Browse Articles
Discover research articles across all indexed journals
Developing quality cancer survivorship care competencies for primary care: A modified Delphi study.
e13780 Background: Most of the growing population of people living with and beyond cancer will access primary care. However, the aspects of cancer survivorship care that primary care clinicians should deliver have not been clearly defined. In 2016, the American Society of Clinical Oncology (ASCO) Core Curriculum for Cancer Survivorship outlined competencies, though these were not specifically targeting the primary care workforce. We aim to define primary care-specific cancer survivorship competencies. Methods: In 2025, the ASCO Core Curriculum was reviewed by the Education Workgroup of the National Cancer Institute’s Interdisciplinary Network for Survivorship and Primary Care Research and Education (INSPiRE). This workgroup, comprising 18 survivorship experts from oncology, primary care, and lived experience, met for one-hour monthly virtual meetings from January to June 2025. The workgroup discussed 11 topics of survivorship care: (1) survivorship; (2) surveillance; (3) treatment effects; (4) chronic disease management; (5) health promotion and disease prevention; (6) psychosocial care; (7) survivors of childhood, adolescent, and young adult cancers; (8) older adults with cancer; (9) survivors living with incurable, advanced, or metastatic cancers; (10) caregivers of survivors; and (11) communication and coordination of care. Members reviewed competencies within these topics, with suggestions for removing, modifying, or adding competencies to be primary care-specific. Transcripts and meeting notes were reviewed, and two authors drafted competency statements that were circulated to the full workgroup for further refinement. To reach consensus, we are conducting a two-step Delphi process to assess the strength of the workgroup’s 52 proposed primary-care specific survivorship competencies by contacting members of INSPiRE, the Society of General Internal Medicine Cancer Research Interest Group, and the North American Primary Care Research Group Cancer Special Interest Group (approximately 270). Results: Available results from the forthcoming Delphi process will be presented at the meeting. Conclusions: Competencies in cancer survivorship tailored for primary care are a critical step needed to train current and future primary care clinicians and thus equip primary care to provide quality cancer survivorship care to the increasing population of survivors.
A real-world analysis of dyslipidemia and attributable risk in lorlatinib-treated ALK-positive lung cancer patients.
e24190 Background: Lorlatinib as a first-line treatment for ALK-positive NSCLC can significantly prolong survival, and the most common adverse reaction is hyperlipidemia, and relevant management data are lacking. This study aims to identify risk factors for poor lipid control in the real world and provide an evidence-based basis for clinical management. Methods: This study retrospectively analyzed patients with stage III-IV ALK-positive advanced NSCLC who received lorlatinib treatment across multiple centers between May 1, 2022, and April 30, 2025. Data on patient baseline characteristics, blood lipid levels (TG,TC,LDL-C) before and after treatment, and lipid-lowering medication use were collected, with further attributable risk analysis of factors associated with poor lipid control. Results: This study enrolled a total of 306 patients from 12 participating centers. Baseline characteristics are detailed in Table 1. Within 9 months of treatment, patients receiving lorlatinib exhibited a marked abnormal elevation in lipid levels (Figure 1), with no significant improvement observed through the 36-week follow-up endpoint, regardless of lipid-lowering medication use (Figure 2). These findings indicate that conventional lipid-lowering interventions fail to effectively ameliorate lorlatinib-associated hyperlipidemia, highlighting an urgent need to optimize clinical management strategies for this drug-induced dyslipidemia. Poor medication adherence was the strongest risk factor, with a Population Attributable Risk Percentage (PAR%) of 26.5%, Attributable Risk Percentage (ARP) of 37.3%, and a population exposure rate of 60.5%. Lack of regular follow-up (patient non-compliance) was the second strongest risk factor (PAR% of 9.2%), with a 100% poor lipid control rate among patients who did not attend follow-up visits. The protective effect of smoking history requires further validation with a larger sample size. Conclusions: This study indicates that poor adherence to lipid-lowering medications and lack of regular follow-up are the most important modifiable risk factors for poor control of hyperlipidemia in lung cancer patients treated with lorlatinib. Addressing these factors may significantly improve lipid management in this population. Patient characteristics. Characteristics Sex Male 144（47.05%） Female 162（52.95%） Age <65 234（76.5%） ≥65 72（23.5%） Smoke Yes 53（17.3%） No 253（82.7%） ECOG 0-1 237（77.5%） ≥2 69（22.%） Health History Diabetes 53（17.3%） Hypertension 40（13.1%） Hyperlipidemia 35（11.4%） Lipid-Lowering Drugs Yes 41（13.4%） No 265（86.6%） Histology Adenocarcinoma 4（1.3%） Squamous cell carcinoma 287（93.8%） Other 15（4.9%） TNM Staging III 19（6.2%） IV 287（93.8%） Treatment Lines 1line 141（46.%） 2line 62（20.3%） 3line 37（12.1%） ≧3line 66（21.6%）
Correlative analysis of ctDNA in CheckMate-9X8 study of frontline treatment with mFOLFOX6, bevacizumab with or without nivolumab in metastatic colorectal cancer.
3584 Background: The CheckMate 9X8 study evaluated ctDNA as biomarker to predict treatment benefit in metastatic colorectal cancer (mCRC) patients receiving standard-of-care therapy with or without nivolumab. Although the phase 2/3 trial did not meet its primary PFS endpoint, the addition of nivolumab appeared to confer clinical benefit in a subset of patients. As predictive biomarkers for first-line nivolumab in mCRC remain unvalidated, this exploratory analysis identified mutation- and methylation-based ctDNA signatures that may help inform future patient stratification and disease management. Methods: Study design and methods were published previously (Lenz, et al. JITC 2024). As part of an exploratory biomarker analysis, paired plasma samples (baseline and on-treatment at C3D1) were evaluated for circulating tumor DNA (ctDNA) using PredicineATLAS (600-gene) comprehensive genomic profiling, PredicineSCORE (Low-pass whole-genome sequencing), and PredicineEPIC (methylation) assays and correlated with baseline patient characteristics and treatment efficacy. Results: Over 98% (138/140) of baseline patients had quantifiable ctDNA with Predicine ATLAS and approximately 83% (116/140) had detectable copy number variants (CNV) with PredicineSCORE. High concordance was observed between ctDNA-derived mutations and tissue-derived KRAS, TP53, and APC mutations and MSI/MSS status at baseline. 97% of patients with KRAS G12D/V/C mutations positively identified by local tissue testing were also positively identified by ctDNA. 6% of subjects considered KRAS wild-type by local tissue testing were identified as having KRAS G12D/V/C mutations by ctDNA. KRAS mutations, wild-type TP53, chromosomal loss of 18q21.33, high methylation scores and hypomethylated Tetraspanin 15 correlated with treatment efficacy. Neo-KRAS WT, or loss of KRAS mutations on treatment, emerged during treatment in 36% of patients with KRAS mutations at baseline and was associated with improved survival. Significant ctDNA level reduction was seen during treatment in both arms of the study irrespective of tumor response. These insights underscore the promise of ctDNA profiling in enhancing precision medicine by enabling real-time, dynamic assessment of tumor biology and therapeutic response with a potential to guide personalized treatment strategies for mCRC patients in the future. Conclusions: This study highlights the potential utility of liquid biopsy in the mCRC setting and provides novel insights from the ctDNA analysis, highlighting genetic and epigenetic ctDNA features that might be associated with clinical efficacy of SOC ± nivolumab in 1L mCRC. The emergence of Neo-KRAS WT in a substantial number of patients indicates a dynamic nature of KRAS alterations and highlights the additional value that longitudinal plasma ctDNA analysis could provide. Clinical trial information: NCT03414983 .
Surufatinib combined with anti-PD-1/PD-L1 antibody in the second line or monotherapy in third line treatment of advanced hepatocellular carcinoma: A single-arm, open-label, multi-center phase II study.
e16172 Background: Surufatinib is a selective inhibitor of VEGFRs, FGFR1, and CSF-1R, demonstrating dual mechanisms of anti-angiogenesis and immunomodulation. Combination therapy with anti-PD-1/PD-L1 antibodies may yield synergistic therapeutic effects. Methods: This single-arm, open-label, multicenter phase II clinical trial (NCT05282433) enrolled patients aged 18-75 years with unresectable hepatocellular carcinoma, BCLC stage B or C, and Child-Pugh class A or B liver function. Patients in Cohort 1 received surufatinib (250 mg orally once daily in 3-week cycles) combined with an anti-PD-1/PD-L1 antibody (administered every 3 weeks) as second-line therapy. Patients in Cohort 2 received surufatinib monotherapy (300 mg orally once daily in 3-week cycles) as third-line therapy. The primary endpoint was progression-free survival (PFS). Secondary endpoints included overall survival (OS), objective response rate (ORR), disease control rate (DCR), and safety. Results: As of April 30, 2025, Cohort 1 enrolled 25 patients (median age 54 years; 20.0% female; 100% BCLC stage C; Child-Pugh A: 84.0%, B: 12.0%, C: 4.0%; 88.0% with elevated AFP, of which 60.0% had AFP ≥400 μg/L; 56.0% with vascular invasion; 88.0% with extrahepatic metastasis). Cohort 2 enrolled 18 patients (median age 58.5 years; 88.9% male; 100% BCLC stage C; 100% Child-Pugh A; 66.7% with elevated AFP, of which 38.9% had AFP≥400μg/L; 50.0% with vascular invasion; 72.2% with extrahepatic metastasis). In Cohort 1, three patients experienced rapid disease progression and discontinued study after only one cycle of immunotherapy, and two patients were not evaluable for efficacy. The ORR and DCR were 8.7% and 69.6%, respectively. The median PFS and median OS were 3.33 months and 7.36 months, respectively. The 16-week PFS rate was 40.0%, with 6-month and 12-month OS rates of 72.0% and 25.0%. In Cohort 2, one patient was not evaluable for efficacy. The ORR and DCR were 5.9% and 82.4% , respectively. The median PFS and median OS were 3.22 months and 13.60 months, respectively. The 16-week PFS rate was 50.0%, with 6-month and 12-month OS rates of 72.2% and 47.1%. Grade ≥3 treatment-related adverse events consisted primarily of thrombocytopenia and leukopenia. No new safety signals were identified. Conclusions: Surufatinib demonstrated promising efficacy and a manageable safety profile in patients with advanced hepatocellular carcinoma. Further studies in broader populations are warranted to validate these findings. Follow-up is ongoing, and additional data are anticipated. Clinical trial information: NCT05282433 .
An indirect comparison of EGFR TKI combination therapies in advanced <i>EGFR</i> -mutated NSCLC: Comparing manual and AI-generated analyses.
e20625 Background: The treatment landscape for stage IV EGFR-mutated non-small cell lung cancer (NSCLC) is rapidly evolving. MARIPOSA (amivantamab plus lazertinib) and FLAURA2 (osimertinib plus chemotherapy) have established combination strategies that improve survival outcomes compared with osimertinib monotherapy. However, the comparative efficacy remains unclear. Additionally, with the growing integration of artificial intelligence (AI) into clinical research, this study aimed to evaluate AI-generated analysis compared to traditional manual analysis. Methods: An indirect comparison analysis was performed using Bucher statistical method, which enables comparative effectiveness estimation between interventions evaluated in separate randomized trials through a shared common comparator. Efficacy and safety outcomes from the MARIPOSA trial were indirectly compared against those of the FLAURA2 trial. Outcomes of interests include PFS, OS, objective response rates (ORR), and adverse events (AEs). Additionally, an independent AI-driven analysis using large language model (LLM) (GPT-5.2) was generated using predefined clinical trial keywords and aggregate trial data. Furthermore, exploratory AI-generated analyses were conducted, including digitization and reconstruction of Kaplan–Meier curves to estimate time-to-event outcomes. Results: In the manual analysis, the hazard ratio (HR) of PFS for Amivantamab-Lazertinib versus Osimertinib + Chemotherapy (as reference) was 1.13 (95% CI 0.83 – 1.53; p = 0.44). The indirect OS HR was 0.97 (95% CI 0.72 – 1.32; p = 0.87). The odds ratio (OR) of ORR in the Amivantamab-Lazertinib arm compared to Osimertinib + Chemotherapy was 0.69 (95% CI 0.39 – 1.22; p = 0.20). Lastly, the risk of experiencing any grade ≥3 AE was similar between treatment arms (OR = 0.81; 95% CI 0.51 – 1.3; p = 0.39). The manual analysis did not find any statistically significant differences between the two treatment arms. The AI-driven analysis produced concordant results. AI generated reconstructed Kaplan–Meier analyses further demonstrated minimal differences in restricted mean survival time at 36 months ( < 1 month) between the two combination regimens, with no statistically significant separation observed. Conclusions: Indirect comparison using both manual and AI LLM-generated approaches demonstrated no significant differences in efficacy or safety between amivantamab–lazertinib and osimertinib plus chemotherapy. Together, these results suggest that both combination strategies provide broadly comparable clinical benefit, supporting treatment selection based on toxicity tolerance and clinical preference. Comparison Value + 95% Confidence Interval P-value Indirect PFS (HR) 1.13 (0.83 – 1.53) 0.44 Indirect OS (HR) 0.97 (0.72 – 1.32) 0.87 Indirect ORR (OR) 0.69 (0.39 – 1.22) 0.20 Indirect grade ≥3 AE (OR) 0.81 (0.51 – 1.31) 0.39
Envafolimab combined with lenvatinib and gemcitabine plus cisplatin in advanced biliary tract cancer as first-line treatment: A multicenter, single-arm, open-label, phase II study (ENLIGHTEN study).
4152 Background: The prognosis of advanced biliary tract cancer (BTC) remains unsatisfactory despite the use of first-line gemcitabine and cisplatin in combination with PD-1/PD-L1 inhibitors. Lenvatinib (LEN), an anti-angiogenic multi-kinase inhibitor, has recently demonstrated promising antitumor activity in various tumors. This study aimed to determine the efficacy and safety of envafolimab (PD-L1 inhibitor) and LEN in combination with gemcitabine and cisplatin as first-line treatment in advanced BTCs. Methods: This investigator-initiated, multicenter, single-arm, phase 2 trial enrolled patients with advanced BTCs from three centers. A Simon's two-stage optimal design was employed: if 4 or more patients achieved partial responses (PR) among the first 10 patients in stage 1, enrollment would expand to a total of 39 patients. The therapy would be considered promising if 15 or more patients achieved PR in total. Including a 10% dropout allowance, the total sample size was set at 43. All patients received subcutaneous envafolimab (400mg, day 1, Q3W) and LEN (8 mg, orally once daily) combined with gemcitabine (1000mg/m2, iv. drip, days 1, 8, Q3W) plus cisplatin (25mg/m2, iv. drip, days 1, 8, Q3W) up for 6-8 cycles, followed by maintenance envafolimab and LEN until disease progression (PD) or unacceptable toxicity. The primary endpoint was objective response rate (ORR) assessed per RECIST 1.1. Secondary endpoints included overall survival (OS), progression-free survival (PFS), disease control rate (DCR), and safety. Results: A total of 43 patients were enrolled in this study from October 2022 to July 2025. However, one patient withdrew consent due to personal reasons after completing one cycle of therapy and was therefore considered off-study. Forty-two patients were included in the final analysis, with a median age of 57 years (range, 29-73). The ORR and DCR were 42.9% (95% CI, 27.7%-59%) and 88.1% (95% CI, 74.4%-96%), respectively. The observed outcomes were 18 patients (42.9%) with a partial response (PR), 19 (45.2%) with stable disease (SD), and 5 (11.9%) with PD, which met the predefined threshold for a positive study endpoint. The median follow-up time, estimated by the reverse Kaplan-Meier method, was 21.5 months (range, 18.7-29.1). The median PFS was 8.8 months (95% CI, 8.0-14.1), and the median OS was 15.9 months (95% CI, 13.4-not reached). The most common grade 3/4 treatment-related adverse events (TRAEs) were platelet decreased (n = 9, 20.9%) and neutropenia (n = 7, 16.3%). No treatment-related deaths occurred. Conclusions: In summary, our study demonstrates that the quadruplet regimen of envafolimab, lenvatinib, gemcitabine, and cisplatin demonstrates promising efficacy and well-tolerated first-line therapy for advanced BTC. Clinical trial information: NCT05410197 .
Group-based versus individualized exercise programs in lung cancer: Results from the randomized crossover LungFit study.
e20101 Background: Patients with lung cancer frequently experience reduced functional capacity and impaired quality of life throughout the disease trajectory, regardless of stage. Exercise interventions have demonstrated benefits in oncology; however, evidence comparing different delivery modalities and their implementation in routine care remains limited. LungFit was designed to evaluate the effects of two structured exercise modalities on functional and patient-reported outcomes in patients with lung cancer. Methods: LungFit is a prospective, randomized, controlled crossover clinical study including patients with lung cancer irrespective of disease stage. Participants are randomized at baseline to receive either a group-based or an individualized structured exercise program for three months. After a 15-day washout period and repeat assessment, participants cross over to the alternate exercise modality for an additional three months. Both interventions consist of supervised, tailored exercise programs targeting aerobic capacity, muscle strength, and functional performance. Assessments are conducted at baseline, after the first intervention period, and after completion of both exercise modalities. The primary endpoint is change in functional capacity. Secondary endpoints include quality of life, symptom burden, adherence, feasibility, and patient-reported outcomes related to humanization of care. Results: In this preliminary analysis, 35 participants completed paired functional assessments after first intervention period of exercise. Mean change in the 6-minute walk test (6MWT) was 15.8 m (95% CI −12.2 to 43.9; p = 0.26). Improvement was greater with the individualized intervention compared with the group-based program (22.8 vs 3.0 m; p = 0.49). Functional performance significantly improved in the 30-second sit-to-stand test, with a mean increase of 2.8 repetitions (p < 0.001). No significant changes were observed in Timed Up and Go or handgrip strength. Among 27 participants with paired psychological assessments, HADS-Anxiety slightly decreased and HADS-Depression remained stable, with no significant differences between assessments. Patient satisfaction with the exercise program was high, with mean scores of 4.74 in the group-based intervention and 4.63 in the individualized intervention, without significant differences between modalities (p = 0.62). Conclusions: Structured exercise programs were feasible, well accepted, and associated with improved lower-limb functional performance in patients with lung cancer. Both group-based and individualized modalities showed high satisfaction and comparable short-term outcomes. Full crossover results will further clarify the comparative impact of both delivery formats.
Phase 1 study of NTX1088, a first-in-class anti-PVR monoclonal antibody, as monotherapy and combined with pembrolizumab, in patients with advanced solid malignancies.
2518 Background: Poliovirus Receptor (PVR; CD155), a transmembrane protein upregulated across multiple solid tumors and associated with poor clinical outcomes and resistance to immune checkpoint inhibitors (CPIs), plays a central role in tumor-mediated immune suppression. PVR suppresses anti-tumor immunity by interacting with the co-stimulatory receptor DNAM1 (CD226) on T and NK cells, leading to its internalization and degradation. PVR also engages the inhibitory immune receptors, TIGIT, CD96, and KIR2DL5A. NTX1088 is a first-in-class monoclonal antibody targeting PVR. Methods: NTX-1088-01 (NCT05378425) is a Phase 1a/1b, open-label, multicenter dose-escalation and expansion study evaluating the safety, tolerability, pharmacokinetics (PK), pharmacodynamics, recommended dose for expansion (RDE), and preliminary efficacy of NTX1088 alone and in combination with pembrolizumab in patients (pts) with advanced solid tumors known to express PVR (prescreening not required). NTX1088 was administered intravenously (IV) every 3 weeks (Q3W) in dose levels (DLs) from 12-1750 mg as monotherapy or from 160-1750 mg in combination with pembrolizumab 200 mg IV Q3W. Phase 1a employed a standard 3+3 dose-escalation design. Phase 1b comprises dose-expansion monotherapy and combination cohorts in gastric, bladder, lung, and other PVR-expressing solid tumors at the RDE. Results: As of January 2026, 81 pts were treated with NTX1088 (24 monotherapy; 57 combination). Median age was 61 years; median lines of prior therapy was 4 and prior CPI exposure was 70%. NTX1088 was well tolerated with no dose-limiting toxicity and 1750 mg was selected as the RDE in monotherapy and in combination with pembrolizumab, based on PK and target occupancy. Preliminary efficacy analysis was focused on active DLs of NTX1088 (1200 and 1750 mg) in combination with pembrolizumab. Forty-eight pts were treated at these DLs and 35 pts were evaluable for response. Of these 35 pts, median age was 57, median lines of prior therapy was 4 and 89% received prior CPI. Six (17%) pts achieved confirmed partial responses (PR) including gastric (2), bladder (1), non-small cell lung (1), squamous cell carcinoma of the head and neck (1), and melanoma (1). All PRs were seen in pts with prior CPI exposure. Sixteen (46%) pts achieved stable disease. At data cutoff, all but one PR pt remain on treatment, with a maximum treatment duration of 20 months. Exploratory analyses of tumor biopsies identified potential predictive biomarkers. Conclusions: NTX1088 demonstrated favorable tolerability and encouraging clinical activity in heavily pretreated pts who had progressed on prior PD-1/PD-L1 inhibitors. These findings support further evaluation of NTX1088 in Phase 2 studies, incorporating less heavily pre-treated pts and biomarker selection. Clinical trial information: NCT05378425 .
Pre-existing mental health diagnosis and colorectal cancer outcomes.
e15711 Background: Mental health disorders are common among patients (pts) with cancer, sometimes associated with delays in diagnosis and treatment. The impact of pre-existing mental health diagnoses (MHD) on colorectal cancer (CRC) presentation and outcomes is not well understood. Methods: We conducted a retrospective study of pts with MHD and CRC treated between 2015 and 2020 at a comprehensive cancer center. Cox proportional hazards models were used to compare overall survival (OS) among pts diagnosed with an MHD either before or after their CRC diagnosis, followed by sensitivity analysis and propensity score matching (PSM). MHD is defined as anxiety, depression, substance use disorder, PTSD, bipolar disorder, and schizophrenia. OS was defined as the time from CRC diagnosis to death from any cause. Statistical significance was set at 0.05. Results: A total of 528 patients with MHD were included;19.2% had pre-existing MHD (pre-MHD). Compared to patients without pre-MHD, those with pre-MHD were older (≥ 60 years) (69.61% vs 44.84%, p < 0.0001) and less likely to present with advanced (stage IV) disease (19.61% vs 35.45%, p < 0.001). Significant predictors of mortality included Black race (HR = 1.72, 95%CI: 1.32–2.24, p < 0.0001), male sex (HR = 1.35, 95%CI: 1.05–1.74, p = 0.019), and stage IV disease (HR vs stage I = 5.32, 95% CI: 3.47–8.16, p < 0.0001). In the unadjusted Cox proportional hazards analysis, pre-MHD was not associated with OS, but was associated with worse OS in the adjusted model (HR = 1.51, 95% CI: 1.09–2.08, p = 0.012). PSM cohort of 204 patients; The propensity of having a pre-MHD was modeled as a function of age, tumor site, and stage. In the unadjusted analysis, the estimated HR of pre-MHD was 1.51 (95% CI 1.00 - 2.28, p = 0.049). However, this association was attenuated in the adjusted model (HR = 1.50, 95% CI 0.99 – 2.28, p = 0.055). Conclusions: Despite presenting with less advanced disease, patients with pre-MHD experienced worse survival. These findings highlight the importance of integrating mental health assessment and support into oncologic care. Validation in larger, prospective cohorts is needed while targeting high-risk groups for intervention. Patient characteristics stratified by Pre-existing Mental Health Diagnosis. Demographics Total (N=528) Pre-MHD(N=102) No pre-MHD(N=426) P-value Site 0.026 Colon 383(72.54) 83(81.37) 300(70.42) Rectum 145(27.46) 19(18.63) 126(29.58) Stage <0.001 1 86(16.29) 31(30.39) 55(12.91) 2 115(21.78) 20(19.61) 95(22.3) 3 156(29.55) 31(30.39) 125(29.34) 4 171(32.39) 20(19.61) 151(35.45) Sex 0.513 Female 311(58.90) 63(61.76) 248(58.22) Male 217(41.10) 39(38.24) 178(41.78) Age at cancer diagnosis <0.001 Median (IQR) 59(50-69) 68(56-82) 58(48-68) Tobacco use 0.189 Yes 72(13.64) 18(17.65) 54(12.68) No or Unknown 456(86.36) 84(82.35) 372(87.32)
Spatial PK/PD profiling in human ex vivo tumor slices to identify exposure-limited niches driving heterogeneous response to a TROP2-targeting ADC in NSCLC.
e15040 Background: TROP2-targeting antibody–drug conjugates (ADCs) have demonstrated promising clinical activity in non-small cell lung cancer (NSCLC); however, substantial inter- and intra-tumoral response heterogeneity remains poorly explained by bulk TROP2 expression alone. We hypothesized that spatial heterogeneity of ADC exposure and payload distribution within intact tumor microenvironments represents a key determinant of pharmacodynamic response. Methods: Fresh human NSCLC tumor specimens (n = 12) were processed into precision-cut ex vivo tissue slices (300–400 µm) and maintained under short-term culture conditions. Tissue slices were treated with a TROP2-targeting ADC or vehicle control for up to 48 hours. Following treatment, slices were fixed, embedded, and re-sectioned into 4–5 µm sections for spatial analysis. ADC localization and target expression were assessed using multiplex immunofluorescence and imaging mass cytometry. Pharmacodynamic responses were evaluated through spatial profiling of DNA damage (γH2AX), apoptosis (cleaved caspase-3), and proliferation (Ki67), together with markers of tumor microenvironmental architecture including vasculature (CD31), stroma (αSMA), and hypoxia (CAIX). Spatial co-localization analyses were performed to map relationships between ADC exposure, tissue compartments, and pharmacodynamic effects. Results: Spatial analysis revealed marked heterogeneity in ADC distribution across tumor regions, with reduced exposure in stromal-rich and hypoxic compartments despite relatively homogeneous TROP2 expression. High-exposure regions exhibited increased γH2AX and cleaved caspase-3 signaling, consistent with treatment-induced DNA damage and apoptosis, whereas low-exposure regions maintained elevated Ki67 expression, indicating persistent proliferative activity. Importantly, bulk TROP2 expression did not correlate with spatial pharmacodynamic response, while compartment-level ADC exposure metrics were strongly associated with treatment-induced biological effects. Conclusions: These findings demonstrate that spatial PK/PD profiling in human ex vivo NSCLC tumor slices provides mechanistic insight into TROP2-ADC efficacy beyond conventional biomarker approaches. Identification of exposure-limited tumor niches may explain heterogeneous responses to ADC therapy and supports the development of spatial biomarkers to guide patient stratification and rational ADC combination strategies.
Prognostic role of ATM in gastric cancer: A systematic review and meta-analysis.
e16342 Background: Gastric adenocarcinoma remains one of the primary causes of cancer-related mortality worldwide. There is conflicting evidence about the Ataxia-Telangiectasia Mutated (ATM) protein's function as a predictive biomarker in gastric cancer, despite the fact that it is crucial for DNA damage repair and genomic stability. Methods: In accordance with PRISMA criteria and PROSPERO registration (CRD20251035762), we performed a systematic review and meta-analysis of cohort studies up to May 2025. Pooled hazard ratios (HRs) with 95% confidence intervals (CIs) were computed to evaluate the impact of ATM protein expression on survival outcomes, including overall survival (OS), disease-free survival (DFS), and disease-specific survival (DSS). Results: A meta-analysis of 9 trials including 2,316 patients indicated that higher levels of ATM expression are substantially linked to better overall survival (HR = 1.87, 95% CI: 1.04-3.39, p = 0.04). While the overall pooled disease-free survival (DFS) was not statistically significant (p = 0.18), subgroup analysis of multivariate models indicated a significant correlation between low ATM expression and poor DFS (HR = 3.27, 95% CI: 1.67–6.42, p = 0.0006). Qualitative synthesis further validated a significant connection between ATM mRNA levels and protein expression. Conclusions: ATM expression is a crucial prognostic marker in gastric cancer, where low expression is associated with reduced survival and an elevated risk of recurrence. These results endorse the utilization of ATM as a biomarker for risk assessment and individualized treatment planning.
Randomized controlled trial of a psychosocial digital application (THRIVE-MM) for patients living with multiple myeloma.
1513 Background: Patients living with multiple myeloma (MM) struggle with considerable quality of life (QOL) impairments, fatigue, and psychological distress throughout their illness course. However, few accessible and scalable interventions address the supportive care needs of these patients. We assessed the efficacy of a self-guided psychosocial digital application, THRIVE-MM, for improving QOL, fatigue, and psychological distress in patients living with MM. Methods: We conducted a single-center randomized controlled trial of THRIVE-MM vs. usual care for adults living with MM. Eligibility criteria included patients newly diagnosed with MM receiving first line therapy, those with relapsed MM receiving second- or third-line therapy, or those with MM on maintenance therapy. Participants were randomly assigned 1:1 to THRIVE-MM or usual care, stratified by line of therapy (first line vs. second/third line vs. maintenance) and were given access to the THRIVE-MM digital app up to 8 weeks after enrollment. THRIVE-MM comprises five modules designed to improve patients’ physical and psychological symptoms via psychoeducation and evidenced-based self-management skills training for behavioral change and coping, delivered through interactive activities, educational games, and videos. Participants completed self-report measures at baseline and at week-8 after enrollment. The primary endpoint was QOL at week-8 assessed by the Functional Assessment of Cancer Therapy (FACT)-MM. Secondary outcomes were fatigue (FACT-Fatigue), depression and anxiety symptoms (Hospital Anxiety and Depression Scale), and symptoms of posttraumatic stress disorder (PTSD Checklist). We used analysis of covariance controlling for baseline criterion value to assess the effect of THRIVE-MM on outcomes. Results: From 12/2023-6/2025, we enrolled 59.1% (120/203) of eligible patients (mean (SD) age = 68.2 (10.0) years; 49.2% women). Mean (SD) use time for THRIVE-MM in the intervention group was 221.9 (197.1) minutes. At week-8, QOL did not differ significantly between the two groups (THRIVE-MM; 125.2 vs. usual care; 125.1, p = 0.977]). THRIVE-MM participants reported lower fatigue (40.6 v. 37.6, p = 0.012) and lower depression symptoms (2.5 vs. 3.4, p = 0.044) at week-8. Anxiety or PTSD symptoms did not differ significantly between the two groups at week-8. Conclusions: THRIVE-MM, a psychosocial digital health intervention, decreased fatigue and depression symptoms but not QOL among patients living with MM. Clinical trial information: NCT06073353 .
A cardio-oncologic burden: Leukemia-associated arrhythmia mortality in the United States (1999-2023).
e18634 Background: Leukemia remains a major contributor to cancer-related mortality in the United States. Cardiac arrhythmias are common in patients with leukemia due to pre-existing cardiovascular disease, myocardial infiltration, chronic inflammation, and treatment-related cardiotoxicity. However, national long-term mortality trends involving both conditions are poorly characterized. We examined temporal patterns and demographic and geographic disparities in leukemia-associated arrhythmia mortality among middle-aged and older adults. Methods: We analyzed U.S. mortality data from the CDC WONDER Multiple Cause of Death database (1999–2023). Deaths among adults aged ≥45 years with leukemia (ICD-10 C91–C95) and cardiac arrhythmias (ICD-10 I47–I49) listed as underlying or contributing causes were included. Analyses were stratified by age, sex, race, census region, state, urbanization, and place of death. Age-adjusted mortality rates (AAMRs) were calculated using the 2000 U.S. standard population. Joinpoint regression estimated annual percent change (APC) and average annual percent change (AAPC); P < 0.05 was considered significant. Results: From 1999–2023, 32,350 deaths occurred among adults aged ≥45 years with leukemia and arrhythmias. Mortality increased gradually from 1999–2018 (APC 2.79%), followed by a marked rise from 2018–2021, with AAMRs increasing from 1.18 to 1.74 (APC 15.9%, P = 0.005), and a subsequent nonsignificant decline after 2021 (APC −2.0%, P = 0.64). Similar surges were observed in males (APC 12.78%) and females (APC 14.86%). AAMRs doubled among adults aged 45–64 years (AAPC 4.35%, P = 0.002) and ≥65 years (AAPC 3.97%, P < 0.001). Mortality increased significantly among White individuals (AAPC 3.90%, P < 0.001) but remained stable among Black individuals (AAPC 2.83%, P = 0.076). All census regions showed rising trends, greatest in the South (AAPC 4.72%) and West (4.58%). Rates increased in both metropolitan and non-metropolitan areas. Most deaths occurred in inpatient facilities (43%), followed by homes (27%). Conclusions: Leukemia-associated arrhythmia mortality has increased substantially over the past 25 years, with a pronounced acceleration between 2018 and 2021. Persistent demographic and regional disparities underscore the need for enhanced cardio-oncology integration, arrhythmia surveillance, and targeted cardiovascular risk management in patients with leukemia.
Trends in ovarian function suppression use among premenopausal women with breast cancer: A longitudinal analysis from 2015-2023.
e12694 Background: Ovarian function suppression (OFS) combined with endocrine therapy (ET) using tamoxifen or an aromatase inhibitor improves outcomes among premenopausal women with hormone receptor-positive breast cancer, particularly those with higher recurrence risk. More than a decade after initial publication of the SOFT and TEXT trials supporting OFS use, real-world adoption patterns and factors associated with OFS uptake over time remain incompletely characterized. Methods: We used the MarketScan healthcare claims database to identify women aged 18-49 with a breast cancer diagnosis who underwent lumpectomy or mastectomy between 2015 and 2023. Analyses were limited to women with commercial insurance and pharmacy coverage from one year before to two years after surgery. Women with a gynecologic cancer diagnosis or prior oophorectomy were excluded. OFS was defined as oophorectomy after surgery or receipt of >2 gonadotropin-releasing hormone agonist (GnRHa) injections. Adherence to ET was defined as proportion of days covered (PDC) >0.80 within the first year. Temporal trends in OFS use were assessed, and multivariable logistic regression was used to identify factors associated with OFS use. Results: We identified 14,949 women who met inclusion criteria. In the overall cohort, 2,988 (20.0%) women received OFS, with uptake increasing significantly over time from 15.5% in 2015 to 25.1% in 2023 (p < 0.001). This trend was driven by increased GnRHa use (8.9% in 2015 to 19.7% in 2023), with oophorectomy rates remaining stable (8.1% to 8.5%). Nearly half of all patients received chemotherapy (46.8%), predominantly in the adjuvant setting (43.4%). OFS use among adjuvant chemotherapy recipients increased from 29.3% to 50.8% over the study period (p < 0.001), again driven by increased GnRHa use (20.6% to 45.2%). Most patients initiated a GnRHa during adjuvant chemotherapy (76.6%), and among these, 84.2% continued the GnRHa after chemotherapy completion. After adjusting for baseline characteristics, OFS use was associated with younger age (35.1% for age 18-39 versus 16.9% for age 40-49; OR 1.94, 95% CI 1.75 – 2.16) and chemotherapy use (OR 11.28, 95% CI 10.05 – 12.67). Among our study population, 10,033 (67.1%) women received ET. Non-adherence to ET was slightly lower among those who underwent OFS (22.8% vs 25.4%, p = 0.01). Conclusions: Uptake of OFS among premenopausal women with breast cancer has increased substantially since 2015, although overall use remains low. By 2023, approximately one-quarter of the overall cohort and one-half of patients who received chemotherapy were treated with OFS. Notably, receipt of OFS was associated with lower rates of non-adherence to ET. Further research is needed to understand patient and treatment factors associated with OFS initiation and ET adherence, in order to optimize OFS use and improve outcomes in young women with breast cancer.
Prospective clinical activity and preclinical basis of panitumumab-based EGFR blockade in SMARCB1-deficient renal medullary carcinoma (RMC): A collaborative multi-institutional study.
4520 Background: RMC is a rare, exceptionally aggressive malignancy that predominantly affects young individuals of African descent with sickle cell trait. RMC is typically refractory to standard kidney cancer therapies, including anti-angiogenic agents and immune checkpoint inhibitors. While cytotoxic chemotherapy is the current standard, objective response rates (ORR) remain low (~29%), with a median progression-free survival (PFS) of ≤ 4 months in the first-line setting. This study explores wild-type EGFR as a therapeutic dependency in RMC. Methods: We characterized EGFR expression and mutation status in primary RMC tumors using whole exome sequencing (WES), CLIA-certified IHC and integrated bulk RNA and histone ChIP-sequencing. Preclinical efficacy was compared between the EGFR antibody panitumumab and the kinase inhibitor erlotinib across two RMC xenograft models. Subsequently, a prospective multi-national registry (N=26) evaluated panitumumab-based therapy (monotherapy or combined with nab-paclitaxel ± carboplatin) in heavily pretreated patients with RMC. Results: No EGFR mutations were detected on WES. RMC tumors demonstrated uniformly high wild-type EGFR protein expression and enhancer-associated activation. In vivo, panitumumab induced profound tumor regressions in both RMC xenograft models, significantly outperforming erlotinib (p < 0.005). Mechanistically, panitumumab triggered definitive lysosomal receptor degradation and suppressed AKT and ERK1/2 signaling. The prospective clinical registry (N=26) included 20 (76.9%) males and 6 (23.1%) females, with a median age of 33.5 years (range, 17–67). At treatment initiation, patients had a median of 4 metastatic disease sites (range, 2–6). The cohort was heavily pretreated with a median of 2 prior systemic therapies (range, 0–4); 96.2% had progressed on prior platinum-based chemotherapy. We observed an ORR of 53.9% (14/26 patients), including 4 (15.4%) complete responses. Median PFS was 5.8 months (95% CI: 4.0–8.2) and median OS was 9.5 months (95% CI: 7.6–NE). Treatment was well-tolerated, with manageable grade 1-2 acneiform rash (80.8%), no grade 3+ rash cases observed, and no treatment-related deaths. Conclusions: These data establish wild-type EGFR dependency as a foundational vulnerability in RMC. Panitumumab-based therapy yields unprecedented responses in heavily pretreated patients and represents a transformative new systemic standard of care for this lethal disease. Clinical outcomes. Total (N=26) Best overall response, n (%) Complete response 4 (15.4%) Partial response 10 (38.5%) Stable disease 7 (26.9%) Progressive disease 5 (19.2%) Depth of response % Median (IQR) -30.9 (-58.9, +3.1)
Time-of-day administration of immunotherapy in early-stage triple-negative breast cancer: Immune-dependent chronotherapy effects in the phase III A-BRAVE trial.
1010 Background: Circadian rhythms regulate immune functions, and morning (AM) administration of immunotherapy (IO) is associated with survival in several solid tumors. However, whether time-of-day (ToDa) of IO delivery influences outcomes in early-stage triple-negative breast cancer (TNBC), and whether this effect depends on immune biomarkers, remains unknown. Methods: A-BRAVE randomized patients with high-risk early-stage TNBC to 1 year of adjuvant avelumab or observation. Infusion time of avelumab was retrieved from eCRF. AM-rate was calculated for each patient as the proportion of the first 4 infusions occurring before 12:30 PM (cohort median ToDa) and patients categorized as AM-dominant (AM-rate ≥50%) or PM-dominant. Tumor-infiltrating lymphocytes (TILs) and PD-L1 (Dako 73-10) were centrally assessed on treatment-naive tumor samples. We analyzed distant disease-free survival (DDFS) and overall survival (OS) by adjusted Cox models. Results: ToDa data were available for 221 avelumab-treated patients (94.0%); TILs and PD-L1 were evaluable in 188 (85%) and 195 (88%) of these, respectively. AM-rate was not associated with outcomes. However, ToDa effects were strongly immune-dependent. Increasing AM-rate was associated with improved DDFS and OS in immune-hot tumors (TILs [≥20%] or PD-L1 [≥21] high), but with worse outcomes in immune-cold tumors (TILs or PD-L1 low) (AM-rate*TILs interaction: DDFS p=0.008, OS p=0.001; AM-rate*PD-L1 interaction: DDFS p=0.030, OS p=0.039). These findings were concordant using AM-dominant vs. PM-dominant categorization (Table). Patients receiving immune-aligned treatment (immune-hot/AM-dominant OR immune-cold/PM-dominant) had superior survival compared with immune-misaligned (immune-hot/PM-dominant OR immune-cold/AM-dominant) and the observational arm (TILs-based: 3-year OS 95.7% vs 75.2% vs 78.0%; p<0.001; PD-L1-based: 3-year OS 93.4% vs 79.4% vs 76.1; p=0.035). Conclusions: In early-stage TNBC, time of IO administration may be a driver of efficacy, with opposing effects according to baseline immune milieu. Immune-informed circadian alignment may represent a previously unrecognized determinant of IO efficacy in TNBC. Clinical trial information: NCT02926196 . Outcome AM-dominant3-yr Rate % (95% CI) PM-dominant3-yr Rate % (95% CI) Log-rank p AM vs PM dominant HR (95% CI) TILs High DDFS 90.1 (81.3–99.8) 72.0 (56.4–91.9) 0.064 0.39 (0.13–1.21) OS 97.5 (92.8–100) 76.0 (61.0–94.7) 0.019 0.35 (0.10–1.20) TILs Low DDFS 63.1 (52.6–75.7) 84.7 (75.5–95.1) 0.003 2.61 (1.23–5.55) OS 74.9 (65.2–86.0) 94.3 (88.3–100) 0.003 3.84 (1.30–11.29) PD-L1 high DDFS 100 (100–100) 61.5 (40.0–94.6) 0.015 0.11 (0.01–0.94) OS 100 (100–100) 76.9 (57.1–100) 0.008 0.11 (0.01–1.03) PD-L1 low DDFS 66.4 (57.3–77.0) 83.0 (74.7–92.2) 0.026 1.82 (1.00–3.31) OS 79.8 (71.9–88.6) 91.5 (85.3–98.2) 0.025 2.05 (0.95–4.43)
Impact of U.S. most-favored-nation (MFN) drug pricing on global oncology access.
11056 Background: U.S. cancer drug prices substantially exceed those in other OECD countries. In May 2025, a U.S. Presidential Executive Order mandated adoption of a Most-Favored-Nation (MFN) pricing policy, linking U.S. drug prices to those of economically comparable countries. In December 2025, the Centers for Medicare & Medicaid Services (CMS) proposed two MFN-based frameworks: the Global Benchmark for Efficient Drug Pricing (GLOBE) for Medicare Part B drugs and Guarding U.S. Medicare Against Rising Drug Costs (GUARD) for Part D drugs. Both require manufacturer rebates when U.S. prices exceed GDP-adjusted international benchmarks, creating potential incentives for manufacturers to alter global oncology pricing and launch strategies. This study assesses which international markets may face pricing pressure or delayed access to new oncology therapies following MFN implementation. Methods: The top 50 best-selling U.S. drugs in 2024 were reviewed, and 12 oncology drugs without generic or biosimilar competition were selected. List prices were collected from the 19 OECD reference countries included in GLOBE and GUARD. Prices were adjusted for GDP per capita using purchasing power parity (PPP). For each drug, the two lowest GDP-adjusted international prices were identified and compared with U.S. prices, using January 2026 Medicare Average Sales Price (ASP) data for Part B drugs and December 2025 Federal Supply Schedule (FSS) prices for Part D drugs. Results: Across all 12 drugs, GDP-adjusted benchmark prices were 48% to 90% lower than U.S. prices. Pembrolizumab, the top-selling oncology drug in 2024, had a U.S. Medicare price of $5,972 per 100 mg; the lowest adjusted prices were observed in South Korea ($2,116) and Japan ($2,271), representing reductions of 65% and 62%, respectively. Daratumumab, the second top-selling oncology drug, was priced at $2,831 per 400 mg in the U.S., compared with $1,467 in South Korea and $1,603 in the Netherlands (48% and 43% lower). The lowest benchmark prices most frequently originated from South Korea (n=5) and the Netherlands (n=4). Norway, Germany, and Japan each provided the lowest benchmark for one drug. Second-lowest benchmark prices were observed in Japan (n=4), South Korea (n=3), the Netherlands (n=2), Germany (n=2), and Norway (n=1). Conclusions: Countries serving as the lowest-price benchmarks- particularly South Korea, Japan, the Netherlands, Norway, and Germany- may face increased risk of upward pricing pressure or delayed access to innovative oncology therapies as manufacturers seek to limit U.S. MFN rebate exposure. Expanded use of confidential discounts, confidential rebates and other managed-entry agreements may help preserve access while reducing MFN benchmark risk.
Construction and validation of a prognostic risk model for lung adenocarcinoma based on pyroptosis and necroptosis-related genes.
e20034 Background: Lung cancer is a leading cause of cancer-related deaths worldwide, and lung adenocarcinoma (LUAD) is the most common subtype. Despite some progress in early detection and treatment, the prognosis remains poor, and the disease burden is substantial. Therefore, identifying new potential therapeutic targets is of great significance. This study aimed to screen pyroptosis and necroptosis-related genes associated with LUAD prognosis, construct a prognostic risk model for LUAD, and identify potential target genes associated with LUAD prognosis. Methods: LUAD data containing survival information were collected from the UCSC Xena and cBioPortal databases. The TCGA cohort, GSE68465, and GSE31210 cohorts were used as the training and validation sets, respectively. Bioinformatic analysis of pyroptosis and necroptosis characteristic genes was performed. Kaplan-Meier curves and log-rank tests were used to calculate the significance of differentially expressed genes. A prognostic risk score model was constructed and validated using LASSO regression. The expression levels and prognostic value of prognostic genes were analyzed based on the TCGA and GTEx databases. Results: Eight pyroptosis and necroptosis characteristic genes (GJB3, PLK1, GAPDH, LDHA, GNG7, PKP2, LYPD3, KRT6A) were used to construct a prognostic risk model for LUAD. Patients in the low-risk group in both the training and validation sets showed significant survival benefits, demonstrating good independent prognostic predictive ability. Analysis of the TCGA and GTEx databases showed that GJB3 among the eight risk genes had a high hazard ratio and significantly impacted the overall survival (OS) and disease-free survival (DFS) of LUAD. Conclusions: A prognostic risk model for lung adenocarcinoma based on pyroptosis and necroptosis-related genes was successfully constructed using data mining techniques. This model serves as an independent prognostic factor for lung adenocarcinoma with good predictive performance. GJB3 is highly and specifically expressed in lung adenocarcinoma and is a potential oncogene affecting prognosis.
First-in-human phase I, open-label, multicenter study of HJ-004, a novel EGFR-PROTAC degrader, in patients with advanced <i>EGFR</i> -mutated non–small cell lung cancer.
TPS8678 Background: Despite remarkable progress with 1 st - to 3 rd -generation epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs), nearly all patients with EGFR-mutated non-small cell lung cancer (NSCLC) eventually develop acquired resistance through secondary EGFR mutations, bypass pathway activation, or histologic transformation. Targeted options that specifically address EGFR-dependent (on-target) resistance after progression on 3rd-generation EGFR-TKIs remain limited. HJ-004 is an orally bioavailable, highly selective proteolysis-targeting chimera (PROTAC) that degrades mutant EGFR by recruiting the cereblon (CRBN) E3 ubiquitin ligase. Preclinically, HJ-004 showed broad activity against classical, uncommon, and osimertinib-resistant EGFR variants with favorable pharmacokinetics (PK) and safety profiles. Methods: This first-in-human, phase I, multicenter, open-label study evaluates the safety, tolerability, PK, and preliminary antitumor activity of HJ-004 in patients with recurrent or metastatic EGFR-mutant non-squamous NSCLC. The study comprises two parts: dose escalation and dose expansion. In the dose-escalation phase, oral HJ-004 is administered once daily in 28-day cycles using accelerated titration followed by a conventional 3 + 3 design (planned doses: 12.5, 25, 50, 87.5, 125, and 162.5 mg). Dose-limiting toxicities (DLTs) are assessed during the single-dose (C0D1-C0D3) and first multiple-dose cycle (C1D1-C1D28) periods. The expansion phase will enroll approximately 20-40 participants in two dose cohorts to further evaluate safety, PK, and preliminary efficacy. Key eligibility criteria include measurable disease per RECIST v1.1 and adequate organ function. The trial is currently ongoing and is expected to enroll up to 76 patients. The phase I study initiated in 2025, and site activation and patient screening are currently underway. No safety or efficacy data are yet available. Enrollment is projected to be completed in 2027, and updated enrollment and PK data will be presented at a future meeting. Clinical trial information: NCT07361237 .
Risk of gastrointestinal (GI) immune-related adverse events (irAEs): Real-world experience.
e23410 Background: Immune checkpoint inhibitors (ICI) improve survival across solid tumors but can cause gastrointestinal (GI) irAEs. Real-world comparative data remains limited, particularly for composite GI outcomes. We compared the risk of GI AEs in patients receiving immunotherapy (IT) versus non-IT systemic therapy. Methods: We conducted a retrospective cohort study of 5,556 adult cancer patients treated at Cleveland Clinic Ohio between 2010–2022. Included cancers were advanced-stages bladder, endometrial, esophageal, gastric, head & neck, renal, bladder, melanoma & lung cancers, stages where IT is approved. We compared IT (± chemo/radiation) vs non-IT regimens; 99.9% of IT was ICI. Follow-up began at systemic therapy start & continued until GI-AE, death or last encounter. GI AEs included diarrhea, colitis, elevated liver enzymes, esophagitis, gastritis, enteritis, enterocolitis, & composite outcomes. Poisson regression estimated incidence rates, Kaplan–Meier methods cumulative incidence (CIR) & time-dependent Cox models (TD-HR) assessed risk with IT as a time-varying exposure. Longitudinal serum liver tests were analyzed using mixed-effects & generalized linear models over one year. Results: Among 5,556 patients, 1,288 (23%) received first-line IT & 951 (17%) second-line IT. Median age was 66 years; 38% were female & 87% White. There were 430 GI-AE & 268 entero-colonic composite events in median follow up of 14 months (CI: 6-34). Incidence rates (/100 person-years) were higher with IT for GI-AE (0.50 vs 0.27) & entero-colonic composites (0.31 vs 0.16) (P < 0.05). IT was associated with decreased esophagitis risk (TD-HR 0.26; P = 0.01). Among colitis cases, 48% were grade 2, 25% grade 3 & 18% grade 4; 89% required treatment, 46% received immunomodulators & 37% permanently discontinued therapy. Longitudinal liver analyses showed IT was associated with increased albumin (β = 0.04; P = 0.04) & decreased INR (β = −0.26; P = 0.04), with no change in ALT, AST, alkaline phosphatase, total bilirubin, or APTT. Baseline values strongly predicted longitudinal changes, & alkaline phosphatase, ALT, total bilirubin & APTT increased over time (P < 0.05). Conclusions: IT was frequently associated with GI AEs, reaching 15% at 4 years, though most were low-to-moderate grade & manageable. IT was associated with selective laboratory changes, while baseline liver values remained the strongest predictors. Lower esophagitis incidence may reflect reduced radiation exposure. Hepatic synthetic markers improved with IT. IT group (n=1,288) Non-IT group (n=4,268) Colitis: absolute number 29 28 Colitis: 2-year CIR / TD-HR TD-HR 2.5% (1.5%, 3.5%) / HR 5.2 (3.0, 9.0) 0.6% (0.4%, 0.9%) / Ref Diarrhea: absolute number 45 157 Diarrhea: CIR / TD-HR 4.3% (3.0%, 5.7%) / HR 1.4 (1.1, 1.9) 4.0% (3.3%, 4.7%) / Ref Elevated liver enzymes: absolute number 41 67 Elevated liver enzymes: CIR / TD-HR 3.8% (2.6%, 5.1%)/ HR 3.2 (2.2, 4.7) 1.7% (1.3%, 2.2%)/ Ref