Guidelines for the welfare and use of animals in cancer research.

Animal experiments remain essential to understand the fundamental mechanisms underpinning malignancy and to discover improved methods to prevent, diagnose and treat cancer. Excellent standards of animal care are fully consistent with the conduct of high quality cancer research. Here we provide updated guidelines on the welfare and use of animals in cancer research. All experiments should incorporate the 3Rs: replacement, reduction and refinement. Focusing on animal welfare, we present recommendations on all aspects of cancer research, including: study design, statistics and pilot studies; choice of tumour models (e.g., genetically engineered, orthotopic and metastatic); therapy (including drugs and radiation); imaging (covering techniques, anaesthesia and restraint); humane endpoints (including tumour burden and site); and publication of best practice.

Therapeutic index: a vital component in selection of anticancer agents for clinical trial.

In an attempt to identify anticancer agents more active in the major groups of solid cancers, the National Cancer Institute has replaced the primary murine tumor systems with an in vitro system using human cancer cell lines. A broader approach is being followed in Europe, where considerable reliance will still be placed on data from animal tumor systems. We argue the case for broader preclinical evaluation using animal model systems that are resistant to standard anticancer agents and thus reflect the clinical disease. Selection of compounds for clinical trials should be based on a therapeutic index that will give a more realistic representation of the margin between antitumor activity and toxicity to normal cells. Furthermore, preclinical evaluation can establish understanding of the factors responsible for treatment efficacy, including pharmacokinetics, metabolism, and drug bioavailability.

Some biological characteristics of transplantable lines of mouse adenocarcinomas of the colon.

During a 5-year period colon tumors induced by 1,2-dimethylhydrazine dihydrochloride treatment in NMRI mice were transplanted sc into isologous mice to obtain transplantable tumor lines that might serve as experimental models for human colon cancer. The resultant 17 serially transplantable tumor lines were adenocarcinomas varying in degree of differentiation and mucin production. Two of the lines passaged for 5 years and used in chemotherapy studies have shown histologic progression toward dedifferentiation with concomitant acceleration of growth rate.

A critical appraisal of the predictive value of in vitro chemosensitivity assays.

The ultimate value of a screening program based on the response of cell lines in vitro will depend on the demonstration of a strong correlation between in vitro and in vivo responses to cytotoxic drugs. However, marked discrepancies in the predictive value of in vitro chemosensitivity assays have been described, suggesting that factors other than the inherent chemosensitivity of tumor cells significantly influence the outcome of chemotherapy in vivo. These factors include the influence of drug pharmacokinetics and metabolism, together with numerous problems associated with the biology of solid tumors in vivo (e.g., drug penetration barriers, proliferation gradients, and microenvironmental conditions). These additional factors may be highly significant in explaining the site-dependent nature of the responses of solid tumors to cytotoxic drugs, the poor prediction of responses in experimental tumor models, and the differences in the responses of multicellular spheroids and monolayers. These discrepancies suggest that the selection of compounds for phase II clinical trials on the basis of disease-specific activity in vitro may be premature.

Transplantation of adenocarcinomas of the colon in mice.

1, 2-Dimethylhydrazine treatment induced multiple colon tumors in 100% of NMRI mice. Many of these tumors were transplanted and yielded five serially transplantable tumor lines. These subcutaneously transplanted neoplasms were all adenocarcinomas varying in degree of differentiation and mucin production. No evidence of dedifferentiation or change in growth rate has been seen in up to six transplant generations. These tumor lines appeared to provide relatively stable, well-differentiated models for colorectal cancer.

Reduction of tumor blood flow by flavone acetic acid: a possible component of therapy.

Flavone acetic acid (FAA) is active against normally refractory murine sc tumors. Clinical studies are disappointing despite achievement of plasma profiles associated with the antitumor murine activity in man. To clarify the mechanism of action, we have followed histologic changes, tumor blood volume, and drug concentrations in a well-differentiated, slow-growing cystic adenocarcinoma in mice. FAA causes massive tumor necrosis beginning 2 hours after treatment. Tumor plasma volumes are reduced by 2 hours after treatment and tumor blood vessels are shutdown, which suggests that tumor vasculature plays a role in the dramatic response of sc tumors in pure-strain male NMRI mice.

Characterization of a transplantable adenocarcinoma of the mouse colon producing cachexia in recipient animals.

MAC16 is a chemically induced, transplantable adenocarcinoma of the colon passaged in inbred NMRI mice. At small tumor burdens (less than 1% of the host weight), weight loss was observed without a reduction in food intake. As the tumor mass increased, weight loss also increased and reached 33% of host body weight in females and 20% in males when compared with the weight of age-matched controls. The reduction in host body weight was directly proportional to the tumor size and was reversible when the tumor was excised. There was a preferential loss of body fat in tumor-bearing animals with an increase in the plasma level of free fatty acids, although there was a minimal elevation of ketone bodies. Tumor growth was accompanied by progressive hypoglycemia and a reduction in the plasma insulin levels. The decrease in plasma insulin may have contributed to the catabolic effects of progressive tumor growth.

Characterization of the major metabolites of flavone acetic acid and comparison of their disposition in humans and mice.

Flavone acetic acid represents a novel chemical structure currently undergoing clinical investigation. Broad spectrum activity has been observed in preclinical animal screens, but at doses close to toxic in mice. Phase I clinical trials have established that equivalent plasma drug levels can be achieved in humans, but to date Phase II trials have not demonstrated significant activity in a range of tumor types. Little is known about the drug's biotransformation, although metabolites have been implicated in proposed mechanisms of action. In this paper, we have purified the two major human metabolites present in urine (also the only two metabolites detected in plasma) and characterized their structure, chemical properties, activity, and pharmacokinetics. Metabolite 1 (M1) was a glucuronide conjugated to the 8-acetic acid grouping (Mr 456), was chemically labile, and showed a strong tendency to undergo chemical rearrangement at mildly alkaline pH. Metabolite 2 (M2) was also a glucuronide (Mr 456) but appeared to be an unusual isomer of M1. Both were noncytotoxic. In patients, biotransformation represented the predominant mechanism of drug clearance with as much as 80% of a low dose (0.5 g/m2) recovered in urine as M1 and M2 after only 6 h. At high dose (4.8 to 8.6 g/m2, 1- to 6-h infusion) the appearance of peak concentrations of metabolites in plasma and urine was delayed, apparently due to saturation of glucuronidation pathways. This resulted in an overall reduction in drug clearance by 3- to 4-fold. Mice cleared flavone acetic acid much more slowly than patients (289 ml/h/m2 after 600 mg/m2 i.p. versus 2.3 liters/h/m2 after 4.8 g/m2-1-h i.v. infusion) without producing M1 or M2. A different metabolite, exhibiting characteristics of a conjugate, was detected at low concentrations in plasma, tissues, and tumor. Extensive metabolism to inactive products followed by their rapid clearance may contribute to the lack of activity so far seen in humans.

Angiogenesis in the hollow fiber tumor model influences drug delivery to tumor cells: implications for anticancer drug screening programs.

The National Cancer Institute uses the hollow fiber assay as part of its screening program for anticancer drug discovery. Angiogenesis to hollow fibers implanted s.c. has not been reported, thereby raising concerns about the efficiency of drug delivery and its subsequent effects on chemosensitivity. By extending postimplantation times beyond the 6-day period presently used, extensive vascular networks develop, resulting in both increased delivery and chemosensitivity to doxorubicin. This study suggests that present protocols used to evaluate compounds may produce false negative results, and additional studies to determine the predictive value of the assay are required.

Pharmacokinetics of PK1 and doxorubicin in experimental colon tumor models with differing responses to PK1.

PK1 is a synthetic N-(2-hydroxypropyl)methacrylamide copolymer-doxorubicin (dox) conjugate currently undergoing Phase II evaluation in the United Kingdom. We have studied the activity of PK1 in three murine colon tumor models that differ in terms of morphology and vascularization in an attempt to determine which factors are most important in the tumor response to PK1. Vascular permeability was evaluated with Evans Blue, and pharmacokinetic studies in MAC15A and MAC26 used high-performance liquid chromatography to monitor both PK1 uptake and dox release in the tumors. Cathepsin B activity was assessed using a specific substrate. PK1 (40 mg x kg(-1) dox equivalent) was significantly more effective than dox alone (10 mg x kg(-1)) was against MAC15A tumors, which possess enhanced perfusion and retention, but not against MAC26 tumors, although MAC15A was also responsive to PK1 when grown as avascular micrometastatic deposits in the lung. Pharmacokinetic studies showed similar levels of PK1 in both tumors. Peak tumor levels of released dox were 7-fold greater in the responsive MAC15A tumor (53 microg x ml(-1)) compared with the less responsive MAC26 tumor (7.7 microg x ml(-1)) and more than 18-fold greater in MAC15A than when free dox was given. These differences in response correlated also with an increased lysosomal activity of cathepsin B. Calculated AUCs for intratumoral dox released were 431 microg x h x g(-1) and 775 microg x h x g(-1) for MAC15A and MAC26, respectively. These AUCs are 4-fold and 7-fold higher, respectively, than when dox is given alone. This study has shown that activity and the pharmacokinetics of PK1 and released dox are dependent on both the vascular properties and enzyme content of the tumors. These studies are likely to have clinical implications as aggressive tumors are known to have increased protease activity.

Inhibition of telomerase activity by cisplatin in human testicular cancer cells.

Telomerase, a ribonucleoprotein, elongates and/or maintains telomeres by adding TTAGGG tandem repeat sequences using the RNA component of the enzyme as a template. Enzyme activity appears to be associated with cell immortalisation and malignant progression as telomerase activity has been found in the majority of human tumours, but not in most somatic cells or tissues. Telomerase inhibition has, therefore, been proposed as a novel and potentially selective target for therapeutic intervention. Since telomeric tandem repeats as well as the human telomerase RNA component (hTR) and its gene are guanosine-rich, we examined whether the sequence specific, G-Pt-G, cross-linking agent cisplatin is capable of inhibiting telomerase activity. The TRAP assay was used to measure telomerase activity in cisplatin treated cell extracts and RT-PCR strategies used to examine hTR expression after drug exposure. Cisplatin reduced telomerase activity in a specific and concentration-dependent manner in human testicular tumour cells, whilst doxorubicin, bleomycin, methotrexate, melphalan and transplatin had no effect. It is proposed that telomerase inhibition might be a component of the efficacy of cisplatin in the treatment of testicular cancer.

DT-diaphorase activity correlates with sensitivity to the indoloquinone EO9 in mouse and human colon carcinomas.

The indoloquinone EO9 exhibits promising in vitro and in vivo antitumour activity. EO9 is metabolised to DNA damaging species by DT-diaphorase in vitro. In the present study DT-diaphorase specific activity was 16 fold higher in the mouse adenocarcinoma MAC 16, a tumour which is quite responsive to EO9 in vivo, compared with levels in the more resistant mouse adenocarcinoma MAC 26. This order of responsiveness is the reverse of that seen with the most active of the clinically used agents in these tumours [chloroethylnitrosoureas and 5-fluorouracil (5-FU)]. In addition, when the in vitro sensitivity of two human colon carcinoma cell lines was compared, EO9 was 15-30 fold more active in the DT-diaphorase rich HT29 line than in the enzyme-deficient BE cell line counterpart. These results are consistent with the hypothesis that DT-diaphorase expression may be a major determinant of the sensitivity of tumours to EO9. This should be considered in the clinical development of the drug.

Cell growth inhibition, G2M cell cycle arrest and apoptosis induced by the imidazoacridinone C1311 in human tumour cell lines.

The cytotoxic activity of the imidazoacridinone C1311 was assessed on two ovarian cancer cell lines (A2780, OAW42) and one osteogenic sarcoma cell line (U2-OS) and their sublines (A2780Cp8, OAW42-MER and U2-OS-R) with experimentally induced resistance to cisplatin. A 1-h exposure to C1311 significantly inhibited the growth of all cell lines, with IC50 values ranging from 0.50 +/-0.11 to 4.10+/-0.36 microM. No or only partial cross-resistance was found between C1311 and cisplatin in the different cell lines. Treatment with equitoxic (IC50) C1311 concentrations consistently induced accumulation of cells in the G2M phase. The cyclin B1-associated p34(cdc2) kinase activity in cells arrested in G2M was superimposable to that of control cells in the OAW42-MER and U2-OS cell lines, whereas a reduction of cdc2 catalytic activity was observed in OAW42 and U2-OS-R cells. Exposure to C1311 (IC50) induced apoptosis in the U2-OS and U2-OS-R cell lines, whereas in the OAW42 and OAW42-MER cell lines there was a negligible percentage of apoptotic cells. In U2-OS, U2-OS-R and OAW42 cells, C1311 induced an increase in p53 expression and an increase in p21waf1 protein, whereas p53 failed to transactivate p21waf1 in OAW42-MER cells. An almost complete abrogation of bcl-2 was observed in U2-OS-R cells in correspondence with the peak of apoptosis induction. Our results indicate that C1311 is active against human ovarian cancer and osteogenic sarcoma cells and is not cross-resistant with CDDP. Moreover, C1311 blocks cells in the G2M phase and induces apoptosis in a small percentage of osteogenic sarcoma cells.

Potentiation of EO9 anti-tumour activity by hydralazine.

EO9[3-hydroxy-5-aziridinyl-1-methyl-2(1H-indole-4,7-dione)prop-bet a-en-alpha-ol] has been selected for phase I evaluation in Europe. Activity has been seen previously in a highly refractory, necrotic mouse adenocarcinoma (MAC 16) but EO9 is shown here to be inactive against early tumours (MAC 15A and MAC 13) and a well vascularised, well-differentiated established adenocarcinoma (MAC 26). EO9 becomes active against MAC 26 tumours when hydralazine (10 mg/kg) is administered 1 min after EO9. Co-administration of hydralazine decreases EO9 plasma clearance and increases plasma area under the curve values (0.053 to 0.115 micrograms h/ml). These pharmacokinetic changes are accompanied by anti-tumour activity but no increase in bone marrow toxicity so this therapeutic gain may be due, at least in part, to microenvironmental changes resulting from hydralazine induced tumour vascular shutdown.

EO9: a novel bioreductive alkylating indoloquinone with preferential solid tumour activity and lack of bone marrow toxicity in preclinical models.

EO9 is a novel and fully synthetic bioreductive alkylating indoloquinone. Although structurally-related to mitomycin C, EO9 exhibits a distinct preclinical antitumour profile and there are also differences in its biochemical activation. In this study, EO9 was found to demonstrate preferential cytotoxicity against solid tumours in vitro as compared to leukaemia cell lines both in the Corbett two-tumour assay and in the disease-oriented human tumour cell line panel of the U.S. National Cancer Institute. In the latter system activity was particularly apparent in colon, melanoma and central nervous system lines, together with some renal and non-small cell lung lines. Preferential cytotoxicity towards hypoxic versus aerobic EMT6 mouse mammary tumour cells was observed. In vivo, EO9 was inactive against the P388 murine leukaemia, while exerting significant antiproliferative effects against several murine and human solid tumours, including the generally resistant MAC mouse colon tumours and gastric, ovarian and breast xenografts. These results confirmed in vitro observations of preferential solid tumour activity. In animal toxicology studies, EO9 induced vascular congestion in the gastrointestinal tract, but no significant bone marrow toxicity. The LD10 value of EO9 after a single intravenous injection into mice was 9 mg/kg (27 mg/m2). A dose of one-tenth of the mouse equivalent LD10 (2.7 mg/m2), the recommended starting dose for clinical phase I studies, was found to be safe in rats. Considering its distinct mechanism of bioactivation as compared to mitomycin C, its preferential solid tumour activity, its excellent activity against hypoxic cells, and lack of significant bone marrow toxicity in animals studies, EO9 has been selected for clinical evaluation within the framework of the EORTC.

Nucleoside analogs. 14. The synthesis of antitumor activity in mice of molecular combinations of 5-fluorouracil and N-(2-Chloroethyl)-N-nitrosourea moieties separated by a three-carbon chain.

5-fluorouracil (5-FU) seco-nucleosdies having as the "sugar" moiety a two-carbon (C2) side chain carrying a N-(2-chloroethyl)-N-nitrosourea group were designed as molecular combinations of antimetabolite and alkylating agent, but hydrolytic release of free 5-FU was not fast enough for significant contribution to the high activity they showed against colon and breast tumors in mice. In the present study of the synthesis of the more reactive C3 seco-nucleosides, it emerged that, of various groups attached to the aldehydic center in the precursor phthalimides, only the alkoxy/uracil-1-yl type was conveniently obtained by the standard method. The methylthio/uracil-1-yl analog required relatively large amounts of reagent methanethiol, and exploration of alternatives involving alpha-chlorination of alkyl methyl sulfide or Pummerer rearrangement of its S-oxide, or successive hydrolysis and methylation of isothiouronium bromide, gave disappointing yields. For successful preparation of the alkoxy/uracil-3-yl compounds, the route used for C2 homologs required considerable experimental modification. In addition to these O,N- and S,N-acetals, some N,N-acetals bearing two 5-FU residues were prepared. The new drugs have been tested against a panel of experimental tumors in mice. Although it is evident from a parallel study that even these C3 seco-nucleosides release free 5-FU too slowly in vivo, several of them have shown impressive anticancer activity. Reviewing their performance in comparison with earlier molecular combinations, a short list of seven [B.4152 (6), B.4015 (5), B.4030 (10), B.3999 (4), B.3995 (2), B.4083 (3), and B.3996 (the N 3-substituted analog of 1)] should be investigated further. This is particularly appropriate in light of the present understanding of the mode of action of chloroethylating agents. Following a prolonged period of clinical impatience with nitrosoureas because of limited selectivity action, a new era is confidently anticipated as these powerful drugs are increasingly studied in combination with O6-benzylguanine and other more efficient inhibitors of repair enzymes like O6-alkylguanine-DNA-alkyltransferase now being developed.

Synthesis, mechanism of action, and biological evaluation of mitosenes.

Mitosenes of both the pyrrolo- and pyrido[1,2-a]indole type have been prepared via modification of these heterotricyclic compounds. Several mitosenes have been studied for their reactions with nucleophiles under reductive conditions. The results of these experiments show that the biological activity of mitosenes is based on the mechanism of bioreductive activation. When both leaving groups at C-1 and C-10 in the mitosene are the same, the nucleophile preferably adds to C-10 under reductive conditions. All mitosenes were studied for their biological activities in vitro against L1210, WiDr, and A204. On the basis of these results a selection of three mitosenes was made for a more detailed biological evaluation. Several tumor model systems were used, viz. P388, human tumor xenografts, MAC 13, and MAC 16. The results of these studies show that mitosenes have a more limited range of activities than mitomycin C. Surprisingly, the in vivo activities of mitosene diol 8b and mitosene diacetate 10b against the gastric human tumor xenograft GXF 97 were very high and comparable with that of mitomycin C.

Genotyping of NAD(P)H:quinone oxidoreductase (NQO1) in a panel of human tumor xenografts: relationship between genotype status, NQO1 activity and the response of xenografts to Mitomycin C chemotherapy in vivo(1).

Pharmacogenetic analysis of polymorphisms in drug metabolizing enzymes is currently generating considerable interest as a means of individualizing patient therapy. Recent studies have suggested that patients that are homozygous for a polymorphic variant (a C to T transition at position 609 of the cDNA sequence) of the enzyme NAD(P)H:quinone oxidoreductase (NQO1) may be resistant to Mitomycin C (MMC). Genotyping of a panel of 54 human tumor xenografts by polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP), classified tumors as wild type (40/54), heterozygotes (11/54), and homozygous mutants (3/54). Previously, 37 of these tumors had been characterized in terms of their response to MMC in vivo, and in this study, a further nine tumor xenografts have been characterized in terms of their response to MMC. No correlation could be found between the NQO1 polymorphic status of xenografts and their response to MMC in vivo. In terms of genotype/phenotype relationships, NQO1 activity in tumors genotyped as wild type, heterozygotes, and homozygous mutants were 311.1 +/- 421.9 (N = 40), 76.9 +/- 109.5 (N = 11), and 0.2 +/- 0.17 (N = 3) nmol/min/mg, respectively. Genotyping of patients may provide a useful initial step in identifying patients who are unlikely to benefit from quinone-based chemotherapy. In the case of MMC, however, the work presented here demonstrates that genotyping of individuals with respect to NQO1 is unlikely to be beneficial in terms of predicting tumor responses to MMC.

Microsatellite instability, chemosensitivity and mutant frequency in a series of 1,2-dimethylhydrazine induced murine colon adenocarcinoma models.

Loss of DNA mismatch repair has been described in a number of tumour types such as colorectal adenocarcinoma and leads to microsatellite instability. This may have clinical relevance due to mismatch repair defects altering chemosensitivity towards certain classes of anti-tumour agent. This study has examined microsatellite instability of eight murine colon adenocarcinoma tumour models induced by 1,2-dimethylhydrazine. Four microsatellite regions were examined suggesting that four of the tumour models exhibit a low level of microsatellite instability. Loss of heterozygosity was found in 5/8 tumours, suggesting that allelic loss may be a relatively common step in the carcinogenesis of these tumour models. Three of the allelic losses involved the D11MIT4 locus which is situated very close to the p53 tumour suppressor locus. Four tumour models are routinely cultured in vitro and these were used to examine whether there was any association between microsatellite instability, mutant frequency and chemo-sensitivity of these tumour models, comparing them with four human adenocarcinoma cell lines of known mismatch repair status. Two cell lines (MAC26 and MAC16) were found to be more chemoresistant towards cisplatin but not 6-thioguanine. No association was found between microsatellite instability and chemosensitivity for either the human or mouse cell lines.

Activity of N-[N'-(2-chloroethyl)-N'-nitrosocarbamoyl]-alanine and derivatives against transplantable adenocarcinomata of the mouse colon (MAC).

The anti-neoplastic activity of N-[N'-(2-chloroethyl)-N-nitrosocarbamoyl(CNC)]-alanine, CNC-alanyl-alanine, CNC-alanine-methylamide and CNC-glycine-methylamide was examined in murine transplantable colon tumours (MAC). The methylamide derivatives were highly active against a solid adenocarcinoma (MAC 13) and an ascitic adenocarcinoma (MAC 15 A). CNC-alanyl-alanine was also highly active against MAC 15 A. Responses of the three latter agents against the ascitic tumour were better than for any previously tested drugs including the nitrosoureas but their eventual usefulness cannot be determined without further toxicological studies.