Live Viruses in Cancer Treatment

Most people think of "viral therapy" as an obscure, experimental approach to the treatment of disease. However, replicating viruses have been used as an effective therapeutic modality for more than 200 years. One of the greatest clinical advances in medical history—the eradication of smallpox—was made with a replicating virus. In 1796, Edward Jenner discovered that pus from the wounds of infected patients contained live cowpox virus, which could be used as an effective vaccine against smallpox.[1] This discovery lead to the vaccination of several million people and world clearance of the disease.[2]

Interestingly, rare reports of complete remission induced in cancer patients in association with smallpox vaccination were sporadically reported.[3] Observations of tumor regression in association with other viral infections have also been described in cancer patients infected with herpes zoster,[4,5] hepatitis virus,[6,7] influenza,[8] varicella,[9] measles,[10-12] and other viruses.[13-15] The first published report of tumor destruction related to replication competent viruses occurred in 1912 when a woman with cervical cancer developed significant tumor necrosis following administration of an attenuated rabies virus for prophylactic treatment after a dog bite.[16,17] Early in vitro demonstration of viral oncolysis was first shown in 1922 with vaccinia virus, which was shown to propagate in several malignant tumor lines.

Following these observations, extensive work was performed investigating the potential use of viral therapy to treat cancer. In 1950, a strain of encephalitis virus was shown to induce a dose-related oncolytic effect in vivo with a mouse sarcoma tumor.[18] Viral replication was shown to correlate with tumor cell lysis; however, viral encephalitis developed in several mice. It was later found that serial passaging of the encephalitis virus in vitro prior to tumor injection in vivo would reduce the proliferative capacity in normal tissues, thereby minimizing the occurrence of encephalitis and enhancing oncolytic capacity. Clinical investigation was stimulated following additional work with Newcastle disease virus (NDV), influenza virus, and other viruses in animal models that showed cessation of ascites tumor (Ehrlich cells) growth and eradication of the malignancy in some animals.[19-23]

Clinical trials were carried out in advanced cancer patients between 1950 and the early 1970s, investigating administration of replication-proficient viruses.[17,24-29] Transient responses were seen. Several mechanisms of action were described, involving direct tumor lysis related to viral proliferation, tumor antigen induced immune activation, modulation of cancer oncogene expression (ie, c-fos, protein kinase C [PKC] modulation with measles injection), apoptosis related to expression of unique viral proteins (ie, E1A),[30,31] release of immunostimulatory cytokines,[32,33] and activation of other antitumor immune responses (ie, natural killer [NK] cell activation).[34] Viruses with low pathogenicity for normal tissue and high oncolytic capacity were investigated. Such viruses include NDV, mumps virus, herpes simplex virus (HSV), Egypt 101 virus, influenza virus, adenovirus serotype 5, vaccinia, and ONYX-015. Historical and current studies will be discussed.

Viral Uptake and Release

Most oncolytic viruses require proliferation in the same species or cell lineage and depend on host factors for successful evolution through life-cycle stages (binding, entry, intracellular transport, genome replication, viral gene expression, assembly and release of progeny). To initiate this process, the virus requires a suitable receptor at the surface of the cell for uptake[35] and transcription factors to bind to the promoter/enhancer elements in the viral genome, to induce expression of viral DNA.

Manipulations of the viral coat protein genes and tumor-specific viral promoters have not adversely affected replication of oncolytic viruses in malignant tissue, but have limited replication capacity in normal tissue. Additional specificity to malignant tissue has been shown following modification of the viral coat protein thereby enabling specific binding to tumor antigens not expressed on normal cell surfaces,[36] and engineering of tumor specific promoter and enhancer regions with the viral genome to generate viruses with selective malignant cell replication capacity.[37,38]

The release of oncolytic virus progeny (up to 104 viruses per cell) coincides with the death of the host tumor cell. The first "burst" of replicating viruses[39] generally occurs less than 24 hours after treatment and may continue as long as conditions are favorable for replication and immune destruction of released virus is limited. Malignant cells are capable of evading immune defenses, and this effect may facilitate local spread of released virus.[40]

Egypt 101 Virus

Egypt 101 virus is a strain of the West Nile virus, which is an adenovirus subtype. Preclinical testing of Egypt 101 virus in the early 1950s showed oncolytic activity in a uterine/cervix cancer cell line (HeLa).[41] Testing of live virus administration via oral and intravenous routes in normal volunteers revealed minimal toxicity (low-grade fever), thereby justifying clinical trials in cancer patients,[42-44] although a small number of patients with hematologic malignancies did develop transient encephalitis.[45] Fever generally occurred within 48 hours after inoculation and often coincided with the detection of live virus in circulation or excretion.[42]

Clinical Trials

In the first such trial, involving 34 cancer patients (27 evaluable for response), tumor regression was observed in 4 patients, stabilization occurred in 5, and 18 showed no response to a single injection of live virus.[42] A subsequent trial involving 30 patients with cervical carcinoma tested several routes of inoculation (direct intratumoral injection, arterial infusion, intravenous infusion).[43,45] Toxicity was limited to low-grade fever, and regression or stabilization of disease was observed in the majority of patients. Unfortunately, most responses were transient (< 3 months), and no patients achieved a durable complete response.

Analysis of cervical tissue and vaginal smears revealed proliferating virus in 77 samples from 20 patients studied. However, with analysis of 140 samples, 10 patients showed no evidence of viral presence. Response did not necessarily correlate with recovery of virus, although patients achieving more extensive necrosis generally harbored detectable virus. Studies to explore more intensive dosing of the virus or combination with other anticancer agents were not pursued.

Mumps Virus

Mumps, a paramyxovirus, has a tight helical RNA inner core enclosed in an outer lipid/protein shell. Oncolytic efficacy of mumps virus was initially demonstrated in a rat sarcoma model.[46]

Clinical Trials

The first clinical trial investigating mumps virus involved 90 patients with advanced cancer and explored several routes of administration including oral, rectal, intratumor, inhalation, and intravenous, depending on the location of the tumor.[25] Initial hematologic response to treatment included leukocytosis followed by lymphopenia. Transient fever, which could be inhibited by prophylactic treatment with low-dose prednisone(Drug information on prednisone), was also observed.

The authors noted that elevated antibodies at baseline were associated with a lesser tumor response.[25] However, three of the four patients achieving an "optimal" response had elevated neutralizing antibodies to mumps virus prior to treatment. Overall, 37 (41%) patients achieved a ³ 50% reduction in tumor size, and 79 patients with stable disease or better demonstrated clinical improvement (improved appetite, reduced pain, increased body weight).

Of the 90 patients, 65 received a combination of local intratumoral injection and/or intravenous infusion of the virus, and 24 (37%) of these 65 patients showed a partial or complete response. Most partial or complete responses occurred in patients with gastric carcinoma. However, the highest proportion of complete or partial responses occurred in cutaneous carcinoma and uterine carcinoma. Additionally, 9 of 10 patients with metastatic pulmonary carcinoma achieved clinical improvement with regression in tumor bulk.

Patients receiving multiple intratumoral injections over a prolonged period achieved a higher response rate and longer duration of response.[46,47] Fifteen patients received intravenous mumps virus alone. Six of these patients received fewer than nine intravenous treatments, and none had a positive response. In contrast, of the nine patients receiving nine or more systemic treatments, six achieved a response (P < .02).[48] This was the first study to suggest that a multitreatment administration schedule may have a clinical advantage.

Further exploration of a systemic route of administration was not performed. A follow-up study involving 200 cancer patients administered mumps virus intratumorally.[49] Toxicity was minimal. Transient tumor regression was noted in 26 patients. Responses were observed in patients with cancer of the breast, rectum, colon, thyroid gland, uterus, and skin. Due to the transient nature of the response and the difficulty in manufacturing a uniform product, further clinical testing of the mumps virus was not pursued.