Browse Articles

Adjuvant versus no adjuvant therapy after major pathological response to neoadjuvant immunotherapy in resectable melanoma: A Swedish nationwide population-based study (NEO-MEL).

9566 Background: Randomized trials have shown that neoadjuvant PD-1–based immunotherapy improves outcomes compared with adjuvant therapy alone in patients with resectable stage III–IV cutaneous melanoma. In trials, patients experiencing a major pathological response (MPR) rarely relapse, and in the NADINA trial, adjuvant therapy was omitted in this group. However, the need for continued adjuvant treatment after MPR has never been prospectively tested, and real-world evidence is limited. Methods: This population-based cohort study included all patients with macroscopic, resectable cutaneous melanoma treated with neoadjuvant immune checkpoint inhibitor(s) across all academic melanoma centers in Sweden between January 1, 2022 and December 30, 2025. Clinical and pathological data were extracted from medical records. MPR was defined as pathological complete response (pCR) or near-pCR. Recurrence-free survival (RFS) was estimated using the Kaplan–Meier method and differences assessed with log-rank test. Results: A total of 279 patients initiated neoadjuvant treatment; 254 (91%) underwent surgery as planned, and 251 had evaluable pathological assessment. Overall, 103 patients (37%) experienced pCR and 17 (6%) near-pCR, yielding 120 patients (43%) with MPR. At a median follow-up of 19 moths, the estimated 24-month RFS for patients with MPR was 92%. Among MPR patients, 52 (43%) received continued adjuvant immune checkpoint inhibitor therapy, while 68 (57%) received no adjuvant treatment. RFS differed significantly between groups, with estimated 24-month RFS of 96% in patients receiving adjuvant therapy versus 80% in those who did not ( P =0.048). In the non-adjuvant group, there were five recurrences (three had pCR and two had near pCR), two were local recurrences and three were distant metastatic events, compared with no recurrence events in the adjuvant group. One unrelated death occurred in the non-adjuvant group and two unrelated deaths in adjuvant group. Conclusions: In this nationwide real-world cohort, patients experiencing MPR after neoadjuvant immunotherapy had excellent outcomes overall. However, a small but clinically relevant risk of relapse was observed among patients who did not receive continued adjuvant therapy after MPR. These findings emphasize the need for individualized assessment when considering adjuvant treatment in patients with MPR. Longer follow-up is needed to determine whether additional events emerge and, importantly, whether relapses can be effectively salvaged with immune checkpoint inhibitor rechallenge.

Role of circulating miR-30a-5p and miR-93-5p as liquid biopsy biomarkers for prediction of prognosis in metastatic non-small cell lung cancer.

e20529 Background: Advanced non-small cell lung cancer (NSCLC) carries a poor prognosis, and reliable non-invasive biomarkers for prognostication remain limited. Circulating microRNAs (miRNAs), such as miR-30a-5p and miR-93-5p, have shown potential in reflecting tumour behaviour, miR-93-5p is often oncogenic, while miR-30a-5p functions as a tumor suppressor. This study aims to evaluate the prognostic value of circulating miR-30a-5p and miR-93-5p in patients with advanced NSCLC. Methods: In this mixed design study with its first part cross sectional and second part prospective cohort study, 60 patients with histologically confirmed advanced NSCLC and 30 healthy controls were enrolled. Serum levels of miR-30a-5p and miR-93-5p were measured using quantitative real-time PCR at diagnosis and after four cycles of platinum-based chemotherapy. Expression patterns were correlated with clinical and pathological parameters. Results: At diagnosis, miR-30a-5p was downregulated in 73.3% and upregulated in 26.7%, while miR-93-5p was downregulated in 60.0% and upregulated in 40.0%. After chemotherapy, downregulation persisted in 85.4% for miR-30a-5p and 54.2% for miR-93-5p. A significant association was found between miR-30a-5p upregulated expression and bone metastasis (p = 0.041). while miR-93-5p showed significant association with its upregulation and lymph node metastasis (p = 0.021). Also, significant association between miR-93-5P upregulation and poor therapeutic response, with higher expression levels predominantly observed in non-responsive patients (P=0.003). Kaplan-Meier survival analysis revealed that up-regulation of serum miRNA-30a-5p and miRNA-93-5p was significantly correlated with poorer patient survival (P=0.002 for both). On multivariate analysis, using Cox’s regression; the only independent factor that significantly affected the overall survival was miRNA-30a-5p with HR 4.1. Conclusions: miR-30a-5p and miR-93-5p shows promise as a prognostic biomarker particularly for predicting bone and lymph node involvement, predict treatment response and overall survival in advanced NSCLC. The persistent downregulation of both miRNAs supports their utility in disease monitoring.

Real-world experience with management of CRS and ICANS following CD3xCD20 bispecific antibody therapy.

e19015 Background: CD3xCD20 bispecific antibodies (BsAbs) have transformed treatment of relapsed/refractory B-cell lymphomas. Immune-mediated toxicities, including cytokine release syndrome (CRS) and immune effector cell–associated neurotoxicity syndrome (ICANS), remain challenging, with variable management. We characterized incidence, severity, timing, and management of CRS and ICANS. Methods: We performed a retrospective, single-center review of adult lymphoma patients treated with CD3xCD20 BsAbs at Mount Sinai Hospital between January 2021 and December 2024. CRS and ICANS were graded per ASTCT criteria. Exploratory comparisons between agents used Fisher’s exact test. CRS/ICANS-related mortality was defined as death with CRS or ICANS as the primary contributing cause. Results: Among 63 patients, 35 received glofitamab, 25 epcoritamab, and 5 mosunetuzumab. Median age was 65 years (IQR 57–73). Most patients had aggressive B-cell lymphomas (43/63, 68.3%), followed by indolent lymphomas (13/63, 20.6%) and other entities (7/63, 11.1%). CRS occurred in 35 patients (55.6%): grade 1 in 18 (28.6%), grade 2 in 13 (20.6%), and grade 3 in 4 (6.3%); no grade 4–5 events occurred. Grade ≥2 CRS occurred in 17 patients (27.0%). CRS incidence by agent was 60.0% with epcoritamab, 20.0% with mosunetuzumab, and 54.3% with glofitamab. All CRS and ICANS events occurred early after treatment initiation (median 1–3 days; range 1–7). Grade 1 CRS was managed with supportive care alone in 10 patients (55.6%), tocilizumab in 6 (33.3%), and corticosteroids in 2 (11.1%). For grade 2 CRS, tocilizumab and corticosteroids were each used in 10 patients (76.9%), with combination therapy in 9; supportive care alone was used in 2 patients (15.4%). All grade 3 CRS events were treated with tocilizumab and corticosteroids. ICANS occurred in 6 patients (9.5%): grade 1 in 1 (1.6%), grade 3 in 4 (6.3%), and grade 4 in 1 (1.6%). All ICANS cases occurred with concurrent CRS. ICANS incidence by agent was 14.3% with glofitamab (5/35), 4.0% with epcoritamab (1/25), and 0% with mosunetuzumab (0/5). All ICANS cases were treated with corticosteroids, and 5 patients (83.3%) received anakinra. There was a trend toward higher rates of grade ≥2 CRS and grade ≥3 ICANS with glofitamab; differences were not statistically significant (all p>0.05). Nine patients (14.3%) required ICU admission. Toxicity resolved in 32 of 35 patients (91.4%). CRS/ICANS-related mortality occurred in 3 of 35 patients (8.6%), all with aggressive B-cell lymphomas treated with glofitamab. Conclusions: CD3xCD20 BsAbs were associated with predominantly low-grade CRS and low ICANS incidence, with agent-specific differences. Severe CRS/ICANS requiring escalation was associated with ICU utilization and mortality, supporting standardized management pathways.

Outcomes of patients with rare cancers treated with combination immune checkpoint inhibitors with and without lung and/or liver metastases (NCI/SWOG S1609).

2633 Background: Limited prior retrospective data, primarily among patients with lung cancer, liver cancer, and melanoma, has indicated that liver metastases are associated with poorer outcomes with chemotherapy, targeted agents, and checkpoint inhibitor therapy. We used a unique clinical trial resource to evaluate whether liver and/or lung metastases were associated with outcomes among patients with rare tumors treated with combination anti-PD-1 and anti-CTLA-4 therapy. Methods: The basket trial, SWOG trial S1609 (NCT02834013, DART trial), treated 655 eligible patients without lung or liver primary cancers. Associations with progression-free and overall survival were evaluated with Cox regression models; multivariable models controlled for age at trial registration, sex, performance status, race, ethnicity, primary tumor organ. Associations with clinical benefit rate (confirmed complete and partial response or stable disease for six months or longer) and treatment-related adverse events were evaluated using Fisher’s exact test. Results: There was no significant difference in rates of grade 3 or higher treatment-related adverse events across the groups. Participants with liver metastases, with or without lung metastases, had a lower likelihood of experiencing clinical benefit (11% and 18%, respectively) compared to participants with lung but not liver metastases or participants with neither lung nor liver metastases (26% and 30%, respectively, p=0.003). On multivariable analysis, both lung and liver metastases were associated with shorter progression-free and overall survival compared to presence of neither liver nor lung metastases. Conclusions: In a diverse cohort of participants with advanced rare tumors treated with combination anti-CTLA-4 and anti-PD-1 therapy, we found that both lung and liver metastases were associated with shorter progression-free and overall survival. Multivariable Cox regression models for progression-free survival and overall survival. Covariate Progression-free survival Overall survival Lung but not liver metastases (N=175) (reference = neither liver nor lung, N=269) 1.40(1.13-1.72) 0.0019 1.24(1.00-1.55)0.053 Liver but not lung metastases N=154) (reference = neither liver nor lung, N=269) 1.73(1.38-2.17) <0.001 1.33(1.05-1.68) 0.017 Liver and lung metastases N=57) (reference = neither liver nor lung, N=269) 1.96(1.44-2.65) <0.001 1.91(1.40-2.61) <0.001 Liver but not lung metastases N=154) (reference = lung but not liver, N=175) 1.24(0.96-1.60)0.095 1.07(0.83-1.38)0.62 Liver and lung metastases N=57) (reference = lung but not liver, N=175) 1.40(1.02-1.92) 0.037 1.54(1.11-2.12) 0.009 Liver and lung metastases N=57) (reference = liver but not lung, N=154) 1.13(0.83-1.55)0.45 1.44(1.04-1.99) 0.030 Hazard ratio (95% confidence interval) p-value reported.

CAR-T in the older adults: Real-world survival and toxicity in patients ≥ 75 years with relapsed/refractory multiple myeloma.

7541 Background: CAR-T cell therapy targeting BCMA has transformed the treatment landscape for relapsed/refractory multiple myeloma (RRMM). However, pivotal clinical trials have limited representation of patients aged ≥75 years, leaving the safety and effectiveness of CAR-T therapy in this older population undercharacterized. In this study, we sought to evaluate the real-world outcomes of CAR-T therapy in patients aged ≥75 years with RRMM. Methods: A retrospective cohort study was conducted using the TriNetX database, which includes data from a large network of healthcare institutions in the US. Adult patients with RRMM who received CAR-T therapy were identified and stratified into two age based cohorts: ≥75 years and <75 years. Propensity score matching (1:1) was performed based on race, gender, comorbidities, and high risk cytogenetics if available. Outcomes included incidence of immune effector cell-associated neurotoxicity syndrome (ICANS), cytokine release syndrome (CRS), tocilizumab (Toci) use, sepsis, 30-day intensive care unit (ICU) admission rates, 90-day mortality rate, and 1 and 3-year overall survival (OS). Cox proportional hazards multivariate analysis was used to assess the impact of age on OS, controlling for confounders. Results: A total of 1,392 patients met study criteria, of which 198 patients were aged ≥75 yrs at the time of infusion. The mean age was 62.6 vs. 77.5 yrs in the younger and older cohort, respectively. Prior to propensity score matching (PSM), there were no significant differences in the rates of CRS, ICANS, Toci, sepsis, or 30-day ICU admissions rate post-infusion. Grade 3 or higher CRS/ICANS were infrequent in both cohorts. Additionally, the 90-day mortality rate did not differ significantly between the older and younger cohorts. Following PSM, these findings remained consistent (Table 1). While there was no significant difference in 1-year OS, the 3-year OS was lower in the ≥75 cohort (69% vs. 55.4%, log rank p=0.51), although the difference was not statistically significant. Cox proportional hazards multivariate analysis, controlling for race, gender, and comorbidities, revealed that age ≥75 was not independently associated with OS (hazard ratio [HR] = 1.04, 95%CI 0.65-1.65, p = 0.87). Conclusions: In this real-world cohort of RRMM patients receiving CAR-T therapy, age ≥75 years was not associated with differences in risks of CRS, ICANS, sepsis, 90-day mortality, or 1-year overall survival. While 3-year overall survival was lower in patients ≥75 years, possibly due to competing mortality risks, these findings overall suggest that age alone should not preclude BCMA-directed CAR T-cell therapy in select older patients. Outcome <75 years(n=197) ≥75 yrs (n=197) p-value CRS 53% 46% 0.27 ICANS 12% 8% 0.22 TOCI use 53% 51% 0.82 ICU admission 3.8% 5.1% 0.39 Sepsis 6% 7.9% 0.36 90 day mortality 3.9% 5% 0.28 1-yr OS 86.2% 86.6% 0.86 3-yr OS 69% 55.4% 0.51

Adoption and inpatient outcomes of chimeric antigen receptor T-cell therapy in lymphomas and leukemias: A national analysis, 2017–2023.

e19104 Background: Chimeric antigen receptor T-cell (CAR-T) therapy has transformed the management of relapsed or refractory hematologic malignancies. Despite increasing outpatient administration, CAR-T therapy remains frequently delivered in the inpatient setting, with substantial resource utilization. Contemporary national data describing real-world inpatient adoption patterns and outcomes across malignancy subtypes are limited. Methods: We analyzed the National Inpatient Sample (NIS) from 2017–2023. CAR-T therapy was identified using ICD-10-PCS procedure codes in any procedure field. The primary cohort included hospitalizations with a principal diagnosis of lymphoma or leukemia (ICD-10-CM C81–C86, C88, C90, C91–C95). Outcomes included in-hospital mortality, length of stay (LOS), and total hospital charges. Survey-weighted analyses accounted for clustering and stratification, with variance estimation using Taylor linearization at the hospital level. Stratified analyses were performed for lymphoma and leukemia admissions. Results: National inpatient CAR-T volume increased from 805 admissions in 2017 to 5,845 in 2021, with 4,310 admissions in 2023. In pooled analyses of lymphoma and leukemia admissions, CAR-T therapy was not associated with higher in-hospital mortality compared with non-CAR-T admissions (risk difference −0.36%, 95% CI −1.35% to 0.62%). CAR-T admissions were associated with significantly longer LOS (+15.0 days, 95% CI 14.0–16.1) and higher hospital charges (+$365,563, 95% CI $331,158–$399,968). In stratified analyses, CAR-T therapy was associated with lower inpatient mortality among lymphoma admissions (risk difference −4.66%, 95% CI −5.05% to −4.26%), while no mortality difference was observed among leukemia admissions. Increased LOS and hospital charges were observed in both strata. Conclusions: In a contemporary national cohort, inpatient CAR-T utilization expanded substantially following regulatory approval and was associated with markedly increased inpatient resource utilization. Inpatient mortality was not increased overall compared with non-CAR-T admissions, with important heterogeneity by diagnosis group. These findings provide real-world context for inpatient CAR-T delivery and may inform future strategies to optimize care pathways and resource allocation.

Effect of ripretinib on the pharmacokinetics (PK) of repaglinide, a sensitive CYP2C8 probe substrate, in adult patients (pts) with advanced gastrointestinal stromal tumor (GIST).

e23511 Background: Ripretinib is a switch-control tyrosine kinase inhibitor (TKI) indicated for the treatment of adult pts with advanced GIST who have received prior treatment with 3 or more kinase inhibitors, including imatinib. Ripretinib and its active metabolite, DP-5439, inhibited CYP2C8 in vitro and met the criteria for a clinical assessment. This open-label study evaluated the effect of ripretinib on the PK of repaglinide, a sensitive CYP2C8 probe substrate, in pts with advanced GIST who progressed on or had intolerance to 2 or more prior TKI therapies. Methods: A single oral 0.5-mg repaglinide dose was administered on cycle 1 day 1 (C1D1) to 13 pts in a fasted state (no food or drink except water for ≥6 hours predose). Ripretinib 150 mg once daily was administered beginning on C1D2. On C1D15, a single 0.5-mg repaglinide dose was coadministered with ripretinib. Ripretinib was continued until disease progression, unacceptable toxicity, or withdrawal of consent. PK samples were collected predose through 24 hours postdose on C1D1 and C1D15 and analyzed for repaglinide. PK parameters were calculated using noncompartmental analysis; ln-transformed PK parameters for repaglinide were compared using ANOVA with treatment (with ripretinib vs alone) as a fixed effect and pt as a random effect. Geometric mean ratios and 90% confidence intervals (CIs) for maximum concentration (C max ), area under the curve from time 0 to last quantifiable concentration (AUC 0-t ), and AUC extrapolated to infinity (AUC 0-∞ ) were computed. Patient safety was monitored. Results: Of 13 pts, 11 (85%) were PK evaluable. Repaglinide AUC 0-t and AUC 0-∞ increased by 24% and 26%, respectively, with ripretinib vs alone. Repaglinide C max and time to reach maximum concentration were comparable with ripretinib vs alone. All 13 pts (100%) had at least 1 treatment-emergent adverse event (TEAE); 4 pts (31%) had Grade 3 TEAEs, with 2 pts (15%) experiencing serious AEs that were Grade 3 and were not related to study treatment. Conclusions: Ripretinib is a weak inhibitor of CYP2C8, with no impact on absorption and a small, clinically insignificant increase in total repaglinide exposure. The safety profile of ripretinib was consistent with its established use in advanced GIST. This study provides the basis for approved concomitant use of ripretinib with CYP2C8 substrates. Clinical trial information: NCT04530981 . Plasma PK parameters of repaglinide by treatment. Parameter Repaglinide Reference n = 11 Repaglinide + ripretinib Test n = 11 Geometric mean ratios (90% CI) C max , pg/mL 9560 (39.5) 9150 (54.9) 0.96 (0.69–1.32) AUC 0–t , h·pg/mL 16,100 (24.9) 20,000 (32.0) 1.24 (1.05–1.46) AUC 0–∞ , h·pg/mL a 16,300 (26.1) 20,000 (32.6) 1.26 (1.05–1.51) Data presented as geometric mean (% geometric coefficient of variation). a n = 10.

Outcomes of hyperthermic intraperitoneal chemotherapy (HIPEC) with cytoreduction in peritoneal metastatic pancreatic adenocarcinoma: Survival benefit and impact of germline genetic mutations.

e16435 Background: Peritoneal metastasis in pancreatic ductal adenocarcinoma (PDAC) is historically associated with a dismal prognosis and limited response to systemic therapy alone. The role of locoregional therapy remains controversial. This study evaluated the survival impact and safety of HIPEC, with or without cytoreductive surgery (CRS), compared to standard-of-care systemic therapy in patients with peritoneal metastatic PDAC. Methods: We conducted a comparative analysis of 130 patients with peritoneal metastatic PDAC, comprising a study cohort treated with HIPEC (n = 33) and a control group treated with systemic chemotherapy alone (Non-HIPEC, n = 97). HIPEC patients received induction chemotherapy (predominantly FOLFIRINOX) followed by HIPEC ± CRS. Co-primary endpoints included 12 months survival, and 6 months progression free survival (PFS). Secondary endpoints evaluated Overall Survival (OS ), safety of CRS, and impact of germline mutation status on survival. Study was approved under IRB protocol 25-008762. Results: The 12-month OS rate was significantly higher in the HIPEC group compared to the chemotherapy-only control group (100% vs. 52.6%, P < 0.001). The 6-month PFS rate in the HIPEC group was 63.0% vs. 47.94 in-only control group (P < 0.001). For secondary endpoints, the HIPEC group demonstrated a significantly longer median OS of 33.0 months vs. 17.0 months for controls (P < 0.001). Within the HIPEC cohort, the addition of CRS (n = 25) significantly improved outcomes compared to HIPEC alone (n = 8): median OS improved from 21.5 to 36.0 months (P = 0.03). Germline testing was performed in 61% of HIPEC patients and 48% of controls, identifying pathogenic mutations in 25% and 17%, respectively. Identified pathogenic mutations in both groups included ATM (n = 4), MUTYH (n = 2), BRCA2 (n = 1), CHEK2 (n = 1), PALB2 (n = 1), and KIT (n = 1). In the control group, specific mutations drove divergent outcomes; ATM median OS: 23.5 months vs. MUTYH : 8.0 months (P = 0.04). Conversely, in the HIPEC+CRS cohort, survival was excellent regardless of genetic profile: mutation carriers achieved a median OS of 33.0 months compared to 31.0 months in wild-type patients (P = 0.85). Despite a 34.8% postoperative complication rate in the HIPEC group, there was no perioperative mortality. Conclusions: In selected patients with peritoneal metastatic PDAC, HIPEC provides a substantial survival advantage over systemic therapy alone, achieving a 100% 12-month survival rate and more than doubling the median overall survival. Importantly, germline mutation status did not preclude survival benefit, with mutation carriers deriving equal efficacy from the multimodal approach. These findings support HIPEC+CRS as a highly effective strategy that may overcome adverse biologic drivers.

Topoisomerase 1 and DNA damage: Pharmacodynamic responses and mechanism of trastuzumab deruxtecan in HER2- expressing advanced solid tumors.

3092 Background: Trastuzumab deruxtecan (T-DXd) consists of deruxtecan, a novel topoisomerase 1 (TOP1) inhibitor, covalently bound via a lysosomal protease-cleavable linker to the human epidermal growth factor receptor 2 (HER2)-targeting antibody, trastuzumab. The mechanism(s) underlying the clinical activity seen in multiple tumor types across various HER2 levels is of ongoing, active interest. To investigate the T-DXd mechanism of action, we developed a pilot clinical trial of T-DXd with a detailed pharmacodynamic analysis from TOP1 target engagement to downstream effects of DNA damage in patients with HER2-expressing (IHC 1-3+, HER2 amplified, or HER2 mutated) advanced solid tumors (NCT04294628). Methods: Based on preclinical studies modeling the trial, research tumor biopsies from consenting patients were collected at three time points: pre-treatment, post-dose Cycle 1 (48-96 hours) and pre-dose Cycle 3. The biopsies were evaluated for TOP1 inhibition, induction of stabilized TOP1covalent complexes (TOP1cc) and induction of downstream DNA damage repair (DDR) markers (RAD51, pNBS1, RPA32) using validated, quantitative multiplex immunofluorescence assays on fixed tumor sections with image analysis methodology. In addition, a retrospective analysis of Schlafen 11 expression in baseline biopsies was performed to determine its predictive value of tumor responses to T-DXd. Results: Twenty-one biopsy pairs (pre-treatment and C1, 48-96h post-dose 1) were evaluable for TOP1 molecular response and downstream DDR marker induction. TOP1 target modulation was detected in 15 (71%) of the on-treatment biopsies and markers of DDR were induced in 18 (86%) biopsy pairs. TOP1 target inhibition, robust induction of downstream DNA damage response including stalled replication fork progression, and DNA breaks were observed in HER2-amplified/2+/3+ tumors, and in HER2 1+ tumors. Conclusions: Our results confirm the intended TOP1 molecular mechanism of action of T-DXd in HER2-amplified/2+/3+ and importantly extend that finding to HER2 1+ tumors, resulting in DNA damage in nearly all cases. Pharmacodynamic biomarker studies are poised to yield important insights into the molecular effects of T-DXd in solid tumors.

Estimation of the impact of chemotherapy-induced biological age acceleration on breast cancer survivorship using multi-scale simulation modeling.

12121 Background: Treatment of breast cancer with chemotherapy improves breast cancer survival. But chemotherapy also increases the accumulation of senescent cells in breast cancer survivors, which induces more senescent cells through paracrine signaling. The increased accumulation of senescent cells increases inflammation and tissue damage, which leads to earlier onset of aging-related diseases in survivors. We developed a simulation model connecting the biology of aging with breast cancer epidemiology to estimate the impact of chemotherapy-induced biological age acceleration on breast cancer survivorship. Methods: We integrated a systems biology model of accumulation of senescent cells with chronological age and an established Cancer Intervention and Surveillance Modeling Network (CISNET) breast cancer simulation model to estimate the impact of biological age-acceleration on breast cancer survivorship outcomes. We used published clinical data on senescence biomarker expression level ( p16 INK4a mRNA expression) in chemotherapy recipient breast cancer survivors to simulate the elevated senescence expression level in the remaining lifetime of breast cancer survivors. The difference in the senescence expression level in the chemotherapy group was evaluated against the senescence expression level in the general female population to quantify the excess biological age, or the biological age acceleration, after chemotherapy. The modified biological age induced by chemotherapy was used to determine the hazard of non-breast cancer mortality in breast cancer survivors. We used the CISNET breast cancer simulation model to simulate multi-birth cohorts of US females diagnosed with breast cancer to estimate survivorship outcomes after chemotherapy. Outcomes included remaining life years after breast cancer diagnosis, absolute number of non-breast cancer deaths, and the time-point at which the risk non-breast cancer mortality starts to dominate breast cancer mortality. Results: Biological age acceleration after anthracycline-based regimes caused a greater loss of life years than anthracycline-free regimens for women diagnosed at ages 30-39 years (median of 11.7 years vs. 2.1 years lost). This loss in life years diminished with increasing age at diagnosis (median of 6.0 years vs. 1.4 years lost for 70-79 year old women). The risk of non-breast cancer mortality exceeded breast cancer mortality up to 10-15 years earlier than expected based on chronological age after anthracycline-based regimens. Conclusions: Integration of computational systems biology modeling with breast cancer population-level modeling helps to identify subgroups of breast cancer survivors who are likely to experience survivorship loss due to biological age acceleration and may need survivorship care targeting non-breast cancer mortality.

Cost-effective colorectal cancer screening in the United States: Comparative analyses of four molecular tests.

10525 Background: Colorectal cancer (CRC) screening in the United States is costly, largely driven by expensive colonoscopy procedures. In 2021, 22.3 million screening tests were performed, costing $27.5 billion; colonoscopy alone accounted for 9.2 million tests and $23.7 billion in expenditure (54.8% of the $43.2 billion spent on five major cancers). This study compared available molecular tests to identify an approach that balances performance and cost as a primary alternative to colonoscopy. Methods: Four molecular tests were evaluated: two FDA-approved stool-based (fecal immunochemical tests [FIT] and multitarget stool DNA test [Cologuard]) and two blood-based (the FDA-approved cfDNA-based Shield and the multiomics-based SimpleScreen). Performance data came from published literature and clinical trials (NCT04136002, NCT01397747, NCTO4369053). Cost analyses used 2021 U.S. screening volumes: FIT, Cologuard, and colonoscopy costs were based on 2021 fee schedules, Shield pricing on Medicare rates; SimpleScreen was excluded due to cost unavailability. Results: For CRC, Cologuard had the highest sensitivity (92.3%) but lower specificity (84.4%), whereas FIT showed lower sensitivity (73.8%) but the highest specificity (93.4%). Shield and SimpleScreen showed intermediate sensitivities (83.1%, 79.2%) and specificities (89.2%, 91.1%). For advanced precancerous lesions, blood-based tests showed minimal detection power, with post-test probabilities (equal to PPV) close to pre-test prevalence (Shield 17.4% vs 14.3%; SimpleScreen 13.5% vs 9.5%; post/pre ratios 1.2 and 1.5). In contrast, stool-based tests modestly increased post-test probabilities: Cologuard 20.7% vs 7.6%, post/pre ratios 4.0; and FIT 27.6% vs 7.6%, post/pre ratios 4.4, indicating preventive potential and supporting their use as alternatives to colonoscopy. Cost analysis further identified FIT as the most cost-effective alternative, with the lowest per-case cost of $57, compared to $651 for Cologuard, $2,571 for colonoscopy, and $1,495 for Shield. Replacing colonoscopy with FIT for the 9.2 million current colonoscopy screenings in 2021 would reduce CRC screening expenditures from $27.5 billion to $4.3 billion (84.2% reduction). Comparatively, Cologuard lower costs to $9.8 billion (64.3% reduction) and Shield to $17.6 billion (36.0% reduction). Nationally, the overall cost for the five major cancers was reduced to $20.0 billion, resulting in a saving of $23.2 billion, which could be reallocated to future multi-cancer screening programs to expand population coverage. Conclusions: FIT provides the best balance of performance, preventive potential, and cost-effectiveness. By reducing colonoscopy demand and national screening costs while maintaining adequate detection, it supports sustainable population-wide screening and warrants consideration in guidelines and reimbursement policies.

Residual-guided hybrid framework for adversarially robust deep learning-based network intrusion detection

The growing sophistication of cyber threats and adversarial attacks poses critical challenges to the security and robustness of machine learning models deployed in real-world systems. While traditional deep learning architectures excel in clean data classification, they often fail under adversarial perturbations, exposing vulnerabilities in sensitive domains such as healthcare, finance, and industrial control. In this paper, we introduce a novel hybrid adversarially-trained deep learning framework that integrates reinforcement learning-inspired robustness adaptation with knowledge-driven regularization to achieve improved resilience against fast gradient sign method (FGSM) and projected gradient descent (PGD) attacks. Our approach simultaneously optimizes clean accuracy and adversarial robustness by balancing cross-entropy and adversarial loss components, while monitoring calibration error, gradient dynamics, and generalization gap to ensure stable convergence. Extensive experiments on reconnaissance, shellcode, and worms datasets demonstrate that the proposed model achieves up to 97.88% accuracy on clean data and maintains 84.9% accuracy under FGSM and 81.75% under PGD attacks, outperforming convolutional neural network (CNN) and long short-term memory (LSTM) baselines by more than 6–10 percentage points in adversarial robustness. Furthermore, the training curves reveal consistent improvements in convergence stability, runtime efficiency, gradient norm decay, and a 30% reduction in expected calibration error, validating the scalability of the framework. This work contributes not only a novel adversarial defense paradigm but also provides insights into the trade-offs between robustness, efficiency, and generalization.

Nanostructured lipid carriers for topical drug delivery: A comprehensive review of design, mechanisms, and therapeutic advances

Enhanced Hydrogen Peroxide Photosynthesis Using X‐Packed Cocrystal Catalysts

ABSTRACT Hydrogen peroxide (H 2 O 2 ) is an essential chemical and potent energy carrier. Its production through solar energy and metal‐free photocatalysts is desirable. Organic semiconductors, as a new generation of semiconductors, can form suitable transition state intermediates showing great ability in enhancing the efficiency and selectivity of photocatalytic H 2 O 2 generation, offering a metal‐free, green, and more economical solution. However, the smaller Frenkel exciton radius and larger exciton Coulomb binding energy lead to a constrained capacity for exciton dissociation, blocking the way of photocatalysts based on organic semiconductors. Here, we overcome the bottleneck by cocrystal engineering. A kind of cocrystal photocatalysts with X‐packing are designed and synthesized, which permit all excited states to be optically allowed due to reduced energy splitting, enhancing the exciton participation in photosynthesis of H 2 O 2 giving much more efficient singlet exciton dissociation and exciton utilization. Indeed, the X‐packed cocrystals show photosynthesis of H 2 O 2 from O 2 and H 2 O at a rate of 2.65 mmol h −1 g −1 and a solar‐chemical energy conversion efficiency of 0.42%, which can be further improved to 13.3 mmol h −1 g −1 with a hole sacrificial agent. This work seems to open a new door for high solar exciton utilization with organic semiconductors by cocrystal engineering.

Preparation and antibacterial activities of silver nanoparticle-loaded hydroxyapatite/collagen bone-like nanocomposite

A coherent structural picture of the interaction of Tau with tubulin provides a link to its aggregation

Age-associated differences in genomic profiles among patients with pancreatic cancer undergoing next-generation sequencing.

e16422 Background: Age is frequently used to guide treatment and clinical trial eligibility in pancreatic cancer. However, limited data exist regarding age-associated differences in genomic profiles. We aimed to evaluate differences in genomic profiles between patients aged 70 years and older, and those aged 69 years and younger. Methods: We conducted a retrospective review of patients with pancreatic cancer treated at Moffitt Cancer Center in 2024. Genomic data were obtained from next-generation sequencing assays (NGS), including FoundationOne, Moffitt STAR, and germline testing platforms such as Ambry Genetics. Detected genomic alterations were recorded, including somatic and germline mutations when available. Pharmacogenomic data for DPYD and UGT1A1 metabolizer status (poor, intermediate, or normal) were also reported. Patients were stratified age into two groups: 70 years and older and 69 years and younger. Results: A total of 284 patients with available NGS were included, including 145 patients aged 69 years and younger and 139 patients aged 70 years and older. KRAS mutations were common in both groups and did not differ significantly by age (54.5% vs 58.3%, p = 0.54). CHIP-associated mutations, including DNMT3A, TET2, and ASXL1, were more frequently observed in patients aged 70 years and older compared with younger patients, although this difference was not statistically significant (15.8% vs 11.0%, p = 0.23). Similarly, the prevalence of pathogenic hereditary cancer gene mutations did not differ significantly between age groups (12.9% vs 14.5%, p = 0.69). Among KRAS-mutant tumors, older patients demonstrated higher reported variant allele frequencies (VAF) compared with younger patients by descriptive analysis. Conclusions: In this cohort of patients with NGS, no statistically significant differences in the prevalence of key genomic alterations were observed between older and younger adults. Older patients demonstrated a higher frequency of CHIP-associated mutations and KRAS VAF by descriptive analysis, while younger patients more frequently harbored pathogenic hereditary cancer gene alterations. These findings suggest that tumor genomic profiles are largely similar across age groups, supporting continued consideration of molecularly informed treatment strategies in older adults. Age-stratified genomic alterations among patients with pancreatic cancer undergoing next-generation sequencing. Feature Age ≤69 (n=145) Age ≥70 (n=139) p value KRAS-mutant tumors, n (%) 79 (54.5) 81 (58.3) 0.54 CHIP-associated mutations‡, n (%) 16 (11.0) 22 (15.8) 0.23 Hereditary cancer gene mutations§, n (%) 21 (14.5) 18 (12.9) 0.69 ‡Clonal hematopoiesis of indeterminate potential (CHIP) defined as DNMT3A, TET2, or ASXL1 mutations. §Hereditary cancer genes include ATM, BRCA1, BRCA2, and PALB2.

SBRT versus surgery for early-stage NSCLC: A propensity-matched systematic review, meta-analysis, and meta-regression.

e20037 Background: For early-stage non-small cell lung cancer (NSCLC), surgery is preferred for operable patients, while stereotactic body radiotherapy (SBRT) is standard for inoperable cases. This meta-analysis compared survival and disease control outcomes using propensity-matched data to minimize confounding. Methods: A systematic review and meta-analysis was conducted per PRISMA guidelines, including propensity-matched studies comparing SBRT and surgery for stage I-II NSCLC through June 2025. Pooled hazard ratios (HRs) and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated for overall survival (OS), cancer-specific survival (CSS), disease-free survival (DFS), local control (LC), distant control (DC), and recurrence. Results: 34 propensity-matched studies comprising 14,224 patients were included. Surgery was associated with improved overall survival (pooled HR = 0.70, 95% CI 0.54–0.91) and cancer-specific survival (HR = 0.72, 95% CI 0.52–0.99). Fixed-time analyses reinforced these findings, showing significantly higher 1-year (OR = 1.33) and 3-year (OR = 1.89) survival with surgery, with the benefit increasing over longer follow-up (P for subgroup difference = 0.039). Surgery also demonstrated superior disease-free survival (OR = 1.99, 95% CI 1.49–2.64) and local control (HR = 0.25, 95% CI 0.09–0.64; OR = 2.57, 95% CI 1.66–3.99). In contrast, distant control did not differ significantly between treatments (OR = 1.30, 95% CI 0.99–1.72). Local recurrence was more common after SBRT (OR = 2.96, 95% CI 1.56–5.61). Meta-regression showed no significant temporal trend in outcomes. Residual demographic differences persisted, with SBRT patients being older on average (73.3 vs. 70.3 years). Conclusions: For operable early-stage NSCLC, surgical resection provides superior OS, CSS, DFS, and LC compared to SBRT, with the survival benefit becoming more pronounced over longer follow-up. SBRT remains a vital, effective option for inoperable patients, offering excellent short-term control with lower procedural risk. The findings underscore the importance of surgical staging and complete tumor removal for maximizing long-term outcomes in fit patients. Multidisciplinary, patient-centered decision-making, considering comorbidities, functional status, and diagnostic needs, is essential for optimal treatment selection. Summary of meta-aAnalysis results. Outcome Hazard Ratio (95% CI) Odds Ratio (95% CI) Favors Overall Survival 0.70 (0.54–0.91) 1.67 (1.43–1.96) Surgery Cancer-Specific Survival 0.72 (0.52–0.99) 1.99 (1.47–2.70) Surgery Disease-Free Survival - 1.99 (1.49–2.64) Surgery Local Control 0.25 (0.09–0.64) 2.57 (1.66–3.99) Surgery Distant Control 0.28 (0.11–0.73) 1.30 (0.99–1.72) Neutral Locoregional Control 0.45 (0.28–0.72) 0.64 (0.14–3.00) Surgery (HR only)

Acute kidney injury in non-variceal upper gastrointestinal bleeding among patients with gastric cancer: Incidence, predictors, mortality risks, and healthcare burden.

e16020 Background: Gastric cancer patients are vulnerable to non-variceal upper gastrointestinal bleeding, a complication that worsens clinical outcomes. Acute kidney injury (AKI) can develop in the setting of gastrointestinal bleeding due to intravascular volume loss and associated hypotension. This study explores the incidence, predictors, mortality, and overall healthcare burden of AKI in gastric cancer patients presenting with non-variceal upper gastrointestinal bleeding(NVUGIB). Methods: The 2016-2022 National Inpatient Sample (NIS) was queried to identify adult hospitalizations with NVUGIB in gastric cancer patients using ICD-10-CM codes. Patients with COVID-19 and those younger than 18 years were excluded. Multivariable logistic regression was performed to identify independent predictors of AKI. Outcomes including in-hospital mortality, need for hemodialysis, length of stay, and hospital charges were compared between AKI and non-AKI groups using appropriate statistical tests. Results: Among 5,135 hospitalized patients with NVUGIB in the setting of gastric cancer, 990 (19.28%) developed AKI. Patients with AKI were significantly older than those without AKI (mean age 71.92 vs 66.46 years, p < 0.001). On multivariable logistic regression analysis, advancing age (AOR 1.029 per year, 95% CI 1.010-1.047, p = 0.002), pre-existing chronic kidney disease(CKD) (AOR 2.572, 95% CI 1.684-3.926, p < 0.001), chronic obstructive pulmonary disease(COPD) (AOR 1.917, 95% CI 1.167-3.150, p = 0.01), and cirrhosis (AOR 1.732, 95% CI 1.015-2.956, p = 0.044) were significant predictors of AKI development. Other demographic and clinical variables including sex, race, insurance status, weekend admission, and comorbidities such as diabetes, hypertension, congestive heart failure, and metastatic cancer were not significantly associated with AKI risk. Patients with AKI had more than three-fold higher odds of in-hospital mortality compared to those without AKI (AOR 3.277, 95% CI 1.728-6.216, p < 0.001). Among patients who developed AKI, 2.53% required hemodialysis during hospitalization. Mean length of stay was significantly longer in the AKI group (7.05 vs 4.64 days, p < 0.001), and mean hospital charges were nearly double ($116,225 vs $63,164, p < 0.001). Conclusions: AKI was present in nearly one-fifth of gastric cancer patients admitted for NVUGIB and was associated with substantial mortality and healthcare utilization. Risk stratification measures in patients with pre-existing chronic kidney disease, COPD, cirrhosis, and advancing age may help mitigate such events and improve outcomes.

Timing of postoperative chemoradiotherapy and survival outcomes in high-risk locally advanced head and neck squamous cell carcinoma: A supplementary analysis of JCOG1008.

6015 Background: For patients with post-operative high-risk locally advanced head and neck squamous cell carcinoma (LA-HNSCC), initiation of chemoradiotherapy (CRT) within 6 weeks after surgical resection is recommended based on previous reports. However, the guideline-defined 6-week threshold for initiating post-operative radiotherapy (PORT) is derived mainly from data in patients treated with PORT alone. Therefore, its relevance in patients receiving post-operative CRT remains uncertain. Methods: This supplementary analysis used data from the JCOG1008 trial (jRCTs031180135), which compared post-operative CRT with weekly cisplatin versus 3-weekly cisplatin. Associations between the surgical-to-PORT interval (S-PORT) and locoregional relapse-free survival (LRFS), overall survival (OS), and relapse-free survival (RFS) were evaluated using restricted cubic spline analyses, Kaplan–Meier methods, and Cox proportional hazards models. Treatment package time (TPT), defined as the sum of S-PORT and radiotherapy duration, was also assessed. Results: Among 261 enrolled patients, 251 were eligible for this analysis (median age, 62 years; 16.3% female; 45.8% with oral cavity primaries). The median S-PORT and TPT were 7.0 weeks (range, 2.9–10.0) and 13.9 weeks (range, 7.6–16.9), respectively. Restricted cubic spline analyses demonstrated minimal variation in risk across the observed S-PORT range, with hazard ratios (HRs) remaining close to 1 for LRFS, OS, and RFS. Kaplan–Meier analyses using the guideline-defined 6-week cutoff showed similar 5-year LRFS between patients with S-PORT ≤6 weeks and >6 weeks (64.0% vs 64.9%; HR 0.91, 95% CI 0.60–1.37; Table), with no significant differences in OS or RFS. In multivariable Cox models adjusting for clinical covariates, S-PORT was not independently associated with outcomes. Findings for TPT closely paralleled those for S-PORT, with no significant associations observed between TPT and treatment. Conclusions: In this supplementary analysis of JCOG1008, no prognostic disadvantage was observed when PORT was initiated within the range observed in this trial (6–10 weeks after surgery) among patients receiving post-operative CRT. These findings support individualized S-PORT timing rather than rigid cutoff-based decision-making in contemporary standardized CRT. Clinical trial information: jRCTs031180135. Kaplan–Meier analyses according to guideline-defined 6-week surgical-to-postoperative radiotherapy interval (S-PORT) cutoff. S-PORT ≤ 6 weeks S-PORT > 6 weeks 5y LRFS (95% CI) 64.0% (52.1%–73.7%) 64.9% (57.3%–71.5%) HR (95% CI), p-value 1 0.91 (0.60–1.37), 0.65 5y OS (95% CI) 65.3% (53.4%–74.9%) 66.6% (59.1%–73.1%) HR (95% CI), p-value 1 0.88 (0.58–1.34), 0.54 5y RFS (95% CI) 60.0% (48.0%–70.1%) 60.2% (52.6%–67.0%) HR (95% CI), p-value 1 0.98 (0.66–1.46), 0.93