Tumor Hypoxia as a Facilitator of Cancer Progression through Immune Evasion
Sceneay J, Chow MT, Chen A, Halse HM, Wong CSF, Andrews DM, et al. Primary tumor hypoxia recruits CD11b+/Ly6Cmed/Ly6G+ immune suppressor cells and compromises NK cell cytotoxicity in the premetastatic niche. Cancer Res; Published OnlineFirst July 2, 2012; doi:10.1158/0008-5472.CAN-11-3873.
Hypoxia is correlated with poor prognosis in many cancers. Accordingly, this process is a key contributor to progression through its ability to facilitate angiogenesis, invasion, and metastasis. These changes in tumor expansion and evolution enabled through hypoxia are mediated primarily by hypoxia-inducible factor-1 (HIF-1). Sceneay and colleagues demonstrate that, in addition to inducing various angiogenic factors, hypoxic conditions also render the immune system ineffective by upregulating immunosuppressive pathways. They examined these phenomena in experimental mouse models of breast cancer (PyMT) and melanoma (B16F10). Remarkably, conditioned medium obtained from hypoxic cells (HCM) induced angiogenic factors, including MCP-1, in driving the premetastatic niche in the lungs of C57BL/6NZeg-enhanced GFP mice. This process was accompanied with recruitment of immunomodulatory CD11b+/Ly6C/LY6G+ bone marrow–derived cells in this niche, which are potential myeloid-derived suppressor cells known to inhibit antitumor immune responses. Moreover, HCM treatment also resulted in the suppression of natural killer (NK) cell cytotoxicity, which was correlated with metastasis formation in secondary organs, in an analogous manner to ablation of NK cells. Conversely, the presence of active NK cells in premetastatic lungs decreased metastatic burden. In total, the present study suggests that primary tumor hypoxia facilitates the production of cytokines and growth factors with potential to create a premetastatic niche by means of recruitment of CD11b+/Ly6C/LY6G+ myeloid cells and a decrease in the cytotoxic effector functions of NK cell populations.
Epidermal Growth Factor Receptor, Inflammation, and Colon Cancer
Inflammatory bowel disease is a chronic illness caused by complex interactions between genetic and environmental factors that damage the epithelium of the colon. This chronic inflammation increases the risk of developing colitis-associated cancer. Dubé and colleagues present important data on the role of epidermal growth factor receptor (EGFR) signaling in this inflammation-induced cancer. Many aspects of homeostasis of the colonic epithelium, including cell proliferation, survival, wound closure, and barrier function, are regulated by EGFR signaling. EGFR transactivation may also be involved in determining inflammatory outcomes. With respect to its homeostatic roles, EGFR signaling could promote wound healing and decreased inflammation, thereby reducing the long-term risk of cancer in patients with colitis. However, EGFR is also widely considered to be a tumor promoter with increased expression in colon adenomas and cancers. Anti-EGFR monoclonal antibodies are also approved treatments for metastatic colon cancer. To address this apparent paradox, Dubé and colleagues studied mice that had a dominant negative mutation in EGFR. Loss of EGFR alone was not sufficient to induce colitis. However, in 2 different colitis models, mice with inactivation of EGFR had greatly increased colonic inflammation compared with wild-type mice, and the incidence and progression of colorectal tumors were accelerated. The authors show that in mice, functional EGFR signaling slows disease progression by optimizing epithelial regeneration, particularly by preventing pathogenic hyperproliferation, dysplasia, and DNA damage. These functions of EGFR are likely to accelerate the rate of wound healing, prevent epithelial barrier dysfunction, and promote the resolution of inflammation. What are the implications of these murine studies for patients with colitis or colon cancer? The data predict that EGFR antagonists would not be effective in patients with colitis-induced cancer either because they might exacerbate the inflammation that is driving tumor progression. Alternatively, EGFR blockade could fail because of inherent resistance in colitis-associated tumors due to inflammation-induced mutations, potentially in the mitogen-activated protein kinase pathway. Conversely, therapies that stimulate EGFR signaling may help patients with colitis [one published trial shows that EGF enemas have therapeutic activity (Sinha A, Nightingale J, West KP, Berlanga-Acosta J. N Engl J Med 2003;349:350–7.)] without increasing their long-term risk of cancer.
Drugging the Undruggable
One of the first identified features of cancer cells was an increase in the size and number of nucleoli, the site of ribosome biogenesis. Several studies have shown that elevated ribosomal DNA (rDNA) transcription by RNA polymerase I (Pol I), a rate-limiting step in ribosome biogenesis, is associated with tumorigenesis. However the precise causal role of accelerated Pol I–mediated rDNA transcription in cancer development remains poorly understood. Although MYC directly regulates Pol I activity in ribosome biogenesis, it has been difficult to address whether increased ribosomal RNA synthesis represents a determinant in Myc oncogenic potential. Bywater and colleagues used a complementary genetic and pharmacologic approach to address this question. Specifically, the authors used a mouse model for lymphoma in which Myc is overexpressed in the B-cell compartment (Eμ-Myc). Premalignant Eμ-Myc B cells exhibited increased cell volume, rates of proliferation, and rates of rDNA transcription compared with wild type. Furthermore, Eμ-Myc B cells displayed elevated expression of Pol I components, including the key activator of Pol I, UBF; the Pol I initiation factor RRN3 (also known as TIF-1A); and POLR1B, one of the largest subunits of Pol I. This activity was sustained from premalignancy to malignancy, suggesting that increased levels of Pol I factors and elevated rates of rDNA transcription were not a consequence of cellular transformation, but rather actively contributed to tumor initiation. Bywater and colleagues knocked down UBF and RRN3 activity in Eμ-Myc lymphoma cells to near-normal levels and showed that these cells exhibited higher levels of apoptosis. Strikingly, the authors demonstrated that a selective Pol I inhibitor (CX-5461) induced cell death in Eμ-Myc lymphoma cells that harbored intact p53 function. Because Pol I–mediated transcription levels ultimately control ribosome production, the authors sought to delineate whether the observed increased apoptosis in Eμ-Myc lymphomas treated with CX-5461 was due to p53-mediated cell death or ribosome insufficiency. Eμ-Myc lymphoma cells analyzed after 1-hour treatment with CX-5461 displayed increased levels of p53 and its downstream targets (e.g., p21), whereas ribosome abundance remained constant, supporting the hypothesis that inhibition of Pol I activity triggered a p53-mediated nucleolar stress response. In this stress response (also known as the nucleolar surveillance pathway), ribosomal proteins RPL5 and RPL11 are released from the nucleolus and bind to the p53 ubiquitin ligase MDM2, which in turn releases p53 to activate its tumor-suppressive program. The authors performed coimmunoprecipitation experiments to show that higher levels of RPL5 and RPL11 were associated with MDM2 in Eμ-Myc lymphoma cells treated with CX-5461, as compared with vehicle-treated cells, thereby suggesting strongly that perturbations in Pol I–mediated transcription in Eμ-Myc lymphoma cells activate the RP–MDM2–p53 nucleolar surveillance pathway. The authors used C57BL/6 mice transplanted with p53 wild-type Eμ-Myc lymphoma to evaluate the clinical efficacy of CX-5461 toward Myc-driven lymphomas. Strikingly, CX-5461 selectively induced p53-mediated death in malignant cells, with no observed cytotoxic effects in non–tumor-bearing mice. Administration of CX-5461 in transplanted cells from a mouse model for acute myeloid leukemia (AML) produced similar levels of apoptotic activity compared with standard drugs currently used to treat AML. Consistent with the authors' in vitro and in vivo findings, human hematologic cancers exhibiting wild-type levels of p53 were more sensitive to the Pol I inhibitor CX-5461 than were tumors harboring mutations in p53. Notably, although cancer cells with mutant p53 displayed similar levels of reduced Pol I activity compared with cells with wild-type p53, cells with intact p53 function displayed higher levels of p53 and p21 along with nucleolar disruption and increased rates of apoptosis. Together, these studies shed new light on a critical role for Pol I–mediated rDNA transcription in aberrant ribosome biogenesis and cancer development. This activity can be targeted with Pol I inhibitors, leading to cell death caused by induction of the RP–MDM2–p53 nucleolar surveillance response. These findings have strong clinical significance, as selectively targeting rDNA transcription in cancers harboring intact p53 with Pol I inhibitors represents a new therapeutic avenue in the treatment of hematologic cancers. This article also highlights a shift in dogma, as what was once considered undruggable (e.g., Pol I activity and rDNA transcription) has been rendered druggable.
Molecular Grading of Primary Cutaneous Melanoma
In cutaneous melanomas, there is a well-known association between tumor thickness and patient outcome, with tumors of greater thickness portending a worse prognosis. Additionally, thicker tumors are associated with elevated mitotic activity and ulceration, both of which are negative prognostic factors. Harbst and colleagues used genomic profiling to examine the biologic implications of these observations. Their previous study of grade 4 melanomas had identified 4 molecular subtypes. One of the subtypes identified, with an immune response signature, was associated with another subtype that displayed a proliferative signature. Although gene expression profiling of primary cutaneous melanomas has represented a technical challenge, the advent of methods to profile small biopsies from formalin-fixed paraffin-embedded (FFPE) material has enabled analysis of these tumors. The authors therefore studied 223 FFPE samples of cutaneous melanoma using Illumina cDNA-mediated annealing, selection, extension, and ligation profiling and placed these tumors in the context of their previously identified molecular subtypes. They were able to classify 98% of the tumors into 1 of the 4 subtypes. Cases in the immune response and normal subtypes had significantly improved overall survival compared with cases classified in the pigmentation and proliferative subtypes. The authors then decided to condense the immune response/normal classes into a “low-grade” group and the remaining 2 classes into a “high-grade” group. They validated the signatures that defined these groups on external data sets and showed significant differences based on patient outcome. In addition, the molecular group was an independent predictor of outcome in multivariate analysis after adjustment for relevant clinical variables. The authors then screened for mutational status of 19 oncogenes and found a higher frequency of BRAF mutations in the low-grade molecular group. In addition, BRAF mutations were more common in tumors that were less than 2 mm thick. NRAS mutations showed a trend toward increased frequency in the high-grade molecular group. However, BRAF mutations alone did not account for the significant survival differences seen in molecular grouping. Overall, the authors defined signatures compatible with FFPE material that may predict patient outcome in primary cutaneous melanoma. These results could potentially be used to guide clinical monitoring and treatment of patients.
Note: Breaking Advances are written by Cancer Research Editors. Readers are encouraged to consult the articles referred to in each item for full details on the findings described.