Our phase Ib neoadjuvant trial demonstrated that oncolytic virus (OV) combined with PD-1 blockade achieved a 77.8

9516 Background: Mucosal melanoma (MM) is a rare and aggressive cancer with a poor prognosis. Unlike its more common counterpart, cutaneous melanoma, MM responds poorly to standard single-agent immunotherapy, such as anti-PD-1 monoclonal antibodies (mAbs), limiting treatment options. This clinical challenge underscores the need for more effective strategies, making combination therapies a clear future direction. The aim is to overcome resistance by targeting multiple pathways simultaneously. The main objective of this study is to explore and compare two distinct immunotherapy backbones—targeting the PD-1/CTLA-4 and VEGF pathways—for their potential in future rational combinations against MM. Methods: This is a phase Ib, dose de-escalation and cohort-expansion, open-label trial. The combination dose of axitinib will utilize a "3+3" dose de-escalation design. The starting dose is 5 mg. If dose-limiting toxicity (DLT) is observed at this dose level, the dose for subsequent cohorts will be reduced to 3 mg. Patients with metastatic melanoma receive 10 mg/kg cadonilimab Q3W (cohort 1) or 20 mg/kg ivonescimab Q3W (cohort 2) in combination with 5 mg axitinib BID. Primary endpoints are rate of dose-limiting toxicities (DLTs) and safety. Secondary endpoints are objective response rate (ORR) by investigator assessment, duration of response (DOR), progression-free survival (PFS), overall survival (OS) and overall safety. Results: As of December 25, 2025, a total of 28 patients have been enrolled in the study (14 cohort 1; 14 cohort 2). Median follow-up of 13 months. The distribution of primary tumor sites was as follows: oral cavity 14.3%, esophagus17.9%, rectum 32.1%, and cervix 35.7%. 21% had elevated LDH. BRAF status, known for 71% of patients, was positive in 25%. No DLT or treatment related death was observed，Axitinib was selected as 5 mg BID. 71.4% patients experienced treatment related AE (TRAE) and 17.9% patients experienced Grade 3-4 TRAEs. In cohort 1，the most common TRAEs include hypothyroidism, hypertension, diarrhea, anorexia and rash. In cohort 2, the most common TRAEs include proteinuria, myalgia, hyperthyroidism, hypertension, abnormal transaminase levels, anorexia and hyperlipidemia. No patient discontinued treatment due to TRAE. ORR, DOR, DCR, and median PFS for all cohorts are shown in Table. Conclusions: The combination therapy involving cadonilimab or ivonescimab plus axitinib demonstrates a promising safety and efficacy profile. These findings suggest that such dual-antibody-based regimens could provide synergistic anti-tumor activity with an improved therapeutic index, warranting further clinical investigation. Clinical trial information: NCT06424626 . Secondary Endpoint Cohort 1 （n=14） Cohort 2 （n=14） PR 5 5 SD 4 5 PD 5 4 ORR 35.7% 35.7% DCR 64.3% 71.4% DOR (mo) 6.7 6.4 mPFS (mo) 5.8 8.7 mOS (mo) NA NA

Accurate sentinel lymph node (SLN) identification is essential for the staging of cutaneous melanoma. Contrast-enhanced ultrasound using Sonazoid (Sonazoid-CEUS), which is based on a perfluorobutane microbubble contrast agent with reticuloendothelial uptake properties, was evaluated for SLN detection in patients with stage Ib–II melanoma. In this prospective single-arm cohort study, 71 eligible subjects underwent perilesional intradermal Sonazoid-CEUS, followed by lymphoscintigraphy and SLN biopsy. The numbers of SLNs detected via CEUS and lymphoscintigraphy were compared by using non-parametric paired analysis (Wilcoxon signed-rank test). The diagnostic accuracy of CEUS for metastatic involvement was evaluated against histopathological results. Sonazoid-CEUS achieved a 97.3

4544 Background: Hypoxia inducible factor-2 alpha (HIF-2α) is a transcription factor and key tumorigenic driver of clear cell renal cell carcinoma (ccRCC). JMKX003948 is a novel, orally administered, selective small-molecule HIF-2α inhibitor that has shown significant antitumor activity in preclinical models of ccRCC. Herein, we present preliminary results from a phase 1 dose-escalation study of JMKX003948 in advanced ccRCC (NCT06321250). Methods: In this study, Patients with advanced ccRCC who had received at least 1 prior standard therapy were enrolled in 40, 80, 120, 180, and 270 mg escalating-dose cohorts sequentially, and received JMKX003948 monotherapy once daily (QD) until progression or unacceptable toxicity. The primary endpoints included safety/tolerability, the maximum tolerated dose (MTD), and the recommended Phase 2 dose (RP2D). Results: As of Dec 23, 2025, 22 patients (17 male) were enrolled. Median age was 57.5 years (range, 32-73 years). 13 patients (59.1%) have received ≥ 2 prior lines of therapy, 20 patients (90.9%) have received immune checkpoint inhibitor, and all patients (100%) have received vascular endothelial growth factor receptor–tyrosine kinase inhibitor (VEGFR-TKI). At data cutoff, the median follow-up was 9.7 months. One dose-limiting toxicity (grade 3 transaminase elevation) was observed at the 120 mg QD dose level, and the MTD was not reached. Treatment-related adverse events (TRAEs) were reported in all patients, with most graded 1-2. The most common TRAEs were anemia (86.4%). Grade 3 TRAEs occurred in 6 patients (27.3%), including hypoxia (13.6%), anemia (9.1%), proteinuria (4.5%), transaminase elevation (4.5%), and hyperkalemia (4.5%). No grade 4 TRAEs occurred. No TRAEs led to treatment discontinuation or death. Of 20 efficacy-evaluable patients, 5 patients (2 patients in 270 mg cohort and 1 patient each in 80, 120, 180 mg cohort) had partial response (PR), and 9 patients had stable disease (SD), objective response rate (ORR) and disease control rate (DCR) per RECIST 1.1 were 25% and 70%, respectively. Mean time to response (TTR) was 3.2 months. Median progression-free survival (PFS) was 7.7 months (95% CI, 1.3-not available). Conclusions: JMKX003948 is well tolerated and shows promising antitumor activity in patients with advanced ccRCC. Clinical trial information: NCT06321250 .

e21536 Background: High-dose interferon-α (HD-IFN) has historically served as a standard adjuvant therapy for resected stage IIB–IIID melanoma, yet relapse remains common. While dendritic cell-cytokine induced killer (DC-CIK) cell therapy has demonstrated antitumor activity across various malignancies, its specific efficacy in the adjuvant setting for melanoma remains unclear. We compared outcomes of DC-CIK plus HD-IFN versus HD-IFN alone in a real-world cohort. Methods: We retrospectively analyzed 903 patients with completely resected stage IIB–IIID melanoma between 2012 and 2016. Patients received either HD-IFN alone (20 MIU for 4 weeks, followed by 10 MIU for 11 months) or HD-IFN plus DC-CIK immunotherapy (at least 3 cycles). For DC-CIK preparation, an HLA-guided antigen loading strategy was employed. Autologous DCs were pulsed with peptides (MART-1/S-100) for HLA-A02/A24+ patients or tumor lysates for others and subsequently co-cultured with CIK cells to generate the effector product. Propensity score matching (1:1) was performed. Endpoints were recurrence-free survival (RFS), overall survival (OS), and toxicities. Results: After propensity score matching, a total of 586 patients were included in the final analysis, with 293 patients in each group. Baseline characteristics were well balanced between the two groups. The cohort comprised 209 patients with stage IIB-IIC disease (35.7%) and 377 patients with stage III disease (64.3%). With a median follow-up of 101.9 months, combination therapy significantly improved RFS compared with HD-IFN alone (median, 23.9 vs. 13.4 months; Hazard Ratio [HR], 0.67; 95% CI, 0.56–0.80; P<0.05). Although no significant difference in OS was observed in the overall population (median, 63.5 vs. 48.2 months; HR, 0.89; 95% CI, 0.72–1.10; P = 0.279), subgroup analysis revealed a significant OS benefit for stage III patients receiving combination therapy (median, 51.5 vs. 35.4 months; P = 0.033). Similarly, the combination group demonstrated superior median RFS in stage III disease (18.7 vs. 9.1 months; P < 0.01) and across specific subtypes, including acral (23.6 vs. 13.8 months; P < 0.01) and cutaneous melanoma (21.4 vs. 12.9 months; P < 0.05). Grade ≥3 adverse events occurred in 15.0% of the combination group and 12.5% of the monotherapy group, with no treatment-related deaths. Conclusions: DC-CIK combined with HD-IFN regimen was associated with significantly prolonged RFS compared with HD-IFN alone in resected stage IIB–IIID melanoma. Notably, this combination therapy demonstrated a significant OS benefit in patients with stage III disease, highlighting a high-risk subgroup that may derive the greatest survival advantage from this strategy. Prospective studies are warranted to confirm these findings and better define the role of DC-CIK–based adjuvant strategies.

2527 Background: QLF31907, a bispecific antibody that simultaneously block PD-1/L1 immunosuppressive pathway on cancer cells and conditionally activate 4-1BB co-stimulatory pathway on tumor-specific T cells, was designed to restrict 4-1BB agonism to the tumor microenvironment, which might overcome resistance to PD-(L)1 inhibitor and reduce hepatoxicity as traditional 4-1BB monoclonal antibodies reported. Although immunotherapy (IO) has revolutionized the treatment of melanoma, a significant proportion of patients (pts), particularly those with mucosal and acral subtypes, either fail to respond initially or experience disease relapse after treatment. Here, we present results of QLF31907 in previously-treated pts with advanced melanoma, including IO-exposed. Methods: This phase 2 trial was comprised of safety observation stage and efficacy expansion stage. Pts with unresectable locally advanced or metastatic melanoma who failed, were intolerable to, or refused standard treatment were recruited and administered QLF31907 via intravenous infusion from 5 mg/kg to 20 mg/kg every 2 weeks (Q2W) or 3 weeks (Q3W). The primary endpoints were dose-limiting toxicity (DLT) and safety in safety observation stage, and was objective response rate (ORR) per RECIST v1.1 assessed by investigator in efficacy expansion stage. Results: As of Dec 31, 2025, 59 pts were enrolled (median age: 57.0 years; male: 47.5%; ECOG PS of 1: 42.4%; stage IV: 93.2%). The mucosal subtype accounted for the most (40.7%), followed by acral (33.9%), cutaneous (non-acral; 18.6%) and primary unknown (6.8%). Median prior lines of therapies were 2.0 (range, 1–5). Fifty-five (93.2%) pts received prior immunotherapy, including 50 (87.7%) pts received prior anti-PD-1/PD-L1 agents. No DLT occurred. Grade ≥3 treatment-emergent adverse events (TEAEs) occurred in 34 (57.6%) pts. The most common grade ≥3 TEAEs (≥10%) were liver injury (18.6%), neutrophil count decreased (15.3%), anemia (13.6%), white blood cell count decreased (11.9%). In 57 efficacy-evaluable pts, seven had partial response. The ORR and disease control rate (DCR) was 12.3% (95% confidence interval [CI], 5.1%-23.7%) and 56.1% (95% CI, 42.4%-69.3%), respectively. The median progression-free survival, duration of response, and overall survival was 2.6 months (95% CI, 1.6-3.7), 5.8 months (95% CI, 2.3-not evaluable [NE]), and 15.3 months (95% CI, 11.6-NE), respectively. Conclusions: QLF31907 showed potential anti-tumor activity and acceptable safety profile in heavily-treated pts with advanced melanoma, including IO-exposed pts. These results warrant validation in further clinical trials. Clinical trial information: NCT05823246 .

Abstract Background: Tumor-infiltrating lymphocytes (TIL) adoptive therapy has shown clinical efficacy for metastatic melanoma refractory to anti-PD-1 therapy. This phase I clinical trial evaluated the safety and efficacy of TILs (LM103) in advanced Chinese patients (pts) with melanoma, and explored correlative biomarkers. Methods: Pts with metastatic melanoma progressing on standard therapies, had both resectable and measurable tumors were eligible to be enrolled. Pts received a lymphodepletion regimen before the intravenous autologous LM103 infusion and then high dose IL-2 for 6 doses (200000IU/Kg, 1 dose per day) to support T cell survival and proliferation. Primary endpoint was safety; secondary endpoints included efficacy and translational analysis. Results: Fourteen pts (4 males; aged 26-70 yrs) were enrolled: 5 acral, 4 cutaneous, 4 mucosal, and 1 unknown primary. Infused autologous TIL contained 3.58-19.47x10^10 viable cells. The most frequent treatment-emergent adverse events (TEAEs) were myelosuppression (100%), fever (85.7%), rash (71.4%), and alopecia (71.4%). Grade 3-4 TEAEs included leukopenia (85.7%), neutropenia (78.6%), lymphopenia (78.6%), fever (71.4%), thrombocytopenia (57.1%), and anemia (28.6%). The ORR and DCR per RECIST v1.1 were 35.7% and 85.7%, respectively. With a median follow-up of 5.51 months, the median PFS and OS were 4.52 months (95% CI, 2.07 to 14.31) and 16.58 months (95% CI, 5.16 to NA), respectively. Immunofluorescence analysis and whole-exome sequencing for the baseline tumors did not reveal any significant difference in the infiltration of multiple immune cell populations and tumor mutation burden, respectively, between responders (Rs) and non-responders (NRs). T cell receptor-sequencing (TCR-seq), flow cytometry analysis and ELISA were performed for the TIL infusion products and the peripheral blood of patients at various time points before and after infusion. The frequency of shared T cell clones between TIL infusion products and post-infusion peripheral blood maintained >60% till 6 month after infusion, indicating that LM103 TILs persisted well in vivo. The frequencies of dominant T cell clones of infused TILs were higher in Rs than in NRs, implying a correlation between TILs persistence and clinical outcome. In addition, the IFN-γ level in the peripheral blood at Week 6 of NRs was significantly higher than that of Rs while other cytokine (such as TNF-α, IL-2, TGF-β1 etc) levels were similar between these two groups. Percentage of CD3+, CD4+ or CD8+ T cells in the peripheral blood was also not significantly different between R and NRs. Conclusion: LM103 TIL therapy demonstrated manageable safety and encouraging efficacy in treatment-refractory melanoma. Clinical response correlated with the persistence and expansion of infused TIL clones, instead of baseline tumour features. Trial registration number: CTR20233999. Citation Format: Chuanliang Cui, Li Zhou, Yan Kong, Xuan Wang, Yue Yang, Jiayong Liu, Zhichao Tan, Zhihong Chi, Lu Si, Bin Lian, Lili Mao, Hui Tian, Yuan Yuan, Yongming Xue, Jun Guo. Updated clinical efficacy and translational biomarker analysis of tumor infiltrating lymphocyte therapy (LM103) in metastatic melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(8_Suppl):Abstract nr LB414.

Mucosal melanoma is an aggressive malignancy with limited neoadjuvant options. We conducted a single arm, phase II study of neoadjuvant pembrolizumab plus lenvatinib followed by surgery and adjuvant pembrolizumab in resectable mucosal melanoma (NCT04622566) along with exploratory biomarker analysis. Primary objective was pathological complete response (pCR) rate; secondary endpoints included relapse-free survival (RFS), overall survival (OS), clinical response, surgical outcomes, and safety. Among 21 surgical patients, the pCR rate was 9.5%, major pathologic response (MPR) rate was 19.0%, and the pathologic response rate was 38.1%. Median RFS was 14.8 months (1-year rate: 61.9%); median OS was not reached. No grade 4-5 treatment-related toxicities or additional perioperative complications occcurred. Spatial profiling revealed a more immune-inflamed baseline microenvironment in responders, with activated CD4⁺/CD8⁺ T cell signatures associated with favorable outcomes. Treatment induced vascular normalization and increased T cell infiltration in non-responders, partially narrowing the immune gap. Responders exhibited higher prevalence of persistent TCR clonotypes and tighter spatial proximity between activated CD4⁺ and CD8⁺ T cells. Although the pre-specified primary endpoint was not met, our findings identify activated CD4+/CD8+ T cell states and TCR persistence as key outcome-associated features, supporting immune-informed optimization of peri-operative therapy in mucosal melanoma.

Cancer gene vaccines combined with immune checkpoint inhibitors hold promise for melanoma therapy, but delivery vehicles remain critical. Here, we report LA-PegPI, a galactose-modified polymeric vector that delivers plasmid DNA specifically to dendritic cells (DCs), achieving higher transfection efficiency than commercial reagents with low cytotoxicity. LA-PegPI/pDNA complexes activate TLR4/NF-κB and NLRP3/caspase-1 pathways, inducing DC maturation and secretion of IL-6, TNF-α, IL-1β, and IL-12. In B16-OVA melanoma models, LA-PegPI/pOVA vaccination synergizes with anti-PD-1 to suppress tumor growth, reduce lung metastasis, and prolong survival, accompanied by increased CD8+ T, NK cells, and Th1 cytokines. This polymer platform integrates targeted delivery and intrinsic adjuvanticity, offering a safe strategy to enhance checkpoint immunotherapy.

INTRODUCTION:Melanomas cause the vast majority of skin-cancer related deaths, despite accounting for only around 2% of skin cancer cases. Moreover, due to genetic and environmental factors, the melanomas that predominate in East Asian populations, generally acral and mucosal melanomas, are extremely rare in Caucasian populations. These melanomas differ from other cutaneous melanomas in clinicopathological features and mutational landscapes, including a lower prevalence of commonly targetable mutations such as BRAF V600, and are usually diagnosed at a more advanced stage with a poor prognosis. The advent of immunotherapies therefore represents a great stride forward in the treatment of melanoma, including in East Asian countries, with immune checkpoint blockade now available as first-line treatment in China. AREAS COVERED:This review briefly outlines the epidemiology and unique clinical characteristics of melanoma in East Asian patients; summarizes existing immunotherapy approvals, treatment landscape, and unmet needs; and aims to enable clinicians to make more informed treatment decisions for patients. EXPERT OPINION:Further research is urgently needed into novel immunotherapeutic combinations, and biomarkers of response, to fully unlock the potential of immunotherapy for East Asian AM and MM and guide future therapeutic strategies.

Combination treatment with anti-angiogenic agents and immune checkpoint inhibitors has demonstrated significant efficacy in advanced renal cell carcinoma (RCC). However, data on treatment-emergent renal injury during combination therapy and its prognostic implications remain limited. This study analyzed renal injury and its association with survival outcomes in patients with advanced RCC receiving first-line benmelstobart combined with anlotinib in the ETER100 trial. All randomized patients in the ETER100 (NCT04523272) trial treated with at least one dose of study treatment were included in this analysis. Baseline characteristics, laboratory test results, and information on objective responses, disease progression, and death events were collected. The primary outcomes of this study included renal function changes during therapy, risk factors for renal function abnormalities, clinical outcomes of patients with renal function abnormalities, and the association between renal function abnormalities and survival outcomes. Compared with sunitinib monotherapy, patients receiving benmelstobart combined with anlotinib showed no significant differences in the incidence of serum creatinine elevation (p = 0.0538), percentage decrease in creatinine clearance (p = 0.2546), absolute decrease in creatinine clearance (p = 0.7343), or proteinuria (p = 0.0728). Multivariate analysis identified a history of nephrectomy and estimated glomerular filtration rate <90 mL/min/1.73 m² as independent predictors of serum creatinine elevation in patients receiving combination therapy. As of January 2024, 64.00

9571 Background: SWOG1801 and NADINA trials suggested that neoadjuvant therapy may provide greater benefits than adjuvant therapy in melanoma. CAP 03-NEO explores the efficacy and safety of camrelizumab combined with apatinib and temozolomide triple-drug regimen as neoadjuvant therapy in patients (pts) with resectable stage II/III acral melanoma (AM). Methods: This two-stage trial (NCT05512481) aimed to enroll 60 pts with resectable stage II/III AM. Pts received two 4-week cycles of neoadjuvant camrelizumab (200 mg IV Q2W), apatinib (250 mg orally QD), and temozolomide (200 mg/m² IV daily D1–5), followed by surgery and 15 cycles of adjuvant camrelizumab (200 mg Q3W). A total of 30 pts were assessed in stage 1, demonstrating stage III AM benefiting more from neoadjuvant therapy (ASCO 2025, abstract 9512). Therefore, stage 2 exclusively enrolled stage III pts. Endpoints included pathological response, EFS, RFS and DMFS. To further evaluate the neoadjuvant benefit, an external contemporary cohort of stage III AM pts treated with upfront surgery and adjuvant PD-1 inhibitor alone was retrospectively collected for comparison. Propensity score matching (PSM 1:1 nearest neighbor method) and inverse probability of treatment weighting (IPTW) were used to minimize bias. Kaplan-Meier method and Cox regression were performed for survival analysis. Stage 2 results and this post-hoc comparison are reported here. Results: As of Jan. 26, 2026, thirty-one pts in stage 2 were enrolled, with a median follow-up of 13 months. The median age was 60 years (IQR: 49–68). Of 28 pts undergoing surgery, 18 (64.3%) achieved any pathological response, including 10 with major pathological response (7 with pCR, 3 with near pCR), and 8 with pPR. Surgery was canceled for 2 pts due to progressive disease. Neoadjuvant therapy was still ongoing for one patient. The 12-month rates of EFS, RFS and DMFS were 74.4%, 80.3% and 84.0%, respectively. All median values were not reached. As compared to external control cohort (adjuvant treatment alone), both unadjusted and adjusted analyses showed that pts treated with additional neoadjuvant therapy achieved significantly better EFS, RFS and DMFS (Table 1, all p values <0.05). The safety profile between stage 1 and stage 2 were similar and no new safety signal was identified. Conclusions: Stage 2 results and external control comparison of CAP 03-NEO further confirmed the significant benefit of neoadjuvant camrelizumab, apatinib and temozolomide in pts with resectable stage III AM. Clinical trial information: NCT05512481 . HR and 95% CI for EFS, RFS, and DMFS (sample size of stage 2 vs. external control). Survival Unadjusted (31 vs. 50) PSM 1:1 (30 vs. 30) IPTW (31 vs. 50) EFS 0.309 (0.129-0.743) 0.246 (0.078-0.778) 0.317 (0.124-0.810) RFS 0.244 (0.086-0.694) 0.241 (0.071-0.812) 0.242 (0.081-0.728) DMFS 0.286 (0.099-0.822) 0.203 (0.051-0.807) 0.290 (0.094-0.893)

As an emerging biomarker, tumor mutational burden (TMB) has attracted increasing attention from clinicians in predicting the efficacy of tumor immunotherapy. Currently, TMB is detected primarily by whole-exome sequencing or targeted panel sequencing on high-throughput sequencing platforms. However, the lack of uniformity in detection methods, threshold settings, and reporting formats, as well as the significant differences in TMB values among different cancer types, have hindered the standardized application of this biomarker in clinical practice. This consensus focuses on the definition, standardization of detection, clinical significance, and limitations of TMB, and provides consensus recommendations for the clinical application of TMB in real-world practice in China. This consensus is aimed at helping clinicians and laboratory personnel understand the clinical significance and testing standards of TMB, promoting more accurate interpretation of test results, and improving patient care.

9537 Background: Homologous recombination deficiency (HRD) describes a tumor cell state characterized by impaired DNA double-strand break repair via the homologous recombination repair (HRR) pathway. Deficiencies in this pathway may serve as predictive biomarkers. However, the relevance of HRR gene mutations remains inadequately characterized in non-cutaneous melanoma. In this study, we aimed to delineate the mutational landscape of HRR genes in this population and evaluate their predictive implications for clinical management. Methods: We conducted a retrospective analysis of 932 patients with non-cutaneous melanoma treated at Beijing Cancer Hospital from January 2008 to January 2026. Tumor genomic profiling was performed using targeted next-generation sequencing with either an 81- (n=852) or a 425-gene (n=80) panel. Associations between HRR mutation status and survival outcomes were assessed using Kaplan-Meier survival estimates and Cox proportional hazards regression models. Results: HRR mutations were identified in 92 patients (9.9%), with the most frequent alterations occurring in ATM (3.5%), ARID1A (3.4%), BRCA2 (2.7%), and BRCA1 (1.8%). Within this HRR-mutated cohort, the primary subtypes were mucosal (44.6%) and acral (29.3%) melanoma. And the disease stages were distributed as follows: stage I, 31 patients (33.7%); stage II, 15 (16.3%); stage III, 22 (23.9%) and stage IV, 24 (25.0%). After a median follow-up of 63.7 months, the median overall survival (OS) was 57.5 months (95% CI: 43.4–71.7), with 1-, 2-, 3-, and 5-year OS rates of 92.1%, 84.9%, 74.7%, and 56.8%, respectively. Multivariate analysis confirmed age ≥65 years as an independent predictor of poorer overall survival (HR 2.7, 95% CI 1.3–5.4, P = 0.005), whereas ATM mutation emerged as a significant protective factor associated with a 68% reduction in mortality risk (HR 0.32, 95% CI 0.1–0.7, P = 0.007). Among 59 HRR-mutated patients who received first-line systemic therapy, 41 were treated with anti-PD-1-based regimens and 18 with non-PD-1-based regimens. In the 54 evaluable patients, median progression-free survival (PFS) was 9.0 months (95% CI: 3.8–14.2) and median OS was 55.0 months (95% CI: 39.0–71.0). The objective response rate (ORR) was 11.1%, and the disease control rate (DCR) was 61.1%. The DCR was 77.8% (21/27) in the mucosal subgroup versus 37.5% (6/16) in the acral subgroup with a statistically significant difference (P = 0.028). In a subgroup analysis, although not statistically significant, anti-PD-1-treated patients trended toward a longer median PFS compared with those not receiving anti-PD-1 therapy (15.0 vs. 6.0 months, P = 0.249). Conclusions: HRR mutations define a distinct molecular subset in non-cutaneous melanoma. ATM mutation serves as an important prognostic biomarker, and patients with HRR-mutated melanoma may derive survival benefit from anti-PD-1 immunotherapy.

Background: Programmed cell death protein (PD-1) inhibitors constitute the standard adjuvant therapy for cutaneous melanoma (CM), but well-established strategies for acral melanoma (AM) remain limited. Objective: To compare the efficacy of adjuvant anti-PD-1 immunotherapy versus high-dose interferon α-2b (HDI) in patients with stage IIB-IV AM and CM. Design: This multicenter, retrospective study enrolled 511 patients with resected stage IIB-IV AM and CM between January 2017 and December 2023. Methods: Patients were divided into four groups by subtype and treatment: patients with CM and treated with anti-PD-1 (CM-PD-1), patients with CM and treated with HDI (CM-HDI), patients with AM and treated with anti-PD-1 (AM-PD-1), and patients with AM and treated with HDI (AM-HDI). Recurrence-free survival (RFS), overall survival (OS), and patient safety were evaluated. Results: This study comprised 362 AM and 149 CM cases. AM cases presented with thicker primary lesions, higher ulceration rates, and fewer BRAF mutations. Median follow-up was 49 months. Among patients with stage IIB/C, median RFS was not reached in any groups. Among stage III/IV patients, median RFS was 14.6 (CM-PD-1), 13.7 (CM-HDI), 13.3 (AM-PD-1), and 11.7 months (AM-HDI), and median OS was 61.6, 40.7, 42.4, and 53.4 months, respectively, without significant intergroup differences. Anti-PD-1 significantly improved RFS in patients with stage III/IV AM with KIT mutations (9.1 vs 5.0 months, p = 0.048) and ⩾4 lymph node metastases (10.5 vs 6.6 months, p = 0.036). Anti-PD-1 had significantly fewer adverse effects than HDI (60.4% vs 88.6%, p < 0.001), including fewer grades 3–4 events (4.6% vs 30.7%, p < 0.001). Conclusion: Adjuvant anti-PD-1 therapy provides RFS comparable to that of HDI in AM and CM, with a superior safety profile. Patients with AM harboring KIT mutations derive greater benefits from adjuvant anti-PD-1 therapy.

LBA9509 Background: Tumor-infiltrating lymphocyte (TIL) therapy is effective in advanced melanoma but hampered by toxicity from high-dose IL-2 and lymphodepletion (cyclophosphamide/fludarabine). We explored a modified TIL therapy (GC101) featuring an IL-2-free protocol and a low-intensity preconditioning regimen. This phase II study evaluates the modified GC101 TIL therapy in anti-PD-1-resistant melanoma. Methods: MIZAR-003 (NCT06703398) is a multicenter, phase II, randomized, open-label trial across 25 sites in China enrolling advanced melanoma patients resistant to PD-1 antibodies. Key eligibility: ECOG 0–1, adequate organ function. Exclusion: uveal melanoma, prior cellular therapy within 6 months. A total of 98 patients will be randomized 1:1 to Arm A (GC101 TIL) or Arm B (investigator-choice chemotherapy). GC101 therapy consists of tumor resection, low-intensity preconditioning (cyclophosphamide 20 mg/kg/d, pre-infusion d -5 to- 3; hydroxychloroquine 600 mg, pre-infusion d -5), followed by infusion of IL-2-free TIL and sintilimab (100 mg at infusion, then 100 mg Q6W × 4). Patients in Arm B with IRC-confirmed disease progression may cross over to receive GC101 TIL. The primary endpoint is IRC-assessed PFS per RECIST v1.1. Secondary endpoints include IRC-assessed ORR, CR rate, DOR, and OS; investigator-assessed ORR, PFS, CR rate, DOR, and PFS2; and safety. Results: In this exploratory interim analysis (70% of expected events), 84 patients were randomized (Arm A: 44; Arm B: 40) between 02/2025 and 01/2026, with a median follow-up of 5.2M (range 1.1-12.0). Baseline characteristics were balanced. As assessed by IRC, Arm A significantly improved mPFS (4.1M [95% CI 2.8–5.6] vs 1.6M [95% CI 1.3-4.0]; HR 0.53 [95% CI 0.30-0.94], p=0.0310), ORR (45.2% vs 5.4%; p=0.0018), and DCR (80.6% vs 43.2%; p=0.0017). OS data are immature. In mucosal melanoma (n=6 per arm), mPFS was not reached vs 1.3M, with DCR of 100.0% vs 33.3%. Grade ≥3 treatment-related AEs (TRAEs) occurred in 36.8% (Arm A) vs 27.5% (Arm B); treatment-related SAEs in 13.2% vs 7.5%. No new safety signals were observed, and no TRAEs led to treatment discontinuation or death. Conclusions: GC101 TIL significantly improved PFS and ORR versus investigator-choice chemotherapy in anti-PD-1-resistant melanoma, with manageable safety, supporting its potential as a new later-line therapy. Clinical trial information: NCT06703398 .

Acral melanoma,the most common melanoma subtype in East Asia,is associated with a poor prognosis.This study aims to comprehensively analyze the genomic characteristics of acral melanoma in East Asians.We conduct whole-genome sequencing of 55 acral melanoma tumors and perform data mining with relevant clinical data.Our findings reveal a unique mutational profile in East Asian acral melanoma,characterized by fewer point mutations and structural variations,a higher prevalence of NRAS mutations,and a lower fre-quency of BRAF mutations compared to patients of European descent.Notably,we identify previously underestimated ultraviolet radiation signatures and their significant association with BRAF and NRAS muta-tions.Structural rearrangement signatures indicate distinct mutational processes in BRAF-driven versus NRAS-driven tumors.We also find that homologous recombination deficiency with MAPK pathway mutations correlated with poor prognosis.The structural variations and amplifications in EP300,TERT,RAC1,and LZTR1 point to potential therapeutic targets tailored to East Asian populations.The high prevalence of whole-genome duplication events in BRAF/NRAS-mutated tumors suggests a synergistic carcinogenic effect that warrants further investigation.In summary,our study provides important insights into the genetic un-derpinnings of acral melanoma in East Asians,creating opportunities for targeted therapies.

Immune checkpoint blockade (ICB) therapy, especially in combination regimens, has significantly increased survival of renal cell carcinoma (RCC) patients. However, the ICB-resistant mechanisms remain largely unclear and require further investigation. Here, an immunosuppressive ecosystem in ICB-resistant tumors is identified, featured by preferential infiltration of MARCO+ tumor-associated macrophages (TAMs) and restrained cytotoxicity of CD8+ cytotoxic T lymphocytes (CTLs). The infiltrated MARCO+ TAMs can obstruct the development of CD8+ CTLs by impairing MHC-I-mediated neoantigen cross-presentation. Mechanistically, MARCO up-regulates the expression of SOCS1, which obstructs the kinase activity of JAK1, thereby downregulating MHC-I expression through the inhibition of the JAK1-STAT1-NLRC5 signaling cascade. Further, MARCO blockade significantly facilitates ICB therapy in in vivo models by recovering tumor recognition and priming anti-tumor CD8+ T cell responses. Taken together, these findings highlight MARCO as a highly desirable target in ICB-refractory individuals for immunorecognition reignition and immunotherapy modulation.