Search

Supplementary data on genomics

Clinical characteristics of melanoma biopsies used in CYTOF

VISTA expression on immune cells in tumor dissociates by flow cytometry

Supplementary table with clinical data

Purpose:Immunotherapies targeting costimulating and coinhibitory checkpoint receptors beyond PD-1 and CTLA-4 have entered clinical trials. Little is known about the relative abundance, coexpression, and immune cells enriched for each specific drug target, limiting understanding of the biological basis of potential treatment outcomes and development of predictive biomarkers for personalized immunotherapy. We sought to assess the abundance of checkpoint receptors during melanoma disease progression and identify immune cells enriched for them.Experimental Design: Multiplex immunofluorescence staining for immune checkpoint receptors (ICOS, GITR, OX40, PD-1, TIM-3, VISTA) was performed on 96 melanoma biopsies from 41 treatment-naïve patients, including patient-matched primary tumors, nodal metastases, and distant metastases. Mass cytometry was conducted on tumor dissociates from 18 treatment-naïve melanoma metastases to explore immune subsets enriched for checkpoint receptors.Results:A small subset of tumor-infiltrating leukocytes expressed checkpoint receptors at any stage of melanoma disease. GITR and OX40 were the least abundant checkpoint receptors, with +/CD103+/CD8+) were enriched for TIGIT (>70%) and other coinhibitory but not costimulatory receptors. The proportion of GITR+ T cells decreased from primary melanoma (>5%) to lymph node (P = 0.04) and distant metastases (P = 0.0005).Conclusions:This study provides the first comprehensive assessment of immune checkpoint receptor expression in any cancer and provides important data for rational selection of targets for trials and predictive biomarker development.

Antibodies used in CYTOF experiment

Purpose: Therapeutic blockade of immune checkpoints has revolutionized cancer treatment. Durable responses, however, occur in less than half of those treated, and efforts to improve treatment efficacy are confounded by a lack of understanding of the characteristics of the cells that initiate antitumor immune response.Patients and Methods: We performed multiparameter flow cytometry and quantitative multiplex immunofluorescence staining on tumor specimens from immunotherapy-naïve melanoma patients and longitudinal biopsy specimen obtained from patients undergoing anti–PD-1 therapy.Results: Increased numbers of CD69+CD103+ tumor-resident CD8+ T cells were associated with improved melanoma-specific survival in immunotherapy-naïve melanoma patients. Local IL15 expression levels strongly correlated with these tumor-resident T-cell numbers. The expression of several immune checkpoints including PD-1 and LAG3 was highly enriched in this subset, and these cells significantly expanded early during anti–PD-1 immunotherapy.Conclusions: Tumor-resident CD8+ T-cell numbers are more prognostic than total CD8+ T cells in metastatic melanoma. In addition, they are likely to initiate response to anti–PD-1 and anti–LAG-3 treatments. We propose that the immune profile of these cells prior to treatment could inform strategies for immune checkpoint blockade. Clin Cancer Res; 24(13); 3036–45. ©2018 AACR.

These are the supplementary data figure legends (S1-S4)

Purpose: Immunotherapies targeting costimulating and coinhibitory checkpoint receptors beyond PD-1 and CTLA-4 have entered clinical trials. Little is known about the relative abundance, coexpression, and immune cells enriched for each specific drug target, limiting understanding of the biological basis of potential treatment outcomes and development of predictive biomarkers for personalized immunotherapy. We sought to assess the abundance of checkpoint receptors during melanoma disease progression and identify immune cells enriched for them. Experimental Design: Multiplex immunofluorescence staining for immune checkpoint receptors (ICOS, GITR, OX40, PD-1, TIM-3, VISTA) was performed on 96 melanoma biopsies from 41 treatment-naïve patients, including patient-matched primary tumors, nodal metastases, and distant metastases. Mass cytometry was conducted on tumor dissociates from 18 treatment-naïve melanoma metastases to explore immune subsets enriched for checkpoint receptors. Results: A small subset of tumor-infiltrating leukocytes expressed checkpoint receptors at any stage of melanoma disease. GITR and OX40 were the least abundant checkpoint receptors, with 70%) and other coinhibitory but not costimulatory receptors. The proportion of GITR+ T cells decreased from primary melanoma (>5%) to lymph node ( Conclusions: This study provides the first comprehensive assessment of immune checkpoint receptor expression in any cancer and provides important data for rational selection of targets for trials and predictive biomarker development.

Purpose: Therapeutic blockade of immune checkpoints has revolutionized cancer treatment. Durable responses, however, occur in less than half of those treated, and efforts to improve treatment efficacy are confounded by a lack of understanding of the characteristics of the cells that initiate antitumor immune response. Patients and Methods: We performed multiparameter flow cytometry and quantitative multiplex immunofluorescence staining on tumor specimens from immunotherapy-naïve melanoma patients and longitudinal biopsy specimen obtained from patients undergoing anti–PD-1 therapy. Results: Increased numbers of CD69+CD103+ tumor-resident CD8+ T cells were associated with improved melanoma-specific survival in immunotherapy-naïve melanoma patients. Local IL15 expression levels strongly correlated with these tumor-resident T-cell numbers. The expression of several immune checkpoints including PD-1 and LAG3 was highly enriched in this subset, and these cells significantly expanded early during anti–PD-1 immunotherapy. Conclusions: Tumor-resident CD8+ T-cell numbers are more prognostic than total CD8+ T cells in metastatic melanoma. In addition, they are likely to initiate response to anti–PD-1 and anti–LAG-3 treatments. We propose that the immune profile of these cells prior to treatment could inform strategies for immune checkpoint blockade. Clin Cancer Res; 24(13); 3036–45. ©2018 AACR.

Purpose: Anti-PD-1 therapy has revolutionized the treatment and improved the survival of stage IV melanoma patients. However, almost half of the patients fail to respond due to immune evasive mechanism. A known mechanism is the downregulation of major histocompatibility complex (MHC) class I expression, which prevents T cell recognition of the tumor. This study determined the relationship between natural killer (NK) cell numbers and clinical response to anti-PD-1 therapy in metastatic melanoma. Experimental Design: Twenty-five anti-PD-1 treated metastatic melanoma patients were categorized into responders (complete response (CR)/partial response (PR)/stable disease (SD) ≥ 6 mo, n = 13) and non-responders (SD < 6 days/progressive disease (PD), n = 12) based on RECIST response. Whole transcriptome sequencing and multiplex immunofluorescent staining were performed on pre-treatment and on a subset of early during treatment tumor samples. Spatial distribution analysis was performed on multiplex immunofluorescent images to determine the proximity of NK cells to tumor cells. Flow cytometry was used to confirm NK phenotypes in lymph node metastases of treatment naive melanoma patients (n = 5). Cytotoxic assay was performed using NK cells treated with anti-PD-1 or with isotype control and co-cultured with 3 different melanoma cell lines and with K562 cells (leukemia cell line). Results: Differential expression analysis identified nine upregulated NK cell specific genes (adjusted p < 0.05) in responding (n = 11) versus non-responding patients (n = 10). Immunofluorescent staining of biopsies confirmed a significantly higher density of intra- and peri-tumoral CD16+ and granzyme B + NK cells in responding patients (p < 0.05). Interestingly, NK cells were in closer proximity to tumor cells in responding PD-1 treated patients compared to non-responding patients. Patients who responded to anti-PD-1 therapy, despite MHC class I loss had higher NK cell densities than patients with low MHC class I expression. Lastly, functional assays demonstrated PD-1 blockade induces an increase in NK cells' cytotoxicity. Conclusions: A higher density of tumoral NK cells is associated with response to anti-PD-1 therapy. NK cells may play an important role in mediating response to anti-PD-1 therapy, including in a subset of tumors downregulating MHC class I expression.

Background: Immunotherapies targeting PD-1/PD-L1 and CTLA4 have revolutionized the treatment of advanced melanoma. Combining the two therapeutics increases the response rates compared to either treatment alone. However, this increased efficacy is accompanied by a higher incidence of severe immune-related adverse events (irAEs). Metrics of the intestinal microbiome are associated with cancer patients’ responses to immunotherapy but the value of microbiome metrics as predictors for irAEs, are unknown. In an effort to reduce irAEs during combination neoadjuvant therapy, the OpACIN-neo trial (Rozeman et al. ESMO 2018) was initiated wherein Stage III melanoma patients were treated with 2 doses each of ipilimumab and nivolumab in the neoadjuvant setting, according to three dosing regimen. Involved lymph nodes were resected after 6 weeks. Aims: (1) To determine whether intestinal microbial components are associated with response to antiPD1/anti-CTLA4 immunotherapy in the neoadjuvant setting or with the development of severe irAEs. (2) To determine the effects of antiPD1/anti-CTLA4 immunotherapy on the microbiome over the 6 week course of treatment. Methods: Of Melanoma Institute Australia patients enrolled in the OpACIN-neo trial (n=38), 68% of patients experienced complete, or near complete pathological responses and 61% experienced at least one irAE of grade 3 or more. Faecal microbiomes at baseline, and at resection (6 weeks immunotherapy), were analyzed using 16S ribosomal gene and metagenomic sequencing and the results compared with patient response and development of G3-G5 irAEs. Results: In baseline samples, Inverse Simpson’s diversity index indicated significantly lower overall microbial diversity in non-responders (p=.014), as well as those who went on to experience severe irAEs (p=.002). Importantly, the group of patients who were both non-responders and experienced severe irAEs had the lowest microbial diversity of all patients (p=.0033). Specific taxa associated with irAEs are distinct from those described for response. The 6 week course of immunotherapy led to a slight increase in microbial diversity but few specific taxa were observed to be significantly altered between the timepoints. Conclusions: The findings suggest that not only are patients with extremely low microbial diversity predisposed to develop severe irAEs but they are also unlikely to respond to combination immunotherapy. Thus, microbial diversity may delineate patients who are likely to have poor outcomes and would benefit from microbial modulation. Citation Format: Marcel Batten, Erin R. Shanahan, Ines P. Silva, Chandra Adhikari, Jordan Conway, Annie Tasker, Alexander M. Menzies, James S. Wilmott, Robyn P. Saw, Andrew J. Spillane, Kerwin F. Shannon, Christian U. Blank, Andrew J. Holmes, Richard A. Scolyer, Georgina V. Long. Low intestinal microbial diversity is associated with severe immune-related adverse events and lack of response to neoadjuvant combination antiPD1, anti-CTLA4 immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2822.

Background A myriad of novel monoclonal antibody based immunotherapies targeting co-stimulating and co-inhibitory receptors have entered clinical trials in melanoma with the aim of increasing response rates and overcoming resistance to standard anti-CTLA-4 and anti-PD-1 immunotherapy. However, little is known about the abundance, co-expression and immune cells enriched for each specific drug target in the various stages of melanoma progression. Therefore, we sought to assess the relative abundance of checkpoint receptors and their expression during melanoma disease progression, as well as the immune cells enriched for each of these molecules. Methods - Multiplex immunofluorescence staining for immune checkpoint receptors (ICOS, GITR, OX40, PD-1, TIM-3, and VISTA) was performed on 95 melanoma biopsies from 41 melanoma patients, including patient matched biopsies for primary, regional lymph node or distant metastases. - Mass cytometry was performed on leukocytes isolated from 18 treatment-naïve melanoma tumors to explore immune subsets enriched for many of the checkpoint receptors currently targeted in clinical trials. Results & Conclusions GITR and OX40 were the least abundant checkpoint receptors in melanoma (p70%) and other co-inhibitory receptors but not co-stimulatory receptors. The proportion of GITR+ T cells decreases from primary melanoma (>5%) to patient matched lymph node ( Note: This abstract was not presented at the meeting. Citation Format: Jarem Edwards, Annie Tasker, Inês Pires da Silva, Camelia Quek, Benjamin M. Allanson, Robyn P. M Saw, John F. Thompson, Alexander M. Menzies, Umaimainthan Palendira, James S. Wilmott, Georgina V. Long, Richard Scolyer. Dynamics of T-cell checkpoint receptor profiles during melanoma progression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3246.

Melanoma patients (pts) with liver mets have a lower response rate (RR), and shorter progression-free (PFS) and overall survival (OS) compared to pts without liver mets when treated with anti-PD-1 (PD1) therapy. The liver microenvironment (ME) induces T-cell tolerance through the interaction of T cells with liver sinusoidal endothelial cells. To explore this further we compared clinicopathological features, circulating cytokines and tumor gene expression (GE) profiles of pts with and without liver mets treated with PD1 combined with ipilimumab (PD1+IPI), and treated with BRAF targeted therapy (TT). Demographics, disease characteristics and outcome data were collected from 140 pts treated with PD1+IPI and from 76 pts treated with BRAF+/-MEKi. Tumor-infiltrating lymphocytes (TILs) immunoreactivity score (TILs density x % of tumor with TILS), % of tumor content, necrosis and fibrosis were assessed by immunohistochemistry in liver and lung samples. Pre-treatment circulating cytokines and tumor GE data (RNA seq) were compared between pts with and without liver mets. In pts treated with IPI+PD1, liver mets had the lowest tumor regression (med -7%) compared to all other sites of disease (med -66%), while in TT-treated pts the response was similar across all sites of disease. Pts with liver mets (n=39) had lower RR (44% v 75%), and shorter median PFS and OS (p Pts with liver mets display distinct clinicopathological features, distinct circulating cytokines and melanoma GE profiles, and are less responsive to PD1+IPI compared to pts without liver mets. The levels of Eotaxin-2, IP-10 and IL-8 are higher in melanoma pts with liver mets compared to pts without liver mets, similar to what is seen in pts with colon cancer liver mets. Liver mets’ ME may hold unique immunosuppressive mechanisms that are amenable to therapeutic targeting. Citation Format: Ines Silva, Annie Tasker, Camelia Quek, Robert Rawson, Su Yin Lim, Kevin Wang, Jordan Conway, Rebecca Velickovic, Tasnia Ahmed, Serigne Lo, Jean Yang, Helen Rizos, James S. Wilmott, Richard A. Scolyer, Alexander M. Menzies, Georgina V. Long. Liver metastases (mets) induce systemic immunosuppression and immunotherapy resistance in metastatic melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 975.

Summary Cancer immunotherapies provide survival benefits in responding patients, but many patients fail to respond. Identifying the biology of treatment response and resistance are a priority to optimize drug selection and improve patient outcomes. We performed transcriptomic and immune profiling on 158 tumor biopsies from melanoma patients treated with anti-PD-1 monotherapy (n = 63) or combined anti-PD-1 and anti-CTLA-4 (n = 57). These data identified activated T cell signatures and T cell populations in responders to both treatments. Further mass cytometry analysis identified an EOMES+CD69+CD45RO+ effector memory T cell phenotype that was significantly more abundant in responders to combined immunotherapy compared with non-responders (n = 18). The gene expression profile of this population was associated with longer progression-free survival in patients treated with single agent and greater tumor shrinkage in both treatments.