Browse Articles
Discover research articles across all indexed journals
Improving cancer education and safety in rural communities through a digital literacy platform.
e23302 Background: Cancer care is increasingly complex, requiring patients to understand diagnoses, treatment options, side effects, and adherence requirements during periods of emotional distress. Limited health literacy, time constraints during clinical encounters, and variability in educational materials contribute to gaps in patient understanding, particularly in rural populations. We developed a clinician-led digital education platform delivering standardized, plain-language cancer education videos integrated into real-world oncology practice. Methods: We evaluated clinician-developed video education in a rural oncology practice. Videos were delivered after the initial physician encounter and before formal chemotherapy education sessions conducted by nurses. Content followed NIH plain-language guidelines and addressed three core questions: what the treatment is, why it is necessary, and how it is administered and managed. Videos were available via Oncology101.org, viewed in the clinic and at home. Patients were grouped by pre-education exposure: video (n = 10), pamphlet (n = 22), or no materials (n = 24). Preparedness and anxiety were assessed using a 10-point Likert scale, and education session duration was recorded. Group differences were analyzed using the Kruskal–Wallis test. Results: Fifty-six patients were analyzed: 10 (18%) reviewed video education, 22 (39%) reviewed pamphlets, and 24 (43%) received no prior materials. Preparedness differed significantly across groups (median [IQR]: 10 [8–10] vs 9 [5–10] vs 9 [5–10]; p = 0.046). Mean ± SD preparedness scores were 8.44 ± 2.92 (video), 7.64 ± 2.77 (pamphlet), and 7.21 ± 3.27 (none). Preparedness was lower by 10.5% in the pamphlet group and 16.7% in the no-resource group compared with the video group. Anxiety scores were similar across groups (5.44 ± 3.09 vs 3.59 ± 3.28 vs 4.17 ± 3.36; p > 0.05), and education session duration did not differ significantly (56.1 ± 24.7 vs 54.9 ± 12.5 vs 53.5 ± 19.5 minutes; p > 0.05). Nurses reported improved patient preparedness, satisfaction, and consistency of education among patients who accessed video content. Conclusions: Clinician-led digital education is feasible and scalable in a rural oncology setting and is associated with improved patient preparedness without increased anxiety. Oncology101.org provides standardized, accessible, physician-designed education that complements clinical encounters. These findings support broader implementation and prospective evaluation of digital clinician-led education as an integral component of contemporary oncology care. The table summarizes the duration of formal education sessions, perceived preparedness, and perceived anxiety across the three groups. Outcome Video Pamphlet None p-value Preparedness 8.44 ± 2.92 7.64 ± 2.77 7.21 ± 3.27 0.046 Anxiety 5.44 ± 3.09 3.59 ± 3.28 4.17 ± 3.36 NS Time (min) 56.1 ± 24.7 54.9 ± 12.5 53.5 ± 19.5 NS
A prospective study of the early treatment of recombinant human thrombopoietin for chemoradiation-induced thrombocytopenia in cervical squamous cell carcinoma patients.
12125 Background: To explore the early intervention timing of recombinant human thrombopoietin (rhTPO) for cancer treatment-induced thrombocytopenia (CTIT) in cervical squamous cell carcinoma (CSCC) patients during concurrent chemoradiotherapy. Methods: Patients with stage I-IVa CSCC who developed CTIT during radical concurrent chemoradiotherapy were prospectively enrolled in our study. rhTPO intervention was used at the first presentation of G1 or G2 CTIT during concurrent chemotherapy respectively. The key indexes, including nadir platelet count, platelet recovery time, platelet transfusion and incidence of G3-5 CTIT were recorded in two groups. χ² test was used for the effective analysis, and univariate and multivariate analyses logistic regression analysis was used to predict the potential factors for the G3-5 CTIT. Results: From February 2021 to June 2024, 204 CSCC patients who developed CTIT during radical chemoradiotherapy at Sichuan Cancer Hospital were prospectively enrolled. 75 patients who occurred G1 and G2 CTIT at the first presentation during concurrent chemoradiotherapy received rhTPO intervention. They were included in the intent-to-treat (ITT) population analysis. According to the timing of the rhTPO intervention, the patients were divided into G1 group (100×10 9 /L>PLT>75×10 9 /L) and G2 group (75×10 9 /L>PLT>50×10 9 /L). For effective analysis, the nadir platelet count in G1 group was greater than that in G2 group (67 × 10 9 /L vs. 54 × 10 9 /L, p = 0.000), and the platelet recovery time in G1 group was shorter than that in G2 group (8 days vs. 14 days, p = 0.003). Moreover, patients in G1 group had a significantly lower incidence of G3-5 CTIT than that in G2 group (14.3% vs. 35%, p = 0.04). In two groups, only 1 patient in G2 group received 4 apheresis platelet units transfusion. Univariate and multivariate analysis showed the early intervention of the rhTPO was a substantial factor for decreasing the incidence of G3-5 CTIT. The sensitivity was 0.67 and the specificity was 0.74. Conclusions: For CSCC patients received with radical concurrent chemoradiotherapy, the early intervention of rhTPO was a substantial factor for decreasing the incidence of G3-5 CTIT. It could significantly improve the nadir platelet count, shorten the platelet recovery time, and reduce the incidence of G3-5 CTIT, thereby ensuring the uninterrupted continuation of treatment. Platelet recovery time and the degree of CTIT. Data Baseline platelet count(×10 9 /L) Days with platelet count recover ≥100×10 9 (Days) Minimal mean platelet count（×10 9 /L） Incidence of G3-5 CTIT(%) G1 group 172±64.92 8±5.86 67±15.81 14.3 G2 group 166±78.43 14±9.59 54±14.37 35 P value 0.76 0.003* 0.000* 0.04* *Statistically significant.
Impact of targeted axillary dissection on axillary lymph node dissection rates compared with sentinel lymph node biopsy in cN1 breast cancer patients after neoadjuvant chemotherapy: A single-centre study.
e12675 Background: In clinically node positive (cN1) breast cancer patients post neoadjuvant chemotherapy (NACT), Axillary de-escalation strategies aim to minimise morbidity without compromising oncological safety. Sentinel lymph node biopsy (SLNB) and targeted axillary dissection (TAD) are commonly employed techniques; however, comparative data evaluating technical failure and the need for axillary lymph node dissection (ALND) remain limited. We compared SLNB and TAD in terms of technical performance and rates of ALND in patients post NACT. Methods: This was a prospective two-centre study including biopsy proven node positive breast cancer (cN1) patients post NACT from January 2023 - November 2025 post NACT who were then converted to ycN0 were included. axillary surgical management differed by participating centre: where one centre patients underwent SLNB , while those treated at another centre underwent TAD. The primary endpoint was the rate of completion ALND in view of technique failure. Technique failure was defined as failed mapping, retrieval of < 3 nodes, failure to retrieve the clipped node. Secondary end points included residual nodal positivity and QOL between the two cohorts. Categorical variables were compared using chi-square or Fisher's exact tests, with statistical significance defined as p < 0.05. Results: Baseline clinicopathological characteristics were comparable between the two groups. A total of 108 patients were included, with 54 in each group. Failed mapping was was uncommon in both cohorts (0% SLNB vs 1.9% TA; p = 0.50). Retrieval of < 3 nodes occurred significantly more in the SLNB group compared with the TAD group (37% vs 11.1%; p = 0.0029). Failure to retrieve clipped node was rare and similar between groups (0% vs 1.9%; p = 0.50). The overall rate of required ALND was significantly higher in the SLNB cohort compared with the TAD cohort (37.0% vs 14.8%; p = 0.0087). Conclusions: In cN1 breast cancer patients after NACT, TAD was significantly associated with lower rates of required ALND due to technical failure compared with SLNB, primarily due to improved nodal retrieval. These findings support TAD as a more reliable strategy that may reduce surgical over treatment without compromising technical adequacy, thus also improving quality of life with lower rates of lymphoedema and shoulder morbidity. Technical failure rates of SLNB vs TAD. SLNB (N=54) TAD (N=54) FAILED MAPPING YES 0 1 (1.85%) <3 NODES 20 (37.04%) 6 (11.11%) CN NOT RETRIEVED YES 0 1 (1.85%) TOTAL RATE OF ALND REQUIRED 20 (37.04%) 8 (14.81%)
Clinical impact and temporal evolution of mutational signatures in metastatic lobular breast cancer.
1141 Background: Invasive lobular carcinoma (ILC) has distinct genomic features, including frequent CDH1 alterations, yet the mutational processes shaping metastatic ILC and their prognostic significance remain poorly defined. We evaluated mutational signatures inferred from targeted sequencing in a large single-center cohort of metastatic ILC and their association with outcomes. Methods: Patients were included if they had ILC histology on early-stage breast biopsy/surgical pathology, and/or CDH1 mutation on metastatic site biopsy and MSK-IMPACT sequencing from a metastatic site. For survival analyses, only biopsies performed within 2 months of metastatic diagnosis were included. Clinicopathologic data were abstracted from electronic medical records. Single base substitution (SBS) signatures were inferred using SigMA; cases with ≤5 SBS were excluded. Overall survival (OS) was estimated using Kaplan–Meier and compared by dominant signature using the log-rank test. Results: 654 patients were included. 467 (71%) had hormone receptor-positive (HR+)/HER2-, 56 (8.6%) had HER2+, and 65 (9.9%) had triple negative disease at metastatic diagnosis. Among 307 (47%) with ILC variant data available, 139 (45%) had classic type, 65 (21%) had pleomorphic, 45 (15%) had mixed, and 58 (19%) had other ILC variants. 385 patients (59%) had MSK-IMPACT testing on a tissue biopsy performed within 2 months of metastatic diagnosis. Among them, the most prevalent mutations included CDH1 (85%), PIK3CA (47%), TP53 (21%), TBX3 (16%), PTEN/AKT1 (14.5%), FOXA1 (12%), ERBB2 (12%), KMT2C (9.9%), CBFB (9.9%), ARID1A (9.6%), MAP3K1 (8.8%), and NF1 (7.5%). 320/385 (83%) MSK-IMPACT tests had ≥5 SBS and were analyzable for mutational signature. The dominant mutational signature was APOBEC in 44%, clock-like (aging) in 38%, Signature 3 (homologous recombination deficiency) in 12%, Signature 17 in 5.6%, and Signature 8 in 1%. Median OS in the total cohort was 4.4 years (95%CI 4.1-4.8) and differed significantly by dominant mutational signature (log-rank p =0.001). Median OS for those with APOBEC-dominant disease was 3.1 years (95%CI 2.5-3.7) and for those with clock-like (aging) signature was 4.4 years (95%CI 3.8-6.5). Among 61 patients who had paired samples analyzable for mutational signature within 2 months of metastatic diagnosis and at another timepoint following metastatic diagnosis, the prevalence of APOBEC as the dominant mutational signature increased (29/61 [48%] to 36/61 [59%]), whereas clock-like (aging) decreased (25/61 [41%] to 17/61 [28%]). Conclusions: Dominant APOBEC mutational signature at metastatic ILC diagnosis was associated with shorter OS, with differences observed between APOBEC- and aging-dominant tumors. Serial testing demonstrating increasing APOBEC dominance over time supports mutational signatures as dynamic, clinically relevant features of metastatic ILC.
Optimizing survivorship transitions utilizing an EHR visual alert.
e13508 Background: There is a large and growing population of cancer survivors at risk for chronic health conditions and secondary malignancies. Despite advances in electronic health record (EHR) technology, many systems lack automated tools to identify patients eligible for transition to survivorship care. In the absence of a streamlined process, primary oncology clinics often become overwhelmed, resulting in care delays for established patients and reduced access for new patients. This project aimed to develop an integrated, non-disruptive EHR-based tool to accurately flag patients eligible for transition to survivorship care. Methods: A multidisciplinary team—comprising survivorship providers, EHR analysts, and key stakeholders from primary oncology departments—collaborated to define criteria for survivorship transition, based on MD Anderson Cancer Survivorship Algorithms. Inclusion criteria considered cancer type, time since treatment completion, and treatment department. Exclusion criteria were designed to omit patients with metastatic disease, those undergoing active treatment, or those enrolled in clinical trials. A series of virtual meetings followed multiple variations of the eligibility registry and team-based chart reviews to validate the tool by assessing the registry’s positive predictive value (PPV) and a false negative rate (FNR). After each round of chart reviews, debrief meetings informed iterative refinements to the EHR report. A non-intrusive visual alert was also developed to appear on provider clinic templates, minimizing workflow disruption. Results: To date, 15 of 24 survivorship eligibility registries are completed and in production. The PPVs ranged between 50% to 100%, while the FNRs were relatively low, ranging from 0-15%. The EHR alert appears as a visual indicator on the clinic template, requiring no additional clicks and avoiding disruptive workflow interruptions. In Fiscal Year 2025, survivorship clinic referrals increased by 33%. Conclusions: This project demonstrates the feasibility and scalability of EHR-based identification of survivorship-eligible patients. The integration of a visual alert system has increased survivorship referrals and reduced pressure on primary oncology clinics. Expansion of this system is ongoing across the institution. Completed eligibility registries. Thyroid CancerPPV – 80% FNR – 6% SarcomaPPV – 85% FNR – 8% Thoracic SurgeryPPV – 83% FNR – 0% Head and Neck SurgeryPPV – 67% FNR – 15% GI-GastricPPV – 85% FNR – 0% Melanoma MedOncPPV – 99% FNR – 4% Lymphoma-HodgkinPPV – 67% FNR – 6% GI-PancreasPPV – 64% FNR – 0% BMO PPV – 87% FNR – 4% NeurosurgeryPPV – 72% FNR – 0% GYNPPV – 76% FNR – 3% GU/Urology-ProstatePPV – 96% FNR – 5% Melanoma SurgOncPPV – 100% FNR – 0% GI-ColonPPV – 83% FNR – 9% GU/Urology-KidneyPPV – 54% FNR – 8%
The real-world outcomes from late-line chemotherapy for advanced colorectal cancer: A population-based analysis of attrition rates, outcomes, and health care contact in a single-payer health system.
e23443 Background: Few population-level real-world studies have evaluated the uptake, health care contact time, and efficacy of later-line treatments for metastatic colorectal cancer (mCRC). We evaluated treatment patterns, attrition rates, survival outcomes, and health care contact time across multiple lines of therapy for patients with mCRC in Ontario, Canada. Methods: We performed a population-based retrospective cohort study of adults with mCRC in Ontario (population 16 million; single-payer public health system) diagnosed between Jan 2010 and December 2019 and receiving at least one line of palliative chemotherapy. Patients were linked across multiple administrative datasets using a unique patient identifier to determine demographics, chemotherapy prescription, outcomes and health care contact time. Health care contact time was calculated as the percentage of contact days per month, accounting for both inpatient and outpatient contact. Overall survival was estimated using Kaplan-Meier methods. Cox proportional hazards regression was used to examine factors associated with survival, and logistic regression for factors association with the receipt of more than one line of therapy. Results: We included 4,541 patients in our study. Excluding patients undergoing metastectomy within 12 months of first chemo (n = 167), median overall survival was 16 months (95% CI 15.6 – 17.04). Approximately 45% (n = 1988) received only 1 line of therapy, 28% (n = 1,221) received only 2 lines of therapy, 19% (n = 819) received only 3 lines of therapy, and 8% (n = 346) received 4 or more lines of therapy. Younger age, fewer comorbidities and higher socioeconomic status were associated with receiving more than one line of therapy, while geographic location and sex showed no clear association. The average proportion of monthly contact time was similar across lines of therapy (11.8% during 1st, 12.6% during 2 nd line, 12.8% during 3 rd line, 12.4% during 4 th line and beyond). Monthly contact days were also similar among the commonly used intravenous first line regimens but were slightly lower among those receiving capecitabine. Most contacts days were outpatient contact days. Conclusions: Median overall survival in this real-world cohort of patients receiving at least one line of palliative chemotherapy is shorter than typically seen in first line clinical trials. Just over half of patients who receive first line chemotherapy for mCRC in Ontario receive a 2 nd line of treatment, illustrating high attrition rates in real world practice. Patients with mCRC treated within the Ontario cancer system should expect to spend approximately 10-13% of their time each month in contact with the health care system, regardless of line of therapy or type of systemic treatment.
Mortality patterns and disparities in primary malignant brain neoplasms among U.S. adults aged ≥75 years: A 25-year retrospective analysis.
e14064 Background: Primary malignant neoplasms of the brain disproportionately affect older adults, yet population-level mortality patterns among individuals aged ≥75 years remain insufficiently characterized. This study aims to evaluate temporal trends and demographic disparities in mortality due to primary malignant brain neoplasms among U.S. adults aged ≥75 years. Methods: Mortality data were obtained from the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research (CDC WONDER) Multiple Cause of Death database for adults aged ≥75 years. Primary malignant brain neoplasms were identified using ICD-10 code (C71), and age-adjusted mortality rates (AAMR) were calculated per 100,000 population. Temporal trends were evaluated using Joinpoint regression to estimate average annual percent change (AAPC) with 95% confidence intervals. Results: Between 1999 and 2023, brain neoplasms among adults ≥75 years led to 101,761 deaths, with the most deaths being recorded at the decedent's home (49.28%). The AAMR rose from 19.9 in 1999 to 22.3 in 2023 with an associated AAPC of 0.59 (p < 0.000001). In 1999, men had an AAMR of 24.9, while women had an AAMR of 16.88. In 2023 men had an AAMR of 27.20, while women had an AAMR of 18.65 (AAPC: 0.53; p < 0.000001 and AAPC: 0.46; p = 0.000038 for men and women respectively). Non-Hispanic White individuals had an AAMR of 24.19 (AAPC: 0.74; p < 0.000001) in 2023, which was almost twice as high as the AAMR of Non-Hispanic Black individuals, who experienced an AAMR of 12.12 in 2023 (AAPC: 0.52; p < 0.027657). This was followed by Hispanic or Latino individuals with an AAMR of 11.57 in 2023 (AAPC:1.26; p = 0.000483). States with the highest AAMRs included South Dakota (27.38). Arkansas (24.97) and Nebraska (24.89); states with the lowest AAMRs included Hawaii (10.65), District of Columbia (12.62) and New York (18.04). Geographically, the South had the highest AAMR of 23.08 in 2023 (AAPC: 0.48; p=0.000046), while the Northeast had the lowest AAMR of 19.99 in 2023 (AAPC: 0.81; p=0.000049). Conclusions: Mortality due to primary malignant brain neoplasms among U.S. adults aged ≥75 years has shown a sustained and statistically significant increase from 1999 to 2023, underscoring a growing public health burden in this rapidly expanding age group. Persistent disparities by sex, race/ethnicity, geography, and state highlight inequities in risk, access to care, and outcomes. Targeted strategies focusing on early recognition, equitable access to specialized neuro-oncologic care, and region-specific interventions are essential to address the rising mortality and demographic disparities. Metric Findings Deaths 101,761; 49% home AAMR 19.9→22.3; AAPC 0.59 Disparities Men, Non-Hispanic White highest Region South highest; Northeast lowest States South Dakota, Arkansas, Nebraska had the highest burden
Long-term trends in pheochromocytoma in the United States, 1975–2022: A SEER-based study.
e22593 Background: Pheochromocytoma is a rare neuroendocrine tumor, and population-level data describing long-term temporal trends are limited. Evaluating national patterns over time may improve epidemiologic understanding and clinical awareness. Methods: We performed a retrospective analysis of pheochromocytoma cases recorded in the Surveillance, Epidemiology, and End Results (SEER) database from 1975 through 2022. Annual case counts were examined overall and by race (White, Black, Other, and Unknown). Temporal trends were assessed using Poisson regression, and annual percent change (APC) was calculated to quantify changes over time. Results: A total of 408 pheochromocytoma cases were identified during the study period. Overall case counts increased significantly over time, with Poisson regression demonstrating a consistent upward trend corresponding to an APC of 5.8% per year (p < 0.001). White individuals accounted for the majority of reported cases, while Black and other racial groups had significantly lower case counts compared with the reference population. Visual assessment of decade-level trends suggested a gradual increase from 1975 through approximately 2010, followed by a sharper rise in more recent years. Conclusions: Pheochromocytoma case counts in the United States have increased substantially over the past five decades based on SEER data. These findings suggest evolving epidemiologic patterns in this rare tumor and emphasize the importance of continued surveillance. Observed trends should be interpreted in the context of potential changes in diagnostic practices, imaging utilization, and cancer registry reporting over time.
RT01: A first-in-class theranostic radionuclide drug conjugate (RDC) with zirconium-89 and lutetium-177 targeting tissue factor.
e15151 Background: Tissue factor (TF) is highly expressed in various solid tumors, including pancreatic, cervical, and lung cancers, and is a key driver of tumor invasion and metastasis. While TF-targeted antibody-drug conjugates (ADCs) have validated this target, their efficacy can be limited by tumor heterogeneity and low antigen expression. The Nanolattix research team developed RT01-Zr89 and RT01-177Lu as a first-in-class oncology theranostic pair targeting TF. This approach is designed to overcome ADC limitations through synergistic therapy and potent killing of tumor cells even in heterogeneous populations. Methods: We evaluated the cytotoxicity, in vivo tumor-suppressive activity, and safety of RT01 using in vitro tumor cell lines (including low TF-expressing models) and tumor-bearing mouse models. Both PET/CT and SPECT/CT imaging analysis were utilized to monitor real-time drug distribution and "dynamic spatial distribution" post-injection, validating the integrated diagnostic and therapeutic functions. Results: In vitro experiments demonstrated that RT01 significantly killed both high and low TF-expressing tumor cells. In tumor-bearing mouse models, a single dose of RT01-177Lu (200 uCi) achieved a maximum mean tumor uptake of 35.29% ID/cc at 96 hours. Tumor shrinkage began on day 6, with a therapeutic effect lasting over 50 days and a tumor inhibition rate of 75.3%. Imaging analysis confirmed specific accumulation in xenograft tumors. RT01-Zr89 studies suggest its effectiveness as a companion imaging diagnosis. Anatomical and pathological examinations revealed no damage to major organs, and no grade 3 or higher adverse reactions were observed. Conclusions: RT01-Zr89 and RT01-177Lu demonstrate outstanding efficacy and safety in preclinical studies as a novel theranostic strategy to overcome resistance in solid tumors. Based on these results, an investigator-initiated clinical trial (IIT) has been initiated.
Outcomes after single-fraction preoperative stereotactic radiosurgery for brain metastases: A systematic review and meta-analysis.
e14016 Background: Postoperative stereotactic radiosurgery (SRS) is utilized following resection of brain metastases (BM), however use is limited by target delineation uncertainty and an increased risk of leptomeningeal disease (LMD). Preoperative SRS allows for precise targeting and is being increasingly utilized in treatment paradigms of resectable BM. Reported outcomes across studies are promising and there are several phase 3 trials which are underway. Therefore, we pooled multiple studies and conducted a meta-analysis to estimate 1-year local control (LC), LMD incidence, and overall survival (OS) following single-fraction preoperative SRS for resectable BM. Methods: We systematically searched PubMed, Embase, and Scopus to identify studies of single-fraction preoperative SRS for resectable BM. For each cohort, data for 1-year LC, 1-year LMD, and 1-year OS, along with corresponding sample sizes, were extracted. A proportion-based meta-analysis was performed to derive pooled estimates and 95% confidence intervals (CIs), with between-study heterogeneity quantified using I². Results: Ten studies published between 2016 and 2025, including both prospective trials and retrospective cohorts, met the inclusion criteria. There were 617 patients treated with single-fraction preoperative SRS. The median age was 60.8 years (range 58-65 years), and the median prescribed dose was 15.8 Gy (range 14.5-17.0 Gy). The pooled 1-year LC rate was 81.9% (95% CI, 76.9-85.9; I² = 43.4%). The pooled 1-year incidence of LMD was 4.6% (95% CI, 2.8-7.4; I² = 0%), and the pooled 1-year OS rate was 60.2% (95% CI, 55.5-64.8; I² = 0%). Sensitivity analyses showed stable pooled estimates, with no single study driving the results. Conclusions: Single-fraction preoperative SRS is associated with low rates of LMD and acceptable LC and OS. In the context of historically reported postoperative SRS outcomes, these results suggest that preoperative SRS represents a promising strategy in selected patients. Randomized trials are needed to define the comparative role and optimal patient selection for preoperative SRS, with ongoing studies (NCT05341739, NCT03741673) expected to clarify its role in the management of resectable BM. Study characteristics and reported outcomes of single-fraction preoperative SRS for BM. Study Study design Number of patients Median age (years) Median dose (Gy) 1-year LTC (%) 1-year LMD (%) 1 year OS (%) Agarwal, 2024 Trial 47 59 NR NR 4.8 NR Higazy, 2025 Trial 35 61 17 82 2.86 NR Li, 2022 Retrospective 24 64 NR 87.6 NR NR Mallela, 2024 Retrospective 26 64.5 NR 77.2 NR NR Murphy, 2024 Trial 35 63 NR 76.6 NR 59 Patel, 2018 Retrospective 12 57.5 NR 49.1 10 74.1 Patel, 2016 Retrospective 66 58.2 14.5 84.1 3.2 NR Prabhu, 2025 Retrospective 124 NR 17 87.2 1.7 64.5 Prabhu, 2025 Retrospective 183 NR 15 86.4 7.3 58.1 Prabhu, 2020 Retrospective 65 NR NR 77.1 4.2 56.3
Impact of neoadjuvant immunotherapy on safety of axillary de-escalation in triple-negative breast cancer (TNBC): A propensity score–matched analysis.
e12674 Background: While the addition of immune checkpoint inhibitors (ICIs) to neoadjuvant chemotherapy (NAC) has improved systemic control and pathologic complete response in TNBC, its effect on de-escalation of axillary surgery and associated overall survival remains unclear. We investigated whether the addition of ICIs to NAC facilitates safe de-escalation of axillary lymph node dissection (ALND) to sentinel lymph node biopsy (SLNB) in patients presenting with clinically node-positive (cLN+) disease. Methods: Women > 18y cN1-3, cM0 TNBC diagnosed 2018-2022 were identified from the National Cancer Database. Patients were stratified by systemic therapy (NAC vs NAC+ICIs) and surgical procedure (SLNB vs ALND). Propensity score matching (PSM) and restricted mean survival time (RMST, truncated at 48 months) were utilized to assess overall survival (OS). Results: 4336 patients were included for analysis (cN1 = 3658 [84.4%], cN2 = 321 [7.4%], cN3 = 357 [8.2%]). NAC+ICI was associated with improved OS in the SLNB group (RMST 46.6 vs 46.1 mos, p = 0.03) but not the ALND group (RMST 46.4 vs 46.3, p = 0.91). In the NAC-alone cohort, SLNB was associated with significantly inferior survival vs ALND (5-year OS ALND 87.8% vs SLNB 83.0%, p = 0.027) and was independently associated with increased risk of mortality (HR 1.63, 95% CI 1.12–2.38, p = 0.01). Notably, in the NAC+ICI cohort, this survival difference did not persist, with SLNB and ALND having comparable survival (5-year OS ALND 93.6% vs SLNB 90.1%, p = 0.99) and identical 48-month RMSTs (46.4 months). Multivariable analysis confirmed that SLNB was not associated with worse survival in the immunotherapy setting (HR 1.19, 95% CI 0.54–2.61, p = 0.67). Conclusions: Among SLNB recipients, ICIs were associated with improved unadjusted OS, but this benefit was not observed among ALND recipients, suggesting the burden of axillary disease necessitating ALND aligns with overall worse prognosis and was less affected by ICI receipt. Patients receiving SLNB after NAC-alone had worse survival vs those receiving ALND, but it is unclear whether this survival difference is due to omission of axillary clearance or patient-level factors associated with the decision to omit ICIs in the first place (e.g., co-morbidities that might also have led to omission of ALND). Notable, however, among those receiving both NAC and ICIs, survival was comparable between SLNB and ALND recipients, suggesting ICIs safely facilitate axillary de-escalation and effectively control tumor burden without the need for axillary clearance beyond SLNB.
Antiviral prophylaxis to prevent hepatitis B reactivation in patients receiving anti–CD20-based chemotherapy and with inactive HBV infection: Systematic review and meta-analysis of randomized controlled trials.
e15034 Background: Patients with history of inactive hepatitis B receiving anti-CD20-based chemotherapy for lymphoma remain at risk for HBV reactivation and reverse seroconversion, which can cause hepatitis and disrupt cancer treatment. We aim to provide randomized evidence on whether routine antiviral prophylaxis is needed in those patients. Methods: We searched PubMed, Embase, Cochrane Library and ClinicalTrials.gov for randomized controlled trials enrolling adults with inactive HBV receiving anti-CD20-based chemotherapy (predominantly rituximab-based; one trial included obinutuzumab) and comparing high-barrier antiviral prophylaxis (entecavir or tenofovir formulations) versus placebo/observation. Inactive HBV was defined as HBsAg-negative and anti-HBc-positive; anti-HBs could be positive or negative. Baseline HBV DNA requirements varied across included trials. Some of them required HBV DNA negativity at enrollment and others permitted low-level detectable HBV DNA. Outcomes were pooled as risk ratios (RR) using a DerSimonian-Laird random-effects model. Primary outcome: HBsAg reverse seroconversion. Secondary outcomes: HBV reactivation, HBV-associated hepatitis, and all-cause mortality. Results: Four RCTs (N = 373) met inclusion. HBsAg reverse seroconversion was not significantly different between the two groups RR 0.42 (95% CI 0.07-2.43; I² = 25.5%). Antiviral prophylaxis reduced the risk of HBV reactivation RR 0.13 (95% CI 0.03-0.57; I² = 0%). There was no difference in HBV-associated hepatitis RR 0.44 (95% CI 0.06-3.09; I² = 0%). All-cause mortality did not differ RR 0.63 (95% CI 0.34-1.16; I² = 0%). Conclusions: High-barrier antiviral prophylaxis in patients receiving anti-CD20-based chemotherapy and have inactive HBV infection reduces HBV reactivation significantly with no significant reduction in reverse seroconversion, HBV-associated hepatitis or all-cause mortality. Trial HBsAg reverse seroconversion HBV reactivation HBV-associated hepatitis All-cause mortality Huang et al, 2013 0/41 vs 4/39 1/41 vs 7/39 0/41 vs 1/39 Not Reported Liu et al, 2018 0/95 vs 2/95 0/95 vs 3/95 0/95 vs 1/95 10/95 vs 16/95 Buti et al, 2017 0/33 vs 1/28 0/33 vs 3/28 0/33 vs 1/28 4/33 vs 5/28 Vakili et al, 2025 2/20 vs 0/22 0/20 vs 0/22 0/20 vs 0/22 0/20 vs 1/22 Total 2/189 vs 7/184 1/189 vs 13/184 0/189 vs 3/184 14/148vs 22/145
Interim analysis of CaboMain: A prospective, single-arm phase 2 clinical trial of cabozantinib as maintenance therapy for patients with “ultra-high-risk” pediatric solid tumors.
10014 Background: There are patients with pediatric solid tumors, including those with relapsed/refractory disease, with “ultra-high” risk (UHR) disease (1-year progression free survival (1-yr PFS) <25%) even when achieving disease control. Most pediatric solid tumors do not have a therapeutically-targetable oncogenic driver, limiting treatment options for maintenance therapy. Cabozantinib is a multitargeted tyrosine kinase inhibitor with demonstrated activity against many solid tumors, with a defined safety profile and dosing in pediatric and young adult patients. We hypothesize that use of cabozantinib for 1 year, after achieving disease control, can improve 1-yr PFS to ≥ 45%. Methods: We are performing a multisite, open-label, single-arm Phase 2 clinical trial for patients with pediatric solid tumors with 4 strata; Stratum A, primary stratum for analysis, enrolls patients with UHR disease. Patients must have completed prior therapy at least 4 weeks and no longer than 12 weeks before enrollment with stable disease or better at time of enrollment. Enrolled patients receive cabozantinib as per dosing nomogram at 40 mg/m 2 /day, taken for up to 365 days from start of therapy. Primary outcome is 1-yr PFS, with Bayesian time-to-event optimal phase 2 design for analysis. Interim analysis follows the rules: assuming a Beta (0.25,0.75) prior distribution for efficacy and 88% statistical power, at interim at least 5/18 evaluable patients must have achieved 1-yr PFS to justify ongoing enrollment. Interim analysis reporting has been recommended by the Data Safety and Monitoring Committee. Results: We report the interim results as of 11/1/2025, which includes all enrolled patients on Stratum A who have reached 1 year after start of therapy (n=13). Eight of thirteen patients reached 1-yr PFS; by diagnosis, neuroblastoma (NBL) = 2, Ewing sarcoma (EWS)= 3, osteosarcoma (OST)= 3. Of those not reaching the one-year time point, 2 patients progressed after voluntarily discontinuing cabozantinib (1 NBL, 1 month after discontinuation; 1 OST, 5 months after discontinuation), 1 NBL elected radiation therapy and was removed from study, and 2 OST progressed on study treatment. Having reached the interim analysis benchmark, enrollment continues. Three additional strata (metastatic EWS, metastatic OST, and fusion-positive metastatic rhabdomyosarcoma, each at end of 1 st line therapy) also continue to enroll, as secondary endpoints. Longer-term PFS and overall survival, patient reported outcomes, and exploratory biological measures continue to be tabulated for later analysis. Conclusions: Use of cabozantinib as maintenance therapy, after prior treatment and achievement of stable disease or better response, has been determined to be feasible to demonstrate improvement of 1-yr PFS by at least 20%, and ongoing enrollment is justified. Clinical trial information: NCT05135975 .
Risk of multiple primary malignancies in mantle cell lymphoma: Insights from a SEER-based study (2000–2021).
e19051 Background: With increasing tumor survival rates and advancements in diagnostics, the incidence of multiple primary malignancies (MPMs), particularly in Mantle Cell Lymphoma (MCL), has risen. This study investigates the incidence and risk factors for MPMs in MCL patients. Methods: We analyzed data from the SEER 17 registries (2000–2021), calculating Standardized Incidence Ratios (SIRs) to assess the relative risk of MPMs. Logistic regression identified key risk factors for developing MPMs. Results: Among 10,880 MCL patients, 7,720 (71%) were male, and 8,575 (78.8%) were non-Hispanic White. Of these, 1,309 developed subsequent primary cancers, yielding an SIR of 1.41 (95% CI, 1.34–1.49) and an excess risk of 72.54 per 10,000 person-years. The risk for hematologic malignancies, especially acute myeloid leukemia (AML), was notably higher (SIR, 7.66 [95% CI, 5.90–9.78]; excess risk, 10.58 per 10,000 person-years). Non-hematologic cancers, particularly thyroid cancer, also showed a significant risk increase (SIR, 3.68 [95% CI, 2.59–5.07]; excess risk, 5.12 per 10,000 person-years). Multivariate analysis revealed that female sex (Exp(B) = 1.284, 95% CI: 1.115–1.479, p < 0.001) and longer survival time (Exp(B) = 1.145, 95% CI: 1.131–1.159, p < 0.001) were significant predictors of MPM development. Factors like race, chemotherapy, radiation therapy, and marital status did not show statistically significant associations (p > 0.05). Conclusions: MCL patients face a significantly elevated risk for MPMs, especially hematologic malignancies like AML and non-hematologic cancers such as thyroid cancer. Female sex and prolonged survival are key predictors. These findings underscore the importance of tailored survivorship care and vigilant monitoring for secondary cancers in high-risk MCL patients.
Treatment with brentuximab vedotin in vitro and CAR T-cell function and viability: Implications for combination or sequential therapeutic approaches for lymphoma.
e15014 Background: Brentuximab vedotin (BV), an antibody-drug conjugate targeting CD30, demonstrates potent cytotoxicity against CD30-expressing cancer cells, and remains a standard of care for lymphoma. Recently, chimeric antigen receptor (CAR) T-cell therapy has shown efficacy as monotherapy for lymphoma and could be a viable option for patients with refractory disease after failure of conventional treatment options. Despite the progress made with CAR T-cell therapy, resistance mechanisms and adverse events associated with CAR T cells remain significant hurdles. Combination of CAR T-cell therapy with BV for lymphoma treatment could be an attractive new therapy paradigm, taking advantage of both modalities to drive responses, particularly in populations with relapsed/refractory disease. However, potential impacts of BV treatment on the viability and function of CAR T cells have not been elucidated. Herein, we describe the effects of BV treatment on CAR T cells in vitro to better understand the potential combinations and/or sequencing with CAR T-cell therapies. Methods: This study investigated the cytotoxic and functional effects of BV treatment on several types of CAR T cells in vitro. Previously frozen, commercially generated CAR T cells targeting BCMA, CD19, CD20, or mesothelin were stimulated with a human CD2/CD3/CD28 activator for 48 h. Following stimulation, CAR T cells were then exposed to increasing concentrations of BV for 96 h. Viability, cytokine production, and surface marker expression of CAR T cells and untransduced donor T cells were measured following BV treatment. A nontargeting vedotin ADC and a CD30-targeting ADC known for increased T-cell cytotoxicity were included as negative and positive controls, respectively. Results: CAR T cells were characterized as a mixture of CD4 and CD8 T cells. BV treatment in a dose titration of 1–10,000 ng/mL, including and exceeding clinical exposures, resulted in little to no cytotoxicity against any CAR T-cell type at all concentrations tested. Primary untransduced T cells from the same CAR T-cell donors also displayed no cytotoxicity when treated with BV in vitro. The positive control ADC targeting CD30 showed increased, dose-dependent T-cell cytotoxicity in the same assay. Functional assays revealed that BV-treated CAR T cells retained cytokine production and cytolytic activity compared with controls, regardless of CD30 expression. Conclusions: Preclinical data from this study demonstrated no adverse changes in viability or functionality of CAR T cells with BV treatment in vitro. Collectively, these results suggest the potential to either combine or sequentially use BV and CAR T-cell therapies. Further investigation into optimal combination dosing and/or sequencing strategies is warranted.
Identifying patients at risk for non-receipt of guideline-concordant treatment in pancreatic cancer using electronic health record data and machine learning.
1576 Background: Non-receipt of guideline-concordant treatment (GCT) for pancreatic cancer is associated with worse outcomes and occurs disproportionately in safety-net settings. Timely identification of patients at risk using traditional approaches remains challenging. We developed machine learning (ML) risk prediction models using electronic health record (EHR) data available at presentation to predict non-receipt of GCT and support targeted navigation. Methods: We conducted a retrospective analysis of patients with pancreatic cancer treated at two safety-net hospitals in Alabama from 2018-2022. GCT was defined based on NCCN guidelines. Candidate predictors included demographic, clinical, laboratory, access-to-care and healthcare utilization variables available at presentation. Primary outcome was non-receipt of GCT. Data were split into stratified 80/20 training and test sets given class imbalance (2/3 rd received GCT). Feature engineering was applied to reduce overfitting from high-cardinality variables. Classification and Regression Tree (CART) analysis was used to explore feature interactions and rank predictors. Multiple ML models were trained using PyCaret with default hyperparameters. Model performance was evaluated on the test set, with recall prioritized to reflect a screening use case in which false positives were acceptable. Results: Of 902 patients, 67.5% (n=609) received GCT. CART identified 30 candidate features. Highly ranked features included Area Deprivation Index, age, albumin, time from symptom onset to start of workup, employment status, presence of primary care physician, BMI and ECOG performance status. Performance metrics of the top 5 models are summarized in Table 1. Across models, test-set accuracy ranged from 72.9% to 86.2%, and recall for non-GCT ranged from 35.6% to 61.0%. Light Gradient Boosting model demonstrated the strongest overall screening performance, achieving the highest recall (61.0%), precision (94.7%), F1 score (0.74), and AUC (0.94). Conclusions: ML models using EHR data readily available at presentation can identify patients at high risk for non-receipt of GCT, driven in part by modifiable access-to-care and social vulnerability factors. Integration of these models into EHR workflows could enable automated risk flagging at diagnosis and trigger early referral to navigation and supportive services, with the potential to improve treatment delivery and outcomes. Test-set performance of machine learning models. Model Accuracy AUC Recall Precision F1 score Decision Tree Classifier 0.79 0.77 0.51 0.77 0.61 Logistic Regression 0.73 0.79 0.36 0.66 0.46 Light Gradient Boosting Machine (LGBM) Classifier 0.86 0.94 0.61 0.95 0.74 Extreme Gradient Boosting (XGB) Classifier 0.81 0.87 0.54 0.80 0.65 Ada Boost Classifier 0.77 0.85 0.46 0.73 0.56
Protective factors supporting caregiver well-being in glioblastoma.
12095 Background: Caregiving in glioblastoma (GBM) means dealing with cancer symptoms, plus the neurological decline that comes with this disease. Becoming a caregiver is rarely straightforward. Caregivers often become essential to the patient’s daily life and well-being, though this may result in neglecting their own health. Most caregivers develop coping strategies (protective factors such as caregiver mastery or spiritual beliefs) to manage the responsibility of caring for a terminally ill loved one. This study looked at how caregiver protective factors shifted across the disease trajectory and whether these factors were connected. Methods: This was a secondary analysis of data from a three-group randomized clinical trial testing a needs-based intervention for neuro-oncology caregivers (NINR # R01 NR013170). Caregivers were included if they were 21 or older, identified by the patient as their unpaid primary caregiver, spoke English, had phone access, were not caring for anyone else (except children under 21), and scored above 6 on the CES-D. Recruitment occurred from March 2014 to July 2016 at the University of Pittsburgh Medical Center, the University of Pittsburgh Cancer Institute, and the MD Anderson Cancer Center. Patients and caregivers gave informed consent prior to participation in the primary study. We conducted an a priori power analysis to determine the sample size required for detecting minimal differences in quality of life (QOL). The final sample included 60 caregivers. We measured outcomes at baseline, two, four, six, and ten months. Analysis used paired samples t-test, bivariate correlations, and linear mixed models. Results: Caregiver QOL changed significantly over time, particularly from baseline to four months ( p = 0.02) and four to six months ( p = 0.01). Protective factors correlated with one another at varying strengths (weak, moderate, and strong). At each time point, religion and spirituality were related to other protective factors. Spirituality had significant positive correlations with caregiver mastery ( r = 0.32-0.65, p = < .001-0.01) and QOL ( r = 0.44-0.70, p = < .001-0.01) throughout the study. After factoring in optimism, we found significant associations for caregiver mastery, positive aspects of caregiving, social support, and religion. Conclusions: Spirituality emerged as a key protective factor, yet most caregiver interventions ignore it. Providers should recognize and support spiritual coping mechanisms throughout GBM care, offering resources that match caregiver values and beliefs. Monitoring caregiver QOL is equally critical as tracking patient outcomes. When QOL declines, early interventions can prevent caregiver burnout and improve care quality for both members of the dyad.
A US real-world study of second-line outcomes in advanced ovarian cancer by first-line PARP inhibitor maintenance exposure.
e17564 Background: As PARPi maintenance (MTx) is widely used after first-line (1L) platinum-based therapy for advanced ovarian cancer (aOC), understanding its impact on subsequent platinum (PLT) sensitivity is critical. This real-world (rw) study described characteristics of aOC patients who received second-line (2L) therapies after 1L PARPi MTx or 1L PLT and evaluated 2L outcomes with 2L PLT versus non-platinum (no-PLT) therapies. Methods: This retrospective cohort study used chart review data from The US Oncology Network electronic medical records. Adult women with Stage III–IV OC initiating 2L systemic treatment (index date) between 1-Jan-2017 and 1-May-2023 following progression after 1L PARPi or 1L PLT were included, with follow-up to 31-Oct-2023. Descriptive statistics were used to summarize patient characteristics and treatment patterns. Kaplan-Meier analysis was used to assess rw treatment-free interval (rwTFI) after 2L treatment, rw platinum-free interval (rwPFI) after 2L PLT, rw progression-free survival (rwPFS), and overall survival (OS) from index date. Results: A total of 165 patients received 2L therapy after 1L PARPi MTx (PARPi cohort) and 83 PARPi-naïve patients received 2L therapy after 1L PLT (PARPi-naïve cohort), selected via stratified random sampling. In the PARPi cohort, 20% (n=33) were PLT resistant (PROC, progressed 1-6 months [mo] after 1L PLT), and 64.2% (n=106) were PLT sensitive (PSOC, progressed >6 mo after 1L PLT), most of whom progressed after PARPi (n=101) and only 5 progressed while on PARPi. In the PARPi cohort, median 2L rwTFI was 1.6 mo with 2L PLT and 1.3 mo with 2L no-PLT; in the PARPi-naïve cohort, median 2L rwTFI was 4.7 mo with 2L PLT and 1.0 mo with 2L no-PLT. Table 1 reports medians and 95% confidence intervals (CIs) of 2L outcomes in each cohort by 2L PLT use by PSOC and PROC. Conclusions: This analysis showed a pattern of numerically more favorable outcomes with 2L PLT than no-PLT, with greater benefit in 1L PARPi-naïve versus PARPi-exposed patients. These descriptive, hypothesis-generating findings suggest that 1L PARPi MTx may impact 2L outcomes, including effectiveness of 2L PLT, regardless of PLT sensitivity, underscoring the need for novel therapies and personalized approaches for patients previously exposed to PARPi. 2L median outcomes in months (95% CI) 1L PARPi MTx PSOC 2L PLT No 2L PLT PROC 2L PLT No 2L PLT 1L PARPi-naive PSOC 2L PLT No 2L PLT PROC 2L PLT No 2L PLT n 79 27 10 23 30 8 9 20 rwTFI 2.0 (1.4,3.2) 1.1 (0.7,2.6) 0.9 (0.0,1.8) 1.5 (1.0,2.1) 5.4 (1.0,8.3) 0.9 (0.1,NR) 1.4 (0.7,NR) 1.1 (0.7,1.4) rwPFI 2.1 (1.0,4.1) NA 0.9 (0.1,2.2) NA 2.6 (0.7,8.3) NA 2.5 (0.7,NR) NA rwPFS 8.3 (7.3,11.5) 5.6 (2.9,6.5) 6.4 (1.2,8.8) 3.7 (1.9,5.2) 9.7 (6.1,18.0) 4.1 (1.8,NR) 8.1 (2.8,15.5) 2.5 (1.9,5.8) OS 20.4 (15.6,34.6) 18.8 (10.6,29.8) 12.3 (1.5,NR) 9.7 (5.7,19.4) 23.2 (13.6,NR) 28.5 (1.8,NR) 15.7 (9.0,NR) 12.2 (3.1,20.5) NA=not available; NR=not reached.
Metric indicators of lymphedema and their impact on symptoms and quality of life in breast cancer patients: A longitudinal real-world study.
e12713 Background: Upper-limb lymphedema is a frequent complication of breast cancer, with impact on symptoms and quality of life (QoL). Although its severity can be objectively quantified, longitudinal real-world data remain limited. The aim of this study was to evaluate associations between lymphedema metric changes, clinical evolution, and QoL in breast cancer patients. Methods: Real-world data from 298 breast cancer patients with lymphedema were analyzed at lymphedema diagnosis (T1) and after rehabilitation program (T2). Upper-limb circumferences were aggregated into proximal (arm, elbow, forearm) and distal (wrist region, hand) indices. Symptoms (pain, fatigue), QoL, muscle strength, and joint range of motion (ROM) were assessed using validated instruments. Longitudinal associations (ΔT2–T1) were analyzed using Spearman correlations with false discovery rate (FDR) adjustment. Results: Although proximal circumference reductions were more frequent, distal metrics showed stronger associations with symptoms and QoL. Reductions in distal upper-limb circumferences were longitudinally associated with reductions in fatigue (ρ≈0.21; q_FDR≈0.001–0.002), improvement in total FACT-G7 score (ρ≈0.14–0.17; q_FDR < 0.05), and increases in global health status score (QLQ-GHS) (ρ≈−0.18; q_FDR < 0.01). At T1, higher muscle strength values were associated with lower levels of pain (ρ≈−0.27; q_FDR < 0.0001) and fatigue (ρ≈−0.25; q_FDR < 0.0001). Longitudinally, increases in muscle strength were correlated with reductions in pain and fatigue between T1-T2. ROM parameters showed significant associations with patient-reported scores. At T1, abduction correlated with physical functioning (QLQ-PF) (ρ = 0.189; q_FDR = 0.006), total FACT-G7 score (ρ = −0.175; q_FDR = 0.006), and QLQ-GHS (ρ = 0.170; q_FDR = 0.007). Flexion correlated with FACT-G7 (ρ = −0.181; q_FDR = 0.006) and QLQ-PF (ρ = 0.128; q_FDR = 0.043). Longitudinally, increases in abduction between T1-T2 were associated with reductions in total FACT-G7 score (ρ = −0.195; q_FDR = 0.006). Conclusions: In real-world practice, upper-limb metric and functional parameters in breast cancer lymphedema are longitudinally associated with pain, fatigue, and QoL. Despite more frequent proximal circumference reductions, distal edema dynamics emerge as a key determinant of QoL outcomes. These results support integrated metric, ROM, and patient-reported assessment pathways and therapeutic approaches for supportive oncology care.
Intrathecal iPSC-derived natural killer cells combined with the Stupp regimen for newly diagnosed WHO grade 4 glioma: A phase I trial.
2072 Background: WHO grade 4 glioma has a poor prognosis despite standard Stupp therapy. Induced pluripotent stem cell–derived allogeneic natural killer cells (iPSC-NK) represent a standardized and scalable cellular immunotherapy with consistent cytotoxic potential. This Phase I study evaluated intrathecal iPSC-NK cells combined with the Stupp regimen in newly diagnosed WHO grade 4 glioma. Methods: Thirteen newly diagnosed WHO grade 4 glioma patients (median age 46 years; IDH-wildtype n=11; MGMT-methylated n=6) received standard chemoradiotherapy followed by intrathecal iPSC-NK cell infusions. iPSC-NK cells were administered at escalating doses (2×10⁷, 6×10⁷, 2×10⁸ cells/cycle) via lumbar puncture (n=8) or Ommaya reservoir (n=5) over four 28-day cycles, with temozolomide priming 5 days before each infusion. The primary endpoint was safety (serious adverse events, ≥Grade 3 toxicities); secondary endpoints included progression-free survival and pharmacokinetic/pharmacodynamic immune analyses. Results: Across 54 intrathecal iPSC-NK infusions, treatment-related toxicities were predominantly low grade: fever occurred as Grade 1 in 35 (64.8%) and Grade 2 in 9 (16.7%) infusions; headache as Grade 1-2 in 19 (35.2%) and Grade 3 in 1 (1.9%) infusion; and cytokine release syndrome was limited to Grade 1 events (44/54, 81.5%), with no ≥Grade 2 CRS or ICANS. Median PFS was 17.41 months, while median OS was not reached at a median follow-up of 22.47 months (data cutoff: January 18, 2026). PK analysis showed detectable iNK cells in CSF on Days 1–3 post-infusion, peaking on Days 1–2 and declining by Day 3. PD analyses demonstrated rapid activation of CD4⁺/CD8⁺ T cells (CD69⁺) and expansion of myeloid populations. Olink proteomics revealed consistent upregulation of cytotoxic (GZMA/B), chemokine (CXCL/CCL), and inflammatory cytokines (TNF, IFN-γ). Conclusions: iPSC-NK cells combined with the Stupp regimen demonstrated a favorable safety profile, induced early immune activation within the central nervous system, and showed encouraging preliminary efficacy in patients with WHO grade4 glioma. Clinical trial information: NCT06147505 .