Comparison of Long-term Survival of Cytomegalovirus Promoter ...

HYBRIDOMA Volume 18, Number 1, 1999 Mary Ann Liebert, Inc.

Survival of Cytomegalovirus Promoter versus Rous Sarcoma Virus Promoter-Driven Thymidine Kinase Gene Therapy in Nude Mice Bearing Human Ovarian Cancer

Comparison

of

Long-term

XIAOWEN TONG,5 DIRK G. ENGEHAUSEN,1 CHRISTIAN T.F. FREUND,4 IRA AGOULNIK,1 MARTIN K. OEHLER,1 TAE-EUNG KIM,1 ANNETTE HASENBURG,1 ZHONGSHENG GUO,2 CHARLES F. CONTANT,3 SAVIO L.C. WOO,2 and DIRK G. KIEBACK12-5

ABSTRACT

cytomegalovirus (CMV) promoter is considered one of the strongest positive regulators leading to expression of higher levels of the thymidine kinase (TK) enzyme than the Rous Sarcoma virus (RSV) promoter in vitro and in vivo. Cell killing efficacy of adenovirus-mediated CMV promoter-driven herpes simplex virus (HSV) TK gene therapy has been found to be 2 to 10 times more effective than RSV driven HSV-TK gene therapy in vitro. In this study the impact of CMV- versus RSV-driven HSV-TK gene therapy on long-term survival of nude mice bearing human ovarian cancer has been evaluated using a prospective randomized experimental design. The experiment was designed to show significance of survival differences from a 50% increase of survived days at ap-value of 0.05 with a power of 80%. All treatment groups showed an increase in median survival compared with control groups. Treatment benefit was ADV/CMV-TK vector dose dependent. At a given viral dose, no significant prolongation of survival was observed comparing CMV- and RSVdriven ADV-TK indicating that simply increasing cell killing efficacy in vitro above a minimal threshold level using a stronger promoter may not lead to prolongation of survival in the HSV-TK/GCV system. The

of the mono-, di-, and tri-phosphates by the cellular enzymes*15~16) and termination of nascent DNA chains whenever a nucleosides-triphosphate molecule is incorporated.'17"19' The tri-phosphorylated nucleoside analog kills cells possibly via apoptosis.(20) One important feature of the HSV-TK/GCV system is the "bystander effect," whereby nearby untransduced cells are also killed via intercellular gap junctional transfer of toxic ganciclovir (GCV) or acyclovir (ACV) metabolites*2'"23> and possibly by uptake of apoptotic vesicles.*24* Encouraging results using this treatment concept have also been achieved in treating a variety of different tumors including brain tumors, colon, liver, and head and neck cancer.*5"6-8"9) This attractive treatment approach is currently under investigation in clinical trials in treating brain tumor, prostate cancer, and ovarian cancer. Our previous studies utilizing an El-deleted adenoviral vector to deliver the HSV-TK gene under transcriptional control

INTRODUCTION

STANDARD debulking chemotherapy.*

therapy for ovarian cancer

surgery

"

and

includes'-radical

platinum-based

combination

Despite recent advances in the nonsurgical treatment, the majority of patients eventually die of their disease.(1) The resistance to standard therapeutic approaches has stimulated the development of new experimental approaches,

One of them is the "suicide gene therapy".*2"" Suicide genes transduced into tumor cells encode foreign protein that convert normally nontoxic compounds into toxic ones resulting in significant antitumor effects in a variety of tumors.*5"1 " One such suicide gene is the herpes simplex virus (HSV) thymidine kinase (TK) gene. The mechanism of the selective cell killing in HSV/TK-positive cells with purine and pyrimidine nucleosides can be considered to be a consequence of selective activation by a viral specified TK,*12"14) conversion

Departments of 'Obstetrics and Gynecology, 2Cell Biology, 3Neurosurgery and 4Urology, Baylor College of Medicine, Houston, TX 77030. department of Obstetrics and Gynecology, University of Frieburg, Frieburg 79106, Germany. 93

94 of the RSV promoter have shown significant efficacy in treating ovarian cancer in vitro and in vivo, resulting in cures of nude mice with this disease.*9""1 These data were later confirmed by another group.*25' Treatment efficacy was shown to be dependent on the viral dose and on the concentration of GCV. The amount of GCV was especially important for the bystander effect that leads to cell death of noninfected cells by cell-tocell transfer of toxic GCV triphosphate generated under the influence of a viral promoter.*10' Despite promising results, several studies have reported tumor reoccurrence after completion of HSV-TK/GCV therapy.*61619' It is therefore important to further refine and improve the HSV-TK delivery system. As treatment efficacy and bystander effect are dependent on TK activity as well as GCV doses, it seemed reasonable to postulate that higher levels of HSV-TK expression per cell could augment the ablation of cancer cells and might increase the number of cells sensitive to GCV treatment by simply manipulating the system driver to produce higher levels of enzyme. Expression levels of introduced genes depend largely on the strength of the enhancer/promoter'26-27' and the transduction efficiency of the gene transfer vector. Therefore, expression vectors used in gene transfer have been designed to utilize strong enhancer/promoters, usually of viral origin, such as the SV40 early promoter, the Rous sarcoma virus (RSV) long terminal repeat (LTR), and the human cytomegalovirus (CMV) immediate early (IE) gene promoter. In addition, some strong cellular promoters, such as that of the human elongation factor la gene, have also been utilized. The properties and activities of these enhancer/promoters have been well characterized in vitro and their relative strengths in driving reporter gene expression have been evaluated in vî'vo.*28' ADV/CMV-TK has been demonstrated to lead to expression of higher levels of the thymidine kinase enzyme than ADV/RSV-TK in vitro and in víW28' To investigate this hypotheses, comparison of ADVCMV/TK versus ADV-RSV/TK-mediated gene therapy in ovarian cancer was performed in vitro using three human epithelial ovarian cancer cell lines with different proliferation patterns. Two to tenfold differences in cell-killing efficacy were observed between the individual cell lines using different combinations of multiplicity of infection (MOI) and GCV dose.*29' The difference in cell-killing efficacy correlated with transduction efficacy: cells that were easy to transduce, like OV-CA-1225 and OV-CA-2774 showed less difference than the cell line SKOV3, which required larger numbers of viral particles for transduction. Promoter-dependent cell-killing efficiency was increased 8 to 10 times in this cell line. Among all three cell lines, any combination of GCV with ADV/CMVTK showed a higher or equal cell-killing efficacy than with ADV/RSV-TK at the same MOI. This was the first evidence that a stronger promoter can achieve higher cell-killing efficacy in ovarian cancer cells without increasing toxicity.*29' Encouraged by these data the impact of CMV- versus RSVdriven HSV-TK gene therapy on long-term survival of nude mice bearing human ovarian cancer was performed in a prospective randomized experimental design with the hope that prolongation of survival could be achieved by simply increasing HSV-TK enzyme levels per tumor cell without increasing toxicity for the host.

TONG ET AL.

MATERIALS AND METHODS Cell culture Three well-characterized human epithelial ovarian cancer cell lines OV-CA-2774 and OV-CA-1225 and SKOV-3 (purchased from ATCC (American Type Culture Collection, Manassas, VA) were selected for this experiment. Cells were maintained in HGDMEM (Dulbecco's Modified Eagle Medium with high glucose, Gibco #56-439-110; Gaithersburg, MD) supplemented with 1% penicillin/streptomycin (100X, Gibco #6005140AG), 1% glutamin (100X Gibco #320-5030AG), and 10% fetal bovine serum (Hyclon, Houston, TX) at 37°C and 5% C02 in a humidified incubator.

Construction

of recombinant adenovirus (ADV)

The construction and isolation of recombinant adenovirus of ADV/RSV-TK and ADV/CMV-TK were described previously.*5,6' The virus titer is based on biological infection (plaque-forming units; PFU). The same preparation of ADV/RSV-TK (containing 5.75 X 1010 PFU and ADV/CMVTK containing 2.55 X 1010 PFU, which have been used in vitro study were used in this in vivo experiment.

In vivo

experiments

Preexperiments. Tumorigenicity of SKOV3 was tested in 6 female CD-I nu/nu mice (Charles River Laboratories, Wilmington, MA) at the age 6-10 weeks. Tumor cells (1 X 108) in a volume of 2 mL Hanks buffer were injected intraperitoneally (IP). To minimize leakage of tumor cell concentrate, injections were performed slowly and the needle was removed only after a 10-sec delay after the actual inoculation. After injection, the mice were weighed daily. Cytological examination of the ascites and histologie examination of heart, lung, diaphragm, liver, spleen, kidney, bowel, uterus, ovary, and omentum were performed immediately after the tumor-induced death of the mice. All six mice died of ovarian cancer 2 to 3 months after tumor cell implantation. While SKOV-3 cells were tumorigenic, this animal model was too time-consuming for long-term survival experiments and the variation of long-term survival between untreated mice was large. We therefore had to abandon the animal model based on SKOV3 cells while it might have been optimal for the purpose of this study because SKOV3 cell showed the largest promoter-dependent difference in the in vitro studies. Another animal model on the basis of OV-CA-2774 was chosen to perform the in vivo study. 1 X 108 tumor cells in a volume of 2 mL Hanks buffer were injected IP into the nude mice (Charles River Laboratories, Wilmington, MA). After injection, the mice were weighed every day. Cytological examination of the ascites and histologie examination of heart, lung, diaphragm, liver, spleen, kidney, bowel, uterus, ovary, and omentum were performed immediately after the tumorinduced death of the mice. One day after IP injection of 1 X 108 cells, malignant ascites containing tumor cells will occur without obvious tumor present in the abdomen. Apparent bloody malignant ascites without macroscopic tumors can be

95

CMV- AND RSV-DRIVEN GENE THERAPY IN OVARIAN CANCER

120 TK1GTK2GTK3GTK4G-

Days survival of mice treated with different dosages of IP ADV-TK gene therapy or 6 days of IP GCV 0 alone 3 days after tumor inoculation with 108 cells. TK1 PFU ADV-TK + GCV (10 mg/kg), IP, b.i.d. for 6 days; TK2 2 X 108 PFU ADV-TK; TK3 6.7 X 108 PFU ADV-TK; TK4 2 X 109 PFU ADV-TK; and G- Negative control: same volume of Hanks buffer as GCV, IP, b.i.d. for 6 days. FIG. 1.

Long-term

=

=

=

=

=

seen in mice sacrificed 3 days after IP injection of 1 X 108 cells. Microscopically, superficial invasion of tumor cells will be observed throughout the abdominal surfaces. Copious bloody malignant ascites with macroscopic tumors dissemi-

nated in the abdomen will be found in mice sacrificed 7 days after tumor implantation. Experimental design. Based on our in vitro and in vivo experience, a prospective randomized treatment plan was designed to evaluate the therapeutic efficiency and toxicity of ADV/RSV-TK- versus ADV/CMV-TK-mediated gene therapy. Randomization was chosen as a means to eliminate potential confounding influences on the data by animal age, cell passage number, viral storage time, and investigators. Survival was chosen as the study endpoint. Each individual treatment and control group was to be comparable to each other with a statistical power of 80%. A 50% difference of survival time was considered the lower limit of statistical discrimination. In this design, 17 mice were needed for each treatment group. Eighty-five female CD-I nu/nu mice (Charles River Laboratories, Wilmington, MA) at the age of 6-10 weeks were split into five groups. Three days after tumor inoculation, 500 ph PBS containing different type and dosage of virus—( 1 ) 6.67 X 106 PFU ADV/CMV-TK; (2) 6.67 X 107 PFU ADV/CMVTK; (3) 6.67 X 108 PFU ADV/CMV-TK; (4) 2 X 109 PFU ADV/CMV-TK; and (5) 6.77 X 108 PFU ADV/RSV-TK— were injected IP followed by administration of GCV (10 mg/kg) at a concentration of 1 mg/mL or Hanks (IP, b.i.d.) for 6 consecutive days. Day 3 was chosen as the proper time for vector injection in this experiment because an ADV/RSV/TK vector dose dependent therapeutic pattern can be observed and no long-term survival has been achieved by using ADV-RSV/TK vector up to 2 X 109 PFU. Seventeen mice inoculated with tumor were treated per week. The logrank test was used to compare survival among the groups.

100?

CMV 6.67 x 10 e6 CMV 6.67 x 10 e7 CMV 6.67 x 10 e8 CMV2x10e9 RSV 6.67 x 10 e8

(0

> > 3

10 20 30 40 50 60 70 80 90 100110120

Days FIG. 2. Comparison of the impact between ADV/CMV-TK and ADV/RSV-TK gene therapy on long-term survival of nude mice bearing human ovarian cancer. Long-term survival of mice treated with different dosages of IP ADV-TK gene therapy and 6 days of IP GCV 3 days after tumor inoculation with 108 cells. CMV 6.67 X 10e6 6.67 X 106 PFU ADV/CMV-TK + GCV (10 mg/kg), IP, b.i.d. for 6 days; CMV 6.67 X 10e7 6.67 X 107 PFU ADV/CMV-TK + GCV (10 mg/kg), IP, b.i.d. for 6 days; CMV 6.67 X 10e8 6.67 X 108 PFU ADV/CMV-TK + GCV (10 mg/kg), IP, b.i.d. for 6 days; CMV 2 X 10e9 2 X 109 PFU ADV/CMV-TK + GCV (10 mg/kg), IP, b.i.d. for 6 days; and RSV 6.67 X 10e8 6.67 X 108 PFU ADV/RSV-TK + GCV (10 mg/kg), IP, b.i.d. for 6 days. =

=

=

=

=

96

TONG ET AL.

RESULTS The median survival of the mice treated with vector alone in the three different dosages and on Day 3 or treated with GCV (10 mg/kg) alone varied from 14.5 to 18 days. The median survival of the mice without any treatment was 16.5 days (Fig. 1). As shown in Fig. 2, all groups showed an increase in median survival compared with control groups. Treatment benefit was ADV/CMV-TK-dose dependent. The survival patterns were quite similar to those observed in the mice treated with ADV/RSV-TK.*9' Mice treated with 6.67 X 108 PFU ADV/CMV-TK and the same dose ADV/RSV-TK had an average survival of 84.7 ± 3 days (median 83 days) and 78.1 ± 12.9 d (median 80 days), respectively. The difference of longterm survival between these two groups was not statistically significant even though there was a trend that mice treated with ADV/CMV-TK survived longer than the mice treated with ADV/RSV-TK. No ADV-CMV/TK viral vector induced toxicity was observed up to 2 X 109 PFU ADV/CMV-TK or 2 X 1010 ADV/CMV-TK particles.

DISCUSSION CMV/TK-mediated gene therapy has been demonstrated to have higher cell-killing efficacy than RSV/TK gene therapy in three human epithelial ovarian cancer cell lines, including OVCA-2774 in vitro. The difference in cell-killing efficacy correlated with transduction efficacy. A doubling of promoterdependent cell-killing efficiency has been observed in OV-CA2774. However, no significant promoter-dependent increase of long-term survival was observed in nude mice even though there was a trend that mice treated with ADV/CMV-TK survived longer than the mice treated with ADV/RSV-TK. These results reflected the results observed in vitro, which showed that cell-killing efficacy was only mildly enhanced by using ADV/CMV-TK in the cell line OV-CA-2774, which is rapidly proliferating. These results indicate that simply increasing cellkilling efficacy in vitro above a minimal threshold level using a stronger promoter in the HSV-TK/GCV system may not lead to prolongation of survival. Because the maximum CMV-promoter-dependent increase in cell killing was observed in SKOV3 cells,*29' this cell line would have been best suited to evaluate the possibly improved long-term survival of nude mice bearing this type of cells when treated with ADV/CMV-TK. However, because of the long time needed for tumor manifestation and death in untreated animals and the large variation in survival, we were forced to chose the OV-CA-2774 animal model with a fast-growing tumor to evaluate the impact of different promoters on the long-term survival of nude mice. While no significant improvement was observed in this animal model by using ADV/CMV-TK in our experimental design that would have required at least a 50% difference in survival to become significant, these results do not exclude the possibility that ADV/CMV-TK may prolong survival in more slowly proliferating tumors. It is to be expected that tumor cells in vivo show a variation in transduction susceptibility similar to cell lines. In tumors behaving like OV-CA-2774 and OV-CA-1225, which are easily transduced, the treatment benefits of using CMV/TK versus

RSV/TK may not be as significant as in tumors that grow analogous to SKOV-3, which was more difficult to transduce and where a high MOI is needed. This may be important for the treatment of those ovarian cancer because high doses of virus will be needed to treat the entire peritoneal cavity. In these cases, the use of a stronger promoter may make a difference leading to higher expression of TK and reducing the necessary number of viral particles while achieving the same or higher GCV triphosphate production and cell-killing efficacy. As we did not observe increased toxicity, we believe that it will be worthwhile to develop the ADV/CMV-TK vector to clinical applicability, if pilot data show promise for TK/GCV as a clinical treatment concept and if the problem of the cost of the development process can be overcome.

REFERENCES 1. McGuire WP, Hoskins WJ, Brady MF, Kucera PR, Partridge EE, Look KY, Clarke-Pearson DL, and Davidson M: Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl J Med

1996;334:1-6. 2. Moolten FL: Tumor chemosensitivity conferred by inserted herpes thymidine kinase genes: Paradigm for a prospective cancer control strategy. Cancer Res 1986;46:5276-5281. 3. Moolten FL: Drug sensitivity ("suicide") genes for selective cancer chemotherapy. Cancer Gene Ther 1994;1:279-287. 4. Freemann SM, McCune C, Ange C, Abraham GN, and Abbound CN: Treatment of ovarian cancer using HSV-TK gene modified vaccine-regulatory issues. Hum Gene Ther 1992;3:342-349. 5. Chen SH, Shine HD, Goodman JC, Grossman RG, and Woo SL: Gene therapy for brain tumors: Regression of experimental gliomas by adenovirus-mediated gene transfer in vivo. Proc Nati Acad Sei USA 1994;91:3054-3057. 6. Chen SH, Chen XH, Wang Y, Kosai K, Finegold MJ, Rich S, and Woo SLC: Combination gene therapy for liver metastasis of colon carcinoma in vivo. Proc Nati Acad Sei USA 1995;92:2577-2581. 7. Moolten FL, Wells JM, Heyman RA, and Evans RM: Lymphoma regression induced by ganciclovir in mice bearing a herpes thymidine kinase transgene. Hum Gene Ther 1990;1:125-134. 8. O'Malley BWJ, Chen SH, Schwartz MR, and Woo SL: Adenovirus-mediated gene therapy for human head and neck squamous cell cancer in a nude mouse model. Cancer Res 1995;55: 1080-1085. 9. Tong XW, Block A, Chen SH, Contant C, Agoulnik I, Blankenburg K, Kaufmann RH, Woo SLC, and Kieback DG: In vivo gene therapy of ovarian cancer by adenovirus-mediated thymidine kinase gene transduction and ganciclovir administration. Gynecol Oncol 1996;61:175-179. 10. Tong XW, Block A, Chen SH, Woo SLC, and Kieback DG: Adenovirus-mediated gene therapy of human epithelial ovarian cancer by transduction with the thymidine kinase gene and exposure to ganciclovir. Anticancer Res 1996;16:1611-1618. 11. Tong XW, Agoulnik I, Blankenburg KP, Contant CF, Hasenburg A, Runnebaum IB, Stickeler E, Kaplan AL, Woo SLC, and Kieback DG: Human epithelial ovarian cancer xenotransplants into nude mice can be cured by adenovirus-mediated thymidine kinase gene therapy. Anticancer Res 1997;17(2A):811-813. 12. Cheng YC, Tsou TY, Hackstadt T, and Mallavia LP: Induction of thymidine kinase and DNase in varicella-zoster virus-infected cells and kinetic properties of the virus-induced thymidine kinase. J Virol 1979;31:172-177. 13. Kit S, Leung WC, Jorgensen GN, and Dubbs DR: Distinctive prop-

97

CMV- AND RSV-DRIVEN GENE THERAPY IN OVARIAN CANCER erties of thymidine kinase isozymes induced by human and avian herpes viruses. Int J Cancer 1974;14:598-610. 14. Kit S, Leung WC, Jorgensen GN, Trkula D, and Dubbs DR: Viral-induced thymidine kinase isozymes. Prog Med Virol 1975; 21:13-34. 15. Miller WH, and Miller RL: Phosphorylation of acyclovir (acycloguanosine) monophosphate by GMP kinase. J Biol Chem

1980;255:7204-7207. 16. Miller WH, and Miller RL: Phosphorylation of acyclovir diphosphate by cellular enzymes. Biochem Pharmacol 1982;31:38793884. 17. Cheng YC, Grill SP, Dutschman GE, Nakayama K, and Bastow KF: Metabolism of 9-(l,3-dihydroxy-2-propoxymethyl)guanine, a new anti-herpes simplex virus compound, in herpes simplex virusinfected cells. J Biol Chem 1983;258:12460-12464. 18. Culver KW, Ram Z, Walbridge S, Ishii H, Oldfield EH, and Blaese RM: In vivo gene transfer with retroviral vector-producer cells for treatment of experimental brain tumor. Science 1992;256:15501552. 19. Ezzeddine ZT, Martuza RA, Platika D, Short MP, Malick A, Choi B, and Breakefield XO: Selective killing of glioma cells in culture and in vivo by retrovirus transfer of the herpes simplex virus thymidine kinase. New Biol 1991;3:608-614. 20. Samejima Y, and Meruelo D: "Bystander killing" induces apoptosis and is inhibited by forskolin. Gene Ther 1995;2:50-58. 21. Elion GB : The biochemistry and mechanism of action of acyclovir. J Antimicrob Chemother 1983;12(Suppl. B):9-17. 22. Bi WL, Parysek LM, Warnick R, and Stambrook PJ: In vitro evidence that metabolic cooperation is responsible for the bystander effect observed with HSV tk retroviral gene therapy. Hum Gene Ther 1993;4:725-731. 23. Elshami AA, Saavedra A, and Zhang H: Gap junctions play a role in the bystander effect of the herpes simplex virus thymidine kinase/ganciclovir system in vitro. Gene Ther 1996;4:725-731.

24. Freeman SM, Abboud CN, Whartenby KA, Packman CH, Koeplin DS, Moolten FL, and Abraham GN: The "bystander effect": Tumor regression when a fraction of the tumor mass is genetically modified. Cancer Res 1993;53:5274-5283. 25. Behbakht K, Benjamin I, Chiu HC, Eck SL, Van Deerlin PG, Rubin SC, and Boyd J: Adenovirus-mediated gene therapy of ovarian cancer in a mouse model. Am J Obstet Gynecol 1996; 175:1260-1265. 26. Serfling E, Jasin M, and Schaffner W: Enhancers and eukaryotic gene transcription. Trends Genetics 1985;1:224-230. 27. Mcknight S, and Tjian R: Transcriptional selectivity of viral genes in mammalian cells. Cell 1986;46:795-805. 28. Guo ZH, Wang LHM, Eisensmith RC, and Woo SLC: Evaluation of promoter strength for hepatic gene expression in vivo following adenovirus-mediated gene transfer. Gene Ther 1996; 3:802-810. 29. Tong XW, Engehausen DG, Freund CTF, Agoulnik I, Guo ZS, Oehler MK, Kim TE, Contant CF, Hasenburg A, Woo SLC, and Kieback DG: THe efficacy of adenovirus-mediated gene therapy of ovarian cancer is enhanced by using cytomegalovirus promoter. Anticancer Res 1998;18:719-725.

Address reprint requests to: Dirk G. Kieback, M.D. Professor and Chairman Departments of Obstetrics and Gynecology

University of Freiburg Hugstetter Street 55 Freiburg 79106

Germany Received for publication June 6, 1998. tion June 8, 1998.

Accepted

for

publica-