Genotoxic effects of five polycyclic aromatic hydrocarbons in human and rat mammary epithelial cells.

Five polycyclic aromatic hydrocarbons (PAHs) of different carcinogenic activities were evaluated for their effects on DNA synthesis (3HTdR labeling index (L.I.] of rat and human mammary epithelial cells (MEC) and for their effects on chromosomes in MEC-mediated sister chromatid exchange (SCE) assays. When compared with DMSO-treated cells, exposures of rat MEC to the two most potent carcinogens (5 micrograms/ml for 24 hr), i.e., 7,12-dimethylbenz(a)anthracene (DMBA) and benzo(a)pyrene (B[a]P), resulted in a 45-62% reduction in the L.I. of rat MEC. Another carcinogen, 20-methylcholanthrene (MCA), produced a 35-48% reduction in L.I., while the noncarcinogenic PAHs, 1,2-benzanthracene (BA) and benzo(e)pyrene (B[e]P), showed no effect. Similarly, exposures of human MEC to DMBA and B[a]P resulted in a 50-90% depression in L.I. while BA was significantly less effective (30% reduction). When co-cultivated with Chinese hamster V-79 cells in the presence of PAH, both rat and human MEC can activate and release the active metabolites to induce SCE in V-79 cells. In the rat MEC-mediated assay for all 5 PAHs, the frequencies of SCE per chromosome in DMBA-, B[a]P-, MCA-, BA-, B[e]P-, and DMSO (solvent control)-treated groups were 6, 3, 1.4, 0.7, 0.4, and 0.3, respectively. DMBA was most effective in increasing SCE, while B[e]P was ineffective. In the human MEC-mediated assay, B[a]P was more effective than DMBA in inducing SCE, and the frequencies of SCE per chromosome were 4.5 and 3.6 in B[a]P- and DMBA-treated groups, respectively. Comparing depression of L.I., SCE, and in vivo carcinogenicity for the 5 PAHs, SCE mediated by rat MEC is better correlated with carcinogenicity in rat than L.I. depression.

Expression of carcinoembryonic antigen, T-antigen, and oncogene products as markers of neoplastic and preneoplastic colonic mucosa.

Several crypt abnormalities have been demonstrated in the mucosa of neoplastic and preneoplastic lesions of the large intestine. In addition, certain tumor markers are expressed in large intestinal carcinoma but not in normal mucosa. To determine whether any correlation exists between tumor marker expression and crypt abnormalities and at what stage markers are expressed, we studied specimens of large intestinal mucosa from 13 patients with preneoplastic conditions (adenomatous polyp, familial polyposis, Crohn's disease, and ulcerative colitis). The tumor markers examined include carcinoembryonic antigen (CEA), the ras gene products p21 and p21ser (mutated form), and beta-D-galactosyl-(1----3)-alpha-N-acetyl-D-galactosamine (gal--gal NAc, also known as T-antigen). Results were compared to those of five cases of adenocarcinoma of colon and three control cases of colonic mucosa obtained at immediate autopsy. All four markers were expressed in three of the five cases of adenocarcinoma, but none were expressed in the control cases. Variable expression of each marker was demonstrated in the dilated, distorted crypts of preneoplastic lesions. CEA and gal--gal NAc appeared to be expressed most frequently, suggesting that these are common markers or are expressed at an earlier stage in the neoplastic process than p21 or p21ser. Demonstration of such markers in preneoplastic conditions may be of use in determining the malignant potential and in monitoring these lesions.

Immunohistochemistry of the cytoskeleton of human prostatic epithelium. Evidence for disturbed organization in neoplasia.

An indirect immunoperoxidase technique was used to evaluate keratin, actin, tubulin, and calmodulin immunoreactivity in histologic sections of normal, hyperplastic, and neoplastic human prostate. Polyclonal as well as monoclonal keratin antibodies produced equivalent and intense staining of normal epithelium. The immunoreactivity of normal prostate with keratin antibodies was more pronounced than with antibodies to the other components of the cytoskeleton. Variation in staining for components of the cytoskeleton was minimal. The same findings applied to hyperplastic prostate. The immunoreactivity of prostate tumors with antibodies to these cytoskeletal proteins differed markedly from normal prostate. Prostatic carcinomas showed reduced keratin immunoreactivity with a panepithelial antibody, but unaltered or enhanced immunoreactivity with tubulin, actin, and calmodulin antibodies. Many tumors were unreactive with a monoclonal keratin antibody that was strongly reactive with tissues that contained cytokeratin 18 (45-kd) and which intensely stained normal and hyperplastic prostate. In addition, prostate carcinomas often yielded heterogeneous patterns of staining with actin, tubulin, and calmodulin antibodies in contrast to normal and hyperplastic prostate, which showed uniform staining. The results suggest that a disturbance in the organization of the cytoskeleton may accompany neoplastic transformation of human prostate.

Immunocytochemical evaluation of primary human esophageal carcinomas and their xenografts for keratin, beta-chorionic gonadotropin, placental lactogen, alpha-fetoprotein, carcinoembryonic antigen, and nonspecific cross-reacting antigen.

The unlabeled antibody peroxidase-antiperoxidase technique was used to examine esophageal neoplasms for the tumor markers beta-human chorionic gonadotropin, human placental lactogen (HPL), alpha-fetoprotein, carcinoembryonic antigen (CEA), and nonspecific cross-reacting antigen (NCA) before and after xenotransplantation to athymic nude mice. In addition, keratin was used as an epithelial cell marker. Immunoreactive beta-human chorionic gonadotropin was detected in four of seven primary tumors and in three of seven xenografts. Two of seven primary tumors contained HPL immunoreactive cells while four of seven tumor xenografts had HPL immunoreactivity. alpha-Fetoprotein was detected in two of seven primary tumors and in one of seven xenografts. NCA and CEA were detected in six of seven primary tumors and in all tumor xenografts. Five of seven primary neoplasms and six of seven tumor xenografts were found to contain both NCA and CEA, while one tumor and its xenografts displayed only NCA immunoreactivity. All seven primary carcinomas displayed keratin immunoreactivity, but only six of the seven xenograft tumors showed keratin positive cells. When a tumor marker was detected in a primary tumor, it was usually found in at least some of the xenografts arising from that tumor. However, marker loss did occur with repeated passage of tumors in some cases. On the other hand, markers were expressed in xenografts which were not present in the corresponding primary tumor. In three instances, HPL was detected in xenografts derived from HPL negative primary carcinomas. This was also true for CEA and NCA in one case. These results show that tumor markers are expressed to varying degrees by tumors growing as xenografts in nude mice. In primary tumors, HPL is associated with poorly differentiated squamous cell carcinomas and this marker was found to appear in HPL negative tumors as the tumor cells became less differentiated while growing as xenografts in nude mice.

Is the corneal posterior cell layer truly endothelial?

The posterior cell layer of the normal human cornea or "endothelium" was investigated by electron microscopy and immunocytochemistry. Ultrastructurally, the cells lacked the characteristic marker for endothelial cells (Weibel-Palade body). Immunoperoxidase studies demonstrated these cells to be negative for factor VIII antigen, but strongly positive for keratine, vimentin, S-100 protein, and neuron-specific enolase. The anterior epithelial cell layer showed identical immunoreactivity. These studies strongly suggest that the posterior cell layer of the cornea lacks ultrastructural and immunocytochemical markers of endothelial cells and both the anterior and posterior cell layers share similar cell markers. The authors propose that the posterior cell layer of the cornea should, therefore, not be misnamed as "endothelium."

Immunocytochemical evaluation of human esophageal neoplasms and preneoplastic lesions for beta-chorionic gonadotropin, placental lactogen, alpha-fetoprotein, carcinoembryonic antigen, and nonspecific cross-reacting antigen.

Human esophageal neoplasms were studied in comparison to normal, uninvolved, and preneoplastic human esophageal epithelium for the presence of human chorionic gonadotropin (HCG), human placental lactogen (HPL), alpha-fetoprotein (AFP), carcinoembryonic antigen (CEA), and nonspecific cross-reacting antigen (NCA) using the unlabeled antibody peroxidase-antiperoxidase technique. HCG immunoreactivity was identified in 10 of 33 squamous cell carcinomas (33%), in 1 of 6 adenocarcinomas (17%), and 1 of 6 preneoplastic esophageal lesions (17%); while 9 of 33 squamous cell carcinomas (33%) and 1 of 6 adenocarcinomas (17%) contained immunoreactive AFP. Immunoreactive HPL was detected in 6 of 33 squamous cell carcinomas (20%), but in none of the adenocarcinomas. Neither AFP nor HPL immunoreactivity was identified in the 6 hyperplastic lesions which were studied. When stained with an antiserum that was able to detect both CEA and NCA, 27 of 33 squamous cell tumors (82%) and 6 of 6 adenocarcinomas (100%) showed positive immunostaining reactions. Of these, 8 squamous cell carcinomas and 1 adenocarcinoma were subsequently shown to contain only NCA immunoreactivity, while 19 squamous cell carcinomas and 5 adenocarcinomas contained both NCA and CEA immunoreactivity. NCA immunoreactivity alone was identified in 3 of 6 preneoplastic lesions and NCA and CEA immunoreactivity in 1 of 6 preneoplastic lesions. None of the markers was detected in 8 specimens of normal esophageal epithelium which were studied as controls, nor in 6 specimens of uninvolved esophageal epithelium obtained from patients with esophageal cancer. Most tumors expressed 2 or 3 markers, and some tumors were identified which expressed up to 4 of the 5 markers investigated. Only 3 tumors failed to express any of the markers studied. No association was found between the degree of tumor differentiation and presence or absence of HCG immunoreactivity. However, HPL immunoreactivity was more common in poorly differentiated squamous cell carcinomas. In contrast, immunoreactive AFP was more common in well-differentiated squamous cell carcinomas than in other tumor types. Similarly, both CEA and NCA were more frequently expressed in well-differentiated squamous cell carcinomas, adenosquamous carcinomas, and adenocarcinomas than in less differentiated tumors. Our results suggest that HCG, HPL, AFP, CEA, and NCA are tumor-associated antigens in esophageal cancer. Therefore, they could be of value in screening tests for esophageal neoplasms and could be useful in subclassification of esophageal neoplasms.

Neuropathology of spiroplasma infection in the rat brain.

This study was designed to demonstrate the neuropathology of persistent spiroplasma infection in the rat brain. GT-48 spiroplasmas were inoculated intracranially into a series of suckling Sprague-Dawley rats. Their brains were evaluated at specific time intervals by microbiologic assay and by morphologic studies including histology, electron microscopy, and immunocytochemistry. The spiroplasmas were observed in the tissues by electron microscopy at peak infection 14 days after intracranial inoculation. At that time they were seen in vacuoles and neuronal processes within the neuropil as filamentous or bleb-like forms. A single tight spiral was identified that closely resembled the spiroplasma-like inclusions previously reported in Creutzfeldt-Jakob disease. The spiroplasmas were shown to spread rapidly throughout the brain tissues presumably by intraneuronal transport. In specimens examined at 25 days after intracranial inoculation and beyond, organisms were localized to gray matter without inflammatory response. The spiroplasmas could not be identified by electron microscopy in the rat brain tissue at late stages of infection. This study has shown an unusual adaptation of spiroplasma infection to the mammalian host brain tissues.

Immunocytochemical evaluation of human prostatic carcinomas for carcinoembryonic antigen, nonspecific cross-reacting antigen, beta-chorionic gonadotrophin, and prostate-specific antigen.

The unlabeled antibody peroxidase-antiperoxidase technique was used to examine human malignant prostatic tissue (primary tumors) for the presence of prostate-specific antigen (PSA), carcinoembryonic antigen (CEA), nonspecific cross-reacting antigen (NCA), and beta-chorionic gonadotrophin (HCG). The results were compared to those obtained with normal and hyperplastic prostate tissue (BPH). All specimens of neoplastic, hyperplastic, and normal prostate tissue showed immunostaining reactions for PSA. Immunostaining for PSA was relatively uniform among samples of normal and BPH tissue, but variations with respect to intensity of PSA immunostaining were noted among prostate tumors as well as between the neoplastic cells of individual tumors. Some areas of normal or hyperplastic prostatic epithelium within tumors showed stronger staining reactions for PSA than the tumor cells themselves. Using an antiserum which was able to detect both NCA and CEA, it was found that 16 of 38 tumors (42%) had positive immunostaining reactions. Of these, 15 were subsequently shown to contain only NCA immunoreactivity, and 1 tumor had both NCA and CEA immunoreactivity. NCA, but not CEA, immunoreactivity was identified in hyperplastic prostate tissue within tumor specimens and in BPH specimens. Neither antigen was detected in normal prostatic epithelium. Three of 38 tumors (8%) were found to contain neoplastic cells with HCG immunoreactivity. HCG immunoreactivity was not identified in BPH or normal prostatic tissue. Therefore, HCG and CEA immunoreactivity appear to be tumor-associated antigens in prostate cancer which are expressed with a low incidence. The results of the study identified prostate tumors with different patterns of immunocytochemical markers: 22 of 38 tumors (58%) contained only PSA immunoreactivity; 13 of 38 tumors (34%) contained PSA and NCA immunoreactivity; 2 of 38 tumors (5%) were positive for PSA, NCA, and HCG immunoreactivity; and 1 of 38 tumors (3%) contained PSA, NCA, HCG, and CEA immunoreactivity. Apart from PSA, which was present in all tumors, the markers studied here appeared to be more frequently expressed in well-differentiated tumors than in less-differentiated tumors. Our results suggest the possibility of subclassifying prostate tumors by means of immunocytochemistry.

Augmentation of immunity of line 10 hepatoma by BCG: comparison of different BCG preparations.

The line 10 hepatoma of strain 2 guinea pigs was used in order to evaluate the adjuvant efficacy of eight Bacillus Calmette-Guérin (BCG) suspensions which differed in strain, method of preservation, or dosage. Lyophilized Tice BCG was consistently the most effective preparation. Each strain had adjuvant activity in at least one experiment. The method of preservation of BCG (fresh-frozen vs. lyophilized) did not have a consistent, a predictable influence on adjuvanticity. A ten-fold increase in the dosage of BCG or of the whole vaccine was not supraoptimal.

Immunotherapeutic effectiveness of BCG inactivated by various modalities.

Factors responsible for the limited effectiveness of Bacillus Calmette-Guérin (BCG) immunotherapy are completely understood. One limitation is that although the effect is dose-related, high-dose administration increases the risk of BCG toxicity, possibly the result of disseminated BCG infection. In the present study, we compared the relative effectiveness of Tice lyophilized BCG which was inactivated by heat, sonication, irradiation, streptomycin, or isoniazid (INH). The model systems were the 13762A rat mammary adenocarcinoma and the line 10 guinea pig hepatoma. In the 13762A system, tumor were injected on day 7 with living or killed BCG preparations, or with Corynebacterium parvum as a positive control. Tumors were excised on day 20. Rats treated with surgery alone usually died within 40--50 days with extensive metastases to lymph nodes, lungs, and viscera. Guinea pig line 10 hepatoma was treated with vaccine containing irradiated tumor cells and BCG. In both the rat and guinea pig models, BCG inactivated by means of irradiation was effective as viable BCG and heat-killed BCG also had a strong effect. Streptomycin treatment diminished the efficacy of the BCG and sonication destroyed BCG effectiveness even though the organisms were not all killed. The INH treatment of tumor-bearing rats did not alter the benefits of single or repeated injections of high-dose viable BCG, irradiated BCG, or C parvum. We conclude that inacativation of BCG with heat, irradiation, or INH host treatment preserves but does not improve the immunotherapeutic benefits of BCG.

Immunotherapy of postoperative metastases of 13762A rat mammary adenocarcinoma: comparative effectiveness of BCG substrains and methods of preparation.

Six Bacillus Calmette-Guérin (BCG) substrains (Phipps, Pasteur, Connaught, Glaxo, RIV, and Tice) and two methods of preparation and storage (pellicle-lyophilized, dispersed-frozen and dispersed lyophilized) were directly compared in a standard immunotherapy protocol. Complete Freund's adjuvant and Bordetella pertussis vaccine were also studied. All experiments included direct comparisons with Corynebacterium parvum. The immunotherapy assay system was the 13762A rat mammary carcinoma in which immunity stimulant were given intratumorally on day 7, followed by excision of primary tumors on day 20. In this protocol, C parvum produced strong inhibition of established metastases and some cures. All of the BCG strains except Glaxo were effective in at least one experiment but none was as strong or as consistent in its effects as C parvum. No strain of BCG or method of preparation was clearly superior to the others. The B pertussis vaccine and complete Freund's adjuvant were ineffective.

Treatment of cancer using Corynebacterium parvum: similarity of two preparations in four animal tumor models.

The tumor inhibitory properties of Corynebacterium parvum obtained from Burroughs Wellcome (CP-BW) or from Institut Merieux (CP-IM) were compared in four animal tumor models: the CaD2 mouse mammary carcinoma treated by intravenous (I.V.) or intratumoral (I.T.) injection of C. parvum; 13762A rat mammary adenocarcinoma treated by I.T. injection of C. parvum either alone or combined with excision of the primary tumor; LSTRA murine leukemia and line 10 cavian hepatoma, each treated with vaccines containing irradiated tumor cells and C. parvum. Both preparations were active against each tumor. In most comparisons the potency of the two materials was not different, but in a few cases the CP-BW was effective at a lower dose than was the CP-IM. These results demonstrate the versatility of C. parvum for use in a variety of immunotherapy procedures and show that the potencies of the two major types of C. parvum are very similar.

The effects of neonatal androgenization on mammary gland mitotic rate and susceptibility of carcinogen-induced mammary dysplastigenesis and tumorigenesis in LEW/Mai rats.

The influence of neonatal testosterone propionate treatment (androgenization) on mammary gland mitotic rate (MR) and susceptibility to 7,12-dimethylbenz(alpha)anthracene (DMBA) carcinogenesis was studied in female LEW/Mai rats. Mammary gland MR in androgenized rats was significantly lower than MR in normal rats at all ages studied. Treatment of androgenized rats with DMBA resulted in a significant increase in mammary gland MR in comparison with untreated androgenized rats. MR in DMBA-treated androgenized rats was similar to MR in DMBA-treated normal rats at most intervals after the introduction of the carcinogen. Although mammary epithelial MR in androgenized rats was significantly lower than that of normal rats of comparable age, no evidence of a decrease in susceptibility of mammary epithelium in androgenized rats to DMBA carcinogenesis was found. Instead, androgenized rats had a higher incidence of DMBA-induced mammary dysplasias, with no change in their morphologic or histologic features, than did normal rats; and there was no change in the incidence, latency, or histopathologic appearance of DMBA-induced mammary tumors in androgenized versus normal rats.

The relationship of terminal duct hyperplasia to mammary carcinoma in 7,12-dimethylbenz(alpha)anthracene-treated LEW/Mai rats.

The evolution of dysplasias and carcinomas in the inguinal mammary glands of LEW/Mai rats given 7,12-dimethylbenz(alpha)anthracene (DMBA) by gastric gavage was studied with the use of stained whole mounts. Two major dysplasias, hyperplastic terminal end buds (HEBs) and hyperplastic alveolar nodules (HANs), developed prior to mammary carcinomas. HEBs were present in the mammary glands of approximately 70% of the rats within 1 week following DMBA. The percentage of rats with these lesions and the incidence of HEBs in the mammary gland decreased prior to the appearance of palpable and microscopic tumors. During the time when tumors first became evident (40-80 days), the percentage of rats with HEBs (11%) paralleled the percentage of rats with mammary tumors (12%). The initial percentage of rats with HEBs ( approximately 70%) paralleled the final tumor incidence (71%) observed in DMBA-treated rats that were allowed to live until they developed tumors. The histologic features of HEBs resembled those of the carcinomas, and HEBs were present in the immediate vicinity of some of the microscopic and palpable tumors. With only one exception, the location of microscopic tumors in the mammary gland was consistent with their derivation from small terminal ducts. These data are compatible with a developmental relationship between HEBs and mammary carcinoma. HANs, on the other hand, developed relatively late (ie, 30 days) following DMBA administration and became more numerous with the passage of time. Over the period of time when mammary carcinomas first became evident, the percentage of rats with HANs (73%) was inconsistent with a developmental relationship between HANs and mammary carcinoma. This conclusion was supported by the absence of HANs in the vicinity of microscopic tumors, by the dissimilarity between the histologic features of HANs and mammary carcinomas, and by their absence from the mammary gland during the time when at least some of the mammary tumors must have arisen. The results implicate terminal duct hyperplasia in the histopathogenesis of rat mammary carcinomas.

Comparative antitumor effects of Corynebacterium parvum, Bordetella pertussis, Bacillus Calmette-Guérin, and levamisole alone or in combination with cyclophosphamide in the CaD2 murine mammary adenocarcinoma system.

The antitumor efficacy of various immune stimulants [Corynebacterium parvum, Bordetella pertussis, and Bacillus Calmette-Guérin (BCG)] and levamisole alone or in conjunction with cyclophosphamide (CY) was studied in the CaD2 mammary adenocarcinoma system using schedules developed previously with C. parvum and CY. Weekly systemic treatment with C. parvum, B. pertussis, or BCG was effective in controlling tumor growth and had equivalent antitumor effects, but weekly treatment (or a single treatment) with levamisole was ineffective. Weekly treatment with B. pertussis was better than treatment given only once, but repeated treatment with C. parvum or BCG was not more effective than a single treatment with these agents. When administered as a single systemic injection, C. parvum was superior to B. pertussis in controlling tumor growth, but a single systemic injection of BCG was as effective as C. parvum. Systemic administration of immune stimulants had variable effects on survival, which were sometimes not correlated with effects on tumor size. Combined treatment with BCG and CY was significantly more effective than CY treatment alone in controlling tumor growth in most trials, as was combination treatment with C. parvum and CY. Combination treatment with B. pertussis and CY was not better in prolonging survival than CY alone. Levamisole rarely improved the antitumor effect of CY chemotherapy and had no effect on survival compared to the effect of CY alone.

Active specific immunotherapy of murine leukemia. I. Irradiated tumor cell vaccine.

This report describes a new animal model for active, specific, systemic immunotherapy of leukemia. Mice were challenged (day 0) with 10(3) viable LSTRA cells and treated later with a vaccine containing 10(7) irradiated LSTRA cells. Four intraperitoneal or intravenous vaccine injections, on days 2, 9, 16 and 23, caused significant prolongation of survival; single injections by the same routes had inconsistent effects. Vaccine was not effective against a higher initial challenge dose, when given later in the course of tumor growth, when given intradermally or when used in weanling mice. The therapeutic effect was specific since a vaccine composed of an antigenically unrelated leukemia was ineffective. These results suggest that LSTRA is suitable for studying active, systemic, specific immunotherapy.

Conditions for effective Bacillus Calmette-Guérin immunotherapy of postsurgical metastases of 13762A rat mammary adenocarcinoma.

We evaluated critical variables in Bacillus Calmette-Guérin (BCG) immunotherapy of residual 13762A rat mammary adenocarcinoma. BCG was given intratumorally on Day 7 of tumor growth and followed by primary tumor excisions on Day 20. Untreated animals died on about Day 40 with axillary nodal and pulmonary parenchymal metastases. BCG-treated animals experienced prolonged survival, and some were cured. The highest dose (5.0 X 10(7) colony-forming units) of BCG was more effective than the lowest (0.5 X 10(7) colony-forming units), but 1,500 micrograms Corynebacterium parvum were more effective than even the highest BCG dose. Previous sensitization to BCG did not improve the effects of BCG treatment. BCG treatment was effective when given on Day 7 and sometimes as late as Day 12 or 17, but C. parvum was ineffective if given after Day 7. Repeated injections of BCG or C. parvum were not more effective than single injections were. Rats cured of residual 13762A tumor by BCG treatment were strongly and specifically immune to rechallenge. We concluded that a high dose (5.0 X 10(7) colony-forming units) of BCG given early (7 days) was the most effective presurgical treatment of 13762A metastases. Repeated injections or host presensitization to BCG did not improve the benefits.

Effects of dose and schedule of immune stimulant on efficacy of combination Corynebacterium parvum-cyclophosphamide treatment for a murine mammary adenocarcinoma.

Certain variables which might influence the outcome of combining cytotoxic drug and immune stimulant therapy were studied to optimize the effectiveness of Corynebacterium parvum combined with cyclophosphamide (CY) as treatment for a murine mammary adenocarcinoma (CaD2). Optimal effects of combined C. parvum-CY treatment in the CaD2 system were obtained when 443 to 1400 microgram of this immune stimulant per mouse were injected 2 to 3 days after CY chemotherapy and when combination treatment was continued on a weekly basis. The most critical factors contributing to the effectiveness of combination treatment in this system were the dose of C. parvum and the treatment frequency. The interval between chemotherapy and immune stimulant therapy was less critical to the outcome of combination treatment. Combination treatment given once or weekly significantly decreased tumor size in comparison to single or weekly CY treatment. A single treatment with CY and C. parvum significantly improved the survival over mice given a single CY treatment, but weekly CY and C. parvum treatment did not increase the survival over mice, given weekly chemotherapy.