Scott I. Abrams, PhD
Department of Immunology
Roswell Park Cancer Institute
Elm and Carlton Streets
Buffalo, NY 14263
Tel: (716) 845-4375
Fax: (716) 845-8906
Figure 1. Model for T Cell-Mediated Immunoselection of Aggressive Tumor Escape Variants. A neoplastic mass or lesion is generally comprised of tumor cell subpopulations of diverse antigenic, molecular and biologic properties (shown in maroon and red), including apoptotic responsiveness, as well as host stromal elements (shown in blue). Such tumor heterogeneity is thought to result from genetic and epigenetic aberrations governing their survival, proliferation and other aspects of the neoplastic process. Nonetheless, an infiltrating antitumor CTL response (shown in green), whether initiated during host immunosurveillance or by immunotherapy, causes destruction of the Ag-bearing, lytically sensitive targets (shown in maroon) within the lesion. Lower frequencies of pre-existing or newly arising Ag-loss variants or lytically resistant neoplastic subpopulations (shown in red) escape CTL-mediated destruction and, consequently, outgrow to reconstitute a more aggressive lesion. CTL-mediated destruction of the lytically sensitive tumor cells may thus serve as a selective pressure that concurrently promotes the reciprocal enrichment of tumor escape variants (TEV) with enhanced malignant characteristics, such as apoptotic resistance. Here, we showed that Faslo-TEV emerge during CTL-tumor interactions in vivo.
Increased resistance of neoplastic subpopulations to cell death is well regarded as a major hallmark of tumor progression.
Therefore, understanding the molecular bases for tumor-cell resistance to cell death is critically important not only to improving our knowledge of cancer biology, but also the overall development of more effective cancer therapies, including immunotherapy.
To that end, our laboratory has focused on the effector-phase of an antitumor CD8+ cytotoxic T lymphocyte (CTL) response to identify relevant tumor regression mechanisms, as well as potential mechanisms underlying lytic (apoptotic) resistance.
During the past several years, our studies have led to the identification of previously unrecognized roles for the Fas/Fas ligand (FasL) system, an important death receptor-initiated pathway, in tumor immunology. Of note is the discovery that the Fas/FasL system has not only positive, but also negative consequences during host-tumor interactions.
We have shown that CTL-mediated destruction of Fas-responsive tumor cells in vivo functions as an integral effector mechanism. Moreover, we have shown that tumor eradication by Fas/FasL interactions concurrently served as a “selective pressure”, resulting in the emergence of pre-existing Fas-refractory or –resistant (Faslo) tumor escape variants (TEV), which displayed enhanced malignant ability. Using matched sets of poorly and highly malignant/metastatic populations in both mouse and human systems, we substantiated the inverse relationship between Fas sensitivity and malignant capacity.
Our findings are consistent with the hypothesis that during the process of neoplastic progression, tumor cell subpopulations down-modulate susceptibility to death receptor-initiated pathways, such as Fas. An infiltrating antitumor CTL response would then cause destruction of the antigen-bearing, lytically sensitive targets (immune selection phase). Lower frequencies of pre-existing or newly arising lytically resistant neoplastic clones would evade CTL-mediated destruction (tumor escape phase) and, consequently, outgrow to repopulate a more malignant mass (tumor progression phase). Our studies showed for the first time that Faslo-TEV emerge during CTL-tumor interactions in vivo, which are summarized in Figure 1.
Subsequently, we have identified two genes integral for this differential Fas death response: (a) interferon consensus sequence binding protein (ICSBP), also known as interferon regulatory factor-8 (IRF-8); and (b) caspase-1 of the Fas signaling pathway. IRF-8 was originally discovered by other laboratories however as an IFN-γ-inducible transcription factor essential for regulating normal myelopoiesis. Interestingly, IRF-8 deficiency has been shown to result in a CML-like syndrome in IRF-8 null mice. In humans, substantially reduced IRF-8 mRNA levels have been reported in CML patients, while the recovery of IRF-8 mRNA levels have been associated with IFN-α-induced cytogenetic clinical responses. Thus, IRF-8 has been implicated as a “tumor suppressor gene” in certain hematopoietic cancers. In our studies, it is important to emphasize that the identification of IRF-8 in solid tumor models was unveiled under pro-inflammatory conditions, which otherwise would not have been self-evident.
Future directions of our laboratory will focus on the mechanism and significance of IRF-8-dependent interactions in previously unrecognized areas of solid tumor biology and tumor immunology. Our underlying hypothesis is that IRF-8 expression plays an important role in the neoplastic process of certain solid malignancies in that its presence favors tumor regression or response to therapy, whereas its absence favors tumor progression or poorer response to therapy. Emphasis will also be on molecular mechanisms that regulate IRF-8 expression in neoplastic cells, and novel strategies to potentially optimize its induction to improve antitumor immune responses in vivo in preclinical mouse models. In summary, these studies may have important translational implications for unraveling some fundamental challenges confronting cancer immunotherapy and underscore the importance of understanding host-tumor interactions as a critical element of therapeutic efficacy.
1. We demonstrated for the first time that CTL attack in vivo led to the outgrowth of pre-existing Fas resistant/refractory tumor cell subpopulations that displayed heightened malignant ability. These studies highlighted the importance of understanding molecular or functional changes occurring within the tumor microenvironment at the level of apoptotic resistance that could impact the efficacy of adoptive or active immunotherapy.
2. Molecular characterization of matched poorly and highly malignant cell line or subline pairs led to the identification of two genes responsible for this differential Fas-mediated death response: ICSBP (also known as IRF-8) and caspase-1, as noted earlier. Therefore, in addition to Fas expression, regulation of IRF-8 may be another significant determinant of tumor regression/progression, which is the subject of future investigations.
3. In a study recently published in PNAS, we described a new mechanism of tumor escape. We examined the role of the adenosine/adenosine receptor system as a negative regulator of antitumor immune responses in vivo. Based on earlier knowledge of this system in immune regulation, we tested the hypothesis that if this pathway could be blocked (e.g., pharmacologically or via RNA interference), this would enable Ag-specific CTL to mediate higher levels of tumor regression in an otherwise adenosine-rich immune suppressive microenvironment. We demonstrated that pharmacological antagonists of A2 receptors, as well as molecular silencing of these receptors on CTL directly, significantly intensified the therapeutic efficacy of antitumor CTL.
4. Our earlier work at the NCI with ras oncogenes as potential targets for immunotherapy led for the first time to the identification of HLA-A2–restricted, CD8+ CTL epitopes reflecting ras codon 12 mutations. This finding laid the foundation for the design of peptide-based vaccines for human malignancies expressing mutated ras genes in NCI clinical trials. The identification of such mutated ras peptide sequences has culminated thus far in patent approval in Europe and is pending approval in the U.S.