Browse Articles

Assessment of efficacy of hereditary cancer didactic training program for advanced practice providers in the community oncology setting.

e13546 Background: Many professional organizations, including ASCO, recommend evaluation of personal and family history of cancer and, when applicable, germline genetic testing as part of comprehensive oncology care (Tung et al, 2024). Additionally, the number of patients qualifying for germline genetic testing based on diagnosis continues to increase. Despite this, standardization of genetics education for oncology care providers, including medical oncologists and advanced practice providers (APPs, consisting of nurse practitioners and physician assistants) is lacking. The Texas Oncology (TXO) genetics program aims to provide this education to our oncology providers. Methods: Since 2012, TXO genetic counselors have provided individualized hereditary cancer education through didactic and clinical training. To support growth and ensure consistency, a standardized program was developed based on NCCN, ASCO, and NSGC core competencies. The CME-accredited program includes over 20 hours of video content organized into 10 topic blocks and weekly 90-minute live sessions for 10–11 weeks to reinforce concepts and apply them clinically. Participants completed a multiple-choice test before training to assess baseline knowledge and repeated the test after training to earn CME. Test results informed revisions of recorded content and live sessions for future trainees. Results: 38 APPs have completed the hereditary cancer education course (18 in fall 2024, 8 in spring 2025, 12 in fall 2025). The post-training test average scores across the 3 groups improved from the pre-training test (Table 1). There is consistency in questions missed across all 3 groups, indicating a gap in teaching and provider knowledge. Conclusions: The overall improvement in post-training test scores indicates that the didactic training provided increases genetics knowledge in 37/38 (97.4%) of providers that completed training. Topics that showed most improvement in understanding included inheritance, use of microsatellite instability and immunohistochemistry testing, management of patients with a hereditary cancer syndrome and pedigree assessment. The topics that did not show overall improvement of understanding included use of somatic test results to guide germline testing and use of breast cancer risk assessment models for personalized management. In the fall 2024 and fall 2025 groups, at least 25% of participants (fall 2024: 5 of 18 participants, fall 2025: 3 of 12 participants) had a post-training test score under 75%. This deserves further exploration into training group size and targeted training based on missed concepts. Training Group Pre-Training Test(% correct) Post-Training Test(% correct) Fall 2024 (n=18) 57.6% (43.3% - 74.6%) 79.3% (56.7% - 95.5%) Spring 2025 (n=8) 56.3% (42.0% - 68.1%) 80.1% (75.3%-89.8%) Fall 2025 (n=12) 58.8% (42.0% - 95.7%) 83.8% (68.1% - 92.8%)

Biogenic synthesis of titanium oxide nanoparticles using Senna auriculata (L.) flower: Antioxidant, anti-arthritic, and antimicrobial Potentials

Real-world experience with inotuzumab ozogamicin in high-risk and R/R B-cell acute lymphoblastic leukemia.

e18507 Background: Inotuzumab ozogamicin (InO) is an effective CD22-directed antibody–drug conjugate for relapsed/refractory (R/R) B-cell acute lymphoblastic leukemia (B-ALL). However, real-world data describing dose-modified InO use in combination or as monotherapy, particularly in post-transplant settings and regions with a high prevalence of metabolic liver disease, is limited. Methods: We retrospectively analyzed consecutive patients with B-ALL treated with InO at our center from October 22,2022 to January 1, 2026. Data collected included demographics, disease biology, indication for InO use, combination with chemotherapy (mini-HCVD), InO dosing and cumulative exposure, response (CR/CRi and MRD), transplant outcomes, and toxicities including SOS/VOD and transplant-associated thrombotic microangiopathy (TA-TMA). Results: 15 patients with B-ALL were included (median age ~17 years; range 6–65); 53% had high-risk biology (TP53, IKZF+, or Ph+). InO was used as bridge to allo-HSCT (47%), post-allo relapse therapy (27%), post-allo maintenance (27%), sequentially with blinatumomab (20%), or as pre-phase stabilization (7%). Inotuzumab was administered in combination with mini-HCVD in 8 of 15 patients (53%), while the remainder received single-agent InO or other combinations based on disease burden and clinical context. All patients received 0.3 mg/m² per dose. The median number of doses was 4 (range 1–12), with a median cumulative dose of ~1.2 mg/m² (range 0.3–3.6 mg/m²), substantially lower than label-based cumulative exposure, reflecting combination therapy, frequent post-transplant use, and regional concern for baseline hepatic vulnerability including underlying fatty liver disease. Overall, CR/CRi was achieved in 80%, with MRD-negative CR in 60%, including patients with TP53-mutated and Ph+ disease. Among transplant-eligible patients, 70% were successfully bridged to allo-HSCT, and durable CR ≥12 months was observed in 40%. Severe SOS/VOD occurred in 1 patient (6.6%), in the post-allo relapse setting, and was fatal. TA-TMA was observed in 27%, clustering in post-allo settings. Conclusions: In this real-world cohort, dose-modified inotuzumab ozogamicin, frequently combined with mini-HCVD, produced high response and MRD-negativity across high-risk B-ALL. Toxicity appeared context-dependent even at lower doses, with rare but severe SOS/VOD and a notable endothelial toxicity signal in post-allo settings. These findings support cautious dosing and timing-based integration of inotuzumab ozogamicin, particularly in transplant-exposed populations and regions with prevalent metabolic liver disease. Parameter Result Patients (B-ALL) 15 Median age (range) ~17 years (6–65) High-risk biology 53% Inotuzumab + mini-HCVD 8 patients (53%) Median InO doses 4 (range 1–12) Median cumulative dose ~1.2 mg/m² CR/CRi 80% MRD-negative CR 60% SOS/VOD 6.6% TA-TMA 27%

COPERNICUS, a pragmatic phase 2b study of subcutaneous (SC) amivantamab (ami) + chemotherapy (chemo) with enhanced dermatologic adverse event (AE) prophylaxis in <i>EGFR</i> -mutated advanced NSCLC: Interim results.

8614 Background: In MARIPOSA-2, intravenous ami + carboplatin-pemetrexed chemo significantly prolonged progression-free survival (PFS) vs chemo (HR, 0.48; P &lt;0.001) in participants (pts) with EGFR -mutated (exon 19 deletion [Ex19del]/L858R) advanced NSCLC after progression on osimertinib. However, longer infusion times, infusion-related reactions (59%), and dermatologic AEs (paronychia [37%], rash [43%]) were observed, with frequent interruptions of ami due to AEs (60%) as a potential result. Numerous studies have since tested ways to optimize ami administration. PALOMA-3/-2 showed reductions in administration-related reactions (ARRs) and administration time with SC ami coformulated with hyaluronidase (rHuPH20), thus enhancing patient experience and leading to approval by the FDA/EMA. COCOON also showed fewer grade ≥2 dermatologic AEs vs standard of care with an enhanced prophylactic regimen. Methods: COPERNICUS (NCT06667076) is the first study to combine SC ami and optimized supportive care, using a pragmatic design to broaden the pt population and better resemble real-world usage. We report planned interim results of Cohort 2 for SC ami every 3 weeks (Q3W) + chemo on/after EGFR TKI progression in US pts with EGFR Ex19del/L858R NSCLC receiving dermatologic AE prophylaxis aligned with the regimen described in COCOON. Pragmatic design included partnering with academic/community sites and less stringent eligibility criteria to enhance pt diversity. Primary endpoint is PFS by investigator. Key secondary endpoints are overall response rate (ORR) and safety, including incidence/severity of dermatologic AEs and ARRs. All comparisons to MARIPOSA-2 are descriptive. Results: As of data cutoff (02 Jan 2026), 29 pts had enrolled in Cohort 2 (target enrollment, 30; median [range] follow-up: 7.6 [0.5+–10.2] mo); 76% were still ongoing in the study. Median age was 62 y, with 45% of pts ≥65 y and 21% ≥75 y; 38% were Asian and 7% African American. Median PFS was 7.4 mo (95% CI, 4.8–NE; Table). AEs were mostly grade 1–2, with no new safety signals; 31% of pts interrupted ami due to AEs. With dermatologic prophylaxis, paronychia and rash occurred in 24% and 14% of pts, respectively, showing numerical reductions vs MARIPOSA-2. ARRs (grouped term) were also numerically lower at 21%. Conclusions: Compared with MARIPOSA-2, SC ami and dermatologic prophylaxis in COPERNICUS led to substantial reductions in ARRs, dermatologic AEs, and ami interruptions, establishing the positive effect of early supportive care interventions. These interim data obtained using a pragmatic design support wide use of SC ami Q3W + chemo post-EGFR TKI progression in a diverse population. Clinical trial information: NCT06667076 . Median PFS, mo (95% CI) 7.4 (4.8–NE) ORR (95% CI) 24.1% (10.3–43.5) Partial response 7 (24.1%) Stable disease 15 (51.7%) Progressive disease 2 (6.9%)

A research-based initiative to coordinate care for patients with basal cell carcinoma and cutaneous squamous cell carcinoma.

e21004 Background: Community teams managing basal cell carcinoma (BCC) and cutaneous squamous cell carcinoma (cSCC) report variation in risk stratification, referrals, and integration of systemic therapy. Methods: A mixed-methods needs assessment (survey n = 62; interviews n = 12; Dec 2024-Feb 2025) informed two ACCME-accredited, online enduring CME activities (1.5 total credits) plus targeted microlearning. Outcomes included participation and pre/post knowledge (3 items), confidence, case-based competence, and intended practice change. Post-activity qualitative interviews are planned to assess perceived educational impact and workflow change (eg, multidisciplinary coordination, referral pathways, and systemic therapy/immune-related AE monitoring). Results: Survey respondents (90% community practice; 55% dermatologists/Mohs surgeons, 45% oncologists) managed a mean of 27 BCC and 26 cSCC patients/month; &gt; 30% managed &gt; = 40/month. Multidisciplinary involvement was inconsistent (15% always, 34% often) and only 34% reported access to cutaneous/dermatologic oncologists. Barriers to integrating systemic therapy included coordinating multi-specialty appointments (61%), lack of clinical guidelines (50%), and patient travel (40%). Challenges included evaluating recurrence/metastasis risk (63%) and referral to oncology when curative surgery/radiation were not feasible (45%). Interviews highlighted heterogeneous definitions of high-risk disease, selective imaging/molecular testing, limited standardized surveillance, and fragmented referral/communication pathways; dermatologists often deferred immune checkpoint inhibitor (ICI) adverse-event monitoring to oncology and requested practical algorithms and patient materials. Education reached 4,041 clinicians (834 enduring CME; 3,207 microlearning). Mean knowledge improved 32% to 63% and confidence identifying ICI-eligible patients improved 19% to 36%. Case-based competence improved for metastatic BCC primary systemic treatment selection (41%-56% gains), borderline resectable cSCC neoadjuvant therapy selection (46%-49% gains), and delayed immune-related adverse-event timing (39%-43% gains); baseline underestimation of delayed events was common (77% dermatology, 81% oncology). Seventy percent of intended learners planned practice changes. Conclusions: Mixed-methods assessment identified actionable community gaps in coordinated BCC/cSCC care. An aligned education initiative improved knowledge and decision-making; planned post-activity interviews will further evaluate perceived practice impact and implementation.

FRAmework-01: A two-part phase 3 study of sofetabart mipitecan versus chemotherapy or mirvetuximab soravtansine in platinum-resistant ovarian cancer (PROC), and sofetabart mipitecan plus bevacizumab versus platinum-based chemotherapy plus bevacizumab in platinum-sensitive ovarian cancer (PSOC).

TPS5644 Background: Folate receptor alpha (FRα) is overexpressed in ovarian cancer and is a validated therapeutic target. Mirvetuximab soravtansine (MIRV), an antibody drug conjugate (ADC) targeting FRα, is approved for the treatment of patients with PROC and high FRα expression (≥75% tumor cells with 2+ and/or 3+ staining intensity). Given MIRV’s limited indication and its association with specific side effects, including ocular adverse events and peripheral neuropathy, there remains a significant medical need to improve outcomes for a broader population of patients with ovarian cancer. Sofetabart mipitecan is an FRα-targeting ADC composed of an Fc-silenced, humanized IgG1 antibody, a novel polysarcosine hydrophobicity masking agent with a dipeptide cleavable linker, and the topoisomerase I inhibitor payload exatecan at a drug-antibody ratio of 8. Data from the Phase 1 study indicate that sofetabart mipitecan monotherapy demonstrates promising and durable clinical activity, with high objective response rates and a tolerable safety profile, in heavily pretreated patients with PROC. Responses were observed across all FRα levels, including in patients who received prior MIRV treatment, and there were no reports of keratopathy or significant peripheral neuropathies. Sofetabart mipitecan was recently granted Breakthrough Therapy Designation by the FDA. Methods: In Part A, 530 participants with PROC who received ≤3 prior lines of systemic cytotoxic therapy (or ≤4 lines if MIRV was included), will be randomized in a 1:1 ratio to receive either sofetabart mipitecan, or the investigator’s choice (IC) of chemotherapy (paclitaxel, pegylated liposomal doxorubicin [PLD], gemcitabine, or topotecan) or MIRV (if eligible per local label). In Part B, 550 participants with PSOC who received ≤2 prior lines of systemic cytotoxic therapy and had disease progression either during or within six months of completing treatment with a poly (ADP‐ribose) polymerase inhibitor (PARPi), will be randomized in a 1:1 ratio to receive either sofetabart mipitecan plus bevacizumab (bev), or the IC of carboplatin-based chemotherapy (paclitaxel, PLD, or gemcitabine) plus bev. Study Parts A and B will enroll participants irrespective of their tumor’s FRα expression level. Each part has progression-free survival as its primary endpoint, while overall survival is a key secondary endpoint. Study parts will be conducted separately, and statistical analyses will be performed independently. Trial Status: The study is actively enrolling participants in Part A and B. Clinical trial information: NCT07213804 .

Efficacy and safety of rivaroxaban compared with LMWH in patients with cancer-associated thrombosis: A systematic review and meta-analysis of randomized controlled trials.

e15136 Background: The optimal anticoagulation strategy for managing cancer-associated thrombosis (CAT) remains uncertain. The choice between low molecular weight heparin (LMWH), the traditional standard of care, and rivaroxaban, a newer oral direct factor Xa inhibitor, lacks clear guidance due to inconclusive evidence. To address this gap, we conducted a systematic review and meta-analysis to compare the safety and efficacy of rivaroxaban versus LMWH in patients with CAT. Methods: We did an extensive search using PubMed, Clinicaltrials.gov and Cochrane Library for studies published from inception to June 2025, evaluating the use of rivaroxaban and LMWH in patients with CAT. Data were extracted and analyzed on RevMan v.5.2. Pooled proportions with 95% confidence intervals (CI) were calculated using a random effects model. Risk ratios (RR) with 95% confidence intervals were used as effect measures for dichotomous variables. The I 2 and χ 2 were used to assess interstudy heterogeneity. A p-value of &lt; 0.05 was considered statistically significant. Results: This meta-analysis included a total of five studies, involving 1,533 patients (622 patients were on rivaroxaban, and 639 patients were on LMWH. The inclusion criteria were patients aged &gt; 18 years with active cancer sites experiencing VTE. Rivaroxaban showed a statistically significant association with a 48% lower risk of recurrent VTE compared to LMWH (RR 0.52, 95% CI 0.30-0.92, I2 = 0%, P = 0.02) and 90% higher risk of minor bleeding compared to LMWH (RR 1.90, 95% CI 1.01-3.55, I2 = 46%, P = 0.05). Rivaroxaban reduces the risk of DVT (RR 0.70, 95% CI 0.44-1.12, I2 = 0%, P = 0.13) and increases the risk of major bleeding (RR 1.47, 95% CI 0.86-2.53, I2 = 0%, P = 0.16) compared to LMWH; however, results are not statistically significant. Conclusions: Our findings suggest that rivaroxaban had 48% lower risk of recurrent VTE, but a 90% higher risk of minor bleeding compared to LMWH. Results for DVT and major bleeding were non-significant, indicating that rivaroxaban may be a reasonable alternative to LMWH for patients with CAT, but caution is required in patients with bleeding risk.

Long term trends in Hodgkin's lymphoma incidence in the United States and Puerto Rico, 1999 to 2022.

e19043 Background: Hodgkin lymphoma (HL) is a highly curable malignancy; however, population-level incidence trends vary across demographic subgroups. Contemporary national data describing long-term HL incidence patterns across age, sex, and race/ethnicity remain limited. We evaluated temporal trends in HL incidence in the United States from 1999 to 2022. Methods: Incident HL cases from 1999 to 2022 were obtained from the United States Cancer Statistics (USCS) Incidence database via CDC WONDER. Age-adjusted incidence rates were calculated per 100,000 population using the 2000 U.S. standard population. Trends were assessed overall and stratified by sex, age (&lt;25, 25–44, 45–64, ≥65 years), race and ethnicity (Non-Hispanic White, NH Black's, NH Asian or Pacific Islander, NH American Indian or Alaska Native, and Hispanic). Joinpoint regression was used to estimate average annual percent change (AAPC) with 95% confidence intervals (CI). Results: A total of 204,884 incident HL cases were identified in the United States between 1999 and 2022. Overall age-adjusted HL incidence declined from 2.79 in 1999 to 2.48 in 2022, corresponding to an AAPC of −0.57% (95% CI, −0.68 to −0.46). Incidence rates were consistently higher among males than females across all years, with steeper declines among males (AAPC −0.63%; 95% CI, −0.79 to −0.43) compared with females (AAPC −0.43%; 95% CI, −0.57 to −0.26). Age-stratified analyses demonstrated stable incidence among individuals aged &lt;25 years (AAPC 0.12%; 95% CI, −0.02 to 0.30), while sustained declines were observed among adults aged 25–44 years (AAPC −0.71%; 95% CI, −0.91 to −0.46), 45–64 years (AAPC −0.61%; 95% CI, −0.86 to −0.36), and ≥65 years (AAPC −0.59%; 95% CI, −0.81 to −0.32). Incidence declined among NH White's (AAPC −0.66%; 95% CI, −0.81 to −0.49) and Hispanic individuals (AAPC −0.55%; 95% CI, −0.80 to −0.27). In contrast, incidence increased among NH Black's (AAPC 0.27%; 95% CI, 0.00 to 0.57) and NH Asian or Pacific Islander's (AAPC 0.75%; 95% CI, 0.38 to 1.21), while no consistent long-term trend was observed among NH American Indian or Alaska Native individuals (AAPC 0.32%; 95% CI, −0.73 to 1.50). Conclusions: Hodgkin lymphoma incidence in the United States declined steadily from 1999 to 2022, driven primarily by reductions among adults aged ≥25 years and White and Hispanic populations. In contrast, increasing incidence among Black or African American and Asian or Pacific Islander individuals highlights emerging demographic divergence in disease burden, warranting further investigation.

Investigating quality of life and decision regret in patients undergoing laryngectomy procedure.

e23267 Background: Total laryngectomy is a surgical procedure for head and neck cancers that involves the removal of the larynx and tongue. This operation negatively impacts key functions such as speech, swallowing, and breathing. The primary purpose of this study is to describe the quality of life among patients who have undergone total laryngectomy with adjuvant radiation or salvage laryngectomy. In addition, the study examines the level of decision regret in these groups. These findings will inform best practices, including improving patient awareness of potential outcomes and challenges following surgery. Methods: The survey was distributed to two groups of patients who had undergone a laryngectomy procedure. The first group included 25 patients who received a primary laryngectomy with adjuvant radiotherapy/treatment, while the 27 patients in the second group underwent salvage laryngectomy. Their quality of life was assessed using the University of Washington Quality of Life Questionnaire (UW-QOL). The Decision Regret Scale was utilized to evaluate the patients’ perspectives on their treatment decisions. Results: Previous studies have revealed that an MDADI score of 60 and above represents an adequate quality of life. According to our analysis, we found that Group 1 patients have a mean MDADI score of 70.28, while Group 2 patients have a lower 63.90 score. The 95% confidence interval (CI) (55.8 – 72.0) of Group 2 patients lies below the benchmark of 60, unlike the CI of the Group 1 patients (64.6 – 75.9). For the Decision Regret, any score over 26 is clinically significant regret, and a score of 40 and above shows total regret of their decision. Group 1 patients reported a decision regret score of 31.4, whereas Group 2 patients had a higher score of 35.37, indicating greater regret. Overall, this shows that while both patient groups report an adequate quality of life with moderate decision regret, Group 1 demonstrates better outcomes compared to Group 2. Conclusions: Our results demonstrate that patients who undergo laryngectomy with adjuvant therapy report a higher quality of life than those who undergo salvage laryngectomy, although both groups meet criteria for an adequate quality of life. Both groups demonstrate clinically significant decision regret, though neither group demonstrated total regret. These findings highlight the importance of fully informing patients of the impact of surgery on quality of life to assist them in their decision-making. Future studies can recruit larger cohorts and investigate the effectiveness of other laryngectomy treatments/procedures.

Effect of HOXB7 on oxaliplatin resistance in colon cancer through glutamine metabolic reprogramming via non-canonical activation of YAP1.

e15732 Background: Colorectal adenocarcinoma (COAD) ranks as the third most commonly diagnosed cancer and the third leading cause of cancer-related mortality worldwide. A major obstacle to successful chemotherapy for this malignancy is the development of resistance to oxaliplatin, a cornerstone chemotherapeutic agent. Intriguingly, recent studies have implicated the homeobox B cluster gene family in both oxaliplatin resistance and metabolic reprogramming. Based on this emerging evidence, we undertook the present study to investigate the specific role of HOXB7 in the development of oxaliplatin resistance in COAD. Methods: To explore the role of HOXB7, we first generated isogenic oxaliplatin-resistant CC cell lines and conducted analyses of clinical specimens. Then, we performed functional loss-of-experiments using knockdown approaches to assess the impact on proliferation, glutamine metabolism, and drug sensitivity. To decipher the underlying mechanism, we investigated the relationship between HOXB7, YAP1, and the deubiquitinase OTUB1 through overexpression rescue experiments and mechanistic studies. Furthermore, utilizing a structure-based virtual screening strategy of a high-throughput compound library, we sought to identify a potential HOXB7 inhibitor. The therapeutic potential of the identified compound, Morin, was then validated in patient-derived xenograft models. Results: Our analyses revealed that HOXB7 expression was significantly elevated in CC tissues compared with adjacent non-tumor tissues. Clinically, high HOXB7 expression was correlated with reduced disease-free survival in COAD patients. Functionally, in vitro knockdown of HOXB7 effectively suppressed CC cell proliferation, impaired glutamine metabolism, and consequently restored oxaliplatin sensitivity. Mechanistically, we discovered that HOXB7 upregulation increased YAP1 protein expression, and critically, overexpression of YAP1 reversed the phenotypic effects of HOXB7 knockdown. We further elucidated that HOXB7 transcriptionally upregulates the deubiquitinase OTUB1. OTUB1, in turn, stabilizes YAP1 protein via deubiquitination, ultimately enhancing glutamine metabolism through Hippo-independent pathways. Importantly, our drug screening efforts identified the FDA-approved drug Morin as a potent HOXB7 inhibitor and oxaliplatin sensitizer. This combination strategy demonstrated a significant enhancement of the antitumor effect of oxaliplatin in patient-derived xenograft models. Conclusions: This study elucidates a novel HOXB7-OTUB1-YAP1 axis that promotes chemoresistance in colon cancer by modulating glutamine metabolism. Our findings posit that targeting this pathway with the identified HOXB7 inhibitor, Morin, represents a promising and readily translatable therapeutic strategy for reversing oxaliplatin resistance in COAD patients.

Co-operation of contralateral prophylactic mastectomy and risk-reducing salpingo-oophorectomy in Korean breast cancer patients with germline <i>BRCA1/2</i> mutation.

e22651 Background: Carriers of germline BRCA1/2 pathogenic variants are considered at high risk for hereditary breast and ovarian cancer. In unilateral breast cancer patients with these variants, contralateral prophylactic mastectomy (CPM) and risk‐reducing salpingo-oophorectomy (RRSO) are recommended; however, undergoing two separate surgeries may pose a substantial burden on carriers. Methods: We retrospectively analyzed BRCA1/2 germline variant carriers who were diagnosed and treated for breast cancer at our institution between 2014 and 2024. Surgical duration, length of hospital stays, and recurrence-free survival (RFS) were compared between patients who underwent cooperative surgery (COS) and those who underwent standalone surgeries (SAS) for CPM and RRSO. Results: Among 554 carriers, 261 did not undergo preventive surgery, 54 underwent CPM only, 132 underwent RRSO only, and 107 underwent both CPM and RRSO. Of the 101 patients who underwent both surgeries at our institution, 83 underwent COS and 18 underwent SAS. The cohort included 44 BRCA1 and 57 BRCA2 carriers, with no significant differences between groups in mutation type, family history, or breast cancer subtype. The age at breast cancer diagnosis and the age at CPM were younger in the SAS group (39.3 years) than in the COS group (48.4 years) (p&lt;0.001), and the age at RRSO was also younger in the SAS group (41.1 years) compared with COS (48.7 years). There were no differences between groups in CPM surgical method, concurrent breast reconstruction, or hysterectomy at the time of RRSO. Total operative time was significantly shorter in the COS group (352.5 minutes) compared with SAS (409.0 minutes) (p=0.035). Length of hospital stay was also shorter in the COS group (6 days) than in the SAS group (16 days). Postoperative breast surgery complications did not differ between groups. Mean RFS was 133.1 months in the COS group and 128.1 months in the SAS group (p=0.039), and median RFS was not reached in either group. In multivariable Cox regression, the only statistically significant factor was age at breast cancer diagnosis (&lt;40 years). Conclusions: Among Korean carriers of BRCA1/2 germline pathogenic variants with breast cancer, COS was associated with shorter operative time and reduced hospitalization compared with SAS, without inferiority in postoperative complications or recurrence-free survival. Further research is warranted to evaluate long-term oncologic outcomes and patient-reported satisfaction.

National trends in opioid-related hospitalizations among adult cancer patients: 2000-2022.

11147 Background: Opioid use disorder is increasingly prevalent among cancer patients. Since mid-2010s, heightened attention to the opioid crisis has led to restrictive opioid prescribing policies; however, national trends in opioid-related hospitalizations (ORH) among cancer patients following these policies remain underreported. This study examines ORH trends from 2000 to 2022 and associated factors among adult cancer patients. Methods: We identified hospitalizations among adult patients (age ≥18 years) with any cancer in the National Inpatient Sample (2000-2022). Outcomes included: 1) primary ORH (PORH), defined as admissions with a primary diagnosis indicating opioid poisoning, dependence or abuse, and 2) any ORH (AORH), capturing these conditions in any diagnostic field. The risks of PORH and AORH were calculated as the number of cases per 10,000 hospitalizations, respectively. Join-point regression assessed trends in risks of PORH and AORH, reporting annual percentage changes (APCs). Multivariable logistic regression identified factors associated with PORH and AORH, with adjusted odds ratio (aORs) reported. Results: Overall, among 59,682,271 weighted hospitalizations in cancer patients (median age 66.8 years, IQR 56.5-76.3), PORH occurred in 5.5 per 10,000 hospitalizations (n = 32,641) and AORH in 200 per 10,000 hospitalizations (n = 1,192,358). The risk of PORH increased non-significantly from 2.2 per 10,000 hospitalizations in 2000 to 4.8 in 2014 (APC = 5.6, p &gt; 0.05), then rose sharply to 10.6 in 2017 (APC = 25.1, p &lt; .001), before declining significantly to 6.8 in 2022 (APC = -9.7, p &lt; .001). The risk of AORH also increased non-significantly from 45 per 10,000 hospitalizations in 2000 to 60.8 in 2014 (APC = 0.2, p &gt; 0.05), then rose to 594.4 in 2019 (APC = 118.2 for 2013-2016 and 11.7 for 2016-2019; both p &lt; .001), before decreasing non-significantly to 476.8 in 2022 (APC = -7.8, p &gt; .05). Higher PORH risk was associated with younger age ( &lt; 65 vs. ≥ 65: aOR = 2.53, p &lt; .001), race and ethnicity (Native American: aOR = 2.07, Black: aOR = 1.85, White: aOR = 1.79, relative to Asian/Pacific Islanders; all p &lt; .001), public or no health insurance, lower income, co-occurring substance use disorders (SUD) (e.g., sedative aOR = 35.19, cocaine aOR = 3.57, both p &lt; .001), depression (aOR = 2.46, p &lt; .001), anxiety, chronic pain, metastases, and care in small, non-teaching, rural hospitals and the West region. Similarly, AORH were associated with younger age, race and ethnicity, insurance, low income, co-occurring SUD, mental health condition, chronic pain, metastases, comorbidity burden, and the West region. Conclusions: Opioid-related hospitalizations decreased substantially in recent years following heightened attention and clinical/policy efforts in the mid-2010s. Elevated risks were primarily associated with patient demographics, co-occurring substance use disorder, mental health conditions, and chronic pain.

Efficacy and safety of mevrometostat (M) in combination with enzalutamide (E) in patients (pts) with metastatic castration-resistant prostate cancer (mCRPC): Data from a phase 1 study.

e17043 Background: M is a potent and selective inhibitor of enhancer of zeste homolog 2. M 1250 mg twice daily (BID) on an empty stomach + E + androgen deprivation therapy showed improved outcomes vs E alone in pts with mCRPC, with a manageable adverse event (AE) profile, in the randomized, dose-expansion part of a phase 1 study (NCT03460977). We report efficacy and safety data from two pt cohorts (2A and 2C) who received M 875 mg BID with food + E. Methods: Pts with mCRPC who received prior abiraterone and/or E (2A) or prior abiraterone (2C), ≤1 prior chemotherapy in any setting, with evidence of progression per modified Prostate Cancer Working Group 3 criteria were included. Primary endpoints were radiographic progression-free survival (rPFS) per investigator assessment and safety. Secondary endpoints included objective response (OR) by RECIST 1.1 (for pts with measurable disease at baseline) and decline in prostate-specific antigen of ≥50% from baseline (PSA 50 ). Results: As of September 1, 2025, 29 pts had received M 875 mg BID with food + E (2A, n = 15; 2C, n = 14). Median (range) age was 74 (61–86) years in 2A and 73 (55–83) years in 2C. In 2A, 9 pts (60.0%) had received abiraterone and 9 pts (60%) had received E; in 2C, 14 pts (100%) received prior abiraterone and were E naive. Efficacy and safety outcomes for 2A and 2C are shown (Table). Six confirmed events (5 progressive disease,1 death) were observed in 2A and 4 (all progressive disease) in 2C; median (90% confidence interval [CI]) rPFS was 14.3 (2.0, not estimable [NE]) months in 2A and NE (6.2, NE) months in 2C. Confirmed PSA 50 was observed in 6 pts (40.0%; 95% CI 16.3, 67.7) in 2A and 6 pts (42.9%; 95% CI 17.7, 71.1) in 2C. In pts with measurable disease at baseline (2A, n = 4; 2C, n = 6), confirmed OR rate (95% CI) was 25.0% (0.6, 80.6; 1 partial response) in 2A and 16.7% (0.4, 64.1; 1 partial response) in 2C. For 2A+2C combined, most common treatment-emergent AEs (TEAEs) were diarrhea (48.3%), thrombocytopenia (48.3%), and decreased appetite (44.8%). Grade ≥3 TEAEs were observed in 41.3% pts (most common: thrombocytopenia, anemia, asthenic conditions, and hypokalemia). Grade ≥3 TEAEs considered related to M were reported in 31.0% pts. There were no treatment-related deaths. Conclusions: M 875 mg BID with food + E shows promising outcomes in pts with mCRPC and a manageable AE profile. Further investigation of M + E in pts with mCRPC is warranted. Clinical trial information: NCT03460977 . 2A(n=15) 2C(n=14) 2A + 2C(n=29) Efficacy Median rPFS (90% CI), months 14.3 (2.0, NE) NE (6.2, NE) – OR (95% CI), % 25.0 (0.6, 80.6) 16.7 (0.4, 64.1) – PSA 50 (95% CI), % 40.0 (16.3, 67.7) 42.9 (17.7, 71.1) – Safety, n (%) Any TEAE 15 (100) 14 (100) 29 (100) Grade ≥3 6 (40.0) 6 (42.9) 12 (41.4) TEAE related to M 15 (100) 13 (92.9) 28 (96.6) Grade ≥3 4 (26.7) 5 (35.7) 9 (31.0)

Patterns of tumor regression after neoadjuvant immunochemotherapy in oral squamous cell carcinoma and their surgical margin implications.

6098 Background: Neoadjuvant immunochemotherapy (nICT) has been increasingly applied in locally advanced oral squamous cell carcinoma (OSCC), demonstrating promising tumor downstaging and improved pathological response rates. However, the spatial patterns of tumor regression after nICT and their implications for surgical margin design and potential de-escalation of surgery remain poorly defined. Methods: We retrospectively analyzed 36 hemiglossectomy specimens from OSCC patients treated with nICT. All specimens were entirely embedded and evaluated using large-format histopathology. Tumor regression patterns were classified based on the distribution of residual viable tumor into central necrotic regression, peripheral regression, and patchy ablation types, each with distinct morphological subtypes. The relationship between pathological residual tumor foci, margin distance (1 mm vs 5 mm), and multicentricity was systematically assessed. Results: Distinct tumor regression patterns were observed following nICT. Most cases exhibited predominantly unicentric residual disease, while a small proportion demonstrated multicentric pathological residuals. Using a 1-mm pathological distance threshold, 1 cases were classified as multicentric, whereas no case met multicentric criteria using a 5-mm threshold. Notably, cases achieving major pathological response (MPR) or near pathological response (NPR) showed minimal residual tumor burden, though isolated satellite foci were occasionally detected. These findings suggest that regression morphology directly influences margin risk distribution and surgical safety. Conclusions: Tumor regression patterns after nICT in OSCC are heterogeneous and play a critical role in determining optimal surgical resection boundaries. While surgical de-escalation may be feasible in selected responders, careful assessment of residual tumor distribution is essential to ensure oncologic safety. Our findings support the development of an integrated preoperative regression assessment and postoperative margin reconstruction framework to guide individualized, function-preserving surgical strategies in OSCC.

Can TGF-β inhibition (galunisertib) enhance response to enzalutamide in metastatic castration-resistant prostate cancer (mCRPC)?: Results from a randomized phase II trial.

5056 Background: Resistance to androgen receptor (AR)-directed therapies is a major challenge in advanced prostate cancer, and enzalutamide shows limited activity after progression on abiraterone. Aberrant transforming growth factor-β (TGF-β) signaling promotes prostate cancer progression and contributes to resistance to AR-directed therapies. Preclinical studies suggest TGF-β inhibition may overcome resistance to enzalutamide. We hypothesized that combining TGF-β receptor I kinase inhibitor, galunisertib, with enzalutamide may delay enzalutamide resistance in patients with mCRPC who have progressed on abiraterone. Methods: This multicenter, randomized phase II trial enrolled chemotherapy-naïve mCRPC patients with disease progression on abiraterone. There was a cycle 1 safety run-in for the first 6 patients, followed by a randomized phase with patients assigned 3:2 to receive galunisertib plus enzalutamide or enzalutamide alone. The primary endpoint was radiographic progression-free survival (rPFS). Secondary endpoints included PSA response, overall survival (OS), safety, and tolerability. rPFS and OS were compared between arms using Cox regression, with medians estimated by Kaplan-Meier method. Results: Of 61 patients randomized, 60 received treatment: 23 with enzalutamide and 37 with galunisertib plus enzalutamide. There was no significant difference in rPFS between the combination and enzalutamide arms (HR 1.23; 95% CI: 0.72, 2.11; p =0.44), with a median rPFS of 5.6 and 7.3 months, respectively. The OS was also similar between the two groups (HR 1.00; 95% CI: 0.55, 1.83, p =1.00), with a median OS of 21.6 months in the combination arm and 24 months in the enzalutamide arm. PSA30 response was seen in 22% of patients in the combination arm vs. 35% in the enzalutamide arm ( p =0.30), and PSA50 response in 19% vs. 26%, respectively ( p =0.50). The most common treatment-related adverse events (AEs) were fatigue, anorexia, and nausea. Grade 3-4 treatment-related AEs occurred in 11% of patients in the combination arm vs. 13% in the enzalutamide arm. Conclusions: The addition of galunisertib to enzalutamide did not improve the rPFS, OS, or PSA response compared to enzalutamide alone in patients with chemotherapy-naïve mCRPC with disease progression on abiraterone. These findings do not support further development of galunisertib plus enzalutamide in patients with mCRPC. Clinical trial information: NCT02452008 . Baseline characteristics of randomized patients. N(n=61) Enzalutamide(n=24) Galunisertib + Enzalutamide (n=37) P Age, median (range) 51 72 (57, 84) 68 (53, 85) 0.06 Race, n (%) 58 1.00 White 17 (81.0) 30 (81.1) Other 4 (19.0) 7 (18.9) Ethnicity, n (%) 60 0.66 Hispanic 0 (0.0) 2 (5.4) Non-Hispanic 21 (91.3) 33 (89.2) Unknown 2 (8.7) 2 (5.4) ECOG, n (%) 55 0.16 0 12 (54.5) 24 (72.7) 1 10 (45.5) 9 (27.3) Baseline PSA, median (range) 60 11.1 (1.0, 2151.3) 18.2 (0.2, 522.4) 0.51

Chemoimmunotherapy with or without locoregional therapy for biliary tract cancer: A propensity score–matched study.

e16185 Background: , Recent phase II studies, including MISPHERE and ABC-07, suggest benefit from combining locoregional therapies such as selective radioembolization with yttrium-90(Y-90) microspheres and external beam radiotherapy with gemcitabine/cisplatin in selected patients with intrahepatic cholangiocarcinoma. However, the benefit of adding locoregional therapy to contemporary chemoimmunotherapy across biliary tract cancer (BTC) subtypes remains unknown. Importantly concurrent locoregional therapy was not permitted in the TOPAZ-1 or KEYNOTE-966. Methods: We conducted a multicenter retrospective cohort study of patients with unresectable or metastatic biliary tract cancer treated with first-line chemoimmunotherapy. Patients were categorized based on receipt of concurrent locoregional therapy at any point during the first-line therapy course including Y-90 radioembolization (Y90-SIRT), external beam radiotherapy (EBRT), transarterial chemoembolization (TACE), or ablation versus chemoimmunotherapy alone. Propensity score (PS) was calculated using logistic regression and PS- matching was performed with 1:1 nearest neighbor for age, gender, Charlson-morbidity index, ECOG status, BTC subtype and tumor burden. Progression free survival and overall survival were assessed using Kaplan–Meier analysis and compared using Cox proportional hazards models. Results: Among 294 patients with advanced BTC, 31 received concurrent locoregional therapy (Y90-SIRT, n = 14; RT, n = 14; TACE, n = 2; ablation, n = 1). Treatment predominantly consisted of gemcitabine-cisplatin and durvalumab (n = 286; 97%). Median follow-up was 18 months. Propensity score matching yielded 62 patients (31 per group), with well-balanced age, gender, Charlson-morbidity index, biliary tract subtype, tumor burden and performance status (standardized mean difference &lt; 0.1). Concurrent locoregional therapy was associated with improved progression-free survival (mPFS, 14.1 vs 7.8 months; HR, 0.45; 95% CI, 0.24-0.87; p = 0.017) and improved overall survival (mOS, 32.0 vs 14.8 months; HR, 0.38; 95% CI, 0.18-0.81; p =0.013) compared to chemoimmunotherapy alone. Results remained consistent in post-matching multivariable Cox models adjusting for biliary tract cancer subtype. Conclusions: In this multicenter propensity score–matched analysis, the addition of locoregional therapy to first-line chemoimmunotherapy was associated with improved progression-free and overall survival in advanced biliary tract cancer. These findings should be interpreted in light of the retrospective design, including the potential for treatment-selection bias, immortal time bias, and residual confounding. Prospective studies are warranted to define the role of concurrent locoregional therapy, with careful attention to patient selection, optimal modality, and timing of treatment intensification.

Efficacy of durvalumab and safety by treatment period in MATTERHORN: A randomized, phase 3 study of durvalumab plus 5-fluorouracil, leucovorin, oxaliplatin, and docetaxel (FLOT) chemotherapy in resectable gastric/gastroesophageal junction (G/GEJ) cancer.

4070 Background: In MATTERHORN (NCT04592913), perioperative durvalumab (D) + FLOT significantly improved event-free survival (EFS) and overall survival vs placebo (Pbo) + FLOT in participants (pts) with resectable G / GEJ cancer. Here, we report EFS by treatment period completion status. Methods: In this global, double-blind, Phase 3 study, pts with histologically confirmed, resectable, untreated G / GEJ adenocarcinoma were randomized 1:1 to D 1500 mg or Pbo every 4 weeks (Q4W) on Day 1 + FLOT every 2 weeks on Days 1 and 15 for 4 cycles (2 cycles each neoadjuvant and adjuvant [adjuvant: 4–12 weeks post-surgery]), followed by D 1500 mg or Pbo Q4W for 10 additional cycles (adjuvant monotherapy). EFS (time from randomization to progression, local or distant recurrence, or death) was assessed by treatment period completion status (Table). Safety was also assessed. Landmark EFS rates were calculated by the Kaplan–Meier method. Hazard ratios and their confidence intervals were estimated by the Cox proportional hazards model. Results: The number of pts who received treatment in each period was similar in the D + FLOT and Pbo + FLOT treatment arms (Table). EFS landmark rates at 12 and 24 months were highest in pts who completed all adjuvant D / Pbo in both arms; EFS was improved in the D + FLOT vs Pbo + FLOT arm irrespective of treatment periods started / completed (Table). A smaller proportion of Grade 3 – 4 adverse events (AEs) and serious AEs (SAEs) were reported in the adjuvant monotherapy period with D + FLOT and Pbo + FLOT (Grade 3 – 4: 21.2% and 20.5%; SAEs: 14.5% and 14.5%, respectively) when compared with the overall treatment period (Grade 3 – 4: 71.6% and 71.2%; SAEs: 48.2% and 44.1%, respectively). In the adjuvant monotherapy period, the rate of any AEs leading to discontinuation of D / Pbo was 5.8% in the D + FLOT vs 2.7% in the Pbo + FLOT arm. Conclusions: EFS landmark rates were greatest for pts who completed the full adjuvant regimen compared with those who discontinued prematurely in both arms. Moreover, the addition of D to FLOT improved EFS regardless of treatment periods started / completed. Safety was manageable and tolerable in the adjuvant monotherapy period. Clinical trial information: NCT04592913 . Received any neoadjuvant treatment but did not complete surgery (N=132) Received any neoadjuvant treatment and completed surgery (N=96)* Received any adjuvant treatment but did not complete D / Pbo (N=223) Completed all adjuvant D / Pbo (N=493) D + FLOT (n=62) Pbo + FLOT (n=70) D + FLOT (n=48) Pbo + FLOT (n=48) D + FLOT (n=116) Pbo + FLOT (n=107) D + FLOT (n=248) Pbo + FLOT (n=245) EFS landmark rate, % 12 mo 16.0 8.3 59.0 41.6 67.4 66.4 99.6 98.8 24 mo 10.0 6.2 49.2 33.3 46.7 29.8 91.8 87.2 EFS hazard ratio (95% CI) 0.80 (0.55–1.18) 0.65 (0.38–1.13) 0.65 (0.46–0.91) 0.63 (0.41–0.97) *These pts did not receive adjuvant treatment.

A subset analysis of clinical activity and pharmacodynamic biomarkers in patients with sarcomas in the phase 1 dose-escalation study of STC-15, a METTL3 inhibitor.

11548 Background: The RNA methyltransferase METTL3 is responsible for the generation of N6-methyladenosine (m6A), the most abundant modification on mRNA and long non-coding RNA. Accumulating evidence suggests numerous roles of METTL3 in cancer initiation and progression highlighting the rationale for targeting this enzyme in oncology. STC-15, a first in class small molecule inhibitor of METTL3 developed by Storm Therapeutics, demonstrated efficacy in pre-clinical models of cancer characterized by their dependence on m6A-mediated mRNA metabolism, including sarcomas that derive from a common mesenchymal progenitor. We report the clinical activity and pharmacodynamic (PD) biomarkers observed in a subset analysis of sarcomas from a first-in-human trial in patients with advanced malignancies. Full results of the study have been previously presented. Methods: This multi-center, open-label, dose escalation study of STC-15 administered oral capsules daily (QD) or three times a week (TIW) in 21-d treatment cycles. Dose escalation followed a 3+3 modified Fibonacci regimen. Primary objectives (safety and pharmacokinetics (PK)) and secondary objectives (efficacy) were previously presented. A subset analysis of sarcoma and selected patients evaluates anticancer activity and PD biomarkers, including target engagement, gene expression analysis, and targeted m6A RNA sequencing using the EpiPlex platform. Results: Of 42 patients enrolled, 35 were evaluable and 13 were sarcoma patients with a mean of 3 prior lines of therapy. Of 13 sarcoma patients with at least 1 on-treatment scan, DCR was 54% at 12 weeks with 1 confirmed PR (8% ORR) in angiosarcoma (DOR 22 months), and mPFS of 5.5 months as of 27 January 2026. Seven sarcoma patients presenting with four subtypes experienced a best response of SD, including tumor regressions. Treatment duration ranged from 15-828 days, with two sarcoma patients remaining on treatment for 590 and 828 days, respectively. An average of &gt;50% reduction in m6A on mRNA in peripheral blood within the first 24h post dosing was observed in dose-escalating cohorts from 60 mg QD to 200 mg TIW, confirming target engagement. Moreover, we present an expanded analysis using the m6A-directed NGS EpiScout platform, where we evaluate both the effects of STC-15 on global m6A mRNA abundance and transcriptional changes that may contribute to an anti-tumor response specific to soft tissue and bone sarcomas. Conclusions: Treatment with STC-15 offers a new paradigm for the treatment of a broad range of sarcomas by halting the cell differentiation process from mesenchymal progenitors. Early biomarker data provide proof of mechanism in target engagement and changes in gene expression, consistent with pharmacological activity and clinical response. Further investigation of efficacy and biomarkers of STC-15 in a Phase 2 sarcoma study is ongoing. Clinical trial information: NCT05584111 .

Trends in disability-adjusted life years from hepatitis C–related liver cancer in the United States: A Global Burden of Disease study.

e16194 Background: Hepatitis C virus (HCV)–related mortality has declined in the United States, largely due to the widespread use of highly effective direct-acting antiviral (DAA) therapies. Despite this progress, the burden of HCV-related liver cancer appears to be rising. We examined long-term trends in disability-adjusted life years (DALYs) attributable to HCV-related liver cancer in the United States. Methods: Data were obtained from the Global Burden of Disease (GBD) 2021 study. DALYs attributable to hepatitis C–related liver cancer were extracted for the United States and analyzed from 1990 to 2021. Temporal trends were assessed using linear regression and the Mann–Kendall trend test. Results: DALYs from HCV-related liver cancer increased steadily throughout the study period. Linear regression demonstrated an average annual increase of 1,646 DALYs per 100,000 population (β = 1,646; p &lt; 0.001; R² = 0.99). The Mann–Kendall trend test confirmed a significant upward monotonic trend (Z = 7.9; p &lt; 0.001), consistent with the regression findings. Conclusions: Despite substantial reductions in HCV-related mortality following the introduction of DAAs, the burden of HCV-related liver cancer in the United States continues to rise. This pattern likely reflects the long latency of hepatocarcinogenesis in chronic HCV infection, particularly among individuals born between 1945 and 1965, who account for the majority of chronic infections. These findings highlight the ongoing need for targeted HCV screening, early treatment, and long-term surveillance in high-risk populations to mitigate future liver cancer burden.

Phase II study of adjuvant nab-paclitaxel plus S-1 after D2 resection in stage III diffuse gastric cancer (NORDICA study).

4085 Background: Stage III diffuse-type gastric cancer (DGC) is highly aggressive, with a substantial risk of postoperative recurrence. Adjuvant chemotherapy with nab-paclitaxel plus S-1 (AS regimen) has shown potential efficacy in early studies, but its safety and effectiveness in this high-risk population remain unclear. Methods: This single-center, prospective phase II trial enrolled patients with stage III DGC who underwent D2 gastrectomy between January 2020 and October 2023. All patients first received one cycle of S-1 monotherapy as a pre-screening tolerance test. Eligible patients then received adjuvant AS therapy: nab-paclitaxel 260 mg/m² on day 1 every 3 weeks (maximum 400 mg) plus S-1 80–120 mg/day on days 1–14 every 3 weeks for 6 cycles, followed by 8 cycles of S-1 monotherapy. The primary endpoint was 1-year disease-free survival (DFS); secondary endpoints included 3-year DFS, overall survival (OS), and safety. To explore potential molecular correlates of treatment response, whole-exome sequencing was performed on tumor specimens obtained from surgical resection in 24 patients, including 12 patients who experienced recurrence within 2 years (resistant group) and 12 patients with DFS of 2 years or longer (sensitive group). Results: The 1-year DFS and OS rates were 84.9% (95% CI, 75.3–95.9%) and 100%, respectively. The 3-year DFS and OS rates were 45.2% (95% CI, 32.5–63.0%) and 62.5% (95% CI, 49.0–79.8%), with a median DFS of 27.7 months (95% CI, 16.5–NR). The most common ≥grade 3 treatment-related adverse events were neutropenia (29.2%), leukopenia (20.8%), rash (6.3%), febrile neutropenia (6.3%), and oral mucositis (4.2%). CDH1 and OBSCN mutations were more frequent in the resistant group, suggesting a potential link to early recurrence. Conclusions: Adjuvant AS therapy is feasible and demonstrates promising efficacy for stage III DGC, with an acceptable safety profile. Molecular findings may help identify patients at higher risk of recurrence, supporting further evaluation in larger, randomized phase III trials. Clinical trial information: NCT03977220 .