The Oncologist, Vol. 7, No. 2, 106-119,
April 2002
© 2002 AlphaMed Press
Viral Oncolysis
John T. Mullena,
Kenneth K. Tanabea,b
a Division of Surgical Oncology and
b MGH Cancer Center, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
Correspondence:
Kenneth K. Tanabe, M.D., Division of Surgical Oncology, Massachusetts General Hospital, Cox 626, 100 Blossom Street, Boston, Massachusetts 02114-2696, USA. Telephone: 617-724-3868; Fax: 617-724-3895; e-mail: ktanabe{at}partners.org
 |
ABSTRACT
|
|---|
The concept of using replicating viruses as anticancer agents is not a new one, but the ability to genetically modify these viruses into increasingly potent and tumor-specific vectors is a recent phenomenon. As more is learned about the functions of viral gene products in controlling the mammalian cell cycle and in disabling cellular defense mechanisms, specific viral functions can be augmented or eliminated to enhance antineoplastic efficacy. In this article, general mechanisms by which oncolytic viruses achieve their antitumor efficacy and specificity are reviewed. The paradoxical roles of the immune response are addressed with respect to oncolytic viral therapy, as it, on one hand, impedes the spread of viral infection, and on the other, augments tumor cell destruction through the recruitment of T cells "vaccinated" against tumor antigens. The most commonly used oncolytic viruses are each reviewed in turn, including adenoviruses, herpes simplex viruses, vaccinia viruses, reoviruses, and Newcastle disease viruses. Special attention is focused on the unique biology of each of these viruses as well as the status of several of these mutants in clinical trials.
Key Words. Oncolysis • Replication-conditional viruses • Gene therapy • Clinical trials
|
INTRODUCTION
|
|---|
The concept of using viruses in the treatment of cancer dates back to the beginning of this century when it was noted that patients with various malignancies experienced spontaneous tumor regression after rabies vaccination or a bout with a viral illness [1, 2]. Animal experiments in the 1920s confirmed that viruses were capable of infecting and lysing experimental murine tumors, and several reports followed in the 1950s demonstrating potent oncolysis of murine tumors by Newcastle disease virus (NDV) and influenza virus [3]. By the late 1940s and early 1950s, studies of oncolytic viruses in human cancer patients were initiated. Perhaps the most recognized of these studies was one reported from the National Cancer Institute in 1956 in which wild-type adenoviruses of different serotypes were injected into patients with cervical carcinomas [4]. More than half of the patients treated with live virus exhibited tumor regression without evidence of toxicity, whereas the control patients treated with inactivated virus showed no response. The initial tumor regression, though, was soon followed by tumor progression in all patients. This apparent lack of antitumor efficacy was mirrored in the other human trials of that day [5–7], thereby leading investigators to abandon this mode of therapy.
These early advocates of viral therapy for cancer were working at a time when virology was in its infancy and molecular biology was an as yet undiscovered field. In addition, these investigators were up against the great challenge of treating patients with the most advanced forms of cancer, in whom any significant improvement in survival was unlikely. These early pioneers set the stage for the current enthusiasm over the therapeutic potential of oncolytic viruses. Advances in tumor biology, genetics, and virology in the intervening years have provided us with the tools necessary to develop oncolytic viral therapy into an effective cancer therapy.
In general, there are two important aspects to oncolytic viral therapy, and they differ with respect to the directness of their antitumor effects. On the one hand, there is the direct treatment of tumors with replicating, oncolytic viral vectors alone or in combination with therapeutic transgene delivery, chemotherapy, or radiation therapy. It is this aspect of oncolytic viral therapy that will be the primary focus of this review. On the other hand, there is the indirect augmentation of antitumor immunity through a modulation of the immune response, as seen with viral oncolysate vaccines, and tumor-protective monoclonal antibodies. We will briefly discuss this aspect of viral oncolysis, but a more thorough presentation of this subject is beyond the scope of this review.
In this review, we first provide a framework and a general discussion of the principles of viral oncolysis, including: A) mechanisms of antitumoral efficacy; B) mechanisms of antitumoral specificity, and C) effects of the immune response. We then review each of the most common classes of oncolytic viruses in turn, focusing special attention on how the unique biology of each of these viruses lends itself to effective oncolysis for cancer therapy.
|
MECHANISMS OF ANTITUMOR EFFICACY
|
|---|
Oncolytic viruses mediate the destruction of tumor cells by several potential mechanisms (Table 1
) [8]. The virus itself is capable of directly lysing the cell as a result of viral replication. This cycle can then repeat itself with progeny virion infecting adjacent cells and destroying them by replication. This feature of replicating viruses provides a continuous amplification of the "input dose," which continues until abrogated by an immune system response or a lack of susceptible cells. As a second mechanism, certain oncolytic viruses generate proteins during their replicative cycle that are directly cytotoxic to the tumor cells. Adenoviruses, for example, express the E3 11.6 kD death protein and the E4ORF4 protein late in the cell cycle, and both of these proteins are cytotoxic to cells [9, 10].
A third mechanism by which oncolytic viruses mediate tumor cell destruction is via induction of nonspecific and specific antitumor immunity. Tumor cells are inherently weakly immunogenic, in part because they exhibit decreased expression of major histocompatibility complex (MHC) antigens and stimulatory signals, including cytokines that activate local immune responses. Infection of tumor cells by a replicating adenovirus with expression of the E1A protein leads to increased sensitivity to tumor necrosis factor (TNF) mediated killing [11]. The E1A protein is expressed early in the adenoviral replication cycle and binds to and inhibits the cellular tumor suppressor retinoblastoma protein (pRB), thereby preventing the host cell from shutting down and allowing protein synthesis and, thus, adenoviral replication to proceed. More importantly, induction of specific antitumor immunity confers the possibility of long-term protection against tumor recurrence. After viral infection of a tumor cell, there is infiltration of lymphocytes and antigen-presenting cells as well as induction of cytokine release. Viral antigens are presented on the cell surface in conjunction with the MHC class I protein, a complex that is then recognized by cytotoxic T lymphocytes (CTLs) that are attracted to the virally transduced tumor. By a mechanism as yet undefined, these CTLs acquire specificity for tumor-specific antigens in the process of viral antigen recognition and cell killing [12]. A systemic, specific antitumor immune response is thereby established in some animal models, as demonstrated with a herpes simplex virus type 1 (HSV-1) mutant [12]. In addition, lysates of virus-infected tumor cells (oncolysates) have been used as active specific immunotherapy in the treatment of patients with melanoma and ovarian carcinoma in clinical models. Lysates of virus-infected allogeneic human tumor cells elicit humoral immune responses against tumor-cell-associated antigens, virus-associated antigens, and antigens that may be virus induced [13], and these immune responses can improve the outcome of patients with melanoma in a surgical adjuvant setting [14].
Oncolytic viruses can engender an increased sensitivity of tumor cells to chemotherapy and radiation therapy. For example, the adenovirus E1A gene product is a potent chemosensitizer, particularly in cells with functional p53 [15]. The E1A gene product can induce high levels of p53 in these cells and render them susceptible to DNA damage from chemotherapy and radiation. Normal, nontransformed cells appear to be unaffected by E1A. Interestingly, the adenovirus E1A gene product can sensitize tumor cells to chemotherapeutic agents even in the absence of functional p53 by an unknown mechanism [16]. Enhanced chemosensitivity following viral infection has been observed in vivo in a phase II clinical trial of intratumoral adenovirus (ONYX-015, formerly dl1520, now known as CI-1042; Park-Davis Pharmaceuticals, Inc.; Ann Arbor, MI) in combination with cisplatin and 5-fluorouracil (5-FU) in patients with head and neck cancer [17]. Compared with either chemotherapy or ONYX-015 alone, the combination therapy produced a higher proportion of complete responses (27%), and by 6 months none of the responding tumors had progressed. None of these head and neck tumors had functional p53, thereby suggesting that ONYX-015 sensitized the tumor cells to the cisplatin and 5-FU by an alternate pathway.
A final mechanism by which oncolytic viruses mediate antineoplastic activity is by the expression of therapeutic transgenes inserted into the viral genome. These armed therapeutic viruses offer a distinct advantage over the replication-incompetent viruses that have been employed in the vast majority of gene therapy applications to date. As the virus amplifies itself through several rounds of replication and infection of neighboring cells, there is a concomitant amplification in transgene expression, which produces an amplified antitumor effect. Some researchers have incorporated prodrug-converting enzymes, such as viral thymidine kinase and bacterial cytosine deaminase (CD), into replication-conditional adenoviruses to augment tumor cell killing via the "bystander effect" [18, 19]. Other groups have introduced various immunostimulatory genes such as interleukins-4 (IL-4) and -12 (IL-12) into oncolytic herpes viruses in an attempt to augment the antitumor immune response of the host [20, 21].
|
MECHANISMS OF ANTITUMOR SPECIFICITY
|
|---|
Chemotherapy and radiation therapy are current mainstays in the treatment of advanced cancers but are limited by tumor cell resistance to these agents and a relatively narrow therapeutic index. Thus, dose-escalation or combination therapies designed to overcome resistance or increase tumor cell kill are limited by toxicity to normal tissues. Oncolytic viral therapy, on the other hand, is capable of increasing the therapeutic index between tumor cells and normal cells when viral replication proceeds preferentially in tumor cells.
How is viral replication specifically targeted to tumor cells? There are two general mechanisms that are employed to achieve tumor-selective viral replication: A) deletion of viral genes that are dispensable upon infection of neoplastic cells but are critical for viral replication in non-neoplastic cells (Table 2), and B) placement of tumor-specific promoters upstream of viral genes that are critical for efficient viral replication. In the first case, one takes advantage of the unique biology of tumors, such as the loss of cell cycle control by mutations of the tumor suppressor proteins p53 or pRb. An elegant example of this strategy is the oncolytic adenovirus ONYX-015, which is an attenuated adenovirus with two mutations in the E1B-55 kD gene. When adenovirus infects a normal cell, p53 levels are upregulated and the cell undergoes either cell-cycle arrest or apoptosis, thereby preventing viral replication. Wild-type adenovirus evades this "cellular suicide" by expressing the E1B-55 kD gene, which encodes a protein that binds to and inactivates p53 and thus allows viral replication to proceed. ONYX-015, however, lacks this gene and so viral replication is markedly attenuated in cells with normal p53 function. Many tumor cell types lack functional p53, and ONYX-015 replicates within and lyses these cells preferentially. Presumably, it is the loss of p53 function that accounts for the tumor-selective replication of ONYX-015. However, several groups challenged this proposed mechanism of tumor-specific replication on the basis of in vitro data that demonstrated that ONYX-015 efficiently replicates in many tumor cell types with wild-type p53 [22, 23]. This apparent contradiction was resolved through examinations of p14ARF, a tumor suppressor gene whose product functionally stabilizes p53 [24]. Loss of p14ARF was identified as a mechanism that allows ONYX-015 replication in tumor cells retaining wild-type p53 [25]. ONYX-015 may be useful then in the treatment of tumors with mutations within the p53 pathway as a whole.
The second strategy commonly employed to achieve tumor-selective viral replication involves insertion of tumor-specific or tissue-specific promoters into the viral genome to regulate the expression of viral genes that are necessary for an effective replication cycle. Paul Hallenbeck and colleagues demonstrated the proof of this principle by developing an oncolytic adenoviral mutant in which the E1A gene, the expression of which is critical for viral replication, is under the control of the tumor-specific 
-fetoprotein (AFP) gene promoter. This mutant, AvE1A04i, replicates preferentially in AFP-expressing cells such as hepatocellular carcinoma (HCC) cells and not in non-AFP-producing cells. AvE1A04i infection of AFP-expressing HCC flank tumors resulted in a greater than 50% long-term survival rate [26].
|
IMMUNE RESPONSE: FRIEND OR FOE?
|
|---|
Following oncolytic viral infection of tumor cells in culture, there is reproducible and effective cell killing due to viral replication. In a rodent or human, however, there are complex virus-host interactions that modulate viral replication, and chief among them is the immune response. One might imagine that immunity to the virus, be it innate or acquired, may restrict viral replication and facilitate viral clearance, thus limiting the antitumor effect of the therapy. On the other hand, one might envision that the immune response against infected tumor cells may enhance tumor destruction, as specific host immunity develops against tumor antigens. Accordingly, is the immune response a welcome friend or a meddlesome foe? Or is it neither? There is evidence to support all of these possibilities.
One of the great challenges of oncolytic viral therapy and of gene therapy in general is the efficient delivery of the viral vector to the tumor. Ikeda et al. [27] have demonstrated that much of this inefficiency of viral delivery is related to inactivation of the virus by the immune system. Interestingly, they found evidence of both innate immunity as well as an elicited humoral response. The innate activity against HSV-1 was seen in both rat and human plasma and was due in part to preimmune IgM as well as complement. This activity was present in both naïve and previously treated mice. A single dose of the B-cell immunosuppressive agent cyclophosphamide (CPA) suppressed this early innate activity and, at later time points, the specific neutralizing antibody response. Rats with intracerebral tumors treated with both CPA and the oncolytic herpes virus, by intravascular injection, had increased viral survival and replication within the tumors, leading to greater tumor regression and improved survival.
Other investigators and we have found that preexisting immunity to HSV has minimal effects on viral oncolytic therapy. Yoon et al. [28] noted that the presence of neutralizing antibodies to HSV-1 in vaccinated mice neither enhanced nor reduced the efficacy of an oncolytic HSV-1 mutant after intraportal administration to mice with diffuse liver metastases. Delman et al. [29] also demonstrated, in a liver metastasis model in immunocompetent mice, that tumor response to an oncolytic herpes virus was minimally affected by immunity to HSV. However, the route of administration of the virus did influence efficacy, with mice treated by intraportal injection exhibiting a greater reduction in liver tumor volume as a result of preexisting immunity to HSV than mice treated by intravenous (tail vein) injection.
There are studies that support the notion that the immune response to viral antigens presented on the surface of tumor cells becomes redirected to tumor cell antigens and, in so doing, enhances antitumor efficacy by generating an antitumor immunity. Toda et al. [12] demonstrated, in a highly immunogenic mouse colon cancer model, that oncolytic HSV replicating within a flank tumor elicits an immune response to specific tumor antigens in addition to the virus itself. In this study, identical mouse colon tumors were established bilaterally in the flanks of syngeneic, immunocompetent mice, and one of the flank tumors was subsequently infected with the oncolytic HSV. The injected tumor promptly regressed, as did the tumor on the contralateral flank despite the fact that the virus had not spread to this tumor. CD8+ CTL activity against a specific tumor cell antigen on the colon cancer cells was observed in these mice. In a separate study, corticosteroid administration did not reduce the oncolytic activity of HSV but did suppress the CTL immune responses to the tumor and the virus with a consequent reduction in the number of tumor cures in a syngeneic mouse neuroblastoma model [30].
Moreover, several clinical studies have supported the use of virus augmented tumor cells (oncolysates) in the active specific immunotherapy of cancer. Infection of human tumor cells by an oncolytic virus enhances the antigenicity of the tumor and, thereby, elicits a systemic immune response against the tumor [13]. On this basis, tumor cells augmented with vaccinia virus [31], NDV [32], and vesicular stomatitis virus [33] have been developed as immunotherapeutic agents for the treatment of cancer, and clinical trials of these agents are under way [14].
The findings in all of the aforementioned studies can be reconciled by the fact that the mode of viral administration and the end points examined are different in each study. Oncolytic viruses can be administered locally, by direct intratumoral inoculation, or systemically, by intravascular (i.e., tail vein, portal venous, etc.) administration. It is possible that the immune system serves to antagonize the effectiveness of oncolytic viruses administered intravascularly by limiting viral delivery to the tumor by virtue of both innate and acquired immunity. On the other hand, once the virus has reached its target and begins replicating within and destroying tumor cells, the immune response can theoretically augment tumor reduction by redirecting the CTL response from viral antigens to tumor antigens. In fact, viruses administered by direct intratumoral inoculation elicit a systemic immune response that prevents tumor formation and causes regression of existing tumors at distant sites [2, 30]. These preclinical data are sufficient for the design of clinical trials for the treatment of both locally advanced and widely disseminated cancers. For this, and other reasons, the role of the immune system in oncolytic viral therapy is an area of intense research.
|
ONCOLYTIC ADENOVIRUSES
|
|---|
Basic Biology
Adenovirus is a nonenveloped, linear, double-stranded DNA virus with a genome size of approximately 38 kb. Adenoviruses are capable of infecting dividing and nondividing cells. In order to efficiently replicate, adenoviruses promote entry into the cell cycle by the infected cell through expression of the adenoviral E1 gene complex. E1A binds to and inhibits the action of the cellular tumor suppressor pRB [34]. This serves to release pRB from the transcription factor E2F and facilitate entry into the G1 phase of the cell cycle. E2F mediates the transcription of several cellular enzymes necessary for DNA synthesis that occurs in S phase. The ultimate effect of immediate E1A expression is to create a cellular environment favorable for the synthesis of multiple copies of the adenoviral genome. E1B-55 kD expression is required to inhibit function of the cellular tumor suppressor protein p53 [35]. If it were not for adenoviral E1B expression, cellular levels of p53 would rise in response to adenoviral infection with resulting apoptosis or cell cycle arrest attenuating viral replication.
Optimizing Tumor-Specific Replication
As mentioned in the introduction to this article, there are two general approaches to achieving tumor-selective viral replication, and both of these approaches are nicely illustrated by oncolytic adenoviruses. The first strategy is to delete gene functions that are critical for viral replication in normal cells but not in tumor cells (Table 2
). Normal cells express high levels of the wild-type tumor suppressor proteins pRB and p53 upon adenoviral infection. In the process of viral replication, adenovirus expresses the E1A and E1B gene products, which inactivate these cellular defenses. Tumor cells very frequently express mutant, non-functional forms of pRB and p53. Thus, the E1A and E1B gene products become dispensable in these tumor cells, as E1A- or E1B-defective adenoviruses can still replicate preferentially in pRb- or p53-defective tumor cells, respectively. In contrast, replication of an E1A- or E1B-defective adenovirus is attenuated in normal cells with intact pRb and p53 pathways. ONYX-015 contains a deletion of the E1B-55 kD gene, and this mutant virus exhibits marked cytopathic effects in p53 mutant cancer cells, but only limited cytotoxicity in normal human fibroblasts and endothelial cells with normal p53 function [36]. ONYX-015 treatment of p53-deficient tumors in various mouse models causes significant tumor regression or inhibition, and this antitumor effect is associated with viral replication [37, 38].
As a result of these promising in vitro and in vivo studies, ONYX-015 is now under study for the treatment of a number of p53-deficient malignancies in phase I and II clinical trials (Table 3). A phase I study in patients with unresectable primary and secondary liver tumors showed that ONYX-015 was well tolerated when administered intratumorally, intravenously, or intraarterially up to a dose of 3 x 1011 plaque-forming units (pfu). A phase II study in this group of patients showed that the combination of ONYX-015 and 5-FU, when infused into the hepatic artery, was well tolerated but did not induce significant liver tumor regression as judged by serial computerized tomography scan measurements [39]. However, in another phase I/II dose-escalation trial of intraarterial ONYX-015 administration to patients with colorectal carcinoma liver metastases, patients who received the highest doses of ONYX-015 experienced better survival than those patients treated with the lower doses [40]. In addition, a phase I trial of intraperitoneal ONYX-015 is under way in the treatment of patients with cisplatin-resistant ovarian carcinoma [38], and the results of a phase I trial of local injection of ONYX-015 into unresectable pancreatic tumors was recently reported [41]. Lastly, phase II trials of intratumoral ONYX-015, with or without systemic chemotherapy, in patients with recurrent squamous cell carcinoma of the head and neck have demonstrated limited treatment-related toxicity and significant antitumor activity (10%-33% complete responses) [42, 43].
A second approach to achieve tumor-selective adenoviral replication is the use of tumor or tissue-specific promoters to drive the expression of an adenoviral gene that is critical for efficient viral replication, such as E1A. By replacing the endogenous viral E1A promoter with a human promoter sequence that is more transcriptionally active in tumor cells, one can restrict viral replication to these tumor cells rather than normal cells, which lack the protein(s) necessary to activate this promoter. One example of this strategy is exemplified by the virus CV706 (formerly CN706; Calydon, Inc.; Sunnyvale, CA), in which the prostate-specific antigen (PSA) gene promoter-enhancer element is inserted upstream of the E1A gene. Because of this modification, CV706 replication is greatest in tissues with high-level PSA expression [44]. This virus is currently in a phase I/II dose-escalation trial of intraprostatic injection in patients with locally recurrent prostate carcinoma following definitive radiotherapy. Results indicate that this treatment has significant clinical antitumor activity and a very favorable safety profile (personal communication, Calydon, Inc.). Calydon has also developed a more potent oncolytic adenovirus, CV787, which contains the prostate-specific rat probasin promoter driving E1A expression and the human prostate-specific enhancer/promoter driving the E1B gene. This virus, unlike CV706, maintains a wild-type E3 region, which encodes proteins that play a role in assisting virus release and in evading host immune responses to the virus. This virus eliminates distant prostate tumor xenografts in athymic mice after a single intravenous tail vein injection [45]. Currently, CV787 is being studied in phase I and II clinical trials in the treatment of patients with either organ-confined prostate carcinoma or hormone-refractory metastatic prostate cancer (Table 3).
Another example of this strategy is the adenoviral vector Ad.DF3-E1, in which the DF3/MUC1 promoter drives expression of E1A, thereby resulting in preferential replication of this virus in MUC1-positive breast cancer cells [46]. These studies elegantly demonstrate the success of targeting viral replication to tumor cells by cell-specific transcriptional regulation of adenoviral genes.
Optimizing Efficacy
In addition to their oncolytic effects, replication-conditional adenoviruses can be engineered to deliver therapeutic transgenes to increase the likelihood of tumor cell destruction through multiple avenues of attack. The advantage of this approach over the use of replication-defective vectors is that with viral replication and spread there is amplification of transgene expression and an increase in the proportion of the tumor mass exposed to the virus or its transgene product. Theoretically, a two-pronged attack of viral oncolysis and therapeutic transgene expression should prove to be more efficacious than either approach alone. Indeed, several investigators have shown this to be the case. Wildner et al. [18] constructed a replication-conditional adenovirus that expresses the HSV-1 thymidine kinase gene (HSV-tk), which confers sensitivity to ganciclovir. HSV-tk phosphorylates ganciclovir to its activated metabolite, which can diffuse to neighboring, uninfected cells through cellular gap junctions. This "bystander effect" increases tumor cell killing above that obtained by direct viral oncolysis. Freytag et al. [19] took this approach a step further by combining double suicide gene therapy with radiation therapy to achieve improved antitumor efficacy. They constructed an adenovirus similar to ONYX-015 but with a CD/HSV-tk fusion gene inserted into the E1B locus. This virus exhibits tumor cell specificity and replication kinetics similar to ONYX-015 but achieves enhanced antitumor efficacy via the suicide gene therapy systems and radiosensitization.
Another approach to optimizing the efficacy of replicating adenoviruses is to combine standard chemotherapy regimens with viral therapy. A phase II clinical trial of ONYX-015 injection combined with cisplatin and 5-FU in patients with recurrent squamous cell cancer of the head and neck was recently completed [17]. Nearly two-thirds of the patients had objective responses, and 27% of the study population had complete responses; toxicities were acceptable. This study as well as others demonstrate that the addition of chemotherapeutic agents, many of which might be expected to curb viral replication, augments the antitumor efficacy obtained with replicating adenoviruses.
|
ONCOLYTIC HERPES VIRUSES
|
|---|
Basic Biology
HSV-1 is an enveloped, double-stranded DNA virus with a genome size of approximately 152 kb. Several features of this virus make it an attractive virus for gene therapy. First, as much as 30 kb of the genome may be replaced by transgenes in replication-defective HSV-1 mutants, allowing for delivery of multiple transgenes and use of heterologous promoters [47]. This represents an advantage over adenovirus, whose much smaller genome limits the size and number of transgenes (Table 4). HSV-1 rarely produces severe medical illness in immunocompetent adults. Moreover, antiherpetic agents, such as acyclovir, are available that provide a safety mechanism to shut off viral replication should systemic toxicity ensue. Finally, HSV-1 does not integrate its genome into the cellular genome, as does the retrovirus for example, and so insertional mutagenesis is not a concern.
Optimizing Tumor-Specific Replication
As with the oncolytic adenoviruses, there are two general strategies that are employed to target HSV-1 replication to cancer cells. The first involves deletion or inactivation of viral genes that are essential for viral replication in normal cells but dispensable in tumor cells, such as HSV-tk, ribonucleotide reductase, and 
134.5. The initial HSV-1 mutant studied for tumor-selective replication, dlsptk, contains a 360-bp deletion within HSV-tk, and it was used to treat malignant gliomas in rodents [48]. Malignant glioma cells are actively dividing and so have high endogenous levels of this enzyme; in contrast, the surrounding normal brain is composed of quiescent neurons and glia. Accordingly, dlsptk replicates well in cultured tumor cells and induces significant growth inhibition of human U87 gliomas growing in the brains of nude mice. However, this HSV-1 mutant has not been examined in clinical trials because it produces neurotoxicity at higher titers, and it is resistant to the antiviral agents acyclovir and ganciclovir by virtue of its disrupted HSV-tk gene. Investigators looked instead at developing HSV-1 mutants that maintain sensitivity to acyclovir and ganciclovir and that exhibit less neurovirulence by deletion of the 134.5 gene. The protein product of 134.5 blocks the shut-off of host protein synthesis in infected cells by interacting with cellular phosphatase 1 to dephosphorylate eIF2 (Fig. 1). This leads to production of more progeny virion from each infected cell. Replication of 134.5-null mutants is attenuated in normal cells, such as adult neurons, thereby minimizing the risk of viral encephalitis. However, 134.5-null mutants replicate in actively dividing cells, such as tumor cells. Without disruption of the HSV-tk locus, these mutants are sensitive to both acyclovir and ganciclovir. One such 134.5-null mutant is G207 (MediGene, Inc.; San Diego, CA), which harbors deletions of both copies of 134.5 and contains an insertional inactivation of the ICP6 gene that encodes a subunit of viral ribonucleotide reductase. The presence of two mutations makes spontaneous reversion to a wild-type strain nearly impossible and so provides an added level of safety. Mammalian ribonucleotide reductase (and thus the level of intracellular nucleotides) is elevated in tumor cells relative to normal cells, so HSV-1 mutants defective in ribonucleotide reductase replicate preferentially in tumor cells [49]. For example, an ICP6-defective HSV-1 mutant replicates preferentially in tumor cells in a mouse model of diffuse colorectal carcinoma liver metastasis [28]. Furthermore, ribonucleotide-reductase-negative HSV-1 mutants are hypersensitive to acyclovir and ganciclovir [50], augmenting this built-in, all-important safety mechanism of HSV-1.
G207 exhibits tumor-cell-specific replication and antitumor efficacy in both in vitro and in vivo models of malignant glioma [51]. Preclinical toxicology evaluation of G207 was undertaken by intracerebral inoculation into owl monkeys, which are exquisitely sensitive to HSV-1 infection [52]. Inoculation of G207 at doses as high as 1 x 109 pfu, well above the efficacious dose in mice, was well tolerated, whereas monkeys that received an inoculation of only 1 x 103 pfu of wild-type HSV-1 died rapidly. These encouraging safety data prompted a phase I clinical study of G207 in the treatment of recurrent malignant glioma in which cohorts of patients were treated with increasing doses of G207. The highest dose examined was 3 x 109 pfu inoculated into five sites [53]. No toxicity related to G207 was seen at any dose. Preparations for phase II clinical trials designed to examine the efficacy of G207 alone and in combination with irradiation in this group of patients are now under way. In addition, G207 has been studied in preclinical models of colorectal cancer [54], ovarian cancer [55], breast cancer [56], and prostate cancer [57].
NV1020 (formerly R7020; MediGene, Inc.) is another genetically engineered oncolytic herpes virus that is being actively studied in clinical trials (Table 3). This virus is now in a phase I trial to evaluate its effectiveness as a vaccine in the treatment of patients with colorectal carcinoma liver metastases. NV1020 was originally designed as a candidate for human immunization against infections with HSV-1 and HSV-2 [58]. It contains a 700-bp deletion in the endogenous HSV-tk gene as well as a 15-kb deletion across the joint region of the long (L) and short (S) components of the HSV-1 genome. The L/S junction of NV1020 contains a 5.2-kb fragment of HSV-2 DNA inserted for previous vaccine studies and an exogenous copy of the HSV-tk gene under the control of the powerful HSV-1 4 promoter. Thus, NV1020 has only one copy of 134.5 deleted and maintains sensitivity to acyclovir and ganciclovir. The virus has demonstrated genetic stability and safety in extensive rodent and primate studies as well as in limited human vaccine trials [58, 59]. Accordingly, NV1020 was tested as a potential oncolytic agent in preclinical models of pancreatic carcinoma [60], transitional cell carcinoma [61], hormone-resistant prostate carcinoma [62], and chemotherapy/radiotherapy-resistant epidermoid carcinoma [63]. In all of these studies, NV1020 replicated to high titer and caused rapid regression of the flank tumor xenografts. Tumor destruction was accelerated with the addition of ionizing radiation to therapy with NV1020 [64]. Given these encouraging preclinical results, a phase I trial of intrahepatic arterial injection of NV1020 in patients with colorectal carcinoma liver metastases is currently under way.
The mutant HSV-1 vectors described thus far replicate preferentially in all types of tumor cells by virtue of the deletion or inactivation of genes (e.g., HSV-tk, ribonucleotide reductase, 134.5) whose absence is functionally complemented in these cells (Table 2). Another strategy to achieve preferential viral replication in tumor cells is by using heterologous tumor-specific or tissue-specific promoters to drive the expression of a viral gene that is critical for viral replication. G92a is one such vector, in which the immediate-early gene ICP4 is under the transcriptional control of the albumin promoter/enhancer [65]. Replication of ICP4-null mutants is markedly attenuated and G92a replicates up to 10,000 times more efficiently in albumin-expressing hepatoma cells than in other nonalbumin-expressing tumor cells and normal cells. In animal models, G92a effectively treats hepatoma but not prostate cancer [66].
Optimizing Efficacy/Potency
Despite the promising results obtained with the oncolytic HSV-1 vectors described thus far, it is likely that a multimodal approach to the eradication of cancer will be more effective. The combination of oncolytic viral therapy with chemotherapy and radiotherapy is a natural and logical next step. Several investigators to date have noted promising synergistic effects upon combining HSV-1-mediated oncolysis with either radiation or cisplatin chemotherapy [57, 64, 67]. In addition, replication-conditional HSV-1 can also be used as a gene transfer vehicle to enhance antineoplastic efficacy by expression of therapeutic transgenes. The choice of the transgene is critical, because expression of transgenes whose products abrogate viral replication is counterproductive. For example, HSV-1 vectors with a wild-type HSV-tk locus confer sensitivity to ganciclovir to infected cells or those that are in the vicinity of an infected cell (bystander effect). Yet, because ganciclovir inhibits herpes viral replication, the combination of HSV-1 oncolysis and ganciclovir is no better than HSV-1 oncolysis alone. Therefore, therapeutic transgenes that do not attenuate viral replication are more likely to enhance overall antineoplastic efficacy.
For example, the HSV-1 mutant rRp450 carries the gene for rat cytochrome p450 oxidase within its ribonucleotide reductase locus. Cytochrome p450 activates prodrugs, such as CPA, into their chemotherapeutically active metabolites, which can also diffuse to neighboring tumor cells and produce bystander killing. Bioactivated CPA only modestly inhibits HSV-1 replication [68]. As described earlier, administration of CPA also suppresses innate and acquired immunity of the host against the virus, thereby enabling much greater viral delivery to the tumor after intravascular administration [27]. Accordingly, administration of the combination of CPA with rRp450 leads to greater antitumor efficacy than that seen with either agent alone in animal brain tumor models [68, 69]. The addition of CPA to rRp450 results in similar improvements in antitumor efficacy in a rat model of hepatocellular carcinoma [70] and in a mouse model of colorectal carcinoma liver metastases [71].
HSV-1 mutants with deletions of both copies of the 134.5 gene and engineered to express cytokines, such as IL-4 and IL-12, display significant antitumor activity and prolong survival of mice with intracranial tumors [20, 21]. These mutants replicate preferentially in tumor cells by virtue of deletions of the 134.5 genes as outlined above and destroy tumors by a two-pronged attack: viral oncolysis and immunologic effects mediated by T cells. Immunohistochemical analysis of brain tumors treated with these mutants reveals significantly more inflammatory infiltration by CD4+ and CD8+ T cells and macrophages than that seen in tumors treated with the parent virus lacking the cytokine gene inserts. Likewise, the oncolytic virus G207 can be used as a helper virus to package a herpes simplex virus-amplicon vector carrying the gene for IL-2 [72]. This single preparation combines the immunostimulatory effects of amplicon-mediated IL-2 expression with the oncolytic effects of viral replication. These studies provide compelling support for the use of oncolytic HSV-1 mutants that deliver therapeutic transgenes to tumors.
|
ONCOLYTIC VACCINIA VIRUSES
|
|---|
Vaccinia virus is best known as the first widely used vaccine, which resulted in eradication of smallpox, and as such, has the longest track record of use in humans. Despite this fact, vaccinia virus has not been widely recognized as a suitable vector for gene therapy applications in humans until recently. Perhaps the observation that vaccinia virus induces a vigorous immune response and can be fatal to immunocompromised hosts explains the modest pace of its development. Enthusiasm for vaccinia in gene therapy applications is based on several observations. First, vaccinia virus exhibits tropism for a wide range of mammalian cell types. In addition, its nearly 200-kb genome allows the insertion of large DNA fragments for gene therapy applications. Also, its immunogenicity can be exploited to augment host immunity against tumor cells. This potential enhancement of therapeutic immune responses is the principle behind the development of poxvirus vaccines for melanoma, some of which are now in clinical trials. Wallack et al. recently presented data from a phase III, randomized, double-blind trial of a vaccinia melanoma oncolysate as an active specific immunotherapeutic agent to increase the disease-free interval or the overall survival of patients with stage III melanoma in a surgical adjuvant setting [14]. In certain subsets of patients, the vaccinia melanoma oncolysate conferred a survival advantage, but there was no difference in survival when all patients were considered together.
Other investigators have generated vaccinia virus mutants that are replication-conditional such that they destroy cancer cells as a by-product of viral replication. The strategy most commonly employed is analogous to one used in HSV-1 mutants in which insertional inactivation of the vaccinia virus thymidine kinase gene limits viral replication to cells with large intracellular nucleotide pools, such as tumor cells. Mastrangelo and colleagues inserted the gene for GM-CSF into the vaccinia virus thymidine kinase gene locus of a wild-type vaccinia virus as a strategy to generate an oncolytic virus that induces antitumor immunity following infection of malignant melanoma. This virus is currently under study in a phase I clinical trial of intralesional administration to patients with refractory, recurrent melanoma. In the first seven patients studied, two patients had a complete response and three other patients had partial responses [73]. Injected lesions showed evidence of viral replication as well as GM-CSF production with concomitant immune infiltrates. As with oncolytic HSV-1 mutants, oncolytic vaccinia viruses harboring cDNAs for cytokines like IL-2 [74] and prodrug-activating enzymes like CD [75] have been engineered to augment antineoplastic efficacy. Furthermore, mutant vaccinia viruses with improved tumor selectivity have been generated by the deletion of the SPI-1 and SPI-2 genes [76]. These viral genes encode serine protease inhibitors that are homologous to human proteins known to be upregulated in cancer and so are dispensable upon infection of tumor cells but are required for successful viral replication in normal cells. These studies offer a glimpse of the exciting possibilities of oncolytic vaccinia viruses as weapons in the fight against cancer.
|
REOVIRUS
|
|---|
Reovirus is a fascinating virus in that it is selectively oncolytic for many tumor cell types, and this selectivity is inherent in the biology of the virus. It is a ubiquitous, nonenveloped double-stranded RNA virus with minimal pathogenicity in humans. Infections in humans are generally mild and restricted to the respiratory and gastrointestinal tracts. For several decades, it was known that reovirus exhibited preferential cytotoxicity for transformed cells compared with normal cells [77, 78], but the mechanism behind this selectivity was not elucidated until recently. Normal mouse fibroblasts (NIH 3T3 cells), which are resistant to reovirus infection, become susceptible to infection following transformation with activated ras [79] or with any activated element of the ras pathway, such as the epidermal growth factor receptor [80] or the v-erbB oncogene [81]. When reovirus infects normal mouse fibroblasts, early viral transcripts activate double-stranded RNA-activated protein kinase (PKR), which then inhibits protein translation by phosphorylation of EIF-2. This cellular response inhibits reovirus replication. However, in cells with an activated ras pathway, PKR phosphorylation and activity are impaired, thereby allowing viral protein synthesis and the lytic cycle to proceed. The mechanism by which ras activation leads to PKR inactivation is unclear; nevertheless, it is clear that a common mechanism by which viruses evade the cellular antiviral defense system is through inhibition of PKR.
Given the facts that reovirus exhibits preferential replication in cells with an activated ras signaling pathway and that some 30% or more of all human tumors possess an activating mutation of ras, reovirus appears to be an ideal oncolytic agent. Coffey et al. [82] examined the efficacy of reovirus against flank tumors established from v-erbB-transformed NIH 3T3 cells and human U87 glioblastoma cells, which overexpress the platelet-derived growth factor receptor and thus have an activated ras pathway. A single intratumoral injection of 1 x 107 pfu of reovirus resulted in tumor regression in 65%-80% of mice. The virus also proved effective in treatment of tumors established from ras-transformed C3H-10T1/2 cells in immunocompetent C3H mice, and preexisting immunity to reovirus did not abrogate the oncolytic effect. More recently, data were presented that intravenous administration of reovirus is effective in reducing tumor burden and prolonging survival of mice bearing Lewis lung carcinoma metastases [83]. Continued investigation into mechanisms by which reovirus exploits cells with an activated ras pathway may provide more clues into better strategies for construction of efficacious and tumor-specific oncolytic viruses.
|
NEWCASTLE DISEASE VIRUS
|
|---|
NDV is a chicken paramyxovirus that was first noted to replicate in and destroy tumor cells in 1955 [84]. The virus is not pathogenic to humans and has been extensively studied as an oncolytic agent in several different human tumor cell lines and tumor models [85, 86]. These studies characterized the most widely known strain of NDV, 73-T, so named because it was passaged through mouse ascites tumor cells 73 times in vitro. It was chosen for trials in humans as both a viral oncolysate, which is a suspension of virus and tumor cells, and as free virus, because this strain demonstrated potent oncolysis and limited toxicity to normal cells. A recent 15-year follow-up of patients with stage III malignant melanoma treated postsurgically with an NDV oncolysate in 1975 as part of a phase II clinical study revealed a 55% overall 15-year survival. The oncolysate, composed of both allogeneic and autologous human melanoma cells infected with live NDV, presumably acts as a tumor vaccine, as the treated patients show evidence of increased numbers of CD8+ T cells [87].
Investigators at Pro-Virus, Inc. (Gaithersburg, MD) have isolated a naturally attenuated strain of NDV, cloned by nonrecombinant methods, that exhibits a broad range of oncolytic activity against human tumors. This strain, designated PV701, is characterized by its in vitro potency, as a ratio of 1 pfu to 10,000 tumor cells results in at least 50% lysis of sensitive tumor cell lines in 5 days. PV701 is also characterized by its tumor selectivity, as 80% of human cancer cell lines are two to four log orders more sensitive than normal human cells to PV701-mediated killing [88]. Intratumoral treatment of a variety of human tumors growing in the flanks of nude mice, including fibrosarcoma, ovarian carcinoma, and melanoma, caused high rates of tumor regression and minimal toxicity. Intravenous administration of PV701 in a dose-escalation study produced partial flank tumor regressions at doses as low as 6 x 105 pfu and complete tumor regressions in greater than 80% of the mice at doses up to 6 x 108 pfu. The antitumor response was associated with evidence of viral replication and although large amounts of virus were recovered from tumor tissue, no virus was isolated from the heart, lung, liver, kidney, or brain tissue [89]. These encouraging preclinical data have led to the initiation of a phase I clinical trial of PV701 administered intravenously to patients with advanced solid cancers who failed conventional therapy. Two partial responses in patients with colon carcinoma and mesothelioma were observed at higher doses of PV701, and six patients with diverse malignancies, including melanoma, colon carcinoma, and pancreatic carcinoma, exhibited measurable tumor reduction [90]. Additional clinical studies of this novel oncolytic virus are now in the planning stages.
|
CONCLUSION
|
|---|
Although the concept of using viruses as oncolytic agents dates back nearly a century, recent advances in the fields of molecular biology, genetics, and virology have enabled investigators to engineer viruses with greater potency and tumor specificity. Further enhancements in this approach involve arming these vectors with therapeutic transgenes, optimally combining the traditional modalities of chemotherapy and radiation therapy with oncolytic viral therapy, and modulating the immune response to minimize antiviral immunity while at the same time maximizing antitumor immunity. A better understanding of mechanisms that viruses use to overcome cellular defenses in order to achieve robust replication within the cell will lead to the development of oncolytic viruses with better tumor specificity and reduced toxicity. It is likely that as the therapeutic window of these new agents widens, oncolytic viruses will become standard agents in the battle against cancer.
|
REFERENCES
|
|---|
-
Dock G. Rabies virus vaccination in a patient with cervical carcinoma. Am J Med Sci 1904;127:563.
-
DePace N. Sulla scomparsa di un enorme cancro vegetante del collo dell’utero senza cura chirurgica. Ginecologia 1912;9:82–89.
-
Sinkovics J, Horvath J. New developments in the virus therapy of cancer: a historical review. Intervirology 1993;36:193–214.[Medline]
-
Smith R, Huebner RJ, Rowe WP et al. Studies on the use of viruses in the treatment of carcinoma of the cervix. Cancer 1956;9:1211–1218.[CrossRef][Medline]
-
Newman W, Southam CM. Virus treatment in advanced cancer. Cancer 1954;7:106–118.
-
Southam CM, Hilleman MR, Werner JH. Pathogenicity and oncolytic capacity of RI virus strain RI-67 in man. Trans NY Acad Sci 1960;22:657–673.
-
Cassel WA, Garrett RE. Newcastle disease virus as an antineoplastic agent. Cancer 1965;18:863–868.[CrossRef][Medline]
-
Kirn DH. Replicating oncolytic viruses: an overview. Expert Opin Investig Drugs 1996;5:753–762.
-
Tollefson AE, Ryerse JS, Scaria A et al. The E3-11.6-kDa adenovirus death protein (ADP) is required for efficient cell death: characterization of cells infected with adp mutants. Virology 1996;220:152–162.[CrossRef][Medline]
-
Shtrichman R, Kleinberger T. Adenovirus type 5 E4 open reading frame 4 protein induces apoptosis in transformed cells. J Virol 1998;72:2975–2982.[Abstract/Free Full Text]
-
Gooding LR. Regulation of TNF-mediated cell death and inflammation by human adenoviruses. Infect Agents Dis 1994;3:106–115.[Medline]
-
Toda M, Rabkin SD, Kojima H et al. Herpes simplex virus as an in situ cancer vaccine for the induction of specific anti-tumor immunity. Hum Gene Ther 1999;10:385–393.[CrossRef][Medline]
-
Savage HE, Rossen RD, Hersh EM et al. Antibody development to viral and allogeneic tumor cell-associated antigens in patients with malignant melanoma and ovarian carcinoma treated with lysates of virus-infected tumor cells. Cancer Res 1986;46:2127–2133.[Medline]
-
Wallack MK, Sivanandham M, Balch CM et al. Surgical adjuvant active specific immunotherapy for patients with stage III melanoma: the final analysis of data from a phase III, randomized, double-blind, multicenter vaccinia melanoma oncolysate trial. J Am Coll Surg 1998;187:69–77.[CrossRef][Medline]
-
Lowe SW, Bodis S, McClatchey A et al. p53 status and the efficacy of cancer therapy in vivo. Science 1994;266:807–810.[Abstract/Free Full Text]
-
Sanchez-Prieto R, Quintanilla M, Cano A et al. Carcinoma cell lines become sensitive to DNA-damaging agents by the expression of the adenovirus E1A gene. Oncogene 1996;13:1083–1092.[Medline]
-
Khuri FR, Nemunaitis J, Ganly I et al. 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 2000;6:879–885.[CrossRef][Medline]
-
Wildner O, Blaese RM, Morris JC. Therapy of colon cancer with oncolytic adenovirus is enhanced by the addition of herpes simplex virus-thymidine kinase. Cancer Res 1999;59:410–413.[Abstract/Free Full Text]
-
Freytag SO, Rogulski KR, Paielli DL et al. A novel three-pronged approach to kill cancer cells selectively: concomitant viral, double suicide gene, and radiotherapy. Hum Gene Ther 1998;9:1323–1333.[Medline]
-
Andreansky S, He B, van Cott J et al. Treatment of intracranial gliomas in immunocompetent mice using herpes simplex viruses that express murine interleukins. Gene Ther 1998;5:121–130.[CrossRef][Medline]
-
Parker JN, Gillespie GY, Love CE et al. Engineered herpes simplex virus expressing IL-12 in the treatment of experimental murine brain tumors. Proc Natl Acad Sci USA 2000;97:2208–2213.[Abstract/Free Full Text]
-
Goodrum FD, Ornelles DA. p53 status does not determine outcome of E1B 55-kilodalton mutant adenovirus lytic infection. J Virol 1998;72:9479–9490.[Abstract/Free Full Text]
-
Rothmann T, Hengstermann A, Whitaker NJ et al. Replication of ONYX-015, a potential anticancer adenovirus, is independent of p53 status in tumor cells. J Virol 1998;72:9470–9478.[Abstract/Free Full Text]
-
Quelle DE, Zindy F, Ashmun RA et al. Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell 1995;83:993–1000.[CrossRef][Medline]
-
Ries SJ, Brandts CH, Chung AS et al. Loss of p14ARF in tumor cells facilitates replication of the adenovirus mutant dl1520 (ONYX-015). Nat Med 2000;6:1128–1133.[CrossRef][Medline]
-
Hallenbeck PL, Chang YN, Hay C et al. A novel tumor-specific replication-restricted adenoviral vector for gene therapy of hepatocellular carcinoma. Hum Gene Ther 1999;10:1721–1733.[CrossRef][Medline]
-
Ikeda K, Ichikawa T, Wakimoto H et al. Oncolytic virus therapy of multiple tumors in the brain requires suppression of innate and elicited antiviral responses. Nat Med 1999;5:881–887.[CrossRef][Medline]
-
Yoon SS, Nakamura H, Carroll NM et al. An oncolytic herpes simplex virus type 1 selectively destroys diffuse liver metastases from colon carcinoma. FASEB J 2000;14:301–311.[Abstract/Free Full Text]
-
Delman KA, Bennett JJ, Zager JS et al. Effects of preexisting immunity on the response to herpes simplex-based oncolytic viral therapy. Hum Gene Ther 2000;11:2465–2472.[CrossRef][Medline]
-
Todo T, Rabkin SD, Sundaresan P et al. Systemic antitumor immunity in experimental brain tumor therapy using a multimutated, replication-competent herpes simplex virus. Hum Gene Ther 1999;10:2741–2755.[CrossRef][Medline]
-
Roenigk HH Jr, Deodhar S, St. Jacques R et al. Immunotherapy of malignant melanoma with vaccinia virus. Arch Dermatol 1974;109:668–673.[CrossRef][Medline]
-
Cassel WA, Murray DR, Phillips HS. A phase II study on the postsurgical management of Stage II malignant melanoma with a Newcastle disease virus oncolysate. Cancer 1983;52:856–860.[CrossRef][Medline]
-
Livingston PO, Albino AP, Chung TJ et al. Serological response of melanoma patients to vaccines prepared from VSV lysates of autologous and allogeneic cultured melanoma cells. Cancer 1985;55:713–720.[CrossRef][Medline]
-
Whyte P, Buchkovich KJ, Horowitz JM et al. Association between an oncogene and an anti-oncogene: the adenovirus E1A proteins bind to the retinoblastoma gene product. Nature 1988;334:124–129.[CrossRef][Medline]
-
Yew PR, Berk AJ. Inhibition of p53 transactivation required for transformation by adenovirus early E1B protein. Nature 1992;357:82–85.[CrossRef][Medline]
-
Bischoff JR, Kirn DH, Williams A et al. An adenovirus mutant that replicates selectively in p53-deficient human tumor cells. Science 1996;274:373–376.[Abstract/Free Full Text]
-
Rogulski KR, Freytag SO, Zhang K et al. In vivo antitumor activity of ONYX-015 is influenced by p53 status and is augmented by radiotherapy. Cancer Res 2000;60:1193–1196.[Abstract/Free Full Text]
-
Heise C, Ganly I, Kim YT et al. Efficacy of a replication-selective adenovirus against ovarian carcinomatosis is dependent on tumor burden, viral replication and p53 status. Gene Ther 2000;7:1925–1929.[CrossRef][Medline]
-
Habib NA, Sarraf CE, Mitry RR et al. E1B-deleted adenovirus (dl1520) gene therapy for patients with primary and secondary liver tumors. Hum Gene Ther 2001;12:219–226.[CrossRef][Medline]
-
Reid TR, Galanis E, Abbruzzese J et al. Intra-arterial administration of a replication-selective adenovirus Ci-1042 (Onyx-015) in patients with colorectal carcinoma metastatic to the liver: safety, feasibility and biological activity. Proc Am Soc Clin Oncol 2001;20:549a.
-
Mulvihill S, Warren R, Venook A et al. Safety and feasibility of injection with an E1B-55 kDa gene-deleted, replication-selective adenovirus (ONYX-015) into primary carcinomas of the pancreas: a phase I trial. Gene Ther 2001;8:308–315.[CrossRef][Medline]
-
Lamont JP, Nemunaitis J, Kuhn JA et al. A prospective phase II trial of ONYX-015 adenovirus and chemotherapy in recurrent squamous cell carcinoma of the head and neck (the Baylor experience). Ann Surg Oncol 2000;7:588–592.[Abstract]
-
Nemunaitis J, Khuri F, Ganly I et al. Phase II trial of intratumoral administration of ONYX-015, a replication-selective adenovirus, in patients with refractory head and neck cancer. J Clin Oncol 2001;19:289–298.[Abstract/Free Full Text]
-
Rodriguez R, Schuur ER, Lim HY et al. Prostate attenuated replication competent adenovirus (ARCA) CN706: a selective cytotoxic for prostate-specific antigen-positive prostate cancer cells. Cancer Res 1997;57:2559–2563.[Abstract/Free Full Text]
-
Yu D, Chen Y, Seng M et al. The addition of adenovirus type 5 region E3 enables Calydon virus 787 to eliminate distant prostate tumor xenografts. Cancer Res 1999;59:4200–4203.[Abstract/Free Full Text]
-
Kurihara T, Brough DE, Kovesdi I et al. Selectivity of a replication-competent adenovirus for human breast carcinoma cells expressing the MUC1 antigen. J Clin Invest 2000;106:763–771.[Medline]
-
Martuza RL. Conditionally replicating herpes vectors for cancer therapy. J Clin Invest 2000;105:841–846.[Medline]
-
Martuza RL, Malick A, Markert JM et al. Experimental therapy of human glioma by means of a genetically engineered virus mutant. Science 1991;252:854–856.[Abstract/Free Full Text]
-
Carroll NM, Chiocca EA, Takahashi K et al. Enhancement of gene therapy sp
Top
Abstract
Introduction
