|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology
Shedding Old Paradigms: Developing Viruses to Treat CancerUniversity of California, San Francisco, CA The template for early testing of new anticancer drugs has always been quite simple: start with a low-dose, advance slowly, treat patients until unacceptable toxicity is encountered, then step back one level for further testing. This classic phase I trial design has been used for decades, and although the minority of agents following this pathway have proven to be successful as therapies, this failure has been blamed on the poor quality of the agents rather than on flawed trial designs. The more we understand about tumor biology and drug development, the more this paradigm has come under criticism. Although this phase I design requires relatively few patients to identify dosing ranges for further analysis, there is concern that we may be exposing patients either to subtherapeutic doses (if enrolled early onto the trial) or to excessive toxicity (if enrolled late onto the trial). New algorithms have been proposed to accelerate these studies and treat even fewer patients at these exploratory doses.1 Other realities are perhaps more disconcerting. This trial design neither accounts for pharmacogenomic factors that may cause great variability in drug metabolism from patient to patient nor for the pharmacodynamic differences that are common among subsets of cancer patients. Even dose administration by body-surface area has been exposed for what it’s worth—not much.2 In other words, although our anticancer armamentarium keeps increasing, our testing system is imperfect. The usual phase I trial design is straightforward compared with the complexities we face with newer biologic agents. Will one ever find the maximum-tolerated dose for drugs that may be nontoxic, or do we need to? What of inconsistent oral bioavailability? What are the surrogates to replace radiographic response rate? What value is acute phase I testing of therapies that may be given for years, not months, so that chronic toxicity also matters? And now to viruses, which raise an entirely new set of issues in phase I testing. The article by Pecora et al,3 in this issue of the Journal of Clinical Oncology reports a complicated phase I trial of PV701, one of many viruses being developed to treat cancer. This report of 79 patients is lengthy and complicated, but by necessity rather than by choice. Whether or not this agent proves effective against cancer, this phase I trial merits careful study, because viral therapies for cancer are emerging, and their optimal development is a major challenge. Pecora et al3 describe a study of PV701, a replication-competent, naturally-attenuated strain of Newcastle disease virus (NDV) in patients with advanced solid tumors. This RNA virus, an avian paramyxovirus, demonstrates relative tumor cell selectivity because of defects in the interferon (IFN)-signaling pathway that are common among many tumor types. Such defects may disable the antiviral function of IFN, thus rendering malignant cells sensitive to infection and replication of viruses like PV701.3 The concept of using oncolytic viruses to treat cancer is not new. Many patients were treated with replication-competent viruses during the 1950s through the 1970s. Wild-type, attenuated, or tissue culture–adapted viruses possessing tumor cell selectivity were used, including adenovirus, mumps virus, vaccinia virus, myxovirus, West Nile virus, NDV, and others.4 Despite some promising results, the advent of recombinant DNA technology led to the abandonment of these viruses in favor of replication-incompetent viruses as vectors for gene therapy. Many vectors are under study, such as recombinant adenovirus, to deliver genes encoding prodrug-activating enzymes, replace wild-type tumor suppressor genes like p53 and p16INK4, and to disrupt oncogene function.4 There is increasing evidence, however, that efficient tumor-cell kill is difficult to achieve with this technology. As a result, there has been renewed interest in replication-competent viruses. At least 10 different replication-competent viral species have entered or are entering clinical trials.5 Because of safety concerns, initial efforts have used viruses which are prevalent in the human population but have low pathogenicity. Most of these fall into one of two basic categories: nonengineered wild-type viruses or recombinant forms with genetic manipulations. Regardless of the specific virus, however, fundamental problems remain that are related to delivery, specificity, infectivity, lateral transmission, and clearance. Nonengineered viruses, such as reovirus, vesicular stomatitis virus, NDV, and autonomous parvovirus, are inherently tumor-selective.5 Natural infections with these viruses are typically mild. NDV and vesicular stomatitis virus exploit tumor-associated defects in the IFN-response pathway to achieve selectivity. Infection with reovirus is tumor-selective by virtue of enhanced replication in cells in which Ras-mediated signal transduction is activated. Transformation also leads to increased sensitivity to killing by the autonomous parvovirus H-1, although the precise mechanisms underlying this phenomenon are unknown. The use of recombinant (engineered) replication-competent viruses for cancer therapy became feasible with the identification of specific genes or gene regions that are necessary for replication or toxicity in normal cells but are expendable in tumor cells. The first generation of these recombinant oncolytic viruses is comprised of adenovirus and herpes simplex virus type 1 gene-deletion mutants, although other viruses have also been targeted through the gene-deletion approach.5,6 To enhance safety, some of these viruses have multiple mutations, thus reducing the chance of reversion to wild-type.4 The largest clinical experience is with the engineered human adenovirus. Loss of all or part of the gene encoding for E1A-CR2, which binds and inactivates pRb family members, confers selectivity for cancer cells with defects in the Rb pathway. Similarly, deletions in the E1B-55-kD gene region, which encodes for a protein responsible for binding and inactivating p53, leads to preferential replication in cells in which the p53 pathway is defective. The most studied example of the E1B-55-kD class of deletion mutants is dl1520 (also known as ONYX-015, CI-1042).5,7 Over 200 patients with a variety of tumor types have been treated with this virus. It seems to be safe and well-tolerated, with no maximum-tolerated dose identified at doses of up to 2 x 1012 particles. Tumor selectivity has been documented in patients after intratumoral, intravenous, intraperitoneal, and hepatic arterial administration. The data from dl1520 has led to important insights into the use of this virus as an anticancer strategy.7-9 First, its single-agent activity seems to be relatively limited with additive, if not synergistic, effects when given in combination with chemotherapy. Second, dependence on p53 mutational status has not been established. Third, neutralizing antibody titers before or after treatment with dl1520 have not been predictive of antitumor activity. The engineering of second- and third-generation oncolytic viruses with improved efficacy is an area of intense interest, and numerous strategies are being used. Three of the approaches under study include the use of specific promoters to control viral replication,5,10 the incorporation of viral coat modifications to enhance tumor cell-specific uptake,11 and arming oncolytic viruses with therapeutic genes.12,13 All of these theoretical constructs are still on the drawing board or are early in development. Meanwhile, the initial clinical experience with unengineered replicating viruses like PV701 so far suggest they are safe but have limited efficacy. There is little experience with these viruses given systemically, however. Until now, all of the treatment protocols have evaluated multiple high-dose local administrations of virus to accessible tumors or multiple doses in combination with chemotherapy. Neutralization of the host immune response and problems related to systemic distribution and intratumoral spread continue to be major limitations of this therapy.10 The role of the immune response is complex, and preclinical studies on the interaction of the immune system with oncolytic viruses like adenovirus have been limited because of poor replication in other species. This complicated phase I trial design reported by Pecora et al3 is full of questions. How does one find an optimal dose of a replication-competent virus? How meaningful are pharmacokinetics? At what time points does one sample for viral levels? How does one account for the immune system? How do we purposely abrogate the antiviral response that may neutralize the anticancer oncolytic virus before it can replicate? And should we worry about shedding of replication-competent NDV or any other virus, given the absence of clear data defining a risk of infection of virus shed in the urine or stool? The answers so far: PV701 can be administered safely, as a single agent, to cancer patients. Many of the anticipated toxicities of a virus, such as fever, malaise, aches, and pains, can be prevented with an introductory desensitizing dose, which seems to develop within 2 days, suggesting that tolerance is not an antibody-based phenomenon. Hematologic and other typical side effects are seen, with significant thrombocytopenia and occasional neutropenia. Although virus is shed in the urine, it is at very low titers and disappears with chronic usage. Finally, virtually every patient develops neutralizing antibodies to the PV701. There are other observations from this study that may infer clinical activity and tumor targeting. Pulmonary toxicity was seen but only in patients with lung involvement by tumor, prompting speculation by the authors that this may represent viral infiltration into target organs. At least one patient, only 2 weeks removed from their last dose of PV701, had extensive tumor sampling, at which time viral particles could be identified at tumor cell membranes with electron microscopy. This phase I study certainly provides some dosing guidance for the future development of PV701. It also warns us that the era of the 20-patient phase I study may be past, and it provides a glimpse into how difficult drug development has become. When a study raises as many, or more, questions than it answers, it allows us to move the field forward, even if very slowly. REFERENCES 1. Mani S, Ratain M: New phase I trial methodology. Semin Oncol 24: 253-261, 1997[Medline]
2.
Mathijssen R, Verweij J, de Jonge M, et al: Impact of body-size measures on Irinotecan clearance: Alternative dosing recommendations. J Clin Oncol 20: 81-87, 2002
3.
Pecora AL, Rizvi N, Cohen GI, et al: Phase I trial of intravenous administration of PV701, an oncolytic virus, in patients with advanced solid cancers. J Clin Oncol 20: 2251-2266, 2002 4. Rancourt C, Curiel D: Conditionally replicative adenoviruses for cancer therapy. Adv Drug Deliv Rev 27: 67-81, 1997[CrossRef][Medline] 5. Kirn D, Martuza R, Zwiebel J: Replication-selective virotherapy for cancer: Biological principles, risk management and future directions. Nat Med 7: 781-787, 2001[CrossRef][Medline] 6. Rampling R, Cruickshank G, Maclean A, et al: Therapeutic replication-competent herpes virus. Nat Med 4: 133, 1998 (letter) 7. Kirn D: Clinical research results with dl1520 (Onyx-015), a replication-selective adenovirus for the treatment of cancer: What have we learned? Gene Ther 8: 89-98, 2001[CrossRef][Medline]
8.
Heise C, Lemmon M, Kirn D: Efficacy with a replication-selective adenovirus plus cisplatin-based chemotherapy: Dependence on sequencing but not p53 functional status or route of administration. Clin Cancer Res 6: 4908-4914, 2000
9.
Nemunaitis J, Ganly I, Khuri F, et al: 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, 2000 10. Alemany R, Balague C, Curiel D: Replicative adenoviruses for cancer therapy. Nat Biotechnol 18: 723-727, 2000[CrossRef][Medline]
11.
Suzuki K, Fueyo J, Krasnykh V, et al: A conditionally replicative adenovirus with enhanced infectivity shows improved oncolytic potency. Clin Cancer Res 7: 120-126, 2001 12. Wildner O, Morris J: The role of the E1B 55 kDa gene product in oncolytic adenoviral vectors expressing Herpes Simplex virus-tk: Assessment of antitumor efficacy and toxicity. Canc Research 60: 4167-4174, 2000 13. Bennett J, Malhotra S, Wong R, et al: Interleukin 12 secretion enhances antitumor efficacy of oncolytic Herpes Simplex viral therapy for colorectal cancer. Ann Surg 233: 819-826, 2001[CrossRef][Medline]
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2002 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
