Browse Articles

Genotype-guided liposome engineering for precision pulmonary therapy: Linking genetic variants to nanocarrier design and clinical translation

High‐Yield Engineering and Identification of Oxygen‐Related Modified Divacancies in 4H‐SiC

ABSTRACT Modified divacancies in the 4H polytype of silicon carbide (SiC) exhibit enhanced charge stability and spin addressability at room temperature, making them attractive for quantum applications. However, their low formation yield and lack of direct structural identification have hindered progress. Here, we demonstrate a controllable method for high‐yield engineering and identification of oxygen‐related modified divacancy color centers in 4H‐SiC via oxygen‐ion implantation. Based on their distinct optical and spin‐resonance characteristics, we experimentally resolve four types of modified divacancies. Furthermore, by measuring isotope‐resolved hyperfine interactions, we identify them as the four crystallographic configurations of oxygen‐vacancy (OV) complexes. Remarkably, single OV centers account for over 90% of the total defect population and exhibit superior optical properties and spin coherence compared with defects created by conventional carbon or nitrogen implantation. We characterize the zero‐phonon lines of these OV centers and reveal distinct temperature‐dependent behavior in spin‐readout contrast. By optimizing implantation dose and annealing temperature, we achieve high‐density ensembles and observe Rabi‐oscillation beating patterns associated with different orientations of basal‐type defects. These results establish a high‐yield route for scalable engineering of these four oxygen‐related modified divacancies in 4H‐SiC and clarify their atomic structure, opening new opportunities for solid‐state quantum technologies.

Cancer care experiences among sexual and gender minority adults: A qualitative descriptive study.

e13731 Background: Sexual and gender minority (SGM) individuals experience disparities across cancer screening, treatment, and survivorship, yet patient-centered data describing SGM experiences and preferences for affirming care remain limited. We explored cancer care experiences among SGM adults, including partner involvement, sexual health, and perceptions of affirming care across the cancer care continuum. Methods: We conducted a qualitative descriptive study using purposive and snowball sampling through an academic oncology program and an SGM support group in the urban Northeastern United States. Adults identifying as SGM with a history of invasive cancer completed a demographic survey and semi-structured telephone interview. Interviews were audio-recorded, transcribed verbatim, and analyzed inductively. Multiple investigators independently coded transcripts and developed themes through consensus. Results: Fifteen SGM participants were interviewed (10 lesbian women, 1 bisexual woman, 3 gay men, 1 transgender man). Mean age at diagnosis was 51.8 years (range 35–81). Diagnoses included breast (n = 11), prostate (n = 2), renal cell carcinoma (n = 1), and central nervous system lymphoma (n = 1). Four themes emerged. First, cancer detection often occurred through self- or partner-detection rather than routine screening, including one breast cancer identified incidentally following gender-affirming chest surgery. Second, partners played central roles in emotional support, advocacy, and healthcare navigation, though misidentification or limited acknowledgment occurred in clinical settings. Third, participants described treatment-related sexual health impacts, including substantial changes in libido, intimacy, and sexual functioning, with limited anticipatory counseling. Fourth, the perceived importance of affirming care varied across the care continuum, with clinical expertise prioritized during diagnosis and treatment, and affirming communication and SGM cultural competence increasingly valued during follow-up and survivorship. Conclusions: SGM patients prioritized high-quality oncologic care during initial treatment. Over time, affirming communication, partner inclusion, and attention to sexual health became increasingly important. Interventions promoting inclusive communication and routine sexual health discussions may improve equity in SGM cancer care. In regions with larger SGM populations, these findings suggest that specialized programs may be a feasible approach to delivering more inclusive and affirming care.

Patterns of artificial intelligence use among physicians in Mexico: Implications for oncology practice.

9002 Background: Artificial intelligence (AI) is increasingly incorporated into medical practice; however, physician-level factors associated with AI adoption and usage patterns, particularly among oncologists practicing in low- and middle-income countries, remain insufficiently characterized. This study aimed to describe patterns of AI use among physicians in Mexico and to explore physician-level factors associated with adoption. Methods: We conducted a cross-sectional, anonymous survey among physicians from multiple specialties in a tertiary care center in Mexico to assess AI use, professional activities (clinical care, research, and teaching), training interests, and perceptions regarding AI. The primary outcome was self-reported AI use. Associations between physician characteristics and AI use were evaluated using nonparametric statistics. Among oncologists, AI use patterns, training interests, and perceptions were summarized descriptively. Results: Between August and September 2025, 170 physicians completed the survey. Overall, 77.1% were < 40 years old, 50% were women, and 91.7% had ≤10 years of professional experience. Clinical specialties predominated (78.8%) over surgical specialties (21.2%). Most respondents were involved in clinical care (94.1%), followed by research (48.4%), and teaching (27.1%). AI use was highly prevalent (88.2%), with generative models being the most commonly used tools (80.0%). Physicians younger than 40 years reported significantly higher AI use than those aged ≥40 years (93.1% vs 71.8%, p<0.001) and were more likely to use AI in clinical practice (77.9% vs 43.6%, p<0.001). No significant age-related differences were observed for research or teaching activities. Use of AI-based data analysis tools varied by specialty and was more frequent among surgical compared with clinical specialties (63.9% vs 37.3%, p=0.004). Among oncologists (n=32), 87.5% reported AI use, primarily in clinical practice (75.0%) and research (50.0%), while use in teaching was less frequent (21.9%). Most oncologists expressed strong interest in formal AI training (84.4%) and ethical considerations (87.5%). The majority agreed that AI could improve quality of care and support clinical decision-making (95.2%), while emphasizing the need for continued physician oversight. Concerns regarding data security were common (84.7%), whereas concerns about loss of medical autonomy were generally moderate (36.4%). Conclusions: AI adoption among physicians in Mexico is strongly associated with age and professional activity. Oncologists demonstrate widespread AI use in clinical care and research, high interest in structured and ethical training, and a consistent emphasis on human oversight. These findings support the development of targeted educational and governance strategies to guide responsible AI integration in oncology.

Temporal trends in mortality involving esophageal cancer with comorbid psychoactive substance-induced mood disorders among adults aged ≥55 years in the United States, 1999–2023: A retrospective CDC WONDER analysis.

e16049 Background: Esophageal Cancer (EC) has strong association with mental and behavioral disorders caused by psychoactive substances, and a substantial portion of the U.S population is addicted to these substances. Due to their carcinogenic effects on mucosa, it significantly increases the prevalence of EC particularly, squamous cell carcinoma. Recent studies indicate delay in diagnosis of EC in psychoactive drug users. This study is aimed to analyze the combined effect of EC and psychoactive drug use to characterize national mortality trends with respect to geodemographic variables. Methods: To analyze the national mortality trends associated with EC (ICD-10: C15), and mental and behavioral disorders caused by psychoactive substances (ICD-10: F10-19), we used CDC WONDER database. Age Adjusted Mortality Rate (AAMR) per 1 million of adults aged ≥ 55 years were extracted stratified by gender, ethnicity, and demographic variables. Joinpoint regression software evaluated Annual Average Percentage Change (AAPC) and 95% Confidence Intervals (CI). Results: A total of 63,694 deaths were reported from 1999 to 2023 with overall AAMRs steeply increasing over the study period (AAPC: 11.2*, 95% CI: 9.0 to 13.4). The males exhibit more significant increase in AAMRs (AAPC: 11.2*, 95% CI: 8.9 to 13.5) as compared to females (AAPC: 8.5*, 95% CI: 4.8 to 12.4). Among different ethnicities, Non-Hispanic (NH) Whites experienced largest increase (AAPC: 12.3*, 95% CI: 9.9 to 14.7) whereas NH Blacks and Hispanics had no statistically significant change in mortality. Geographically, the Northeast region demonstrated highest increase in AAMRs (AAPC: 12.6*, 95% CI: 9.1 to 16.2) with inclining trends observed in rural counties (AAPC: 15.1*, 95% CI: 11.7 to 18.6). Conclusions: National mortality trends due to EC and psychoactive substances have substantially increased over the past two decades with most profound increase in males, NH Whites, Northeast region and rural counties. These alarmingly high mortality trends highlight the need for integrated cancer screening and control interventions, and substance abuse prevention strategies particularly targeting high risk populations. Variable Deaths AAMR 1999 AAMR 2023 AAPC (95% CI) Overall 63694 3.9 36.7 11.2*(9.0-13.4) Gender:FemaleMale 1085952835 1.67 11.966.2 8.5*(4.8-12.4)11.2*(8.9-13.5) Census region:NortheastMidwestSouthWest 12269201942082210409 2.444.74 30.856.933.628.1 12.6*(9.1-16.2)10.8*(2.8-19.4)9.9*(6.9-12.8)8.6*(6.0-11.4) Race:HispanicsNH WhitesNH Blacks 1782570275573 5.33.311.4 10.840.823.1 -0.4(-1.8-0.9)12.3*(9.9-14.7)2.7(-2.4-8.1) Urbanization (1999-2020):MetroNon-Metro 3734111641 1999 3.75 2020 34.359.8 11.0*(4.8-17.6)15.1*(11.7-18.6) *Indicates p < 0.05.

CURE AI as a predictor of TIGIT as a therapeutic target in pediatric cancers.

e22000 Background: Clinical trials are not able to study individual treatment effects as methods to do so have not yet matured into practical use. For decades, we have studied groups of patients enrolled on a trial, comparing groups of patients as large cohorts and losing signal for important, complex features that define individuals. Foundation models take a different approach. Using the CURE AI foundation model generated from clinical and multi-omics data from hundreds of thousands of patients, complex, non-linear biological patterns can be found in new datasets that can provide insights into disease biology. We previously identified a complex signature predictive of immunotherapy response relative to chemotherapy response by analysis of a set of non-small cell lung cancer clinical trials (Weiss et al., AI in Precision Oncology, 2025). The immunotherapy benefit signature, applied to other cancer types such as colorectal cancer, identified known immunotherapy biomarkers such as MSI status as predictors of immunotherapy response. In the current study, we theorized that we could apply predictors of immunotherapy response/nonresponse from adult clinical trials to pediatric cancer patients. Methods: CURE AI was utilized to define the clinicogenomic features of therapeutic benefit by comparing tiragolumab/atezolizumab (anti-TIGIT/anti-PD-L1) versus atezolizumab (anti-PD-L1) on the Phase II CITYSCAPE first-line metastatic non-small lung cancer trial. CURE AI was fine-tuned on CITYSCAPE patient clinicogenomic features using the methodology described in the above cited publication. The model defined an omics-based benefit score that predicts benefit to TIGIT/PD-L1 combination therapy over PD-L1 monotherapy that is valid in held-out adult non-small cell lung cancer datasets. From this prediction, we then asked whether we could use this score to stratify pediatric patients from the St. Jude Cloud Genomics Platform by predicted benefit to anti-TIGIT therapy based on their baseline tumor RNA-sequencing. Results: Pediatric solid tumors were classified as having more benefit to TIGIT/PD-L1 combination therapy or as having more benefit to PD-L1 monotherapy. Multiple pediatric tumors including neuroblastoma, Wilms tumor, osteosarcoma, and rhabdomyosarcoma had major proportions of patients who are predicted to benefit more from TIGIT-containing combination therapy than PD-L1 monotherapy. When comparing patients who are predicted to benefit more from TIGIT-containing combination therapy than PD-L1 monotherapy, specific molecular signatures that are predictive of TIGIT-specific benefit were defined. Conclusions: As an indication expansion strategy, our approach to connect completed adult clinical trials to pediatric cancer treatment prediction could improve pediatric cancer trial design strategies.

Integrated clinical and genomic analysis of mediastinal germ cell tumors (GCT) with somatic transformation (SM): A multi-institutional cohort study.

5029 Background: SM of GCT is a rare but lethal entity largely described in the context of testicular primaries. The distinct clinical and genomic features of SM that arise from anterior mediastinum are less defined. Methods: A multi-institutional study was conducted by integrating data from Indiana University and the MD Anderson Cancer Center, including patients(pts) with SM arising from mediastinal GCT treated between 2016 and 2025. Electronic health records were queried to obtain clinical, pathologic and genomic data when available. SM was classified as “de novo” if detected prior to disease relapse; all others were “evolved”. In the absence of elevated markers or concomitant GCT elements, isochromosome12p testing confirmed GCT ancestry. Descriptive statistics were used. Cox regression and Kaplan Meier analysis were used for overall survival(OS) and reverse KM for follow-up. OS measured from the time of transformation to date of last follow-up. The study was under protocol PA16-0736. Results: 40 pts were identified. All were male; median age at GCT diagnosis was 25 yrs (IQR:22-35); First detection of SM was by diagnostic biopsy in 14(35%) instances and surgical excision in 24(60%). Sarcoma was the most common histology detected in 33(82%), notably rhabdomyosarcoma and angiosarcoma- 8(20%) each. More than one SM pathology in 6(15%) pts and hematolymphoid SM was detected in 4(10%). Of 23/40 (57%) pts with genomic testing, 17(74%) were performed on the primary specimens, 5(22%) on metastatic deposits and 1(4%) on peripheral blood. TP53 and PTEN/AKT pathway mutations were seen in 16(70%) and 14(61%) respectively. De novo SM was seen in 32(80%). 9(28%) had baseline metastatic disease. Elevated AFP and beta-hCG in 28/32(84%) and 14/32(44%); 6(19%) were marker negative. 24(75%) received conventional GCT regimens. All marker negative pts had histology-driven therapy. Response data in Table. 26(81%) received surgical consolidation. For full cohort, median follow-up was 2.9 yrs, 2yr OS- 62%(95%CI:46.9-81.9) and 5yr OS- 52%(95%CI: 35.6-75.4). For de novo SM, surgical consolidation was associated with longer survival (mOS 8 yrs vs 0.6 yrs, HR-0.09, 95%CI:0.02-0.3, p-0.0003). Conclusions: The vast majority of mediastinal SM are sarcomas, that can be detected de novo. Conventional tumor markers are frequently elevated at presentation and may reflect concomitant non-teratoma GCT. TP53 and PTEN/AKT pathway mutations characterize the most common alterations and can be detected in primary and metastatic specimens. Response to first-line therapy GCT Therapy (n= 24) (%) Histology driven therapy (n=7) (%) Complete Remission (without resection) 4 (17) 1 (14) Partial Remission 8 (34) 3 (42) Stable Disease 5 (22) 2 (29) Progressive Disease 6 (26) 1(14) Not available 1 (4) Marker Response Elevated at presentation 23 (96) 1 (14) Normalization of markers 15 (63) 1 (14)

Incorporating tumor pathological subtype into a multiparameter model: Improved accuracy in predicting prostate cancer extraprostatic extension.

e17117 Background: Extraprostatic extension (EPE) at radical prostatectomy (RP) is associated with worse biochemical recurrence-free survival and informs nerve-sparing decisions in localized prostate cancer. Existing nomograms and MRI-based scoring systems provide moderate accuracy, and no prior model has incorporated tumor pathological subtype. This study evaluated whether adding pathological subtype improves EPE prediction beyond MRI-based parameters. Methods: We retrospectively analyzed 343 patients who underwent preoperative multiparametric MRI and RP (August 2023-August 2025). EPE was present in 138 patients. Cases treated after January 2025 formed a temporal validation cohort (n=82), with earlier cases assigned to training (n=183) and test (n=78) cohorts. Candidate predictors included MRI-based parameters, clinical features, and pathological subtype. Logistic regression–based models and machine-learning algorithms were developed; model selection was based on multivariable performance. Results: The final model included pathological subtype (acinar adenocarcinoma, PAA), ESUR score, PSA density (PSAD), capsular contact length (CCL), and positive core rate. The model demonstrated superior discrimination compared with MRI-based systems, achieving an AUC of 0.829 in the training cohort, 0.795 in the test cohort, and 0.849 in the temporal validation cohort. Decision-curve analysis showed greater net clinical benefit across threshold probabilities. SHAP analysis identified pathological subtype as a major contributor to prediction. Conclusions: A pathology-centric multiparameter model incorporating PAA improves the accuracy and generalizability of EPE prediction. Pathological subtype contributes independent predictive value beyond MRI-based measures. This model may assist preoperative risk stratification and guide nerve-sparing decisions in patients undergoing radical prostatectomy. Comparison of AUC values among different models. Model Training Set AUC Test Set AUC Temporal Validation Set AUC EPE Grade 0.758 0.698 0.731 ESUR Score 0.774 0.745 0.765 Our Model 0.829 0.795 0.849 EPE Grade=MRI-based EPE grade. ESUR Score=The scoring scheme proposed by the European Society of Urogenital Radiology. AUC=Area Under the Curve.

Immune checkpoint blockade (ICB) in advanced/metastatic leiomyosarcoma (LMS): Real-world cohort from MD Anderson.

e23545 Background: LMS is an aggressive soft-tissue sarcoma with an immunologically “cold” tumor microenvironment and historically poor responsiveness to ICB. With limited benefit from cytotoxic therapy beyond first-line and low ICB monotherapy activity, defining real-world ICB outcomes and identifying patients (pts) who may benefit remain an unmet clinical need. Methods: Adults with advanced/metastatic LMS treated with ICB at MD Anderson (2010–2025) were retrospectively identified via the pharmacy database. Response was assessed per trial RECIST v1.1 when available; otherwise, partial response/stable disease/progressive disease (PR/SD/PD) were extracted from documentation with radiology impressions. Disease control rate (DCR) was defined as PR+SD, and clinical benefit rate (CBR) as PR or SD ≥6 months (mo). Toxicity was graded per CTCAE v5. Subgroups were compared using chi-square/Fisher’s exact tests. Progression-free survival (PFS) was estimated by Kaplan–Meier, compared by log-rank, and evaluated with Cox regression. Results: Among 41 pts (median age 59; 76% female), 90% (n = 37) had metastatic disease at ICB start, 59% (n = 24) had ≥2 prior systemic therapy lines; 90% (n = 37) had doxorubicin; 76% (n = 31) had gemcitabine. Metastases involved lung in 83% (n = 34), liver in 66% (n = 27), and bone in 32% (n = 13). Primary sites were uterus in 29% (n = 12), retroperitoneum in 39% (n = 16), and other soft-tissue sites in 32% (n = 13). Overall, 85% (35/41) received ICB on a trial and 90% (37/41) received combination therapy: 57% (21) dual ICB, 24% (9) investigational targeted combinations, 11% (4) with concomitant radiotherapy, and 8% (3) with TKI. Best response was PR in 2 pts (5%), SD in 24 (59%), and PD in 15 (37%), yielding DCR 63% and CBR 22%. CBR differed by LMS type: 0/12 uterine LMS (uLMS) vs 9/29 (31%) for soft-tissue LMS (ST-LMS; p = 0.04) and by primary site (retroperitoneum 19% [3/16] vs other sites 46% [6/13]; Fisher’s exact p = 0.008). Median PFS was 3 mo overall and was longer with DCR vs PD (4.5 vs 1.5 mo; p < 0.001). PFS was longer with lung metastases (4.1 vs 1.3 mo; p < 0.001) and shorter with bone metastases (2.7 vs 4.1 mo; p = 0.02) and with prior gemcitabine exposure (2.8 vs 4.5 mo; p = 0.03), reflecting heavier pretreatment. In multivariable Cox regression, lung metastases (aHR 0.06; p < 0.001), bone metastases (aHR 4.8; p < 0.001), prior gemcitabine exposure (aHR 2.4; p = 0.04), and trial participation (aHR 0.2; p = 0.002; reflecting selection/regimen differences) remained independently associated with PFS. Toxicity occurred in 51%, with grade ≥3 events in 24%. Conclusions: In this real-world cohort of ICB-treated advanced LMS, trial participation and lung metastases predicted better PFS while bone metastases and prior gemcitabine exposure predicted worse PFS. ST-LMS had greater CBR than uLMS. Translational efforts are ongoing to identify LMS-specific biomarkers of ICB response.

Septicemia in acute lymphoblastic leukemia: Disparities in incidence, resource utilization, and mortality between pediatric and adult populations.

e18510 Background: Acute lymphoblastic leukemia (ALL) in adults has substantially worse outcomes than pediatric ALL, with cure rates of 40-50% versus 80-90% respectively. Both groups are at high risk for infection during treatment, particularly during prolonged neutropenia. We aim to analyze age-based differences in sepsis incidence, clinical profiles, and outcomes in ALL patients. Methods: We identified active ALL admissions via the National Inpatient Sample (NIS) between 2016 and 2022, excluding patients in remission and those with COVID-19. Patients were stratified as adults (≥18 years) and children (<18 years). We evaluated the incidence of septicemia and compared demographics, vasopressor use, palliative care encounters, all-cause mortality, length of stay, and total hospital charges between adults and children with ALL-associated septicemia. Results: Among 254,380 patients with active ALL (108,570 children and 145,810 adults), septicemia incidence was significantly higher in adults (14.12% vs 5.77%, p<0.001). The study included 26,855 ALL patients with septicemia (pediatric: 6,265; adult: 20,590). Adults were more likely to be covered by Medicare (28.3% vs 0.2%, p<0.001) while children predominantly had Medicaid coverage (50.8% vs 25.4%, p<0.001). Adults were more frequently admitted to rural hospitals (2.0% vs 0.5%) and urban non-teaching facilities (8.4% vs 1.1%), while pediatric patients were more often treated at urban teaching centers (98.4% vs 89.6%, p<0.001). Race distribution differed significantly (p<0.001), with adults having higher proportions of White (49.4% vs 43.0%) and Black patients (8.7% vs 6.9%) while pediatric patients had more Hispanic patients (39.2% vs 30.7%). Sex distribution was similar (44.2% vs 44.4% female, p=0.938). Weekend admission rates and median household income showed no meaningful differences. Adults had higher rates of vasopressor use (7.9% vs 2.7%, p<0.001) and palliative care encounters (14.6% vs 4.7%, p<0.001). Pediatric patients had longer hospital length of stay (24.6 vs 16.9 days, p<0.001) and higher total charges ($610,623 vs $337,771, p<0.001). Pediatric patients had significantly lower mortality (9.26% vs 17.03%, p<0.001), with adults having 1.60 times the odds of mortality (aOR 1.60, 95% CI 1.27-2.01, p<0.001). Conclusions: Our study revealed that adults with ALL had higher incidence of sepsis and higher odds of mortality when compared with children with ALL. Despite longer hospital stays, the pediatric cohort had favorable outcomes signifying differences in biology, treatment approaches and access to specialized leukemia care warranting strategic steps to intervene.

Demographic patterns and survival outcomes in acute myeloid leukemia with t(6;9)(p23;q34); DEK–NUP214.

e18553 Background: Acute myeloid leukemia (AML) with t(6;9)(p23;q34); DEK–NUP214 is a rare, adverse-risk cytogenetic subtype characterized by aggressive clinical behavior and poor outcomes. Due to its low incidence, representing 0.6 - 1.7% of AML cases in children and 1% of AML cases in adults, population-level survival patterns and demographic predictors of prognosis remain poorly defined. We aimed to evaluate overall survival (OS), cancer-specific survival (CSS), and the impact of demographic and socioeconomic factors in this high-risk AML subtype. Methods: Adult patients with AML harboring t(6;9)(p23;q34) (DEK–NUP214) were identified from 17 SEER registries between 2000 and 2021. Cases with microscopic confirmation and non-missing survival data were included. Variables analyzed included age at diagnosis, sex, race, household income, and year of diagnosis. OS was defined as death from any cause and CSS as death attributable to AML. Kaplan–Meier methods were used to estimate survival. Multivariable Cox proportional hazards models were performed to evaluate associations with OS and CSS. Results: A total of 83 patients were included. Median survival was 17 months (IQR 9–36; mean 29.3 months). Females comprised 51% of the cohort. Most patients were White (75%), and the majority were in the middle-income category (56%), followed by high-income (38%) and low-income (6%). During follow-up, 47 patients (57%) died from any cause and 39 (47%) experienced AML-related death. Age ≥60 years was associated with worse overall survival (HR 2.56, 95% CI 1.33–4.91; P=0.0048) and cancer-specific survival (HR 2.15, 95% CI 1.06–4.37; P=0.035). Sex was not significant. Patients from racial/ethnic groups other than White showed a non-significant trend toward poorer OS (HR 1.36, 95% CI 0.66–2.81; P=0.401) and CSS (HR 1.55, 95% CI 0.71–3.38; P=0.269). Middle- and low-income patients did not differ significantly from high-income patients. Later year of diagnosis showed a modest, non-significant improvement in CSS. Conclusions: AML with t(6;9)(p23;q34); DEK–NUP214 is associated with early mortality. Advanced age remains the strongest predictor of adverse outcomes, while race and income showed non-significant trends. Findings should be interpreted cautiously due to the limited SEER sample and warrant validation in larger cohorts. Multivariable Cox regression for overall and cancer-specific survival in t(6;9) AML (DEK-NUP214). Characteristic HR (95% CI, OS) P value (OS) HR (95% CI, CSS) P value (CSS) Age ≥ 60 vs < 60 2.56 (1.33-4.91) 0.005 2.15 (1.06-4.37) 0.035 Male vs Female 0.73 (0.39-1.37) 0.328 0.72 (0.36-1.43) 0.353 Racial/ethnic group other than White vs White 1.36 (0.66-2.81) 0.401 1.55 (0.71-3.38) 0.269 Income: Middle vs High 1.43 (0.72-2.81) 0.306 1.37 (0.65-2.86) 0.408 Income: Low vs High 1.12 (0.3-4.14) 0.869 0.87 (0.18-4.14) 0.864 Year of diagnosis (per year) 0.93 (0.83-1.03) 0.168 0.91 (0.81-1.03) 0.127

Development and early implementation of a statewide comprehensive childhood cancer network to support pediatric oncology care and research.

10056 Background: Although pediatric cancer survival has improved substantially, families continue to face fragmented access to navigation, clinical trials, survivorship care, and psychosocial support. Florida’s pediatric oncology system is geographically dispersed and resource-variable, highlighting the need for coordinated statewide infrastructure. Methods: The Live Like Bella Comprehensive Childhood Cancer Network (CCCN) was established as a statewide, multi-institutional initiative to address gaps in access, care coordination, and research participation. Phase 1 activities emphasized infrastructure development, stakeholder engagement, and needs assessment. A caregiver survey was distributed to 2,478 families who received pediatric cancer support services between April 2024 and August 2025. Descriptive analyses evaluated barriers to care, information gaps, survivorship education, and clinical trial awareness. Missing survey responses were handled using complete-case analysis for each item, with denominators reported per question. In parallel, CCCN convened children’s hospitals, academic centers, advocacy organizations, and researchers to develop collaborative frameworks, digital navigation tools, and consortium-based initiatives. Results: Over 170 caregivers responded. Eighty percent reported significant non-clinical care challenges (e.g., housing, utilities, transportation), 67% reported work-related disruptions, and 52% identified unmet caregiver support needs. Thirty-three percent reported difficulty locating services. Only 50–53% rated information about childhood cancer and diagnosis-specific education as comprehensive. Forty-eight percent of families reported being offered a clinical trial; among those, 71% participated. Survivorship education gaps were common, with 22.7% and 26% reporting limited or no information on post-treatment and survivorship care. CCCN established leadership and advisory governance, launched development of a digital navigation and clinical trial matching portal, hosted a statewide research symposium (>250 participants), and initiated multi-institutional grant proposals focused on precision oncology and navigation workforce development. Conclusions: Early CCCN implementation demonstrates the feasibility of a statewide, family-centered model to identify care gaps, support research participation, and align pediatric oncology stakeholders. Ongoing evaluation will assess impacts on access, survivorship, and clinical trial engagement.

Competing risks of relapse and major organ failure after allogeneic hematopoietic cell transplantation.

e18579 Background: Outcomes after allogeneic hematopoietic cell transplantation (allo-HCT) are often summarized using overall survival (OS) or treatment related mortality (TRM), which may obscure distinct biologic and toxicity driven pathways of failure. Major organ toxicities are important causes of morbidity and mortality but are rarely evaluated as competing events alongside relapse. We sought to characterize relapse and major organ failure as distinct competing pathways after allo-HCT. Methods: We analyzed adult allo-HCT recipients from the CIBMTR RT17-01 publicly available dataset (2008-2016). The primary endpoint was the first post transplant failure pathway, defined as relapse, major organ failure (MOF) requiring dialysis or diagnosed as thrombotic microangiopathy (TMA), veno-occlusive disease (VOD), or idiopathic pneumonia syndrome (IPS), or death without prior relapse or MOF. Competing risk methods estimated cumulative incidence, and cause specific Cox models evaluated associations between conditioning intensity and outcomes, adjusting for clinical and transplant related covariates, with the renal component of the hematopoietic cell transplantation comorbidity index (HCT-CI) removed. Results: The cohort included 13,187 adult recipients with a median age of 58.6 years, of whom 59.3 percent were male. Conditioning intensity was myeloablative (MAC) in 6,134 patients, reduced intensity (RIC) in 4,795, and non myeloablative (NMA) in 2,221. The first failure pathway was relapse in 5,157 patients, MOF in 1,511, and death without relapse or MOF in 2,535. At one year, cumulative incidence was 32.4 percent for relapse, 10.4 percent for MOF, and 13.1 percent for death without either event. IPS and dialysis requiring renal failure were the most frequent MOF events. Compared with MAC, RIC was associated with higher relapse risk (adjusted hazard ratio [aHR] 1.14; 95 percent confidence interval [CI] 1.07-1.23) and lower MOF risk (aHR 0.72; 95 percent CI 0.64-0.82). NMA conditioning showed a further shift toward relapse (aHR 1.42; 95 percent CI 1.30-1.54) with greater reduction in MOF risk (aHR 0.55; 95 percent CI 0.46-0.66). Conclusions: Relapse and MOF represent distinct competing pathways of failure after allo-HCT. Conditioning intensity influences which pathway predominates, with less intensive regimens shifting risk from organ toxicity toward relapse. Pathway based competing risk analyses provide insight beyond traditional survival endpoints and may support more individualized transplant decision making.

Prognostic validation of artificial intelligence (AI)–based Stratipath breast risk stratification in TAILORx.

555 Background: Stratipath Breast is an AI-based prognostic medical device for risk stratification of early-stage breast cancer using routine H&E-stained histopathology whole slide images (WSIs). AI-based analyses of histopathology slides offer an alternative to costly and logistically demanding genomic assays. In this study, the prognostic performance of Stratipath Breast was validated in the TAILORx trial (NCT00310180). Methods: WSIs from patients enrolled in TAILORx were analyzed by Stratipath Breast. After histopathology quality control and availability of clinical endpoints and WSIs, 5,519 patients were included. Stratipath Breast binary risk category, multi-level risk group, and continuous risk score were evaluated. The prognostic performance was analyzed by Kaplan-Meier statistic and log rank test, as well as concordance index (C-index). Multivariable Cox Proportional Hazard (PH) model adjusting for age, tumor size, histologic subtype and continuous Oncotype DX recurrence score (RS), both without and with histologic grade, was used to assess independent prognostic value. Recurrence-free interval (RFI) and distant recurrence-free interval (DRFI) were evaluated. Results: 53.9% (2,975/5,519) of patients were classified as Stratipath low risk and 46.1% (2,544/5,519) as high risk. A significant association of Stratipath risk category and group with RFI and DRFI was confirmed (p < 0.05). In multivariable Cox PH analyses, Stratipath high-risk category was an independent prognostic factor associated with worse outcomes (RFI HR = 1.54, 95%CI:1.29-1.84, p < 0.05; DRFI HR = 1.61, 95%CI: 1.30-1.99, p < 0.05). Stratipath risk category remained significant in multivariable analysis when histologic grade was included, whereas grade did not. RS remained prognostic significant in multivariable analyses together with Stratipath Breast, indicating potentially complementary prognostic information. C-index improved with inclusion of Stratipath Breast together with clinical variables and RS, exceeding that of grade. Conclusions: In the TAILORx trial Stratipath Breast was found to provide significant prognostic value. Stratipath Breast also provided independent prognostic value in multivariable analysis adjusting for standard clinicopathologic factors and RS. These findings support the clinical relevance of AI-driven morphology-based biomarkers for breast cancer risk stratification.

Hearing/vestibular dysfunction and longitudinal changes in psychosocial well-being among breast cancer survivors.

12084 Background: Breast cancer treatment, in some cases, can cause hearing/vestibular dysfunction and negatively affect patients’ quality of life. The breast cancer survivor population is both growing and aging. Despite evidence that other cancer survivors have higher rates of hearing/vestibular problems than the general population, little is unknown about the impact of hearing/vestibular dysfunction on long-term changes in psychosocial well-being (PSWB) among breast cancer survivors. Methods: We surveyed the Chicago Multiethnic Epidemiologic Breast Cancer Cohort regarding whether patients had ever been told by a doctor that they had tinnitus, hearing loss, and/or vertigo between July and September 2023. PSWB was assessed repeatedly using the 10-item PROMIS on psychological distress/social isolation in 2020, 2021, 2022, and 2025. Total scores range from 0 to 40, with higher scores reflecting better PSWB. We fit separate linear mixed-effects models for hearing/vestibular dysfunction, controlling for age, race/ethnicity, marital status, income, insurance, comorbidity, AJCC stage, and treatment modality. Adjusted coefficients ( β ) and 95% CIs were calculated. Results: 1,466 breast cancer survivors were included: the mean age at survey was 63.5 years (SD 11.7); 69.6% identified as White, 22.6% as Black, 3.7% as Asian or Pacific Islander, and 3.6% as Hispanic. Overall, 16.5%, 17.4%, and 8.6% had tinnitus, hearing loss, and vertigo, respectively. The distribution of PSWB scores and trend over time are presented in the table. Survivors with hearing/vestibular dysfunction had lower mean scores than those without. After covariate adjustment, tinnitus ( β –1.12; 95% CI –1.93, –0.30), hearing loss ( β –1.12; 95% CI –1.93, –0.30), and vertigo ( β –1.34; 95% CI –2.44, –0.24) were associated with significantly lower PSWB scores. In all subgroups, the PSWB score increased significantly over the years. Furthermore, Hispanic ethnicity, Medicaid, and being single were correlated with lower scores; whereas higher income levels were associated with higher scores. Conclusions: In this diverse cohort, breast cancer survivors with hearing/vestibular dysfunction experienced suboptimal PSWB over the study period. PSWB also varied by socioeconomic status. Oncology programs should consider screening for hearing/vestibular dysfunction and unmet sensory healthcare needs to improve PSWB and quality of life among breast cancer survivors. PSWB score, mean (SD) Tinnitus Hearing loss Vertigo Survey year No Yes No Yes No Yes 2020 28.7 (6.0) 28.4 (6.0) 28.8 (6.0) 28.2 (5.9) 28.7 (6.0) 28.3 (6.1) 2021 29.4 (5.6) 28.9 (5.9) 29.4 (5.6) 29.0 (5.5) 29.5 (5.6) 27.8 (6.0) 2022 30.1 (5.6) 29.2 (6.1) 30.0 (5.7) 29.7 (6.0) 30.1 (5.6) 28.5 (6.5) 2025 33.3 (5.6) 32.1 (6.4) 33.3 (5.6) 32.4 (6.2) 33.2 (5.6) 31.8 (6.3) Trend per year ( p -value) 0.89 (<0.001) 0.75 (<0.001) 0.87 (<0.001) 0.86 (<0.001) 0.89 (<0.001) 0.66 (<0.001)

Efficacy and safety of giredestrant (GIRE) in patients (pts) with estrogen receptor–positive, HER2-negative early breast cancer (ER+, HER2– eBC) in the phase III lidERA BC clinical trial: Results by menopausal status.

502 Background: The lidERA BC trial (NCT04961996) demonstrated a statistically significant, clinically meaningful improvement in invasive disease-free survival (IDFS) with GIRE (a next-generation oral SERD and full ER antagonist) vs standard-of-care endocrine therapy (SOC ET) in ER+, HER2–, Stage I–III eBC (Bardia SABCS 2025). We report efficacy and safety in pre-menopausal (PRE-M) vs post-menopausal (POST-M) pts. Methods: Pts with ER+, HER2– eBC who had BC surgery and (neo)adjuvant chemotherapy (if indicated) were randomized 1:1 to once-daily 30 mg GIRE or SOC ET (anastrozole/letrozole/exemestane [aromatase inhibitors; AIs] or tamoxifen [TAM]) for 5 years (y) of treatment (tx). PRE-M pts on GIRE and on AIs received an LHRH. Results: In the efficacy-evaluable population (n = 4170), 40.7% of pts were PRE-M and 59.3% POST-M. 58.0% of pts on TAM received LHRH. PRE-M pts generally had a slightly higher baseline risk of recurrence vs POST-M pts (high risk, 70.5% vs 66.4%; medium risk, 28.4% vs 32.5%; higher use of [neo]adjuvant chemotherapy, 91.2% vs 78.9%). GIRE showed IDFS benefit and higher 3-y rates in both PRE-M and POST-M subgroups; a similar trend was observed for distant recurrence-free interval (DRFI) (Table). Adverse events (AE) were comparable in both subgroups and across txs (Table). Fewer pts discontinued GIRE due to AEs compared with pts receiving an AI, regardless of menopausal status. Discontinuation due to musculoskeletal pain occurred in 1.6% vs 4.5% of pts receiving GIRE vs AI, respectively. More pts switched to alternative ET in the SOC ET arm vs GIRE (10.1% vs 5.7%), mainly due to AEs for both arms. Conclusions: Adjuvant GIRE improved IDFS and DRFI vs SOC ET; benefit was consistent irrespective of menopausal status. PRE-M pts experienced a 42% reduction in the risk of developing metastatic disease and POST-M pts a 24% reduction. Safety was comparable between menopausal groups and there were few discontinuations, regardless of menopausal status, although PRE-M and POST-M pts receiving GIRE had fewer tx discontinuations than those receiving an AI or TAM in the PRE-M and POST-M settings. Clinical trial information: NCT04961996 . GIRE PRE-M n = 849 SOC ET PRE-M n = 838 GIRED POST-M n = 1220 SOC ET POST-M n = 1236 IDFS hazard ratio (HR) 0.65 0.74 3-yr rate, % 94.0 91.5 91.3 88.3 DRFI HR 0.58 0.76 3-yr rate, % 95.5 92.6 93.6 91.6 PRE-M n = 1672 POST-M n = 2440 Pts, % GIRE n = 840 SOC ET n = 832 SOC ET (AI) n = 563 SOC ET (TAM) n = 269 GIRE n = 1209 SOC ET n = 1231 SOC ET (AI) n = 1183 SOC ET (TAM) n = 48 All Grade/Grade 3–4 AE 95.5/18.2 95.3/17.3 96.3/16.9 93.3/18.2 94.5/20.7 93.7/18.4 94.1/18.6 85.4/14.6 AE leading to tx discontinuation 4.9 7.8 7.5 8.6 5.7 8.5 8.2 16.7 Musculoskeletal pain* 58.1 53.5 59.7 40.5 50.8 53.6 54.6 29.2 HRs are unstratified vs SOC ET. *Includes other related terms.

Characteristics of patients with resectable early-stage non–small cell lung cancer treated with neoadjuvant chemoimmunotherapy or surgery ± adjuvant therapy.

8061 Background: The integration of immune checkpoint inhibitors (ICIs) into neoadjuvant, perioperative, and adjuvant treatment strategies for early-stage NSCLC (eNSCLC) has been associated with improved patient outcomes; however, despite these advances, uptake of ICIs in the neoadjuvant/perioperative setting remains limited. This study aimed to characterize differences in patient characteristics between those receiving neoadjuvant chemoimmunotherapy (neoadjuvant ICI+CTx) and those receiving upfront surgery ± adjuvant therapy. Methods: Adult patients with stage II-IIIB(N2) NSCLC (AJCC 8 th edition) diagnosed between March 2022 and six months before data cut-off (November 2025) were identified in the Flatiron Health database. Patients who received ICI+CTx prior to surgery were assigned to the neoadjuvant ICI+CTx group, and patients that underwent upfront resection following an NSCLC diagnosis with no previous treatment, with or without adjuvant therapy, were assigned to the surgery± adjuvant therapy group. The Pearson chi-square or fisher exact tests were conducted to evaluate differences in baseline characteristics. All statistical analyses were two-sided. Results: A total of 1,807 patients were identified with resectable stage II-IIIB(N2) NSCLC, of which 204 (11%) initiated neoadjuvant ICI+CTx and 1,603 (89%) received surgery± adjuvant therapy (583 (36%) with adjuvant therapy; 1,020 (64%) without adjuvant therapy). Of the 204 patients, 144 (71%) underwent surgery and 76 (53% of those undergoing surgery) subsequently received adjuvant therapy. Significant differences were observed in the baseline characteristics between the two groups. Patients who received neoadjuvant ICI+CTx had higher proportion of stage III disease, cN2 status, evenly split histology, higher biomarker (PD-L1, EGFR) testing, PD-L1 positive expression levels, and fewer EGFR mutations while patients who received surgery± adjuvant therapy had higher proportion of stage II disease, cN0 status, non-squamous histology, lower biomarker testing, negative or low PD-L1 expression levels, and higher EGFR mutations (all p < 0.001). Conclusions: Despite the clear benefit across clinical trials of neoadjuvant and perioperative ICI strategies to treat resectable eNSCLC, their utilization in clinical practice remains limited. This study further demonstrates significant differences in baseline characteristics that might contribute to patient selection for neoadjuvant ICI+CTx versus upfront surgery ± adjuvant therapy as a primary therapeutic approach. Tailored interventions to further educate the medical community on recent clinical trial evidence may facilitate wider implementation of neoadjuvant and perioperative ICI strategies and broaden appropriate patient selection in resectable eNSCLC.

Long-term cognitive impairment after CAR-T therapy compared with autologous stem cell transplantation.

7032 Background: Chimeric antigen receptor T-cell (CAR-T) therapy has transformed the management of relapsed and refractory hematologic malignancies, challenging the established role of autologous stem cell transplantation (ASCT). However, the chronic neurocognitive effects of CAR-T therapy remain unclear. This study compares the long-term risk of cognitive impairment between CAR-T and ASCT therapy recipients. Methods: We conducted a retrospective cohort study using US hospital data from the TriNetX research network. Patients exclusively receiving either CAR-T therapy or ASCT were propensity score-matched 1:1 and adjusted for age, sex, primary cancer diagnosis, major comorbidities, neuropsychiatric comorbidities, dexamethasone use, and laboratory values (BMI, LDH if available). Participants with prior cognitive impairment diagnosis within 1 year before index or who received multiple treatment modalities during the study period were excluded. The primary endpoint was time-to-event cognitive impairment, defined by ICD-10 codes for mild cognitive impairment, memory loss, and difficulty with concentration. Secondary outcomes included time to neurological dysfunction (delirium, encephalopathy, seizures), mood and stress-related disorders (depression, anxiety, sleep disorder), fall-related injuries, mobility impairment, and secondary malignancies. Results: After propensity score matching, 3,067 CAR-T patients were compared with 3,067 ASCT patients. CAR-T recipients demonstrated a 58% higher risk of cognitive impairment (HR 1.583; 95% CI 1.390-1.80) and a 57% higher risk of neurological dysfunction (HR 1.567; 95% CI 1.403-1.751) compared to ASCT patients. Encephalopathy was the most prominent individual complication, with CAR-T patients showing a 2-fold higher hazard compared to ASCT patients (HR 2.035; 95% CI 1.727-2.398). To account for acute ICANS, events occurring within the first 7 days were excluded, and CAR-T patients maintained higher hazard of cognitive impairment (HR 1.695; 95% CI 1.466-1.959). The risk was highest within the first 30 days (HR 4.224; 95% CI 3.227-5.528) and remained persistently elevated throughout the follow-up period, even at 2500 days (HR 1.761; 95% CI 1.535- 2.021). Among CAR-T products, ciltacabtagene autoleucel showed the highest hazard (HR 2.013; 95% CI 1.092-3.708), while tisagenlecleucel showed the lowest (HR 1.530; 95% CI 1.030-2.271). All p<0.001. Conclusions: Compared to ASCT, CAR-T cell therapy was associated with a significantly increased risk of long-term cognitive impairment and neurological dysfunction. These findings have important implications for incorporating neurocognitive assessment into treatment discussion, informed consent, and the need for long-term monitoring.

Establishing a rapid diagnostic clinic at a tertiary care cancer center: The Indiana University Simon Comprehensive Cancer Center (IUSCCC) experience.

e13602 Background: Rapid diagnostic clinics (RDCs) are designed to accelerate evaluation and treatment for patients with suspected malignancy. We assessed the performance of the newly established RDC at the Indiana University Simon Comprehensive Cancer Center (IUSCCC) with respect to referral volume, diagnostic yield, cancer distribution, and key time-to-care metrics. Methods: We conducted a retrospective cohort study of all patients referred to the IUSCCC RDC between November 1, 2021, and November 1, 2025. Patient demographics, cancer diagnosis status, primary tumor site distribution, initial treatment modality among patients diagnosed with cancer, and time-to-care intervals were summarized. Time metrics were reported as median (range) and included: suspected cancer finding to RDC appointment, referral to first RDC visit, suspected cancer finding to confirmed diagnosis, RDC visit to subspecialist consultation, and suspected cancer finding to treatment initiation. Results: A total of 697 referrals were evaluated. Median age was 63 years (range, 17-95), and 48.3% were male and 51.7% female. Racial distribution was 84.1% White, 12.6% Black/African American, 2.9% Asian, and 0.1% each American Indian/Alaska Native, other, or unknown. Overall, 361 patients (51.8%) were diagnosed with cancer, while 336 (48.2%) did not have a malignancy identified after comprehensive evaluation, and 7 of the latter were lost to follow-up. Among patients diagnosed with cancer, initial management included systemic therapy in 52.1%, surgery in 18.8%, hospice/palliative care in 11.6%, surveillance in 6.6%, and radiation therapy in 4.4%. Median time from suspected cancer finding to RDC appointment was 8 days (range, 0-281), and from referral to first RDC visit was 5 days (range, 0-42). Median time from suspected cancer finding to confirmed diagnosis was 17 days (range, 0-304), and to treatment initiation was 24 days (range, 2-267). Median time from RDC visit to subspecialist consultation was 15 days (range, 0-263). The most common primary tumor origins were thoracic (28.0%), gastrointestinal (20.8%), and hematologic (18.6%). Conclusions: In this large RDC cohort, approximately half of referred patients were ultimately diagnosed with cancer. The IUSCCC RDC demonstrated rapid access to diagnostic evaluation and timely initiation of treatment, with a median of 24 days from suspected cancer finding to therapy. These findings support the value of RDC models in expediting cancer diagnosis and care delivery at tertiary cancer centers.

Analysis of an oral anticancer repository program and the impact on access to care.

1506 Background: Patients with cancer face many barriers to timely access of oral anticancer agents, including financial toxicity and delays in insurance and patient assistance program (PAP) approval. The Ohio State University Comprehensive Cancer Center (OSUCCC) launched a pharmacist-led Oral Oncology Drug Donation Repository (OODDR) program in 2020 to allow patients to donate unused oral anticancer medications with the intent to re-dispense to other patients in need. The goal of this study was to describe our OODDR experience and its impact on access to anticancer care. Methods: A retrospective chart review was conducted for OSUCCC patients who received at least one dispense of oral anticancer medication from the start of the program in February 2020 through August 2025. Data were collected on patient characteristics, anticancer therapy, barriers to access, and long-term medication access outcomes. Results: A total of 392 medication access encounters were identified for 376 patients through OODDRR dispense reports. Four encounters were excluded for missing data or not starting treatment, leaving 388 encounters for analysis. The median age of patients was 65 years, majority were female (51%), white (83%), and had commercial insurance (38%), Medicare (39%), or were uninsured (18%). The most common treatment indications included breast (23%), leukemia (17%), lymphoma (15%), and gastrointestinal (13%) malignancies. The OODDR dispensed 53 unique medications, mostly for on-label indications (76%), with 91% of encounters dispensing one treatment cycle/month or less (median 14 days [IQR 10-28]). Patients accessed long-term drug supply primarily via manufacturer PAP (42%) or insurance approval (31%), while 28 patients (7%) received the full course of treatment with OODDR supply. Reasons that repository supply was utilized included (multiple may apply): delay in prior authorization or appeal process (25%), delay in drug delivery from pharmacy (22%), delay in PAP approval and/or supply delivery (30%), and loss or change of insurance/assistance (19%). The OODDR notably improved the time from treatment decision to initiation with a median start of 6 days (IQR 1-13 days) compared to 14.5 days (IQR 6-30 days) to non-repository drug access. The average retail cost of drug dispensed per encounter was $10,789.28. Utilizing repository supply when patients required inpatient treatment (19%) resulted in hospital savings of $207,654.78. In total, the OODDR provided more than $4.1 million worth of anticancer treatment at no charge. Conclusions: The OSUCCC OODDR is an instrumental resource for patients and care teams to remove barriers and improve access to anticancer medications. Patients were able to initiate therapy sooner using the OODDR than would have otherwise been possible using routine insurance and assistance programs. Future efforts to improve medication access and awareness of medication assistance platforms are vital.