Viral oncolysates in patients with advanced ovarian cancer

GYNECOLOCIC

ONCOLOGY

29, 337-347 (1988)

Viral Oncolysates

in Patients with Advanced Ovarian Cancer’

RALPH S. FREEDMAN,*” CREIGHTON L. EDWARDS,* JAMES M. BOWEN,? EVA LOTZOVAA RUTH KATZ,§ ERROLLEWIS,% NEELY ATKINSON,(~ AND RITA CARSETTI* Departments of *Gynecology; fTutnor Biology; ZGeneral Surgery, Laboratory of Immunogenetics; §Pathology; llRadiology; and j/Biomathematics, The University of Texas M. D., Anderson Hospital and Tumor Institute at Houston, Texas 77030

Received October 28, 1987 Viral oncolysates (VO) derived from two cultured ovarian carcinoma cell lines infected with influenza A/PR8/34 were administered intraperitoneally (IP) to 40 patients with advanced ovarian carcinoma, including 31 with late-onset ascites and 5 with pleural effusions. PR8 virus-specific antigens and ovarian tumor-associated antigens have been demonstrated on two oncolysates designated OVOI and OV02. Thirty-five patients received 9 mg of a 1: 1 mixture of OVOI and OV02, 5 patients received one or the other. During the first month three IP schedules were evaluated, i.e., single, biweekly, and weekly, which were followed by monthly injections. Intrapleural (IPI) injections of a 3.0-mg 1: 1 mixture of OVI and OV2 were administered to 3 patients concurrently with initial IP injections and to 2 patients following later development of pleural effusions. In 7 patients ascites disappeared; in 5 of these the number of cytologically detected malignant cells was markedly reduced, in 1 pleural effusion disappeared, and in 3 tumor masses were reduced. Tumor masses shrank also in 2 patients without ascites. Tumor reduction conformed to standard response criteria in 2 of the 5 patients. Response duration in the 9 responding patients lasted from 3 to 19 months and survival durations 4 to 42 months. Disease symptoms in 7 patients improved noticeably. Two of the 9 responders later developed unilateral pleural effusions that responded for 7 and 15+ months to a single IPI injection. Seventeen patients experienced one or more treatment side effects including fever, nausea or anorexia, malaise, abdominal pain, and arthralgia, but in only 2 patients, both on the weekly schedule, was toxicity severe enough to require treatment withdrawal. Humoral responses to viral and tumor cell-surface antigens were frequently observed in patients demonstrating clinical activity. o 1988 Academic Press, Inc.

INTRODUCTION

Virus augmentation is a biologic response modifier (BRM) approach by which the immunogenicity of tumor cell extracts is enhanced by previously infecting the tumor cells with suitable viruses, including influenza virus, vesicular stomatitis virus (VSV), Newcastle disease virus (NCDV), and vaccinia [l-5]. ’ Presented at the Felix Rutledge Society Meeting, Louisville, KY, May 21-25, 1986. This work was supported by a grant from the Blum Kovler Foundation. * To whom reprint requests should be addressed at Department of Gynecology, UT M. D. Anderson Hospital and Tumor Institute, 6723 Bertner Avenue, Houston, TX 77030. 337 0090-8258/88$1.50 Copyright 0 1988 by Academic Press, Inc. All rights of reproduction in any form reserved.

338

FREEDMAN

ET AL.

The virus-augmentation effect was first demonstrated in animals. Researchers compared the reactions of animals protected from transplanted tumors by prior immunization with viral oncolysates (VO) to those of animals that received unmodified tumor cell extract, virus alone, or an admixture of virus and extract. Intensified delayed-type hypersensitivity reactions (DTHR) were observed in animals injected with VO. Furthermore, some animals that had established tumors survived longer than expected after injection with VO [2,4]. Humans who received allogeneic extracts of cultured tumor cells similar to their own tumors also showed intensified DTHR [l]. We observed similar reactions in patients with ovarian cancer: When injected with VO derived from two ovarian carcinoma cultures their DTHR were augmented above those of patients injected with either influenza virus or unmodified tumor cell extracts alone [6]. Although possible therapeutic applications for VO have been investigated in various tumors [7-151, our study reports the first attempts to administer VO by the intraperitoneal (IP) and intrapleural (IPl) routes. Because of its regional spread pattern, ovarian cancer may be well suited to regional BRM approaches. Moreover, there is increasing evidence that the natural immunity is impaired in patients with ovarian cancers which is seen most clearly in lymphocytes obtained from the peritoneal cavity [16,17]. In an earlier report we observed a significant increase in ascitic-fluid natural killer (NK) cell cytotoxicity after administering IP ovarian VO to three patients with ovarian cancer in a weekly schedule that escalated the dose from 1.5 to 12.0 mg [5]. Persistent anorexia and malaise were, however, experienced by two of these three patients. In the present study, 40 additional patients received IP ovarian VO under an Institutional Review Board-approved protocol. Clinical activity, toxicity, and humoral responses to IP VO were evaluated in this patient group. Five of the 40 patients also received IPl VP injections, and their responses to the treatment were evaluated. MATERIALS AND METHODS Patients. The 40 patients eligible for the study had advanced disease that was refractory to conventional therapies. Their median age was 55 years, the range 26-79 years. Serous carcinoma was diagnosed in 28 patients, and other variants of epithelial carcinoma, including mutinous, endometrioid, clear cell, undifferentiated, and mixed patterns, were present in the remaining 12 patients. Tumor was moderately to poorly differentiated in 35 patients and well differentiated in 5. Clinically obvious malignant ascites of late onset was present in 31 patients. Repeated therapeutic paracenteses had been necessary in 23 patients. Measurable abdominal or pelvic tumor masses were present in 35 patients. All patients previously had received cisplatin, and 31 had received two or more systemic therapy regimens in addition to surgery. Preparation and characteristics of viral oncolysates. Two cultured and characterized ovarian carcinoma cell lines, MDAH 2774 and CaOV3, were used in preparing ovarian VO, as described previously [6]. CaOV3 was kindly provided by J. Fogh of the Sloan Kettering Hospital for Medical Research; MDAH 2774 was originally provided to Ralph S. Freedman by J. Sinkovics of St. Joseph’s

OVARIAN

VIRAL

ONCOLYSATES

IN

ADVANCED

339

CANCER

TABLE 1 HUMAN LEUKOCYTE ANTIGEN CHARACTERIZATION OF OVARIAN CARCINOMA CELL LINES Cell line MDAH 2774 CaOV3

HLA-A 3,24 28,29

HLA-B 34,w57 49,W61

HLA-CW 1,X 4,x

HLA-DR

2S 5,W8

MT MTI, MTI,

TDll 4

Community Cancer Center in Tampa, Florida. The cells underwent isozyme analysis by M. Sicilian0 of the University of Texas M. D. Anderson Hospital and Tumor Institute at Houston (UT MDAH) [18]. Both lines were independent of HeLa-cell contamination, since they had the b phenotype for G6PD. The cell line MDAH 2774 was heterozygous [1,2] for EsD, and CaOV3 was only type 1 for that locus, which indicates that the origins of the lines were independent of each other. Genotypes at certain other loci may be informative for identifying these cell lines in the future, since they were not the most frequent in the human population: MDAH 2774 was heterozygous [I ,2] for PGP and had the less common form (type 1) for GLOI; CaOV3 had the less common forms, 2 and 1, for ME2 and GLOl, respectively. Human leukocyte antigen analysis of the cell lines was performed by H. A. Fischer of UT MDAH and, on a standardized two-stage NIH microcytotoxicity assay [ 191, revealed the additional characteristics shown in Table 1. Surface-antigen characterization of tumor cells used in preparing ovarian VO was performed with murine monoclonal antibodies (MoAb) to blood group and tumor-associated antigens by fluorescence-activated cell- sorter analysis. The Ca-125 antigen, commonly associated with nonmucinous ovarian carcinoma tissues, was detected on the surface membrane of 89% of MDAH 2774 cells and 33% of CaOV3 cells, whereas blood groups A, B, and Lewis a and b were notably absent. Control antibody stained 2% of MDAH 2774 cells and 3% of the CaOV3 cells. The PR8/34 strain of type A influenza was used to infect the tumor cells, as previously described [6]. Lysates of the two virus-infected ovarian carcinoma cell lines, MDAH 2774 and CaOV3, were designated, respectively, OVOl and OV02. An aliquot of each extract was adjusted to a protein content of 4.5 mg and suspended in 5 ml N saline for IP injections. PR8 influenza-specific hemagglutinins and neuraminidase were demonstrated in more recent preparations of OVOl and OV02, using murine MoAbs against HA:H37-45-4, HA: H36-4-5, and NA: 11l-106 that were obtained from M. Herlyn and W. Gerhard of the Wistar Institute [20]. One microgram aliquot of each VO preparation was allowed to react with MoAbs in a standardized enzyme-linked immunoabsorbance assay 1211. The same assay procedure demonstrated that OVOl was reactive with MoAb 3 174, derived from and reactive with MDAH 2774, and OV02 was reactive with MoAb 3174 and MT334. MT334 was obtained from M. J. Mattes of the Sloan Kettering Institute for Medical Research and has been reported to be reactive with tissue sections of epithelial ovarian carcinoma and CaOV3 [22]. Supernatants of OVOl and OV02 were also assayed for interferons (IFN),

340

FREEDMAN

ET AL.

using radioimmunoassay detection kits (Roche Diagnostics, Nutley, NJ, for IFN(Y,and Centocor, Malvem, PA, for IFN-a), and a spectrum of cytotoxic cytokines detectable in a standardized bioassay by J. Kostergaard [23]. No measurable levels of either IFN or other cytokines were found in these extracts. IFN assays were performed by H. Fritsche in the Department of Laboratory Medicine at UT MDAH. Aliquots from each VO batch were also assayed for bacterial sterility by a standardized procedure [24]. Viral oncolysate administration, dose, and schedule. OVOl and OV02 aliquots were mixed and suspended in 10 ml nonpyrogenic saline for IP injection, usually through a No. 15gauge angiocatheter. Each IP dose was equivalent to 9.0 mg. No concomitant chemotherapy was administered. Thirty-five of the 40 patients received 9.0-mg doses of a 1: 1 combination of OVOl and OV02 by IP injection. Four patients received 9.0 mg OVOl alone and 1 received 9.0 mg OV02 alone, because the accompanying extract was unavailable at that time. During the first month, 28 patients received a single injection, 6 received one injection 2 weeks apart, and 6 received weekly injections. A semipermanent catheter (Tenckhoff type) was already in place from previous IP chemotherapy in 3 patients: catheters were placed in 2 patients just prior to administering IP VO. An ultrasound-guided technique was used to identify the peritoneal cavity in patients without ascites or in those whose ascites disappeared following treatment with IP VO. We used three IP injection schedules, i.e., weekly, biweekly, or monthly, during the first month to determine treatment tolerance. Subsequent injections were given at monthly intervals. Five patients received additional IPl injections of a 1: 1 mixture of OVl and OV2 equivalent to 3.0 mg. Injections were repeated in 1 patient. IPl injections were administered simultaneously with initial IP injections in 3 patients who had presented with pleural effusions and ascites, and at later dates to 2 patients who developed pleural effusions while undergoing IP OVO injections. Evaluation of clinical response. Clinical activity, documented by two or more observers, was indicated by disappearance of malignant ascites and stabilization or decrease in tumor masses for 8 or more weeks from the start of treatment. Standard criteria for tumor response were employed in patients without ascites. Disappearance or significant decrease in malignant cells following VO injection was considered a further indication of clinical activity. Subjective responses were assessed by the Zubrod scoring system. The frequency and type of treatment side effects were evaluated in all patients. Cytopathology of ascitic fluid. The proportion of malignant cells present in cytocentrifuge preparations of ascitic fluid was measured by one of the authors (R.K.) using nuclear and cytoplasmic criteria previously established in the Department of Cytopathology. Cytocentrifuge preparations were fixed immediately in 95% alcohol and stained with the Papanicolaou technique. A total of 500 cells were counted under 400x magnification. Malignant cell clusters were counted as single units because of the inability to count individual cells comprising the clusters. Pretreatment samples were compared to post-treatment samples taken before ascitic fluid disappeared. Serologic response to viral oncolysates. Before injection and at l-2 months

OVARIAN

VIRAL

ONCOLYSATES

IN

ADVANCED

CANCER

341

after injection, serum antibody titers to viral hemagglutinin and surface membrane antigens were determined on the cultured ovarian carcinoma cells used in preparing OVOl and OV02. Virus hemagglutination and protein A surface-binding assays were used, as previously described [8,25]. RESULTS

Nine of the 40 patients showed evidence of clinical activity after receiving IP or IP plus IPl OVO. No patient required a thoracostomy tube. Characteristics of this group of responding patients are shown in Table 2. Ascites regressed in 7 patients. In 5 of the 9 patients, 3 with ascites and 2 without, tumor masses decreased; the decrease in 2 patients (No. 5. and No. 8) conformed to standard criteria for a partial response. The response times in these 9 patients ranged from 3 to 19 months, and their survival durations ranged from 4 to 42 months. In 7 of the 9, patient’s subjective improvement was equated with a resumption of most normal activities. Single and clustered malignant cells became undetectable prior to ascites disappearance in 4 responding patients and declined significantly in 2 other patients: 1 of the latter (not shown in Table 2) had a transitory decrease in ascites but showed no clinical improvement. Cytologic changes were accompanied by inflammatory responses, which were notably mononuclear cell in type. Treatment doses were administered as follows: 2 or fewer doses, 13 patients; 3-5 doses, 15 patients; 6-10 doses, 8 patients; and more than 10 doses, 4 patients. Six of the 9 responding patients received a single injection each during the first month, and 3 received 2 injections each. No responses occurred in patients who received weekly injections, but 2 in this group withdrew because of anorexia and malaise. Three patients received IPl VO concurrently with the first IP injection. The effusion of one (patient 3, Table 2) disappeared for 7 months. Before treatment with IPl OVO, this patient had undergone an unsuccessful attempt at tetracyclineinduced pleurodesis and was bedridden and receiving continuous oxygen. Two patients (No. 2 and No. 5 in Table 2) developed pleural effusions at 12 and 6 months, respectively, after receiving IP injections of OVO. Pleural effusions regressed in both for 15+ and 7 months, respectively, after they received IPl VO. In an additional patient, not shown in Table 2, a pleural effusion remained for 5 months following repeated IPl injections. Side effects and toxicity. One or more side effects were experienced by 17 patients. The frequency of each side effect is shown in Table 3. Temperature elevations occurred after 6 hr and were occasionally as high as 39°C. Defervescence usually occurred after 24 to 48 hr. Other symptoms included nausea or anorexia, occasional vomiting, diarrhea, fatigue, and abdominal pain. In most patients these symptoms abated within 24 to 48 hr, but in 3 the malaise, abdominal pain, or both persisted for several days, requiring hospitalization for 2 patients and leading to treatment withdrawal for 2 patients. Weekly IP injection schedules were poorly tolerated because of persistent anorexia and fatigue. No hematologic, liver, or renal function abnormalities could be attributed to IP or IPI VO injections. Serologic responses to viral oncolysates. Serum antibody titers to the PR8

5

5.5

5.5

III

III

70 Serous

71 Serous

1.00

III

60 Adenocarcinoma

4.25

II

I.25

Duration (years)

32 Serous

Grade

I

Pathologic designation

39 Mutinous

No. Age

CLINICAL

FOLLOWING

Recurrent ascites, pelvic mass (2) Recurrent ascites, pelvic/abdominal mass (2) Recurrent ascites, pleural effusion, supraclavicular nodes (3) Recurrent ascites, weekly paracentesis, poor appetite, fatigue, pelvic mass (2) Ascites, CTdocumented liver and para-aortic metatases, fatigue, weight loss (2)

Pretreatment status (Zubrod scale)

ACTIVITY

VIRAL

Para-aortic mass regressed, liver stable on CT for 12 months (NI)

Ascites regressed, tumor stable (2)

Pleural effusion, regressed, nodes stable (1)

Ascites regressed, mass smaller (0) Ascites regressed, masses stable (0)

6’

5

7’

12”

19

13

10

7.5

21

42

Survival (months)

INJECTIONS

Duration (months)

ONCOLYSATE

Response (Zubrod scale”)

TABLE 2 IP AND IPI

MC and MCC (1W

MC (100) MCC (99)

ND

MC and MCC’ (loo) MC and MCC (100)

Cytology (% decrease)

10

10

6

21

17

1

2

1

I

1

Dose No. Scheduleb

51 Serous

26 Serous

50 Serous

7

8

9

III

I

III

III

5.0

2.4

5.0

2.6

Ascites, abdominal Ascites regressed, mass, subacute obstruction temporarily bowel obstruction (3) relieved, mass stable (I) Recurrent ascites, Ascites regressed, <50% regression abdominal, pelvic, of pelvic mass (I) and subcutaneous masses (3) Abdominal pelvic >50% regression of abdominal, pelvic masses, recent bowel masses (0) bypass surgery (I) Vulvar and pelvic <50% regression masses (I) pelvic mass, vulvar lesions stable (NI) NI ND

I5 9

5

I2

MC and MCC (100)

13

4

ND

4.4

3

8

I8

8

6

I

2

I

2

” Zubrod Scale: 0 = no symptoms; I = symptoms, fully ambulatory; 2 = requires nursing assistance or equivalent, bedridden 50% of normal day; 3 = bedridden 50% of normal day; 4 = unable to get out of bed; NI = no improvement. ’ Injection frequency during first month. ’ MC = single malignant ascitic cells; MCC = malignant cell clumps; ND = no data. ’ Left-sided pleural effusion at I2 months; disappeared after single IPI VO injection. c Combined IP and UPI VO injection. ’ Left pleural effusion at 6 months disappeared after single IPI VO injection.

79 Serous

6

344

FREEDMAN

ET AL.

TABLE 3 TOXICITY FROM INTRAPERITONEALVIRAL ONCOLYSATEAS RELATED TO SCHEDULE

Toxic symptoms Schedule“

Fever and chills

Nausea/ anorexia

Monthly Biweekly Weekly Total

8 3 2 13

WY 3

2(2) 9

Vomiting

Diarrhea

Fatigue

WI 1 1 7

5 1 1 7

W)

Abdominal pain

Arthralgia

10 3 l(1) 14

1 2 0 3

3

w 13

a Injection schedule during first month. ’ ( ) = moderate to severe, causing disability or withdrawal.

virus and surface membrane of MDAH 2774 and CaOV3 are shown before and after treatment in Fig. 1. Virus antihemagglutinin titers increased fourfold or greater in 10 of 23 patients, including 7 of the 9 patients shown in Table 2. Before their injections with VO, serum-activated binding of protein A-coated red blood cells to MDAH 2774 was observed in 1 of 21 patients and 5 patients showed reactivity against CaOV3. After treatment, 16 patients developed fourfold or more titer elevations to MDAH 2774 and 9 patients to CaOV3. In 7 of the 9 patients shown in Table 2, titers were significantly elevated against one or both cell lines.

P F F 4

409620481024512256-

00 0

wo

0

0

aeoo

0

0

9

128-

WO

co

64-

mxx3

2 8

32-

E G ti

I42O-

883

0

CCD

0

0

00 0

0 a300

uoo 030

0

0

0

0

0

0

00

0

00

gig 00

0 (7) 8 (13) POST

PRE VlFtlJs-~m

(23 PATiENfS)

03

PRE

POST

ANTISOOY TO SURFACE

WV POST

PRE ANnsoDY

TO !3UFACE

MmFtAEAN-lEa

-ANTtGEN

ON MMH 2774 (21 PATIENTS) l -Respcndll patients shown in TaMe 2

(6) (10)8

ON caov3 (21 PATIENTS)

0 -Patients withoutclinical activity

FIG. 1. Serologic responses before and after viral oncolysate treatment. Post-treatment samples were obtained between 1 and 2 months.

OVARIAN

VIRAL

ONCOLYSATES

IN

ADVANCED

CANCER

345

DISCUSSION Immune reactive preparations of ovarian VO that have been partially characterized were derived from cultured ovarian carcinoma cells that had been modified by influenza A/PR8/34, as described. VO was administered by IP and IPl routes to patients with advanced recurrent carcinoma of the ovary considered to be refractory to conventional therapy. Regression of malignant effusions and reduction of some tumor masses indicated a modest degree of clinical activity in this group of patients. Malignant effusions that result from recurrent ovarian and nonovarian malignancies have been reported to respond to several biologic response modifiers, including IP or IPl Corynebacterium parvum [17,26-281, IP interferon [28], and IPl lyophilized extract of Streptococcus pyogenes [29]. More recent studies have also described C. parvum and interferon activity in patients who have ovarian cancer accompanied by persistent peritoneal carcinomatosis with microscopic or small macroscopic disease [30,31]. Similar patients were not included in this study. Side effects of IP or IPl ovarian VO were similar to those of other biologic response modifiers discussed, but few patients experienced incapacitating problems with malaise and anorexia. More frequent injections, e.g., once a week, may be associated with increased frequency and severity of side effects, although clinical activity was observed on the less frequent injection schedule. Several viruses can result in virus augmentation, but the myxo viruses influenza and NCDV have properties that may favor their use in therapeutic studies. Influenza and NCDV are strongly antigenic surface-budding viruses, and VO prepared with influenza virus appear to be more immunogenic than VSV oncolysates [3]. Furthermore, VO prepared with influenza are effective in prolonging the survivals of animals with established tumors [3,14], and NCDV VO have been observed to cause regression of established tumors [4]. An attenuated strain of influenza virus such as PR8/A/34 fulfills the principal requirements for virus augmentation and has proved safe in our current and past clinical trials [6,8,9]. Mechanisms responsible for virus augmentation could involve both cell-mediated and humoral responses. Delayed-type hypersensitivity responses, which are considered T cell mediated, were observed after intradermal ovarian VO injection [61. We have also observed non-T or natural killer cell cytotoxicity following IP VO injections, which may remain at high levels even at 21 days following a single IP VO injection [ 16,321. Inoculation of live attenuated influenza virus into human subjects may also augment NK cell cytotoxicity but the augmentation appears to be of short duration [33]. Moreover, the virus-replication time and inactivation steps currently in use have markedly reduced the quantity of live influenza virus in the oncolysate preparations. Intracavitary injections of C. parvum, interferon, and streptococcal antigen may be associated with an increase’ in NK cell cytotoxicity [29,30], although others have not observed such augmentation following IP C. parvum injection in patients with malignant ascites [171. Boone and co-workers have proposed that the virus augmentation phenomenon

346

FREEDMAN

ET AL.

involves virus-induced antigens placed in juxtaposition to weaker tumor-associated antigens [2]. Antigenicity of the virus is considered important since the immunogenic effect of VO may be abrogated by inducing tolerance to the virus. Our patients developed antibody titers both to the virus and tumor cell-surface membrane determinants after VO injections. Moreover, in related studies following VO injections, postimmunization sera have appeared to recognize a variety of tumorassociated antigens [34]. We have not yet determined whether this humoral response is of clinical importance in our patients, but we have made progress toward cloning the serologic response through the production of human monoclonal antibodies that react with antigen determinants on the cultured ovarian carcinoma cells used in the preparation of VO, and with ascitic tumor cells derived from nonimmunized patients with ovarian cancer. These experiments may ultimately help to detemine if antibody-dependent cytotoxicity mechanisms are involved, and the biochemical characteristics of the virus-augmented antigens. REFERENCES 1. Austin, F. C., and Boone, C. W. (Eds.) Virus augmentation of the antigenicity of tumor cell extracts, in Advances in cancer research, Academic Press, New York, Vol. 30, pp. 301-345 (1979). 2. Boone, C. W. Augmented immunogenicity of tumor cell homogenates infected with influenza virus, Recent Results Cancer Res. 47, 394-400 (1974). 3. Gillette, R. W., and Boone, C. W. Augmented immunogenicity of tumor cell membranes produced by surface budding viruses: Parameters of optimal immunization, Znt. J. Cancer 18, 216-222, (1976). 4. Heicappell, R., Schirrmacher, V., Von Hoegen, P., Ahlert, T., and Appelhans, B. Prevention of metastatic spread by postoperative immunotherapy with virally modified autologous tumor cells. I. Parameters for optimal therapeutic effects, Znt. J. Cancer 37, 569-577 (1986). 5. Lindenmann, J., and Klein, P. A. Viral oncolysis: Increased imunogenicity of host cell antigen associated with influenza virus, .Z. Exp. Med. 126, 93-108 (1967). 6. Freedman, R. S., Bowen, J. M., Atkinson, E. N., Scott, W., and Wagner, S. Virus-augmented delayed hypersensitivity skin tests in gynecological malignancies, Cancer Zmmunol. Zmmunother. 17, 142-146 (1984). 7. Cassel, W. A., Murray, D. R., and Phillips, H. S. A phase II study on the postsurgical management of stage II malignant melanoma with Newcastle disease virus oncolysate, Cancer S&856-860 (1983). 8. Freedman, R. S., Bowen, J. M., Herson, J. H., Wharton, J. T., Edwards, C. L., and Rutledge, F. N. Immunotherapy for vulvar carcinoma with virus-modified homologous extracts, Obstet. Gynecol. 152, 707-714 (1983). 9. Freedman, R. S., and Rutledge, F. N. Adjunctive immunotherapy with VMTCE in patients with high risk squamous carcinoma of the uterine cervix, Amer. J. C/in. Oncol. 6, 155-156 (1983). 10. Green, A. A., Pratt, C., Webster, R. G., and Smith, K. Immunotherapy of osteosarcoma patients with virus-modified tumor cells. Ann. N.Y. Acad. Sci. 277, 396-411 (1976). 11. 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, 680-686 (1977). 12. Sauter, C., Cavalli, F., Lindemann, J., Gmur, J. P., Berchtold, W., Alberta, P., Obrecht, P., and Senn, H. J. Viral oncolysis: Its application in maintenance treatment of acute myelogenous leukemia, in Immunotherapy of cancer: Present status trials in man (W. D. Terry and D. Windhorst, Eds.), Raven Press, New York, pp. 355-363 (1978). 13. Sinkovics, J. G. Immunotherapy with viral oncolysates for sarcoma. (Letter). J. Amer. Med. Assoc. 237, 869 (1977). 14. Sinkovics, J. G. Viral oncolysates for the immunotherapy of human tumors, Proceedings, 13th International Congress of Chemotherapy, Vienna, Austria, pp. 225-243 (1983).

OVARIAN

VIRAL

ONCOLYSATES

IN ADVANCED

CANCER

347

IS. Wallack, M. K., Meyer, M., Bourgoin, A., Dore, J. F., Leftheriotis, E., Carcagne, J., and Koprowski, H. A preliminary trial of vaccinia oncolysates in the treatment of recurrent melanoma with serologic responses to the treatment, J. Biol. Response Modif 2, 586-596 (1983). 16. Lotzova, E., Savary, C. A., Freedman, R. S., and Bowen, J. M. Natural killer cell cytotoxic potential of patients with ovarian carcinoma and its modulation with virus-modified tumor cell extract, Cancer Zmmunol. Zmmunother. 17, 124-129 (1984). 17. Mantovani, A., Allavena, P., Sessa, C., Bolis, Cl., and Mangioni, C. Natural activity of lymphoid cells isolated from human ascitic ovarian tumors, Znt. J. Cancer 25, 573-582 (1980). IS. Siciliano, M. J., Barker, P. E., and Cailleau, R. Mutually exclusive genetic signatures of human breast tumor cell lines with a common chromosomal marker, Cancer Res. 39, 919-921 (1979). 19. Pollock, M. S., Heagney, S. G., and Fogh, J. HLA typing of cultured human tumor lines: The detection of genetically appropriate HLA-A, B, C, and DR alloantigens, Transplant. Proc. 12, 134-137 (1980). 20. Staudt, L. M., and Gerhard, W. Generation of antibody diversity in the immune response of BALB/c mice to influenza virus hemagglutinin, J. Exp. Med. 157, 687-704 (1983). 21. Chan, J. C. A monoclonal antibody generated against MSV-induced transformation related proteins (MC), Monoclonal Antibody News 2, 7-9 (1983). 22. Mattes, M. J., Cordon-Cardo, C., Lewis, J. L., Jr., Old, L. J., and Lloyd, K. 0. Cell surface antigens of human ovarian and endometrial carcinoma defined by mouse monoclonal antibodies, Proc. Natl. Acad. Sci. USA 81, 568-572 (1984). 23. Klostergaard, J. A rapid extremely sensitive, quantitative microassay for cytotoxic cytokines, Lymphokine Res. 4, 309-317 (1985). 24. General biological standards, Fed. Regist. 38, 32065, Section 610.12 (1973). 25. Pfreundschuh, M., Shiku, H., and Takahashi, T. Serological analysis of cell surface antigens of malignant human brain tumors, Proc. Nat/. Acad. Sci. USA 75, 5122-5126 (1978). 26. Currie, J. L., Gall, S., Weed, J. C., and Creasman, W. T. Intracavitary Corynebacterium parvum for treatment of malignant effusions, Gynecol. &co/. 16, 6-14 (1983). 27. Webb, H. E., Oaten, S. W., and Pike, C. P. Treatment of malignant ascites and pleural effusions parvum, &it. Med. J. 1, 339-340 (1978). with Corynebacterium 28. Rambaldi, A., Introna, M., Colotta, F., Landolfo, S., Columbo, N., Mangioni, C., and Mantovani, A. Intraperitoneal administration of interferon beta in ovarian cancer patients, Cancer 56,294301 (1985). 29. Uchida, A., and Micksche, M. Intrapleural administration of OK432 in cancer patients: Activation of NK cells and reduction of suppressor cells, Znt. J. Cancer 31, l-5 (1983). 30. Berek, J. S., Knapp, R. C., Hacker, N. F., Lichtenstein, A., Jung, T., Spina, C., Obrist, R., Grifhths, C. T., Berkowitz, R. S., Parker, L., Zighelboim, J., and Bast, R. C., Jr. lntraperitoneal immunotherapy of epithelial ovarian carcinoma with Corynebacterium parvum, Amer. J. Ubstet. Gynecol. 152, 1003-1010 (1985). 31. Berek, J. S., Hacker, N. F., Lichtenstein, A., Jung, T., Spina, C., Knox, R. M., Brady, J., Greene, T., Ettinger, L. M., Lagasse, L. D., Bonnem, E. M., Spiegel, R. J., and Zighelboim, J. Intraperitoneal recombinant alpha-interferon for “Salvage” immunotherapy in stage III epithelial ovarian cancer: A Gynecologic Oncology Group study, Cancer Res. 4.5, 4447-4453 (1985). 32. Lotzova, E., Savary, C. A., and Freedman, R. S. Involvement of NK cells in human ovarian cancer, in Diagnosis and treatment strafegiesfor gynecologic cancer. (F. N. Rutledge, D. M. Gershenson, and R. S. Freedman, Eds.), Univ. of Texas Press, Austin, in press. 33. Ennis, F. A., Meager, A., Baere, A. S., Qi, Y. H. U., Riley, D., Schwarz, G., Schild, G. C., and Rook, A. H. Interferon induction and increased natural killer-cell activity in influenza infections in man, Lancet 2, 891-893 (1981). 34. Savage, H. E., Rossen, R. D., Hersh, E. M., Freedman, R. S., Bowen, J. M., and Plager, G. 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. 46, 2127-2133 (1986).