The exquisite sensitivity of the prostate gland to androgenic steroids has provided a foothold for the development of systemic prostate cancer therapy for more than seventy years. A sustained strategic approach that focused on inhibiting this unique signaling pathway led to the use of androgen-deprivation and antiandrogenic therapies for Cancer Cell Previews advanced prostate cancer. These therapies continue to serve as the standard of care, although, unfortunately, antiandrogenic therapies are not curative; new approaches are needed. With the advent of targeted therapies for cancer, antiandrogenic agents have continued to form the base on which combination therapies-including those that target common oncogenic signaling activities- can be developed.

In the case of prostate cancer, this has proved particularly challenging because of the extremely heterogeneous nature of the genetic alterations that underlie this disease. A prominent molecular target for prostate cancer therapy is the PI3K-AKT signaling pathway. A recent study of 218 prostate cancer tumors showed that 42% of the primary tumors and 100% of the metastases harbored genomic aberrations in that pathway.

The best-characterized genetic alteration in this pathway is in PTEN, which has been shown to be mutated and/or exhibit loss of heterozygosity in approximately 15% of localized prostate cancer and 30% of metastatic disease. Multiple small-molecule inhibitors of PI3K-AKT signaling have been developed and tested clinically. Although the results of early clinical trials are inconclusive, the therapeutic activities of PI3K-AKT inhibitors as single agents have generally been modest in patients with advanced prostate cancer. Thus, there is considerable effort to rationally integrate PI3K-AKT inhibitors into combination therapy protocols.

In recent issues of Cancer Cell, both report on having identified reciprocal feedback regulation between AR and PTEN loss/PI3K-AKT signaling in prostate cancer. By making effective use of the PB-Cre;Ptenlox/lox mouse model and carefully annotated human prostate cancer tissue samples, these two groups of investigators have made a seminal contribution to our understanding of the regulation of growth and survival signaling in prostate cancer cells and, by extension, to the rationale for use of specific combination therapy for advanced prostate cancer. Using similar experimental approaches, the loss of PTEN function sets into motion a series of molecular events that establish a linkage between two expansive signaling networks that exert control over the growth, survival, and differentiation of prostatic epithelial cells. Activation of PI3K-AKT signaling as a result of Pten mutation in the PB-Cre;Ptenlox/lox mouse leads to suppression of AR signaling.

Transcriptome analysis revealed substantial overlap of up- and downregulated genes between intact male Pten/mice and castrated wild-type mice and also demonstrated that PTEN loss is associated with reduced AR signaling in PTEN-deficient human prostate tumors. These results, together with those of previous studies, show that the loss of PTEN function and activation of PI3K-AKT signaling plant the seeds for androgen-independent prostate cancer growth by establishing a castrate genetic program. Using both pharmacologic and genetic approaches, different mechanisms contribute to the repression of AR output. The PI3K-AKT, but not MEK signaling, is responsible for inhibiting AR signaling, and that this inhibition depends on upstream HER kinase inhibition. Using a PTEN re-expression approach, PTEN loss may suppress androgen-responsive genes through upregulation of Egr1 and c-Jun transcriptional coregulators and the catalytic subunit of Polycomb repressive complex 2, Ezh2. Thus, PTEN loss can lead to repression of AR signaling on two levels: upstream suppression of MAPK-stimulated HER kinase, and suppression/subversion of AR-mediated transcription through increased expression of transcriptional coregulators and a histone methyltransferase. Probing the castration response in PBCre; Ptenlox/lox mice, PB-MYC mice, and androgen-sensitive prostate cancer cells and analyzing a double-knockout mutant, PB-Cre; Ptenlox/lox;Arlox/Y, mouse and human prostate cancer samples led to the second crucial surprising finding-that castration or AR loss increased AKT phosphorylation.

An important note is that these two experimental approaches independently led to the identification of a reciprocal negative-feedback signal in thePB-Cre;Ptenlox/loxmodel and in androgen-sensitive human prostate cancer cell lines; that signal is AR-stimulated, FKBP5-mediated activation of the AKT phosphatase PHLPP, which suppresses AKT activities. On the basis of their results, both groups hypothesized that prostate cancers in a castrate state (or with low AR levels) have greater dependency on PTEN loss/ PI3K-AKTsignaling. Totest this hypothesis in vivo, in scientific synchrony, Carver and colleagues showed that a combination of BEZ235 (a dual PI3K and mTOR inhibitor) and castration resulted in dramatic reductions in tumor volume, in contrast to no effect of single-pathway therapy, in LNCaP xenografts and near-complete pathologic responses in the PB-Cre;Ptenlox/lox model; Mulholland and colleagues demonstrated that rapamycin (an mTOR inhibitor) treatment of castrated PB-Cre;Ptenlox/lox; Arlox/Y mice harboring prostate cancer resulted in significantly reduced proliferation and tumor burden when compared with castration alone. The reciprocal negative feedback that links the AR and PTEN loss/PI3K-AKT signaling networks is intriguing on many levels. However, the gene expression analysis does not exclude PI3K-AKT-independent, PTEN loss-mediated signaling as a mechanism underlying upregulation of EGR1, c-JUN, and EZH2, extending the linkage between the androgenic and PTEN loss/PI3K-AKT signaling.

It is well established that AR signaling promotes the growth and differentiation of prostate epithelial cells. The precision and coordination involved in androgenic regulation of prostatic growth, morphogenesis, and cytodifferentiation depends to a large extent on AR target gene activities, which are modulated by numerous coregulators.

A recent study showed that the TMPRSS2-ERG gene fusion product can disrupt androgenic signaling in prostate cancer cells through multiple mechanisms, including binding to AR target genes and induction of EZH2 expression, which in turn can suppress prostate cell differentiation. In addition, under some conditions, PI3K-AKT signaling can enhance AR activities and induce AR target genes, such as p21WAF/CIP, which is associated with androgen-independent growth of prostate cancer. In light of the new knowledge about this mechanistic framework that has resulted from the discovery of reciprocal negative feedback linking the AR and PI3K-AKT signaling networks, it may be possible to better characterize and delineate additional signaling pathways andidentifyadditional transcriptional coregulators and chromatin modifiers that underlie specific AR target gene functions related to androgen-dependent prostatic growth and/or differentiation and to androgen-independent growth in prostate cancer. The inexorable process of selection through which cancer cells develop resistance to all types of anticancer agents presents research and clinical oncologistswith a daunting task. Through their discovery of important reciprocal negative feedback involving AR and PTEN loss/PI3K-AKT signaling in prostate cancer.

Communication between the immune system and tumor cells takes place via both cell-cell contact-dependent receptor-ligand interactions and released cytokines/chemokines. T lymphocytes of the adaptive immune system learn in the thymus to distinguish various self or altered self-structures from non-self-structures presented as peptides bound to major histocompatibility complex (MHC) class I or class II antigens (pMHC). pMHC molecules represent the antigenic universe to ab-T lymphocytes, both the self and the non-self-repertoire.

Gd-T lymphocytes recognize small phosphorylated molecules or non-classical MHCI antigens in a non-MHC restricted manner, and B lymphocytes recognize tertiary or quaternary structures of antigens using immunoglobulins (Ig). A third type of recognition is used by cells of the innate immune system: natural killer (NK) and natural killer T (NKT) cells recognize lack of expression of self (missing self), i.e., absence or low cell-surface levels of MHCI and/or MHCII molecules Furthermore, antigen-presenting cells (APC) such as dendritic cells (DC), macrophages, and granulocytes can recognize non-self-structures via toll-like receptors (TLR) or C-type lectin receptors (CLRs).

Activation of T cells, NKT, or NK cells happens via the interaction of activating receptors (KAR) associated with signaling molecules expressing immuno-tyrosineactivation- motif (ITAM) signal-motifs in their cytoplasmic region. T lymphocytes and NKT cells express T-cell receptor (TCR)/CD3 complexes, and T, NKT, or NK cells express KARs such as NKG2D associated with DAP10 signal-transduction molecules and Ly49D associated with DAP12 molecules. It appears that the high number (~10) of ITAMs associated with TCR molecules is necessary to avoid autoimmunity. Inhibition of cell activation by inhibitory receptors (KIRs) such as NKG2A/ CD94 is related to the expression by KIRs of ITIM inhibitory motifs in their cytoplasmic tails

Helper T (Th) cells trigger differentiation of precursor cells into CD8+ cytotoxic T lymphocytes (CTL) or antibody-producing B cells. The so-called Th1 cells induce preferentially the production of IL-2 and IFNg and the differentiation of CTL, whereas Th2 cells induce mostly the production of IL-4, IL-5, and IL-13 and the differentiation of B cells into antibody-producing plasma cells. When administered to elicit specific immune responses and memory but not tolerance, the antigens have to be presented in the form of cells, particles, or aggregates, or emulsified in adjuvants such as Freund’s adjuvants or aluminum salt precipitates. A unifying concept of these phenomena was proposed by late Charlie Janeway.

The innate immune system, NKT cells, NK cells, DC, macrophages, and granulocytes are activated by pathogenassociated molecular patterns (PAMPs) by means of TLR or CLRs. The DC differentiates from immature, phagocytosing cells to mature, non-phagocytosing cells with increased levels of co-stimulatory molecules such as MHCII, CD40, CD80, and CD86 and enhanced antigen-presenting activity. Cells from the innate immune system release inflammatory cytokines that induce the priming of CD4+ Th subpopulations (Th1, Th2, and Th17). Immediately after stimulation, NKT cells release preformed IFNg and IL-4 which direct Th1 and Th2 cell differentiation, respectively. IFNg triggers DC to produce IL-12, which induces preferential Th1-priming and NK cell production of IFNg and cytotoxicity. A subset of DCs, CD8+CD205+ dendritic cells, produces endogenous TGFb and is specialized to induce Foxp3+ Treg cells, whereas another subset, CD8 CD205 and DCIR-2+ (DC inhibitory receptor-2), participates in Th2 responses.

Although stimulation of the innate immune system may greatly help the initiation of adaptive cellular and humoral immune responses, over-activation of the innate immune system represents a risk due to a possible “fatal cytokine storm” However, the cytokine storm is prevented by CD4+ Th or CD8+ CTL, which down-regulate the activity of the innate immune cells by a cell-cell contact, MHC-dependent mechanism. It seems clear that memory T cells are derived from effector T cells by avoiding antigen-induced cell death. In contrast, the subdivision of CD4+ T cells into Th1, Th2, Th9, Th17, and Treg cells is more malleable and demonstrates more functional plasticity than previously thought.

Inexperienced, naive T cells appear to receive different types of “secondary education” when they encounter antigen at various regional sites including tumor micro environments. Besides possessing different effector functions, DC and T cells should be able to migrate to and within tissues. Adhesion via L-selectin induces rolling, activation, and transmigration via chemokine receptors such as CCR7. The interaction between CCR7 and its ligands, CCL19 and CCL21, may balance immunity and tolerance T lymphocytes do not react with self-structures that are expressed in normal physiological conditions, i.e., the organism is tolerant to self-structures.

However, thymus negative selection (central tolerance) is not infallible, and self-reactive Concepts and Ways to Amplify the Anti tumor Immune Response 99 T cells with low-avidity TCR do emigrate from the thymus. Such self-reactive T cells are regulated by CD4+CD25+ T regulatory (Treg) cells (peripheral tolerance). Thus, since many anti tumor immune responses are “autoimmune” reactions, these are often weak in both quantity and quality. It is important to distinguish between the following three levels of low responsiveness to malignant cells:

(1) tumor-specific T cells with high-avidity TCR have been eliminated in the thymus;

(2) tumor cells or their products may induce tolerance (in the sense of unresponsiveness), i.e., tumor-specific T cells are present but are rendered anergic;

(3) immune effectors recognize tumor cells but are prohibited from performing their natural function due to tumor-associated inhibitory molecules and/or cells.

The latter two phenomena are linked to the fact that resting DC or macrophages in the tumor induces and maintains peripheral tolerance and functional anergy in CD8+ T cells, NKT cells, and NK cells through PD-1 and CTLA-4-related mechanisms.