JOURNAL

OF SURGICAL

52,448-453

RESEARCH

Newcastle

(1992)

Disease Virus Selectively Kills Human Tumor Cells’

KIRK W. REICHARD, M.D.,* ROBERT M. LORENCE, M.D., PH.D.,?’ CHRISTOPHER J. CASCINO, M.D.,$ MARK E. PEEPLES, PH.D.,§ ROBERT J. WALTER, PH.D.,* MICHAEL B. FERNANDO, B.S.,? HERNAN M. REYES, M.D.,* AND JOHN A. GREAGER, M.D.* *Department

of Surgery,

Presented

University of Illirwis College of Medicine, Cook County Hospital, and Hektoen Institute and Departments of tpathology, $Neurosurgery, and $Immunology/Microbiology, Rush Presbyterian St. Luke’s Medical Center, Chicago, Illinois 60612-3864

at the Annual

Meeting

of the Association

for Academic

Newcastle disease virus (NDV), strain 73-T, has previously been shown to be cytolytic to mouse tumor cells. In this study, we have evaluated the ability of NDV to replicate in and kill human tumor cells in culture and in athymic mice. Plaque assays were used to determine the cytolytic activity of NDV on six human tumor cell lines, fibrosarcoma (HT1080), osteosarcoma (KHOS), cervical carcinoma (KBEl-5-11), bladder carcinoma (HCV29T), neuroblastoma (IMR32), and Wilm’s tumor (G104), and on nine different normal human fibroblast lines. NDV formed plaques on all tumor cells tested as well as on chick embryo cells (CEC), the native host for NDV. Plaques did not form on any of the normal fibroblast lines. To detect NDV replication, virus yield assays were performed which measured virus particles in infected cell culture supernatants. Virus yield increased lO,OOO-fold within 24 hr in tumor and CEC supernatants. Titers remained near zero in normal fibroblast supernatants. In vivo tumoricidal activity was evaluated in athymic nude Balb-c mice by subcutaneous injection of 9 X 10’ tumor cells followed by intralesional injection of either live or heat-killed NDV (1 .O x 10’ plaque forming units [PFU]), or medium. After live NDV treatment, tumor regression occurred in 10 out of 11 mice bearing KB8-5- 11 tumors, 8 out of 8 with HT1080 tumors, and 6 out of 7 with IMR-32 tumors. After treatment with heat-killed NDV no regression occurred (P < 0.01, Fisher’s exact test). Nontumor-bearing mice injected with 1.0 X 10’ PFU of NDV remained healthy. These results indicate that NDV efficiently and selectively replicates in and kills tumor cells, but

1 Supported in part by the Eleanor B. Pillsbury Fellowship, University of Illinois, and the Margaret M. Bailey Endowment for Cancer Research of the University of Illinois at Chicago Department of Surgery (K.W.R.) and Public Health Service grant AI-29606 (M.E.P.). The work of R.M.L. is dedicated to Michael Platt. ’ To whom all correspondence should be addressed at Department of Pathology, Rush-Presbyterian-St. Luke’s Medical Center, 1653 W. Congress, Chicago, IL 60612-3864. 0022-4804/92 $4.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form

448 Inc. reserved.

Surgery, Colorado

Springs, Colorado,

for Medical

Research;

Nov. 20-23, 1991

not normal cells, and that intralesional NDV causes complete tumor regression in athymic mice with a high therapeutic index. 0 1992 Academic Press, Inc.

INTRODUCTION Cancer treatment using viruses has been discussed in the clinical literature for years as anecdotal case reports and uncontrolled clinical trials. The earliest references described patients treated with live-attenuated viral vaccines who experienced tumor regression [ 1, 21. Several authors [3-lo] have since described regression of tumors in patients during measles and mumps infections, including one study of 90 patients intentionally infected with live mumps virus, followed by at least some regression of a variety of different tumors in 79 cases [lo]. Serious sequelae of infection with these human pathogens, however, were a major concern. Therefore, Cassel and co-workers began investigating Newcastle disease virus (NDV), an avian paramyxovirus not known to be pathogenic in humans [ 111. Initial results using a mouse-tumor-adapted strain (73-T) suggested that NDV was directly oncolytic to murine tumor cells both in vitro and in uiuo [ll]. However, rigorous testing of this effect on a wide variety of human tumor cells was never performed. Attention was instead focused upon the use of viral oncolysates as immunoenhancing agents. This approach was employed with some success in several clinical trials on patients with advanced melanoma [ 12-141. Interest in virus-mediated immunocytolysis has been continued by Schirrmacher and colleagues using tumor cells that were modified by infection with an avirulant strain of NDV [15, 161. We have undertaken the current study to evaluate the direct cytolytic effects of NDV strain 73-T on a variety of human tumor cells, both in vitro and in viuo, and to establish NDV’s specificity for tumor versus normal cells.

REICHARD

MATERIALS

ET AL.:

AND

NEWCASTLE

DISEASE

VIRUS

KILLS

HUMAN

METHODS

Virus Newcastle disease virus strain 73-T was donated by Dr. William Cassel (Emory University, Atlanta, GA). The original stock was amplified by passage through loday-old chick embryos. Forty-eight hours after inoculation of 1000 plaque forming units (PFUs), the allantoic fluid was harvested and ultracentrifuged for 18 hr at 24,000g. The viral pellet was then resuspended in Hank’s balanced salt solution (HBSS) and separated from the egg proteins on a discontinuous sucrose gradient (20 and 55%) at 18,000g for 1 hr [17]. The partially purified virus stock was harvested from the interface and stored at -80°C. Cells Human fibrosarcoma cell line HT-1080, human osteosarcoma cell line KHOS, human Wilm’s tumor cell line G104, and human neuroblastoma cell line IMR-32 were obtained from American Type Culture Collection (Rockville, MD). Normal human fibroblasts cell lines Fb-101, Fb-103, Fb-104, and Fb-106 were primary cultures started from skin biopsy specimens. Human bladder carcinoma cell line HCV29T and human multidrug resistant cervical carcinoma cell line KB 8-5-11 were donated by Dr. John Coon, Department of Pathology, Rush-Presbyterian-St. Luke’s Medical Center, Chicago, IL. Normal fibroblast cell lines FBB, FBD, FBL, FBR, and NHF were donated by Dr. Warren Knudson, Department of Biochemistry, Rush-Presbyterian-St. Luke’s Medical Center, Chicago, IL. Chick embryo cells (CEC) were harvested from fresh lo-day-old chick embryos. All cells were maintained at 37°C in 5% CO, in Dulbecco’s modified eagle’s medium (Flow Laboratories, Inc., McLean, VA) enriched with nonessential amino acids, glutamine, penicillin/streptomycin, and 10% heat-inactivated fetal bovine serum. Animals Female Balb-c athymic nude mice (nu/nu), aged 6-8 weeks, were obtained from Life Sciences, Inc. (St. Petersburg, FL). They were housed in an isolated room in sterile, filtered cages, and obtained food and water ad libitum. Virus-infected waste and animal carcasses were double-bagged and autoclaved before disposal. Experimental protocols were conducted according to guidelines established by the Institutional Animal Care Committee.

TUMOR

CELLS

TABLE

1

449

Plaque Sizes on NDV-Treated Monolayers of Diverse Classes of Human Tumor Cells, Chick Embryo Cells, and Normal Human Fibroblasts

Cell type Cervical CA (KB8-5-11) Bladder CA (HCV29T) Wilm’s tumor (G104) Fibrosarcoma (HT-1080) Osteosarcoma (KHOS) Neuroblastoma (IMR-32) Chick embryo cells (CJW Normal fibroblasts (n = 9)

Size (mm) 2 days after live NDV

Size (mm) 3 days after live NDV

Overlapping

Overlapping

1.3 t 0.1

2.3 + 0.2

0.9 t 0.1

1.5 k 0.1

1.9 t 0.1

3.3 + 0.7

0.9 f 0.1

1.6 f 0.2

2.0 f 0.1

2.9 iz 0.2

2.0 f 0.1

3.3 I 0.3

0

0

Size (mm) 3 days after Ht-killed NDV

Note. Cultured monolayers were infected with live NDV or heatkilled (h&killed) NDV, covered with agar, and incubated. After fixing and staining, plaques were measured. All tumor cell cultures developed plaques within 2 days, as did the CEC cultures. None of the fibroblast cultures developed any plaques.

perature, medium containing 0.9% Bacto-Agar (Difco, Detroit, MI) was added and allowed to solidify, thereby limiting viral diffusion. These dishes were then incubated at 37°C in 5% CO, for 2 days. After removal of the agar, the monolayers were fixed with 100% methanol and stained with 0.2% crystal violet. Plaques were then counted and measured. Virus Yield Assay Confluent cell monolayers on 6-well plastic tissue culture plates (35 mm per well) were infected with NDV (multiplicity of infection = 0.2) in HBSS for 45 min at room temperature. The cells were then washed thoroughly with HBSS to remove all unadsorbed virus, and medium was added. At specified times (2-48 hr) the medium was harvested, and the titer of virus was quantified using the plaque assay with HT-1080 human fibrosarcoma cells as targets. These cells are extremely sensitive to the cytolytic effects of NDV and are simpler to work with than chick embryo cells (Cascino, Lorence, Reichard, and Peeples, unpublished observations).

Plaque Assay Confluent cell monolayers on 50-mm plastic tissue culture dishes were incubated with serial dilutions of virus in HBSS. Controls were treated with HBSS alone. After allowing viral adsorption for 45 min at room tem-

Tumor Regression Athymic nude mice were divided into three groups and injected subcutaneously with 9.0 X lo6 tumor cells suspended in 0.1 cc culture medium, causing a visible wheal.

450

JOURNAL

OF SURGICAL

RESEARCH:

VOL.

52, NO. 5, MAY

TABLE

log pfu/ml

Regression

of Subcutaneous after NDV

Cell type Fibrosarcoma (HT-1080) Neuroblastoma (IMR-32) Cervical carcinoma (KB&5-11) 0

4

8

1992

12

18

20

24

28

32

38

40

44

Fbl

-

CEC

-

Nb

-

Fb2

-+

Tumors Treatment

in Athymic

Mice

3 weeks after 1 x lo6 PFU live NDV

3 weeks after 1 x lo6 PFU H&killed NDV

8outof8 6 out of 7 10 out of 11

Ooutof5 Ooutof5 0 out of 10

48

time (hrs) -

2

Sarc

FIG. 1. Replication assay. Virus titers in cell culture supernatants, expressed in a logarithmic scale as PFU/ml, are shown for each time point measured. Titers were measured using the plaque assay (see Materials and Methods). Virus titers in supernatants from HT1080 fibrosarcoma (Sarc), IMR-32 neuroblastoma (Nb), and chicken embryo cell (CEC) cultures peaked at 24 hr, while virus titers from the two normal human fibroblast cultures (Fbl, Fb2) remained low.

Three minutes later, 0.1 cc culture medium containing either 1.0 X lo6 PFU of live NDV, heat-killed NDV (6O’C for 1 hr) or medium alone was injected into the subcutaneous wheal. The animals were placed back into their cages and observed for tumor growth. Other animals were prepared similarly, except that 1.0 X lo6 tumor cells were injected, followed by 1.0 X lo6 PFU of NDV. These animals were sacrificed at 8 and 48 hr for histologic examination. Nontumor-bearing athymic mice were injected with 1.0 X lOa PFU of NDV and observed for evidence of disease. RESULTS

In Vitro Cytolysis Chickens are the natural host for NDV. To examine the ability of NDV to kill human tumor cells, fibrosarcoma (HT-1080), osteosarcoma (KHOS), cervical carcinoma (KBB-5-11), bladder carcinoma (HCV29T), Wilm’s tumor (G104), neuroblastoma (IMR-32), and chick embryo cells (CEC) were each infected in a plaque assay. Plaques appeared on all tumor cell culture plates within 48 hr, as well as on the native host CEC plates. In contrast, no plaques appeared on any of the nine normal human fibroblast culture plates, even after 5 days of treatment with concentrated virus (Table 1).

Note. Athymic nude mice were injected with 9 X lo6 cells followed by 1 X 10’ PFU of either live or heat-killed (h&killed) NDV. Tumor regression 3 weeks after treatment is shown. Differences between live and heat-killed groups are significant for all tumors (P < 0.01, Fisher’s exact test).

In Vitro Virus Yield from Cultured

Cells

The ability of NDV to replicate in tumor cells was next examined. Supernatants from cell culture monolayers were sampled at specified times after infection, and virus particle release was quantified using the plaque assay, as shown in Fig. 1. Virus titers in the supernatants of HT-1080 and IMR-32 cultures increased lO,OOO-fold over a 24-hr period as did those from the host CEC cultures. NDV was not detected in significant amounts in the supernatants from normal fibroblast cells treated with the same amount or 100 times more virus (Fig. 1). In Vivo Tumor Regression To determine if NDV exhibits specific tumoricidal activity in vivo, athymic nude mice were injected subcutaneously with 9.0 X 10’ tumors cells. The tumor site was then injected with normal or heat-inactivated NDV, or with culture medium. All animals exhibited tumor growth for the first 5 days, attaining a mean tumor diameter of 3.0 mm. By the end of 3 weeks, the tumors in animals treated with live NDV had completely regressed, while the tumors in those animals treated with heat-inactivated NDV or with culture medium alone continued to grow (Table 2). There were no differences between the tumor diameters of the two control groups. Histologic sections from animals injected with 1.0 X lo6 tumor cells, followed by 1.0 X lo5 PFU of NDV, were examined. At 8 hr postinfection, these sections revealed numerous multinucleated giant cells as well as significant cellular necrosis. By 48 hr, no viable tumor cells were present in any of the sections, and only limited

FIG. 2. Histology. Athymic nude mice were injected with 1 X lOa KBS-5-11 cervical carcinoma cells followed by 1 X 10’ PFU of NDV. Histologic sections from animals sacrificed after 8 hr showed a typical squamous carcinoma appearance in tumors injected with heat-killed virus (a, 35~; b, 140x). Tumor from animals injected with live NDV (c, 280X) revealed multinucleated giant cells (arrow) and necrosis. After 48 hr (d, 35X), there was no viable tumor present, only individual necrotic tumors cells and a mild inflammatory infiltrate (see arrow).

REICHARD

ET AL.:

NEWCASTLE

DISEASE

VIRUS

KILLS

HUMAN

TUMOR

CELLS

451

452

JOURNAL

OF SURGICAL

RESEARCH:

inflammatory cell infiltrates remained (Fig. 2). None of the animals treated with live NDV (1.0 X lo6 PFU, 1.0 X lOa PFU) showed any signs of illness. All NDV-treated animals continued to grow and eat normally during an observation time of at least 6 weeks, and no NDVtreated animal suffered a tumor recurrence after regression. DISCUSSION Three principle mechanisms have been proposed to explain virus-induced tumor regression [3,18]. (1) Viral proteins associated with either tumor cell membrane fragments or intact cells may enhance the immunogenicity of tumor antigens, thereby augmenting host cellularand humoral-mediated destruction of the tumor. This mechanism, referred to as immunologic cytolysis, can occur in response to in vivo infection of tumors with lytic or nonlytic viruses, or in response to vaccination with virus-lysed cell preparations (oncolysate). Clinical trials employing each of these modalities have shown some efficacy [ 12-161. (2) Certain viruses may enhance host production of cytokines, e.g., interferon and tumor necrosis factor-a [19-211. These first two mechanisms require participation by the host’s immune system. (3) Finally, viruses may be directly and specifically cytotoxic to tumor cells [ll, 181. Results of the present study provide support for the last mechanism as described below. Data from the plaque assays, from the virus yield studies and from the athymic animal experiments all suggest that tumor cells are susceptible to NDV replication and virus-mediated cytolysis, while normal cells are not. The selective sensitivity of tumor cells to NDV may be due to the unique presence of a viral membrane receptor or to differential uptake, replication, or release of the virus. It is unlikely that the effects are due solely to selective infection of rapidly growing cells, since NDV had little effect on rapidly dividing leukemia cells (Lorence and Peeples, unpublished data). Nor is it likely that the cell of origin determines the sensitivity to NDV, since tumor lines representing three diverse classes (epithelial, mesenchymal, and neural) were examined and were all susceptible to the effects of NDV. In the plaque assay, virus is unable to diffuse in the agar so that each plaque represents the effect of one infective virus particle that has progressively replicated and lysed adjacent cells. In addition, virus yield increased substantially in all tumor cell supernatants harvested after 24 hr and was similar to levels obtained in the virus’ natural host, chick embryo cells. Even after 5 days, no viral titers were detected in normal human fibroblast supernatants, reenforcing the hypothesis that NDV is selective for tumor cells. The potential significance of these observations, especially regarding systemic treatment, is that only a small number of infective NDV particles is required to reach the tumor. Subsequent amplification, which takes place in the presence of

VOL.

52, NO. 5, MAY

1992

tumor cells, creates a high local concentration of tumoritidal virus. The selective efficiency of NDV is further supported by the data from athymic animals, in which subcutaneous lesions injected simultaneously with live NDV completely regressed. Conventional chemotherapy follows log cell kill kinetics, in which a proportional number of cells will be killed with a given dose of drug regardless of the tumor burden [22]. NDV, in contrast, seems to be effective at lysing all cells present in the area. Histology confirmed early cell-to-cell fusion and lysis, as demonstrated by the presence of multinucleated giant cells and of tumor cell necrosis. Complete tumor regression in these immunodeficient mice, combined with the minimal inflammatory cell infiltrate seen histologically even after 48 hr, suggest that these effects were principally due to direct viral oncolysis. This is in contrast to other workers [ 11, 12, 181, who have postulated a primarily immune-mediated mechanism. Safety of NDV is an important consideration. Cassel and Garrett [ll], in their original studies, noted that very large doses of NDV strain 73-T were nonpathogenic in a variety of animal models including the newborn mouse, even when directly inoculated into the CNS [ 111. In addition, they were able to treat a large number of patients over several years with NDV strain 73-T and never encountered an untoward reaction [13,14]. Others have had similar experience with NDV [ 15,161. Further, chicken farmers (exposed to NDV during outbreaks in their flocks) and laboratory workers (exposed to high titers of several strains of NDV) have exhibited only occasional mild conjunctivitis without cornea1 involvement [23,24]. These observations have been confirmed by current results which showed that doses of up to 100 times that used to treat subcutaneous tumors caused no illness in nontumor-bearing athymic mice. In summary, we have shown that Newcastle disease virus strain 73-T selectively replicates in and kills a variety of tumor cell types in vitro. Normal fibroblast cultures, however, seem to be resistant. Further, small human subcutaneous tumors in athymic nude mice undergo viral-mediated lysis without evidence of recurrence and with a therapeutic index of at least 100. Further work is necessary to identify the putative tumorspecific components that allow for selective NDV infection and replication and to treat more locally advanced and metastatic tumors in athymic mice. REFERENCES 1. 2. 3. 4.

DePace, N. G. Sulla scomparsa di un enorme cancro vegetante de1 callo dell’utero senza cura chirurgica. Ginecologia 9: 82,1912. Salmon, P., and Baix. Vaccine variolique dans le cancer. Compt. Rend. Sot. Biol. 86: 819, 1922. Webb, H. E., and Smith, C. E. G. Viruses in the treatment of cancer. Lancer, 1: 1206, 1970. Moore, A. E. Effects of viruses on tumors. Annu. Rev. Microbial. 8: 393, 1954.

REICHARD

ET AL.:

NEWCASTLE

DISEASE

5.

Southam, C. M. Present status of oncolytic N. Y. Acad. Sci. 22: 657,196O.

6.

Pasquinucci, G. Possible effect of measles on leukemia. Lancet 1: 136,197l. Gross, S. Measles and leukemia. Lancet 1: 397, 1971. Bluming, A. Z., and Zeigler, J. L. Regression of Burkitt’s Lymphoma in association with measles infection. Lancet 2: 105, 1971. Csatary, L. K. Viruses in the treatment of cancer. Lancet 2: 825, 1971. Asada, T. Treatment of human cancer with mumps virus. Cancer 34: 1907, 1974. Cassel, W. A., and Garrett, R. E. Newcastle Disease Virus as an antineoplastic agent. Cancer 18: 863, 1965.

7. 8.

9. 10. 11. 12.

13.

14.

15.

16.

VIRUS

virus studies. Trans.

Cassel, W. A., Murray, D. R., Torbin, A. H., Olkowski, Z. L., and Moore, M. E. Viral oncolysate in the management of malignant melanoma. I. Preparation of the oncolysate and measurement of the immunologic responses. Cancer 40: 672, 1977. Murray, D. R., Cassel, W. A., Torbin, A. H., Olkowski, Z. L., and Moore, M. E. Viral oncolysate in the management of malignant melanoma. II. Clinical studies. Cancer 40: 680, 1977. Cassel, W. A., Murray, D. R., and Phillips, H. S. A phase II study on the postsurgical management of stage II malignant melanoma with a Newcastle Disease Virus oncolysate. Cancer 52: 856, 1983. Schirrmacher, V., von Hoegen, P., Ahlert, T., and Heicappell, R. Design of a virus-modified autologous tumor vaccine for activespecific immunotherapy of cancer metastasis. Arch. Geschwulstforsch. 61: 23, 1991. Bohle, W., Schlag, P., Liebrich, W., Hohenberger, P., Manasterski, M., Moller, P., and Schirrmacher, V. Postoperative active

17.

18.

19.

20. 21.

22.

23.

24.

KILLS

HUMAN

TUMOR

CELLS

453

specific immunization in colorectal cancer patients with virusmodified autologous tumor-cell vaccine. First clinical results with tumor-cell vaccines modified with live but avirulent Newcastle disease virus. Cancer 66: 1517, 1990. Faaberg, K. S., and Peeples, M. E. Strain variation and nuclear association of Newcastle disease virus matrix protein. J. Virol. 62: 586,1988. Schirrmacher, V., Ahlert, T., Heicappell, R., Appelhans, B., and von Hoegen, P. Successful application of a non-oncogenic virus for anti-metastatic cancer immunotherapy. Cancer Reu. 5: 19, 1986. Lorence, R. M., Rood, P. A., and Kelly, K. W. Newcastle disease virus as an antineoplastic agent: Induction of tumor necrosis factor-alpha and augmentation of its cytotoxicity. J. Nat. Cancer Inst. 80: 1305, 1988. Ho, M. Factors influencing the interferon response. Arch. Int. Med. 126: 136, 1970. Merigan, T. C., De Clercq, E., Finkelstein, M. S., Clever, L., Walker, S., and Waddell, D. S. Clinical studies employing interferon inducers in man and animals. Ann. NY Acad. Sci. 173: 746, 1970. DeVita, V. T. Principles of chemotherapy. In V. T. DeVita, S. Hellman, and S. A. Rosenberg (Eds.), Cancer: Principles and Practice of Oncology. Philadelphia: J.B. Lippincott, 1989. Pp. 276-300. Ginsberg, H. S. Paramyxovirus. In B. D. Davis, R. Dulbecco, H. N. Eisen, and H. S. Ginsberg (Eds.), Microbiology, Including Immunology and Molecular Genetics, 3rd Ed. Hagerstown: Harper & Row, 1980. Pp. 1139-1159. Alexander, D. J. Newcastle disease: Methods of spread. In D. J. Alexander (Ed.), Newcastle Disease. Boston: Kluwer Academic Publishers, 1988. Pp. 256-272.

Newcastle disease virus selectively kills human tumor cells.

Newcastle disease virus (NDV), strain 73-T, has previously been shown to be cytolytic to mouse tumor cells. In this study, we have evaluated the abili...
932KB Sizes 0 Downloads 0 Views