9501 Background: Advances in the neoadjuvant (neo) setting of locoregionally advanced melanoma have recently transformed practice. However, there continues to be a need to enhance efficacy while deescalating the duration of systemic therapy for those who achieve major pathologic (path) responses (MPR). Recent data support an important role for triplet INR checkpoint inhibitor therapy targeting CTLA4, PD1 and LAG3 in the treatment of advanced melanoma. Methods: A phase II trial of INR in patients (pts) with resectable clinical AJCC8 stages IIIB-D. It planned to enroll 20 evaluable pts to neo ipilimumab 1 mg/kg x1 dose combined with nivolumab-relatlimab (NR) 480/160 mg every 4 weeks x2 doses, followed by definitive surgery. Primary endpoint was MPR consisting of complete path response (pCR; 0% viable tumor) and near-pCR (< 10%). Secondary endpoints included other path responses, recurrence free survival, overall survival, event-free survival [EFS: disease progression (PD), recurrence or death] and safety. Adjuvant therapy was given in the absence of MPR. Biospecimens for research from consenting pts were banked at baseline, during treatment, at surgery and in follow up. Results: The study enrolled 20 pts August 2024-January 2026, including 7 female, 13 male, 16 cutaneous (2 acral) and 4 unknown primary. Median age was 65 (63-80). Baseline clinical stages were 9 IIIB, 9 IIIC, 2 IIID, including 4 pts with in-transit metastases. Among 19 patients who completed the neoadjuvant phase, all received ipilimumab and the median number of neo NR doses was 2. Among 19 pts who initiated treatment, 5 pts (26%) experienced Gr3 related AEs including colitis (1), encephalitis (1), headache (1), hyperglycemia (1), rash (1), and one pt experienced an event of Gr5 myocarditis. Among 18 pts who completed radiologic tumor response assessment, there were 13 (72%) partial response (PR), 4 stable disease (SD) and 1 disease progression (PD) preoperatively. There was 1 event of death before surgery. One patient declined surgery following deep PR, 1 pt is pending surgery and 1 pt continues in the neo phase. Among 16 pts who had surgery, median time from enrollment to surgery was 73 days. Pathologic responses were 5 path non-response (pNR), 1 near-pCR (microscopic residual disease) and 10 pCR (63%; 95% CI, 38.64 – 81.52). MPR rate (pCR + near-pCR) was 11/16 (69%; 95% CI, 44.40 – 85.84) and none of these patients received adjuvant therapy. Median follow up time from enrollment was 9.8 months. Among all pts, no pt experienced recurrence after surgery to date. Conclusions: Neoadjuvant INR demonstrated encouraging clinical activity in pts with resectable clinical stages IIIB/IIIC/IIID melanoma, warranting further investigation. The toxicity profile was consistent with reported data of INR and other combinations in this setting. Mechanistic and biomarker studies related to response, resistance and toxicity are underway. Clinical trial information: NCT06295159 .

Supplementary Fig. S1: Patient level heatmaps by assay. Supplementary Fig. S2: Overview of modalities in database. Supplementary Fig. S3: Computational pipeline for single cell RNA-seq data processing. Supplementary Fig. S4: Quality control metrics of tissue single cell data set. Supplementary Fig. S5: Cell type clusters in single-cell PBMC dataset. Supplementary Fig. S6: Myeloid cells in Merkel cell carcinoma. Supplementary Fig. S7: Response to ICB is associated with tumor cell upregulation of an interferon-γ signature. Supplementary Fig. S8: GSEA on tumor cells controlling for proliferation. Supplementary Fig. S9: Dot plot of RNA markers in CD8 clusters (tissue). Supplementary Fig. S10: Clonotype distribution between CD8 T cell lineages. Supplementary Fig. S11: CD8 lineage characterization. Supplementary Fig. S12: Clonality and neoantigen metrics of metrics of CD8 T cells. Supplementary Fig. S13: Comparisons of cell type signatures and cell proportions separated by sample origin. Supplementary Fig. S14: Epitope screening results. Supplementary Fig. S15: Public epitopes. Supplementary Fig. S16: Expression of naïve-like markers on Vδ1 T cells. Supplementary Fig. S17: GeoMx quality control for regions of interest. Supplementary Fig. S18: GeoMx image and differential expression analysis. Supplementary Fig. S19: CosMx quality control for fields of view. Supplementary Fig. S20: CosMx degree of co-localization sensitivity test shows B cell, T cell, pDC, and DC interactions. Supplementary Fig. S21: Pre-post genes. Supplementary Fig. S22: MIXCR analysis of pre-post ICB clonal dynamics. Supplementary Fig. S23: Single cell analysis of pre-post ICB clonal dynamics. Supplementary Fig. S24: Functional convergence of CD8 and γT cells. Supplementary Fig. S25: Graphical model of immune checkpoint response in MCC.

9553 Background: Metastatic uveal melanoma (mUM) has a poor prognosis with modest response to immune checkpoint blockade (ICB). Inactivating mutations in BRCA-1 associated protein 1 ( BAP1 ) are common in mUM leading to deficient homologous recombinant DNA repair and increasing reliance on alternate repair pathways, including poly[ADP-ribose] polymerase (PARP). We investigated if the PARP inhibitor olaparib could improve objective response to pembrolizumab (PEM) in mUM (NCT05524935). Methods: Key eligibility included mUM with measurable disease, ECOG 0-1, and adequate organ function. Prior liver directed therapy was allowed. Patients (pts) received PEM 200mg IV every 21 days plus oral olaparib 300mg BID till progression, toxicity, or completion of 2 years of treatment. In a Simon 2-stage optimal design, > 1 response by RECIST 1.1 in 12 evaluable patients would permit accrual to expand to 37 pts. Results: Twelve eligible mUM pts, 6 male and 6 female, median age 59 years (42, 84) were treated in stage 1 of this trial. Most pts (11/12; 92%) had both liver and extra-hepatic metastatic disease; 5 pts had elevated serum LDH at baseline. Seven pts (58%) were naïve to prior systemic therapy; 4 had prior liver directed therapy. There was no objective response observed; best response was stable disease (SD, n=5), remainder (n=7) with progression; disease-control rate was 42%. One patient continues treatment at 13 months with ongoing SD. Of 4 pts who received prior tebentafusp, three had SD lasting greater than 6 months. At median follow-up of 13 months, the median progression-free survival (PFS) was 3.6 months (95% CI: 2.2, 8.3), and median overall survival (OS) was 13.8 months (95% CI: 2.5, 24.3). Six-month PFS and OS were 0.42 (95% CI: 0.15, 0.67) and 0.83 (95% CI: 0.48, 0.96) respectively; one year OS was 0.73 (95% CI: 0.37, 0.91). Most common (≥ 25%) all grade treatment related adverse events (AEs) were fatigue, arthralgia, diarrhea, nausea/vomiting, dyspnea, anemia, abdominal pain, bloating, anorexia, muscle weakness, pruritus, rash, vitiligo, dyspnea, cough, hypertension, muscle weakness, ↑ AST, ↑ alkaline phosphatase, ↓ lymphocyte count, ↑ glucose, ↑ potassium and ↓ albumin. Grade 3 AEs were infrequent; nausea, anorexia, dehydration, hypotension, and muscle weakness (all n=1); there was no treatment related death. 5/12 (42%) pts required a dose reduction for olaparib. Conclusions: The addition of olaparib to pembrolizumab was well tolerated but did not result in any objective response in molecularly unselected mUM. Landmark survival results appear clinically meaningful and likely related to disease control with stable metastatic burden. Ongoing biomarker studies of tumor tissue (BAP1, PD-L1, PARP1, TIL) and blood may help discern which patients could benefit from this combination regimen. Optimal sequencing of ICB with tebentafusp in HLA-A*02:01+ mUM should be explored further. Supported by Merck, Inc. Clinical trial information: NCT05524935 .

Supplementary Table S1. Patient cohort: Clinical data and assays Supplementary Table S2. Sample cohort: Clinical data and assays Supplementary Table S3. Pre-ICB bulk RNA-seq cohort, 63 samples Supplementary Table S4. Pre-Post ICB matched bulk RNA-seq, 28 samples, 14 patient pairs Supplementary Table S5. Number of reads for each bulk RNA-seq sample (85 samples) Supplementary Table S6. Tissue single-cell RNA-seq quality control and sample metadata (54 samples) Supplementary Table S7. Blood single-cell RNA-seq quality control and sample metadata (55 samples) Supplementary Table S8. Multiplex immunofluorescence sample cohort (44 samples) Supplementary Table S9. GeoMx sample cohort (15 samples) Supplementary Table S10. CosMx sample cohort (13 samples) Supplementary Table S11: Genes differentially expressed between MCC that respond to anti-PD-1/PD-L1 therapy (n=31) and MCC that do not respond to anti-PD-1/PD-L1 therapy (n=32) in the pre-ICB cohort (n=63). Supplementary Table S12. Enrichr results of differentially expressed genes (FDR < 0.1) in responders and non-responders prior to immunotherapy in the bulk RNA-seq dataset Supplementary Table S13. Wilcoxon markers of single-cell RNA-seq clusters for all cells in the tissue dataset Supplementary Table S14. Pseudobulk markers of tumor cells comparing immunotherapy responders to non-responders Supplementary Table S15. GSEA results of differentially expressed genes from responders and non-responders in tumor cells pseudobulk dataset. Supplementary Table S16. SCENIC tumor markers of response and non-response to immunotherapy Supplementary Table S17. Genes in single-cell derived gene signatures Supplementary Table S18. Average antibody derived tag expression for cells in CD8 T cell clusters Supplementary Table S19. Wilcoxon markers of Tcirc and Trm CD8 T cells Supplementary Table S20. Wilcoxon markers of CD8 T cell clusters Supplementary Table S21. TCR clonotype linkage of CD8 T cell clusters Supplementary Table S22. Gini index of each CD8 cluster split by sample Supplementary Table S23. Small T and large T antigens peptide pools for TCR epitope screen Supplementary Table S24. GLIPH2 results of CD8 TCRs Supplementary Table S25. CD8 T cells with TCRs that match public databases with known antigens (VDJdb, McPAS-TCR, TRAdb) Supplementary Table S26. Wilcoxon RNA markers of each cell type, split by δ chain, in the γδ/NK cell subset. Supplementary Table S27. Wilcoxon antibody derived tag markers of each cell type, split by δ chain and cellular source, in the γδ/NK cell subset. Supplementary Table S28. Wilcoxon antibody derived tag markers of δ1 versus δ2 T cells from the tissue dataset Supplementary Table S29. Genes with significant non-linear fits along Vδ1 pseudotime Supplementary Table S30. GeoMx: Differentially expressed genes between responders and non-responders in tumor regions of interest Supplementary Table S31. GeoMx: Differentially expressed genes between responders and non-responders in stroma regions of interest Supplementary Table S32. CosMx: Differentially expressed genes among cell types identified by InSituType. Supplementary Table S33. CosMx: SpatialTime matrices of degree of co-localization between cell types Supplementary Table S34. CellChat receptor ligand interaction with CD8 exhausted T cells (Tex) as targets Supplementary Table S35. Differential expressed genes between bulk RNA-seq samples pre/post immunotherapy treatment split between responders (n=7) and non-responders (n=7) Supplementary Table S36. Enrichr results of differentially expressed genes (FDR <0.1) post-immunotherapy responders in the pre/post matched bulk RNA-seq dataset

The groundbreaking SWOG S1801 and NADINA trials herald the use of neoadjuvant immune checkpoint blockade (NICB) as the standard of care for stage III-IV melanomas. The shift from radiographic to pathologic response scoring offers an unprecedented window of opportunity to interrogate ‘on-treatment’ biospecimens for the immune mechanisms of response and resistance at curative-intent surgeries. To date, single-cell/bulk RNA studies have demonstrated that TCF7+ stem-like CD8+ T cells and tertiary lymphoid structures (TLS) are positive predictors of ICB response. However, how these immune cells organize and communicate within the spatial context of the neoadjuvant tumor microenvironment remains poorly understood. We assembled a retrospective cohort of 68 patients with stage 3 melanoma: 35 NICB (Ipilimumab-nivolumab, nivolumab-relatlimab, nivolumab) (14 complete response, 5 partial response, 14 non-response) and 33 treatment-naive (TxN). We dissected the spatial molecular architecture of all tumor beds (n= 68) and 10 uninvolved LNs (5 post-PD-1/5 TxN) using integrative multiplexed error-robust fluorescence in situ hybridization (MERFISH, 305 genes), Vectra multiplex IF and RNAscope multiomics in 38 whole slides and 3 TMAs (77 cores). We also performed single-cell-FFPE-seq using matched blocks (n=12) as an orthogonal benchmark for the MERFISH data. To analyze these high-dimensional datasets, we built a computational framework that includes 1) cell-typing with scFFPEseq validation, 2) novel spatial quantification methods to compute receptor-ligand (R-L) interactions that account for cell-cell distance and chemical signaling, 3) a graph-based clustering that quantifies germinal centers (GCs)/TLS, B-cell follicles and other spatial topography, and 4) MERFISH/ scRNA imputation via generative AI. Here, we showed that the quantity and size of spatially resolved GC/TLS and B-cell follicles surrounded by TCF7+ stem-like CD4+ and CD8 + T-cells, plasma cells, Tregs and myeloids are associated with positive response to neoadjuvant ICB. Our R-L analysis further revealed the critical importance of CXCR4-CXCL12 between TCF7+ T-cells and M2-myeloids and CCR7-CCL19/21 between TCF7+ T-cells in organizing the immune hubs in NICB response. Lastly, we incorporated scFFPEseq and MERFISH data into a unified latent embedding by environmental variational inference (ENVI) allowing us to impute melanoma phylogenetics, geographically accentuated hypoxic foci, and immune hubs within the spatial MERFISH images. We leveraged cutting-edge spatial -omics technologies and novel computational methods to resolve the immunologic hallmarks of NICB response in metastatic melanoma. We believe our approach will usher in a new paradigm for the investigation of other cancer biospecimens in the new era of standard of care neoadjuvant immunotherapy. Zichao Liu, Xiaofei Song, Jodi Balasi, Wei-Shen Chen, Jiang He, Justin He, Jonathan Nguyen, Carlos Moran-Segura, Joseph Johnson, Chaomei Zhang, Jane Messina, Zena Sayegh, Nan Sun, Douglas Marchion, Sean Yoder, Vernon K. Sondak, Jeffrey H. Chuang, Pei-Ling Chen. Spatially resolved immunologic hallmarks of response to neoadjuvant immune checkpoint blockade in metastatic melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5809.

9511 Background: Predictive biomarkers that identify melanoma tumors capable of generating therapeutically effective tumor-infiltrating lymphocytes (TIL) remain inadequately defined. We conducted an integrated spatial and histopathologic evaluation of resected melanoma specimens to characterize tumor microenvironmental features associated with durable clinical benefit (≥12 months) and improved survival following TIL therapy. Methods: Data were extracted from early-phase clinical trials conducted at Moffitt Cancer Center enrolling patients with advanced melanoma treated with TIL. Tumor specimens from 45 of 50 patients undergoing TIL manufacturing were evaluable based on tissue availability. Histopathology, multiplex immunofluorescence, and NanoString GeoMx spatial transcriptomics were performed on procured tumors. Tertiary lymphoid structures (TLS) were defined as organized aggregates of CD3⁺ T cells and CD20⁺ B cells and classified as immature or mature based on CD21/CD23 expression. Associations with ex vivo TIL expansion, infused product composition, clinical response, progression-free survival (PFS), and overall survival (OS) were assessed. Results: Procurement sites included subcutaneous/soft tissue (78%), lymph node (18%), and lung (4%). Viable tumor content and necrosis were not associated with ex vivo TIL expansion or response. TLS were present in 56% (25/45) of tumors and were strongly associated with clinical response (18/45 responders overall; response rate 56% [14/25] in TLS-positive tumors vs 20% [4/20] in TLS-negative tumors; p=0.014), improved PFS after adjusting for procurement site (HR=0.32; p=0.005) and improved OS (p<0.0001). Evaluation of ex vivo TIL expansion demonstrated that TLS presence did not affect the total number of expanded TIL (p = 0.3) but was significantly associated with a higher proportion of CD8⁺ T cells within the infusion product (83% vs 56%, p=0.027). Spatial transcriptomics of 132 immune-enriched regions of interest demonstrated that responders exhibited marked upregulation of antigen-presentation and interferon-response programs, including MHC I/II genes, STAT1/2, IFI6, and CXCL9 . A patient-level immune activation score derived from the top 25 responder-associated transcripts correlated with improved PFS (p<0.0001). To independently validate the significance of this immune gene set score, we analyzed an external dataset from the Lauss et al., 2017 cohort. Consistent with our findings, the gene set score was significantly higher in TIL responders compared with non-responders (p=0.048). Conclusions: TLS and spatial immune activation signatures in procured melanoma tumors are strongly associated with response and survival following TIL therapy. These findings support the integration of tumor microenvironment biomarkers into patient selection and optimization strategies for TIL therapy.

LBA9505 Background: Advances in the neoadjuvant (neo) setting of locoregionally advanced melanoma have recently transformed practice. However, there continues to be a need to enhance efficacy while minimizing systemic toxicity. Preliminary data support an important role for vidutolimod (V), a CpG-A TLR9 agonist packaged within a virus-like particle given intratumorally (IT) in combination with IV anti-PD1. Methods: A U.S. intergroup randomized phase II trial of pembrolizumab (P) vs P + V in patients (pts) with resectable clinical AJCC8 stages IIIB-D. It planned to randomize ~60 pts (for 54 evaluable) 1:1 to Arm A (neo P 200 mg IV Q3W x3) or Arm B (neo P 200 mg IV Q3W x3 + V 5 mg SC x1 then 10 mg IT QW x6) followed by definitive surgery then adjuvant P 400 mg IV Q6W x8, stratified by stage (IIIB/C vs IIID). Primary endpoint was pathologic (path) complete response (pCR) on each arm. Secondary endpoints included path responses, recurrence, overall survival, event-free survival [EFS: disease progression (PD), recurrence or death] and safety. Results: EA6194 enrolled 57 pts March 2021-March 2024, 19 female, 38 male, all cutaneous primary (1 acral on Arm B), median age 64 (26-88), 29 on Arm A [10 IIIB (8N1b, 1N1c, 1N2b), 19 IIIC (3N1b, 1N1c, 3N2b, 3N2c, 7N3b, 2N3c)] and 28 on Arm B [13 IIIB (11 N1b, 2 N2b), 13 IIIC (5 N1b, 2N2b, 2N2c, 3N3b, 1N3c), 2 IIID (N3b)]. The median numbers of neo P/adjuvant P doses were similar for both arms at 2/7. Median number/total dose of V were 7/60 mg. Among pts who initiated treatment, highest grade related AEs (Gr 3/4) were 25% in Arm A (N=28) and 29% in Arm B (N=28) including in Arm B diarrhea (1), injection site reaction (1), cytokine release (1), wound dehiscence (1), lymphocytopenia (1), pain (1), headache (1), hypertension (1), hypotension (1), all Gr 3 and one Gr 4 hyperglycemia. Median time to surgery from randomization 2.5 months. Median follow up time from enrollment 19 months. There were 3 deaths on Arm A and 1 on Arm B. On Arm A 25 pts had surgery, 6 path non-response (pNR), 2 partial (pPR), 3 near-pCR, 14 pCR (56%; 95% CI, 35 - 76). MPR (pCR + near-pCR) was 17/25 (68%; 95% CI, 46 - 85). Among these, 3 had recurrence after surgery (1 pNR, 1 near-pCR, 1 pCR). On Arm B 27 pts had surgery, 3 pNR, 2 pPR, 2 near-pCR, 20 pCR (74%, 95%, CI 54 - 89). MPR 22/27 (79%; 95% CI 62 - 94). Among these, 2 had recurrence after surgery (1 pNR, 1 pPR). Table 1 summarizes efficacy data including all enrolled pts. Conclusions: Neoadjuvant P + Vdemonstrated acceptable safety and encouraging clinical activity in pts with resectable clinical stages IIIB/IIIC/IIID melanoma when considering Arm A and historical controls, warranting further investigation. Clinical trial information: NCT04708418 . Arm A:P (N=29*) Arm B:P + V (N=28*) pCR (%; 95% CI) 14 (48; 29 - 67) 20 (71; 51 - 87) MPR (%; 95% CI) 17 (59; 39 - 76) 22 (79; 59 - 92) 1-year EFS (95% CI) 75% (59 – 91) 89% (78 - 100) *4 on Arm A and 1 on Arm B did not have surgery due to PD.

9516 Background: Identification of the TIL therapy product (TILp) characteristics associated with objective response to therapy is critical to improve future adoptive TIL-based therapy. In this study we performed comprehensive phenotypic analysis of TILp in association with TIL therapeutic outcomes in consecutive patients with MM. Methods: Data were extracted from early-phase clinical trials of MM patients treated with TIL only, TIL plus ipilimumab, TIL plus nivolumab, and TIL plus vemurafenib at Moffitt Cancer Center. The immunophenotypic features of TILp were evaluated by flow cytometry using antibodies against checkpoints, costimulatory molecules, T cell subsets and TCRβ sequencing. Tumor reactivity was measured using HLA-matched cell lines. The relationship between TILp characteristics and both objective response and progression-free survival (PFS) was assessed in treated patients. Results: A total of 50 patients, 21 female (42%) and 29 male (58%), median age 49 [IQR 40–55] received lymphodepleting chemotherapy followed by TIL and interleukin-2 (IL-2). Median numbers [IQR] of infused TIL and IL-2 dose were 59e 9 [42-84e 9 ] and 5 [4–6], respectively. Patients with objective response had a significantly higher total number of infused TIL, total number of infused CD8 + TIL, and proportion of CD8 + cells in the infusion product (p < 0.05), with high CD8 + TIL being associated with improved PFS (p = 0.0001). The total number and proportion of infused stem cell-like memory CD8 + T cells (T SCM, CD8 + CD45RA + CCR7 + CD62L + CD95 + ) were significantly higher in responders and associated with improved PFS (p < 0.01). TILs from responders had distinct patterns of co-inhibitory and co-stimulatory receptors’ expression and were characterized by significantly higher proportion of LAG3 + and LAG3 + TIGIT + co-expressed TIL of total CD3 + TIL (p < 0.05). Proportions of LAG3 + CD8 + and TIGIT + CD8 + cells were also increased in responders (p < 0.05) with no significant differences observed in PD1 + CD8 + , BTLA + CD8 + , and TIM3 + CD8 + cells. There was an increased proportion of OX40 + CD8 + and OX40 + 4-1BB neg CD8 + cells among responders (p < 0.01). T cell clonal analysis using top 20 clones revealed high persistence in responders (p < 0.01) measured by TCRβ overlap between ACTP and post-treatment peripheral blood. There were no significant differences in clonality, diversity and evenness in responders vs. non-responders. Using HLA-matched cell lines there was a trend towards HLA-matched reactive TIL and improved PFS (p = 0.058). Conclusions: Response to TIL therapy is associated with TIL persistence and distinct TILp immunophenotypic features characterized by enhanced proportion of CD8 + TIL, T SCM CD8 + cells, and high surface expression of LAG3 + TIGIT + and OX40 + . Novel strategies to modulate ex vivo TIL expansion toward this optimal TIL phenotype may result in increased response for future trial design.

9535 Background: Simultaneous blockade of lymphocyte activation gene-3 (LAG-3) and programmed death-1 (PD-1) pathways enhances antitumor activity in patients (pts) with mel. While efficacy of nivo-rela has been demonstrated in a large cohort of therapy-naïve advanced mel, it remains important to assess clinical activity of this combination in real-world settings focusing on resectable mel and on tumor microenvironment (TME) factors impacting response to nivo-rela. In this study we report the clinical activity of nivo-rela and the relation of TME characteristics to response to therapy. Methods: This study included pts with mel treated with first-line nivo-rela in the neoadjuvant or advanced settings between 2022-2024 at Moffitt. Safety, progression-free (PFS), overall survival (OS) and pathologic response rates were evaluated in relation to mel characteristics. To assess the impact of TME on response to nivo-rela in advanced mel, we performed NanoString GeoMx proteomic analysis using immune cell profiling, immune cell typing, and IO drug target panels. Differential proteomic analysis (fold change ≥0.05, FDR <0.05) was performed using selected baseline samples (n=16) from responders (R) and non-responders (NR). Results: The study included 128 mel pts treated with nivo-rela in the advanced (n=101, [C1]) and neoadjuvant (n=27, [C2]) settings, 48 female (38%) and 80 male (62%), median age 71 [IQR 62–78]. In C1, with a median follow up of 13 months, mPFS and mOS with 95% CI were 19m (10-NR), and NR (25-NR), respectively. In C2, 6 (22%) of pts did not undergo definitive surgery due to disease progression (n=4) and clinical response to therapy (n=2). Among path-evaluable pts (n=21), major pathologic response rate was 10 (48%) including 6 complete (29%) and 4 near-complete responses 19%), with partial response in 3 (14%) and non-response in 8 patients (38%). Adverse events included adrenal insufficiency and myocarditis in 12 (9%) and 4 (3%) of all pts. TME assessment revealed no significant differences in expression of LAG3, PD-1, PDL-1, CD8 + among R vs NR. Baseline tumors from NR were characterized by high expression of B7-H3 in both tumor and immune compartments. No significant correlations were found between B7-H3 and PDL-1, LAG-3, and CTLA-4. High B7-H3 ROIs were enriched by CD163 + , CD68 + , CD14 + , and fibroblast activation protein alpha. Conclusions: In a real-world setting, first-line nivo-rela in advanced mel resulted in potentially better PFS while in resectable pts the major pathologic response rate was lower than previously reported in clinical trials. Mel TME with high B7-H3 protein expression was enriched with M2-skewed macrophages and fibroblasts in association with non-response to nivo-rela. This finding warrants further investigation of B7-H3 targeting agents in mel pts.

The 5th edition of the World Health Organization Classification of Tumours (WCT) serves as a foundation for global diagnostic standards in tumour pathology. Similar to other volumes in this series, the Skin Tumours (Skin5) edition follows a standardized approach. This edition introduces two new chapters: 'Tumours of the nail unit' and 'Metastases to skin', along with new entities across relevant chapters. This review article provides an overview of the updates in Skin5 based on currently published evidence, with emphasis on newly introduced chapters and newly described entities that involve the skin, in particular, epidermal, melanocytic and appendageal tumours.

BACKGROUND:Next-generation sequencing (NGS) is becoming more commonly used for diagnosis in dermatopathology. It's critical to appraise its efficacy and limitations. Distinguishing benign deep penetrating nevi (DPN) from deep penetrating like-melanoma (DPN-M) is a challenging diagnostic scenario even for experienced dermatopathologists. METHODS:We sent a two-phase survey (pre-and postgenomics) to 32 experienced dermatopathologists to evaluate 39 diagnostically challenging cases from the DPN/WNT-activated family of melanocytic neoplasms. RESULTS:With NGS data, interobserver agreement improved from 0.41 to 0.51 (p < 0.0001) in distinguishing DPN-M from nonmelanoma cases. Overall diagnostic accuracy improved, mostly driven by a 16% increase in accurate diagnosis of DPN-M. However, in two cases, the inclusion of genomics shifted the majority vote from a correct to an incorrect diagnosis. A total of 218 diagnostic changes occurred between Survey 1 and 2. Among the changes, 132 votes moved toward the correct diagnosis while 86 moved toward an incorrect diagnosis. The shift in voting which resulted in improved diagnostic accuracy was statistically significant (p = 0.0001). CONCLUSIONS:NGS has the potential to improve interobserver agreement and diagnostic accuracy. We provide guidance on the utilization of bioinformatic data to maximize its benefits and improve diagnostic accuracy and interobserver agreement.

Next-generation sequencing (NGS) is increasingly being utilized as an ancillary tool for diagnostically challenging melanocytic neoplasms. It is incumbent upon the pathology community to perform studies assessing the benefits and limitations of these tools in specific diagnostic scenarios. One of the most challenging diagnostic scenarios faced by skin pathologists involves accurate diagnosis of desmoplastic melanocytic neoplasms (DMNs). In this study, 20 expert melanoma pathologists rendered a diagnosis on 47 DMNs based on hematoxylin and eosin sections with demographic information. After submitting their diagnosis, the experts were given the same cases, but this time with comprehensive genomic sequencing results, and asked to render a diagnosis again. Identification of desmoplastic melanoma (DM) improved by 7%, and this difference was statistically significant ( P <0.05). In addition, among the 15 melanoma cases, in the pregenomic assessment, only 12 were favored to be DM by the experts, while after genomics, this improved to 14 of the cases being favored to be DM. In fact, some cases resulting in metastatic disease had a substantial increase in the number of experts recognizing them as DM after genomics. The impact of the genomic findings was less dramatic among benign and intermediate-grade desmoplastic tumors (BIDTs). Interobserver agreement also improved, with the Fleiss multirater Kappa being 0.36 before genomics to 0.4 after genomics. NGS has the potential to improve diagnostic accuracy in the assessment of desmoplastic melanocytic tumors. The degree of improvement will be most substantial among pathologists with some background and experience in bioinformatics and melanoma genetics.

Abstract Introduction: IT TAVO-EP (tavokinogene telseplasmid delivered by electroporation) results in localized expression of IL12 in the tumor microenvironment (TME). This study (NCT04526730) evaluated NeoAd TAVO-EP and NIVO. All patients provided a written informed consent (Advarra IRB Pro00041794). Biomarkers of tumoral and systemic immune responses were conducted and correlated with clinical outcomes. Methods: NeoAd phase comprised up to 3 × 4-week cycles of IT TAVO-EP and iv NIVO followed by surgery. Endpoints included pathologic complete response (pCR), near pCR, major response (pMR; pCR + near pCR) and nonresponse (pNR). Biospecimens were collected at screening, C1D8, C2D1 (~30 days from treatment start), pre-surgery (~90 days). Slides cut from FFPE tissue were analyzed by chromogenic immunohistochemistry (IHC) for CD8+ tumor infiltrating lymphocytes (TIL) and PD-L1 (22C3 CDx assay). RNA extracted from FFPE tissue was subjected to NanoString’s IO360 transcriptomic analysis, including Tumor Inflammation Signature (TIS). PBMCs were stained and analyzed via flow cytometry on an LSR Fortessa X-20. Serum samples were analyzed for cytokine levels using Luminex® MAGPIX® platform (15-plex). Results: 16 pts were treated; 1 with PR declined surgery, 1 with early distant PD did not have surgery, 14 had surgery: 2 pNR, 3 near pCR, 9 pCR; pMR rate 12/15 (80%). In tumor, at baseline, 9/11 pts tested had <20% CD8+ TIL, 7/11 <10% PD-L1 tumor proportion score (TPS) and 6/10 TIS of ≤0, predicting non-response to anti-PD-1. In 5 pts with evaluable matched tissue at baseline and C2D1: 5/5 had increased peritumoral CD8+ T cells, 4/5 increased CD8+ TIL, 2/5 increased PD-L1, 4/5 increased TIS at C2D1 compared to baseline. Tumoral transcriptome profile revealed significant upregulation of genes involved in innate and adaptive immune responses (IFN-gamma, APM, granzymes, CD8, PD-L1, JAK1, chemokines, others) at C2D1. In blood, proliferating Ki-67+/PD-1+/CD8+ T cells expanded at C2D1 while total PD1+/CD8+ cells decreased. Decrease of PD1+CD8+ cells and other subtypes in blood at C2D1 coincided with TME infiltration by CD8+ cells. Serum effector cytokines including IL12, IFN-gamma and IL2, showed no biologically meaningful changes following treatment or differences between responders and non-responders. Conclusions: At baseline, most patients exhibited low CD8+ TIL, PD-L1 and TIS, with enhanced immune activation following treatment in the TME and blood including increased immune-related gene expression, CD8+ TIL, peritumoral CD8+ T cells and TIS. Four of 5 pts with negative predictive baseline biomarkers [CD8+TIL/PD-L1/TIS]low experienced pCRs supporting activity of IL12/anti-PD1 based regimens in this setting. Citation Format: Ahmad A. Tarhini, Zeynep Eroglu, Jonathan S. Zager, Ricardo J. Gonzalez, Amod A. Sarnaik, C Wayne Cruse, Deanryan B. De Aquino, Edith Abraham, Diana M. Acevedo, Allison Richards, Michael J. Schell, Denise Kalos, Pei-Ling Chen, Jane L. Messina, David A. Canton, Vernon K. Sondak. Tumoral and systemic immune modulation with neoadjuvant (NeoAd) intratumoral (IT) TAVO-EP (plasmid IL-12 electro gene transfer) and nivolumab (NIVO) in patients (pts) with operable locoregionally advanced melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3734.

PURPOSE:Intratumoral tavokinogene telseplasmid delivered by electroporation (TAVO-EP) results in localized expression of IL-12 within the tumor microenvironment (TME). This study evaluated neoadjuvant TAVO-EP combined with intravenous nivolumab followed by surgery and adjuvant nivolumab in patients with operable, locoregionally advanced melanoma. PATIENTS AND METHODS:The neoadjuvant phase comprised up to 3 × 4-week cycles during which TAVO-EP was given intratumorally on days 1, 8, and 15 (optional) concurrently with 480 mg nivolumab intravenously on day 8 of each 4-week cycle. Surgery followed, and adjuvant nivolumab was initiated after surgery. The primary endpoint was pathologic complete response (pCR). Secondary endpoints included major pathologic response (MPR; pCR or near pCR). RESULTS:Sixteen patients were enrolled, and the preoperative radiological response rate was 63%. One patient declined surgery after experiencing a significant clinical response. Among the remaining 15 patients, the pCR rate was 60% and the MPR was 80%. No patient with MPR has had disease recurrence with a median follow-up from the date of surgery of 15.4 months. At baseline, most patients exhibited low CD8+ tumor-infiltrating lymphocytes, PD-L1, and IFN-γ gene expression signature. There was enhanced immune activation following treatment in the TME and blood, including increased immune-related gene expression, CD8+ tumor-infiltrating lymphocytes, and proliferating immune cell subsets. CONCLUSIONS:The clinical efficacy of neoadjuvant intratumoral TAVO-EP + nivolumab is promising with 80% of patients achieving an MPR. Evidence of potent immune activation both systemically and within the TME along with a favorable safety profile supports the activity of local IL-12 and anti-PD-1 based regimens.

Background: Spiradenocylindroma is a benign tumor of skin adnexal origin with overlapping features of two distinct neoplasms: spiradenoma and cylindroma. This cutaneous tumor typically presents on the head and neck and extracutaneous presentations are uncommon. The presentation described below involves a spiradenocylindroma within a mature ovarian teratoma is very rare. Aim: The aim of this article is to portray the diagnostic process of this unusual spiradenocylindroma presentation. Case presentation: A 65 year-old female with a left adnexal mass underwent ultrasonography and magnetic resonance imaging (MRI) which showed a left ovarian multiseptated lesion, with mural calcifications and projections into the mass. Excisional surgery was performed and histopathological examination revealed a spiradenocylindroma. Conclusion: Spiradenocylindroma is rare, hard to identify, and often misdiagnosed. Our study described the process of diagnosis and depicts the rare presentation of this lesion arising within a mature teratoma.

Background While the prognostic role of tertiary lymphoid structures (TLS) has been well studied in solid cancers, the prevalence and impact of immature precursor lymphoid structures known as lymphoid aggregates (LA) remain unresolved in relation to the disease process. In this study, we examined characteristics and the prognostic utility of LA and TLS status in histological samples from patients with melanoma.Methods We assessed The Cancer Genomic Atlas-skin cutaneous melanoma digital slides and melanoma specimens from the University of Pittsburgh for the presence of LA and TLS using H&E staining, multiplex immunofluorescence (mIF) and transcriptomic analyses. Cox proportional hazard regression models were used to assess the prognostic value associated with the presence of lymphoid structures in melanomas.Results A total of 278 evaluable samples were analyzed and split into primary melanomas in skin (N=195) and metastatic melanomas involving skin/subcutaneous/soft tissue sites (N=83). 72% of tumor specimens contained histologically defined LA located in peritumoral (34%), intratumoral (5.6%) or stromal (6.1%) locations, with the remaining samples (54.3%) exhibiting LA in multiple locations. In contrast to LA which tended to form more commonly in primary melanoma samples, TLS with germinal centers predominantly formed in peritumoral (45.2%) or stromal (35.5%) locations in metastatic melanomas (p=0.02), with TLS observed in 11% of all melanoma specimens evaluated. mIF analyses revealed cellular heterogeneity of lymphoid structures, with CD20+ (B) cells present in nodule-shaped and stromal locations where they exhibited a high degree of colocalization with CD4+ and CD8+ T cells. A previously defined 12-chemokine gene expression score was significantly higher in samples with evidence of LA versus none (p<0.001), and samples without LA/TLS were enriched with pigmentation/neural network gene signatures. The presence of LA was significantly associated with tumor-free regional lymph node status (p=0.002). In multivariable analysis, after adjusting for age, sex, sample type, and stage, the presence of LA was associated with improved patient overall survival (OS) (HR=0.52, 95% CI 0.31 to 0.87, p=0.01).Conclusion Melanoma frequently contains LA, which tends to form in diverse locations in the tumor microenvironment in association with improved overall survival and tumor-free regional lymph node status in patients with primary disease.

PURPOSEThe purpose of this study was to develop recommendations for the diagnostic evaluation and surgical management of cutaneous melanoma (CM) and atypical Spitz tumors (AST) and non-Spitz melanocytic tumors (melanocytomas) in pediatric (age 0-10 years) and adolescent (age 11-18 years) patients.METHODSA Children's Oncology Group-led panel with external, multidisciplinary CM specialists convened to develop recommendations on the basis of available data and expertise.RESULTSThirty-three experts from multiple specialties (cutaneous/medical/surgical oncology, dermatology, and dermatopathology) established recommendations with supporting data from 87 peer-reviewed publications.RECOMMENDATIONS(1) Excisional biopsies with 1-3 mm margins should be performed when feasible for clinically suspicious melanocytic neoplasms. (2) Definitive surgical treatment for CM, including wide local excision and sentinel lymph node biopsy (SLNB), should follow National Comprehensive Cancer Network Guidelines in the absence of data from pediatric-specific surgery trials and/or cohort studies. (3) Accurate classification of ASTs as benign or malignant is more likely with immunohistochemistry and next-generation sequencing. (4) It may not be possible to classify some ASTs as likely/definitively benign or malignant after clinicopathologic and/or molecular correlation, and these Spitz tumors of uncertain malignant potential should be excised with 5 mm margins. (5) ASTs favored to be benign should be excised with 1- to 3-mm margins if transected on biopsy. (6) Re-excision is not necessary if the AST does not extend to the biopsy margin(s) when complete/excisional biopsy was performed. (7) SLNB should not be performed for Spitz tumors unless a diagnosis of CM is favored on clinicopathologic evaluation. (8) Non-Spitz melanocytomas have a presumed increased risk for progression to CM and should be excised with 1- to 3-mm margins if transected on biopsy. (9) Re-excision of non-Spitz melanocytomas is not necessary if the lesion is completely excised on biopsy.

Abstract Merkel cell carcinoma (MCC) is an aggressive neuroendocrine skin cancer with a ∼50% response rate to immune checkpoint blockade (ICB) therapy. To identify predictive biomarkers, we integrated bulk and single-cell RNA sequencing (RNA-seq) with spatial transcriptomics from a cohort of 186 samples from 116 patients, including bulk RNA-seq from 14 matched pairs pre- and post-ICB. In nonresponders, tumors show evidence of increased tumor proliferation, neuronal stem cell markers, and IL1. Responders have increased type I/II interferons and preexisting tissue resident (Trm) CD8 or Vδ1 γδ T cells that functionally converge with overlapping antigen-specific transcriptional programs and clonal expansion of public T-cell receptors. Spatial transcriptomics demonstrated colocalization of T cells with B and dendritic cells, which supply chemokines and costimulation. Lastly, ICB significantly increased clonal expansion or recruitment of Trm and Vδ1 cells in tumors specifically in responders, underscoring their therapeutic importance. These data identify potential clinically actionable biomarkers and therapeutic targets for MCC. Significance: MCC serves as a model of ICB response. We utilized the largest-to-date, multimodal MCC dataset (n = 116 patients) to uncover unique tumor-intrinsic properties and immune circuits that predict response. We identified CD8 Trm and Vδ1 T cells as clinically actionable mediators of ICB response in major histocompatibility complex–high and –low MCCs, respectively.