Faculty Bibliography

Metastatic cancers produce exosomes that condition pre-metastatic niches in remote microenvironments to favor metastasis. In contrast, here we show that exosomes from poorly metastatic melanoma cells can potently inhibit metastasis to the lung. These "non-metastatic" exosomes stimulate an innate immune response through the expansion of Ly6Clow patrolling monocytes (PMo) in the bone marrow, which then cause cancer cell clearance at the pre-metastatic niche, via the recruitment of NK cells and TRAIL-dependent killing of melanoma cells by macrophages. These events require the induction of the Nr4a1 transcription factor and are dependent on pigment epithelium-derived factor (PEDF) on the outer surface of exosomes. Importantly, exosomes isolated from patients with non-metastatic primary melanomas have a similar ability to suppress lung metastasis. This study thus demonstrates that pre-metastatic tumors produce exosomes, which elicit a broad range of PMo-reliant innate immune responses via trigger(s) of immune surveillance, causing cancer cell clearance at the pre-metastatic niche.

Current staging guidelines are insufficient to predict which patients with thin primary melanoma are at high risk of recurrence. Computer-assisted image analysis may allow for more practical and objective histopathological analysis of primary tumors than traditional light microscopy. We studied a prospective cohort of stage IB melanoma patients treated at NYU Langone Medical Center from 2002 to 2014. Primary tumor width, manual area, digital area, and conformation were evaluated in a patient subset via computer-assisted image analysis. The associations between histologic variables and survival were evaluated using Cox proportional hazards model. Logistic regressions were used to build a classifier with clinicopathological characteristics to predict recurrence status. Of the 655 patients with stage IB melanoma studied, a subset of 149 patient tumors (63 recurred, 86 did not recur) underwent computer-assisted histopathological analysis. Increasing tumor width (hazard ratios (HR): 1.17, P=0.01) and digital area (HR: 1.08, P<0.01) were significantly associated with worse recurrence-free survival, whereas non-contiguous conformation (HR: 0.57, P=0.05) was significantly associated with better recurrence-free survival. The novel histopathological classifier composed of digital area, conformation, and baseline variables effectively distinguished recurrent cases from non-recurrent cases (AUC: 0.733, 95% confidence interval (CI): 0.647-0.818), compared to the baseline classifier alone (AUC: 0.635, 95% CI: 0.545-0.724). Primary tumor cross-sectional area, width, and conformation measured via computer-assisted analysis may help identify high-risk patients with stage IB melanoma.Modern Pathology advance online publication, 21 July 2017; doi:10.1038/modpathol.2017.64.

Circular RNAs are a novel class of non-coding RNAs with functions that remain poorly characterized in normal and pathological conditions. CDR1as is a non-canonical circRNA observed to act as a sponge for miR-7 in brain tissues. Analysis of RNA-seq data of melanocytes and melanoma cell lines and short-term cultures revealed loss of CDR1as expression as a hallmark of melanoma cells. We confirmed silencing of CDR1as in melanoma cells and tissues by RT-qPCR using divergent primers. Clinically, we observed CDR1as loss associated with metastatic progression and poor patient outcomes in a cohort of fresh-frozen melanoma tissue samples. Depletion of CDR1as in melanoma cell lines enhanced invasion in vitro and lung metastasis in vivo, demonstrating functional significance of CDR1as silencing. Surprisingly, CDR1as depletion had no clear effect on miR-7 activity in melanoma cells, and miR-7 inhibition was insufficient to rescue CDR1as silencing-induced invasion. Moreover, GSEA analyses of proteomic profiling of melanoma cells depleted of CDR1as revealed reductions of proteins involved in oxidative phosphorylation (OXPHOS) and mitochondrial function, suggesting CDR1as loss may alter metabolism of melanoma cells. Mining of CLIP-Seq data sets and subsequent RIP-PCR revealed direct interactions of CDR1as with the IGF2BP family of proteins and TAR

Background: There are no predictive biomarkers of ipilimumab (IPI) toxicity. Of metastatic melanoma (MM) patients (pts) receiving IPI (3mg/kg), 35% require systemic therapies to treat immune-related adverse events (irAEs) and 20% must terminate treatment (Horvat et al., JCO 2015). Here we tested the hypothesis that a pre-existing autoantibody (autoAb) profile is predictive of IPI irAEs. Methods: We measured autoAb levels in pre- and post-treatment sera from mm pts who received IPI (3mg/kg) monotherapy on a proteome microarray containing ~20,000 unique full-length human proteins (HuProt array, CDI Laboratories). Clinical data were prospectively collected with protocol-driven follow-up. IrAEs were categorized by CTCAE guidelines as none (grade 0), mild (grade 1-2), or severe (grade 3-4). AutoAb levels were standardized using median quantile normalization and considered positive hits if > 2-SD above the peak array signal and differed by >=2 fold with p < 0.05 between toxicity groups (Non-parametric Analysis/Wilcox test). Results: Seventy-eight sera from 37 mm pts were analyzed. Antibodies against CTLA-4 were significantly elevated post IPI treatment (p < 0.0001), validating the assay. The pre-treatment levels of 190 IgG autoAbs were significantly diferent in pts who experienced irAEs (n = 28) compared to those with no irAEs (n = 9). Comparison of severe irAE (n = 9) and no irAE (n = 9) groups revealed 129 IgG autoAbs that significantly differed in pre-treatment sera. Localization and pathway analysis (UniProt, KEGG, Reactome) showed 81/190 (43%) of the autoAbs targeted nuclear and mitochondrial antigens and were enriched in metabolic pathways (p = 0.015). AutoAbs associated with irAEs did not correlate with treatment response. Conclusions: AutoAbs to antigens enriched in metabolic pathways prior to treatment may predict IPI-induced toxicity in MM. The subcellular localization of targeted antigens could explain the autoimmune toxicities associated with IPI. Studies in larger cohorts and in pts receiving other checkpoint inhibitors and/or combination therapies are essential to determine the validity of the data. If validated, our results would support the discovery of the first toxicity predictor in cancer immunotherapy

Background: Recently, tumor mutation burden (TMB) has been shown to increase the presentation of neoantigens that stimulate immune tumor recognition, resulting in improved immunotherapy (IT) outcomes in melanoma and other cancers. As melanoma is highly immunogenic, here we tested whether TMB associates with immune recognition during tumor progression, hence impacting melanoma overall survival (OS), independently of IT treatment. Methods: We have generated somatic mutation data from 314 IT-naive metastatic melanomas from The Cancer Genome Atlas (TCGA). In the TCGA cohort, TMB has been calculated for 210 genes (200GS) previously established from TMB studies of anti-CTLA4 and anti-PD1/PD-L1 IT. For validation, we have sequenced exonic regions of 20 genes (20GS) with the highest TMB among 200GS in 89 IT-naive metastatic melanomas ascertained at New York University Langone Medical Center. The TMB was defined using total number of somatic, non-synonymous mutations in either 200GS (TCGA discovery) or 20GS (validation), respectively. For discovery and validation cohorts, OS from primary diagnosis of samples with high TMB was compared against low TMB, using thresholds established in previous studies. Results: We found that total TMB predicts better OS (p = 0.03, HR = 2.64) in TCGA melanomas. Restricting the analysis only to the established 200GS, this association became more significant in all patients (p = 0.01, HR = 2.67) as well as in patients without IT (p = 0.01, HR = 2.67). In the validation stage of 89 melanomas without prior IT treatment, a high TMB in a subset of 20GS accurately determined favorable OS (p = 0.02, HR = 2.69) and confirmed TCGA observations from the 200GS. Conclusions: Here we show, for the first time, that in addition to IT, high TMB predicts more favorable OS in patients that never received IT, potentially serving as a novel marker of prognosis of melanoma and likely other immunogenic tumors at early stages. In addition, our study suggests that TMB test can be robust when applied to only a small subset of genes that trigger significantly higher immunogenicity. This may also eventually assist with accurate sub-selection of early stage patients likely to respond to IT regimens

Background: Unlike other solid tumors, the impact of primary HS on melanoma survival and response to systemic therapy is not well studied. Nodular melanoma (NM) has a worse prognosis than superficial spreading melanoma (SSM), which is usually attributed to thicker primary tumors. Herein, we examine the hypothesis that HS might have an impact on MSS independent of thickness and that NM and SSM exhibit different mutational landscapes that associate with response to checkpoint inhibitor immunotherapy (IT) and BRAF targeted therapy (TT) in the metastatic setting. Methods: Primary NM and SSM patients prospectively enrolled at NYU (2002 - 2016) were compared to the most recent SEER cohort (1973 - 2012) and analyzed with respect to MSS. Next-Generation Sequencing (NGS) was performed on a subset of matched tumor-germline pairs, allowing a comparison of the mutational landscape between NM and SSM. In the metastatic setting, survival analyses were used to compare outcomes and responses to treatment across HS. Results: The NYU cohort of 1,621 patients with either NM (n = 510) or SSM (n = 1,111) was representative of the analogous SEER cohort (21,339 NM, 97,169 SSM), with NM presenting as thicker, more ulcerated, and later stage (all p < 0.001). Among the NYU cohort, NM was found to have lower rates of TIL (p = 0.047), higher mitotic index (p < 0.001), and higher rates of NRAS mutation (p < 0.001). In multivariate Cox models, NM was a significant predictor of worse MSS, independent of thickness and stage (p = 0.01). NM had a significantly lower mutational burden across the exome (p < 0.001). Some of the most under-mutated genes noted in NM were NOTCH4, BCL2L12 and RPS6KA6 (all p < 0.01). Among patients treated with TT (n = 56), NM remained a significant predictor of worse MSS (p = 0.004). However, there was no difference in response to IT. Conclusions: NM and SSM show divergent mutational patterns which may contribute to their different clinical behaviors and responses to BRAF targeted therapy. More studies are needed to better understand the key molecular and cellular processes driving such differences. Integration of HS data into prospective clinical trial reporting is needed to better assess its impact on response to treatment

Background: Efforts to identify targeted therapies that can improve treatment outcome in metastatic ALM have been unsuccessful. In a previous genomic screening, we identified copy number amplification of the histone methyltransferase EZH2 in 47% of ALM cases, a higher frequency than previously reported in cutaneous melanomas (CM) (5%). Here, we tested the hypothesis that increased EZH2 expression contributes to ALM progression and may confer selective sensitivity to EZH2 inhibition. Methods: EZH2 expression was examined by immunohistochemistry (IHC) in 51 primary (21 stage I, 13 Stage II and 17 Stage III) and 23 metastatic (11 in transit, 8 nodal and 4 visceral) ALM cases with extensive clinicopathological data and protocol-driven follow up. Colony formation and cell proliferation was assessed following treatment of ALM and CM cell lines with three EZH2 inhibitors, including GSK126, currently in clinical trials. The effect of GSK126 on H3K27me3 and downstream EZH2 targets was analyzed by western blotting. Results: EZH2 is commonly overexpressed in both primary (30/51; 65%) and metastatic (20/23; 87%) ALM cases, with a significant increase in mean IHC score between primary and metastatic tumors (1.9 vs. 2.7, respectively, p = 0.047). EZH2 expression increased in 6/10 metastatic ALM tumors compared to their matched primary tumors. ALM tumors with EZH2 gene amplification showed increased EZH2 protein expression; however more cases showed overexpression with no amplification suggesting a potential epigenetic component of EZH2 regulation. GSK126 significantly suppressed ALM colony formation at lower doses compared to CM (1 muM vs. 5 muM, respectively). EZH2 inhibition also increased expression of the downstream tumor suppressor E-cadherin in ALM but not in CM cell lines. Finally, ALM cell lines had significantly lower basal H3K27me3 levels than CM cell lines, suggesting an additional, histone methyltransferase-independent function of EZH2 in ALM. Conclusions: Our data demonstrate thatEZH2 upregulation is common in ALM, and suggest that it may play a role in ALM's metastatic progression that requires further investigation. Selective sensitivity of ALM cell lines to EZH2 inhibitors supports the therapeutic potential of EZH2-targeted therapy in ALM