XR5944 (MLN944) is a novel DNA targeting agent with potent antitumor activity, both in vitro and in vivo, against several murine and human tumor models. We have used an ATP-tumor chemosensitivity assay to assess the ex vivo sensitivity of a variety of solid tumors (n = 90) and a CCRF-CEM leukemia cell line selected with XR5944. Differences in gene expression between the parental CCRF-CEM and the resistant subline were investigated by quantitative reverse transcription-PCR. Immunohistochemistry for topoisomerases I and IIα and multidrug resistance (MDR1) protein was done on those tumors for which tissue was available (n = 32). The CCRF-CEM XR5944 line showed increased mRNA levels of MDR1, major vault protein, and MDR-associated protein 1 compared with the parental line, whereas the expression of topoisomerases I, IIα, and IIβ was essentially unchanged, suggesting that XR5944 is susceptible to MDR mechanisms. The median IC90 and IC50 values for XR5944 in tumor-derived cells were 68 and 26 nmol/L, respectively, 6-fold greater than in resistant cell lines. XR5944 was 40- to 300-fold more potent than the other cytotoxics tested, such as doxorubicin, topotecan, and paclitaxel. Breast and gynecologic malignancies were most sensitive to XR5944, whereas gastrointestinal tumors showed greater resistance. A positive correlation (r = 0.68; P < 0.0001) was found between the IC50 values of XR5944 and P-glycoprotein/MDR1 staining but not with either topoisomerase I or IIα immunohistochemistry index. These data support the rapid introduction of XR5944 to clinical trials and suggest that it may be effective against a broad spectrum of tumor types, especially ovarian and breast cancer.

The novel bis-phenazine XR5944 (MLN944) is an extremely potent cytotoxic agent both in vitro and in vivo (1). Against a panel of human cell lines in vitro, the IC50 of XR5944 was 0.04 to 0.4 nmol/L, and this potency translated well to human xenograft models in vivo where XR5944 induced complete tumor regression in the H69 small cell lung carcinoma model (1). Although XR5944 originated from a program to generate dual topoisomerase I and II inhibitors (2), recent data suggest that cell death is not mediated via topoisomerase inhibition, but the precise mechanism of action is still being elucidated. XR5944 has been reported to bind strongly and intercalate into DNA (2) and can stabilize topoisomerase-dependent cleavage complexes as visualized by electrophoresis using linearized, labeled plasmid DNA and purified topoisomerases I and II. XR5944 also induced cleavage complex formation for topoisomerases I,IIα, and IIβ in human leukemic K562 cells visualized using the Trapped in Agarose DNA Immunostaining assay (3). Although these observations suggested a topoisomerase-mediated mechanism for XR5944, the increase in enzyme-mediated DNA cleavage required relatively high concentrations of XR5944 and, in the K562 cells, long incubation times. Furthermore, data have been presented recently demonstrating a topoisomerase-independent mechanism of action for XR5944. In yeast models, the cytotoxicity of XR5944 was not dependent on the presence of either topoisomerase I or II and the potency was not attenuated in mutant strains unable to repair dsDNA breaks (4). Cell cycle analysis also differentiated XR5944 from both topoisomerase I and II inhibitors, as XR5944 treatment induced a G1 and G2 arrest in contrast to the G2-M arrest noted with either doxorubicin (a topoisomerase II inhibitor) or camptothecin (a topoisomerase I inhibitor; ref. 4). Finally, functional genomic data have also differentiated XR5944 from known topoisomerase inhibitors. Transcript profiling of XR5944 in yeast cells indicated up-regulated expression of RNA polymerase subunits as well as genes involved in rRNA processing; importantly, DNA damage response genes seemed to be unaffected (4).

Previous in vitro studies showed that XR5944 is probably a substrate for both multidrug resistance (MDR)–associated protein (MRP) and P-glycoprotein (P-gp; ref. 1), the product of the MDR1 gene. Although the potency of XR5944 is somewhat attenuated in cell lines overexpressing these drug efflux proteins, XR5944 remains a very active cytotoxic with IC50 values similar to or better than other chemotherapeutic agents such as topotecan and paclitaxel in drug-resistant cells.

Many compounds show excellent activity on cell lines, but this does not always translate to clinical efficacy. Despite the many advantages of using cell lines (cost, rapidity, and reproducibility), there are also some disadvantages. In particular, in contrast to human tumors, cell lines consist of a homogenous population of rapidly proliferating cells that grow reproducibly in culture (5, 6). Proliferating cells also show an enhanced response to anticancer drugs (7, 8). This is reflected in the response rates of rapidly proliferating cancers to anticancer drugs (9) but also applies to cell lines derived from tumors by selection for cells that grow rapidly in serum-containing cell culture.

As XR5944 is in phase I clinical trials (10), it is potentially important to show that the compound is effective against cells derived from clinical tumor samples to help design phase II trials. It has been proposed that the ATP-tumor chemosensitivity assay (ATP-TCA) can be used in the development of new agents and combinations for use in cancer patients (11). Although chemosensitivity testing has been used previously to help guide decision-making in the pharmaceutical industry, the ATP-based assay has considerable advantages in terms of reproducibility and sensitivity compared with other assay types (12–14). As an example, this method has been employed previously to assess the ex vivo activity of a novel, combined inhibitor of topoisomerase I and II inhibitors, XR5000 (15): the assay showed that although XR5000 was effective against melanoma as well as ovarian cancer ex vivo, the concentrations required were unlikely to be achieved in patients (16).

In the present study, we aimed to determine the ex vivo activity of XR5944 in a variety of solid tumors to investigate possible mechanisms of resistance and identify clinical indications for further development of this new agent.

Patients and Samples

A total of 90 tumors were tested with XR5944, with local ethics committee approval for the use of tissue or cells not required for diagnosis. The median age of the patients was 59 years (range 35–88; 17 male, 73 female; Table 1). The ovarian cancer patients subgroup were all treated previously with carboplatin alone or with carboplatin plus taxanes first-line followed in four cases by an anthracycline-containing regimen and in two cases by etoposide. Briefly, tumor tissue that was not required for diagnosis was taken by a histopathologist under sterile conditions, with patient consent, and transported to the laboratory in DMEM (Sigma Chemical Co., Poole, United Kingdom) with antibiotics (100 units/mL penicillin and 100 μg/mL streptomycin, Sigma Chemical) at 4°C.

Table 1.

Tumors tested with XR5944 showing the average age (range), sex ratio, and previous treatment

Tumor TypenAge (y)SexPrevious Treatment
Ovarian carcinoma 44 58 (38–78) 0 male, 44 female Primary: platinum + taxane (n = 27), platinum (n = 17); second line: anthracycline, etoposide, carboplatin + gemcitabine 
Cutaneous melanoma 14 58 (35–78) 7 male, 7 female Primary: melphalan (n = 3), vindesine (n = 2), cyclophosphamide (n = 1) 
Breast carcinoma 57 (39–61) 0 male, 5 female Primary: cyclophosphamide + anthracycline (n = 3), antihormonal agent (n = 2) 
Colorectal carcinoma 71 (39–88) 4 male, 5 female Primary: irinotecan (n = 1) 
Esophageal carcinoma 55 (52–72) 3 male, 0 female Primary: epirubicin + cisplatin + 5-fluorouracil (n = 2) 
Endometrial carcinoma 60 (55–66) 0 male, 2 female Primary: carboplatin (n = 2); second line: paclitaxel (n = 1) 
Lung (non–small cell lung carcinoma) 58 0 male, 1 female Primary: cisplatin + vinorelbine (n = 1) 
Sarcoma 70 (57–82) 0 male, 2 female None 
Unknown primary 10 63 (45–78) 3 male, 7 female Primary: carboplatin (n = 1) 
Total 90 59 (35–88) 17 male, 73 female  
Tumor TypenAge (y)SexPrevious Treatment
Ovarian carcinoma 44 58 (38–78) 0 male, 44 female Primary: platinum + taxane (n = 27), platinum (n = 17); second line: anthracycline, etoposide, carboplatin + gemcitabine 
Cutaneous melanoma 14 58 (35–78) 7 male, 7 female Primary: melphalan (n = 3), vindesine (n = 2), cyclophosphamide (n = 1) 
Breast carcinoma 57 (39–61) 0 male, 5 female Primary: cyclophosphamide + anthracycline (n = 3), antihormonal agent (n = 2) 
Colorectal carcinoma 71 (39–88) 4 male, 5 female Primary: irinotecan (n = 1) 
Esophageal carcinoma 55 (52–72) 3 male, 0 female Primary: epirubicin + cisplatin + 5-fluorouracil (n = 2) 
Endometrial carcinoma 60 (55–66) 0 male, 2 female Primary: carboplatin (n = 2); second line: paclitaxel (n = 1) 
Lung (non–small cell lung carcinoma) 58 0 male, 1 female Primary: cisplatin + vinorelbine (n = 1) 
Sarcoma 70 (57–82) 0 male, 2 female None 
Unknown primary 10 63 (45–78) 3 male, 7 female Primary: carboplatin (n = 1) 
Total 90 59 (35–88) 17 male, 73 female  

Cell Lines

CCRF-CEM lines were grown in RPMI 1640 (Sigma Chemical) supplemented with 2 mmol/L l-glutamine (Sigma Chemical) and 10% fetal bovine serum (Labtech International, Ringmer, United Kingdom). A resistant subline was obtained by stepwise treatment with XR5944 up to 10 nmol/L.

Drugs

XR5944 bis-mesylate salt was supplied by Xenova Ltd. (Slough, United Kingdom) as powder. It was dissolved in DMSO to give a stock solution of 1 mg/mL and aliquots were stored at −20°C. Cisplatin, treosulfan, vinorelbine (Navelbine), paclitaxel (Taxol), doxorubicin, etoposide (Vepesid), and topotecan (Hycamptin) were obtained from the pharmacy at Queen Alexandra Hospital. Cisplatin and paclitaxel were stored at room temperature, vinorelbine was kept in the refrigerator, and all other drugs were stored at −20°C as reported previously (17). Not all samples were tested with all drugs.

ATP-Tumor Chemosensitivity Assay

Cells were obtained from solid tumors by gentle enzymatic dissociation, usually 0.75 mg/mL collagenase (Sigma Chemical), overnight. Viable tumor-derived cells were separated from dead cells and debris by density centrifugation (Histopaque 1077-1, Sigma Chemical), washed, counted, and resuspended to 100,000 cells/mL in case of ascitic specimens or 200,000 cells/mL for solid biopsies. The cells were used to set up ATP-TCA plates according to Andreotti et al. (12). Briefly, cells were placed in 96-well polypropylene microplates (Corning-Costar, High Wycombe, United Kingdom) at 10,000 to 20,000 cells per well with each drug/combination at six doubling dilutions in triplicate from 200% to 6.25% test drug concentration. Test drug concentrations were 72.5 nmol/L for XR5944, 10.0 μmol/L for cisplatin, 71.9 μmol/L for treosulfan, 11.1 μmol/L for vinorelbine, 15.9 μmol/L for paclitaxel, 2.5 μmol/L for doxorubicin, 81.6 μmol/L for etoposide, and 1.64 μmol/L for topotecan. The plates were then incubated at 37°C in 5% CO2 for 6 days. The degree of cell inhibition at the end of this period was assessed by measurement of the remaining ATP in comparison with negative control (no drug, MO) and positive control (maximum inhibitor, MI) rows of 12 wells each. ATP was extracted from the cells and measured by light output in a microplate luminometer (Berthold Detection Systems GmbH, Pforzheim, Germany) following addition of luciferin-luciferase.

Immunohistochemistry

Of the 90 tumor samples studied, material for immunohistochemistry was available for 32 tumors (12 skin melanomas, 10 ovarian carcinomas, 3 colorectal carcinomas, 3 esophageal carcinomas, 2 unknown primaries, 1 breast carcinoma, and 1 sarcoma). The monoclonal antibodies P-gp (NCL-JSB1), topoisomerase I clone 1D6 (NCL-TOPO1), and topoisomerase IIα clone 3F6 (NCL-TOPO2A) from Novacastra Laboratories Ltd. (Newcastle upon Tyne, United Kingdom) were detected using the Vectastain Universal Alkaline Phosphatase kit (Vector Laboratories Ltd., Peterborough, United Kingdom). High-temperature antigen retrieval (pressure cooking) using sodium citrate buffer (pH 6.0) for 2 minutes was used to reveal the antigen-presenting sites blocked by formalin fixation (18). Following incubation in the primary antibody and rinsing in TBS, the slides were incubated for 30 minutes with diluted biotinylated universal secondary solution, rinsed, and incubated for 30 minutes with Vectastain ABC-AP reagent. To visualize the reaction, the slides were incubated for 20 minutes in Vector Red Alkaline Phosphate Substrate kit. The slides were counterstained with Gill's hematoxylin, dehydrated, and cleared using the Leica XL slide staining machine [Leica Microsystems (UK) Ltd., Milton Keynes, United Kingdom]. The concentration of each antibody was determined by titration on positive control material and was made up to its optimal dilution in TBS (pH 7.6). A positive control section was run with each batch of staining (renal proximal tubules for P-gp; tonsil or appendix for topoisomerases). A duplicate of each test section was included as a negative control by omitting the antibody and replacing with TBS.

Quantitative Reverse Transcription-PCR

CCRF-CEM lines were harvested by centrifugation at 400 × g and washed with PBS. Total RNA was extracted from at least 107 cells with a commercially available kit (NucleoSpin RNA II Mini, Macherey-Nagel, Germany) according to the manufacturer's instructions. The protocol included a DNase digestion step to prevent carryover of genomic DNA in further analysis. Reverse-transcribed RNA (Promega, Southampton, United Kingdom) was amplified by real-time quantitative PCR on an iCycler instrument (Bio-Rad Laboratories, Hemel Hampstead, United Kingdom). As transcript profiling of XR5944-treated cells previously showed up-regulated expression of RNA polymerase subunits as well as genes involved in rRNA processing (19), we have used HPRT1, PBGD, and TBP as housekeeping genes and not the more commonly employed 18S. The internal reference genes were selected due to their low abundance in normal tissue and the lack of pseudogenes (20). The housekeeping genes were amplified parallel to the target genes BCRP, MDR1, MRP1, MRP2, MVP, TOPO I, TOPO IIα, and TOPO IIβ in separate vessels. The primer sequences were as follows: HPRT1 (NM000194) 5′-TCAGGCAGTATAATCCAAAGATGGT-3′, 5′-AGTCTGGCTTATATCCAACACTTCG-3′; PBGD (NM000190) 5′-CTGCACGATCCCGAGACTCT-3′, 5′-GCTGTATGCACGGCTACTGG-3′; TBP (X54993) 5′-CACGAACCACGGCACTGATT-3′,5′-TTTTCTTGCTGCCAGTCTGGAC-3′; BCRP (AF098951) 5′-CACAACCATTGCATCTTGGC-3′, 5′-GCTGCAAAGCCGTAAATCCA-3′; MDR1 (AF016535) 5′-TGGTTCAGGTGGCTCTGGAT-3′, 5′-CTGTAGACAAACGATGAGCTATCACA-3′; MRP1 (L05628) 5′-CAATGCTGTGATGGCGATG-3′, 5′-GATCCGATTGTCTTTGCTCTTCA-3′; MRP2 (NM000392) 5′-TGCAGCCTCCATAACCATGAG-3′, 5′-GATGCCTGCCATTGGACCTA-3′; MVP (NM017458) 5′-CAGCTGGCCATCGAGATCA-3′, 5′-TCCAGTCTCTGAGCCTCATGC-3′; TOPO I (J03250) 5′-CTCCACAACGATTCCCAGAT-3′, 5′-TTATGTTCACTGTTGCTATGCTT-3′; TOPO IIα (NM001067) 5′-GTAATTTTGATGTCCCTCCACGA-3′, 5′-TCAAGGTCTGACACGACACTT-3′; and TOPO IIβ (NM001068) 5′-GCAGCCGAAAGACCTAAATA-3′, 5′-AATCATTATTGTCATCATCATCATC-3′. The constituents of each PCR reaction (25 μL) were 1 μL cDNA (or H2O); 400 nmol/L of each primer; 200 μmol/L of each dATP, dCTP, and dGTP; 400 μmol/L dUTP; 4.0 mmol/L MgCl2; 0.125 units AMPErase UNG; 0.625 units AmpliTaq Gold DNA polymerase; and 1× SYBR Green PCR buffer (all reagents were from Applied Biosystems, Foster City, CA). Product amplification was done up to 45 PCR cycles after uracil removal (2 minutes at 50°C) and polymerase activation (10 minutes at 95°C). Each two-step PCR cycle comprised denaturing (15 seconds at 95°C), annealing, and extending (1 minute at 60°C). At the end of each run, a final melt curve cycle (cooling to 50°C and increasing stepwise to 95°C) was done to exclude the presence of primer-dimer artifacts. A positive control (pooled cDNA from a variety of human tumors, including breast, ovarian, colorectal, and esophageal carcinoma) and negative controls with no template and reverse transcriptase negative as template were added in every experiment. All assays were run in triplicate. Validation experiments were run to show that the efficiencies of the target and reference gene amplifications were approximately equal. The PCR cycle number that generated the first fluorescence signal above a threshold (threshold cycle, Ct; 10 SD above the mean fluorescence generated during the baseline cycles) was determined, and a comparative Ct method was then used to measure relative gene expression (21). The following formula was used to calculate the relative amount of the transcript in the sample: 2−ΔΔCt, where ΔCt is the difference in Ct between the gene of interest and the mean of the three reference genes, and ΔΔCt = ΔCt of the parental CCRF-CEM line − ΔCt of XR5944-resistant subline.

Data Analysis

The percentage inhibition for each drug concentration was calculated as 1 − [(Test − MI) / (MO − MI)] × 100 using an Excel 2000 spreadsheet (Microsoft). For each drug-concentration curve, the IC50 and IC90 were calculated as described previously (13). Assessment of slides was done using the Hscore. Staining intensity (none, 0 points; weak, 1 point; moderate, 2 points; strong, 3 points) and percentage of positive tumor cells were multiplied to achieve a score between 0 and 300. A H score of ≥100 was regarded as positive. The correlation coefficients were calculated by the method of the least squares, and the correlation between IC90 and IC50 values and immunohistochemistry indices was assessed using univariate linear regression (StatsDirect). The calculated and descriptive data were entered into an Access 2000 database (Microsoft) and analyzed using a Wilcoxon two-tailed paired rank sum test or the Mann-Whitney U test for unpaired data, as appropriate (StatsDirect).

Activity of XR5944 against CCRF-CEM Cell Lines

Using the ATP-TCA, the IC50 values for XR5944 were 0.27 ± 0.05 and 4.24 ± 0.25 nmol/L in the CCRF-CEM parental and XR5944-resistant sublines, respectively.

The quantitative PCR technique produced extremely reproducible results with an intra-assay and interassay coefficient of variation of <1.5% and <5%, respectively. The XR5944 line showed 10- and >100-fold increase in mRNA levels of major vault protein and MDR1 (Fig. 1), respectively, when compared with the parental line. Quantitative PCR also showed a modest increase in the expression of MRP1 (2−ΔΔCt = 2.0). The levels of BCRP, MRP2, and topoisomerase isoforms were essentially unchanged in the resistant subline.

Figure 1.

Difference in gene expression between CCRF-CEM parental and XR5944-treated sublines determined by quantitative reverse transcription-PCR.

Figure 1.

Difference in gene expression between CCRF-CEM parental and XR5944-treated sublines determined by quantitative reverse transcription-PCR.

Close modal

Activity of XR5944 against Tumor-Derived Cells

Each of the agents tested produced a wide range of tumor inhibition, as can be seen in Table 2. This is consistent with considerable heterogeneity of chemosensitivity as published previously (22–24). XR5944 showed a steep concentration response curve in most tumors. Overall, XR5944 had median IC50 and IC90 values of 26 and 68 nmol/L, respectively (Table 2). The IC90 values well correlated with the IC50 values (r = 0.8055; P < 0.0001; nonparametric Spearman correlation). In all samples, XR5944 was more potent than other common chemotherapeutic agents currently used in the clinic (Table 2; Fig. 2). The IC90 of XR5944 was above the 1.0 μmol/L threshold in only 1 of 90samples (Table 2): this ovarian tumor was relatively refractory to chemotherapy, as shown also by the extremely high IC90 and IC50 values for doxorubicin (53.8 and 29.9 μmol/L, respectively; Table 2).

Table 2.

Median IC 90 and IC 50 values (range) for the cytotoxics tested in the ATP-TCA

DrugnIC 50 (μmol/L)IC 90 (μmol/L)
Ovarian carcinoma    
 XR5944 (MLN944) 44 0.023 (0.003–0.253) 0.060 (0.016–1.315) 
 Cisplatin 44 13.4 (0.4–80.4) 25.9 (7.6–144.8) 
 Treosulfan 42 39.5 (4.2–390.8) 129.8 (23.9–703.4) 
 Etoposide 43 73.4 (3.3–932.1) 176.1 (16.8–1678.3) 
 Topotecan 43 0.98 (0.07–10.02) 3.60 (0.29–18.04) 
 Doxorubicin 43 1.49 (0.17–29.88) 2.56 (0.45–53.78) 
 Paclitaxel 43 8.78 (0.55–32.79) 26.59 (3.50–59.03) 
 Vinorelbine 21 2.32 (0.41–14.35) 12.59 (7.37–30.91) 
Skin melanoma    
 XR5944 (MLN944) 14 0.031 (0.002–0.064) 0.071 (0.015–0.157) 
 Cisplatin 13 8.8 (0.9–32.6) 23.2 (14.9–58.6) 
 Treosulfan 14 37.0 (2.9–92.8) 108.7 (25.2–175.4) 
 Etoposide — N/A* N/A 
 Topotecan — N/A N/A 
 Doxorubicin 0.29 (0.10–1.06) 1.02 (0.30–2.35) 
 Paclitaxel 14 9.07 (0.62–27.26) 18.28 (5.73–35.61) 
 Vinorelbine 14 1.49 (0.44–34.49) 18.59 (4.67–62.08) 
Unknown primary    
 XR5944 (MLN944) 10 0.029 (0.007–0.054) 0.066 (0.029–0.125) 
 Cisplatin 10 12.1 (1.5–38.4) 23.5 (21.3–69.2) 
 Treosulfan 55.0 (3.6–94.9) 129.4 (82.7–154.8) 
 Etoposide 58.1 (3.3–112.9) 191.2 (172.9–209.6) 
 Topotecan 0.54 (0.07–2.48) 3.26 (1.30–4.35) 
 Doxorubicin 1.20 (0.61–3.76) 2.49 (1.41–7.55) 
 Paclitaxel 10 6.77 (0.80–25.63) 19.04 (10.98–42.03) 
 Vinorelbine 10 1.67 (0.52–10.79) 15.11 (9.79–21.02) 
Colorectal carcinoma    
 XR5944 (MLN944) 0.126 (0.009–0.221) 0.249 (0.160–0.399) 
 Cisplatin 19.6 (14.0–23.1) 35.8 (26.4–41.6) 
 Treosulfan 60.4 (18.0–61.1) 142.4 (132.3–146.0) 
 Etoposide 71.8 203.9 
 Topotecan 1.21 4.62 (4.03–5.20) 
 Doxorubicin 3.10 (2.87–45.74) 7.89 (5.58–82.32) 
 Paclitaxel 9.71 (3.82–18.79) 20.54 (14.49–29.93) 
 Vinorelbine 5.34 (1.00–13.34) 19.46 (10.01–28.69) 
Breast carcinoma    
 XR5944 (MLN944) 0.023 (0.004–0.076) 0.055 (0.036–0.164) 
 Cisplatin 21.2 (6.3–53.2) 37.0 (24.2–95.8) 
 Treosulfan 28.4 (4.1–86.3) 105.8 (87.0–177.1) 
 Etoposide — N/A N/A 
 Topotecan 1.51 (1.45–1.57) 3.82 (3.76–3.88) 
 Doxorubicin 0.84 (0.19–1.81) 2.22 (0.52–15.86) 
 Paclitaxel 7.82 (5.81–17.19) 30.12 (20.69–43.78) 
 Vinorelbine 7.93 (0.45–11.01) 18.60 (3.91–20.79) 
Esophageal carcinoma    
 XR5944 (MLN944) 0.066 (0.017–0.097) 0.178 (0.070–0.197) 
 Cisplatin 46.8 (13.3–80.4) 84.9 (25.1–144.8) 
 Treosulfan 31.3 (16.5–46.1) 93.7 (60.4–127.0) 
 Etoposide — N/A N/A 
 Topotecan 1.98 3.93 
 Doxorubicin 2.79 (2.27–3.31) 6.83 (5.77–7.89) 
 Paclitaxel 2.71 14.65 
 Vinorelbine 0.89 15.68 
Sarcoma    
 XR5944 (MLN944) 0.034 (0.012–0.057) 0.115 (0.110–0.120) 
 Cisplatin 13.0 24.5 
 Treosulfan 28.8 62.6 
 Etoposide — N/A N/A 
 Topotecan — N/A N/A 
 oxorubicin 3.90 (0.25–7.55) 7.20 (0.80–13.60) 
 Paclitaxel 13.02 (9.33–16.72) 22.05 (14.65–29.45) 
 Vinorelbine 0.44 3.11 
Endometrial carcinoma    
 XR5944 (MLN944) 0.025 (0.024–0.026) 0.097 (0.044–0.150) 
 Cisplatin 7.7 (5.6–9.8) 30.3 (28.9–31.7) 
 Treosulfan 28.6 (8.9–48.3) 141.4 (116.4–166.4) 
 Etoposide 163.6 (143.9–183.3) 301.2 (272.5–329.9) 
 Topotecan 0.89 (0.80–0.98) 3.58 (3.56–3.59) 
 Doxorubicin 0.73 (0.44–1.02) 6.31 (2.00–10.62) 
 Paclitaxel 10.46 (9.02–11.91) 28.61 (27.61–29.61) 
 Vinorelbine — N/A N/A 
Lung (non–small cell lung carcinoma)    
 XR5944 (MLN944) 0.017 0.033 
 Cisplatin 30.9 55.7 
 Treosulfan — N/A N/A 
 Etoposide 141.6 169.6 
 Topotecan — N/A N/A 
 Doxorubicin 1.79 3.46 
 Paclitaxel — N/A N/A 
 Vinorelbine 0.37 0.67 
Total    
 XR5944 (MLN944) 90 0.026 (0.002–0.253) 0.068 (0.015–1.315) 
 Cisplatin 81 13.4 (0.4–80.4) 25.9 (7.6–144.8) 
 Treosulfan 77 39.6 (2.9–390.8) 116.4 (23.9–703.4) 
 Etoposide 49 74.2 (3.3–932.1) 177.1 (16.8–1678.3) 
 Topotecan 53 1.00 (0.07–10.02) 3.62 (0.29–18.04) 
 Doxorubicin 74 1.20 (0.10–45.74) 2.49 (0.30–82.32) 
 Paclitaxel 84 8.99 (0.55–32.79) 25.39 (3.50–59.03) 
 Vinorelbine 58 2.17 (0.37–34.49) 16.85 (0.67–62.08) 
DrugnIC 50 (μmol/L)IC 90 (μmol/L)
Ovarian carcinoma    
 XR5944 (MLN944) 44 0.023 (0.003–0.253) 0.060 (0.016–1.315) 
 Cisplatin 44 13.4 (0.4–80.4) 25.9 (7.6–144.8) 
 Treosulfan 42 39.5 (4.2–390.8) 129.8 (23.9–703.4) 
 Etoposide 43 73.4 (3.3–932.1) 176.1 (16.8–1678.3) 
 Topotecan 43 0.98 (0.07–10.02) 3.60 (0.29–18.04) 
 Doxorubicin 43 1.49 (0.17–29.88) 2.56 (0.45–53.78) 
 Paclitaxel 43 8.78 (0.55–32.79) 26.59 (3.50–59.03) 
 Vinorelbine 21 2.32 (0.41–14.35) 12.59 (7.37–30.91) 
Skin melanoma    
 XR5944 (MLN944) 14 0.031 (0.002–0.064) 0.071 (0.015–0.157) 
 Cisplatin 13 8.8 (0.9–32.6) 23.2 (14.9–58.6) 
 Treosulfan 14 37.0 (2.9–92.8) 108.7 (25.2–175.4) 
 Etoposide — N/A* N/A 
 Topotecan — N/A N/A 
 Doxorubicin 0.29 (0.10–1.06) 1.02 (0.30–2.35) 
 Paclitaxel 14 9.07 (0.62–27.26) 18.28 (5.73–35.61) 
 Vinorelbine 14 1.49 (0.44–34.49) 18.59 (4.67–62.08) 
Unknown primary    
 XR5944 (MLN944) 10 0.029 (0.007–0.054) 0.066 (0.029–0.125) 
 Cisplatin 10 12.1 (1.5–38.4) 23.5 (21.3–69.2) 
 Treosulfan 55.0 (3.6–94.9) 129.4 (82.7–154.8) 
 Etoposide 58.1 (3.3–112.9) 191.2 (172.9–209.6) 
 Topotecan 0.54 (0.07–2.48) 3.26 (1.30–4.35) 
 Doxorubicin 1.20 (0.61–3.76) 2.49 (1.41–7.55) 
 Paclitaxel 10 6.77 (0.80–25.63) 19.04 (10.98–42.03) 
 Vinorelbine 10 1.67 (0.52–10.79) 15.11 (9.79–21.02) 
Colorectal carcinoma    
 XR5944 (MLN944) 0.126 (0.009–0.221) 0.249 (0.160–0.399) 
 Cisplatin 19.6 (14.0–23.1) 35.8 (26.4–41.6) 
 Treosulfan 60.4 (18.0–61.1) 142.4 (132.3–146.0) 
 Etoposide 71.8 203.9 
 Topotecan 1.21 4.62 (4.03–5.20) 
 Doxorubicin 3.10 (2.87–45.74) 7.89 (5.58–82.32) 
 Paclitaxel 9.71 (3.82–18.79) 20.54 (14.49–29.93) 
 Vinorelbine 5.34 (1.00–13.34) 19.46 (10.01–28.69) 
Breast carcinoma    
 XR5944 (MLN944) 0.023 (0.004–0.076) 0.055 (0.036–0.164) 
 Cisplatin 21.2 (6.3–53.2) 37.0 (24.2–95.8) 
 Treosulfan 28.4 (4.1–86.3) 105.8 (87.0–177.1) 
 Etoposide — N/A N/A 
 Topotecan 1.51 (1.45–1.57) 3.82 (3.76–3.88) 
 Doxorubicin 0.84 (0.19–1.81) 2.22 (0.52–15.86) 
 Paclitaxel 7.82 (5.81–17.19) 30.12 (20.69–43.78) 
 Vinorelbine 7.93 (0.45–11.01) 18.60 (3.91–20.79) 
Esophageal carcinoma    
 XR5944 (MLN944) 0.066 (0.017–0.097) 0.178 (0.070–0.197) 
 Cisplatin 46.8 (13.3–80.4) 84.9 (25.1–144.8) 
 Treosulfan 31.3 (16.5–46.1) 93.7 (60.4–127.0) 
 Etoposide — N/A N/A 
 Topotecan 1.98 3.93 
 Doxorubicin 2.79 (2.27–3.31) 6.83 (5.77–7.89) 
 Paclitaxel 2.71 14.65 
 Vinorelbine 0.89 15.68 
Sarcoma    
 XR5944 (MLN944) 0.034 (0.012–0.057) 0.115 (0.110–0.120) 
 Cisplatin 13.0 24.5 
 Treosulfan 28.8 62.6 
 Etoposide — N/A N/A 
 Topotecan — N/A N/A 
 oxorubicin 3.90 (0.25–7.55) 7.20 (0.80–13.60) 
 Paclitaxel 13.02 (9.33–16.72) 22.05 (14.65–29.45) 
 Vinorelbine 0.44 3.11 
Endometrial carcinoma    
 XR5944 (MLN944) 0.025 (0.024–0.026) 0.097 (0.044–0.150) 
 Cisplatin 7.7 (5.6–9.8) 30.3 (28.9–31.7) 
 Treosulfan 28.6 (8.9–48.3) 141.4 (116.4–166.4) 
 Etoposide 163.6 (143.9–183.3) 301.2 (272.5–329.9) 
 Topotecan 0.89 (0.80–0.98) 3.58 (3.56–3.59) 
 Doxorubicin 0.73 (0.44–1.02) 6.31 (2.00–10.62) 
 Paclitaxel 10.46 (9.02–11.91) 28.61 (27.61–29.61) 
 Vinorelbine — N/A N/A 
Lung (non–small cell lung carcinoma)    
 XR5944 (MLN944) 0.017 0.033 
 Cisplatin 30.9 55.7 
 Treosulfan — N/A N/A 
 Etoposide 141.6 169.6 
 Topotecan — N/A N/A 
 Doxorubicin 1.79 3.46 
 Paclitaxel — N/A N/A 
 Vinorelbine 0.37 0.67 
Total    
 XR5944 (MLN944) 90 0.026 (0.002–0.253) 0.068 (0.015–1.315) 
 Cisplatin 81 13.4 (0.4–80.4) 25.9 (7.6–144.8) 
 Treosulfan 77 39.6 (2.9–390.8) 116.4 (23.9–703.4) 
 Etoposide 49 74.2 (3.3–932.1) 177.1 (16.8–1678.3) 
 Topotecan 53 1.00 (0.07–10.02) 3.62 (0.29–18.04) 
 Doxorubicin 74 1.20 (0.10–45.74) 2.49 (0.30–82.32) 
 Paclitaxel 84 8.99 (0.55–32.79) 25.39 (3.50–59.03) 
 Vinorelbine 58 2.17 (0.37–34.49) 16.85 (0.67–62.08) 
*

N/A, not assessed.

Figure 2.

Example of ATP-TCA results for an ovarian adenocarcinoma.

Figure 2.

Example of ATP-TCA results for an ovarian adenocarcinoma.

Close modal

XR5944 showed its activity on a wide variety of different tumors. Unknown primaries, breast and gynecologic malignancies have proven highly sensitive to this new agent. In particular, 37 of 44 (84%) ovarian cancer samples had an IC50 value below 50 nmol/L and 33 of 44 (75%) had an IC90 below 100 nmol/L. When compared with other agents used in second line for ovarian cancer, XR5944 was at least 40-fold more potent than doxorubicin or topotecan. We observed the best median activity in the small group of breast tumors tested (n = 5), in which XR5944 had an IC50 of 23 nmol/L and an IC90 of 55 nmol/L. In this tumor type, XR5944 had an IC50 value 340-fold better than paclitaxel. XR5944 retained excellent activity in skin melanoma, a tumor that is usually little sensitive to chemotherapy. Eleven of 14 (79%) melanoma samples had an IC50 value below 50 nmol/L and 10 of 14 (71%) had an IC90 value below 100 nmol/L. XR5944 had slightly decreased potency in colorectal and esophageal samples, in which the median IC50 values were 126 and 66 nmol/L, respectively. Only 2 of 9 (22%) colorectal samples had an IC50 value below 50 nmol/L and none (0%) had IC90 values below 100 nmol/L. However, in colorectal samples, we recorded the highest median IC50 and IC90 values for most drugs, with the exception of paclitaxel. Paclitaxel was least active in the ovarian cancer subgroup probably due the high prevalence of patients pretreated with a taxane-based regimen (Table 1).

Correlation of XR5944 Activity with P-gp Immunohistochemistry

Immunostaining for P-gp was positive in 14 of 32 (44%) samples tested with XR5944. The P-gp-expressing samples consisted of five ovarian tumors, three skin melanomas, three esophageal cancers, two colorectal tumors, and one unknown primary carcinoma. The median XR5944 IC50 values for P-gp negative and positive samples were 23 and 58 nmol/L, respectively (P < 0.05, Mann-Whitney U test, P = 0.0143), whereas the median IC90 values were 68 and 142 nmol/L, respectively (P < 0.05, Mann-Whitney U test, P = 0.0113). The expression of P-gp shifted to the right of the drug-response curve of XR5944 (Fig. 3A). A positive correlation by linear regression analysis was found between P-gp staining and the IC50 of XR5944 (r = 0.65; P < 0.0001) and the IC90 (r = 0.54; P < 0.005; Fig. 3B) but not with either topoisomerase Iα or IIα immunohistochemistry index (not significant; data not shown).

Figure 3.

A, activity of XR5944 in P-gp-negative (n = 18) and P-gp-positive (n = 14) samples. Points, mean; bars, SE. B, correlation between XR5944 activity (IC50) and intensity of P-gp immunostaining.

Figure 3.

A, activity of XR5944 in P-gp-negative (n = 18) and P-gp-positive (n = 14) samples. Points, mean; bars, SE. B, correlation between XR5944 activity (IC50) and intensity of P-gp immunostaining.

Close modal

The ex vivo profile of XR5944 in human solid tumors showed it to be an exceptionally potent cytotoxic agent. In all samples tested, XR5944 was at least 10-fold more active than all other compounds tested, including topotecan, doxorubicin, and paclitaxel (Table 2). In keeping with the results of many other drugs, the potency of XR5944 is attenuated in tumor-derived cells compared with cell lines, but this probably reflects the homogeneity and the fast-growing status of the CCRF-CEM cells used here and the cell lines used in previous studies (1). In addition, tumor-derived cells in the ATP-TCA were seeded in round-bottomed well polypropylene plates to create a spheroid-like culture system, which is more chemoresistant than cells assayed in monolayer (6). Moreover, the use of a serum-free, well-defined medium results in a preponderance of quiescent cells that more resembles the proliferating status of most solid tumors.

Although the expression levels of MDR1 were increased at least 100-fold after prolonged and gradual XR5944 exposure of the CCRF-CEM cells, the IC50 value in the resistant line showed a modest 15-fold increase when compared with the IC50 of the parental line. The relevance of this resistant model in the clinical setting may be questionable. In fact, it should be pointed that resistant lines generated in vitro after prolonged treatment might not reflect the in vivo situation, as factors that allow cell survival following acute cytotoxic drug exposure may differ from mechanisms selected for by chronic drug exposure. Further, a recent study (25) has shown that, in K562 myelogenous leukemic cells, only long-term, and not short-term, drug exposure was able to overcome a translational block so that MDR1 mRNA was translated and P-gp was overexpressed.

It is remarkable that XR5944 showed great activity in the subset of ovarian cancer samples. In fact, all the ovarian patients had relapsed after receiving a platinum-based regimen, which might have resulted in a gradual enrichment of resistant cells in those samples. For example, platinum treatment may cause up-regulation of drug detoxifying enzymes, such as the glutathione S-transferases (26), and loss of mismatch repair mechanisms in a proportion of patients (27), therefore rendering the cells more resistant to several chemotherapeutic agents (28). Our data suggest that XR5944 may not be sensitive to common resistance mechanisms in ovarian cancer.

The median IC90 and IC50 values of XR5944 in colon carcinoma samples were 4-fold greater compared with the median of all tumor samples. All but one of these patients were chemotherapy naive, excluding the possibility that previous treatment could have altered tumor response in these cases. On the other hand, colorectal carcinoma is well known to overexpress MDR1 and other pump proteins and is generally refractory to chemotherapy. It should be noted that, even in those samples overexpressing P-gp, XR5944 showed better activity than the other cytotoxic agents tested.

Although the expression of P-gp could play a role in conferring resistance to XR5944, resistance is very often multifactorial and other mechanisms may also be important. Our data on the CCRF-CEM and other cell lines (1) suggest that other MDR transporters, such as major vault protein and MRP1, may mediate chemoresistance to this new agent. No immunostaining for these proteins was done in this study, and further investigations may be needed to evaluate the importance of these newer pumps in the clinical setting.

In vitro and in vivo reports have shown levels of topoisomerase IIα to correlate with the activity of topoisomerase inhibitors as doxorubicin and etoposide (29, 30). Our data showed that XR5944 activity does not correlate with the expression of either topoisomerase I or IIα and that the levels of these enzymes are unaltered in a drug-selected leukemia cell line. Although these findings could be explained by inhibition of both enzymes, it is probable that XR5944 acts through an alternative mechanism of action, distinct from topoisomerase inhibition, as more recently suggested (4).

Multiple resistance mechanisms (including those not mentioned here or yet undiscovered) may explain the heterogeneity of chemosensitivity between tumor types and individual tumors within the same tumor type, which has been observed in our series. However, in this study, the cell line data, the decreased potency in gastrointestinal tumors, and the better correlation of P-gp with the IC50 of XR5944 than with the IC90 all suggested that P-gp may be a mechanism of resistance to very low concentrations of XR5944. The results of this study may help further clinical development of XR5944. If phase I studies prove the safety of this compound, then our results may be useful for the design of phase II trials.

Grant support: Xenova Ltd. UK.

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.

Notes:Possible conflicts of interest (including financial and other relationships) for each author include the following: Dr. D. Norris (Xenova employee); Dr. P.A. Charlton (Xenova employee until September 2003); and I.A. Cree (director of Cantech Ltd.; received funding from Xenova).

We thank Lisa Mills, Penny Johnson, and Alison Parker for assistance with immunohistochemistry and all the oncologists and surgeons who submitted material for testing, particularly Drs. Lamont, Hindley, Osborne, Allerton, Khoury, and Weaver.

1
Stewart AJ, Mistry P, Dangerfield W, et al. Antitumor activity of XR5944, a novel and potent topoisomerase poison.
Anticancer Drugs
2001
;
12
:
359
–67.
2
Gamage SA, Spicer JA, Finlay GJ, et al. Dicationic bis(9-methylphenazine-1-carboxamides): relationships between biological activity and linker chain structure for a series of potent topoisomerase-targeted anticancer drugs.
J Med Chem
2001
;
44
:
1407
–15.
3
Jobson A, Willmore E, Tilby M, et al. Characterisation of the roles of topoisomerase I and II in the mechanism of action of novel anti-tumor agents XR11576 (MLN576) and XR5944 (MLN944) [abstract 86].
Eur J Cancer
2002
;
38
:
31
.
4
Sappal DS, McClendon AK, Fleming JA, et al. Biological characterization of MLN944: a potent DNA binding agent.
Mol Cancer Ther
2004
;
3
:
47
–58.
5
Phillips RM, Bibby MC, Double JA. A critical appraisal of the predictive value of in vitro chemosensitivity assays.
J Natl Cancer Inst
1990
;
82
:
1457
–68.
6
Andreotti PE, Linder D, Hartmann DM, Cree IA, Pazzagli M, Bruckner HW. TCA-100 tumor chemosensitivity assay: differences in sensitivity between cultured tumor cell lines and clinical studies.
J Biolumin Chemilumin
1994
;
9
:
373
–8.
7
Drewinko B, Patchen M, Yang LY, Barlogie B. Differential killing efficacy of twenty antitumor drugs on proliferating and nonproliferating human tumor cells.
Cancer Res
1981
;
41
:
2328
–33.
8
Dewys WD. A quantitative model for the study of the growth and treatment of a tumor and its metastases with correlation between proliferative state and sensitivity to cyclophosphamide.
Cancer Res
1972
;
32
:
367
–73.
9
Valeriote F, van Putten L. Proliferation-dependent cytotoxicity of anticancer agents: a review.
Cancer Res
1975
;
35
:
2619
–30.
10
Cooper M, Rankin E, Bissett D, et al. Initial phase I study of XR5944, a novel DNA and RNA targeting agent [abstract 2100].
Proc Am Assoc Clin Oncol
2004
;
23
.
11
Cree IA, Kurbacher CM. ATP based tumor chemosensitivity testing: assisting new agent development.
Anticancer Drugs
1999
;
10
:
431
–5.
12
Andreotti PE, Cree IA, Kurbacher CM, et al. Chemosensitivity testing of human tumors using a microplate adenosine triphosphate luminescence assay: clinical correlation for cisplatin resistance of ovarian carcinoma.
Cancer Res
1995
;
55
:
5276
–82.
13
Cree IA, Kurbacher CM. Individualising chemotherapy for solid tumors—is there any alternative?
Anticancer Drugs
1997
;
8
:
541
–8.
14
Cree IA. Luminescence-based cell viability testing. In: LaRossa RA, editor. Bioluminescence methods and protocols. Methods in molecular biology. Vol. 102. Totowa, NJ: Humana Press; 1998. p. 169–77.
15
Neale MH, Charlton PA, Cree IA. Ex vivo activity of XR5000 against solid tumors.
Anticancer Drugs
2000
;
11
:
471
–8.
16
Cree IA. Chemosensitivity testing as an aid to anti-cancer drug and regimen development.
Recent Results Cancer Res
2003
;
161
:
119
–25.
17
Hunter EM, Sutherland LA, Cree IA, et al. The influence of storage on cytotoxic drug activity in an ATP-based chemosensitivity assay.
Anticancer Drugs
1994
;
5
:
171
–6.
18
Norton AJ, Jordan S, Yeomans P. Brief, high temperature heat denaturation (pressure cooking): a simple and effective method of antigen retrieval for routinely processed tissues.
J Pathol
1994
;
173
:
371
–9.
19
Fleming JA, Blackman RK, Thoroddsen V, et al. Using yeast to probe the mechanism of action of MLN944 (XR5944), a novel bis-phenazine with potent anti-tumor activity [abstract 6574].
Proc AACR
2003
;
44
.
20
Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes.
Genome Biol
2002
;
3
:
0034.1
–11.
21
ABI PRISM 7700. Relative quantitation of gene expression (P/N 4303859B).
Sequence Detect Syst User Bull
2001
;
2
:
11
–5.
22
Cree IA, Neale MH, Myatt NE, et al. Heterogeneity of chemosensitivity of metastatic cutaneous melanoma.
Anticancer Drugs
1999
;
10
:
437
–44.
23
Whitehouse PA, Knight LA, Di Nicolantonio F, et al. Heterogeneity of chemosensitivity of colorectal adenocarcinoma determined by a modified ex vivo ATP-tumor chemosensitivity assay (ATP-TCA).
Anticancer Drugs
2003
;
14
:
369
–75.
24
Mercer SJ, Somers SS, Knight LA, et al. Heterogeneity of chemosensitivity of esophageal and gastric carcinoma.
Anticancer Drugs
2003
;
14
:
397
–403.
25
Yague E, Armesilla AL, Harrison G, et al. P-glycoprotein (MDR1) expression in leukemic cells is regulated at two distinct steps, mRNA stabilization and translational initiation.
J Biol Chem
2003
;
278
:
10344
–52.
26
Cheng X, Kigawa J, Minagawa Y, et al. Glutathione S-transferase-π expression and glutathione concentration in ovarian carcinoma before and after chemotherapy.
Cancer
1997
;
79
:
521
–7.
27
Samimi G, Fink D, Varki NM, et al. Analysis of MLH1 and MSH2 expression in ovarian cancer before and after platinum drug-based chemotherapy.
Clin Cancer Res
2000
;
6
:
1415
–21.
28
Irving JAE, Hall AG. Mismatch repair defects as a cause of resistance to cytotoxic drugs.
Expert Rev Anticancer Ther
2001
;
1
:
149
–58.
29
Dingemans AMC, Witlox MA, Stallaert RALM, et al. Expression of DNA topoisomerase II and topoisomerase II genes predicts survival and response to chemotherapy in patients with small cell lung cancer.
Clin Cancer Res
1999
;
5
:
2048
–58.
30
Koshiyama M, Fujii H, Kinezaki M, et al. Immunohistochemical expression of topoisomerase IIα (Topo IIα) and multi-drug resistance-associated protein (MRP), plus chemosensitivity testing, as chemotherapeutic indices of ovarian and endometrial carcinomas.
Anticancer Res
2001
;
21
:
2925
–32.