Mitomycin C (MMC) is a clinically used anticancer drug that is reduced to cytotoxic metabolites by cellular reductases via a process known as bioreductive drug activation. The identification of key enzymes responsible for drug activation has been investigated extensively with the ultimate aim of tailoring drug administration to patients whose tumors possess the biochemical machinery required for drug activation. In the case of MMC, considerable interest has been centered upon the enzyme DT-diaphorase (DTD) although conflicting reports of good and poor correlations between enzyme activity and response in vitro and in vivo have been published. The principle aim of this study was to provide a definitive answer to the question of whether tumor response to MMC could be predicted on the basis of DTD activity in a large panel of human tumor xenografts. DTD levels were measured in 45 human tumor xenografts that had been characterized previously in terms of their sensitivity to MMC in vitro and in vivo (the in vivoresponse profile to MMC was taken from work published previously). A poor correlation between DTD activity and antitumor activity in vitro as well as in vivo was obtained. This study also assessed the predictive value of an alternative approach based upon the ability of tumor homogenates to metabolize MMC. This approach is based on the premise that the overall rate of MMC metabolism may provide a better indicator of response than single enzyme measurements. MMC metabolism was evaluated in tumor homogenates(clarified by centrifugation at 1000 × gfor 1 min) by measuring the disappearance of the parent compound by HPLC. In responsive [T/C <10% (T/C defined as the relative size of treated and control tumors)] and resistant (T/C >50%) tumors, the mean half life of MMC was 75 ± 48.3 and 280 ± 129.6 min, respectively. The difference between the two groups was statistically significant (P < 0.005). In conclusion, these results unequivocally demonstrate that response to MMC in vivo cannot be predicted on the basis of DTD activity. Measurement of MMC metabolism by tumor homogenates on the other hand may provide a better indicator of tumor response, and further studies are required to determine whether this approach has real clinical potential in terms of individualizing patient chemotherapy.

The ability to tailor chemotherapy to individual patients has been a major objective in cancer therapy for many years, but despite extensive studies using a variety of chemosensitivity tests (1, 2), no predictive assay is in widespread use in the clinic today. Individualizing patient chemotherapy remains a major issue in current cancer therapy, and attention is now being paid to the identification and evaluation of various markers of tumor response at the molecular level (3). Within the field of bioreductive drug development, the ability to select patients who will benefit from this treatment was recognized at an early stage and forms one of the major objectives of a concept known as “enzyme-directed bioreductive drug development” (4, 5). The identification of drugs that are activated by specific reductases and the selection of drugs for individual patients based upon the activity of specific reductase enzymes within the tumor represent the principle objectives of this concept. The fundamental requirements, therefore, for the successful clinical application of this concept are the development of drugs where key enzymes involved in drug activation are known, coupled with evidence of a strong correlation between the response of tumors(in vitro but particularly in vivo) and enzyme activity.

MMC3is a clinically active antineoplastic agent used to treat a variety of tumors and is regarded as the prototypical bioreductive drug (6, 7). Its mechanism of action is complex involving several reductase enzymes, some of which have yet to be identified(8, 9, 10, 11, 12, 13). It is generally believed that the enzyme DTD[NAD(P)H:quinone oxidoreductase; EC 1.6.99.2] is the major enzyme responsible for bioreductive activation of MMC under normal oxygenated conditions, whereas under hypoxic conditions, other enzymes (such as cytochrome P-450 reductase) assume a prominent role(14, 15, 16). Attempts to elucidate the fine details of MMC activation in terms of identifying enzymes that determine cellular response have, however, generated conflicting and controversial results. This is particularly true in the case of DTD, where MMC has been shown to be a substrate for DTD at acidic pH, whereas under normal physiological pH conditions, MMC is not only a poor substrate but is also an inhibitor of enzyme activity (17, 18). Attempts to clarify the role of DTD in the activation of MMC using a variety of experimental models (e.g., transfected cells, MMC-resistant cell lines, and the use of dicumarol as an inhibitor of DTD) have not been successful in that conflicting reports for and against a major role for DTD in MMC activation have been published(19, 20, 21, 22, 23). In terms of predicting responses to MMC in vitro based upon DTD activity, reports of good correlations(24) conflict with reports of poor correlations(25). Only limited studies have been conducted in vivo, although a good correlation between DTD activity and antitumor activity has been reported in a panel of eight non-small cell lung cancer and small cell lung cancer xenografts (26). There are, however, reports of poor correlations between DTD activity and MMC activity in vivo(27); therefore, the issue of whether tumor response to MMC in vivo can be predicted on the basis of DTD activity remains confused.

Recent studies by Cummings et al.(28, 29) have suggested that the concept of enzyme-directed bioreductive drug development may need to be remodeled, on the basis that the mechanism of action of compounds such as MMC and the structurally related indoloquinone EO9 are too complex to allow for accurate predictions of response based upon the activity of a single enzyme. An alternative approach based upon the ability of homogenates of tumor tissue to metabolize bioreductive drugs has been proposed by Cummings et al.(28, 29), and good correlations between response and the rate of reduction of EO9 have been reported in a limited number of tumors (29). To assess the relative merits of predicting tumor response on the basis of DTD levels or metabolism of MMC by tumor homogenates, this study has compared both end points in a large panel of human tumor xenografts. These xenografts have been established within the European Organization for Research and Treatment of Cancer as an in vivo-based screening facility to identify novel anticancer drugs and have been extensively characterized in terms of their response to several anticancer drugs including MMC(30). In addition, in vitro chemosensitivity studies using the soft agar clonogenic assay have been conducted with the aim of assessing the relationship between DTD activity and chemosensitivity in vitro.

Drugs.

MMC was purchased from Medac (Hamburg, Germany) and dissolved in normal saline. 5-Fluorouracil was obtained from Sigma (Deisenhofen, Germany). Porfiromycin was a gift from Dr. J. Brown (Department of Pharmacy,University of Bradford). Culture media and supplements were from Life Technologies, Inc. (Karlsruhe, Germany), and plastics were from Costar and Falcon (Schubert Laboratories, Germany).

Tissue Collection.

Tissues of human tumor xenografts growing s.c. in thymus aplastic nude mice (NMRI background) were collected from the Freiburg Tumor Bank. This bank comprises over 350 human tumor xenograft models that were established in serial passage in vivo, of which 60 are intensively characterized with respect to morphology/histology,chemosensitivity patterns (in vitro and in vivo),and molecular targets (30). Approximately 500 mg of fresh tissue of each tumor was subjected to the clonogenic assay for in vitro chemosensitivity testing against MMC, and 1 g of tumor was flash frozen in liquid nitrogen immediately after removal. Later,tissues were stored at −80°C for determination of DTD enzyme activity and MMC metabolism.

Clonogenic Assay/in Vitro MMC Sensitivity Testing.

Freshly removed xenograft tissues were minced and then incubated with an enzyme cocktail (collagenase, 1.2 units/ml; DNase, 375 units/ml;hyaluronidase, 29 units/ml) at 37°C for 30 min. Tumor homogenates were washed twice with PBS, passed through sieves (200–50 μm), and viable tumor cells were counted. The assay was performed according to a modified two-layer soft agar assay in 24-well plates (31). In brief, 4 × 104 to 8 × 104 cells were added to 0.2 ml Iscoves’modified Dulbecco’s medium/20% FCS containing 0.4% agar and plated on top of the base layer (0.75% agar containing Iscoves’ modified Dulbecco’s medium plus 20% FCS). After 24 h, an additional 0.2 ml of medium (control) or medium containing MMC was added. Each plate contained six untreated control wells, three vehicle controls, and six different drug concentrations (0.1 ng-10 μg/ml) in triplicate. 5-Fluorouracil was used as positive control in a single concentration(1000 μg/ml). Cultures were incubated at 37°C, 7%CO2 for 5–15 days, and monitored closely for colony growth. Vital colonies were stained with 50 μl/well 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride (1 mg/ml) 24 h prior to evaluation, and colonies >50 μm were counted with an automated image analysis system (Omnicon FAS IV;Biosys). Results were expressed as the concentration required to induce 70% growth inhibition/cell kill (IC70). Assays were considered evaluable if control groups produced >20 colonies with a diameter of >50 μm, and initial plate colony counts on days 0 or 2 were <20% of the final control group colony count. Plating efficiencies were ∼0.1% and are consistent with the low plating efficiencies reported for primary human tumor cell cultures(1).

Assessment of in Vivo Activity of MMC in Human Tumor Xenograft Models.

The response of 43 human tumor xenografts to MMC in vivo has been described previously in detail elsewhere (30) and are summarized below. Master stocks of all tumors in the Freiburg panel are maintained in liquid nitrogen, and all chemotherapy studies are conducted on tumors within 10 passages from recovery from master stocks(32). Although the original chemotherapy studies were conducted prior to 1992 (30), this policy should ensure that the response and biochemical properties remain stable enough to allow a valid retrospective study to be conducted. Nevertheless, to ensure that chemosensitivity profiles to MMC were stable, chemotherapy studies were repeated for 13 human tumor xenografts according to the methodology described previously (30). Tumors were implanted s.c. into both flanks of outbred athymic nude mice of NMRI genetic background, and treatment started when the tumors reached a median diameter of 6 mm. At this time (day 0), mice were randomly assigned to either treatment or control groups with five to six mice per group, and MMC was administered i.v. at the maximum tolerated dose of 2 mg/kg on days 1 and 15. Antitumor effects were determined by two-dimensional caliper measurements that were normalized relative to tumor volume at day 0. Experiments were terminated when tumors reached a size of ∼1.5 cm in diameter. Activity was expressed in terms of percentage of optimal T/C (i.e., relative volume of treated tumors divided by the relative volume of control tumors × 100 at the time of maximal drug effect) and classified as complete regression (T/C <10%, +++), partial remission (T/C 11–25%,++), minimal remission (T/C 25–50%, +), resistant (T/C >50%, −). The response of these tumors was very similar (within 95% confidence intervals) to tumor responses obtained in the original study, thereby validating the experimental design of this study (data not shown). All animal experiments were performed in accordance with German Animal License Regulations (Tierschutzgesetz) identical to United Kingdom Co-ordinating Committee on Cancer Research Guidelines for the Welfare of Animals in Experimental Neoplasia (33).

Measurement of DTD Activity.

Tissues were homogenized (10% w/v homogenate) in sucrose (0.25 m) using an Ultra Turrax blender. Cytosolic fractions were prepared by centrifugation of the homogenate at 18,000 × g for 4 min, followed by further centrifugation of the supernatant at 110,000 × g for 1 h at 4°C in a Beckman Optima TL ultracentrifuge. Activity of DTD in the supernatant was determined spectrophotometrically (Beckman DU650 spectrophotometer) by measuring the dicumarol-sensitive reduction of DCPIP at 600 nm (34). Each reaction contained NADH (200μ m), DCPIP (40 μm),dicumarol (20 μm, when required), and a cytosolic fraction of tissues (50 μl per assay) in a final volume of 1 ml of Tris-HCl buffer (50 mm, pH 7.4)containing BSA (0.7 mg/ml). Rates of DCPIP reduction were calculated from the initial linear part of the reaction curve (30 s), and results were expressed in terms of nmol DCPIP reduced/min/mg protein using a molar extinction coefficient of 21 mm/cm for DCPIP. Protein concentration was determined using the Bradford assay(35).

COMPARE Analysis.

We have developed a COMPARE algorithm based on the differential activity of drugs against human tumors growing in soft agar in vitro analogous to the National Cancer Institute-DTP COMPARE computer program (36). This tool allows for comparison of possible structural similarity and molecular targets (37). Thus, DTD levels in human tumor xenografts were ranked and related to their in vitro sensitivity against MMC using the Spearman rank coefficient test (38).

Metabolism of MMC by Tumor Homogenates.

A limited panel of xenografts were selected for this aspect of the study. The selection criteria were simply based upon extremes of tumor response to MMC in vivo [i.e., sensitive (T/C<10%) and resistant (T/C >50%)], and sufficient tumor specimens were included in each group so that the full range of DTD activity was represented (Table 3). Tumors were homogenized in ice-cold Tris-HCl (5 mm, pH 7.4) containing EDTA (0.5 mm) and sucrose (250 mm)using an Ultra Turrax blender (Janke and Kunkel). Samples were centrifuged at 1000 × g for 1 min to remove tumor fragments. Each reaction consisted of tumor homogenate (10 mg/ml protein), MMC (200 μm), NADH and NADPH (2 mm) in a final volume of 0.2 ml of homogenizing buffer. Samples were incubated at 37°C, and at various time intervals after the addition of MMC, 30 μl of reaction mix were removed and added to 90 μl of acetonitrile containing the internal standard,porfiromycin (50 μm), samples were mixed, the solvent was evaporated in a Jouan evaporator, and the residue was resuspended in 100 μl of mobile phase. Chromatographic separation of MMC was achieved using a RP-18 end-capped LiChrospher column (5 μm,250 × 4 mm; Phenomenex, Cheshire, United Kingdom). The HPLC system consisted of a system Gold Beckman 126 Programmable Solvent Module (Beckman Instruments UK Ltd., High Wycombe, United Kingdom), a Beckman 507 Autosampler, which was cooled to 4°C by the Grant Cooling Unit LTD6 (Grant Instruments, Cambridge, United Kingdom), and a Beckman 168 Photo Diode Array Detector. Dual wavelength detection used 365 nm for MMC and 310 nm for the detection of metabolites with a flow rate of 1.2 ml/min, and data were processed using System Gold software(Beckman). The mobile phase consisted of an 18 mmphosphate buffer (pH 6.4):methanol mixture. The gradient program to separate MMC from metabolites was 95% A up to 10 min and then 5% A by 30 min, where A was 95% buffer and B was 76% buffer. The half life of MMC was determined from least squares log linear regression analysis using the equation T1/2 = 0.693/Kd where Kd is the decay rate constant (the slope of the regression analysis × 2.303).

DTD Activity and the Relationship between Enzyme Activity and Tumor Responses to MMC in Vivo.

The activity of DTD in a panel of 58 human tumor xenografts studied is presented in Table 1. There was a broad spectrum of DTD activity both throughout the panel of tumors and within each tumor type (with the exception of gastric cancers where the range of DTD activity was 510.6–980.5 nmol/min/mg). Responses to MMC have been determined for 43 human tumor xenografts,and a broad spectrum of response exists with lung tumors (T/C <10% in five of seven non-small cell lung cancer) being particularly sensitive to MMC (Table 1). No correlation between DTD activity and the response of xenografts to mitomycin C exists (Fig. 1).

DTD Activity and the Relationship between Enzyme Activity and Tumor Responses to MMC in Vitro.

Thirty-eight xenograft tissues that were subjected to DTD activity measurements were also tested for their response to MMC treatment in the tumor stem cell/clonogenic assay in vitro. In terms of the relationship between DTD activity and chemosensitivity in vitro (Fig. 2), a poor correlation exists (regression coefficient, 0.055). Similarly,by using a COMPARE algorithm based on the rank of DTD levels and IC70 of MMC, the poor relationship between DTD activity and the response of xenografts to MMC was confirmed, which is reflected in a Spearman rank coefficient of 0.24 (Table 2).

Metabolism of MMC by Tumor Homogenates.

Representative chromatograms of MMC metabolism by LXFL 529 tumor homogenates are presented in Fig. 3. No metabolites of MMC were present immediately after the addition of MMC (Fig. 3,A). After 60 min incubation at 37°C, MMC was barely detectable, and several metabolites of MMC were found. 2,7-DAM was identified (this coeluted with the products formed as a result of the reaction of MMC with DTD and NADPH) and a very polar metabolite (unknown identity) were the most prominent species detected(Fig. 3,B). Two other metabolites were tentatively identified as 1,2-cis- and 1,2-trans-1-hydroxy-2,7-diaminomitosene, because these two metabolites coeluted with the hydrolysis products formed after the incubation of MMC with HCl (0.1 n) as reported by Cummings et al.(39). No obvious correlation between the presence of these metabolites and antitumor activity(i.e., 2,7-DAM and the polar metabolite could be found in poorly responding tumors that metabolized MMC) was observed (data not shown). A broad spectrum of rates of MMC metabolism was observed in the panel of tumors studied (Table 3), and examples of MMC metabolism by homogenates derived from resistant(RXF 393) and sensitive tumors (LXFL 529) are presented in Fig. 4. Half-lives of MMC in homogenates derived from RXF 393 and LXFL 529 tumors were 234 ± 52 and 34 ± 4 min,respectively. The concentrations of MMC used in this study (200μ m) are above the plasma levels of MMC found in vivo (∼20 μm). This concentration was necessary to detect metabolites of MMC and in the case of LXFL 529 tumors, no differences in terms of MMC half-lives were observed when starting MMC concentrations of 200 or 20μ m were used (data not shown).

Relationship between the Antitumor Activity of MMC, DTD Activity, in Vitro Chemosensitivity, and MMC Metabolism in a Panel of Human Tumor Xenografts.

A panel of xenografts was selected for analysis of drug metabolism. The selection criteria were predominantly based upon two factors. The first was their response to MMC in vivo (i.e., good or poor responders), and the second was based upon DTD activity, where the aim was to have a broad spectrum of high and low DTD activities in both the good- and poor-responding tumor groups. The relationship between antitumor activity, DTD activity, and MMC metabolism by tumor homogenates is presented in Table 3 and Fig. 5. In both good (T/C <10%, score +++) and poor (T/C >50%, score −)responders to MMC, a broad spectrum of DTD activity exists (Table 2). Mean DTD activities for good and poor responders were 326.15 ± 340.21 and 355.4 ± 231.5 nmol/min/mg,respectively. No significant differences exist between DTD activity and response to MMC in this panel of tumors (Fig. 5,A). Similarly, no correlation existed between the rate of MMC metabolism by tumor homogenates (expressed as T1/2values) and DTD activity (Fig. 5,B) in the tumor xenografts studied. MMC half-lives in tumor homogenates derived from sensitive and poorly responsive xenografts were 75 ± 48.3 and 280 ± 129.6 min, respectively. A significant difference[P < 0.005 (two tailed t test)]existed between rates of MMC metabolism in poor and good responders to MMC, although a subset of tumors exist that have the ability to metabolize MMC yet do not respond well to MMC in vivo (Fig. 5,C). Within this panel of tumors, in vitrochemosensitivity data demonstrated that the majority of responsive tumors had low IC70s with the exception of LXFA 289, which was relatively resistant to MMC in vitro (Fig. 5,D and Table 3).

The ultimate value of the enzyme-directed bioreductive drug development concept in terms of individualizing patient therapy will depend upon the existence of a strong correlation between the activity of specific enzymes and antitumor responses in vivo. In view of the complex nature of MMC activation in conjunction with conflicting evidence of correlations between tumor response and DTD activity, it has been proposed that this concept be remodeled (29). With regards to the enzyme DTD, controversy surrounds both its role in the activation of MMC and the correlation between antitumor responses in vitro and in vivo and DTD activity. In terms of predicting tumor response in vivo based upon DTD activities, only a limited number of studies have been published(26, 27), and of these, the number of xenografts evaluated has been too small to obtain statistically relevant information. This study has used a large panel of human tumor xenografts that have a broad spectrum of both DTD activity and antitumor response to MMC(Table 1). The results clearly demonstrate that tumor responses in vivo to MMC cannot be predicted on the basis of DTD activity alone (Fig. 1). It should be noted that this conclusion is based upon the assumption that tumor response to MMC should be proportional to DTD activity. Recent studies using the BE human colon carcinoma cell line transfected with NQO1 have demonstrated that both a lower threshold of DTD activity is required to initiate toxicity to streptonigrin and RH1 (>22 nmol/min/mg) and an upper threshold (>77 nmol/min/mg) beyond which no further increase in toxicity occurs(40). The results presented in Fig. 1 clearly demonstrate that the response of tumors to MMC in vivo is independent of DTD activity and that thresholds of DTD activity cannot be applied in this case with any degree of certainty. A comparison between in vitro chemosensitivity and DTD activity also demonstrates that responses to MMC at the cellular level cannot be predicted on the basis of DTD levels (Fig. 2). These results support the findings of other groups that the response of cells to MMC in vitro cannot be forecast on the basis of DTD activity (25). Recent studies using BE cells transfected with NQO1 have, however, suggested that the influence of DTD activity on chemosensitivity in vitro may be affected by the drug exposure conditions (i.e.,acute or chronic) used in vitro(41). For example, in the case of the aziridinyl benzoquinone compounds, MeDZQ and RH1, a marked cytotoxic potentiation in DTD-rich BE cells occurred when acute (24 h) drug exposures were used, whereas chronic exposures(96 h) showed much less potentiation. In the case of MMC, however, the duration of drug exposure is unlikely to influence the correlation between MMC toxicity in vitro and DTD activity because conflicting reports of good and poor correlations have been generated using similar chronic drug exposure conditions (24, 25). In addition, the poor correlation between the rank order of responses in vitro and DTD activity suggest that the use of short-term drug exposures would not improve the predictive value of the assays. The results of this study, in conjunction with the controversy surrounding MMC activation by DTD, therefore support the view expressed by Cummings et al.(29) that the concept of enzyme-directed bioreductive drug development needs to be remodeled in the case of MMC. It is important to stress that this conclusion applies only to MMC, and the concept may still be applicable for other bioreductive drugs. MMC has a complex mechanism of action involving several enzymes (29, 42, 43), but for compounds that have a simpler mechanism of action where one enzyme predominates in the activation process, this concept may still be valid.

The alternative approach as set out by Cummings et al.(29) is to determine the ability of tumor homogenates to metabolize MMC on the basis that bioactivation of the drug is determined by various enzymes present in the tumor. In a selected panel of tumors that represent the extremes in terms of antitumor response to MMC [i.e., responsive (T/C <10%), n = 7 and nonresponsive (T/C >50%), n = 11], the half-life of MMC in tumor homogenates was 75 ± 48.3 and 280 ± 129.6 min, respectively. The difference in means between the two groups was statistically significant (P < 0.005) and represented a marked improvement over the correlation between DTD and antitumor response in this panel of tumors (Fig. 5,A). There was also a poor relationship between DTD levels and MMC metabolism (Fig. 5,B), which provides indirect evidence to suggest that other enzymes are involved in MMC reduction. Cummings et al.(29) have proposed a model whereby several enzymes compete for MMC on the basis of protein level as opposed to enzyme kinetics. This is based upon the fact that Michaelis Menton affinity constants for MMC are similar for various enzymes (8, 10, 29, 44). If this were the case, then DTD-rich tumors should metabolize MMC rapidly, but this does not hold true for all DTD-rich tumors studied(Table 3). In addition, it is of considerable interest to note that some poorly responsive tumors (Table 3) have the ability to metabolize MMC (e.g., PRXF DU145, PAXF 736, and MEXF 535). There are several possible explanations for this including the fact that within these tumors, the disappearance of the parent compound is the result of a detoxification pathway as opposed to bioreductive activation. In all of these tumors, however, 2,7-DAM (which is a marker for bioactivation)was detectable at levels comparable with tumors that were sensitive to MMC (data not shown). Alternatively, other morphological features of the tumor (i.e., poor blood supply resulting in poor drug delivery) or cellular defense mechanisms (i.e., drug resistance) may have a significant bearing on the outcome of chemotherapy. With regard to possible drug resistance mechanisms, a recent paper by Belcourt et al.(45) have demonstrated that the bacterial MCRA protein (which acts as a hydroquinone oxidase, thereby oxidizing the reduced MMC back to the parent compound) causes profound resistance to MMC under aerobic conditions in Chinese hamster ovary cells transfected with the mcrA gene. However, it is not clear whether resistance to MMC could be caused by a MCRA-like mechanism in mammalian cells. Intrinsic drug resistance may well be the case for PRXF DU145,which has relatively high IC70s (24 ng/ml; Table 3) but not for PAXF 736, which is quite sensitive to MMC in vitro (IC70, 5 ng/ml). Further studies are required to determine why these tumors are resistant to MMC, despite their inherent ability to metabolize MMC.

The results of this study have demonstrated that antitumor responses to MMC cannot be forecast on the basis of either DTD activity or in vitro chemosensitivity, whereas a better correlation between MMC metabolism by tumor homogenates and tumor response in vivoexists. It is of interest to note that a discrepancy exists between the measures of chemosensitivity in vitro and metabolism by tumor homogenates in vitro in terms of predicting responses in vivo because conceptually, these end points should give similar results. There are, however, significant differences between the two methodologies that may explain these findings. The major difference relates to the fact that in vitro metabolism studies are short-term assays performed on tumor homogenates, whereas in vitro chemosensitivity assays described in this paper are relatively long term (conducted over 5–15 days). In the case of the later, it is conceivable that many biological and biochemical parameters may change during the culture period that may have either direct or indirect effects on the outcome of chemotherapy in vitro. For example, the selective growth of the anchorage-independent clonogenic cell population in soft agar effectively removes tumor cells from contact with normal stromal cells and the extracellular matrix. There is now a considerable body of evidence (46) pointing to the fact that many biological processes are influenced by the tumor microenvironment(i.e., the extracellular matrix and stromal components of tumors), and therefore, the results of the clonogenic assay in vitro may not reflect the response of tumors in vivo(as a result of altered cell kinetics or biochemical properties). In addition, the activity of several drug-metabolizing enzymes(e.g., members of the cytochrome P-450 family) is markedly decreased within a few hours of isolation from fresh tissue(47). In vitro metabolism studies, on the other hand, are conducted within a time scale where changes in biochemical and biological parameters are unlikely to occur. The short-term nature of in vitro metabolism studies may therefore represent a more realistic “snapshot” of tumor biology/biochemistry compared with the chronic nature of the clonogenic assay described in this report.

In conclusion, this study has clearly demonstrated that antitumor responses to MMC cannot be predicted on the basis of DTD levels alone. In view of the size of the panel of tumors used, together with the fact that a broad spectrum of DTD levels exists in both responsive and nonresponsive tumors, it is clear that individualizing chemotherapy on the basis of DTD levels is not feasible with regard to MMC. Measurement of MMC metabolism by tumor homogenates, on the other hand, can distinguish between responsive and nonresponsive tumors in the majority of cases. No correlation was seen in terms of the major metabolite of MMC (2,7-DAM) and tumor response in this study, which would appear to conflict with the hypothesis put forward by Cummings et al.(29). It is important to stress, however, that the generation of reactive metabolites would be technically challenging to measure accurately in view of the fact that these metabolites will bind covalently to cellular macromolecules. On the basis of this study,measurement of the disappearance of the prodrug maybe feasible in terms of predicting tumor response in vivo. If a sensitivity threshold for MMC metabolism (in terms of T1/2 values) of 200 min were imposed, the predictive value of the assay is: 7/7 true-positive predictions, 9/11(81.8%) true-negative predictions, and 2/11 (18.8%) false-positive predictions. This represents a substantial improvement over an enzymological end point, but a subset of tumors exists that is capable of metabolizing MMC but does not respond to MMC in vivo. Tumor responses are determined by many factors (48), and it is unlikely that any ex vivo assay will mimic all of these conditions. The key condition that has to be achieved by any predictive assay is that the incidence of false-negative predictions must be low. The results of this study suggest that by measuring the ability of a tumor homogenate to metabolize MMC, it could be possible to identify those tumors that have a good probability of responding. From a technical standpoint, the assay described in this report could be adapted to biopsy material because samples as small as 200 mg can be assayed reproducibly. Further studies are required using clinical material to assess whether this approach to predicting MMC activity has real clinical applications.

Fig. 1.

The relationship between DTD activity and the response of a panel of 43 human tumor xenografts to MMC in vivo.

Fig. 1.

The relationship between DTD activity and the response of a panel of 43 human tumor xenografts to MMC in vivo.

Close modal
Fig. 2.

The relationship between DTD activity and in vitro chemosensitivity (IC70) after continuous exposure to MMC.

Fig. 2.

The relationship between DTD activity and in vitro chemosensitivity (IC70) after continuous exposure to MMC.

Close modal
Fig. 3.

Representative chromatograms showing gradient reversed-phase HPLC separation of MMC and the internal standard,porfiromycin (PMC) at t = 0 (panels A at 365 and 310 nm) and t = 60 min (panels B at 365 and 310 nm) after the addition of MMC to LXFL 529 tumor homogenates. Dotted line, gradient profile for mobile phase B (A = 95% phosphate buffer, 18 mm,pH 6.4:methanol and B = 76% phosphate buffer).

Fig. 3.

Representative chromatograms showing gradient reversed-phase HPLC separation of MMC and the internal standard,porfiromycin (PMC) at t = 0 (panels A at 365 and 310 nm) and t = 60 min (panels B at 365 and 310 nm) after the addition of MMC to LXFL 529 tumor homogenates. Dotted line, gradient profile for mobile phase B (A = 95% phosphate buffer, 18 mm,pH 6.4:methanol and B = 76% phosphate buffer).

Close modal
Fig. 4.

Metabolism of MMC in tumor homogenates derived from a MMC-sensitive (LXFL529, ⊡) and -resistant (RXF393, ⊙) human tumor xenografts. Half-lives of MMC in LXFL 529 and RXF 393 tumor homogenates were 34 and 234 min, respectively.

Fig. 4.

Metabolism of MMC in tumor homogenates derived from a MMC-sensitive (LXFL529, ⊡) and -resistant (RXF393, ⊙) human tumor xenografts. Half-lives of MMC in LXFL 529 and RXF 393 tumor homogenates were 34 and 234 min, respectively.

Close modal
Fig. 5.

Relationship between DTD activity, in vitrochemosensitivity, MMC metabolism, and antitumor responses to MMC in vivo. A, lack of correlation between DTD activity and antitumor activity in tumors that are resistant[NR (−)] and sensitive [R (+++)] to MMC. B, lack of correlation between the rate of MMC metabolism and DTD activity in sensitive (⊙) and resistant (⊡)tumors. C, relationship between MMC metabolism and tumor response to MMC. A significant difference (P < 0.005) exists between MMC metabolism by tumor homogenates derived from responsive [R (+++)] and nonresponsive [NR (−)]xenografts in vivo. D, relationship between in vitro chemosensitivity and antitumor response in vivo in responsive (R) and nonresponsive (NR) tumor xenografts.

Fig. 5.

Relationship between DTD activity, in vitrochemosensitivity, MMC metabolism, and antitumor responses to MMC in vivo. A, lack of correlation between DTD activity and antitumor activity in tumors that are resistant[NR (−)] and sensitive [R (+++)] to MMC. B, lack of correlation between the rate of MMC metabolism and DTD activity in sensitive (⊙) and resistant (⊡)tumors. C, relationship between MMC metabolism and tumor response to MMC. A significant difference (P < 0.005) exists between MMC metabolism by tumor homogenates derived from responsive [R (+++)] and nonresponsive [NR (−)]xenografts in vivo. D, relationship between in vitro chemosensitivity and antitumor response in vivo in responsive (R) and nonresponsive (NR) tumor xenografts.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported by the Cancer Research Campaign and the Association for International Cancer Research.

3

The abbreviations used are: MMC, mitomycin C;DTD, DT-diaphorase; DCPIP, dichlorophenolindophenol; HPLC,high-performance liquid chromatography; 2,7DAM,2,7-diaminomitosene.

Table 1

Relationship between DTD activity and the response of a panel of human tumor xenografts to MMC (2 mg/kg i.v. days 1 and 15)

TumoraDTD activity (nmol/min/mg)In vivo response (% T/C)
XF 575 1843.8 ± 321.2 ND 
OVXF 899 1375.4 ± 250.0 29.8 (+) 
PRXF DU145 1075.1 ± 53.9 79.6 (−) 
GXF 251 980.5 ± 250.0 31.1 (+) 
LXFL 1072 800.3 ± 95.6 ND 
RXF 1220 799.5 ± 72.4 86.4 (−) 
GXF 209 660.5 ± 107.0 6.3 (+++) 
GXF 97 582.8 ± 53.0 2.6 (+++) 
LXFE 839 564.4 ± 64.4 6.4 (+++) 
LXFA 289 554.4 ± 86.7 8.3 (+++) 
GXF 214 510.6 ± 66.9 59.4 (−) 
PAXF 736 415.6 ± 16.6 ND 
CEXF 633 370.9 ± 37.8 2.0 (+++) 
BXF 1299 362.5 ± 7.5 88.64 (−) 
LXFL 529 322.2 ± 6.7 5.2 (+++) 
CXF HT29X 313.4 ± 62.1 82.7 (−) 
PAXF 546 300.9 ± 17.2 49.1 (+) 
CXF 280 238.9 ± 31.3 5.2 (+++) 
BXF 1036 233.6 ± 50.6 ND 
MEXF 989 229.4 ± 44.5 21.7 (++) 
LXFA 418 228.0 ± 25.5 ND 
CXF 609 197.0 ± 15.2 59.4 (−) 
BXF 1352 169.2 ± 16.3 ND 
HNXF 536 159.3 ± 12.1 74.4 (−) 
CXF HCT116 155.0 ± 3.9 40.0 (+) 
PRXF PC3MX 144.4 ± 26.4 100 (−) 
MEXF 514 134.4 ± 13.5 76.2 (−) 
CXF 158 120.3 ± 23.9 67.5 (−) 
RXF 488 120.5 ± 4.7 ND 
LXFS 538 125.4 ± 15.2 1.5 
CXF 883 93.3 ± 2.3 ND 
BXF 439 88.1 ± 16.6 ND 
LXFE 211 64.6 ± 4.8 8.3 (+++) 
CXF 1103 61.8 ± 2.6 57.8 (−) 
OVXF 1353 58.3 ± 7.5 100 (−) 
CXF DLD1LX 55.3 ± 13.7 ND 
SXF 1410 25.4 ± 2.9 ND 
RXF 423 23.0 ± 0.4 74.2 (−) 
LXFA 526 20.5 ± 2.3 76.9 (−) 
PRXF 1369 16.2 ± 5.0 69.2 (−) 
TXF 881 15.9 ± 2.9 31.4 (+) 
LXFE 397 15.2 ± 0.8 67.8 (−) 
BXF 1258 11.9 ± 1.1 41.6 (+) 
CNXF 498 9.7 ± 4.8 ND 
BXF 1301 6.0 ± 7.9 ND 
RXF 486 6.1 ± 4.2 88.0 (−) 
LEXF HL60X 4.7 ± 4.9 100 (−) 
OVXF 1023 4.3 ± 0.1 7.9 (+++) 
SXF 627 3.6 ± 6.3 ND 
LXFS 650a 1.9 ± 0.7 57.2 (−) 
RXF 944LX 1.8 ± 2.1 ND 
LYXF 1189 1.7 ± 2.9 63.2 (−) 
MEXF 1341 1.5 ± 1.7 54.4 (−) 
MAXF 857 1.4 ± 0.6 79.2 (−) 
LXFE 409 1.1 ± 2.0 3.6 (+++) 
MAXF 1162 0.4 ± 0.6 70.9 (−) 
LXFS 650b <0.1 57.2 (−) 
MAXF 449 <0.1 8.1 (+++) 
MEXF 535 <0.1 85.3 (−) 
TumoraDTD activity (nmol/min/mg)In vivo response (% T/C)
XF 575 1843.8 ± 321.2 ND 
OVXF 899 1375.4 ± 250.0 29.8 (+) 
PRXF DU145 1075.1 ± 53.9 79.6 (−) 
GXF 251 980.5 ± 250.0 31.1 (+) 
LXFL 1072 800.3 ± 95.6 ND 
RXF 1220 799.5 ± 72.4 86.4 (−) 
GXF 209 660.5 ± 107.0 6.3 (+++) 
GXF 97 582.8 ± 53.0 2.6 (+++) 
LXFE 839 564.4 ± 64.4 6.4 (+++) 
LXFA 289 554.4 ± 86.7 8.3 (+++) 
GXF 214 510.6 ± 66.9 59.4 (−) 
PAXF 736 415.6 ± 16.6 ND 
CEXF 633 370.9 ± 37.8 2.0 (+++) 
BXF 1299 362.5 ± 7.5 88.64 (−) 
LXFL 529 322.2 ± 6.7 5.2 (+++) 
CXF HT29X 313.4 ± 62.1 82.7 (−) 
PAXF 546 300.9 ± 17.2 49.1 (+) 
CXF 280 238.9 ± 31.3 5.2 (+++) 
BXF 1036 233.6 ± 50.6 ND 
MEXF 989 229.4 ± 44.5 21.7 (++) 
LXFA 418 228.0 ± 25.5 ND 
CXF 609 197.0 ± 15.2 59.4 (−) 
BXF 1352 169.2 ± 16.3 ND 
HNXF 536 159.3 ± 12.1 74.4 (−) 
CXF HCT116 155.0 ± 3.9 40.0 (+) 
PRXF PC3MX 144.4 ± 26.4 100 (−) 
MEXF 514 134.4 ± 13.5 76.2 (−) 
CXF 158 120.3 ± 23.9 67.5 (−) 
RXF 488 120.5 ± 4.7 ND 
LXFS 538 125.4 ± 15.2 1.5 
CXF 883 93.3 ± 2.3 ND 
BXF 439 88.1 ± 16.6 ND 
LXFE 211 64.6 ± 4.8 8.3 (+++) 
CXF 1103 61.8 ± 2.6 57.8 (−) 
OVXF 1353 58.3 ± 7.5 100 (−) 
CXF DLD1LX 55.3 ± 13.7 ND 
SXF 1410 25.4 ± 2.9 ND 
RXF 423 23.0 ± 0.4 74.2 (−) 
LXFA 526 20.5 ± 2.3 76.9 (−) 
PRXF 1369 16.2 ± 5.0 69.2 (−) 
TXF 881 15.9 ± 2.9 31.4 (+) 
LXFE 397 15.2 ± 0.8 67.8 (−) 
BXF 1258 11.9 ± 1.1 41.6 (+) 
CNXF 498 9.7 ± 4.8 ND 
BXF 1301 6.0 ± 7.9 ND 
RXF 486 6.1 ± 4.2 88.0 (−) 
LEXF HL60X 4.7 ± 4.9 100 (−) 
OVXF 1023 4.3 ± 0.1 7.9 (+++) 
SXF 627 3.6 ± 6.3 ND 
LXFS 650a 1.9 ± 0.7 57.2 (−) 
RXF 944LX 1.8 ± 2.1 ND 
LYXF 1189 1.7 ± 2.9 63.2 (−) 
MEXF 1341 1.5 ± 1.7 54.4 (−) 
MAXF 857 1.4 ± 0.6 79.2 (−) 
LXFE 409 1.1 ± 2.0 3.6 (+++) 
MAXF 1162 0.4 ± 0.6 70.9 (−) 
LXFS 650b <0.1 57.2 (−) 
MAXF 449 <0.1 8.1 (+++) 
MEXF 535 <0.1 85.3 (−) 
a

BXF, bladder; CXF, colorectal;GXF, gastric; LXFA, lung (adenocarcinoma); LXFE, lung (epidermoid);LXFL, lung (large cell); LXFS, lung (small cell lung cancer; MAXF,mammary; MEXF, melanoma; OVXF, ovarian; RXF, renal; TXF, testis; SXF,sarcoma; CNXF, central naval system; CEXF, cervical; PRXF, prostate;HNXF, head and neck; LYXF, lymphoma; LEXF, leukaemia; PAXF, pancreatic;XF, mixed histology.

b

Values in parentheses denote activity ratings where (−) = T/C values taken from Fiebig et al (30). DTD-specific activities represent the mean of three independent experiments ± SD.

Table 2

Relationship between DTD levels and the response in vitro of primary explants derived from human tumor xenografts to MMC: Results of COMPARE analysis

TumoraIC70 (ng/ml)RankaDTD (nmol/min/mg)Rankb
MAXF 449 15 24.5 <0.1 
MAXF 1162 10.0 0.4 
LXFE 409 3.5 1.1 
MAXF 857 10.0 1.4 
MEXF 1341 26 32.0 1.5 
LXFS 650b 6.0 1.9 
SXF 627 10.0 1.4 
OVXF 1023 15 24.5 4.3 
BXF 1301 45 35.0 6.0 
CNXF 498 10.0 9.7 10 
LXFE 397 16.0 15.2 11 
TXF 881 3.5 15.9 12 
LXFA 526 11 17.0 20.5 13 
RXF 423 15 24.5 23 14 
OVXF 1353 15 24.5 58.3 15 
CXF 1103 15 24.5 61.8 16 
LXFE 211 15.0 64.6 17 
CXF 158 83 37.0 120 18.5 
RXF488 15 24.5 120 18.5 
MEXF 514 12 18.0 134 20 
CXF HCT116 50 36.0 155 21 
HNXF 536 10.0 159 22 
CXF 609 15 24.5 197 23 
MEXF 989 15 24.5 229 24 
CXF 280 1.5 238 25 
PAXF 546 15 24.5 300 26 
LXFL 529 5.0 322 27 
BXF 1299 15 24.5 362 28 
CEXF633 10.0 10.0 29 
PAXF 736 10.0 415 30 
LXFA 289 92 38.0 554 31 
GXF 97 1.5 582 32 
GXF 209 14.0 660 33 
RXF 1220 37 33.0 799 34 
LXFL 1072 15 24.5 800 35 
GXF 251 15 243.5 980 36 
PRXF DU145 24 31.0 1075 37 
OVXF 899 15 34.0 1375 38 
TumoraIC70 (ng/ml)RankaDTD (nmol/min/mg)Rankb
MAXF 449 15 24.5 <0.1 
MAXF 1162 10.0 0.4 
LXFE 409 3.5 1.1 
MAXF 857 10.0 1.4 
MEXF 1341 26 32.0 1.5 
LXFS 650b 6.0 1.9 
SXF 627 10.0 1.4 
OVXF 1023 15 24.5 4.3 
BXF 1301 45 35.0 6.0 
CNXF 498 10.0 9.7 10 
LXFE 397 16.0 15.2 11 
TXF 881 3.5 15.9 12 
LXFA 526 11 17.0 20.5 13 
RXF 423 15 24.5 23 14 
OVXF 1353 15 24.5 58.3 15 
CXF 1103 15 24.5 61.8 16 
LXFE 211 15.0 64.6 17 
CXF 158 83 37.0 120 18.5 
RXF488 15 24.5 120 18.5 
MEXF 514 12 18.0 134 20 
CXF HCT116 50 36.0 155 21 
HNXF 536 10.0 159 22 
CXF 609 15 24.5 197 23 
MEXF 989 15 24.5 229 24 
CXF 280 1.5 238 25 
PAXF 546 15 24.5 300 26 
LXFL 529 5.0 322 27 
BXF 1299 15 24.5 362 28 
CEXF633 10.0 10.0 29 
PAXF 736 10.0 415 30 
LXFA 289 92 38.0 554 31 
GXF 97 1.5 582 32 
GXF 209 14.0 660 33 
RXF 1220 37 33.0 799 34 
LXFL 1072 15 24.5 800 35 
GXF 251 15 243.5 980 36 
PRXF DU145 24 31.0 1075 37 
OVXF 899 15 34.0 1375 38 
a

For descriptions of tumor types, see Table 1.

b

Spearman rank coefficient, 0.24. n = 38.

Table 3

Relationship between rumor response to MMC, DTD activity, in vitro chemosensitivity, and rate of MMC metabollism in tumor homogenates

TumorDTD activity (nmol/min/mg)Response to MMC in vivo (%T/C)In vitro chemosensitivity (IC7070, ng/ml)aMMC metabolism by tumor homogenates (T1/2, min)a
PRXF DU145 1075 79.6 (−) 24 80 
RXF 1220 510 86.4 (−) 37 218 
GXF 214 510 59.4 (−) NDb 586 
PAXF 736 415.6 82.0 (−) 213 
RXF 393 243.5 78.5 (−) ND 234 
BXF 1299 362.5 88.6 (−) 15 341 
HNXF 536 159.3 74.4 (−) 327 
LXFA 526 20.5 76.9 (−) 11 273 
LXFS 650 1.9 57.2 (−) 397 
MAXF 1162 0.4 70.9 (−) 279 
MEXF 535 <0.1 85.3 (−) ND 135 
GXF 209 660.5 6.3 (+++) 109 
GXF 97 582.8 2.6 (+++) 68 
LXFA 289 554 8.3 (+++) 92 168 
LXFL 529 322.2 5.2 (+++) 34 
CXF 280 238.9 5.2 (+++) 90 
LXFS 538 125.4 1.5 (+++) ND 37 
OVXF 1023 4.3 7.9 (+++) 15 19 
TumorDTD activity (nmol/min/mg)Response to MMC in vivo (%T/C)In vitro chemosensitivity (IC7070, ng/ml)aMMC metabolism by tumor homogenates (T1/2, min)a
PRXF DU145 1075 79.6 (−) 24 80 
RXF 1220 510 86.4 (−) 37 218 
GXF 214 510 59.4 (−) NDb 586 
PAXF 736 415.6 82.0 (−) 213 
RXF 393 243.5 78.5 (−) ND 234 
BXF 1299 362.5 88.6 (−) 15 341 
HNXF 536 159.3 74.4 (−) 327 
LXFA 526 20.5 76.9 (−) 11 273 
LXFS 650 1.9 57.2 (−) 397 
MAXF 1162 0.4 70.9 (−) 279 
MEXF 535 <0.1 85.3 (−) ND 135 
GXF 209 660.5 6.3 (+++) 109 
GXF 97 582.8 2.6 (+++) 68 
LXFA 289 554 8.3 (+++) 92 168 
LXFL 529 322.2 5.2 (+++) 34 
CXF 280 238.9 5.2 (+++) 90 
LXFS 538 125.4 1.5 (+++) ND 37 
OVXF 1023 4.3 7.9 (+++) 15 19 
a

Values presented represent the means of at least three independent experiments. SDs(n = 3) for in vitrochemosensitivity tests and in vitro metabolism studies were less than 20 and 15%, respectively.

b

ND, not determined.

1
Selby P. J., Buick R. N., Tannock I. A critical appraisal of the human tumor stem cell assay.
N. Engl. J. Med.
,
308
:
129
-134,  
1983
.
2
Von Hoff D. D. He’s not going to talk about in vitro predictive assays again is he?.
J. Natl. Cancer Inst.
,
82
:
96
-101,  
1990
.
3
McLeod H. L., Murray G. I., Mollison J., McKay J., Cassidy J. Selection of markers to predict tumor response or survival: description of a novel approach.
Eur. J. Cancer
,
35
:
1650
-1652,  
1999
.
4
Workman P., Walton M. I. Enzyme directed bioreductive drug development Adams G. E. Breccia A. Fielden E. M. Wardman P. eds. .
Selective Activation of Drugs by Redox Processes
,
:
173
-191, Plenum Publishing Corp. New York  
1990
.
5
Workman P. Enzyme directed bioreductive drug development revisited: a commentary on recent progress and future prospects with emphasis on quinone anticancer drugs and quinone metabolizing enzymes, particularly DT-diaphorase.
Oncol. Res.
,
6
:
461
-475,  
1994
.
6
Beretta, G. (ed.). Mitomycin C: Current Status. Turin: Edizioni Minerva Medica, 1996.
7
Sartorelli A. C., Hodnick W. F., Belcourt M. F., Tomaz M., Haffty B., Fisher J. J., Rockwell S. Mitomycin C: a prototype bioreductive agent.
Oncol. Res.
,
6
:
501
-108,  
1994
.
8
Pan S-S., Andrews P. A., Glover C. J., Bachur N. R. Reductive activation of mitomycin C and mitomycin C metabolites by NADPH-cytochrome P450 reductase and xanthine oxidase.
J. Biol. Chem.
,
259
:
959
-966,  
1984
.
9
Siegel D., Gibson N. W., Preusch P. C., Ross D. Metabolism of mitomycin C by DT-diaphorase: role in mitomycin C induced DNA damage and cytotoxicity in human colon carcinoma cells.
Cancer Res.
,
50
:
7483
-7489,  
1990
.
10
Gustafsson D. L., Pritsos C. A. Bioactivation of mitomycin C by xanthine dehydrogenase from EMT6 mouse mammary carcinoma tumors.
J. Natl. Cancer Inst.
,
84
:
1180
-1185,  
1992
.
11
Hodnick W. F., Sartorelli A. C. Reductive activation of mitomycin C by NADH: cytochrome b5 reductase.
Cancer Res.
,
53
:
4907
-4912,  
1993
.
12
Joseph P., Xu Y., Jaiswal A. K. Non-enzymatic and enzymatic activation of mitomycin C: identification of a unique cytosolic activity.
Int. J. Cancer
,
65
:
263
-271,  
1996
.
13
Spanswick V. J., Cummings J., Smyth J. F. Enzymology of mitomycin C metabolic activation in tumor tissue: characterization of a novel mitochondrial reductase.
Biochem. Pharmacol.
,
51
:
1623
-1630,  
1996
.
14
Krishna M. C., DeGraff W., Tamura S., Gonzalez F. J., Samuni A., Russo A., Mitchell J. B. Mechanism of hypoxic and aerobic toxicity of mitomycin C in Chinese hamster V79 cells.
Cancer Res.
,
51
:
6622
-6628,  
1991
.
15
Belcourt M. F., Hodnick W. F., Rockwell S., Sartorelli A. C. Differential toxicity of mitomycin C and porfiromycin to aerobic and hypoxic Chinese hamster ovary cells overexpressing human NADPH: cytochrome P-450 reductase.
Proc. Natl. Acad. Sci. USA
,
93
:
456
-460,  
1996
.
16
Ross D., Beall H. D., Siegel D., Traver R. D., Gustafson D. L. Enzymology of bioreductive drug activation.
Br. J. Cancer
,
74(Suppl.27)
:
1s
-8s,  
1996
.
17
Schlager J. J., Powis G. Mitomycin C is not metabolized by but is an inhibitor of human kidney NAD(P)H: (quinone acceptor) oxidoreductase.
Cancer Chemother. Pharmacol.
,
22
:
126
-130,  
1988
.
18
Siegel D., Beall H. D., Kasai M., Arai H., Gibson N. W., Ross D. pH dependent inactivation of DT-diaphorase by mitomycin C and porfiromycin.
Mol. Pharmacol.
,
44
:
1128
-1134,  
1993
.
19
Belcourt M. F., Hodnick W. F., Rockwell S., Sartorelli A. C. Bioactivation of mitomycin C antibiotics by aerobic and hypoxic Chinese hamster ovary cells overexpressing DT-diaphorase.
Biochem. Pharmacol.
,
51
:
1669
-1678,  
1996
.
20
Dulhanty A. M., Whitmore G. F. Chinese hamster ovary cells resistant to mitomycin C under aerobic but not hypoxic conditions are deficient in DT-diaphorase.
Cancer Res.
,
51
:
1860
-1865,  
1991
.
21
Powis G., Gasdaska P. Y., Gallegos A., Sherill K., Goodman D. Overexpression of DT-diaphorase in transfected NIH 3T3 cells does not lead to increased anticancer quinone sensitivity: a questionable role for the enzyme as a target for bioreductively activated anticancer drugs.
Anticancer Res.
,
15
:
1141
-1146,  
1995
.
22
O’Dwyer P. J., Perez R. P., Yao K-S., Godwin A. K., Hamilton T. C. Increased DT-diaphorase expression and cross resistance to mitomycin C in a series of cisplatin resistant human ovarian cancer cell lines.
Biochem. Pharmacol.
,
52
:
21
-27,  
1996
.
23
Plumb J. A., Workman P. Unusually marked hypoxic sensitization to indoloquinone EO9 and mitomycin C in a human colon tumor cell line that lacks DT-diaphorase.
Int. J. Cancer
,
56
:
134
-139,  
1994
.
24
Fitzsimmons S. A., Workman P., Grever M., Paull K., Camalier R., Lewis A. D. Reductase enzyme expression across the National Cancer Institute tumor cell line panel: correlation with sensitivity to mitomycin C and EO9.
J. Natl. Cancer Inst.
,
88
:
259
-269,  
1996
.
25
Robertson N., Stratford I. J., Houlbrook S., Carmichael J., Adams G. E. The sensitivity of human tumor cells to quinone bioreductive drugs: what role for DT-diaphorase?.
Biochem. Pharmacol.
,
44
:
409
-412,  
1992
.
26
Malkinson A. M., Siegel D., Forrest G. L., Gazdar A. F., Oie H. K., Chan D. C., Bunn P. A., Mabry M., Dykes D. J., Harrison S. D., Jr., Ross D. Elevated DT-diaphorase activity and messenger RNA content in human non-small cell lung carcinoma: relationship to the response of lung tumor xenografts to mitomycin C.
Cancer Res.
,
52
:
4752
-4757,  
1992
.
27
Nishiyama M., Saeki S., Aogi K., Hirabayashi N., Toge T. Relevance of DT-diaphorase activity to mitomycin C (MMC) efficacy on human cancer cells: differences in in vitro and in vivo systems.
Int. J. Cancer
,
53
:
1013
-1016,  
1993
.
28
Cummings J., Spanswick V. J., Gardiner J., Ritchie A., Smyth J. F. Pharmacological and biochemical determinants of the antitumor activity of the indoloquinone EO9.
Biochem. Pharmacol.
,
55
:
253
-260,  
1998
.
29
Cummings J., Spanswick V. J., Tomaz M., Smyth J. F. Enzymology of mitomycin C metabolic activation in tumor tissue.
Implications for enzyme directed bioreductive drug development. Biochem. Pharmacol.
,
56
:
405
-414,  
1998
.
30
Fiebig H. H., Berger D. P., Dengler W. A., Wallbrecher E., Winterhalter B. R. Combined in vitro/in vivo test procedure with human tumor xenografts Fiebig H. H. Berger D. P. eds. .
Immunodeficient Mice in Oncology
,
:
321
-351, S. Karger AG Contrib. Oncol. Basel  
1992
.
31
Hamburger A. W., Salmon S. E. Primary bioassay of human tumor stem cells.
Science (Washington DC)
,
197
:
461
-463,  
1977
.
32
Fiebig, H. H., Dengler, W. A., and Roth, T. Human Tumor xenografts. Predictivity, characterisation and discovery of new anticancer drugs. In: H. H. Fiebig and A. M. Burger (eds.), Relevance of Tumor Models for Anticancer Drug Development, Vol. 54, pp. 29–50. Basel: S. Karger AG Contrib. Oncol., 1999.
33
Workman, P., Twentyman, P., Balkwill, F., Balmain, A., Chaplin, D., Double, J. A., Embleton, J., Newell, D., Raymond, R., Stables, J., Stevens, T., and Wallace, J. United Kingdom Co-ordinating Committee on Cancer Research (UKCCCR) guidelines for the welfare of animals in experimental neoplasia (ed. 2). Br. J. Cancer, 77: 1–10, 1998.
34
Traver R. D., Horikoshi T., Dannenberg K. D., Stadlbauer T. H. W., Dannenberg P. V., Ross D., Gibson N. W. NAD(P)H: quinone oxidoreductase gene expression in human colon carcinoma cells: characterization of a mutation which modulates DT-diaphorase activity and mitomycin sensitivity.
Cancer Res.
,
52
:
797
-802,  
1992
.
35
Bradford M. M. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding.
Anal. Biochem.
,
72
:
248
-254,  
1976
.
36
Paull K. D., Shoemaker R. H., Hodes L., Monks A., Scudiero D. A., Rubinstein L., Plowman J., Boyd M. R. Display and analysis of patterns of differential activity of drugs against human tumor cell lines: development of mean graph and COMPARE algorithm.
J. Natl. Cancer Inst.
,
81
:
1088
-1092,  
1989
.
37
Weinstein J. N., Myers G. T., O’Conner P. M., Friend S. H., Fornace A. J., Kohn K. W., Fojo T., Bates S. E., Rubinstein L. V., Anderson L. N., Buolamwini J. K., van Osdo W. W., Monks A. P., Scudiero D. A., Sausville E. A., Zaharevitz D. W., Bunow B., Viswanadhan V. N., Johnson G. S., Wittes R. E., Paull K. D. An information-intensive approach to the molecular pharmacology of cancer.
Science (Washington DC)
,
275
:
343
-349,  
1997
.
38
Sachs, L. Angewandte Statistik: Anwendung statistischer Methoden, Ed. 8, pp. 510–515. Berlin: Springer-Verlag, 1997.
39
Cummings J., Chirrey L., Willmott N., Halbert G. W., Smyth J. F. Determination of mitomycin C, 2,7-diaminomitosene, 1,2-cis- and 1,2-trans-1-hydroxy-2,7-diaminomitosene in tumor tissue by high performance liquid chromatography.
J. Chromatogr. B
,
612
:
105
-113,  
1993
.
40
Winiski S. L., Swann E., Hargreaves R. H. L., Butler J., Moody C. J., Ross D. Relationship between NAD(P)H: quinone oxidoreductase 1 (NQO1) levels in a series of stably transfected cell lines and susceptibility to antitumor quinones.
Clin. Cancer Res.
,
5
:
3765S
1999
.
41
Kelland L. R., Sharp S. Y., Valenti M. R., Brunton L. A., Workman P. An isogenic human colon model for NQO1 and its application in determining the role of DT-diaphorase in the antitumor activity of a range of quinone based agents.
Clin. Cancer Res.
,
5
:
3818S
1999
.
42
Cummings J., Spanswick V. J., Smyth J. F. Re-evaluation of the molecular pharmacology of mitomycin C.
Eur. J. Cancer
,
31A
:
1928
-1933,  
1995
.
43
Ross D., Siegel D., Beall H., Prakash A. S., Mulcahy R. T., Gibson N. W. DT-diaphorase in activation and detoxification of quinones.
Bioreductive activation of mitomycin C. Cancer Metastasis Rev.
,
12
:
83
-101,  
1993
.
44
Walton M. I., Sugget N., Workman P. The role of human and rodent DT-diaphorase in the reductive metabolism of hypoxic cell cytotoxins.
Int. J. Radiat. Oncol. Biol. Phys.
,
22
:
643
-647,  
1992
.
45
Belcourt M. F., Penketh P. G., Hodnick W. F., Johnson D. A., Sherman D. H., Rockwell S., Sartorelli A. C. Mitomycin resistance in mammalian cells expressing the bacterial mitomycin C resistance protein MCRA.
Proc. Natl. Acad. Sci. USA
,
96
:
10489
-10494,  
1999
.
46
Wernert N. The multiple roles of tumor stroma.
Virchows Arch.
,
430
:
433
-443,  
1997
.
47
Patterson L. H., McKeown S. R., Robson T., Gallagher R., Raleigh S. M., Orr S. Antitumor prodrug development using cytochrome P450 (CYP) mediated activation.
Anti-Cancer Drug Des.
,
14
:
473
-486,  
2000
.
48
Phillips R. M., Bibby M. C., Double J. A. A critical appraisal of the predictive value of in vitro chemosensitivity assays.
J. Natl. Cancer Inst.
,
82
:
1457
-1468,  
1990
.