden, with a significant increase between 2013 and 2017 (APC: 5.47). By race/ethnicity, Non-Hispanic Asian individuals demonstrated an overall increasing trend (AAPC: 1.20), with a sharp rise from 2013 to 2018 (APC: 8.08). Regionally, mortality declined across all U.S. Census regions, most prominently in the Northeast (AAPC: −1.61). Both metropolitan and non-metropolitan areas showed declining trends, with the steepest reduction in metropolitan areas from 1999 to 2008 (APC: −4.23). Conclusions: Mortality related to thoracic involvement in metastatic breast cancer shows substantial temporal and demographic heterogeneity in the United States. Early declines have plateaued, with recent increases observed after 2017, particularly among older adults and non-Hispanic Asian females. These findings suggest a potential reversal of prior gains and highlight persistent geo-demographic disparities, emphasizing the need for improved surveillance, earlier detection, and equitable multidisciplinary care.
Promise or predicament? Real-world analysis of molecular targets and therapy approaches in uro-oncology.
e17022 Background: Despite ongoing research and evolving therapeutic landscapes, advanced urological malignancies such as prostate cancer (PC), bladder/urothelial cancer (BC), renal cell carcinoma (RCC), and penile cancer (PeC) remain associated with an unfavorable prognosis. While molecularly targeted therapies in PC are already clinically established, their use in other urological entities remains limited. The aim of this study was to analyze the real-world implementation of molecular diagnostics and its therapeutic consequences in a single-center tertiary setting. Methods: In a retrospective database analysis, all patients presenting at our university-based interdisciplinary tumor board (ITB) between 2018 and 2024 were included. Cases with a recommendation for molecular diagnostics were identified and categorized according to tumor entity. Molecular analyses were performed using disease-specific or pan-oncological next-generation sequencing (NGS) panels. Treatment recommendations were evaluated by the Molecular Tumor Board (MTB). Molecular data (including mutations, therapy recommendations, level of evidence, and administered molecular therapies), were analyzed. Results: Out of a total of 11,562 cases discussed in the ITB, molecular diagnostics followed by presentation at the MTB were recommended in 332 (2.9%) cases. Molecular analyses were actually performed in 113 (29.4%, overall cohort: 1.5%) patients, including 64 PC, 24 BC, 22 RCC, and three PeC patients, as well as 66 cases with isolated BRCA testing. The most frequent genetic alterations were PTEN in PC, HER2 and FGFR in BC, VHL in RCC, and EGFR in PeC. BRCA1/2 mutations were detected in 13.6%. Molecularly targeted therapies were initiated in 14 patients including PARP inhibition. Conclusions: Molecular diagnostics is gaining increasing importance in urological oncology; however, in routine clinical practice it rarely translates into therapeutic consequences. BRCA testing and PARP inhibition were the most common treatment. These findings highlight the need for structured molecular testing, prospective data collection, and further development of evidence-based treatment options to better exploit the potential of personalized therapies in uro-oncology.
Real-world outcomes of cabazitaxel plus carboplatin versus cabazitaxel alone in metastatic castration-resistant prostate cancer.
e17044 Background: Cabazitaxel is an established treatment for metastatic castration-resistant prostate cancer (mCRPC). In real-world practice, carboplatin is often added for aggressive disease phenotypes despite limited comparative effectiveness data. We evaluated the real-world effectiveness and safety of cabazitaxel plus carboplatin versus cabazitaxel alone. Methods: We conducted a retrospective comparative effectiveness study using the TriNetX US Collaborative Network, including data from 69 U.S. healthcare organizations. Adult men (≥18 years) with mCRPC treated with cabazitaxel were identified and categorized into cabazitaxel plus carboplatin or cabazitaxel alone cohorts. The index date was defined as first exposure to cabazitaxel, with outcomes assessed from 90 days post-index. Patients with hematologic malignancies were excluded. Propensity score matching (1:1) was performed using demographics and key comorbidities. Outcomes included PSA response, all-cause mortality, treatment-related toxicities, and healthcare utilization. Results: After propensity score matching, 1,618 patients were included (809 per cohort), with a median age at index of 68 years in both groups. The matched population was racially and ethnically similar between cohorts, comprising approximately 74% White patients, 15% Black or African American patients, and 7% Hispanic or Latino patients (approximately 81% non-Hispanic). Baseline comorbidity burden was comparable after matching, including hypertension (~62%), chronic kidney disease (~14%), and heart failure (~7%) (all p > 0.05). Patients treated with cabazitaxel plus carboplatin were significantly more likely to achieve a deep biochemical response, defined as PSA < 4 ng/mL, compared with cabazitaxel alone (14.5% vs 9.5%; risk ratio 1.52; p = 0.002). In contrast, rates of partial PSA response (PSA 4–20 ng/mL) were similar between groups (19.2% vs 17.3%; p = 0.33), indicating that the PSA benefit was driven by deeper PSA suppression rather than modest PSA reductions. All-cause mortality was comparable between cohorts (65.4% vs 65.5%), with no significant difference in overall survival. Combination therapy was associated with higher rates of hematologic toxicity, including bone marrow suppression (54.0% vs 45.0%; p < 0.001), anemia (71.7% vs 65.4%; p = 0.006), and thrombocytopenia (33.6% vs 26.0%; p = 0.001). Rates of fatigue (41.8% vs 39.7%), hospital admissions (44.6% vs 42.8%), and analgesic utilization (90.5% vs 89.0%) were similar between treatment groups. Conclusions: In this large real-world cohort, adding carboplatin to cabazitaxel was associated with higher rates of deep PSA response but no overall survival benefit and increased hematologic toxicity. These findings highlight the need for improved patient selection and prospective studies to clarify the clinical role of this combination in mCRPC.
The HOUSES index: A novel, patient-level measure of socioeconomic status linking real-world housing data to overall survival in metastatic pancreatic ductal adenocarcinoma.
e13719 Background: Previous research has demonstrated disparities in survival outcomes for patients with pancreatic ductal adenocarcinoma (PDAC), related to social determinants of health (SDoH). A major limitation of cancer disparities research is the reliance on census-level measures of socioeconomic status (SES). These measures ignore individual-level variability within the same zip code, risking ecological fallacy and limiting clinical applicability. In contrast, the Housing-Based Socioeconomic Status (HOUSES) index is a patient-level validated measurement of SES. This study aims to assess the relationship of HOUSES with overall survival (OS) amongst patients with metastatic pancreatic ductal adenocarcinoma (mPDAC). Methods: We retrospectively identified 189 patients with a diagnosis of mPDAC between 2019-2024 from the institutional database of a high-volume academic center. Supplemental data was abstracted from the electronic medical record (EMR). The HOUSES index was calculated using each patient’s home address at the time of diagnosis to geocode housing characteristics obtained from government assessor offices. These housing features include assessed value, square footage, and number of bedrooms and bathrooms. OS was defined as the time between diagnosis and the date of death or last follow up on 1/1/2026. A multivariable Cox model was performed adjusting for age, sex, tobacco use, ECOG performance status, income, insurance, cancer treatment, national-percentile area deprivation index, and rural-urban residence. Results: Of the 189 patients, the median age was 67 (IQR 58-73), 52.9% were male, and 92.1% were self-identified non-Hispanic white. The median OS was 10.3 months (IQR 4.9-22.9), and 85.3% received cancer directed treatment. Compared to the lowest SES Q1, higher HOUSES quartiles had significantly lower hazard ratios (HR), HOUSES Q2 (HR: 0.58 [95%-CI: 0.39–0.85]), Q3 (HR: 0.51 [95%-CI: 0.35–0.75]), and Q4 (HR: 0.64 [95%-CI: 0.45–0.89]), demonstrating a protective effect on OS. Other factors independently associated with OS included Medicare compared to commercial insurance (HR: 0.18 [95%-CI: 0.10–0.32]), age (HR: 1.04 [95%-CI: 1.02–1.06]), and ECOG performance status (HR: 2.14 [95%-CI: 1.31–3.50]). Conclusions: Our study demonstrates that mPDAC patients in higher SES quartiles of the HOUSES index have a longer OS when compared to the most disadvantaged quartile. These trends exist while controlling for other SDoH variables, demonstrating that HOUSES provides additional information on patient SES and health barriers. Overall, these results demonstrate the potential of HOUSES as a novel, patient-level sensitive measure of SDoH that can be valuable for both cancer research, and in clinical practice as a method to identify patients in need for further support.
Lambert-Eaton myasthenic syndrome in United States veterans with small cell lung carcinoma (SCLC).
e20148 Background: Lambert-Eaton Myasthenic Syndrome (LEMS) is a rare autoimmune disorder characterized by muscle weakness, fatigue, and autonomic dysfunction. 50-60% of LEMS cases present as paraneoplastic, mostly with SCLC. US Veterans have higher rates of SCLC incidence due to increased tobacco use and exposure to service-related carcinogens. The true prevalence, survival rate, and treatment of veterans with SCLC and LEMS remains uncertain, further research is necessary for improving clinical recognition of LEMS and to provide critical insights into the burden of this paraneoplastic condition in a high-risk population. Methods: Veterans with SCLC between October 1, 1999 and December 31, 2024 were retrospectively identified from the national Veterans Affairs (VA) Corporate Data Warehouse by ICD-O-3 codes. ICD-9 and ICD-10 codes for LEMS, Myasthenic Syndrome, or myoneural disorders were used to identify patients with a probable diagnosis of LEMS. The presence of LEMS was confirmed in patients via text matching in hematology, oncology, or neurology clinic notes. We used a binomial logistic regression to analyze the frequency of LEMS in SCLC patients over 5-year time periods. The overall survival of patients with SCLC compared to patients with SCLC and LEMS was analyzed using Cox proportional hazards ratio (HR) and Kaplan-Meier survival model. Results: We identified 25,953 distinct patients with SCLC for the period queried. Our screening method yielded TIU notes for 80 SCLC patients. We could confirm a diagnosis of LEMS in 71 cases, probable LEMS in 6 cases, and uncertain LEMS in 3 cases. The crude total count (confirmed plus probable cases) of LEMS was 2.97 per 1000 veterans with SCLC. Amifampridine (3,4-Diaminopyridine) was involved in the treatment for 7 of the 71 confirmed LEMS cases while 42 of these cases were treated with pyridostigmine. There was not a significant correlation between LEMS diagnosis and the 5-year time period frequency rate. The Cox HR for SCLC-LEMS to SCLC was 0.72 (95% CI: 0.55-0.93). Overall survival of patients in SCLC was significantly higher for the group with LEMS (p-value = 0.011). Conclusions: Among US veterans with SCLC, LEMS is a rare disorder consistently observed over the past 25 years with a slight negative, non-significant incidence trend. The majority of LEMS patients were treated with pyridostigmine rather than amifampridine. SCLC patients with LEMS exhibited improved survival compared to those with SCLC alone. LEMS 5-year time period frequency. Dates Number of SCLC Patients Number of confirmed LEMS cases (Percentage) 1999 - 2005 6397 21 (0.33%) 2006 - 2010 5563 19 (0.34%) 2011 - 2015 5583 13 (0.23%) 2016 - 2020 4980 14 (0.28%) 2021 - 2024 3435 4 (0.12%) Total 25953 71 (0.27%)
Indirect comparison of the effectiveness of tepotinib and immunotherapy in NSCLC with METex14 skipping alterations: Final propensity-scored analysis of eight pooled real-world datasets (TOGETHER) vs the VISION trial.
e20751 Background: Tepotinib demonstrated clinical efficacy in NSCLC patients with MET ex14 skipping alterations in the single-arm VISION trial. Comparative effectiveness estimates against the contemporary therapeutic option of immunotherapy (IO) with or without chemotherapy (CT), were generated using external data. Methods: Eight international real-world datasets, which included patients who started treatment between 2010 and 2022 were pooled to form a real-world cohort. Inclusion and exclusion criteria based on VISION were applied to ensure generalizability and propensity scoring used to address differences in observed characteristics, stratifying by previous treatment status. Variables used were mean age, advanced vs. metastatic disease, sex, adenocarcinoma histology, and smoking history. Reweighted data were used to facilitate comparisons of real-world progression-free (PFS) and overall survival (OS) outcomes between therapy regimens in the first line (1L), second or later line (2L+), or line agnostic setting (LAs). Results: TOGETHER yielded information on 45 and 24 patients receiving IO or IO+CT in 1L, 83 and 6 patients in 2L+, respectively. In LAs 146 patients received IO±CT. TOGETHER patients were reweighted to match baseline characteristics of patients in VISION. As shown in the table (survival time in months), in 1L median (m) PFS was longest and the 24-month survival proportion was highest for tepotinib, compared to IO and IO+CT. In 2L+, due to the single digit sample size for those receiving IO+CT, only IO and IO±CT could be compared, with tepotinib showing longer mPFS and higher 24-month PFS. Due to the effects of prior or subsequent treatment lines which cannot be accounted for, 1L and 2L+ OS estimates are confounded; this notwithstanding, analyses of LAs showed the following: mOS for tepotinib was 19.2 months, compared to 19.0 for IO and 16.7 for IO±CT. Conclusions: Although MET ex14 skipping alterations are rare in NSCLC, pooling real-world datasets provided sufficient patient numbers to allow for comparative effectiveness estimates. In propensity score weighted analyses, tepotinib demonstrated longer PFS versus IO, as monotherapy or in combination with CT. The short PFS across comparators underlines the poor prognosis for patients with existing therapies, while also confirming tepotinib as an effective treatment option in this rare, biomarker-driven NSCLC subtype. Tepotinib IO IO+CT IO±CT 1L n 164 45 24 67 mPFS [95% CI] 8.7 [8.2, 12.6] 5.2 [2.7, 12.4] 8.0 [3.0, 17.5] 4.3 [2.7, 9.7] 24-month PFS 30% 13% Not reached 8% 2L+ n 149 83 6 88 mPFS [95% CI] 8.3 [7.0, 11.0] 4.8 [3.2, 7.7] - 4.8 [3.2, 7.7] 24-month PFS 19% 12% - 11% Line agnostic n 313 121 29 146 mOS [95% CI] 19.2 [16.3, 22.3] 19.0 [14.9, 22.4] 13.6 [12.3, 23.1] 16.7 [13.7, 21.7] 24-month OS 41% 33% 7% 32%
Cladribine/cyclophosphamide lymphodepletion efficacy and toxicity in CD19 CAR-T therapy for B-ALL.
6549 Background: During an international fludarabine (Flu) shortage, our center adopted cladribine (Cla) as a substitute for Flu with cyclophosphamide (Cy) for lymphodepletion (LD) prior to CAR-T therapy in patients with relapsed/refractory (R/R) B-cell acute lymphoblastic leukemia (B-ALL). We evaluated clinical outcomes and toxicity with Cla/Cy LD. Methods: We conducted a retrospective, single-center analysis of R/R B-ALL patients who received CD19-directed CAR-T therapy between January 2018 and April 2025. The Cla/Cy regimen substituted Flu 30 mg/m² with Cla 5 mg/m²/day on days −5 to −3. ALL-HT risk was calculated (Nair et al, 2025). CRS and ICANS were graded per ASTCT; cytopenias per CTCAE v5.0. Early (<30 days) and late (31–100 days) infections were assessed. Overall survival (OS), Event free survival (EFS, event=progression or death) and Minimal residual disease (MRD) status by clonoSEQ assay at Day 30 bone marrow biopsy were assessed. Results: We analyzed 53 patients (brexu-cel n=49; tisa-cel n=4). 28 received Flu/Cy and 25 received Cla/Cy. Baseline characteristics are outlined in Table 1. Three patients died before day 30 and two before day 100. Ten patients underwent allo-SCT within 6 months of CAR-T. Rates of grade ≥3 CRS (4% vs 12%, p=0.29) and ICANS (40% vs 24%, p=0.25) did not differ. MRD negativity at day 30 was comparable (Cla/Cy 81.8% vs Flu/Cy 75.0%, p=0.71). Early infection rates were similar; however, late infections were more frequent with Flu/Cy (48% vs 17.4%, p=0.034), with a persistent trend after ALL-HT adjustment (adjusted OR 3.54, p=0.073). Grade ≥3 thrombocytopenia (35.7% vs 8.0%, p=0.022) and neutropenia (25.0% vs 4.0%, p=0.05) at day 90 were higher with Flu/Cy, with similar trends after adjustment (adjusted OR 4.71, p = 0.082 and adjusted OR 7.88, p = 0.067, respectively). Neutrophil recovery (ANC >1000) time was comparable (median 17 vs 18 days, p=0.10). With a median follow-up of 14.6(Flu/Cy) and 13.1 months(Cla/Cy), EFS (6.8 vs 18.1 months, p=0.45) and OS (33.9 months vs not reached, p=0.25) were similar. Conclusions: Cla/Cy demonstrated comparable efficacy to Flu/Cy with lower toxicity, including fewer late infections and severe cytopenias. These findings suggest Cla/Cy may represent a safer alternative lymphodepletion strategy. Prospective studies are warranted to validate these observations. Baseline characteristics of study cohort. Characteristic Flu/Cy (n=28) Clad/Cy (n=25) p-value Age, median (IQR) 38.0 (30.2–58.5) 47.0 (26.0–64.0) 0.77 Male sex, n (%) 18 (64.3) 13 (52.0) 0.53 White race, n (%) 25 (89.3) 20 (80.0) 0.35 KPS ≥80, n (%) 24 (85.7) 21 (84.0) 1.00 Marrow blasts pre-LD, median (IQR) 2.0 (1.0–3.5) 1.5 (0.5–11.0) 0.67 Ph+ ALL, n (%) 11 (39.3) 12 (48.0) 0.78 Prior allo-HSCT, n (%) 6 (21.4) 7 (28.0) 0.56 Prior lines of therapy, median (IQR) 2 (2–3) 3 (2–3) 0.96 High-risk ALL-HT, n (%) 12 (42.9) 3 (12.0) 0.016
Cancer-associated left ventricular thrombus in hospitalized oncology patients: National burden and inpatient outcomes.
e23230 Background: Left ventricular thrombus is an uncommon, but high-risk cardiac finding that may reflect acute cardio-oncologic decompensation during hospitalization. Contemporary national data describing its inpatient burden and associated outcomes in hospitalized cancer populations are limited. This study evaluated the prevalence of left ventricular thrombus and its association with inpatient outcomes and resource utilization during cancer hospitalizations. Methods: A serial cross-sectional analysis was conducted using adult hospitalizations with a principal diagnosis of malignancy in the 2018 to 2023 Healthcare Cost and Utilization Project National Inpatient Sample with discharge-level survey weighting. Left ventricular thrombus was identified using ICD-10-CM code I51.3 in any diagnosis field. Outcomes included in-hospital mortality, mechanical ventilation (ICD-10-PCS 5A1935Z, 5A1945Z, 5A1955Z), shock, all types (R57*), stroke (I63 or I61), pulmonary embolism (I26*), length of stay, and hospitalization cost using cost-to-charge ratios. National estimates accounted for survey stratification and clustering. Survey-weighted multivariable regression adjusted for demographics, year, payer, ZIP-code income quartile, elective status, hospital teaching status, region, and All Patient Refined Diagnosis Related Group severity. Results: Among cancer hospitalizations, left ventricular thrombus prevalence was 0.13%, corresponding to approximately 7,620 hospitalizations nationally from 2018 to 2023. In-hospital mortality was higher among hospitalizations with left ventricular thrombus compared with those without (9.12% vs 4.36%). Left ventricular thrombus was associated with higher rates of mechanical ventilation (5.58% vs 2.57%), shock (4.72% vs 1.47%), stroke (5.77% vs 1.57%), and pulmonary embolism (14.83% vs 2.03%). Resource utilization was greater, with longer length of stay (9.73 vs 6.86 days) and higher mean hospitalization cost ($40,064 vs $30,537).After weight adjustment, left ventricular thrombus was not independently associated with in-hospital mortality (adjusted odds ratio 1.15, 95% CI 0.70 to 1.87) but remained independently associated with shock (adjusted odds ratio 2.45, 95% CI 1.32 to 4.52). Conclusions: Left ventricular thrombus is an uncommon but clinically meaningful inpatient finding among hospitalized cancer patients and is associated with increased intubation, thromboembolic complications, and resource utilization. The persistent association with shock supports left ventricular thrombus as an important prognostic marker for hospitalized cancer patients leading to significant morbidity and mortality.
Healthcare system gaps in access to HER2-directed chemotherapy in early-stage HER2-positive breast cancer: A scoping review.
e13031 Background: HER2-directed targeted treatment with Trastuzumab and chemotherapy has been the standard therapy for early-stage HER2-positive breast cancer, increasing survival rates. However, many studies have shown gaps in disease management and care of the disease. Many eligible patients do not receive this therapy due to healthcare system-related factors such as type of treatment facility, insurance limitations, referral timings, and locational convenience. Some studies have shown that the underuse of HER2-directed targeted treatment with chemotherapy can be correlated with lack of care coordination, especially for economically disadvantaged populations. Acknowledging and understanding these limitations in management is essential to provide proper and equal healthcare amongst all patients. Methods: This scoping review looked at studies with the date range of 2005-2026 and found in PubMed, EMBASE, and Google Scholar, when Trastuzumab became the standard of care for early-stage HER2-positive breast cancer. Inclusion criteria included studies examining health system factors that influenced initiating, receiving, and completing HER2-directed management among patients that were diagnosed with early-stage HER2-positive breast cancer. We synthesized findings narratively due to inconsistencies in methodology and reported outcomes. Results: Twenty studies that met the inclusion criteria were reviewed. One study illustrates that out of 418 patients, one third of the prescriptions were filled with financial help and among 48 prescriptions never filled, one third was due to being unable to afford the therapy. Additionally, delayed oncology referrals worsen the outcomes of early-stage breast cancer therapy. Another study illustrates that waiting more than 12 weeks to initiate Trastuzumab had worse overall survival rates while another study explains waiting more than 6 months increased the chance of relapse. Additionally, women with lower annual incomes were more likely to discontinue therapy compared to women making an annual income of $50,000 or more. Collectively, the studies reviewed highlight that lower sociodemographic status, insurance and financial barriers, and delayed oncology referrals can negatively impact access to treatment for early-stage HER2-positive breast cancer. Conclusions: Health system barriers limit the care and access to HER2-directed targeted treatment. Changes that focus on care management, decreased delay in referrals, more geographical access to treatment centers, and decreased insurance-related delays can improve delivery of HER2-directed chemotherapy. Selection bias due to used keywords and inclusion criteria, and findings being reflective only of studies found in the mentioned bibliographic databases are the main limitations of this review.
Development of a shared neoantigen peptide library to promote personalized vaccine strategy in patients with non–small cell lung cancer.
e20624 Background: To establish and validate a shared neoantigen peptide library based on NSCLC hotspot mutations for efficient personalized neoantigen identification and to explore its application in combination therapy for EGFR-TKI-resistant patients. Methods: A peptide library was constructed from hotspot mutations in key NSCLC genes (EGFR, TP53,KRAS,CTNNB1,CDKN2A,HER-2 ) and common Chinese HLA types using NetMHCpan for epitope prediction. Tumor mutations were identified via NGS/RT-PCR, and HLA genotyping was performed. PBMCs from 51 matched patients were stimulated with corresponding peptides. T-cell activation (IFN-γ, CD137) and cytokine profiles were assessed by ELISPOT/flow cytometry, and neoantigen-specific CTL cytotoxicity was confirmed. A clinical trial evaluated the safety and efficacy of the shared neoantigen vaccine combined with chemotherapy and PD-1 blockade in advanced, TKI-resistant NSCLC. Results: The library comprised 44 high-affinity peptides. Immunogenicity testing identified 29 immunogenic peptides inducing IFN-γ/CD137 expression and specific target cell lysis. In 7 enrolled TKI-resistant patients, the combination therapy resulted in 1 complete response, 3 partial responses, and 3 stable disease (DCR = 100%). Median PFS was 7.5 months and median OS was 23.6 months. Vaccine-related adverse events were manageable (e.g., injection-site reactions, fatigue). Conclusions: The validated shared neoantigen peptide library provides a novel platform for rapid neoantigen discovery and represents a safe and effective combination therapy strategy for NSCLC patients with acquired TKI resistance. Clinical trial information: NCT06095934 .
Low-dose nivolumab in patients at high risk of immune-related adverse events: A retrospective cohort study.
2656 Background: Immune checkpoint inhibitors (ICIs) are relatively contraindicated in patients with prior severe immune-related adverse events (IRAEs) and/or have pre-existing serious auto-immune diseases (AID). Such patients are generally excluded from immunotherapy trials due to toxicity concerns, and data on ICI (re)exposure is limited. Our correlative data [Tachiki L 2024 SITC] suggest that low-dose (LD) nivolumab (Nivo; 40 mg) achieves PD-1 receptor occupancy comparable to standard-dose (SD) Nivo (240 or 480 mg), suggesting a potentially similar risk of developing IRAEs. However, faster serum clearance observed with LD Nivo may allow easier management of IRAEs when they occur. Based on this rationale, we have offered LD Nivo to high-risk patients after careful clinical discussion. In this study, we present the outcomes of our institutional experience. Methods: This single institution, retrospective cohort study included patients with advanced skin cancers who received LD Nivo (40 mg) due to a high-risk of IRAEs, including those with a previous history of IRAEs from SD ICIs or ICI-naïve patients with pre-existing AID. We analyzed efficacy and safety endpoints, including best objective response rates (BORR) per RECIST v1.1 and rates of IRAE with SD and LD ICIs. Outcomes were analyzed using descriptive statistics and stratified Cox models. Results: From 2015 – 2025, 23 patients with advanced skin cancers (22-Melanoma; 1-Merkel cell carcinoma) received LD Nivo due to an elevated IRAE risk. Sixteen patients had a history of treatment-limiting IRAEs on SD ICIs (“Prior-IRAE” cohort: 7 SD anti-PD-1 monotherapy; 9 SD combination ICI), and 7 patients were ICI-naïve with pre-existing AID (“AID” cohort). In the Prior-IRAE cohort, treatment hold/discontinuation due to IRAEs occurred in 100% with SD ICI versus 43.8% with LD Nivo (p = 0.032). Among the 7 patients previously treated with SD anti-PD-1 monotherapy, the incidence of grade ≥2 IRAEs was 100% (median duration 75 days; range, 12–217) on SD therapy versus 42.8% (median duration 43 days; range, 23–146) on LD Nivo. In the AID cohort, 100% patients experienced at least one grade 2 IRAE, but no grade 3 or 4 events were observed. In patients with evaluable disease (measurable and progressing at LD Nivo initiation), BORR was 57.1% (4/7; 1 complete response [CR], 3 partial responses) in the Prior-IRAE cohort and 33.3% (2/6; both CR) in the AID cohort. Conclusions: LD Nivo demonstrates biological activity, as evidenced by both anti-tumor responses and toxicities, in patients at high risk of IRAEs. Compared to SD ICI, LD Nivo may offer an advantage in toxicity management, potentially enabling more consistent treatment delivery with fewer interruptions and reduced need for immunosuppressive interventions. This strategy warrants further evaluation in prospective clinical trials.