Introduction

Over the past 30 years, the understanding of the complexity and origins of cancer at the genetic level has changed dramatically. These advances have been translated into more targeted cancer treatments in the form of small molecules like Iressa (gefitinib, ZD1839; Astra Zeneca, Manchester, UK) and Gleevec (imatinib mesylate, STI571; Novartis, Basel, Switzerland) and antibodies like Zevalin (ibritumomab tiuxetan; Biogen-Idec Pharmaceuticals, Cambridge, MA, USA); Rituxan (rituximab; Genentech, South San Francisco, CA, USA), Herceptin (tratuzumab; Genentech), and Avastin (bevacizumab; Genentech); Mylotarg (gemtuzumab oxogamicin; Wyeth, Cambridge, MA); Campath (alemtuzumab; Ilex Pharmaceuticals, San Antonio, TX, USA; Berlex Laboratories, Richmond, CA, USA; Schering AG, Berlin, Germany); and Panorex (Mab17-1a, edrecolomab; GlaxoSmithKline, Middlesex, UK). If genetically-based therapies are to join these validated cancer therapy platforms, our field can certainly benefit from the study of their discovery and development pathways to success. At the same time, genetically based agents for cancer have distinct and significant differences from validated therapeutic platforms, and each faces its own unique challenges. In this review, we will summarize the research and development paradigms that have accompanied major therapeutic advances in cancer over the past decade. In addition, we use our own personal experiences and perspectives in the development of genetically based therapies to identify issues that are unique to some, if not all, of the agents from this platform. In particular, we will focus this commentary primarily on oncolytic virus development; nevertheless, the issues that are highlighted have broad relevance to the entire field. Therefore, by discussing changes in research and development methodologies that have influenced the therapeutic platforms of small molecules and antibodies, we will be able to compare and contrast the research and development of a range of molecular therapies for cancer. It is our hope that this review will facilitate the discovery and development of promising genetically based cancer therapeutics.

From target identification to investigational new drug candidate

Target discovery

In the target identification stage, investigators seek to identify a molecular or cellular target critical to the pathophysiology of cancer. Ideally, this target would be preferentially expressed and/or activated in cancer cells versus normal cells so that it could be exploited for the development of a cancer therapeutic. Target discovery has, through history, paralleled our understanding of cancer genetics. Perhaps the earliest molecular target identified was the estrogen receptor in breast carcinoma; therapies targeting this protein have been successfully developed for decades. Many initial cancer-specific targets were identified through the study of the transforming proteins associated with an assortment of oncogenic viruses. This line of investigation has proven rich in targets, revealing the receptor tyrosine kinase KIT; the non-receptor tyrosine kinases ABL, LCK, and SRC; the serine/threonine kinases AKT and RAF; the guanine nucleotide exchange factor Ras; and the nuclear proteins and transcription factors FOS, JUN, and MYC as potential targets for drug development. The study of nonrandom chromosomal rearrangements, including translocations, inversions, deletions, and gene amplifications, has also been a fertile resource for therapeutic targets. Most notable among these has been the BCR-ABL translocation associated with CML and the gene amplification associated with the overexpression of c-erbB-2 (HER-2/Neu). These two targets have led to the generation of the small molecule inhibitor Gleevec and the antibody-based breast cancer treatment Herceptin, respectively. The study of mouse cancer modifier loci is also a genomic-based approach to identifying target genes. Mouse cancer modifier loci are regions with allele-specific effects on cancer development (both positive and negative, reviewed in ^1,2,3). It is assumed that human homologues of mouse cancer modifier loci exist and may provide targets for the development of diagnostic, preventative and therapeutic strategies for cancer.

The Human Genome Project has had a significant impact on the target discovery process. The application of new genomic technologies has allowed for extensive comparisons of expression profiles in tumor cells and normal cells. Gene expression microarrays ^4,5 represent one of the genomic technologies directing target discovery. Gene expression profiling is finding use not only in target identification, but also in lead optimization, toxicology, biomarker and mechanism of action identification, and tumor classification and treatment selection.

While gene expression profiles generate potentially useful data, this methodology should be complemented by protein expression assays. Indeed, RNA levels and protein abundance and activity may differ significantly ^6,7,8. Proteomics is the study of all the proteins and their posttranslational modifications expressed from a target genome, with a goal to characterize the protein pathways and networks necessary for information flow within the cell and, on a larger scale, within the organism (reviewed in ^9,10,11,12, including excellent reviews of the various technologies associated with proteomics). Like the gene expression microarrays, the application of proteomics in drug development extends far beyond target identification. An alternative approach to target identification and drug development is to screen large, complex compound libraries against tumor cell lines to identify efficacious compounds without a preconceived notion as to the target ¹³.

As our understanding of the complexity of human tumors has increased, it has become clear that tumor progression is dependent upon a variety of noncancerous cells (e.g., endothelial cells, stromal cells) and factors associated directly or indirectly with them (e.g., matrix metalloproteinases, growth factors). Consequently, noncancerous cells and these associated factors should also be considered as a source for cancer therapy targets. The validity of this type of approach is supported by the recent approval of an anti-VEGF antibody, bevacizumab (Avastin), for the treatment of human colon cancer in combination with 5-fluorouracil-based chemotherapy. The approval of this antiangiogenic treatment has bolstered activity to test additional antiangiogenic cancer agents and to identify novel antiangiogenic and antivascular targets ^{14,15,16,17,18,19}. Additional noncancerous cell-associated targets are actively being pursued and identified, including those targeted toward tumor-associated extracellular matrix components, stromal cells, and proteases (reviewed in ^{20,21,22,23,24,25,26,27}).

How do these advances in target identification specifically affect small-molecule-, antibody-, and gene therapy-based drug approaches? There is no doubt that genomic and proteomic studies will generate new and exciting information and advances both now and in the future. Consequently, it is likely that we will have an overabundance of targets for directed therapies. More important, then, will be the statistical methods and algorithm development that must accompany the advances in data accumulation. The data analysis will need to move at a rate equal to the data accumulation occurring through genomic and proteomic analysis. In this fashion, the immense amount of information being generated can be discriminated into useful data that identifies and prioritizes the relevant targets. It is likely that this will move toward an enhanced understanding of the interconnecting signaling pathways and networks within cells, with the hope that the information flow within and outside of the cell can be linked so that critical pathways or networks can be targeted effectively. Consequently, it is likely that therapies will evolve away from narrowly targeted, protein site-specific therapies that act on a single component in a complex pathway or network toward therapies that act on more than one target and thereby affect multiple critical pathways to treat the cancer progression more effectively.

Target validation

Target validation involves demonstrating that the desired biological effect is achieved when the target is engaged in an appropriate in vitro and/or in vivo model system. In vitro, this has traditionally involved the use of several different assay systems, including antisense oligonucleotides ^28,29,30, ribozymes (reviewed in ³¹), targeted or dominant negative mutants, and antibodies, if available. Protein mutants are also particularly useful in target validation. Loss-of-function, gain-of-function, or specifically targeted mutations to determine that the selected domain is actually relevant to the function of the protein are highly desirable. For all of these systems, target validation is dependent upon the inactivation or elimination of the target protein and a concurrent readily measurable outcome, including phenotypic properties (dedifferentiation), proliferation, migration, or survival. In vivo, nude mice carrying human tumor xenografts and transgenic mice genetically engineered to express the target gene aberrantly (over- or underexpress) have been the mainstays of traditional in vivo target validation. For example, direct evidence for the involvement of VEGF in tumorigenesis was first demonstrated using monoclonal antibodies to the VEGF protein that were shown to inhibit human tumor xenograft growth ³², target validation studies that eventually resulted in the recently approved VEGF antibody-based drug Avastin.

More recently, additional and more sophisticated methods have been developed to complement these existing target validation mainstays. These include screening large, highly complex libraries of oligonucleotides (aptamers) and peptides (peptide aptamers) to identify molecules that bind to a target molecules with high affinity and specificity (reviewed in ^33,34,35). A method garnering a great deal of attention as a tool for in vitro target validation is RNA interference (RNAi). This is a simple, rapid, and effective method of silencing gene expression (10–100 more potent than antisense) that exploits a natural process of sequence-specific, posttranscriptional gene regulation common to all cells of eukaryotic organisms (reviewed in ^36,37). Briefly, there are two major steps to its mechanism of action: the first involves the degradation of double-stranded RNA to form small interfering RNAs (siRNAs), 19 to 25 nucleotides long, by an RNase III enzyme called DICER ³⁸. In the second step, these siRNAs interact with a complex termed RISC (RNA-induced silencing complex) to facilitate the targeting of homologous RNA for degradation, thus creating a sequence-specific nuclease ³⁹. To generate prolonged or permanently altered cells, there have been several reports on the effective use of short hairpin RNAs (shRNAs). The shRNAs contain a short target sequence followed by a hairpin spacer, followed by a sequence complementary to the target but in reverse orientation. This molecule folds back upon itself to form the double-stranded short hairpin that enters the RNAi pathway and operates in the same manner as an siRNA for gene silencing. Recently, a set of retroviral vectors containing over 23,000 distinct shRNAs, targeting over 7900 different human genes for suppression, has been constructed and represents a powerful tool to facilitate large-scale loss-of-function genetic screens in mammalian cells ⁴⁰.

For in vivo functional target validation, animal modeling has continued to evolve both in complexity and in relevance. Now multiple events within the same cells can be induced and tracked by intercrossing independently derived mouse strains altered to express different target oncogenes and/or to lack tumor suppressor protein function(s). This provides a powerful genetic approach for determining specific collaborating events and defining essential networks in tumorigenesis. To examine the impact of oncogene expression during distinct developmental stages, and to explore the requirement for sustained oncogene activity, several investigators have pioneered the use of reversible (e.g., tetracycline inducible, reviewed in ⁴¹) and conditional gene expression strategies (e.g., Cre–Lox system reviewed in ^42,43,44). Through these various techniques, elegant models now exist for some of the major cancer indications (e.g., prostate ^45,46), for bone metastasis ⁴⁷, and for angiogenesis (reviewed in ^48,49), and these can be complemented, in some cases, by spontaneous tumor models in dogs and cats ⁵⁰. The development of miniaturized imaging equipment and sensitive marker genes has allowed investigators to monitor tumor growth/regression in a noninvasive fashion in vivo (reviewed in ^51,52,53) with technologies ranging from charge-coupled device cameras for optical detection to three-dimensional reconstructing positron emission tomography (PET) and X-ray computed tomography.

Tissue microarrays are also being used for expression target validation. Since potential target genes may have been identified using material derived from a heterogeneous primary tumor, it may be important to distinguish which of the multiple different cell types within the tumor is expressing the target gene. These clinically defined tissue samples, available through nonprofit organizations such as the National Human Genome Research Institute (National Institutes of Health, Bethesda, MD, USA: http://research.nhgri.nih.gov/tma), consist of clinical material and are ideal for in situ tissue technologies such as immunohistochemistry, RNA in situ hybridization, and fluorescence in situ hybridization to validate expression or location of expression within the context of the tumor ^54,55,56. In this method, up to 1000 different minute tissue samples can be assayed simultaneously in situ on one microscope glass slide.

As was noted for target identification, target validation should be performed with complementing technologies to increase the probability that the observed phenotype is relevant to the target and not a phenomenon of the testing technique. For example, identical results with antisense RNA, through its use of a different enzyme (RNase H) and compartment (nucleus), and siRNA, through its use of a completely different enzyme (DICER) and compartment (cytoplasm), will substantially increase the confidence about the information generated by the target validation outcome.

Lead discovery

The lead discovery phase in the drug development pathway is the initial step in the identification of a chemical or biological entity that interacts with the target in a fashion that generates the desired therapeutic outcome. For the small-molecule-, antibody-, and genetic therapy (the collective term for DNA-, RNA-, viral-, or plasmid-based cancer therapeutics)-based approaches to developing cancer therapies, this step is marked by significant differences in the starting materials, but they share a common starting point, philosophically—the desire to maximize opportunities to make an optimal lead discovery.

For small molecules, lead compounds have traditionally been and continue to be identified by screening thousands of different chemical compounds in cell-free or cell-based assays, using specific predefined parameters to identify target "hits." As experience with small molecules has grown with in vitro and in vivo models and in the clinic, it has become possible to predict, to varying degrees, some of the properties critical to successful drug development (reviewed in ^57,58,59). Consequently, absorption, distribution, metabolism, and excretion (ADME) and toxicology, parameters that traditionally had minor consideration during the discovery phase, are now are being used to identify "drug-like" chemical fragments that can serve as the basis for the initial starting materials (reviewed in ^58,60). Aiding in the screening process is virtual screening, in which the 3D structure of the target protein binding site is used to prioritize compounds by their likelihood of binding to the protein or by using one or more proteins known to bind to the protein and using a similarity criterion to identify potential leads ^61,62.

For antibodies, lead candidates were originally identified by their high-affinity binding to the tumor-associated cell surface target antigen. The first antibodies were rodent monoclonal antibodies, and many of the limitations to the murine-based antibody technologies (highly immunogenic, short half-life, inability to trigger effector functions) were not appreciated initially. It was also thought that, if the antibody coated the target tumor cell, elimination would be inevitable. However, most cells of the body have defenses that protect them from attack by complement and other cellular effectors and it is known that these defenses can be enhanced in malignant cells (reviewed in ⁶³). With the recognized limitations of murine antibodies, lead antibodies now will be fully human, coming from human antibody libraries expressed in a phage display format, from plants, or from transgenic animals that have had their endogenous antibody genes replaced by the equivalent human sequences ^{64,65,66,67,68}. In addition, lead antibody screens will need to identify the interaction with the target to understand fully the liabilities or strengths associated with the binding. For example, antibody binding to the target protein may induce or block internalization of the target or signaling pathways necessary for survival or progression.

In the case of genetic therapies, the traditional initial lead discovery candidate is composed of two components, a delivery vector (e.g., plasmid, virus-based vector) and a therapeutic gene designed to lead, directly or indirectly, to the destruction of the target tumor cell. Traditionally, the therapeutic gene classes included tumor suppressors (e.g., p53) restored to arrest or induce apoptosis in the mutant tumor cell, immunoregulatory genes (e.g., GM-CSF, IL-2, or TNF) selected to stimulate or elicit an immune response to the tumor, prodrug converting enzymes (e.g., thymidine kinase) that would selectively convert a nontoxic prodrug to a toxic chemotherapeutic at the tumor site, or antibodies, as discussed previously. With time, it has become clear that both components have contributed to the limited efficacy seen with this approach. In the case of the vectors, it has become clear from both preclinical and clinical experience that none of the vectors currently available (e.g., virus, bacterium, plasmid) will infect all the cells of a tumor mass. For viral vectors, this can be due to the specificity of these systems for their receptor that may or may not be present on the target tumor. More important, however, has been the complexity of the tumor. At the time of diagnosis, human solid tumors are generally marked by cellular heterogeneity and regions of high interstitial pressure and hypoxia due to the compromised vasculature of the tumor. These properties of the tumor make it difficult to deliver vectors effectively to all cells in the tumor mass. As a result, the therapeutic gene approaches that require the infection of all the cells in the tumor mass have failed as single agents to date. In addition, the initial genetic therapy approaches were dominated by the use of a single gene as the therapeutic. As the appreciation of the complexity and heterogeneity of human tumors has increased and clinical experience with a wide array of gene- and protein-based therapeutics has continued, it has become clear that a single "magic bullet" will likely not result in significant inhibition of advanced cancers (as defined by prolongation of patient survival) or a cure as monotherapy.

Consequently, this field has moved toward enhancing the capacity of the genetic-based therapy to address this limitation. One popular approach has been the use of replicating agents (e.g., viruses and bacteria, collectively referred to as macrotherapeutics) that are naturally or genetically altered to replicate selectively in the human tumor. This approach gets around the limitation of trying to infect each cell within the tumor mass at the time of treatment, utilizing the replicating agent's ability to proliferate and spread within the tumor to effect tumor clearance. For purposes of the remainder of this review, the oncolytic virus will be followed as the example of a macrotherapeutic with the understanding that similar approaches could be applied to other replication-competent anti-cancer agents. Like for small molecules and antibodies, lead discovery can take place by screening the oncolytic virus candidates against panels of tumor cell lines representative of the target indication and relevant primary normal cells (e.g., epithelial, endothelial) with the differential lytic potential for tumor and normal cells serving as the basis to define the agent's therapeutic index.

Importantly, while the oncolytic virus approach has proven popular, it is still in its infancy relative to the small-molecule and antibody approaches. Consequently, unlike small molecules, for which "drug-like" properties have been identified, or antibodies, for which experience has moved the field to humanized antibodies and binding characterization, there is a plethora of candidate oncolytic viruses but not enough clinical experience to direct the field to an ideal virus or viruses for specific clinical indications. Consequently, viruses or virus families will need to be screened for drug-like properties (the ability to target tumors effectively after intravenous administration without dose-limiting toxicity, ability to replicate efficiently and spread within the context of the target tumor, and the ability to synergize with chemotherapy and/or radiation) to identify better their oncolytic potential. It will also be important to understand if any of the viruses being considered as candidates for the development of oncolytic viruses (e.g., adenovirus, herpes simplex virus, vaccinia virus, vesicular stomatitis virus, New Castle disease virus) have a natural tropism for any of the target tumors. It is possible that different classes or families of viruses may have properties that make them inherently better suited to treat a particular type of cancer. As biological agents, it should also be possible to derive replicating agents with enhanced desired properties (e.g., potency, selectivity) for target human tumor indications using Darwinian-type selection approaches. This has recently been demonstrated with human adenovirus ⁶⁹ and human herpes simplex virus ⁷⁰, for which mutagenesis followed by passage on a targeted tumor cell line generated viruses with enhanced potency. The strength of this type of approach is that it is nonprejudiced and is directed toward an outcome (e.g., in developing a more lytic virus) and thus allows for any parameter or combination of parameters of the viral life cycle (e.g., infectivity, replication cycle, shedding, promoter alterations) to be altered to generate the desired outcome.

In summary, lead discovery marks the stage at which the common data used for tumor target identification and validation are applied to develop therapeutics from the different classes, each with their unique strengths and challenges.

Lead optimization

Lead optimization is the stage at which the initial "hit" is transformed into an agent with pharmacological, toxicological, and pharmaceutical properties optimized for the introduction of the agent into the clinic for testing. While the goal of lead optimization is shared between the therapeutic modalities, the path to achieving this is distinctly different for each agent class. This is a pivotal point in the drug development process as the lead is developed into the drug candidate, and with it comes an associated increase in cost as the candidate drug progresses toward and into clinical development and testing.

Traditionally, drug optimization for small molecules has been driven by studies of the small molecule hit in an attempt to design improvements rationally into lead compounds. With improvements in both in vitro and in vivo models, data accumulation and systems biology, pharmaceutical profiling information can be used to select optimal candidates from a series of lead compounds, assisting in predicting drug metabolism and toxicity potentials and identifying sites amenable to optimization for enhanced potency, selectivity, or pharmacokinetic properties (reviewed in ^71,72). These types of approaches to drug optimization are especially critical to the development of successful small molecules since approximately 40% of development compounds that enter clinical trials fail to reach the market due to poor pharmaceutical properties (e.g., poor solubility, permeability, or metabolic stability, reviewed in ⁷³).

Antibodies have long held appeal as cancer therapeutics due to their specificity, high affinity, and ability to create antitumor effects (e.g., complement-mediated cytolysis and antibody-dependent, cell-mediated cytotoxicity). Like for small molecules, optimization centers on the target whose properties may differ significantly (e.g., shed, internalizing or noninternalizing, glycosylation status, abundance, role in tumorigenesis). For example, Ig isotype choice can be made based on the desired therapeutic outcome, with the IgG1 isotype the Ig of choice for triggering effector cascades (e.g., complement fixation and antibody-dependent cell-mediated cytotoxicity) and the IgG4 isotype the desired Ig for neutralizing antibodies since it is incapable of triggering these cascades. In another example, it has been suggested that whole immunoglobulins are suboptimal for solid tumor therapy due to their poor tumor penetration ⁷⁴. Antibody fragments (Fab or scFv) may represent a viable optimization step due to their reduced size and subsequent increased tumor penetration. Where target expression is low but tumor specific, antibody conjugates (e.g., radioisotopes, protein toxin–enzyme fusions) represent an optimization step to create a more potent alternative antibody-based anti-cancer approach. Optimization must also take into account manufacturing issues. For example, antibody toxin conjugates may have the advantage over radioisotopes in that they can be engineered directly into the antibody-constant region, reducing the number of manipulations to the antibody needed and the manufacturing cost. Consequently, like small molecules, antibody leads are amenable to various optimization steps to address potential pitfalls of the initial lead candidate.

Lead optimization with macrotherapeutics can take many forms to address the issues of selectivity, potency, and PK. This stems from the fact that replicating agents, especially viruses, are generally well studied and genetically pliable, with established methodologies for their random or targeted genetic manipulation. Consequently, strategies to increase potency ^{75,76,77,78,79,80,81,82}, enhance selectivity ^83,84,85, and decrease immunogenicity ⁸⁶, for instance, can be employed in a rational fashion either through genetic manipulation of the agent's endogenous genes or through the addition of exogenous therapeutic genes (reviewed in ^79,87) to assist the agent, directly or indirectly, in its ability to eradicate the tumor and associated metastasis. The capability to synergize with chemotherapy (e.g., ⁸⁸) and radiation (e.g., ⁸⁹) coupled with the capability of many of these agents naturally to circulate systemically (e.g., as evidenced by the appearance of skin blisters throughout the body for poxvirus family members) may make these agents uniquely capable of efficiently seeking out and confronting the complexity and heterogeneity associated with solid tumors and their associated metastatic lesions. Several efforts have also been made to enhance the systemic delivery of these agents through genetic approaches (reviewed in ⁹⁰), by the attachment of chemical moieties (reviewed in ⁹¹), or by using cells as carriers for these agents (reviewed in ⁹²). Consequently, lead optimization of macrotherapeutics can take place both at a genetic and at a chemical modification level. However, as mentioned in the lead discovery stage, each virus will need to be tested individually for its ability to meet the therapeutic optimum of systemic delivery, selectivity, and potency for the target tumor; capacity to express exogenous genes; and ability to synergize with radiation and standard of care chemotherapy for each tumor type pursued.

From investigational new drug candidate to the clinic

Regulatory and institutional submissions

Once discovery research has identified a product candidate for clinical testing, the product development process begins. A number of regulatory and institutional submissions must be made prior to treating the first patient (Table 1). Specific submissions must be made for all experimental therapeutics; virotherapy and genetic therapy agents require additional submissions, as well. For all experimental therapeutics, an Investigational New Drug (IND) application must be made to the Food and Drug Administration (FDA) in the United States (and to its counterparts in other countries). Due to the nature of viral and genetic therapy, with its potential for broader public health implications than nongenetic therapeutics such as monoclonal antibodies or small molecules, additional submissions must be made to institutional biosafety review boards. In addition, for clinical trials involving National Institutes of Health (NIH)-funded institutions, an application must be made to the NIH Office of Biotechnology Activities for review by the Recombinant DNA Advisory Committee.

Table 1 - Clinical development and regulatory milestones: viral and genetic therapies.

Full table

Pharmacology–Toxicology–Biodistribution data

Preclinical pharmacology

Pharmacology studies are used to demonstrate efficacy from in vitro (biochemical and cell-based) and in vivo cancer models. Typically, mechanism-of-action studies are also included ⁹³. Historically, athymic (nude) mouse–human tumor xenografts have frequently been used; tumors were generally grown subcutaneously. In contrast, viral and genetic therapies often require more sophisticated and more predictive pharmacology models and endpoints ⁹⁴. In vivo modeling may require orthotopic tumors (i.e., tumors growing in the appropriate anatomical location) and/or de novo tumors to reflect the appropriate growth pattern, extracellular milieu/matrix, and blood supply ^95,96,97,98. In addition, species-specific differences in biological effects may require that efficacy studies be performed with murine versions of an active transgene ⁹⁴. The species specificity of viral and genetic therapies may be greater than with small-molecule therapeutics, for example, because the interaction between virus and host immune response has potentially evolved over millions of years. For oncolytic replication-selective viruses, for example, pharmacokinetics and pharmacodynamics will be significantly affected by species-specific replication permissivity and immune effector interactions.

Cell lines passaged in the lab for years may not reflect tumors that develop in situ. Murine models in which tumors develop de novo may be more predictive for the clinical setting; examples include tumor suppressor gene knockout mice, conditional knockout mice, and transgenics (see above) ^95,96,99. Primary human tumor tissue explants ¹⁰⁰ or spheroids ¹⁰¹ have also been developed to mimic more closely cancer genetics and/or growth in a patient. Such model systems may be particularly important for genetic and viral therapeutics; for example, critical differences have been demonstrated between cell lines in vitro and primary in vivo tissue samples in receptor expression and/or accessibility (e.g., CAR for adenoviruses).

Good laboratory practice (GLP) toxicology and biodistribution

IND-supporting toxicology studies are typically done under GLP guidelines ⁹⁴. For small-molecule development, for example, these studies have historically been done in at least two species, including one nonrodent species (e.g., dog). Given species-specificity issues (above), however, this has not been as common a practice with viral and genetic therapies and has been assessed by FDA reviewers on a case-by-case basis. Virotherapy and genetic therapy agents pose unique points to consider since their biology and pharmacodynamics may be influenced by the species used. For example, certain human transgenes and/or oncolytic virotherapy agents will not be fully functional in nonprimate species ¹⁰². Toxicology studies with murine versions of an active transgene, for example, may be considered ¹⁰³.

Small-molecule therapeutics undergo evaluation for ADME. In contrast, viral and genetic therapy agents require evaluation of their initial tissue biodistribution and subsequent clearance in relevant animal models (typically murine) ^94,104. Germ-cell transmission data are of particular interest ¹⁰⁴. Infectious units, viral genomes, and/or gene expression can be monitored. The route of administration can mirror the intended clinical route exactly (e.g., intraperitoneal) and/or reflect a "worst case scenario" for systemic distribution (e.g., intravenous for an intratumoral injection approach). Once again, species-specific differences in biodistribution and clearance should be considered.

Process development, manufacturing, quality assurance, and control

A detailed description of Good Manufacturing Practice manufacturing and product release testing of viral and genetic therapy agents is beyond the scope of this review. Readers are referred to an excellent review article by Wisher et al. ¹⁰⁵. Of note, the development of a manufacturing and purification process must often take place in parallel with the development of assays for product characterization, safety, and release. Unique issues for oncolytic viruses arise when standard in vitro and in vivo safety screening tests are not evaluable because of nonspecific viral toxicity in the assays; neutralizing antibodies can be used to inactivate the therapeutic virus and allow for outgrowth of adventitious viruses if present. Virus-specific PCR-based assays are used to rule out the presence of known adventitious pathogenic viruses (EBV, HBV, HCV, HIV, and others).

Clinical development: phase I through III

The patient populations enrolled, the dosing regimens, and the endpoints studied vary depending on the phase of clinical development and on the product being tested. It is important to note that the targeted clinical indication for approval should be used to guide planning for Phase III approval trials, which in turn will drive the design of Phase II and Phase I trials (Table 2).

Table 2 - Phases of clinical development in oncology.

Full table

Phase I trials

The primary objectives for phase I trials are typically to determine (1) the safety and (2) the maximally tolerated dose of the agent. Secondary objectives frequently are to determine (3) the pharmacokinetics, (4) the antitumoral efficacy, and (5) the pharmacodynamics (i.e., the effects on biological endpoints in the patient) of the agent. Phase I trials study escalating doses of the agent. Whereas in noncancer indications the trial might study a single dose in normal volunteers, Phase I trials for anti-cancer agents typically study a multidose, potentially therapeutic regimen in advanced cancer patients. Cohorts of patients (typically three to six per cohort) are enrolled sequentially and are given progressively higher doses of the therapeutic. Patients typically have advanced, treatment-refractory cancers that have failed standard therapy.

New targeted agents have led to significant changes in the methods used in Phase I trials. First, given the targeted and highly selective nature of these agents, a maximally tolerated (i.e., toxic) dose may not be defined. The dose level for current and future efficacy trials may therefore not be the maximum tolerated dose. The optimal dose may therefore be either (1) the highest feasible dose (based on manufacturing capacity and/or the cost of goods) or (2) the dose at which maximal target inhibition is achieved (as defined by pharmacodynamic testing of patient samples). In addition, patients may be enrolled based on the genetic features of their tumors (e.g., overexpression of an oncogene product) rather than on their histologic type (e.g., breast versus colon). Finally, greater importance is being placed on mechanism-of-action studies. Safety and some hint of efficacy are no longer enough to move to Phase II trials. Instead, biological proof of concept for the approach and its mechanism of action are critical. These can be obtained through functional imaging studies (e.g., PET scan for tumor metabolism or dynamic magnetic resonance imaging (MRI) for tumor blood flow) or through analysis of patient tissues (e.g., normal peripheral blood mononuclear cells or tumor biopsies). The activity, distribution, and shedding of viral and genetic therapies can be assessed through analysis of blood, tumor, urine, etc. Investigators often analyze viral genomes, infectious units, or gene expression ^{106,107,108,109,110,111,112}. In addition, for these agents it is often important to study both the innate (e.g., cytokines, complement, natural killer cells) and the acquired (humoral, cell-mediated) immune responses to the virus and/or transgene ^106,110,113.

Phase II clinical trials

The primary objectives for phase II trials are typically (1) to determine the efficacy and toxicity profile of the agent in the appropriate dose and treatment regimen, (2) to make a "go–no go" decision regarding initiation of large and expensive Phase III trials, (3) to determine which efficacy endpoints to use in Phase III, and (4) to determine the necessary number of patients to accrue in the Phase III trial. Efficacy endpoints frequently include objective response rate (complete and partial), time-to-tumor progression, progression-free survival, overall survival, clinical benefit (e.g., predefined improvements in pain, weight, performance status), and quality of life (QOL) (e.g., using previously validated QOL instruments). The relevant endpoints, and a clinically meaningful effect on such endpoints, should be determined by clinical investigators for the specific cancer patient population. These will vary greatly based on the specific tumor type, stage, and prior treatments. Phase II trials can be single arm or randomized. If randomized, the arms can be experimental agent vs standard of care (e.g., placebo) or two different doses of the experimental agent. Randomization is important if, for example, the behavior of the control group is unpredictable or the optimal dose for Phase III is unclear. Randomized Phase II trials are not necessarily large enough to show statistically significant results (this is saved for Phase III). They can be used to design the large, well-powered (i.e., adequately large) Phase III randomized trial(s) and to make the go–no go decision for Phase III ¹¹⁴. In contrast, single-arm Phase II trials are used when the experimental agent causes tumor responses in a patient population with no available therapy (i.e., terminal patients), since untreated controls would not have tumor responses. Phase II trials can also be in combination with standard therapy (e.g., radiotherapy, chemotherapy, monoclonal antibodies, etc.) ^88,115. Randomization can be to standard therapy alone vs standard therapy plus experimental agent.

Many novel cancer treatment agents cause cancer stabilization rather than tumor shrinkage and are therefore termed "static" ¹¹⁶. Go–no go decisions for Phase III with static agents are particularly difficult ¹¹⁷. Since long-term disease stabilization can occur in some patients without therapy, stable disease is difficult to interpret from nonrandomized Phase II trials ¹¹⁸. Randomized Phase II trials have the potential problems of requiring many more patients and the requirement that some patients receive inactive control treatment. Ratain and colleagues designed an innovative approach to this clinical problem termed the "randomized discontinuation design" ¹¹⁹. With this design, all patients receive active therapy and at a predetermined time point in treatment are classified as responders, stable disease, or progressive disease. Responders continue on the therapy, whereas patients with progression go off study. However, patients with stable disease are randomized to continuation of experimental therapy or discontinuation. The time to tumor progression is then compared between the two groups to determine whether the treatment significantly causes tumor stabilization. Patients progressing off of therapy can have treatment reinitiated. This design is representative of innovative clinical trial designs being developed for evaluation of novel therapies for cancer.

Phase II-B (accelerated) and Phase III approval trials

Two general approaches are available for the approval of cancer agents ¹²⁰. Standard, full regulatory approval typically requires a Phase III randomized trial showing a statistically significant and clinically meaningful advantage to the trial arm with the experimental therapy; endpoints usually include survival and/or clinical benefit/QOL (see endpoints above) ¹²⁰. Rarely, very large studies can be done to demonstrate equivalent efficacy and reduced toxicity, although this approach is difficult to use due to the relatively large number of patients required statistically. A second approach can lead to "accelerated" approval if no effective therapies exist for the terminal patient population enrolled ^120,121,122. In this case approval would be based on a single-arm trial in which efficacy that would be expected to result in patient benefit is demonstrated (e.g., objective responses in >15% of patients). In this case, sponsors must agree with the FDA to perform subsequent confirmatory trials with the agent.

It is important to note that clinically meaningful endpoints are independent of the mechanism of action of the agent or its functional class. Although novel imaging modalities (e.g., PET scanning; dynamic contrast-enhanced MRI) or blood markers (e.g., prostate-specific antigen for prostate cancer) may be available ¹¹⁶, these will not be the basis for approval until they are validated clinically. All clinical trial designs and analyses should be done in close partnership with the FDA and other international regulatory agencies.

Approval trials of novel therapeutics for cancer are far more likely to demonstrate efficacy if patients most likely to benefit are enrolled. For example, the monoclonal antibody trastuzumab (Herceptin; Genentech) has a response rate of approximately 15% in breast cancer patients whose tumors markedly overexpress erbB/HER2 (20% of all breast cancer patients), in contrast to a response rate of <5% in those that do not have overexpression. Therefore, the expected response rate in a nonselected patient population would be <10%. Interestingly, the small-molecule epidermal growth factor (EGF) receptor inhibitor gefitinib (Iressa; Astra Zeneca) appears to have selective activity in EGF receptor-mutated tumors, whereas overall expression levels of EGF receptor do not appear to correlate with response.

Combination treatment regimens in clinical trials

Most standard treatment regimens for cancer involve the simultaneous or sequential application of distinct therapeutic agents (e.g., chemotherapy combinations) or distinct therapeutic modalities (e.g., surgery followed by adjuvant chemotherapy, chemotherapy–radiotherapy combinations, antibody–chemotherapy combinations). Efficacy can potentially be increased, and the development of resistance reduced, by combining agents with different mechanisms of action. Therefore, combination clinical trials often play a central role in the clinical development of an agent. Ideally, combinations are designed to optimize chances for therapeutic synergy and avoid overlapping toxicities. Issues to be addressed with this approach are numerous, however. First, the optimal agent(s) to combine with the novel therapy must be defined, ideally based on both a mechanistic rationale and efficacy data in vivo. However, for many cancer indications the standard of care will drive the design of the combination regimen. The optimal sequencing of agents should be defined as early on in development as possible to avoid large-scale Phase III failures (e.g., small-molecule epidermal growth factor receptor inhibitors and chemotherapy combinations in non-small-cell lung cancer). Phase I combination trials typically study the approved standard of care (e.g., chemotherapy) at the standard dose and schedule while performing a dose escalation of the experimental therapy up to the maximum tolerated dose. Efficacy and toxicity analyses on such trials must take into account the efficacy and toxicities expected with the standard agent(s) alone. As such these trials often require either a control group (standard agent only) or a very large patient population compared to single-agent Phase II trials. One approach to identify expeditiously combination efficacy that is dependent on the experimental agent itself is to enroll patients who are proven refractory to the standard agent already (see C225/Erbitux antibody in combination with irinotecan in colorectal carcinoma ¹²³; ONYX-015 plus cisplatin/5-fluorouracil in chemotherapy-refractory patients ⁸⁸). Proving cancers are refractory prior to study entry must be carefully planned, however. Another unique approach to this issue when testing therapy that is locally but not systemically active (e.g., local injection of genetic agents) is to inject one tumor mass while leaving other(s) uninjected ⁸⁸. In this way the response of injected vs matched uninjected tumors can be compared following standard treatment with chemotherapy (e.g., ONYX-015 in a Phase II trial in patients with squamous cell carcinoma of the head and neck) or radiotherapy (e.g., TNFerade in a Phase I trial in patients with multiple solid tumor types; Genvec Inc., Gaithersburg, MD, USA) ¹²⁴.

Summary

The development of a drug from inception of the concept and identification of the target to the realization of a clinically beneficial drug is long, highly labor intensive, and extremely costly. The approaches to developing these agents have changed considerably over time and will continue to evolve in the future. The drug development process outlined briefly in this review is meant to serve as a roadmap to how highly innovative ideas and approaches across various drug platforms have converged toward standard rigorous methodologies to develop therapeutics in a time and cost-efficient manner. Investigators in the molecular therapeutics field can learn a great deal from the experiences of their colleagues working on different platforms.

References

1. Balmain, A. (2002). Cancer as a complex genetic trait: tumor susceptibility in humans and mouse models. Cell 108: 145–152. | Article | PubMed | ISI | ChemPort |
2. Demant, P. (2003). Cancer susceptibility in the mouse: genetics, biology and implications for human cancer. Nat. Rev. Genet. 4: 721–734. | Article | PubMed | ISI | ChemPort |
3. Dragani, T. A. (2003). 10 years of mouse cancer modifier loci: human relevance. Cancer Res. 63: 3011–3018. | PubMed | ISI | ChemPort |
4. Cunningham, M. J. (2000). Genomics and proteomics: the new millennium of drug discovery and development. J. Pharmacol. Toxicol. Methods 44: 291–300. | Article | PubMed | ISI | ChemPort |
5. Clarke, P. A., te Poele, R., Wooster, R. and Workman, P. (2001). Gene expression microarray analysis in cancer biology, pharmacology, and drug development: progress and potential. Biochem. Pharmacol. 62: 1311–1336. | Article | PubMed | ISI | ChemPort |
6. Chen, G. et al. (2002). Discordant protein and mRNA expression in lung adenocarcinomas. Mol. Cell. Proteom. 1: 304–313. | ChemPort |
7. Anderson, L. and Seilhamer, J. (1997). A comparison of selected mRNA and protein abundances in human liver. Electrophoresis 18: 533–537. | Article | PubMed | ISI | ChemPort |
8. Gygi, S. P., Han, D. K., Gingras, A. C., Sonenberg, N. and Aebersold, R. (1999). Protein analysis by mass spectrometry and sequence database searching: tools for cancer research in the post-genomic era. Electrophoresis 20: 310–319. | Article | PubMed | ISI | ChemPort |
9. Figeys, D. (2003). Novel approaches to map protein interactions. Curr. Opin. Biotechnol. 14: 119–125. | Article | PubMed | ChemPort |
10. Petricoin, E. F., Zoon, K. C., Kohn, E. C., Barrett, J. C. and Liotta, L. A. (2002). Clinical proteomics: translating benchside promise into bedside reality. Nat. Rev. Drug Discovery 1: 683–695. | ChemPort |
11. Tyers, M. and Mann, M. (2003). From genomics to proteomics. Nature 422: 193–197. | Article | PubMed | ISI | ChemPort |
12. Zhu, H., Bilgin, M. and Snyder, M. (2003). Proteomics. Annu. Rev. Biochem 72: 783–812. | Article | PubMed | ISI | ChemPort |
13. Chanda, S. K. and Caldwell, J. S. (2003). Fulfilling the promise: drug discovery in the post-genomic era. Drug Discovery Today 8: 168–174. | Article | PubMed | ChemPort |
14. Alessi, P., Ebbinghaus, C. and Neri, D. (2004). Molecular targeting of angiogenesis. Biochim. Biophys. Acta 1654: 39–49. | PubMed | ChemPort |
15. Bicknell, R. and Harris, A. L. (2004). Novel angiogenic signaling pathways and vascular targets. Annu. Rev. Pharmacol. Toxicol. 44: 219–238. | Article | PubMed | ISI | ChemPort |
16. Carver, L. A. and Schnitzer, J. E. (2003). Caveolae: mining little caves for new cancer targets. Nat. Rev. Cancer 3: 571–581. | Article | PubMed | ISI | ChemPort |
17. Nanda, A. and St Croix, B. (2004). Tumor endothelial markers: new targets for cancer therapy. Curr. Opin. Oncol. 16: 44–49. | Article | PubMed | ISI | ChemPort |
18. Siemann, D. W., Chaplin, D. J. and Horsman, M. R. (2004). Vascular-targeting therapies for treatment of malignant disease. Cancer 100: 2491–2499. | Article | PubMed | ISI | ChemPort |
19. Sawyer, T. K. (2004). Cancer metastasis therapeutic targets and drug discovery: emerging small-molecule protein kinase inhibitors. Expert Opin. Invest. Drugs 13: 1–19. | Article | ISI | ChemPort |
20. Bissell, M. J. and Radisky, D. (2001). Putting tumours in context. Nat. Rev. Cancer 1: 46–54. | Article | PubMed | ChemPort |
21. Coussens, L. M., Fingleton, B. and Matrisian, L. M. (2002). Matrix metalloproteinase inhibitors and cancer: trials and tribulations. Science 295: 2387–2392. | Article | PubMed | ISI | ChemPort |
22. Erickson, A. C. and Barcellos-Hoff, M. H. (2003). The not-so innocent bystander: the microenvironment as a therapeutic target in cancer. Expert Opin. Ther. Targets 7: 71–88. | Article | PubMed | ChemPort |
23. Ramnath, N. and Creaven, P. J. (2004). Matrix metalloproteinase inhibitors. Curr. Oncol. Rep. 6: 96–102. | PubMed |
24. Sanz, L. and Alvarez-Vallina, L. (2003). The extracellular matrix: a new turn-of-the-screw for anti-angiogenic strategies. Trends Mol. Med. 9: 256–262. | Article | PubMed | ChemPort |
25. Singer, C. F., Hudelist, G., Galid, A. and Kubista, E. (2003). Pharmacological modulation of local feedback mechanisms as a therapeutic approach in breast cancer treatment. Drugs Today (Barcelona) 39: 917–926. | ChemPort |
26. Sung, S. Y. and Chung, L. W. (2002). Prostate tumor-stroma interaction: molecular mechanisms and opportunities for therapeutic targeting. Differentiation 70: 506–521. | Article | PubMed | ChemPort |
27. Vihinen, P. and Kahari, V. M. (2002). Matrix metalloproteinases in cancer: prognostic markers and therapeutic targets. Int. J. Cancer 99: 157–166. | Article | PubMed | ISI | ChemPort |
28. Baker, B. F. and Monia, B. P. (1999). Novel mechanisms for antisense-mediated regulation of gene expression. Biochim. Biophys. Acta 1489: 3–18. | PubMed | ChemPort |
29. Crooke, S. T. (1999). Molecular mechanisms of action of antisense drugs. Biochim. Biophys. Acta 1489: 31–44. | Article | PubMed | ISI | ChemPort |
30. Koller, E., Gaarde, W. A. and Monia, B. P. (2000). Elucidating cell signaling mechanisms using antisense technology. Trends Pharmacol. Sci. 21: 142–148. | Article | PubMed | ISI | ChemPort |
31. Khan, A. U. and Lal, S. K. (2003). Ribozymes: a modern tool in medicine. J. Biomed. Sci. 10: 457–467. | Article | PubMed | ChemPort |
32. Kim, K. J. et al. (1993). Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumour growth in vivo. Nature 362: 841–844. | Article | PubMed | ISI | ChemPort |
33. Brody, E. N. and Gold, L. (2000). Aptamers as therapeutic and diagnostic agents. J. Biotechnol. 74: 5–13. | Article | PubMed | ChemPort |
34. Burgstaller, P., Jenne, A. and Blind, M. (2002). Aptamers and aptazymes: accelerating small molecule drug discovery. Curr. Opin. Drug Discovery Dev. 5: 690–700. | ChemPort |
35. Butz, K. et al. (2001). Peptide aptamers targeting the hepatitis B virus core protein: a new class of molecules with antiviral activity. Oncogene 20: 6579–6586. | Article | PubMed | ChemPort |
36. Hannon, G. J. (2002). RNA interference. Nature 418: 244–251. | Article | PubMed | ISI | ChemPort |
37. McManus, M. T., Petersen, C. P., Haines, B. B., Chen, J. and Sharp, P. A. (2002). Gene silencing using micro-RNA designed hairpins. RNA 8: 842–850. | Article | PubMed | ISI | ChemPort |
38. Bernstein, E., Caudy, A. A., Hammond, S. M. and Hannon, G. J. (2001). Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature 409: 363–366. | Article | PubMed | ISI | ChemPort |
39. Hammond, S. M., Boettcher, S., Caudy, A. A., Kobayashi, R. and Hannon, G. J. (2001). Argonaute2: a link between genetic and biochemical analyses of RNAi. Science 293: 1146–1150. | Article | PubMed | ISI | ChemPort |
40. Berns, K. et al. (2004). A large-scale RNAi screen in human cells identifies new components of the p53 pathway. Nature 428: 431–437. | Article | PubMed | ISI | ChemPort |
41. Baron, U. and Bujard, H. (2000). Tet repressor-based system for regulated gene expression in eukaryotic cells: principles and advances. Methods Enzymol. 327: 401–421. | Article | PubMed | ISI | ChemPort |
42. Jonkers, J. and Berns, A. (2002). Conditional mouse models of sporadic cancer. Nat. Rev. Cancer 2: 251–265. | Article | PubMed | ISI | ChemPort |
43. Lewandoski, M. (2001). Conditional control of gene expression in the mouse. Nat. Rev. Genet. 2: 743–755. | Article | PubMed | ISI | ChemPort |
44. Ostrand-Rosenberg, S. (2004). Animal models of tumor immunity, immunotherapy and cancer vaccines. Curr. Opin. Immunol. 16: 143–150. | Article | PubMed | ISI | ChemPort |
45. Russell, P. J. and Voeks, D. J. (2003). Animal models of prostate cancer. Methods Mol. Med. 81: 89–112. | PubMed | ChemPort |
46. Powell, W. C., Cardiff, R. D., Cohen, M. B., Miller, G. J. and Roy-Burman, P. (2003). Mouse strains for prostate tumorigenesis based on genes altered in human prostate cancer. Curr. Drug Targets 4: 263–279. | Article | PubMed | ChemPort |
47. Rosol, T. J., Tannehill-Gregg, S. H., Corn, S., Schneider, A. and McCauley, L. K. (2004). Animal models of bone metastasis. Cancer Treat. Res. 118: 47–81. | PubMed |
48. MG, D. A., Afanasieva, T. and Aguzzi, A. (2000). Angiogenesis in transgenic models of multistep carcinogenesis. J. Neurooncol. 50: 89–98. | PubMed |
49. Hanahan, D., Christofori, G., Naik, P. and Arbeit, J. (1996). Transgenic mouse models of tumour angiogenesis: the angiogenic switch, its molecular controls, and prospects for preclinical therapeutic models. Eur. J. Cancer 32A: 2386–2393.
50. Hansen, K. and Khanna, C. (2004). Spontaneous and genetically engineered animal models: use in preclinical cancer drug development. Eur. J. Cancer 40: 858–880. | Article | PubMed | ISI | ChemPort |
51. Budinger, T. F., Benaron, D. A. and Koretsky, A. P. (1999). Imaging transgenic animals. Annu. Rev. Biomed. Eng. 1: 611–648. | Article | PubMed | ISI | ChemPort |
52. Lewis, J. S., Achilefu, S., Garbow, J. R., Laforest, R. and Welch, M. J. (2002). Small animal imaging. current technology and perspectives for oncological imaging. Eur. J. Cancer 38: 2173–2188. | Article | PubMed | ISI |
53. Weissleder, R. (2002). Scaling down imaging: molecular mapping of cancer in mice. Nat. Rev. Cancer 2: 11–18. | Article | PubMed | ISI | ChemPort |
54. Simon, R. and Sauter, G. (2003). Tissue microarray (TMA) applications: implications for molecular medicine. Expert Rev. Mol. Med. 2003: 1–12. | Article | PubMed |
55. Bubendorf, L., Nocito, A., Moch, H. and Sauter, G. (2001). Tissue microarray (TMA) technology: miniaturized pathology archives for high-throughput in situ studies. J. Pathol. 195: 72–79. | Article | PubMed | ISI | ChemPort |
56. Kononen, J. et al. (1998). Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nat. Med. 4: 844–847. | Article | PubMed | ISI | ChemPort |
57. Oprea, T. I. (2002). Chemical space navigation in lead discovery. Curr. Opin. Chem. Biol. 6: 384–389. | Article | PubMed | ChemPort |
58. Kerns, E. H. and Di, L. (2003). Pharmaceutical profiling in drug discovery. Drug Discovery Today 8: 316–323. | Article | PubMed | ChemPort |
59. Wenlock, M. C., Austin, R. P., Barton, P., Davis, A. M. and Leeson, P. D. (2003). A comparison of physiochemical property profiles of development and marketed oral drugs. J. Med. Chem. 46: 1250–1256. | Article | PubMed | ISI | ChemPort |
60. Fattori, D. (2004). Molecular recognition: the fragment approach in lead generation. Drug Discovery Today 9: 229–238. | Article | PubMed | ChemPort |
61. Green, D. V. (2003). Virtual screening of virtual libraries. Prog. Med. Chem. 41: 61–97. | PubMed | ChemPort |
62. Lengauer, T., Lemmen, C., Rarey, M. and Zimmermann, M. (2004). Novel technologies for virtual screening. Drug Discovery Today 9: 27–34. | Article | PubMed | ChemPort |
63. Fishelson, Z., Donin, N., Zell, S., Schultz, S. and Kirschfink, M. (2003). Obstacles to cancer immunotherapy: expression of membrane complement regulatory proteins (mCRPs) in tumors. Mol. Immunol. 40: 109–123. | Article | PubMed | ISI | ChemPort |
64. Houdebine, L. M. (2002). Antibody manufacture in transgenic animals and comparisons with other systems. Curr. Opin. Biotechnol. 13: 625–629. | Article | PubMed | ChemPort |
65. Kellermann, S. A. and Green, L. L. (2002). Antibody discovery: the use of transgenic mice to generate human monoclonal antibodies for therapeutics. Curr. Opin. Biotechnol. 13: 593–597. | Article | PubMed | ChemPort |
66. Kretzschmar, T. and von Ruden, T. (2002). Antibody discovery: phage display. Curr. Opin. Biotechnol. 13: 598–602. | Article | PubMed | ISI | ChemPort |
67. Lonberg, N. et al. (1994). Antigen-specific human antibodies from mice comprising four distinct genetic modifications. Nature 368: 856–859. | Article | PubMed | ISI | ChemPort |
68. van Dijk, M. A. and van de Winkel, J. G. (2001). Human antibodies as next generation therapeutics. Curr. Opin. Chem. Biol. 5: 368–374. | Article | PubMed | ISI | ChemPort |
69. Yan, W. et al. (2003). Developing novel oncolytic adenoviruses through bioselection. J. Virol. 77: 2640–2650. | Article | PubMed | ISI | ChemPort |
70. Taneja, S., MacGregor, J., Markus, S., Ha, S. and Mohr, I. (2001). Enhanced antitumor efficacy of a herpes simplex virus mutant isolated by genetic selection in cancer cells. Proc. Natl. Acad. Sci. USA 98: 8804–8808. | Article | PubMed | ChemPort |
71. Theil, F. P., Guentert, T. W., Haddad, S. and Poulin, P. (2003). Utility of physiologically based pharmacokinetic models to drug development and rational drug discovery candidate selection. Toxicol. Lett. 138: 29–49. | Article | PubMed | ChemPort |
72. Bugrim, A., Nikolskaya, T. and Nikolsky, Y. (2004). Early prediction of drug metabolism and toxicity: systems biology approach and modeling. Drug Discovery Today 9: 127–135. | Article | PubMed | ChemPort |
73. Venkatesh, S. and Lipper, R. A. (2000). Role of the development scientist in compound lead selection and optimization. J. Pharm. Sci. 89: 145–154. | Article | PubMed | ISI | ChemPort |
74. Colcher, D., Pavlinkova, G., Beresford, G., Booth, B. J., Choudhury, A. and Batra, S. K. (1998). Pharmacokinetics and biodistribution of genetically-engineered antibodies. Q. J. Nucl. Med. 42: 225–241. | PubMed | ISI | ChemPort |
75. Barnett, B. G., Tillman, B. W., Curiel, D. T. and Douglas, J. T. (2002). Dual targeting of adenoviral vectors at the levels of transduction and transcription enhances the specificity of gene expression in cancer cells. Mol. Ther. 6: 377–385. | Article | PubMed | ISI | ChemPort |
76. Doronin, K., Toth, K., Kuppuswamy, M., Ward, P., Tollefson, A. E. and Wold, W. S. (2000). Tumor-specific, replication-competent adenovirus vectors overexpressing the adenovirus death protein. J. Virol. 74: 6147–6155. | Article | PubMed | ISI | ChemPort |
77. Harrison, D., Sauthoff, H., Heitner, S., Jagirdar, J., Rom, W. N. and Hay, J. G. (2001). Wild-type adenovirus decreases tumor xenograft growth, but despite viral persistence complete tumor responses are rarely achieved—deletion of the viral E1b-19-kD gene increases the viral oncolytic effect. Hum. Gene Ther. 12: 1323–1332. | Article | PubMed | ISI | ChemPort |
78. Mi, J., Li, Z. Y., Ni, S., Steinwaerder, D. and Lieber, A. (2001). Induced apoptosis supports spread of adenovirus vectors in tumors. Hum. Gene Ther. 12: 1343–1352. | Article | PubMed | ChemPort |
79. Hermiston, T. W. and Kuhn, I. (2002). Armed therapeutic viruses: strategies and challenges to arming oncolytic viruses with therapeutic genes. Cancer Gene Ther. 9: 1022–1035. | Article | PubMed | ISI | ChemPort |
80. Sauthoff, H., Heitner, S., Rom, W. N. and Hay, J. G. (2000). Deletion of the adenoviral E1b-19kD gene enhances tumor cell killing of a replicating adenoviral vector. Hum. Gene Ther. 11: 379–388. | Article | PubMed | ISI | ChemPort |
81. Suzuki, K., Fueyo, J., Krasnykh, V., Reynolds, P. N., Curiel, D. T. and Alemany, R. (2001). A conditionally replicative adenovirus with enhanced infectivity shows improved oncolytic potency. Clin. Cancer Res. 7: 120–126. | PubMed | ISI | ChemPort |
82. Fu, X., Tao, L., Jin, A., Vile, R., Brenner, M. K. and Zhang, X. (2003). Expression of a fusogenic membrane glycoprotein by an oncolytic herpes simplex virus potentiates the viral antitumor effect. Mol. Ther. 7: 748–754. | Article | PubMed | ISI | ChemPort |
83. Stojdl, D. F. et al. (2003). VSV strains with defects in their ability to shutdown innate immunity are potent systemic anti-cancer agents. Cancer Cell 4: 263–275. | Article | PubMed | ISI | ChemPort |
84. Bischoff, J. R. et al. (1996). An adenovirus mutant that replicates selectively in p53-deficient human tumor cells. Science 274: 373–376. | Article | PubMed | ISI | ChemPort |
85. Rodriguez, R., Schuur, E. R., Lim, H. Y., Henderson, G. A., Simons, J. W. and Henderson, D. R. (1997). Prostate attenuated replication competent adenovirus (ARCA) CN706: a selective cytotoxic for prostate-specific antigen-positive prostate cancer cells. Cancer Res. 57: 2559–2563. | PubMed | ISI | ChemPort |
86. Wang, Y. et al. (2003). E3 gene manipulations affect oncolytic adenovirus activity in immunocompetent tumor models. Nat. Biotechnol. 21: 1328–1335. | Article | PubMed | ISI | ChemPort |
87. Hermiston, T. (2000). Gene delivery from replication-selective viruses: arming guided missiles in the war against cancer. J. Clin. Invest. 105: 1169–1172. | PubMed | ISI | ChemPort |
88. Khuri, F. R. et al. (2000). A controlled trial of intratumoral ONYX-015, a selectively-replicating adenovirus, in combination with cisplatin and 5-fluorouracil in patients with recurrent head and neck cancer. Nat. Med. 6: 879–885. | Article | PubMed | ISI | ChemPort |
89. Freytag, S. O., Rogulski, K. R., Paielli, D. L., Gilbert, J. D. and Kim, J. H. (1998). A novel three-pronged approach to kill cancer cells selectively: concomitant viral, double suicide gene, and radiotherapy. Hum. Gene Ther. 9: 1323–1333. | PubMed | ISI | ChemPort |
90. Galanis, E., Vile, R. and Russell, S. J. (2001). Delivery systems intended for in vivo gene therapy of cancer: targeting and replication competent viral vectors. Crit. Rev. Oncol. Hematol. 38: 177–192. | Article | PubMed | ChemPort |
91. Green, N. K. and Seymour, L. W. (2002). Adenoviral vectors: systemic delivery and tumor targeting. Cancer Gene Ther. 9: 1036–1042. | Article | PubMed | ISI | ChemPort |
92. Harrington, K. et al. (2002). Cells as vehicles for cancer gene therapy: the missing link between targeted vectors and systemic delivery?. Hum. Gene Ther. 13: 1263–1280. | Article | PubMed | ISI | ChemPort |
93. Heise, C., Sampson-Johannes, A., Williams, A., McCormick, F., Von Hoff, D. D. and Kirn, D. H. (1997). ONYX-015, an E1B gene-attenuated adenovirus, causes tumor-specific cytolysis and antitumoral efficacy that can be augmented by standard chemotherapeutic agents. Nat. Med. 3: 639–645. | Article | PubMed | ISI | ChemPort |
94. Pilaro, A. M. and Serabian, M. A. (1999). Preclinical development strategies for novel gene therapeutic products. Toxicol. Pathol. 27: 4–7. | PubMed | ChemPort |
95. Inoue, K., Chikazawa, M., Fukata, S., Yoshikawa, C. and Shuin, T. (2002). Frequent administration of angiogenesis inhibitor TNP-470 (AGM-1470) at an optimal biological dose inhibits tumor growth and metastasis of metastatic human transitional cell carcinoma in the urinary bladder. Clin. Cancer Res. 8: 2389–2398. | PubMed | ISI | ChemPort |
96. Hingorani, S. R. et al. (2003). Preinvasive and invasive ductal pancreatic cancer and its early detection in the mouse. Cancer Cell 4: 437–450. | Article | PubMed | ISI | ChemPort |
97. Heise, C., Ganly, I., Kim, Y. T., Sampson-Johannes, A., Brown, R. and Kirn, D. (2000). Efficacy of a replication-selective adenovirus against ovarian carcinomatosis is dependent on tumor burden, viral replication and p53 status. Gene Ther. 7: 1925–1929. | Article | PubMed | ChemPort |
98. Yamasaki, T. et al. (1996). Natural resistance against tumors grafted into the brain in association with histocompatibility-class-I-antigen expression. Int. J. Cancer 67: 365–371. | Article | PubMed | ChemPort |
99. Nikitin, A. Y. et al. (2004). Classification of proliferative pulmonary lesions of the mouse: recommendations of the Mouse Models of Human Cancers Consortium. Cancer Res. 64: 2307–2316. | Article | PubMed | ChemPort |
100. Buskens, C. J. et al. (2003). A genetically retargeted adenoviral vector enhances viral transduction in esophageal carcinoma cell lines and primary cultured esophageal resection specimens. Ann. Surg. 238: 815–824. (discussion 825-816.) | PubMed |
101. van Beusechem, V. W. et al. (2003). Conditionally replicative adenovirus expressing a targeting adapter molecule exhibits enhanced oncolytic potency on CAR-deficient tumors. Gene Ther. 10: 1982–1991. | Article | PubMed | ChemPort |
102. Hallden, G. et al. (2003). Novel immunocompetent murine tumor models for the assessment of replication-competent oncolytic adenovirus efficacy. Mol. Ther. 8: 412–424. | Article | PubMed | ISI | ChemPort |
103. Rasmussen, H. et al. (2002). TNFerade biologic: preclinical toxicology of a novel adenovector with a radiation-inducible promoter, carrying the human tumor necrosis factor alpha gene. Cancer Gene Ther. 9: 951–957. | Article | PubMed | ISI | ChemPort |
104. Paielli, D. L. et al. (2000). Evaluation of the biodistribution, persistence, toxicity, and potential of germ-line transmission of a replication-competent human adenovirus following intraprostatic administration in the mouse. Mol. Ther. 1: 263–274. | Article | PubMed | ISI | ChemPort |
105. Wisher, M. (2002). Biosafety and product release testing issues relevant to replication-competent oncolytic viruses. Cancer Gene Ther. 9: 1056–1061. | Article | PubMed | ChemPort |
106. Reid, T. et al. (2002). Hepatic arterial infusion of a replication-selective oncolytic adenovirus (dl1520): phase II viral, immunologic, and clinical endpoints. Cancer Res. 62: 6070–6079. | PubMed | ISI | ChemPort |
107. Reid, T. et al. (2001). Intra-arterial administration of a replication-selective adenovirus (dl1520) in patients with colorectal carcinoma metastatic to the liver: a phase I trial. Gene Ther. 8: 1618–1626. | Article | PubMed | ISI | ChemPort |
108. Makower, D. et al. (2003). Phase II clinical trial of intralesional administration of the oncolytic adenovirus ONYX-015 in patients with hepatobiliary tumors with correlative p53 studies. Clin. Cancer Res. 9: 693–702. | PubMed | ISI |
109. Markert, J. M. et al. (2000). Conditionally replicating herpes simplex virus mutant, G207, for the treatment of malignant glioma: results of a phase I trial. Gene Ther. 7: 867–874. | Article | PubMed | ISI | ChemPort |
110. Mastrangelo, M. J. et al. (1999). Intratumoral recombinant GM-CSF-encoding virus as gene therapy in patients with cutaneous melanoma. Cancer Gene Ther. 6: 409–422. | Article | PubMed | ISI | ChemPort |
111. DeWeese, T. L. et al. (2001). A phase I trial of CV706, a replication-competent, PSA selective oncolytic adenovirus, for the treatment of locally recurrent prostate cancer following radiation therapy. Cancer Res. 61: 7464–7472. | PubMed | ISI | ChemPort |
112. Freytag, S. O. et al. (2002). Phase I study of replication-competent adenovirus-mediated double suicide gene therapy for the treatment of locally recurrent prostate cancer. Cancer Res. 62: 4968–4976. | PubMed | ISI | ChemPort |
113. Nemunaitis, J. et al. (2000). Selective replication and oncolysis in p53 mutant tumors with ONYX-015, an E1B-55kD gene-deleted adenovirus, in patients with advanced head and neck cancer: a phase II trial. Cancer Res. 60: 6359–6366. | PubMed | ISI | ChemPort |
114. Stadler, W. M. and Ratain, M. J. (2000). Development of target-based antineoplastic agents. Invest. New Drugs 18: 7–16. | Article | PubMed | ChemPort |
115. Freytag, S. O. et al. (2003). Phase I study of replication-competent adenovirus-mediated double-suicide gene therapy in combination with conventional-dose three-dimensional conformal radiation therapy for the treatment of newly diagnosed, intermediate- to high-risk prostate cancer. Cancer Res. 63: 7497–7506. | PubMed | ISI | ChemPort |
116. Hermiston, T. and Kirn, D. (2004). AACR-NCL-EORTC International Conference on Molecular Targets and Cancer Therapeutics. Mol. Ther. 9: 144–146. | PubMed | ChemPort |
117. Roberts, T. G.Jr. , Lynch, T. J.Jr. and Chabner, B. A. (2003). The phase III trial in the era of targeted therapy: unraveling the "go or no go" decision. J. Clin. Oncol. 21: 3683–3695. | Article | PubMed | ISI | ChemPort |
118. Johnson, D. H. and Arteaga, C. L. (2003). Gefitinib in recurrent non-small-cell lung cancer: an IDEAL trial?. J. Clin. Oncol. 21: 2227–2229. | Article | PubMed | ISI | ChemPort |
119. Rosner, G. L., Stadler, W. and Ratain, M. J. (2002). Randomized discontinuation design: application to cytostatic antineoplastic agents. J. Clin. Oncol. 20: 4478–4484. | Article | PubMed | ISI | ChemPort |
120. Johnson, J. R. et al. (2003). Approval summary: imatinib mesylate capsules for treatment of adult patients with newly diagnosed Philadelphia chromosome-positive chronic myelogenous leukemia in chronic phase. Clin. Cancer Res. 9: 1972–1979. | PubMed | ChemPort |
121. Kane, R. C., Bross, P. F., Farrell, A. T. and Pazdur, R. (2003). Velcade: U.S. FDA approval for the treatment of multiple myeloma progressing on prior therapy. Oncologist 8: 508–513. | Article | PubMed | ISI |
122. Cohen, M. H. et al. (2004). United States Food and Drug Administration drug approval summary: gefitinib (ZD1839; Iressa) tablets. Clin. Cancer Res. 10: 1212–1218. | Article | PubMed | ISI | ChemPort |
123. O'Dwyer, P. J. and Benson, A. B. III (2002). Epidermal growth factor receptor-targeted therapy in colorectal cancer. Semin. Oncol. 29: 10–17. | Article | PubMed | ChemPort |
124. Senzer, N. et al. (2004). TNFerade biologic, an adenovector with a radiation-inducible promoter, carrying the human tumor necrosis factor alpha gene: a phase I study in patients with solid tumors. J. Clin. Oncol. 22: 592–601. | Article | PubMed | ISI | ChemPort |