Browse Articles
Discover research articles across all indexed journals
Trends in toxicity-associated hospitalizations and inpatient outcomes among gastrointestinal malignancies in the United States, 2016–2023.
e23237 Background: Advances in systemic therapy for gastrointestinal (GI) malignancies have expanded rapidly, including immune checkpoint inhibitors, targeted agents, and combination regimens. While outpatient survival has improved, it remains unclear whether inpatient outcomes during cancer-related hospitalizations have improved in parallel. We evaluated national trends in toxicity-associated hospitalizations and inpatient outcomes among GI malignancies across contemporary treatment eras. Methods: A retrospective, survey-weighted analysis of the National Inpatient Sample (2016–2023) was conducted. Adult hospitalizations with any gastrointestinal malignancy (ICD-10-CM C15–C20, C22–C25) were included, excluding encounters with palliative care coding (Z51.5). Treatment era was defined as 2016–2017 versus 2018–2023. Toxicity-like syndromes were identified using diagnosis proxies, including neutropenic sepsis (D70 with A40/A41), acute kidney injury (N17), colitis or severe diarrhea (K52, A09), respiratory failure or pneumonitis (J96, J80), myocarditis or shock (I40, R57), and electrolyte disorders (E87). Outcomes included in-hospital mortality, ICU escalation proxies (mechanical ventilation or shock), length of stay, hospitalization cost, and rescue outcomes defined as mortality among ICU, sepsis, and AKI hospitalizations. Results: The cohort included 748,734 weighted hospitalizations. The prevalence of toxicity-associated syndromes increased from 42.4% in 2016 to 53.4% in 2023. ICU escalation increased modestly from 4.0% to 4.3%, while inpatient mortality declined from 3.21% (95% CI, 3.06–3.37) to 2.53% (95% CI, 2.39–2.67). Compared with the transition era, the modern era demonstrated higher toxicity-associated hospitalizations (50.1% [95% CI, 49.8–50.4] vs 43.3% [95% CI, 42.9–43.8]) with minimal mortality difference (2.91% [95% CI, 2.85–2.97] vs 3.12% [95% CI, 3.02–3.23]). Mean LOS increased from 6.01 to 6.36 days, and mean cost increased from $21,318 to $24,547. Among toxicity-associated hospitalizations, mortality declined modestly from 6.16% (95% CI, 5.96–6.38) to 5.23% (95% CI, 5.12–5.33). Mortality among ICU hospitalizations remained high across eras (30.8% [95% CI, 29.7–32.0] vs 29.9% [95% CI, 29.3–30.6]). Conclusions: Among U.S. hospitalizations involving GI malignancies, toxicity-associated organ dysfunction and ICU escalation increased substantially in the modern treatment era, while inpatient mortality improved only modestly and remained high following critical illness. These findings highlight a growing inpatient care burden accompanying contemporary GI oncology treatment and underscore the need to align therapeutic advances with improved inpatient rescue capacity.
Online peer support for breast cancer survivors: A decentralized multicenter non-blinded parallel-group pilot randomized controlled trial (HOPE-BC study).
1668 Background: Peer support effectiveness for breast cancer survivors has reportedly varied, and no study has examined patient–supporter matching. This decentralized, multicenter, non-blinded pilot randomized controlled trial aimed to assess online peer support effectiveness and explore factors for a matching algorithm. Methods: Breast cancer survivors within three years of completing initial treatment were recruited and provided electronic informed consent. Based on the sample size estimation, the target sample size was set at 50. Participants were randomly assigned either to the peer support group, which scheduled an online peer support session immediately after registration, or to the waitlist group, which scheduled their session two weeks later. Participants could choose their preferred time and date through a web-based scheduling system. Peer support sessions were conducted online by two trained peer supporters. The primary outcome was the UCLA Loneliness Scale score, assessed one week after the session in the peer support group and at the end of the 2-week waiting period in the waitlist group. Secondary outcomes included satisfaction with peer support and other psychosocial measures, all collected through an electronic patient-reported outcome system. Effect sizes for outcomes were calculated to inform sample size estimation for future trials. Analyses were performed to explore factors associated with satisfaction. The study protocol was registered in the UMIN Clinical Trials Registry (UMIN000056741). Results: From February to July 2025, 54 participants enrolled and 52 individuals completed the study (27 peer support group, 25 waitlist group). At the primary endpoint, the mean (standard deviation) UCLA Loneliness Scale scores were 38.4 (11.2) in the peer support group and 41.3 (13.3) in the waitlist group (Cohen’s d = 0.24; t-test, P = 0.40). The change from baseline corresponded to a Cohen’s d = 0.32. Among the secondary outcomes, the changes in Generalized Anxiety Disorder-7 scores differed between groups (mean difference [MD] = -1.3; 95% CI = -2.5 to -0.1), and the changes in Patient Health Questionnaire-9 scores tended to be greater in the peer support group (MD = -1.2; 95% CI = -2.5 to 0.1). The global health status on EORTC QLQ-C30 did not differ between groups (MD = 1.0; 95% CI = -6.6 to 8.7). The median (IQR) satisfaction score was 46 (41.5, 50) on a 50-point scale. A linear mixed-effects model indicated that sharing ≥1 topic and age proximity between the participant and peer supporter were significantly associated with higher satisfaction. Conclusions: Results demonstrate the feasibility of future decentralized randomized trials on online peer support and provide effect size estimates to guide sample size planning. Shared topics and age proximity between participants and peer supporters may be considered when scheduling sessions.
Alliance A072301: A phase III trial of radiotherapy followed by adjuvant temozolomide in combination with vorasidenib vs placebo in <i>IDH</i> -mutated newly diagnosed grade 3 astrocytoma (VORTEX).
TPS2103 Background: Treatment for newly diagnosed IDH-mutant astrocytoma, WHO grade 3, is based on the results of the CATNON study, demonstrating benefit of radiotherapy followed by 12 months of adjuvant temozolomide. While temozolomide and radiotherapy significantly improved survival, the prognosis for these patients remains limited and more effective therapies are required. This trial will evaluate the benefit of the addition of the IDH1 and IDH-2 inhibitor vorasidenib to adjuvant temozolomide versus placebo following standard radiotherapy to evaluate possible improvement in outcomes. Methods: Alliance A072301 (NCT07215910) is a multicenter, double-blinded phase 3 randomized study to determine if the addition of vorasidenib to adjuvant temozolomide significantly improves progression-free survival (PFS), based on blinded central review in patients with newly diagnosed, IDH-mutant astrocytoma, WHO grade 3. Key inclusion criteria include age ≥ 12 years old, histological confirmation of astrocytoma (absence of 1p/19q codeletion), WHO grade 3, presence of any IDH mutation, plan for radiation and chemotherapy and surgery within 6 months. The presence of CDKN2A/B homozygous deletion, spinal or leptomeningeal disease, as well as prior chemotherapy, cranial irradiation or IDH inhibitor therapy, is exclusionary. 408 patients will be randomized 1:1 to receive vorasidenib 40 mg or placebo daily with 12 cycles of temozolomide following radiotherapy and then continued as monotherapy until disease progression or unacceptable toxicity. Stratification factors include age (< 40 years old vs ≥ 40 years old) and residual disease (< 2 cm vs ≥ 2 cm). Tumor assessments with MRI will be every three months for the first two years, every four months for the next two years and then every six months thereafter as per blinded independent central review using the Response Assessment in Neuro-Oncology (RANO) 2.0 criteria. Key secondary endpoints include PFS by local review, time to next treatment, safety and tolerability, quality of life measures. Exploratory measures will include seizure frequency. The trial will have 85% power to detect a hazard ratio of 0.625 with one-sided type I error rate of 0.025. Enrollment for the trial commenced February 2026. Support: U10CA180821, U10CA180882, U24CA196171. https://acknowledgments.alliancefound.org.
Real-world outcomes of epcoritamab vs glofitamab in relapsed/refractory diffuse large B-cell lymphoma: A TriNetX retrospective study.
7037 Background: Glofitamab and epcoritamab are CD20-directed bispecific antibodies that engage CD3-positive T cells and are approved for relapsed/refractory diffuse large B-cell lymphoma (R/R DLBCL). While trials reported high response rates, real-world comparative data on survival, safety, and treatment complications are limited. We conducted a propensity score-matched analysis to compare the effectiveness and safety of glofitamab and epcoritamab in R/R DLBCL. Methods: We conducted a retrospective cohort study of adult relapsed/refractory diffuse large B-cell lymphoma patients treated with glofitamab or epcoritamab through January 2026 using the TriNetX network. Eligible patients had confirmed R/R DLBCL, ≥2 prior systemic therapies and ≥30 days follow-up. Data from 134 million patients across 70 US healthcare organizations were analyzed. Propensity-score matching was done for age, sex, race, comorbidities, prior CAR T, and prior therapies. Outcomes included overall survival, all-cause mortality, hematologic adverse events, CRS, ICANS, infections, and ICU admissions. Results: A total of 369 epcoritamab and 337 glofitamab patients met the inclusion criteria. After 1:1 propensity-score matching, 205 patients were included in each cohort. Median age was 68 years for glofitamab and 67.8 years for epcoritamab, with 62.9% vs 64.9% male. Median follow-up was 6.8 months vs 5.8 months. At 6 months, mortality was 29.3% versus 39.0% (RR 0.75, 95% CI 0.57-0.99; p=0.037), with estimated overall survival 68.1% versus 55.8% (HR 0.68, 95% CI 0.49-0.95; p=0.023). This benefit persisted at 12 and 24 months (estimated overall survival 54.0% vs 43.8% and 37.5% vs 33.3%, HR 0.73-0.75), though fixed-time risk differences were not significant. Any-grade CRS occurred in 22.4% vs 30.2% (RR 0.74, 95% CI 0.53-1.03; p=0.073), with grade ≥3 CRS not significantly different between cohorts. ICANS occurred in 7.3% vs 11.7% (RR 0.63, 95% CI 0.34-1.16; p=0.130), and tocilizumab use was comparable (20.0% vs 22.0%). Hematologic adverse events, including neutropenia (21.5% vs 26.8%), thrombocytopenia (34.1% vs 35.1%), anemia (35.1% vs 40.0%), and ICU admissions (19.0% vs 25.4%), were comparable and not significantly different. Infections were significantly lower with glofitamab (38.0% vs 48.8%, RR 0.78, 95% CI 0.62-0.98; p=0.028), supporting a favorable safety profile. Conclusions: In this real-world analysis, glofitamab and epcoritamab had similar safety profiles, with glofitamab showing lower early mortality and higher overall survival. Hematologic adverse events, CRS, ICANS, and ICU admissions were comparable, while infections were significantly lower with glofitamab. The selection of CD20×CD3 bispecific antibodies should consider infection risk and early survival outcomes until longer follow-up clarifies durability and late toxicities.
Window of opportunity study of the Notch inhibitor AL101 in Notch1 activated adenoid cystic carcinoma (ACC): Biological effects and biomarker correlates.
6119 Background: ACC with NOTCH1 pathway activation (ACC-N1) is associated with aggressive clinical behavior and poor outcomes. AL101, an inhibitor of gamma secretase-mediated Notch signaling, previously demonstrated modest clinical activity in metastatic NOTCH -mutant ACC. We conducted a window-of-opportunity study to evaluate the biological effects of AL101 in ACC-N1 and to identify biomarkers to inform rational combination strategies. Methods: Patients with ACC-N1 received AL101 (4 mg weekly) for 4 to 8 weeks prior to surgery. Eligibility required Notch1 pathway activation by cleaved Notch (NICD1) immunohistochemistry (≥ 70% nuclear staining). Treatment-related adverse events (AEs) were assessed per CTCAE v5.0 and radiographic response per RECIST v1.1. Whole-exome sequencing and RNA sequencing were performed on baseline samples, with paired analysis (pre- and post-treatment) in 12 patients. This report focuses on biomarker analysis; clinical endpoints and feasibility were previously presented. Results: 13 patients were enrolled between Nov/21 and Dec/23; 8 were newly diagnosed. The median number of AL101 doses were 6 (range: 4–7), the most common primary site was maxillary sinus (n=4). There were no grade 3-5 AEs. One patient achieved a partial response (ORR 7.7%), 11 had stable disease (including 2 with >20% tumor shrinkage), and one had progression in a non-target lesion. Post-treatment NICD1 expression was not significantly reduced (p=0.8). Genomic profiling revealed NOTCH1 activating mutations in 9/13 tumors and MYB-NFIB fusions in 8/13, with co-occurrence in 6 patients. MYB-NFIB fusion with MYB overexpression, irrespective of NOTCH1 mutation status, was associated with tumor shrinkage (p=0.02). Higher baseline NOTCH signaling activity by RNA sequencing (NOTCH signaling signature) correlated with greater tumor shrinkage (p=0.025). Although NICD1 IHC levels did not significantly change, AL101 treatment resulted in significant downregulation of NOTCH signaling activity, and the magnitude of signature reduction correlated with tumor shrinkage (p=0.037). Post-treatment transcriptomic analysis demonstrated upregulation of potentially druggable oncogenic pathways, providing biologic rationale for future combination therapeutic strategies. Conclusions: In this first window-of-opportunity study in ACC, AL101 demonstrated biological target modulation. Importantly, MYB overexpression and NOTCH signaling activity rather than NICD1 modulation or NOTCH1 mutation correlated with tumor shrinkage. These findings provide translational insights and support biomarker-driven development of rational combination strategies in NOTCH1-activated ACC. Clinical trial information: NCT04973683 .
Individualized estimation of benefit from adjuvant docetaxel plus S-1 in stage III gastric cancer using deep learning–based counterfactual survival analysis of the START-2 trial.
4095 Background: The START-2 trial demonstrated that docetaxel plus S-1 (DS) was superior to S-1 alone in terms of relapse-free survival (RFS) and overall survival (OS) as adjuvant chemotherapy for patients (pts) with stage III gastric cancer (GC) (J Clin Oncol 2019; Gastric Cancer 2022). We conducted the START-2 AR study, a retrospective analysis, to develop a neural network–based model for predicting RFS and estimating individual-level treatment benefit. Methods: Among 912 pts enrolled in the START-2 trial, 599 provided written informed consent to participate in the START-2 AR study. Pts were randomly divided into training (70%) and test (30%) cohorts. Missing data were addressed using multiple imputation. A deep learning–based Cox proportional hazards model was trained in the training cohort and evaluated in the test cohort. Using this model, we estimated individual-level treatment benefit among pts in the DS group, quantified as the difference in restricted mean survival time (dRMST) between counterfactual survival curves comparing DS with S-1 alone. Subsequently, a LightGBM regressor was applied to identify clinicopathological factors associated with treatment benefit using SHapley Additive exPlanations (SHAP). In an exploratory analysis, the top-ranked variables were used to identify a subgroup with limited additional benefit from DS. Results: Among 599 pts (training cohort, n = 419; test cohort, n = 180), 249 RFS events were observed (174 and 75 events, respectively). The pooled Antolini C-index was 0.686 (95% confidence interval [CI], 0.643–0.726) in the training cohort and 0.628 (95% CI, 0.565–0.697) in the test cohort. The estimated dRMST in the DS group was 2.71 months (95% CI, 2.58–2.84). SHAP analysis of dRMST indicated that a lower number of metastatic lymph nodes, a higher platelet count, a higher number of dissected lymph nodes, and differentiated histology were the key factors associated with a smaller dRMST, indicating limited treatment benefit. When continuous variables were dichotomized at their median values and pts were stratified by the cumulative number of these four factors, the subgroup with < 3 factors (n = 454) showed significantly improved RFS with DS compared with S-1 alone (hazard ratio [HR], 0.67; 95% CI, 0.51–0.88; p = 0.004). In contrast, no significant benefit was observed in the subgroup with ≥3 factors (n = 145; HR, 1.32; 95% CI, 0.70–2.50; p = 0.396). Conclusions: Using a deep learning–based counterfactual dRMST framework, we identified a subgroup of pts with resected stage III GC who derived limited additional benefit from adjuvant DS therapy. Stratification based on four clinicopathological factors may facilitate personalized de-escalation of DS treatment. External validation in independent cohorts is warranted. Clinical trial information: UMIN000011438.
Clinical outcomes among patients with metastatic non–small cell lung cancer (mNSCLC) who received immune checkpoint inhibitors (ICI) in first (1L) and second line (2L).
e20578 Background: The use of ICIs, as monotherapy (mono) or with platinum-based chemotherapies, is the 1L standard of care for patients with mNSCLC without actionable genomic alterations. However, many patients progress during or shortly after ICI therapy, and the benefit of ICI retreatment in 2L is not clear. This study investigated 2L outcomes among responders and non-responders to 1L ICI. Methods: This retrospective study used ConcertAI Patient360 NSCLC dataset and included patients who received any ICI from the start of 2L who also received an ICI in 1L (from 01/2020-02/2025), based on a progression-based line of therapy algorithm. Patients were defined as 1L responders (first disease assessment was complete/partial response) or non-responders (first disease assessment was progression). Patients with stable disease only or no response assessment were excluded. Median real-world overall survival (rwOS) and progression-free survival (rwPFS) were compared with Kaplan-Meier analysis, with the start of 2L as index date. Results: 459 patients met eligibility criteria: 243 (53%) were classified as responders and 216 were non-responders. Patients were non-Hispanic (88%), White (83%), and 49% were male. 71% had non-squamous NSCLC, and most had ECOG 0-1 (67%), were current/former smokers (90%), and treated in community centers (80%). Of the 79% of patients tested for PD-L1 prior to 2L start, approximately 25% had TPS ≥50%. Responders tended to be older than non-responders and have de novo mNSCLC (86% vs 79%). Treatments were similar among the two groups; overall, 43% received pembrolizumab (pembro)+platinum doublet as their first treatment, 22% received pembro mono. The most common 2L regimens were pembro mono (31%) and pembro+pemetrexed (19%), with 29 additional unique regimens observed. No significant difference was seen in median 2L rwPFS for responders (5.3 months, 95% Confidence Interval [CI]: 4.4, 5.9) compared to non-responders (4.6 months, 95% CI: 4.0, 5.5), or 2L rwOS for responders (12.9 months, 95% CI: 11.2, 14.9) versus non-responders (11.7 months, 95% CI: 9.8, 14.3). However, among the responders, patients with >2 months treatment-free interval from end of 1L to progression (n=63) had significantly longer median 2L rwOS (15.9 months, 95% CI: 14.2, 20.4) than patients who progressed earlier (n=180; 11.8 months, 95% CI: 9.3, 13.1; p=0.02). Conclusions: Response to 1L ICI was not associated with patient outcomes on 2L ICI. However, patients who maintained their response for >2 months after 1L ICI discontinuation had significantly longer rwOS than those who progressed earlier. Further research is needed to identify which patients with 1L ICI are most likely to respond to ICI in 2L and to compare outcomes with patients receiving other agents in 2L.
AVATAR: Efficacy of personalized tumorogram-based therapy in breast cancer established from patient-derived organoids.
TPS1157 Background: Despite therapeutic advances in advanced breast cancer (ABC), predicting drug efficacy for individual patients remains difficult. Patient-derived tumor organoids (PDTOs) could enable drug efficacy testing within a timeframe compatible with clinical decision-making. We hypothesize that personalized treatment based on PDTO-guided tumorograms can improve outcomes for patients with HER2-negative ABC. Methods: AVATAR (NCT06459791) is a single-arm multicenter phase II trial assessing tumor response rate within 6 months as the primary endpoint in patients with HER2-negative ABC treated with tumorogram-guided chemotherapy. A biopsy of a tumor lesion is performed and transferred to the laboratory for PDTO establishment. While waiting for the result of the tumorogram, the patient receives standard-of-care treatment. A drug screening of standard-of-care drugs is performed on the PDTO (~5-10 drugs per patient), with selection adapted to the clinical setting. A multidisciplinary team makes therapeutic recommendations based on the tumorogram. A tumorogram is considered informative if it identifies at least one chemotherapy drug that is considered effective in the PDTO model. When possible, patients with an informative tumorogram receive one of the recommended treatments as their next line of therapy. Otherwise, they receive standard treatment. Eligible patients include those with HER2-negative ABC who are treated with chemotherapy (endocrine-resistant for HR+/HER2-), have a performance status of 0-1, and have a tumor lesion that can be biopsied. Patients who have received more than three lines of chemotherapy in the metastatic setting or who have progressive brain or leptomeningeal metastases are excluded. We hypothesize that the prescription of chemotherapy based on an informative tumorogram will be clinically meaningful if the tumor response rate is at least 25% (p0). With a one-sided α of 9%, a power of 90%, and an alternative hypothesis of 45% (p1), 39 patients should be treated according to an informative tumorogram, and 110 patients need to be included (assuming a minimum of 40% informative tumorograms and accounting for dropouts). Enrollment began in December 2024 and is ongoing. Clinical trial information: NCT06459791 .
The use of generative AI tools by cancer patients and caregivers: A cross-sectional survey of awareness, attitudes, and experiences.
1612 Background: Generative artificial intelligence (AI) tools offer novel avenues for patient support, yet data on their adoption in oncology remain limited. This study evaluated the awareness, attitudes, and usage patterns of these tools among cancer patients and caregivers to inform clinical integration. Methods: A descriptive, cross-sectional survey was conducted at a tertiary oncology center. Adult cancer patients and caregivers completed an anonymous, structured questionnaire assessing sociodemographics, digital literacy, AI usage patterns, and perceptions. Multivariate logistic regression identified independent predictors of AI use. Results: A total of 400 participants (68% patients, 32% caregivers; mean age 59 years) were included. While 75% were aware of AI chatbots, only 24% used them for cancer-related information. Multivariate analysis revealed that university-level education (adjusted odds ratio [aOR] = 4.47, p < 0.001), younger age (≤55 years; aOR = 2.64, p < 0.001), and recent diagnosis (<1 year; aOR = 2.16, p = 0.003) were independent predictors of AI use (Table 1). Notably, 69.1% of users did not disclose their AI use to their physician. Despite this, no user reported decreased trust in their physician; in fact, 26.8% reported an increase. A substantial 73.3% of all participants expressed a willingness to use a physician-verified AI tool. Subgroup analysis revealed divergent usage patterns: caregivers used AI more regularly (weekly/daily: 54.3% vs. 30.6%; p = 0.022) and focused on researching medical terminology (62.9% vs. 38.7%; p = 0.022) and alternative medicine (34.3% vs. 16.1%; p = 0.040), whereas patients were significantly more likely to seek emotional support (22.6% vs. 5.7%; p = 0.032). Conclusions: Generative AI use among cancer patients and caregivers remains moderate and reflects a marked digital divide driven by age and education. The prevalence of “silent” AI use raises significant patient safety concerns related to the risks of unverified information. Nevertheless, the strong demand for physician-endorsed tools highlights a clear opportunity for clinical integration. Oncologists should proactively address AI usage to mitigate misinformation risks and ensure safe integration into clinical practice. Independent predictors of generative AI use for cancer information (n=400). Parameter aOR 95% CI p value Age (≤55 years vs. >55 years) 2.64 1.594 - 4.381 <0.001 Education level (University or higher vs. lower) 4.47 2.195 - 9.109 <0.001 Time since diagnosis (<1 year vs. ≥1 year) 2.16 1.289 - 3.625 0.003 Abbreviations: aOR, adjusted odds ratio; CI, confidence interval.
Association between adverse childhood experiences (ACEs) and common cancers in the All of Us database.
e24109 Background: ACEs are reportedly associated with increased cancer risk in adulthood, yet their relationship with specific cancer types is unclear. We investigated the association between high ACEs and the four most common cancers in the U.S: breast, prostate, lung, and colorectal cancer. Methods: We conducted a propensity-score-matched case-control study using the All of Us dataset v8 (n = 393,596 with EHR data). The sample included 36,367 adults over 18 years, who completed the 11-item Behavioral Risk Factor Surveillance System ACE module. Exposure was dichotomized into low (0–3 ACEs) and high (≥4 ACEs) scores, consistent with established thresholds. Outcomes were EHR-based diagnoses of breast, prostate, lung, and colorectal cancer. Participants were matched 1:1 on age, race, BMI, smoking, income, and education using nearest-neighbor matching (caliper = 0.2), yielding 7,624 matched pairs (n = 15,248) with good balance (SMD < 0.1). Multivariable logistic regression and Cox proportional hazards (PH) regression were conducted. Results: Among 36,367 participants (mean age 63.2 ± 15.2 years; 64.8% female; 90.6% White, 9.8% Black/Black-American, 1.5% Hispanic, 3.7% other race), high ACE prevalence was 30.5% (n = 8,495). On average, individuals with high ACE scores were younger (58.0 ± 14.8 vs 64.8 ± 15.0 years), more frequently females (75.0% vs 61.7%), non-White (15.3% vs 7.6%), smokers (46.5% vs 33.2%), less likely to be college graduates (57.5% vs 74.6%), and more likely to have BMI > 25 kg/m 2 (81.3% vs 74.8%). Although the group with high ACE scores had lower overall cancer prevalence (5.8% vs 8.2%; p < 0.001), PH analysis showed that high ACE scores were associated with earlier time to diagnosis after age 18 years for breast cancer [HR = 1.59, 95% CI: 1.40–1.81, p < 0.001] and lung cancer [HR = 1.73, 95% CI: 1.24–2.43, p = 0.001], but not prostate cancer [HR = 0.82, 95% CI: 0.67–1.01, p = 0.067] and colorectal cancer [HR = 1.08, 95% CI: 0.80–1.46, p = 0.609]. After matching, the prevalence of breast (aOR = 1.18, 95% CI: 0.93-1.50, p = 0.183), prostate (aOR = 0.85, 95% CI: 0.57-1.26, p = 0.415), and lung (aOR = 1.33, 95% CI: 0.76-2.36, p = 0.37) cancer was similar between ACE groups. However, colorectal cancer rates were lower in the high ACE group (aOR = 0.55, 95% CI: 0.33-0.92, p = 0.025). Conclusions: ACEs disproportionately impact women and non-White individuals, and associated socioeconomic and lifestyle risk factors for cancer. Lower overall prevalence of cancer in the high ACE group was likely related to the average younger age of that group, especially as the differences for breast, lung, and prostate cancer were attenuated by matching. Notably, high ACE scores were linked to earlier time to diagnosis for breast and lung cancer. However, high ACE was also associated with a lower likelihood of colorectal cancer, even after matching and adjustment, indicating that ACEs may impact the risk of varying types of cancers differently.
Real-world safety of adoptive cellular therapy with lifileucel in patients with advanced melanoma.
9510 Background: Although acute toxicities associated with adoptive cellular therapy using lifileucel are well described and largely occur within the first 2 weeks of treatment, adverse events beyond this early window remain incompletely characterized. We sought to characterize treatment-related adverse events (AEs) with emphasis on those starting or persisting ≥2 weeks after lifileucel infusion. Methods: We conducted a retrospective multicenter study across three centers (Moffitt Cancer Center, Massachusetts General Hospital, and Mayo Clinic). Patients with advanced melanoma treated with lifileucel were included. The primary endpoint was incidence and characterization of AEs occurring or persisting ≥2 weeks after standard of care lifileucel infusion. Results: Among 79 patients (median follow-up 11 months [95% CI, 9.6–15]), median age was 63 years (range 30–79), 59% were male (n = 47), 41% were female (n = 32), and ECOG PS was 0–1 in 98%. Patients received a median of 5 IL-2 doses (range 0–6). Of 79 patients, 78 experienced treatment-related adverse events that either persisted or occurred >2 weeks after cell infusion. A total of 308 delayed treatment-related adverse events were identified; 96 were new events (31%) and 212 (69%) were continued toxicities that started < 2 weeks after cell infusion. Lymphopenia was the most common delayed or persistent toxicity, occurring in a majority of patients (n = 68/79, 86%). Treatment-associated anemia (n = 59/79, 75%), fatigue (n = 56/79, 71%) and thrombocytopenia (n = 24/79, 30%) were also common. Viral infections occurred in 9.0% (n = 7), including CMV/EBV viremia (n = 1) and respiratory or gastrointestinal viral infections (metapneumovirus, parainfluenza, SARS-CoV-2, and norovirus). Hemophagocytic lymphohistiocytosis (HLH) occurred in 2 patients (n = 2/79, 3%) and was fatal in both cases. No other toxicity related deaths were noted. Autoimmune toxicities directly attributable to TIL included vitiligo in 9% (n = 7; median onset 41 days), uveitis in 5% (n = 4; median onset 26 days), and sensorineural hearing loss in 1% (n = 1; onset 28 days). All patients underwent TIL infusion in the inpatient setting; a total of 14 of 79 patients (18%) were re-admitted after discharge due to treatment-related toxicity (median time from infusion 19 days). Conclusions: In this multicenter real-world cohort, toxicities persisting beyond two weeks were largely driven by lymphodepleting chemotherapy–associated immunosuppression and cytopenias. These findings highlight the importance of structured post-discharge monitoring and support ongoing efforts to de-intensify lymphodepleting chemotherapy to improve safety while maintaining clinical benefit.
Efficacy and safety results of a prospective phase II study (IPUtrial) of inetetamab and pyrotinib in combination with utidelone for first- or second-line treatment of HER2-positive metastatic breast cancer.
1050 Background: Despite standard anti-HER2 regimens with chemotherapy, therapeutic challenges persist in HER2-positive metastatic breast cancer (MBC). Inetetamab (anti-HER2 monoclonal antibody) and pyrotinib (tyrosine kinase inhibitor) demonstrate established efficacy in advanced disease. Utidelone, a novel epothilone analog, exhibits antitumor activity with a differentiated safety profile. This phase II study assessed the triplet combination in first- or second-line HER2-positive MBC. Methods: This multicenter single-arm phase II trial enrolled HER2-positive MBC patients for first- or second-line therapy. Patients who have previously used trastuzumab must meet the followings: for first-line treatment, neoadjuvant therapy was effective, and relapse occurred over 1 year after the end of adjuvant therapy; for second-line treatment, progressed in previous adjuvant/neoadjuvant therapy or progressed within 1 year of completion of trastuzumab therapy; disease progression occurred after 6 months of treatment with trastuzumab in advanced setting. Patients with stable brain metastases were enrolled with or without prior local treatment. ECOG PS 0-1, and measurable disease (RECIST v1.1). Eligible patients received inetetamab (8 mg/kg cycle 1, then 6 mg/kg, IV, day 1), pyrotinib (400 mg, PO, daily), and utidelone (30 mg/m², IV, days 1-5) every 21 days until progression or intolerance. Primary endpoint was investigator-assessed ORR. Secondary endpoints included PFS, OS, and safety. Results: By data cutoff (December 31, 2025), the recruitment has been completed. 94 patients were enrolled. Median treatment cycles were 11 (range 4-57). 56 (65.9%) and 29 (34.1%) patients were in first- and second-line settings, respectively. 85 patients were evaluable for response with 6 CR, 64 PR and 7 SD. The ORR was 82.4% (95% CI, 72.6%-89.8%). Median PFS was 13.1 months (95% CI, 10.420-15.780). Median OS was not reached. The OS rate of 12-month, 26-month and 35-month was 94.6% (95% CI, 89.5%-99.7%), 85.2% (95% CI, 76.6%-93.8%) and 78.5% (95% CI, 67.7%-89.3%), respectively. Predominant Grade ≥3 TRAEs were diarrhea (28.2%) and peripheral neuropathy (4.7%). Other Grade ≥3 TRAEs included nausea (3.5%), transaminitis (2.4%) and neutropenia (1.2%). Treatment discontinuations due to TRAEs occurred in 6 patients (6.6%). No treatment-related deaths occurred. Conclusions: The regimen demonstrated clinically meaningful activity of this triplet regimen for the first/second-line treatment of HER2-positive MBC, evidenced by high ORR (82.4%), durable median PFS (13.1 months), with encouraging 35-month OS rate (78.5%). Predominant toxicities (diarrhea, neuropathy) were controllable through supportive measures and dose adjustments. These data support phase III evaluation of this combination strategy.
Safety signal dilution in cancer trials: Estimation of toxicities through term fragmentation.
e24134 Background: To harmonize global clinical reporting, standardized frameworks are used to classify the severity of adverse events (AEs). Despite standardization, a single AE can be represented by multiple distinct codes. When a clinical event is distributed across synonymous terms, the resulting 'signal dilution' may lead to the underreporting of toxicities and a distorted safety profile. We conducted a proof-of-concept analysis to evaluate a potential signal dilution in AE reporting, focusing on renal AEs in oncology trials. Methods: We systematically reviewed phase 3 oncology randomized controlled trials (RCTs) published in NEJM, The Lancet, and JAMA between 2016 and 2021. We assessed AE coding variability using the Common Terminology Criteria for Adverse Events (CTCAE) and Medical Dictionary for Regulatory Activities (MedDRA). Signal dilution was quantified by analyzing: 1) Minimum AE frequency reporting thresholds and 2) Term fragmentation, defined as the number of distinct terms used for clinically related events. Renal toxicity was used as a case study, aggregating related terms to compare pooled vs. reported incidence rates. Results: A total of 94 phase 3 RCTs yielding 107 unique treatment comparisons met the inclusion criteria. Renal AEs were reported in 46 (43%) comparisons. Median reporting thresholds for all-grade AEs were 10% (IQR 5-10) in primary manuscripts and 5% (IQR 1.5-10) in supplements. The most frequent related terms were blood creatinine increased (37.5%), acute kidney injury (37.5%), nephritis (25%), renal failure (16.7%) and renal impairment (14.6%). An average of 1.5 (range 1-4) related terms were used per trial; 29% (14/46) of trials utilized ≥2 related terms. Notably, there was a significant 40% increase in term fragmentation within experimental vs. control arms (mean 1.4 vs. 1.0 terms, respectively; p = 0.002). Pooling related terms increased cumulative renal toxicity incidence by an average of 6.3% (range 0-33.5%) relative to the single highest reported term. Furthermore, 22% (6/27) of trials did not report renal AEs because individual terms fell below the mandatory reporting threshold in the main text, despite the aggregate pooled signal exceeding those limits. Modeling a scenario with one additional related term under trial-specific thresholds indicates that the rate of unreported aggregate signals would increase to 74% (20/27) of trials and mask a cumulative AE incidence of 13.5% (range 0-43.5%). Conclusions: Current oncology trial reporting conventions lead to safety signal dilution. By splitting clinically related toxicities into distinct terms, the true incidence of AEs, particularly in experimental arms, is systematically underestimated. To ensure evidence-based risk-benefit trade-offs, regulatory and editorial standards should mandate the reporting of pooled, clinically related terms and lower the reporting thresholds for toxicities.
Characterizing patient outcomes and tumor genetics in perinatal early-onset colorectal cancer patients.
e15738 Background: The overall incidence of colorectal cancer (CRC) in the United States is declining, however, the incidence of early onset CRC (EOCRC, < age 50) has increased in the past decade. Hereditary cancer syndromes only explain a small proportion of EOCRC cases and ongoing research is investigating the epidemiologic characteristics of EOCRC compared to classical onset ( > age 50) CRC. The University of California San Diego has observed an increase in the number of EOCRC cases in the perinatal period (1 year before and after delivery) across the last 8 years, warranting further investigation into its unique molecular and genetic characteristics during pregnancy and its influence on patient treatment. Methods: We conducted a single-institution retrospective cohort study of EOCRC patients diagnosed in the perinatal period and age-matched controls. Endpoints include patient demographics, systemic therapy plans, molecular and genetic characteristics, and survival outcomes. Patient identification for inclusion and data collection is ongoing. Results: A total of 35 EOCRC patients were identified between 1/1/2018 and 3/1/2025; 9 (25.7%) were diagnosed during or within a year of pregnancy. The average age at diagnosis was 35.6 years old for cases and 37.3 years old for controls (p = 0.26). Since April 2021, across a 54-month period, a total of 4 patients were deceased, with an average survival of 22.8 months for cases vs 24.56 months for control (ns). One case (11.1%) presented with metastatic disease compared to 4 controls (15.4%). Of the 9 EOCRC perinatal patients, 7 received systemic therapy; 2 (22%) received treatment prior to delivery (both receiving a 5-Fluorouracil backbone), in one case (11%) systemic treatment was delayed until after delivery, and in 4 cases (44%) diagnosis was made after delivery (average 8.9 months after). Germline testing was done in 32 (91.4%) of patients. Two cases (22%) had germline mutations, in MLH1 and MSH2. Three controls (11%) had germline mutations, including BRCA1, APC, and MUTYH. The most identified somatic gene mutations were APC with 29 total mutations, 66% vs 46% patients contained at least 1 mutation in APC (10 mutations total in the perinatal group vs. 19 in controls), followed by KRAS with 17 total, 44% vs 50% (4 vs 13), TP53 with 15 total, 33% vs 46% (3 vs 12), SMAD4 with 6 total, 11% vs 19% (1 vs 5), ARID1A with 6 total, 44% vs 8% (4 vs 2), PIK3CA with 5 total, 22% vs 12% (2 vs 3), and PTEN with 3 total, 0% vs 12% (0 vs 3). Conclusions: The overwhelming majority of EOCRC patients undergo germline hereditary cancer screening. Nearly half of our perinatal cohort were diagnosed with CRC after delivery, indicating that there may be a delay in recognizing symptoms of CRC during pregnancy. In this small EOCRC cohort, some tumor mutations among perinatal cases appear over-represented (ARID1A) while others were under-represented (KRAS and TP53). Further data collection and reanalysis are ongoing.
First-in-human study of in vivo CAR-T therapy GT801 in adults with relapsed or refractory CD19-positive B-cell hematologic malignancies.
7079 Background: Despite high initial cure rates, patients with B-cell hematologic malignancies, including B-ALL and B-NHL, experience substantial relapse, with markedly reduced disease-free and overall survival after second-line therapy. GT801 is a first-in-human, in vivo CAR-T therapy that uses antibody-displayed lipid nanoparticle (LNP) to deliver CD19 CAR mRNA directly to T cells, enabling in vivo reprogramming without ex vivo cell manufacturing or lymphodepletion. We report initial safety and efficacy data from an open-label, single-arm phase I study (NCT07205315). Methods: GT801 was evaluated using a modified 3+3 dose-escalation design across three dose levels (0.5, 1.5, and 3.0 mg). Patients received up to 4 treatment cycles, each consisting of a single intravenous administration of GT801 on Day 1 of a 7-day cycle. The primary endpoint was safety, including treatment-emergent adverse events (TEAEs), graded per CTCAE v5.0. Secondary endpoints included 3-month overall response rate (ORR), best overall response (BOR), duration of response (DOR), progression-free survival (PFS), and overall survival (OS), assessed per Lugano 2014 criteria. Results: As of January 13, 2026, 3 patients were enrolled (median age 60 years; median three prior therapies). One patient received 3 doses at 0.5 mg, and two patients received 4 doses at 1.5 mg. All patients had B-NHL. Premedication included dexamethasone, NSAIDs, antihistamines. Most AEs were Grade 1-2. Common TEAEs included cytokine release syndrome (CRS), cytopenia, and transient liver enzyme elevations. Grade ≥3 events were hematologic toxicities and CRS. No neurotoxicity or organ failure was observed. At Week 4, all patients demonstrated a treatment response. Peripheral blood flow cytometry showed rapid CAR-T generation, with CD8 + CAR-T cells detectable within 4 hours post-infusion and peak levels on Day 1. Reprogramming efficiency reached up to 93% and was maintained with repeat dosing. Minimal off-target CAR expression was detected in monocytes. Complete peripheral B-cell depletion occurred within 4 hours after the first dose and was sustained for at least 7 days; after the third dose, B-cell depletion and CAR-T detection were observed in both peripheral blood and bone marrow. Conclusions: GT801 demonstrated a manageable safety profile and early signs of antitumor activity in heavily pretreated B-NHL patients. The ability to generate functional CAR-T cells in vivo without lymphodepletion supports further clinical evaluation of GT801 as a promising therapeutic approach. Clinical trial information: NCT07205315 . Research sponsor: Vivacta Biotechnology (Shanghai) Co., Ltd.
Influence of cannabis use on relapse and remission rates in patients with acute lymphoblastic leukemia (ALL).
e18526 Background: There is a paucity of data regarding the impact of cannabis on relapse and remission rates, as well as other clinical outcomes, in patients who received chemotherapy for acute lymphoblastic leukemia (ALL) based on real world data. This retrospective observational study aims to identify the potential effect of cannabis use on treatment, mood, and other clinical outcomes in patients with ALL who received chemotherapy. Methods: We queried the TriNetX Global collaborative network from January 2006 to December 2021 for patients aged 18 years or older diagnosed with ALL. Patients were stratified based on cannabis use. Propensity scores matching done for sociodemographics, other malignancies, lab values, and chemotherapeutic medications resulted in 287 patients in each cohort. The primary outcome was achieving remission from acute leukemia. Secondary outcomes were acute leukemia relapse, all-cause mortality, ICU admission, sepsis, acute kidney injury (AKI), chronic pain, anxiety disorders, and depression. Odds ratios obtained via conditional logistic regression were used to compare outcomes over a 4-year follow-up period. Results: Age at index event was 35.2 years in the cannabis cohort and 37.2 years in the control arm. The cannabis cohort was associated with a significantly lower odds of the primary outcome (OR 0.56; 95% CI: 0.35-0.91, p=0.018). Cannabis use was also associated with higher odds of some secondary outcomes, including acute leukemia relapse (OR 1.91; 1.05-3.47, p=0.031), chronic pain (OR 1.70; 1.03-2.80, p=0.035), and anxiety disorders (OR 2.59; 1.54-4.34, p<0.0001). No significant difference was seen between cannabis users and the control cohort for all-cause mortality (OR 1.18; 95% CI: 0.80-1.73, p=0.414), ICU admission (OR 1.13; 0.69-1.86, p=0.632), respiratory failure (OR 1.02; 95% CI: 0.60-1.74, p=0.945), sepsis (OR 1.10; 95% CI: 0.72-1.71, p=0.650) and AKI (OR 0.97; 95% CI: 0.58-1.62, p=0.899) and depression (OR 1.64; 95% CI: 0.98-2.75, p=0.059). Conclusions: Among patients with ALL who received chemotherapy, cannabis use was associated with a significantly lower odds of remission, higher odds of relapse, chronic pain, and anxiety disorders, but no association with all-cause mortality, depression, or respiratory failure.
Outcomes of malignant melanotic nerve sheath tumors (MMNST): A retrospective cohort study.
e23556 Background: MMNST is a rare and aggressive peripheral nerve sheath tumor characterized by melanocytic differentiation and frequently misdiagnosed as melanoma due to overlapping histologic and immunophenotypic features. MMNST is commonly associated with PRKAR1A loss-of-function mutations and Carney complex, though it can also occur sporadically. Due to its rarity, optimal management and prognostic factors remain poorly defined. We performed a retrospective cohort study to characterize clinicopathologic features, genomic alterations, treatment patterns, and outcomes in patients with MMNST. Methods: We retrospectively identified patients with pathologically confirmed MMNST treated at MD Anderson Cancer Center between 2016-2025. Clinical data including demographics, tumor location, stage, treatments received, recurrence patterns, and survival outcomes were abstracted from medical records. Survival outcomes were estimated using the Kaplan-Meier method. Results: A total of 13 patients with MMNST were identified. Median age at diagnosis was 39 years (range 29-72), and 61.5% were male. Primary tumors most commonly arose in the paraspinal (46.2%, n = 6), head/neck (30.8%, n = 4), thoracic (15.4%, n = 2), and abdominopelvic (7.7%, n = 1) locations. Next generation sequencing testing was completed in 84.6% (11/13) of cases, and 72.7% (8/11) had PRKAR1A loss-of-function mutation. 7.7% (n = 1) of patients had metastatic disease at presentation. Surgical resection was performed in 83.3% (10/12) with localized disease, with radiation and/or systemic chemotherapy administered in 66.7% (8/12). At a median follow-up of 29.3 months, 66.7% (8/12) of patients with localized disease at presentation developed local recurrence (2/12) or distant metastases (6/12), with median recurrence-free survival of 25.9 months (95% CI 15.1-29.7 months). 88.9% (8/9) of patients with metastatic disease received systemic therapy, and the most common regimen was ipilimumab plus nivolumab (7/9); responses to different regimens will be detailed at the meeting. Median overall survival from the time of distant metastasis development was 50.6 months. Median overall survival in the whole cohort was 81.4 months. Conclusions: MMNST is a rare malignancy most commonly arising in paraspinal and head/neck locations. Despite aggressive local management, most patients with initially localized disease experienced recurrence or metastasis by about two years. These findings highlight the need for improved local and systemic treatment strategies and prospective, collaborative studies for this rare malignancy.
Systemic anticancer therapy near the end of life: A binational multicenter analysis in Argentina and Uruguay.
e24047 Background: Systemic anticancer therapy (SACT) near the end of life is considered a negative quality-of-care indicator, associated with limited survival benefit, increased toxicity, higher health care utilization, and delayed palliative care integration. Despite international recommendations, SACT close to death remains common. Multicenter Latin American data across different health care settings are scarce; a binational approach may help identify modifiable factors to improve the quality and value of end-of-life cancer care. Methods: This retrospective multicenter binational observational study (Argentina–Uruguay) included adults with advanced cancer who died between January 2022 and December 2023 and received systemic anticancer therapy during the last 90 days of life. The primary objective was to describe SACT exposure within 30 and 90 days before death as markers of therapeutic intensity. Clinical and treatment-related variables, as well as between-country differences, were analyzed. Results: A total of 249 patients were included (Argentina n = 173; Uruguay n = 76). Clinical characteristics and SACT use by country are shown in Table 1. Median age at death was similar between countries; however, patients in Uruguay more frequently had preserved ECOG performance status and were receiving first-line SACT, whereas patients in Argentina more often had ECOG PS ≥2 and received SACT in later treatment lines. In Uruguay, end-of-life SACT was mainly concentrated in tumors traditionally considered chemo-responsive (breast, colon, lung, gynecologic), while in Argentina it was distributed across a broader range of tumor types, affecting older and more heavily pretreated patients. Conclusions: SACT near the end of life was frequent in this binational cohort. Variability between Argentina and Uruguay suggests differences in health care contexts and clinical decision-making processes, with patterns of treatment intensity not explained solely by performance status or line of therapy. The high proportion of patients receiving first-line SACT followed by death shortly thereafter may reflect late diagnosis or delayed recognition of poor prognosis in the Uruguayan cohort. Conversely, more frequent SACT use in later lines in Argentina may be influenced by longer therapeutic trajectories and delayed integration of palliative care. These findings highlight the need to improve prognostic assessment, shared decision-making, and earlier palliative care integration to enhance the quality and value of end-of-life cancer care. Clinical characteristics and use of TOES by countr. Characteristics Argentina (n=173) Uruguay (n=76) Median age at death, years (range) 63 (23–92) 60 (30–85) ECOG PS 0–1 at last SACT (%) 42.4 64.5 ECOG PS ≥2 at last SACT (%) 57.2 35.5 First-line SACT at last treatment (%) 25.4 75.0 SACT ≤30 days before death (%) 31.2 39.5 SACT ≤90 days before death (%) 100 100
Clinical and genetic spectrum of 59 Birt-Hogg-Dubé patients in a single-institution cohort.
e22663 Background: Birt-Hogg-Dubé (BHD) syndrome is a rare autosomal dominant disorder caused by germline pathogenic variants in the tumor suppressor gene FLCN , predisposing affected individuals to tumorigenesis. Clinically, it is characterized by fibrofolliculomas, pulmonary cysts with risk of spontaneous pneumothorax, and an increased risk of renal tumors. Early recognition and genetic confirmation are important to guide surveillance and family screening. Methods: We performed a retrospective review of 59 individuals with suspected or confirmed BHD disease evaluated and managed at Massachusetts General Hospital (Boston, MA) between 2008 and 2025. Demographic data, clinical manifestations, imaging findings, family history, and FLCN genetic testing results were extracted from medical records. Clinical BHD diagnoses were assigned based on the presence of at least two out of three characteristic clinical features when molecular confirmation was negative or unavailable. Results: The cohort included 59 individuals (44% male) with a median age of 55 years (IQR 44-68) at the time of data collection. FLCN genetic testing was performed in 55 of 59 individuals: 45 (82%) had a pathogenic/likely pathogenic FLCN variant, and 10 (18%) had no pathogenic variant identified. The remaining four individuals did not undergo genetic testing but were diagnosed clinically based on clinical manifestations and/or significant family history and were managed with BHD-directed surveillance. Overall, 42 of 59 individuals had clinical manifestations of the disease (71%), with a median age at first symptom or diagnosis of 38 years (IQR 25-50). Among those with clinical disease, 31 had a pathogenic/likely pathogenic FLCN variant (74%), and eight had no pathogenic variant identified (19%). Twenty-four had fibrofolliculomas/angiofibromas (57%), 20 had a history of pneumothorax (48%), and 18 developed renal tumors (43%). A positive family history of BHD or related manifestations was present in 47 of 59 individuals (80%), and 27 of 59 had genetically confirmed affected family members (46%). Consistent with this, the most common reason for referral was family history, followed by targeted or incidental germline detection and clinical suspicion of BHD. Conclusions: In this single-center cohort of individuals with suspected or confirmed BHD disease, renal tumors, pneumothoraces, and fibrofolliculomas were common among those with clinical disease. Most individuals with clinical disease carried a pathogenic/likely pathogenic FLCN variant. Family history was a major driver of diagnosis, emphasizing the value of genetic testing and longitudinal surveillance. Early recognition and coordinated management remain key to reducing BHD-related complications.
Update of the EORTC QLQ-Ovarian Cancer Module (EORTC-QLQ-OV28) phase I and II.
5596 Background: Ovarian cancer is the eighth most common cancer in women worldwide. The 5-year survival rate is 90.0% for early-stage disease and 30.2% for advanced disease. The introduction of new treatment options, such as PARP inhibitors, immunotherapy and bevacizumab, has led to improved recurrence-free intervals and an increased number of long-term survivors. However, new therapies have also brought new side effects that are not sufficiently covered by the EORTC QLQ OV-28, which was developed in 1997. There is therefore a strong clinical and research need to update the questionnaire to reflect contemporary treatments and their impact on QoL. This work presents the results of Phases 1 and 2 of the EORTC module development study updating the EORTC QLQ-OV28. Methods: The EORTC QLQ-OV28 module was adapted in accordance with the EORTC QLG guidelines for methodology for updating modules focused on identifying QoL issues relevant to ovarian cancer patients. This involved a systematic literature search, a review of questionnaires , investigator brochures, focus groups, and research group meetings (Phase 1).An issue was added to the list if it had a patient and/or HCP relevance rating of at least 2 (on a scale of 1 to 4) and a priority rating of at least 30%. The relevant issues were then translated into items, resulting in a preliminary questionnaire (Phase 2). Results: The mixed methods approach identified 57 issues related to QoL in ovarian cancer patients. An issue list was created and used for interviews with 58 patients and 31 HCPs, who rated each issue according to its relevance using a four-point response format: 'not at all', 'a little', 'quite a bit', 'very much' and priority (yes/no). Of the initial 57 issues, 20 (as rated by patients) and 34 (as rated by HCPs) met both inclusion criteria. Sixteen of these were covered by the existing EORTC QLQ-OV28, and five new issues were identified: weight loss, numbness in the hands and feet, anxiety, concerns about acceptance of their condition, and impaired sexuality. Twelve issues from the existing questionnaire did not fulfil all the inclusion criteria.In Phase II, these issues were translated into items and used to create a preliminary 39-item questionnaire to assess QoL in ovarian cancer patients. Conclusions: The results of Phases I and II indicate that the current EORTC QLQ-OV28 module does not adequately address all the relevant QoL issues for patients with ovarian cancer. Five new issues are not addressed by the existing module, and twelve existing issues did not fulfil all the inclusion criteria. Therefore, the EORTC QLQ-OV28 module should be updated. A preliminary questionnaire will be tested in Phase III of the EORTC module development process. Note: This document represents preliminary findings of the module development and is subject to revision; subsequent modifications of the module in the next stages may reflect significant changes.