Organization of the Organ Systems Branch
Overview, organization, staff listing and contact information

Information for Applicants
Information for researchers interested in applying for the SPORE program

Information for the Public
Information for the public interested in activities of the SPORE program

IntraSPORE Communications
(access restricted to SPORE Investigators)

OVERALL ABSTRACT
Principal Investigator(s): Kirby I. Bland, M.D.

This Breast Cancer SPORE builds upon and enhances programs of the Comprehensive Cancer Center of The University of Alabama at Birmingham (UAB) in the areas of breast cancer prevention, genetics and therapy. The Research Projects described in this application significantly enhance well-established translational research programs, which were developed and are directed by basic and clinical scientists. The proposed breast cancer prevention project provides for the development of less toxic and more effective 9-cis retinoids and for their study in translational pre-clinical and Phase I clinical trials. The biologic effects of GKLF, a novel oncogene that modifies integrin expression in epithelial cells, are to be explored in premalignant and malignant breast tissues. A murine xenograft model will be utilized to explore the molecular mechanisms responsible for angiogenesis growth factor-induced resistance to tamoxifen and the use of angiogenesis inhibitors to restore tamoxifen sensitivity. The considerable experience of UAB investigators in experimental therapeutics is to be focused on the translational development of new breast cancer therapies including adenovirus-mediated anti-angiogenesis gene therapy, pre-targeted radioimmunotherapy utilizing a DOTA-Biotin isotope delivery system and unique polynucleotide vaccine therapies to generate potent antitumor immunity in breast cancer patients. These translational research projects are supported by Core Resources that are integrated with pre-existing Cancer Center shared facilities and will provide the administrative, biostatistical, tissue procurement, immunopathology and clinical trials expertise necessary for the successful completion of the proposed projects. UAB is providing strong institutional support for the Career Development of junior and established translational breast cancer research scientists and for a broad variety of promising Developmental Projects. Collaborations with Breast Cancer SPORES at Duke and Georgetown Universities and with the UAB Ovarian Cancer SPORE not only will enable UAB breast cancer investigators to translate basic research findings into cutting-edge clinical therapies, but also will provide for new and innovative translational breast cancer research opportunities.

Project 1
Preclinical/Clinical Development of Novel Retinoids
Principal Investigator(s): Donald D. Muccio, Ph.D.
Co-Investigator(s): Kirby I. Bland, M.D.; Samuel W. Beenken, M.D.; Wayne J. Brouillette, Ph.D.; Clinton J. Grubbs, Ph.D.; Donald H. Hill, Ph.D.; Ruiwen Zhang, M.D., Ph.D.

Nearly 44,000 women died of breast cancer in 1999. Although screening with physical exam and mammography have decreased breast cancer mortality by 30% in the 50-69 year age group, reports of mortality reduction in the pre-menopausal age group are conflicting (1). This is in part due to increased breast density in pre-menopausal women, which decreases mammography sensitivity (2,3). For many young women, current screening techniques are inadequate (4). For older women, screening compliance may be a more important issue (5). However, each year approximately 20,000 women over 50 have a negative mammogram at the time their breast cancer is palpable by physical exam.

Breast cancer incidence rates begin to rise sharply at age 30 with a very steep slope between 40 and 50 (6). A pertinent question is, "Can breast cancer be prevented by administering chemopreventive agents to women in whom early detection techniques are inadequate?"

Selective estrogen receptor modulators or SERMS are a class of compounds that exhibit antiestrogen effects in breast tissue but variable estrogen agonist or antagonist effects in other estrogen responsive organ systems (7). Most of the antitumor effects of antiestrogens are modulated directly or indirectly through the estrogen receptor. Tamoxifen, which itself is a partial estrogen agonist/antagonist, has effects on various endocrine and protein markers not directly related to carcinogenesis.

To date, Phase II and III breast cancer chemopreventive trials have utilized the antiestrogen tamoxifen or retinoids or both. Tamoxifen in doses of 20 to 40 mg per day given adjunctively to women with Stage I and II breast cancer for 2 to 5 years has been found to reduce the odds of development of a contralateral breast cancer by approximately 40% (8). A Phase III Breast Cancer Prevention trial was recently completed by the National Surgical Adjuvant Breast and Bowel Project (NSABP P-1). In that trial 13,388 women at increased risk for breast cancer were randomized to tamoxifen 20 mg vs. placebo for five years (1992 to 1997). The women enrolled in the trial were 60 years of age or older, 35-59 years of age with a five-year predicted risk of breast cancer of at least 1.66% based on the Gail's algorithm, or had a history of lobular carcinoma in situ. At a median participant follow-up time of 45 months, tamoxifen reduced the risk of invasive breast cancer and noninvasive breast cancer by 50% in comparison to the control arm. The decreased risk was noted across all age groups (9).

In an effort to reduce toxicities associated with Tamoxifen use, other SERMS are being tested in clinical trials. We are focusing our research efforts on retinoids, some of which have been shown to be effective in preventing breast cancer in animal models. A long-term goal is to study the toxicity profile and chemoprevention efficacy of appropriate retinoids when given in combination with Tamoxifen (or another SERM) in human chemoprevention trials. Based upon literature reports, two classes of retinoids are the most promising for in vivo prevention of breast cancers: (A) amide or ester derivatives of all-trans-retinoic acid (RA), with the most studied example being fenretinide (4-hydroxyphenyl retinamide or 4-HPR), and (B) 9-cis-retinoic acid (9cRA) and related analogs, like Targretin (LGD1069), that are selective agonists for the nuclear retinoid X receptor (RXR). Each of these classes exhibits much lower toxicity than the natural hormones, RA or 9cRA. Further, RXR-selective retinoids enhance the efficacy of Tamoxifen combination therapy.

At UAB we have designed a novel series of RXR-selective retinoid agonists. In preliminary studies we have shown that one example, 9cUAB30, prevents mammary cancer in vivo with no signs of toxicity. Our research team has also confirmed that Tamoxifen and 9cRA act additively in an in vivo mammary cancer prevention model. An optimal combination of antiestrogens and RXR-selective retinoids could decrease toxicity and enhance chemopreventive efficacy. In this proposal we will conduct further studies with 9cUAB30 and its analogs to optimize breast cancer chemopreventive efficacy and minimize toxicity, using both single and combination therapy regimens. The most promising agent(s) will be moved into Phase I human clinical trials. The specific aims are:

1. Select UAB Retinoids with Potential for High Chemopreventive Activity and Low Toxicity. Our synthetic program has provided small quantities of new UAB retinoids. These retinoids and newly designed analogs will undergo in vitro nuclear receptor binding/activational studies, and each year three UAB retinoids that are selective RXR-agonists will be chosen for synthesis at the 10 gram level for preliminary toxicity studies (Aim 2). The UAB retinoid(s) with the lowest toxicity that maintains a favorable RXR-selective receptor binding and activational profile will be synthesized at the 50-100 g level for chemoprevention efficacy studies (Aim 3). 9cUAB30, which is highly efficacious and completely nontoxic at 200 mg/kg of diet in an in vivo chemoprevention model, will be tested for efficacy at higher doses during year 01 of this proposal.

2. Determine the Oral Toxicity and Pharmacology of RXR-Selective UAB Retinoids. The maximum tolerated dose (MTD) of the three UAB retinoids (Aim 1) will be determined orally in rats. The most promising UAB retinoid(s) will undergo large-scale synthesis (50-100 g) for in vivo efficacy studies (Aim 3). The oral bioavailability and pharmacokinetics of any UAB retinoids that exhibit significant mammary cancer chemopreventive activity will be determined.

3. Determination of the Effective Dose for Cancer Chemoprevention in the MNU-model. One or two UAB retinoids each year will be evaluated orally in the N-methylnitrosourea (MNU)-model for mammary cancer chemopreventive efficacy in rats. The most effective, nontoxic mammary cancer chemoprevention agent will be selected for further evaluation in combination with the estrogen antagonist Tamoxifen. Tissue distribution and metabolite identification for the UAB retinoid will be determined and correlated between treated and control groups. The effects of retinoids on preneoplastic lesions in a MNU-model of mammary cancer will be determined by evaluating changes in surrogate endpoint biomarkers. These endpoint biomarkers will include markers of proliferation (e.g., PCNA), apoptosis (TUNEL), and specific growth factors and proto-oncogenes (TGF-alpha, EGF-R, and p185erbB-2) (Core B).

4. Phase I Clinical Trials of New Retinoids. We will advance the most active UAB retinoids that have low toxicity and favorable pharmacology to Phase I clinical trials in humans. A RAPID proposal has already been submitted to NCI to move 9cUAB30 as a chemopreventive agent into human clinical trials (Appendix). Specific objectives are: (A) To determine the pharmacokinetics and bioavailability of an orally administered, single dose novel 9-cis UAB retinoid; (B) To determine first and final day pharmacokinetics of an orally administered novel 9-cis UAB retinoid during a one-month daily course of treatment; (C) To identify and quantify any toxicities produced by chronic oral administration of a novel 9-cis UAB retinoid; (D) To determine the effect of a one month oral course of a novel 9-cis UAB retinoid on apoptosis, proliferation, and expression of growth factors/receptors in breast carcinoma and normal appearing breast tissue from women with breast cancer.

Project 2
Hormonal Resistance: Mechanisms and Reversal
Principal Investigator(s): Francis G. Kern, Ph.D.
Co-Investigator(s): John T. Carpenter, Jr., M.D.; Marina S. Manuvakhova, M.D.; Shanti K. Samuel, Ph.D.; Yulia Y. Maxuitenko, Ph.D.

Adjuvant hormonal therapies are as effective as more toxic chemotherapies for both premenopausal and postmenopausal women with estrogen receptor positive tumors (1-3). Tamoxifen is also useful for patients with ER+ tumors who present with advanced disseminated disease (4-6). Although instances of remission can last up to ten years, an inevitable occurrence in all tamoxifen treated patients with advanced disease is the outgrowth of populations of cells that no longer respond (6,7). The low toxicity and relatively mild side effects associated with hormonal therapies have spurred the development of second line hormonal therapies. While these therapies deserve further development, there are also indications that these will only have limited effectiveness in a portion of patients who have failed or relapsed following tamoxifen treatment (5,8). Consequently, efforts aimed at gaining a better understanding of the mechanisms underlying tamoxifen resistance are warranted.

Our preliminary and published data indicate that paracrine effects of angiogenic growth factor overexpression by ERa+ breast cancer cells can reduce the effectiveness of tamoxifen as a cytostatic or cytotoxic agent (9-14). These paracrine effects may also increase tamoxifen's ability to function as an agonist of the estrogen receptor. A better understanding is needed of the mechanisms by which these growth factors increase the agonistic properties of tamoxifen, alter the balance of proliferation and cell death, or bypass the need for estrogen receptor activation. This will lead to increased effectiveness of selective estrogen receptor response modifiers (SERMS) in the adjuvant and possibly prevention treatment settings. The transfected cell lines and model systems we have developed provide our laboratory with unique resources that allow us to address these important issues through the following specific aims.

1. We will derive cell lines in which the overexpression of VEGF is regulated by doxycycline to examine the effect of temporal regulation of VEGF overexpression on growth and maintenance of xenografts in nude mice in the presence of estrogen or tamoxifen. MCF-7 cells will be transfected with an improved version of a tetracycline regulated reverse transactivator expression vector that should permit continued stable expression in the majority of the population. Clones stably expressing this tet repressor – VP16 transactivator fusion protein will be retransfected with an expression vector in which a minimal promoter containing tet operator sequences upstream of a VEGF165 cDNA allows inducible VEGF expression in the presence of doxycycline. The effects of the presence and absence of VEGF overexpression on tumor growth in tamoxifen treated and estrogen treated nude mice in will be assessed by providing and removing doxycycline from the drinking water at different time points.
   

2. We will determine the mechanism by which angiogenic growth factor overexpression permits the growth of ER+ human breast carcinoma cells as a xenograft in tamoxifen treated nude mice. Tumors from control transfected and VEGF transfected MCF-7 cells will be analyzed by Annexin V, Ki67, and BrdU staining to test the hypothesis that angiogenic growth factor overexpression alters the balance in cell proliferation and apoptosis in tamoxifen treated nude mice. We will explore the hypothesis that VEGF mediated neoangiogenesis decreases the rate of apoptosis, thereby permitting the agonistic properties of tamoxifen to be manifested as a net increase in tumor volume. Quantitative RT-PCR assays and Western Blots with available antibodies will be used to assess the effects of angiogenesis on the expression of antiapoptotic proteins bcl-2, bcl-XL, BAG-1, and Mcl-1 and the proapoptotic proteins bax, bad, bcl-Xs, bid and bak. Effects of VEGF overexpression on the levels of phospho-bad, phospho p38 MAPK, and phospho-AKT within xenografts will be monitored. We will also examine whether the transcription of estrogen receptor regulated genes, the S phase fraction, the BrdU or Ki67 labeling index or ERK MAPK or RSK activity are increased in tumors from tamoxifen treated mice in the context of VEGF overexpression. We will establish cocultures of VEGF treated endothelial cells or carcinoma activated stromal fibroblasts and MCF-7 cells to determine if soluble factors or tumor cell -stromal cell contact mediate tamoxifen stimulated breast cancer cell growth.
   

3. We will explore whether mechanisms proposed for acquired tamoxifen resistance in other systems are operative in our model in the context of VEGF overexpression. RT- PCR and Western Blotting assays will be used to test the hypothesis that angiogenesis alters the ratio of transcriptional coactivators to corepressors within tumors thereby increasing the agonistic properties of tamoxifen. We will also determine whether angiogenesis affects the level of AP-1 activity and increases the ratio of ERb to ERa in MCF-7 cells, thereby permitting tamoxifen bound ERb to cooperate with AP-1 in the induction of AP-1 regulated growth stimulating genes. TGF-b1 and TGF-b2 expression will also be assessed. The activity of ERK and RSK kinases that phosphorylate regulatory serine residues within the ERa AF-1 domain and increase the agonistic properties of tamoxifen will be also examined.
   

4. We will assess the ability of VEGFR antagonists to restore tamoxifen sensitivity in MCF-7 VEGF transfected human breast cancer xenografts and in MNU induced rat mammary tumors. To provide the conceptual framework for a clinical trial in metastatic breast cancer patients with acquired tamoxifen resistance, small molecule tyrosine kinase inhibitors and a monoclonal blocking antibody targeted to the Flk-1/KDR receptor will be used in a preclinical study. These agents will allow us to determine if inhibition of VEGF mediated neoangiogenesis can inhibit the development of tumors in tamoxifen treated animals or restore tamoxifen sensitivity to established VEGF overexpressing MCF-7 cell tumors. The small molecule inhibitor will also be used to determine whether VEGFR or VEGFR/FGFR inhibition, when used in combination with tamoxifen, can further decrease the frequency of new tumors or reduce the size of established tumors that arise after tamoxifen treatment in an N-nitrosomethylurea (NMU) induced rat mammary tumor prevention model.
   

5. We will initiate a pilot clinical study to determine if antagonists of VEGF signaling can restore tamoxifen sensitivity to patients who have failed on tamoxifen or relapsed following adjuvant tamoxifen treatment. Patients with ER+ tumors who present with metastatic disease and acquire resistance to tamoxifen or patients with ER+ tumors who relapse following or while receiving adjuvant tamoxifen therapy will be treated with a combination of a VEGFR antagonist and tamoxifen and response to treatment will be monitored.

Project 3
Biology and Intermediate Marker Role
of a Novel Breast Cancer Oncogene, GKLF
Principal Investigator(s): J. Michael Ruppert, M.D., Ph.D.
Co-Investigator(s): Susan L. Bellis, Ph.D.; Andra R. Frost, M.D.

A subset of oncogenes, including wild-type alleles of c-MYC or GLI, and activated alleles of RAS or -catenin, can transform cultured diploid epithelial cells in vitro (RK3E cells). These genes are frequently activated in human tumors by mutation of upstream regulatory molecules. Expression of GLI mRNA is activated in nearly all basal cell carcinomas of the skin through mutation of the tumor suppressor PTC1 or the PTC1-associated protein, SMO, while c-MYC is activated in a majority of colorectal carcinomas through mutation of APC or -catenin. Neither GLI nor c-MYC exhibit frequent point mutation, gene amplification, or gene rearrangement, perhaps because expression of the mRNA is normally tightly controlled and limits the overall activity of the pathway.

Our preliminary data identify GKLF/KLF-4 as a new member of this important subset of oncogenes (1). GKLF is expressed in the nonproliferating, integrin-negative suprabasal squamous cell layers, and plays an essential role in maturation of stratified squamous epithelium (2,3,4). Expression of GKLF in vitro is not increased in association with cellular proliferation or by transformation with other oncogenes (3,1). In colorectal carcinomas, which frequently exhibit activation of RAS and/or the APC-- -catenin--c-MYC pathway, GKLF mRNA expression is reduced in tumor cells compared with normal colonic mucosa (5). However, expression of GKLF mRNA and protein are greatly increased early during progression of most cases of ductal carcinoma of the breast and oral squamous cell carcinoma. These tumor-types do not frequently exhibit activation of other oncogenes such as -catenin, RAS or GLI. Based upon these results, we proposed that dysregulation of GKLF mRNA expression may be a consequence of genetic alterations in upstream regulatory proteins (1).

Our results identify GKLF as a potent inhibitor of integrin expression. Unlike RK3E cells transformed by other oncogenes, cells transformed by GKLF exhibit markedly reduced expression of specific integrins, and reduced rates of attachment to their ligands collagen and laminin. In contrast, GKLF activates the vitronectin receptor, previously implicated in the metastatic phenotype of breast cancer cells (6,7).

Integrins encode receptors for the basement membrane components laminin and collagen and are required for cellular adhesion to the basement membrane and for epithelial polarization and ordered morphogenesis. Expression of integrin mRNA (8) and protein (9,10) is reduced in dysplastic epithelium and in tumors, and markedly reduced expression of integrins in breast tumors is a predictor of metastasis to axillary lymph nodes (11,12). Blocking integrin function in vitro or in vivo blocks epithelial polarization and ordered morphogenesis of the breast (13,14). Restoring integrin expression in breast cancer cell lines induces multiple characteristics of normal breast epithelia (15). While loss of integrin expression has been proposed to be important in the pathogenesis of breast cancer (12,16), mechanisms responsible for integrin loss remain unclear.

An additional consequence of GKLF activation in RK3E cells is increased expression of the tumor marker clusterin. Clusterin may play a role in cell-cell or cell-matrix interactions, and increased expression of this molecule is associated with a poor prognosis in other tumor types such as prostate cancer (17).

Expression of a GKLF transgene in the basal cell layer of skin induces loss of the apical-basal polarization of these cells and other features of dysplasia. Taken together, our preliminary data identify GKLF as a regulator of integrins and clusterin and as a candidate effector of tumor progression for tumor-types such as breast cancer that do not exhibit frequent activation of previously characterized oncogenes.

Aim 1: To determine the expression pattern of GKLF, integrins, and clusterin in ductal hyperplasia (DH), atypical ductal hyperplasia (ADH), ductal carcinoma-in-situ (DCIS), and invasive ductal carcinoma of the breast (IDC).

Hypothesis: GKLF expression is activated early during tumor progression and is inversely correlated with integrin expression.

• mRNA in situ hybridization and immunohistochemical analyses will be used to examine GKLF expression in normal breast tissue and in sections of breast containing uninvolved epithelium, DH, ADH, DCIS, and IDC.

• Immunohistochemical analysis of the same clinical samples will be used to correlate expression of GKLF with expression of integrins 2 1, 3 1, 6 4, 6 3, and clusterin.

• Results of these studies will be correlated with clinical outcome for over 200 cases for which tissue, treatment history, and clinical outcome data are currently available.

Aim 2: To characterize the role of GKLF in neoplastic progression of the breast using mice transgenic for constitutive or inducible GKLF transgenes.

Hypothesis: GKLF transgenic mice are predisposed to tumor progression and provide a genetically well-defined model of human tumor progression.

• Female mice expressing GKLF in breast epithelium using a constitutive MMTV promoter or a doxycycline-inducible promoter will be analyzed for altered morphology of breast epithelium and for expression of integrins, clusterin, and specific markers of proliferation and differentiation.

• We will determine whether GKLF can promote tumor progression when expressed alone or in combination with polyoma middle T antigen, Her-2/neu, or TGF.

Aim 3: To characterize the regulation of integrin expression and function by GKLF in RK3E epithelial cells and human mammary epithelial cells (HMECs).

Hypotheses: Carcinoma oncogenes inhibit integrin expression or function by distinct mechanisms. GKLF represses transcription of specific integrin family members.

• RK3E cells and existing cell lines transformed by RAS, GLI, GKLF, and c-MYC will be assessed for expression of integrin family members by immunoblot and northern analysis.

• HMECs and GKLF- or c-MYC-transduced populations will be examined for altered expression and function of integrins, and for ability to form branching ducts and alveolar structures when cultured in a three-dimensional gel.

• Integrin function will be further assessed for each cell line by determining the rate of attachment to specific components of the extracellular matrix (ECM).

• Cell lines will be assessed for rate of spreading on specific components of the ECM, overall migration rate, and invasion through a model basement membrane.

• A modified doxycycline-inducible strategy will be used to determine the time course of GKLF-induced alterations in vitro.

• A promoter-reporter assay will be used to identify GKLF-responsive elements in specific integrin promoters and to investigate whether GKLF can directly regulate integrin promoters.

• We will determine whether repression of integrin expression is necessary for GKLF-induced transformation.

Project 4
Gene Therapy Specifically Directed at Tumor Vasculature
Principal Investigator(s): David T. Curiel, M.D.
Co-Investigator(s): Victor Krasnykh, Ph.D.

As advances are made in identifying the molecular mechanisms of neoplastic growth, it has become apparent that neovascularization plays a key role in development of tumors. This is especially relevant for carcinoma of the breast, whereby the tumor angiogenesis has been shown to be strictly linked to tumor dissemination, thereby leading to the disease progression to the later stages at which currently available therapeutic modalities are inefficient. To this end, the reliance of tumor progression on active neoangiogenesis, and the identification of numerous factors modulating this process, has provided a rationale for a number of novel anti-cancer strategies based on eradication of tumors via abrogation of their blood supply. These studies have unambiguously demonstrated that it is possible to significantly slow the tumor growth and even cause its necrosis by interrupting the chain of intermolecular interactions required for vascularization of tumors. These strategies have involved the delivery of recombinant proteins which interact specifically with growth factor arcs involved in neoangiogenesis. In addition, gene therapy strategies have been developed along these same lines to achieve delivery of soluble factors via gene based methods. In both of these general approaches, significant inhibition of tumor development and progression have been achieved. Unfortunately, these responses have been less than complete. It has been hypothesized that these limited responses have reflected the inability of the aforementioned delivery systems to achieve high, local concentrations of the angiogenesis factors at the site of neoangiogenesis. Thus, in the context of gene therapy strategies, it has been suggested that specific delivery of antiangiogenesis genes to the sites of neovascularization may represent the means to realize therapeutic benefit. On this basis, there is a recognized need for gene delivery vector capable of this specific and selective transduction of tumor vasculature endothelium.

It is thus the goal of this proposal to facilitate novel therapies for abrogation of tumor angiogenesis in the context of carcinoma of the breast by developing a transfer vector capable of efficient and specific gene delivery to tumor vasculature. Recombinant adenovirus vectors are uniquely suited for this role due to their systemic stability and ability to achieve efficient gene transfer in vivo. However, widespread distribution of the cellular receptor for the fiber of human adenovirus serotype 5 prevents the cell-specific gene delivery by the present generation of vectors. We have shown recently that genetic modification of adenovirus fiber protein can be successfully used to expand the tropism of the vector. However, the systemic context of tumor vasculature targeting applies more stringent requirements for the vector design, thereby necessitating the derivation of the vector uniquely targeted to specific receptor molecules. This, in turn, requires dramatic structural changes to the fiber protein in order to ablate endogenous tropism of the vector. By using our genetic approach, we have obtained preliminary results, which strongly suggest, that a replacement of the entire fiber protein in an adenovirus virion with a chimeric T4 fibritin molecule may result in the derivation of the desired vector. Based on these findings, we hypothesize that it will be possible to employ the advantages of our genetic strategy in order to create truly targeted adenovirus vector capable of selective transduction and subsequent eradication of tumor vasculature cells. The specific aims of this proposal are thus:

Specific Aim #1. To develop targetable adenovirus vector by employing the fiber replacement strategy, which will be capable of targeted gene delivery to the endothelial cells of tumor vasculature.

Specific Aim #2. To employ this novel vector to demonstrate tumor vasculature targeting in in vivo model of disseminated carcinoma of the breast.

Specific Aim #3. To create the derivatives of this viral vector expressing anti-angiogenic factors and to employ them in murine model of disseminated carcinoma of the breast.

The realization of these specific aims would result in a systemically stable vector capable of efficient transduction and eradication of tumor vasculature cells. By addressing a key issue which limits the translation of present gene therapy strategies into clinical trials, the development of such a vector system proposed herein would therefore represent a major technological advance in gene therapy for breast cancer. Moreover, such vector would be of great utility in a wider field of gene therapy.

Project 5
Polynucleotide Vaccine Therapy of Breast Cancer
Principal Investigator(s): Theresa V. Strong, Ph.D.
Co-Investigator(s): Robert M. Conry, M.D.; Denise R. Shaw, Ph.D.

Enthusiasm over tumor immunization strategies is increasing with the identification of multiple cloned tumor-associated antigen targets with defined immunodominant peptide epitopes. Similarly, interest in polynucleotide immunization is rapidly growing with numerous demonstrations of efficacy in animal models of infectious diseases and cancer including nonhuman primate trials. However, the initial trials of polynucleotide vaccines in humans have produced more modest immune responses, suggesting the importance of incorporating augmentation strategies into second generation trials of this modality. We believe that it is impractical to evaluate the wide array of available augmentation strategies in clinical trials. We, therefore, further believe that a syngeneic, human CEA-expressing adenocarcinoma model in mice transgenic for human CEA represents the best available model for examination and comparison of these augmentation strategies to allow translation of the most promising strategies to clinical breast cancer trials. Although we have selected CEA as our initial tumor-associated antigen, the technology described herein is relevant to any protein tumor-associated antigen, cost effective, and applicable to a wide variety of clinical settings.

1. To test the hypothesis that administration of a plasmid DNA vaccine as a sustained release preparation will more effectively break immunological tolerance than conventional administration in saline.

   1. To evaluate biodegradable polymer polylactide-co-glycolide (PLG) microspheres as a delivery system for DNA vaccination via intramuscular and intradermal routes.

   2. To evaluate the sucrose acetate isobutyrate (SABER®) delivery system for DNA vaccination via intramuscular versus intradermal routes.

2. To test the hypothesis that plasmid DNA vaccines encoding fusion proteins consisting of a tumor antigen fused to an antigen presenting cell targeting moiety will enhance immune response and antitumor effects compared to a plasmid encoding the antigen alone.

   1. Plasmid DNA encoding CEA-CD40 ligand (CD154) fusion.

   2. Plasmid DNA encoding CEA-GM-CSF fusion.

   3. Plasmid DNA encoding CEA-hsp70.

3. To compare codelivery of plasmid-encoded cytokines, chemokines or growth factors to augment immune response and antitumor effects of a DNA vaccine encoding a tumor-associated self-antigen to identify the optimal strategy for translation to clinical trials.

   1. pCEA + pIL-12

   2. pCEA + pGM-CSF

   3. pCEA + pCD40L

   4. pCEA + pflt3L

   5. pCEA + pLymphotactin

   6. pCEA + phsp70

4. To test the hypothesis that Sindbis virus-derived naked replicative RNA vaccines with or without cytopathic effects will induce immune responses and antitumor effects superior to conventional plasmid DNA vaccines.

5. To conduct a phase Ib clinical trial in patients with hormone responsive metastatic breast cancer of a CEA polynucleotide vaccine using the optimal vector, delivery method, fusion protein and/or plasmid-encoded augmentation molecule based upon the outcome of Specific Aims 1-4.

Project 6
Pre-targeting Radioimmunotherapy of Metastatic Breast Cancer
Principal Investigator(s): Ruby F. Meredith, M.D., Ph.D.
Co-Investigator(s): Donald J. Buchsbaum, Ph.D.; Albert F. LoBuglio, M.D.; M.B. Khazaeli, Ph.D.

The overall goal of this project is to develop novel radioimmunotherapy programs for patients with metastatic breast carcinoma. We propose to use a pre-targeting strategy developed by our collaborators at NeoRx in which we will explore a variety of pertinent strategic issues in animal models with rapid translation to early phase I and II trials in patients. Our specific aims are:

1. To determine the ability of a novel recombinant tetrameric targeting molecule (sFv4/SA) directed at the TAG-72 or MUC-1 antigen on tumor cells to accomplish eradication of antigen positive mammary fatpad xenografts via Biotin-DOTA-90Y tumor delivery in a pre-targeting radioimmunotherapy strategy.

2. To examine the benefit of using a mixture of two pre-targeting recombinant molecules each of which target a different surface antigen for incremental isotope tumor delivery without a need to increase the dose of radioisotope administration.

3. To further explore this pre-targeting radioimmunotherapy strategy in models of hepatic, lung and bone metastases (overt and microscopic) utilizing 90Y, 177Lu and 213Bi. This deals with the issue of multiple small tumor deposits and varying isotope radiation characteristics.

4. To carry out phase I and phase II trials of pre-targeting radioimmunotherapy using recombinant tetrameric targeting molecules and Biotin-DOTA-radioactive isotope reagents in patients with metastatic breast cancer.