Joseph, Richard W.

9558 Background: Desmoplastic melanoma (DM) is a rare subtype of melanoma characterized by a dense fibrous stroma, resistance to chemotherapy and no actionable driver mutation for targeted therapy. We investigated the efficacy of PD-1/L1 inhibitors and correlation with genetic landscape and tumor immune microenvironment in DM. Methods: Retrospective analysis of 1054 pts with melanoma treated with anti-PD-1/L1, resulting in 57 pts with unresectable or metastatic DM. Available baseline biopsies were analysed by digital quantitative immunohistochemistry (IHC) for CD8 and PD-L1 and by whole exome sequencing (WES), compared to available tissue from non-DM pts treated with anti-PD1/L1 at UCLA. Results: At a median follow up of 20 mo, 40 pts (70%, 95% CI 57-82) had an objective response by RECIST 1.1 criteria, including 18 (45%) CRs with no relapse observed to date. Responses were similar in DM subsets (23 pure, 29 mixed and 5 indeterminate). Kaplan-Meier estimated 1-year and 2-year overall survival were 85% (95% CI 78-98) and 74% (95% CI 64-89). WES revealed a median of 1282 (interquartile range 517-1692) non-synonymous somatic mutations per tumor in DM tumors (n = 17), significantly higher (p = 0.02) than the median of 462 (interquartile range 230-1150) in non-DM (n = 23). Mutations in NF-1 were the most common (13/17) followed by loss-of-function TP53 and ARID2, and > 82% of single nucleotide mutations were UV damage signatures. IHC analysis from 19 DM and 13 non-DM revealed a strikingly higher percentage of PD-L1 positive cells in the tumor parenchyma in DM (p = 0.04), highly associated with CD8 density and PD-L1 expression in the tumor invasive margins, indicating an active adaptive immune response. No genetic mechanisms known to cause constitutive PD-L1 expression were detected in these samples. Conclusions: Patients with advanced DM derive significant clinical benefit from PD-1/L1 inhibitors, likely related to the high mutational burden and a highly active adaptive immune response as the main mechanism of immune escape prior to therapy. Our results challenge the general conception that dense fibrous stroma around the malignant cells interferes with immune cell infiltration and efficacy of immunotherapy.

e21057 Background: BRAF, NRAS, CDKN2A and P53 mutations are commonly observed pathogenic mutations in cutaneous melanoma. The influence of mutations such as BRAF, NRAS, CDKN2A and P53 on treatment response in the PD1 inhibitor era has not been well defined. Methods: This study included patients with melanoma with a cutaneous or unknown primary. All patients had genomic profiling with the Mayo Clinic 50-gene solid tumor mutation panel. Treatment response to first line immunotherapy with PD1, CTLA-4, or combination PD1/CTLA-4 inhibitors were assessed by response rate and time-to-progression (TTP) rate at 12 months. Each patient outcome was stratified by non-mutually exclusive gene mutation status into BRAF, NRAS, P53, CDKN2A, or quadruple negative (N4) groups. Results: Genomic profiling was performed on 208 patients, of whom 102 received first line immunotherapy. Most patients received a PD1 inhibitor (n = 62), followed by CTLA-4 inhibitor (n = 34), and combination therapy (n = 6). A trend towards improved outcomes was observed in CDKN2A and BRAF mutated patients with worse outcomes noted in P53, N4, and NRAS patients. The percent without progression at 12 months were BRAF 52.2%, CDKN2A 58.2%, NRAS 30.5%, P53 37.7%, and N4 26.7% (see table below). Response rates were BRAF 46.4%, CDKN2A 38.5%, NRAS 22.2%, P53 40.0%, and N4 33.3%. Conclusions: There is a trend towards improved outcomes with immunotherapy in BRAF and CDKN2A mutated patients while P53, NRAS and N4 patients have inferior outcomes. [Table: see text]

4505 Background: While inhibiting VEGF improves outcomes in mRCC pts, most develop resistance, often within a year. Here, we report results from a Ph II study of atezo (anti–PD-L1) and bev (anti-VEGF) vs and following atezo or sun (TKI) in mRCC pts. Methods: Pts with untreated mRCC were enrolled in the hypothesis generating IMmotion150 study (NCT01984242) and randomized to atezo 1200 mg IV q3w + bev 15 mg/kg IV q3w, atezo alone or sun 50 mg PO QD 4 wk on/2 wk off. After progression on atezo or sun, crossover to atezo + bev was allowed. PD-L1 status was scored on tumor-infiltrating immune cells (IC, SP142 IHC assay). The primary analysis was modified prior to final analysis to reflect the coprimary endpoints of IRF-assessed PFS (RECIST v1.1) in ITT pts and pts with PD-L1 expression on ≥ 1% of IC (PD-L1+). Results: 54% of pts were PD-L1+. In PD-L1+ pts 1L treatment resulted in a PFS hazard ratio (HR) of 0.64 for atezo + bev vs sun (table). After 1L treatment, 78% of sun and 60% of atezo pts who progressed subsequently received atezo + bev and achieved ORRs of 28% and 24%, respectively (table). Safety was comparable to the known individual profiles of atezo and bev. Additional clinical, safety and biomarker data will be presented. Conclusions: Atezo + bev resulted in encouraging antitumor activity in 1L pts with PD-L1+ mRCC. Preliminary activity in the 2L setting was demonstrated in pts who crossed over to atezo + bev, regardless of prior therapy. 1L atezo + bev vs sun is being evaluated in the ongoing Ph III study IMmotion151 (NCT02420821). Clinical trial information: NCT01984242. [Table: see text]

9517 Background: Although PD-1/L1 axis directed therapies induce durable responses in some mm patients (pts), biomarkers of response remain elusive. We hypothesized that quantifying key immune suppression mechanisms within the tumor microenvironment would provide superior predictors of response to anti-PD-1 compared with single marker assessment. Methods: Pre-treatment tumor biopsies from 124 mm pts treated with anti-PD-1 at 7 academic centers were fluorescently stained with multiple immune markers in discovery (n = 24) and validation (n = 100) cohorts. Selected biomarker signatures, PD-1/PD-L1 interaction score (proportion of PD-1+ cells co-localized with PD-L1) and IDO1/HLA-DR co-expression were evaluated for anti-PD-1 treatment response and survival. Slides were imaged using Vectra; biomarker positive cells and their co-localization were objectively quantified in pathologist-selected regions using novel Automated Quantitative Analysis (AQUA) algorithms. Results: In the discovery cohort, high levels of PD-1/PD-L1 interaction score and/or IDO1/HLA-DR coexpression was strongly positively associated with response to anti-PD-1 (p = 0.0005). In contrast, other individual biomarkers (PD-1, PD-L1, CD8) were not associated with response or survival (p > 0.10). This finding was replicated in the validation cohort: pts with high PD-1/PD-L1 and/or IDO1/HLA-DR were more likely to respond to treatment (p = 0.009). These pts also experienced a three-fold increase in progression free survival (hazards ratio (HR) = 0.33; p = 0.003) and overall survival (HR = 0.34; p = 0.004). Multivariate analyses revealed that these findings were independent of BRAF mutation, stage, LDH and prior therapy. In the combined cohort (n = 124), 80% of responding pts had higher levels of PD-1/PD-L1 interaction scores and/or IDO1/HLA-DR. In contrast, PD-L1 expression alone (≥1% or ≥50%) was not predictive of PFS or OS (p > 0.1). Conclusions: This novel multiplexed method profiling key tumor-immune suppression pathways identified mm pts likely to respond to anti-PD-1 therapy. This method could help stratify patients for PD-1 monotherapy and be useful in guiding future clinical trials.