Association of gene expression profile in metastatic melanoma and survival to a dendritic cell-based vaccine

Abstract

9002 Background: Emerging data suggests that features of the melanoma tumor microenvironment may determine the clinical outcome to immunotherapies. We recently have observed a gene expression signature that correlated with a favorable clinical outcome in response to an IL-12-based melanoma vaccine. Increased expression of chemokine genes and T cell transcripts, and decreased expression of genes associated with aggressive tumor biology, were observed in the favorable group. To determine whether these patterns were reproducible, gene expression profiling was performed from an independent vaccine clinical trial. Methods: Patients with advanced melanoma were treated with autologous, mature monocyte-derived dendritic cells loaded with a combination of melanoma antigen peptides. Pretreatment biopsies were cryopreserved for RNA extraction and gene expression profiling. Patients were categorized into “long survival” (> 24 months) or “short survival” outcomes. Supervised hierarchical clustering was performed to identify genes differentially expressed in the two outcome groups. Results: RNA that passed quality control was obtained from 17 stage IV patients, 5 with a short survival and 12 with a long survival. 408 genes were differentially at least 2- fold. Consistent with previous observations, tumors from favorable outcome patients expressed higher levels of several T cell-specific genes, including Thy1 and CD28; chemokines, including CCL19, CXCL12, and CXCL14; and other immune genes, including LTβ, IL-1R, IFNαR2, IL27R, CD69, and FcRs. Conversely, tumors from unfavorable outcome patients expressed higher levels of pro- angiogeneic genes, including Flt1; anti-apoptotic genes, including SerpinH1 and Serpine1; and multiple collagens. Conclusions: Our results confirm that a subset of transcripts expressed in melanoma metastases may be useful as a predictive biomarker for response to melanoma vaccines. The categories of genes identified point toward new opportunities for overcoming resistance mechanisms. Future studies should integrate gene expression profiling of pre-treatment biopsies as a stratification or enrichment factor in immunotherapy trials. No significant financial relationships to disclose.

An immunologic portrait of cancer

“…These findings suggest that tumors of the inflamed phenotype do not discriminate among various components of the immune response as they can sustain immune effector and immune regulatory functions simultaneously; it is probably the overall balance between the two vectors that determines the ultimate fate of these cancers individual cancers. More recently, the same group identified a similar pattern predictive of favorable outcome in patients vaccinated with dendritic cell loaded with multiple tumor antigen-derived peptides [ 28 ]. Similar findings were observed by GSK-Biologics in non small cell lung cancer and melanoma patients undergoing vaccination with MAGE-3 protein [ 29 , 30 ].…”

Identifying tumor antigens and immuno‐subtyping in colon adenocarcinoma to facilitate the development of mRNA vaccine

“… 51 Inspired by these results, the following study has analyzed the immune gene profiles of the pre‐treatment tumor biopsies from a second vaccine trial, which revealed a positive association between inflamed gene expression profiles and clinical benefits from a DC‐based vaccine. 52 Furthermore, a clinical trial that validating the effectiveness of therapeutic cancer vaccine Theratope in patients with ovarian, breast or colorectal cancer demonstrated the correlations between pre‐existing inflammation and clinical benefits. However, the patients were not screened or stratified according to their pre‐vaccination immune status in the subsequent phase III clinical trial.…”

Primer on tumor immunology and cancer immunotherapy

“…In one preclinical study of metastatic melanoma, the expression of a subset of chemokines were associated with CD8 + T cell infiltration [ 34 ]. In patients with metastatic melanoma, the expression of T cell markers and chemokines correlated with response to a DC-based vaccine [ 35 ]. Likewise, a pro-inflammatory gene expression profile within the tumor microenvironment was associated with survival following administration of a protein-based vaccine in patients with metastatic melanoma [ 36 ].…”