Previously, we obtained gene expression profiles from primary prostate tumors resected from 33 African-American and 36 European-American men. Analyzing the resulting datasets, we identified gene expression differences that portray the existence of a distinct tumor microenvironment for these two patient groups. Many of the differently expressed genes were immune-regulatory. Perhaps most significant was the presence of a distinct interferon signature in many of the African-American tumors. As later discovered by us, this signature is almost identical with an interferon-related DNA damage resistance signature (IRDS) that causes resistance to chemotherapy and radiation. Therefore, the distinct immunobiology of prostate tumors in African-American men may not only influence their response to immune-based therapies, as was proposed by us, but may also make them more resistant to standard therapy including radiation therapy because of IRDS. In addition, IRDS has been linked to increased cancer metastasis and appears to be partly induced by interactions between tumor cells and stromal cells. Thus, IRDS may promote both the metastatic process and therapy resistance. Subsequently, we completed the analysis of a second dataset with gene expression profiles from African-American and European-American tumors, which confirmed that IRDS is significantly more common in tumors from the African-American patients than in tumors from European-American patients. To further explore the origin of IRDS, we also interrogated the expression profiles of isolated primary human prostate epithelial cells from 14 African-American patients and 13 European-American patients for the presence of IRDS. This analysis led to the observation that IRDS was present in these cultured cells in 5 out of 14 (36%) isolates from African-American patients versus 2 out of 13 (15%) isolates from European-American patients, thus persisting in cancer cells after being removed from the tumor microenvironment. The latter finding is consistent with previous observations that interferon-induced IRDS can persist and become intrinsic after withdrawal of interferon. We think that the heightened prevalence of IRDS in cancer cells from African-American patients could be clinically very significant and warrants further investigations into the origin of this signature, and also how this signature can be targeted. These ongoing investigations are pursued in two collaborations. The multi-institutional collaboration between investigators at the Cleveland Clinic, the University of Chicago, and the NCI will investigate the mechanisms of IRDS induction in human prostate cancer cells and candidate therapies to target IRDS, and the relationship between IRDS in prostate tumors and disease outcome in African-American and European-American patients. Here, we are taking advantage of a large tissue repository (including 700 prostate tumors from African-American patients with long-term follow-up) at the Cleveland Clinic. In a second collaboration with Ludmila Prokunina-Olsson at the NCI, we will explore the relationship between IRDS and a recently discovered gene variant that is linked to the expression of a novel interferon gene. Preliminary data suggest that this variant is linked to the increased prevalence of IRDS in the prostate tumors from African-American patients. In an additional research approach, we are testing the hypothesis that the immune signature in African-American tumors is immune suppressive and involves recruitment of immune regulatory cells into the cancerous prostate. As part of a pilot study, we have started with the evaluation of immune cell profiles in peripheral blood mononuclear cells from African-American and European-American patients and age-matched population-based controls. This investigation will focus on immune cell subpopulations that have immune-regulatory functions in cancer biology [myeloid-derived suppressor cells, dendritic cells, regulatory T cells (Treg), regulatory B cells (Breg), gamma/delta T cells, and Th2 cells]. It is the hypothesis of this research that immune cell subpopulations that have an immune-regulatory function in cancer biology are different in abundance between African-American and European-American men. According to our hypothesis, those sub-populations with an immune suppressive activity are more abundant in African-American patients. If we would find that these immune cell profiles are in fact different between African-American and European-American patients, we will analyze tumors for differences in these cells and will also profile plasma samples for cytokine and chemokine levels that are typically produced by the same cells. In 2013, we completed the analysis of buffy coats from 25 African-American and 25 European-American patients and 25 African-American and 25 European-American population-based controls. These results revealed only marginal differences in the studied immune cell populations by race/ethnicity, suggesting that significant differences in blood-based immune profiles representing immune cell subpopulations with immune-regulatory functions do not exist between African-American and European-American men. In contrast, several significant differences where observed between cases and controls in the combined analysis without stratification by race/ethnicity. The presence of an interferon gene signature in prostate tumors also suggests a possible involvement of either a viral infection in disease pathology or the reactivation of endogenous retroviruses in the tumor microenvironment. This hypothesis was further supported by our finding that the interferon signature in prostate tumors coincides with a gene signature of retroviral activation. Thus, we started a project exploring the presence of viral infections and the reactivation of endogenous retroviruses in tumors from African-American and European-American patients. In a pilot study, we investigated tumor samples by immunohistochemistry and peripheral blood mononuclear cell (PBMC) isolates by quantitative PCR for expression of a family of endogenous retroviruses, HERV-K. This project is a collaboration with Drs. Feng Wang-Johanning at MD Anderson Cancer Center and Sharon Glynn at the University of Galway, Ireland. The findings showed that the HERV-K envelope protein can be detected in prostate tumors by immunohistochemistry and is more commonly detected in tumors from African-American men, indicating that HERV-K encoded proteins are expressed in these tumors, and that HERV-K gag message expression is elevated blood mononuclear cells from prostate cancer patients when compared with a non-cancer control population. As a next step, we started to analyze blood samples from men prospectively recruited into the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO). We analyzed 30 blood samples blinded to the case-control status (15/15). Blood samples have been drawn from men that developed prostate cancer at and prior to diagnosis and from men without the disease. The results showed that HERV-K expression is more commonly detected in blood samples from men with prostate cancer at or prior to diagnosis than in men without the disease, corroborating findings from our first study. These encouraging results will now be followed up in a much larger investigation supported by PLCO consisting of blood samples from men without disease, with localized disease, and with advanced disease. Again, we will analyze blood drawn at time of diagnosis and up to two years prior to diagnosis to corroborate our preliminary observations that HERV-K expression in PBMCs is an early disease detection marker that is elevated prior to disease diagnosis.
Showing the most recent 10 out of 18 publications
