Cytokine Gene Vaccine Therapy for Treatment of a Brain Tumor.

A glioma is a malignant brain tumor with a poor prognosis. Attempts at the surgical removal of the tumor are the first approach, but additional treatment strategies, including radiation therapy and systemic or local chemotherapy, are necessary. Furthermore, the treatments are often associated with significant adverse side effects. Normal and malignant cells generally have antigenic differences, and this is the rationale for clinical immunotherapeutic strategies. Cytokines such as IL-15 or IL-2, which stimulate an anti-tumor immune response, have been shown to have a particularly high potential for use in immunotherapy against various tumors. In this review, treatments with either a poxvirus, genetically engineered to secrete IL-15, or allogeneic fibroblasts, transfected with tumor DNA and engineered to secrete IL-2, are shown to be effective strategies in extending the survival of mice with malignant brain tumors upon intracerebral injection of the treatment cells. Future studies with these treatment strategies in patients with intracerebral tumors are urgently needed.

Immune evasion in cancer: Mechanistic basis and therapeutic strategies.

Cancer immune evasion is a major stumbling block in designing effective anticancer therapeutic strategies. Although considerable progress has been made in understanding how cancers evade destructive immunity, measures to counteract tumor escape have not kept pace. There are a number of factors that contribute to tumor persistence despite having a normal host immune system. Immune editing is one of the key aspects why tumors evade surveillance causing the tumors to lie dormant in patients for years through "equilibrium" and "senescence" before re-emerging. In addition, tumors exploit several immunological processes such as targeting the regulatory T cell function or their secretions, antigen presentation, modifying the production of immune suppressive mediators, tolerance and immune deviation. Besides these, tumor heterogeneity and metastasis also play a critical role in tumor growth. A number of potential targets like promoting Th1, NK cell, γδ T cell responses, inhibiting Treg functionality, induction of IL-12, use of drugs including phytochemicals have been designed to counter tumor progression with much success. Some natural agents and phytochemicals merit further study. For example, use of certain key polysaccharide components from mushrooms and plants have shown to possess therapeutic impact on tumor-imposed genetic instability, anti-growth signaling, replicative immortality, dysregulated metabolism etc. In this review, we will discuss the advances made toward understanding the basis of cancer immune evasion and summarize the efficacy of various therapeutic measures and targets that have been developed or are being investigated to enhance tumor rejection.

Immunogene therapy.

Antigenic differences between normal and malignant cells of the cancer patient form the rationale for clinical immunotherapeutic strategies. Because the antigenic phenotype of neoplastic cells varies widely among different cells within the same malignant cell-population, immunization with a vaccine that stimulates immunity to the broad array of tumor antigens expressed by the cancer cells is likely to be more efficacious than immunization with a vaccine for a single antigen. A vaccine prepared by transfer of DNA from the tumor into a highly immunogenic cell line can encompass the array of tumor antigens that characterize the patient's neoplasm. Poorly immunogenic tumor antigens, characteristic of malignant cells, can become strongly antigenic if they are expressed by highly immunogenic cells. A DNA-based vaccine was prepared by transfer of genomic DNA from a breast cancer that arose spontaneously in a C3H/He mouse into a highly immunogenic mouse fibroblast cell line, where genes specifying tumor-antigens were expressed. The fibroblasts were modified in advance of DNA-transfer to secrete an immune augmenting cytokine and to express allogeneic MHC Class I-determinants. In an animal model of breast cancer metastatic to the brain, introduction of the vaccine directly into the tumor bed stimulated a systemic cellular antitumor immune response measured by two independent in vitro assays and prolonged the lives of the tumor-bearing mice. Furthermore, using antibodies against the various T-cell subsets, it was determined that the systemic cellular antitumor immunity was mediated by CD8+, CD4+ and NK/LAK cells. In addition an enrichment strategy has also been developed to increase the proportion of immunotherapeutic cells in the vaccine which has resulted in the development of enhanced antitumor immunity. Finally regulatory T cells (CD4+CD25+Fox p3+-positive) were found to be relatively deficient in the spleen cells from the tumor-bearing mice injected intracerebrally with the enriched vaccine. The application of DNA-based genomic vaccines for the treatment of a variety of brain tumors is being explored.

Investigation of immunosuppressive mechanisms in a mouse glioma model.

The development of an immune competent mouse model for the study of immunosuppressive mechanisms is important for improving the efficacy of brain tumor immunotherapy. In the present study we investigated regulatory T cells (Tregs), TGF-beta1 and other putative immunosuppressive cytokines using GL261 mouse glioma in C57BL mice. We explored whether tumor growth factor-beta1 (TGF-beta1) is expressed and secreted by glioma cells constitutively or in response to a T-cell mediated immunity (simulated by conditioned media from T cells (TCM) activated by anti-CD3 antibody). We also investigated TGF-beta1's role in Treg mediated immunosuppression by quantifying TGF-beta1secretion from T regulatory cells (Tregs) co-incubated with GL261 cells as compared to Tregs alone. Finally, we studied other newly identified cytokines that were secreted preferentially by glioma cells in response to CD3 activated TCM versus cytokines secreted by glioma cells in absence of T-cell activation (naïve TCM). TGF-beta1expression was studied using RT-PCR and secretion was quantified using ELISA. A 308 protein cytokine array was used to identify and quantify cytokine expression. TGF-beta1expression and secretion from glioma cells was found to be up-regulated by conditioned media from CD3-activated T cells, suggesting that this immunosuppressive cytokine is not secreted constitutively but in response to immunity. TGF-beta1 was not found to be differentially secreted by Tregs co-incubated with glioma cells as compared to Tregs alone. This data suggest that TGF-beta1immunosupppression may not be a Treg dependent mechanism in this glioma model. Finally, the cytokine array elucidated several other cytokines which were up-regulated or down-regulated by CD3-activated TCM. These results have several implications for enhancing immunotherapy treatment, including the potential benefit of TGF-beta1inhibition in conjunction with immunotherapy, as well as the illumination of several other potential cytokine targets to be explored as shown by the cytokine array.

PPAR-gamma Thiazolidinedione Agonists and Immunotherapy in the Treatment of Brain Tumors.

Thiazolidinediones (TZDs) are selective agonists of the peroxisome proliferator-activated receptor (PPAR) gamma, a transcription factor belonging to the superfamily of nuclear hormone receptors. Although activation of PPARgamma by TZDs has been best characterized by its ability to regulate expression of genes associated with lipid metabolism, PPARgamma agonists have other physiological effects including modulating pro- and anti-inflammatory gene expression and inducing apoptosis in several cell types including glioma cells and cell lines. Immunotherapeutic approaches to reducing brain tumors are focused on means to reduce the immunosuppressive responses of tumors which dampen the ability of cytotoxic T-lymphocytes to kill tumors. Initial studies from our lab show that combination of an immunotherapeutic strategy with TZD treatment provides synergistic benefit in animals with implanted tumors. The potential of this combined approach for treatment of brain tumors is reviewed in this report.

Enhanced immunity to intracerebral breast cancer in mice immunized with a cDNA-based vaccine enriched for immunotherapeutic cells.

Structural differences between malignant and nonmalignant cells of the same individual form the basis of clinical immunotherapeutic strategies. Previously, we reported the therapeutic properties of a vaccine prepared by transfer of a cDNA-expression library from breast cancer cells into a highly immunogenic allogeneic fibroblast cell line where genes specifying an array of breast cancer antigens were expressed. Here, we report the application of this cell-based vaccination strategy for breast cancer metastatic to the brain. As relatively few cells in the vaccine were expected to have incorporated cDNA fragments specifying breast cancer antigens (most specify normal cellular constituents), an enrichment strategy was employed to increase the proportion of immunotherapeutic cells. Enhanced immunity to the neoplasm mediated predominantly by CD4+, CD8+, and NK/LAK cells was generated in the spleens of mice injected intracerebrally into the tumor bed with cells from the enriched vaccine, which translated into prolonged survival. Regulatory T cells (CD4+CD25+Foxp3+-positive) were relatively deficient in the spleen cells from tumor-bearing mice injected intracerebrally with the enriched vaccine.

Prolonged survival of mice with established intracerebral glioma receiving combined treatment with peroxisome proliferator-activated receptor-gamma thiazolidinedione agonists and interleukin-2-secreting syngeneic/allogeneic fibroblasts.

OBJECT: In this study the authors explored the benefits of treating C57B1/6 mice with an established intracerebral glioma by combining immunotherapy with interleukin (IL)-2-secreting syngeneic/allogeneic fibroblasts administered into the tumor bed along with the chemotherapeutic agent pioglitazone, a thiazolidinedione (TZD). The TZDs are agonists of the peroxisome proliferator-activated receptor-gamma. They have been found to exert antiproliferative effects on several transformed cell lines. Data from prior studies by these authors have revealed the immunotherapeutic properties of the IL-2-secreting fibroblasts in treating intracerebral gliomas in mice. METHODS: The sensitivity of GL261 glioma cells and primary astrocytes to pioglitazone was determined in vitro by incubating the cells with increasing amounts of the drug. Viability was assessed by measuring lactate dehydrogenase release, and effects on metabolism were determined by measuring superoxide production and levels of superoxide dismutase. The GL261 cells were injected intracerebrally into C57B1/6 mice, followed by treatment with pioglitazone either orally or intracerebrally into the tumor bed. The effect of the combined therapy was determined by injecting C57B1/6 mice with an established intracerebral GL261 glioma with IL-2-secreting allogeneic fibroblasts and pioglitazone directly into the tumor bed through a unique cannula system. Pioglitazone was found to induce cell death in GL261 glioma cells grown in vitro while causing only modest damage to astrocytes. The application of pioglitazone also resulted in a significantly greater induction of cellular superoxide in glioma cells than in astrocytes, which can activate apoptotic pathways. Pioglitazone administered intracerebrally (p < 0.05) but not orally was found to prolong survival in mice harboring an intracerebral glioma. Synergistic effects of combination therapy on prolonging survival were found in mice receiving both pioglitazone and IL-2-secreting fibroblasts (p < 0.005, compared with untreated animals). Pioglitazone induces metabolic and oxidative stresses that are tolerated by astrocytes but not glioma cells, which could account for selective vulnerability and increased sensitivity to IL-2, suggesting potential for the use of this Food and Drug Administration-approved drug in the treatment of brain tumors. CONCLUSIONS: The data indicate the beneficial effects of combination therapy using pioglitazone and immunotherapy in mice harboring intracerebral glioma.

Differential effects of PPARgamma agonists on the metabolic properties of gliomas and astrocytes.

Recent studies show that thiazolinediones (TZDs), agonists of the peroxisome proliferator-activated receptor gamma (PPARgamma), induce apoptosis in glioma and glioblastoma cells. Here we compared the effects of troglitazone (Trog), a TZD with low affinity for binding to PPARgamma but with potent metabolic effects, on survival and metabolism in GL261 glioma cells versus primary astrocytes. Trog dose-dependently induced cell death in GL261 cells (with over 90% death at 30 microM) but did not cause any toxicity in astrocytes at the same doses. Measurements of glucose and lactate levels after incubation with Trog (30 microM) indicated an overall increase of glucose consumption and lactate production in both cell types. In astrocytes the ratio of lactate produced to glucose utilized was not significantly altered by Trog, while in glioma cells this ratio was decreased by about 40%. Trog dose-dependently reduced mitochondrial membrane potential (DeltaPsi(m)) in both cell types; and the loss of DeltaPsi(m) was greater in the tumor cells (90% loss at 20 microM) than in astrocytes (70% loss at 20 microM). These results suggest that differences in metabolic responses could contribute to the selective resistance of astrocytes to cytotoxic effects of Trog. TZDs such as Trog should therefore be considered for testing in treatment of gliomas.

Antigenic Differences Between Normal and Malignant Cells as a Basis for Treatment of Intracerebral Neoplasms Using a DNA-Based Vaccine.

Antigenic differences between normal and malignant cells of the cancer patient form the rationale for clinical immunotherapeutic strategies. Because the antigenic phenotype of neoplastic cells varies widely among different cells within the same malignant cell-population, immunization with a vaccine that stimulates immunity to the broad array of tumor antigens expressed by the cancer cells is likely to be more efficacious than immunization with a vaccine for a single antigen. A vaccine prepared by transfer of DNA from the tumor into a highly immunogenic cell line can encompass the array of tumor antigens that characterize the patient's neoplasm. Poorly immunogenic tumor antigens, characteristic of malignant cells, can become strongly antigenic if they are expressed by highly immunogenic cells. A DNA-based vaccine was prepared by transfer of genomic DNA from a breast cancer that arose spontaneously in a C3H/He mouse into a highly immunogenic mouse fibroblast cell line, where genes specifying tumor-antigens were expressed. The fibroblasts were modified in advance of DNA-transfer to secrete an immune augmenting cytokine and to express allogeneic MHC class I-determinants. In an animal model of breast cancer metastatic to the brain, introduction of the vaccine directly into the tumor bed stimulated a systemic cellular anti-tumor immune response measured by two independent in vitro assays and prolonged the lives of the tumor-bearing mice. Furthermore, using antibodies against the various T-cell subsets, it was determined that the systemic cellular anti-tumor immunity was mediated by CD8(+), CD4(+) and NK/LAK cells. The application of DNA-based genomic vaccines for the treatment of a variety of brain tumors is being explored.

Early experience from the application of a noninvasive magnetic resonance imaging-based measurement of intracranial pressure in hydrocephalus.

OBJECTIVE: The decision for surgical intervention in hydrocephalic patients presenting with symptoms suggesting raised intracranial pressure (ICP) is challenging because radiographic ventricular size often lacks the specificity to predict abnormal ICP. An early assessment of the potential clinical usefulness of a noninvasive magnetic resonance imaging-based measurement of ICP (MR-ICP) in symptomatic hydrocephalic patients is reported. METHODS: Twenty-seven symptomatic hydrocephalic patients (18 with shunts and 9 without shunts) underwent brain magnetic resonance imaging-based studies that included measurements of cerebrospinal fluid and blood flows to and from the cranial vault, from which measurements of ICP were derived using a previously described algorithm. The predictive values of the MR-ICP measurement were determined on the basis of whether or not the patient underwent a surgical treatment of a shunt placement or shunt revision within a 3-month period after the magnetic resonance imaging-based study. RESULTS: MR-ICP values in these patients spanned a much wider range than in healthy control subjects. However, the majority of the patients (20 out of 26 patients) had MR-ICP values within the normal range. The short-term follow-up of patients who had normal MR-ICP measurement reveals that only one of the 20 patients required surgery. Consequently, the MR-ICP measurement has a strong negative predictive value (95% for all patients and 100% for patients without a shunt). CONCLUSION: A finding of a normal MR-ICP value in hydrocephalic patients presenting with symptoms suggestive of abnormal ICP is a strong predictor for resolution of symptoms or stable outcome without surgical intervention.

Advantages of a unique DNA-based vaccine in comparison to paclitaxel in treatment of an established intracerebral breast cancer in mice.

In this study we compared the benefits of treating C3H/He mice with an established intracerebral breast carcinoma by immunization with a unique DNA-based vaccine to chemotherapy with paclitaxel. Prior studies revealed the immunotherapeutic properties of a vaccine prepared by transfer of genomic DNA from breast cancer cells into a highly immunogenic cell line. Here, C3H/He mice with an established intracerebral breast cancer were treated either by injection into the tumor bed through a unique cannula system with the cell based vaccine or with paclitaxel administered intraperitoneally. Both treatment strategies were effective in prolonging survival and stimulating a systemic anti-tumor immune response (p< 0.025). However, unlike mice treated with the vaccine, the animals that received paclitaxel alone displayed significant toxic side effects. No additional therapeutic advantage was detected when these two treatment strategies were combined. The vaccine tended to provide a somewhat better therapeutic and clearly better systemic immunologic effect based on two independent spleen cell assays in comparison to paclitaxel.

Immunogene therapy as a treatment for malignant brain tumors in young mice.

OBJECT: New and innovative forms of effective treatments for malignant brain tumors in children are urgently needed. The authors have previously shown that intracerebral injection into the tumor bed of allogeneic fibroblasts genetically engineered to secrete interleukin-2 (IL-2) results in prolongation of survival and an antitumor immunocytotoxic response in adult mice that harbor intracerebral gliomas. The first goal of this study was to determine if malignant gliomas (GI261) could be treated in mice (C57BL/6) in the pediatric age group (weanlings [2-3 weeks old] and adolescents [3-4 weeks old]). The second goal was to determine the effectiveness of using IL-2-secreting allogeneic fibroblasts as a protective vaccine to prevent the development of intracerebral gliomas in these young mice. METHODS: Using GI261 glioma cells derived from a spontaneously arising glioma in C57BL/6 immunocompetent mice, animals 2 to 4 weeks of age received an intracranial injection of 5 x 10(4) tumor cells into the right frontal lobe through a bur hole. The treatment vaccine consisted of 10(6) allogeneic IL-2-secreting fibroblasts, given at the time of tumor injection (treatment experiments) or at three weekly intervals prior to tumor injection (protection experiments). Control groups received either medium or nonsecreting allogeneic fibroblasts. The effects of this treatment on survival and long-term immunity were investigated. The results demonstrate a significant prolongation of survival in animals harboring intracerebral gliomas that were treated with intracerebral injections of IL-2-secreting allogeneic fibroblasts (p < 0.05). Morbidity and mortality rates did not increase as a result of intracerebral immunization. Compared with naive controls, long-term survivors demonstrated immune memory, as evidenced by prolongation of survival when they were rechallenged with tumor cells. The results of the protection experiment demonstrate a significant delay (p < 0.005) in the development of gliomas in the animals pretreated with either allogeneic nonsecreting or allogeneic IL-2-secreting fibroblasts prior to the introduction of tumor cells. In addition, in 78% of these animals a tumor did not develop when rechallenged. CONCLUSIONS: These results demonstrate the efficacy and safety of using intratumoral injection of IL-2-secreting allogeneic fibroblasts as a treatment or protective vaccine in young mice. It is hoped that these preclinical studies will lead to a clinical trial for the treatment of malignant brain tumors in children.

Intratumoral injection of IL-secreting syngeneic/allogeneic fibroblasts transfected with DNA from breast cancer cells prolongs the survival of mice with intracerebral breast cancer.

Prior studies have revealed the immunotherapeutic properties of a vaccine prepared by transfer of genomic DNA from breast cancer cells into a highly immunogenic cell line. The rationale for this type of vaccine is that genes specifying an array of weakly immunogenic, unique tumor antigens associated with the malignant cells will be expressed in a highly immunogenic form by the transfected cells. Here, the immunotherapeutic properties of a vaccine prepared by transfection of mouse fibroblasts with DNA from a breast carcinoma (SB-5b) that arose spontaneously in a C3H/He mouse (H-2Kb) were tested in mice with intracerebral breast cancer. To augment their nonspecific immunogenic properties, before DNA transfer, the fibroblasts (of C3H/He mouse origin) were modified to express allogeneic MHC class I H-2Kb-determinants and to secrete IL-2, IL-18 or GM-CSF. The results indicate that C3H/He mice injected intracerebrally (i.c.) with the breast cancer cells and syngeneic/allogeneic-transfected fibroblasts modified to secrete IL-2 survived significantly longer (P < .005) than mice in various control groups, including mice injected i.c. with the breast cancer cells alone. The immunotherapeutic properties of transfected fibroblasts modified to secrete IL-18 or GM-CSF were less efficacious. The results of two independent in vitro cytotoxicity assays indicate that systemic cellular antitumor immunity was generated in mice injected i.c. with the transfected cells, and the immunity was mediated predominantly by CD8+ T cells.

Evaluating the effect of decompression surgery on cerebrospinal fluid flow and intracranial compliance in patients with chiari malformation with magnetic resonance imaging flow studies.

OBJECTIVE: To quantify the effect of decompression surgery on craniocervical junction hydrodynamics and on global intracranial compliance (ICC) in patients with Chiari I malformation by use of magnetic resonance measurements of cerebrospinal fluid and blood flow. Studying the effect of decompression surgery may improve our understanding of the pathophysiological characteristics of Chiari I malformation and aid in identifying patients who will benefit from the procedure. METHODS: Twelve patients were studied with a 1.5-T magnetic resonance imaging scanner before and after decompression surgery. Cine phase contrast magnetic resonance images were used to quantify maximum cord displacement, maximum systolic cerebrospinal fluid velocity and volumetric flow rate, and overall ICC. ICC was derived by use of a previously reported method that measures small changes in intracranial volume and pressure that occur naturally with each cardiac cycle. RESULTS: After surgery, changes were documented both in the local hydrodynamic parameters and in ICC. However, only the change in ICC, an average increase of more than 60%, was statistically significant. Increased ICC, which was associated with improved outcome, was measured in 10 of the 12 patients, no significant change was documented in 1 patient, and decreased ICC was measured in 1 patient whose symptoms persisted after surgery. CONCLUSION: An increase in the overall compliance of the intracranial compartment is the most significant and consistent change measured after decompression surgery. Changes in cord displacement, cerebrospinal fluid velocities, and flow in the craniospinal junction were less consistent and less affected by the operation. Thus, ICC may play an important role in the outcome of decompression surgery related to improving symptoms and restoring normal neurological hydrodynamics in patients with Chiari I malformations.

Cytokine immuno-gene therapy for treatment of brain tumors.

The prognosis for patients with an intracerebral (i.c.) neoplasm is poor. Conventional treatments such as surgery, radiation therapy and chemotherapy have done little to affect long-term survival, and new methods of treatment are urgently needed. In this report approaches involving cytokine gene therapy in treatment of malignant brain tumors are reviewed and contrasted to a strategy developed in this laboratory involving the use of allogeneic cells genetically modified to secrete cytokines. In our studies, mice with an i.c. glioma, melanoma or breast carcinoma treated solely by intratumoral injections with allogeneic cells genetically modified to secrete interleukin-2 (IL-2) were found to survive significantly longer than mice in various control groups. The anti-tumor response was mediated predominantly by T-cell subsets (CD8+ and NK/LAK cells). The injections resulted in the killing of only the neoplastic cells; non-neoplastic cells were unaffected. Experiments involving treatment of animals with i.c. tumor using subcutaneous injections of cytokine-secreting allogeneic cells in the presence of tumor antigens demonstrated no effect in prolonging survival in spite of the development of a vigorous systemic anti-tumor immune response. Of special interest, mice injected intracerebrally with the cytokine-secreting allogeneic cells alone exhibited no neurologic defect and there were no adverse effects on survival. The injection of cytokine-secreting allogeneic cells into the microenvironment of an i.c. tumor is hypothesized to induce an anti-tumor immune response capable of prolonging survival. This pre-clinical animal data directly translates into clinical treatments for patients with a malignant i.c. tumor.

Treatment with allogeneic interleukin-2 secreting fibroblasts protects against the development of malignant brain tumors.

We have reported that mice with an intracerebral (i.c.) malignant glioma or breast cancer treated with i.c. injection of allogeneic fibroblasts genetically engineered to secrete interleukin-2 (IL-2) survived longer than mice in various control groups. The goal of the present study was to determine the effectiveness of utilizing IL-2 secreting allogeneic fibroblasts as a protective treatment to prevent the development of an i.c. glioma or breast carcinoma. Using an intracranial microcannula system that we developed, the animals received weekly injections of the cellular vaccine prior to the introduction of tumor cells via the cannula. The results demonstrate a significant delay (P < 0.005) in the development of glioma in the animals pre-treated with either allogeneic non-secreting or IL-2-secreting fibroblasts prior to introduction of tumor cells. In addition, 50% of the animals pre-treated with IL-2 secreting allogeneic fibroblasts injected subsequently with G1261 glioma cells did not develop a tumor, while all of the animals injected with glioma cells alone and 92% of those treated with non-secreting fibroblasts eventually died. The long-term survivors from the pre-treatment group were subsequently re-challenged with tumor and compared to naive controls. There was evidence that long-term immunity was established in the pre-treated animals, since there was a significant prolongation of survival (P < 0.01). In similar experiments using breast cancer cells, 62% of the animals pre-treated with non-secreting allogeneic fibroblasts and 75% of the animals pre-treated with allogeneic IL-2 secreting fibroblasts subsequently injected with SB-5b breast carcinoma cells did not develop tumors and had a significant prolongation of survival. These data suggest that i.c. injection of allogeneic IL-2 secreting fibroblasts are effective as a protective treatment in the prevention of the development of a brain tumor when the fibroblasts are introduced into the same site where the tumor is subsequently injected.

Genomic expression discovery predicts pathways and opposing functions behind phenotypes.

Discovering states of genetic expression that are true to a high degree of certainty is likely to predict gene function behind biological phenotypes. The states of expression (up- or down-regulated) of 19200 cDNAs in 10 meningiomas are compared with normal brain by an algorithm that detects only 1 false measurement per 192000; 364 genes are discovered. The expression data accurately predict activation of signaling pathways and link gene function to specific phenotypes. Meningiomas appear to acquire aberrant phenotypes by disturbing the balanced expression of molecules that promote opposing functions. The findings expose interconnected genes and propose a role of genomic expression discovery in functional genomics of living systems.

Mathematical modeling of noise and discovery of genetic expression classes in gliomas.

The microarray array experimental system generates noisy data that require validation by other experimental methods for measuring gene expression. Here we present an algebraic modeling of noise that extracts expression measurements true to a high degree of confidence. This work profiles the expression of 19 200 cDNAs in 35 human gliomas; the experiments are designed to generate four replicate spots/gene with switching of probes. The validity of the extracted measurements is confirmed by: (1) cluster analysis that generates a molecular classification differentiating glioblastoma from lower-grade tumors and radiation necrosis; (2) By what other investigators have reported in gliomas using paradigms for assaying molecular expression other than gene profiling; and (3) Real-time RT-PCR. The results yield a genetic analysis of gliomas and identify classes of genetic expression that link novel genes to the biology of gliomas.