Ahmad A. Tarhini, MD, PhD (Clinical and Translational Science)
Walter Storkus, PhD (Basic Science)
Lisa H. Butterfield, PhD (Basic Science)
John M. Kirkwood, MD (Clinical Science)
William LaFramboise, PhD (Basic Science)
The current paradigm of adjuvant therapy in melanoma involves the indiscriminate treatment of all patients clinically considered at high risk for melanoma recurrence and mortality, despite data showing that only a small proportion of patients will benefit. The primary goal of Project 1 is to identify baseline and/or early on-treatment predictive biomarker(s) capable of classifying patients according to the degree of benefit they will receive from treatment with ipilimumab or interferon alpha (IFNα). Secondary goals are to identify biomarkers of significant prognostic value capable of identifying patients at highest risk of recurrence and mortality. During the last SPORE funding period, we tested the prognostic and predictive value of autoimmunity and candidate biomarkers in relation to IFNα as part of two Eastern Cooperative Oncology Group (ECOG)-led adjuvant trials (E1697 and E1694). We also studied biomarkers predictive of CTLA4 blockade therapeutic benefit within metastatic and neoadjuvant trials that laid the groundwork for this modified continuation of Project 1. We identified a series of potentially therapeutically predictive biomarkers of a pro-inflammatory immune response and of immunosuppression in both tumor tissue and in circulating blood, which we will evaluate simultaneously due to the common systems biology, in our quest for a predictive biomarker model.
The renewal Project 1 will be nested within the ongoing ECOG E1609 trial that is testing adjuvant ipilimumab (at high dose and standard dose) versus IFNα in patients with operable stage IIIB/C and M1a/b melanoma who are at high risk for recurrence and death. The large sample size (n=1600) and diverse biorepository for E1609 make it an ideal platform through which to evaluate these interrelated markers for their immunotherapeutic predictive value with regard to ipilimumab in reference to IFNα, and disease prognostic value, as assessed individually and in combination. We will also develop a therapeutic predictive model that links circulating markers of the pro-inflammatory response and immunosuppression with the tumor microenvironment.
We hypothesize that a baseline pro-inflammatory biomarker signature will be predictive of therapeutic benefit. We expect to discover unique biomarker differences between ipilimumab and IFNα due to the different mechanisms of action but overlapping predictive models, given the common pro-inflammatory biological impact. Thus, our renewal Project 1 will assess and validate candidate biomarkers predictive of relapse-free survival (RFS, primary outcome) and overall survival (OS, secondary outcome) of melanoma patients in the context of E1609 and will develop and validate a predictive model for clinical outcome using these markers.
Specific Aim 1: Circulating biomarkers. Based on preliminary data from the last funding period, we will examine:
- (1a) Cellular populations. We will test the hypothesis that baseline and/or early on-treatment IFNg+CD4+ and IFNg+CD8+ antigen specific T cell immunity as well as specific host suppressor elements (defined populations of regulatory T cells and MDSC) predict therapeutic benefit (RFS and OS). We will also assess the overall lymphocyte count as a potential predictive marker.
- (1b) Serum proteins. We will test the hypothesis that baseline and/or early on-treatment pro-inflammatory cytokine and chemokine profiles predict therapeutic benefit (RFS and OS). Secondarily, we will examine candidate serum biomarkers known to be associated with the host cellular immune response (in Aim 1a) to confirm preliminary findings from the last funding period.
Specific Aim 2: Tumor and tumor microenvironment biomarkers. Based on multiple recent reports, we will first test the hypothesis that a pretreatment, pro-inflammatory, tumor microenvironment (high baseline expression levels of immune-related genes) predicts therapeutic benefit (RFS and OS). As secondary sub-aims, we will test the association of the tumor mutational status (BRAF, NRAS, and WT status for both) and clinical outcome. We will also assess the methylation levels of immune-related genes.
Specific Aim 3: Prediction model development and validation. Using known clinical covariates and significant markers identified in Aims 1 and 2, we will use the entire data set from E1609 to develop and validate models capable of identifying which patients belong to different therapeutic predictive groups.
Project 1 has the potential to significantly impact the field of adjuvant immunotherapy, transforming our current model of care from a nonspecific population approach to a personalized and targeted therapeutic strategy. Further, our results will provide urgently needed mechanistic insights into the complex interactions between host and tumor by linking elements of inflammation and immunosuppression in circulation with the tumor microenvironment, ultimately leading to improved therapeutic targeting and patient outcomes. Our findings should provide the foundation for assessing these biomarkers in other stages of melanoma, and for other immunotherapies (e.g., IL-2), other immune checkpoint inhibitors, and more specific (vaccine) modalities.