Clinical Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Raymond, E.
Right arrow Articles by Von Hoff, D. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Raymond, E.
Right arrow Articles by Von Hoff, D. D.
Clinical Cancer Research Vol. 7, 1171-1180, May 2001
© 2001 American Association for Cancer Research

Clinical Trials

Effects of Hydroxyurea on Extrachromosomal DNA in Patients with Advanced Ovarian Carcinomas1

Eric Raymond2, Sandrine Faivre, Geoffrey Weiss, John McGill, Karen Davidson, Elzbieta Izbicka, John G. Kuhn, Craig Allred, Gary M. Clark and Daniel D. Von Hoff3

Institute for Drug Development-Cancer Therapy and Research Center, San Antonio, Texas 78245-3217


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Purpose: In vitro low concentrations of hydroxyurea eliminate double-minute chromosomes (dmins) containing amplified drug-resistance genes and oncogenes from cancer cells. This clinical trial investigated whether a noncytotoxic dose of oral hydroxyurea could reduce the number of dmins in cancer cells in patients with advanced ovarian carcinomas.

Experimental Design: The high frequency of ascites associated with ovarian cancer facilitated the monitoring of cytogenetic variations with minimal discomfort in patients who required frequent abdominal paracentesis. Sixteen patients with advanced ovarian carcinomas resistant to conventional cisplatin-based and/or paclitaxel chemotherapy and with ascites requiring frequent abdominal paracentesis were entered in this study. A course of treatment consisted of a single oral dose of 80 mg/kg hydroxyurea every 3 days for 6 weeks. Blood and i.p. levels of hydroxyurea were determined. We monitored the variations of dmins in tumor cells taken from serial abdominal paracenteses.

Results: The median number of courses administered to the patients was 1 (range, 1–9). In ascites, hydroxyurea concentrations were 610.3 ± 76.3, 219.8 ± 85.6, and 86.1 µmol/liter at 4, 24, and 30 h after oral administration, respectively. Eleven (78.6%) of 14 patient specimens contained dmins before therapy. The number of spreads with tumor cells containing dmins were reduced by more than 50% in 5 (45%) of 11 and 3 (60%) of 5 patients at the completion of the first and second course of chemotherapy, respectively. Using tumor cells taken directly from the patients and grown in soft agar, we documented that concentrations of hydroxyurea in ascites were too low to have any cytotoxic effects. No grade 3–4 hydroxyurea-related toxicities nor any objective responses were observed. However, despite the utilization of a low noncytotoxic dose of hydroxyurea, two patients had prolonged stabilization of their disease for 6 and 10 months, respectively, with concomitant decreases in the number of dmins that remained until progression.

Conclusions: This study showed that, in some circumstances, a noncytotoxic dose of hydroxyurea given to patients with ovarian cancer can decrease the number of metaphase spreads containing dmins in cancer cells.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Recent molecular cytogenetic techniques have permitted intensive exploration of cancer progression (1) . Early cytogenetic manifestations of gene amplification in human cancer cells are dmins4 and expanded HSRs (2 , 3) . Previous reports indicated that dmins are the predominant manifestation of gene amplification in tumor tissues taken directly from patients and can be found in 95% of cancers (4) including 88% of ovarian carcinomas (5) . It has been proposed that submicroscopic extrachromosomal circular structures called "episomes" are precursors of dmins (6 , 7) . Episomes are rapidly induced DNA molecules that multimerize over time (8 , 9) and that contain amplified drug-resistance genes (10, 11, 12, 13) and oncogenes (9 , 14 , 15) . The presence of amplified myc, Her-2/neu, mutated p53, and mdm2 are important factors in cancer progression (1 , 14 , 16) . Therefore, therapies that can eliminate episomes and dmins from cancer cells will likely affect genes that encode for aggressive phenotypes and drug resistance in advanced cancer.

Several previous studies have documented that amplified genes in dmins are unstable and could gradually be lost in the absence of selective pressure (17) . Furthermore, other studies have provided evidence that dmins can be eliminated in cultured cells exposed to drugs and that elimination of extrachromosomally amplified oncogenes from cancer cells reduces tumorigenicity (18, 19, 20, 21, 22) . Low concentrations of hydroxyurea demonstrated preclinical activity through eliminating dmins from human tumor cells in culture and in xenografts without cytotoxic effects and toxicity (19 , 20 , 22) . The precise mechanism for accelerated hydroxyurea-induced loss of extrachromosomal amplified genes remains unclear but does not seem to involve the inhibition of ribonucleotide reductase or DNA synthesis that accounts for hydroxyurea cytotoxicity (19) . Evidence does suggest that the mechanism by which hydroxyurea affects the loss of dmins may involve the preferential capture of dmins within micronuclei (20) .

Strategies to overcome drug resistance and/or delay tumor progression are urgently needed for patients with advanced ovarian cancer resistant to cisplatin- and/or paclitaxel-based chemotherapy (23) . On the basis of animal and human pharmacokinetic parameters of hydroxyurea (24, 25, 26, 27) , we conducted a pilot clinical trial to test the effects of administering low-dose hydroxyurea on dmins in cancer cells in patients with advanced ovarian carcinomas. Clinical evaluation of a noncytotoxic anticancer strategy that is not expected to induce tumor shrinkage remains a major problem. Usually, no reliable clinical end points such as the measurement of tumor volumes are plausible, and tumors are not easily accessible for biological studies. Moreover, there is still some concerns about the validity of CA125 as an alternative end point for studying the anticancer effects of new drugs in ovarian cancer. In this study, the high frequency of ascites associated with ovarian cancer offered the unique opportunity to implement the monitoring of cytogenetic variations with minimal discomfort in patients who required frequent abdominal paracentesis.

Our study is the first clinical attempt to eliminate dmins in patients with advanced cancers using a noncytotoxic strategy. The objectives were to determine whether hydroxyurea can accumulate in malignant ascites after oral administration and decrease the amount of extrachromosomal DNA in ovarian cancer cells. Despite the small number of patients included in this study, we showed that a noncytotoxic dose of hydroxyurea could reduce the number of dmins in cancer cells in some patients. This study suggests that this noncytotoxic strategy to eliminate dmins could be useful in delaying ovarian cancer progression.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Patient Eligibility.
Patients with histologically documented advanced ovarian carcinoma who were refractory to previous chemotherapy were candidates for this study. Eligibility criteria also included: (a) cytologically documented malignant ascites requiring frequent abdominal paracentesis for patient comfort; (b) objectively measurable disease evaluated by computed tomography scan and measuring at least 0.5 cm in diameter; (c) age >= 18 years; (d) WHO performance status <= 3 (capable of self-care); (e) life-expectancy >= 12 weeks (enabling the completion of at least two courses of treatment); (f) no surgery within 2 weeks or no radiotherapy and/or biological therapy within 4 weeks of entering the protocol; (g) adequate hematological [WBC count >= 4,000/µl; absolute neutrophil count (ANC) >= 1,500/µl; hemoglobin >= 9.0 g/dl; and platelet count >= 100,000/µl], hepatic (total bilirubin <= 2.0 mg/dl; aspartate aminotransferase <= 2 times ULN or <= 3 times ULN in the presence of liver metastases; or alkaline phosphatase <= 2 times ULN or <= 5 times ULN with liver or bony metastases), and renal (serum creatinine <= 1.5 mg/dl and a creatinine clearance estimated using the Cockcroft equation with correction for body surface area >= 60 ml/min) functions; (h) no concurrent medical problem unrelated to the malignancy and no coexisting infection that would significantly limit compliance with the study or expose the patient to undue risk; and (i) no brain or leptomeningeal metastases. Before treatment, all of the patients signed written Institutional Review Board- approved informed consent, indicating that they were aware of the investigational nature of the study, according to federal and institutional guidelines.

Treatment and Clinical Follow-Up.
Hydroxyurea was given p.o. as a single dose of 80 mg/kg every 3 days for 6 weeks. A course of treatment was defined as 6 weeks of treatment.

Before therapy, all of the patients had a complete history and physical examination. Resistance to cisplatin was defined according to Markman et al. (23) Patients were followed at least once weekly for signs and symptoms of toxicity. Toxicity was assessed according to the WHO criteria. Dose reduction was allowed in cases of grade 3–4 hematological toxicity. Assessment of antitumor response was performed at each treatment course by a physical examination, serum CA 125 determination, and computed tomography scan. A complete response was defined as the disappearance of all of the measurable lesions lasting at least 4 weeks with no new lesions. A partial response was defined as at least a 50% reduction in size of all of the measurable tumors as measured by the sum of all of the products of the greatest diameters, lasting at least 4 weeks without any appearance of a new lesion. Progressive disease was defined as an increase of more than 25% in the size of measurable tumors or the appearance of any new lesion. Stable disease was defined as a lesion that failed to qualify for response or progression.

Pharmacokinetic Analysis.
Prior to drug administration, an indwelling heparin lock was placed in one arm for serial blood specimen collection. Blood samples were collected in heparinized tubes just before hydroxyurea administration, then 10 min, 30 min, 60 min, 2 h, 3 h, 4 h, 24 h, and 48 h after the dose of hydroxyurea. At each sampling time, one milliliter of blood was withdrawn and discarded to assure that the heparin solution did not dilute the sample. Then, 10 ml of whole blood were removed for analysis. Samples of ascites were obtained 1 week before administration of hydroxyurea, then 1 h after each administration of hydroxyurea. Additional samples were obtained from patients who required abdominal paracentesis as a comfort measure. The measurement of hydroxyurea in blood and ascites was performed using a modified colorimetric assay originally designed by Fabricius and Rajewsky (28) .

Collection and Preparation of the Tumor Cell Specimens.
Specimens from ascites were collected by sterile standard paracentesis as part of routine clinical procedures. Samples were immediately delivered and processed in the laboratory. Preservative-free heparin (10 units/ml; O’Neill, Johns and Feldman, St. Louis, MO) was added immediately after the collection of the ascites to prevent coagulation. Tumor cells in the specimens were washed twice in McCoy’s 5A medium containing 5% horse serum (Sigma Chemical Co.), 10% heat-inactivated FCS (Hyclone, Logan, UT), 2 mM sodium pyruvate, 2 mM glutamine, 90 units/ml penicillin, 90 µg/ml streptomycin, and 35 µg/ml L-serine (all Life Technologies, Inc., Long Island, NY).The viability of cells (typically >50%) was determined on a hemocytometer with trypan blue. The percentage of tumor cells in the specimens was determined by cytological parameters including nucleo:cytoplasmic ratio and number of nucleoli in the cells. Mucin-containing and mesothelial cells were stained using carcinoembryonic antigen and 8A11 monoclonal antibodies, respectively. The number of micronuclei was evaluated per 1000 tumor cells.

Culture of Cancer Cells in Soft Agar.
The ability of cells to clone in soft agar, as well as the in vitro effect of hydroxyurea, was evaluated using a tumor colony-forming unit system as reported previously (29) . Tumor cells were suspended in 0.3% agar in enriched Connaught Medical Research Laboratories medium 1066 (Life Technologies, Inc.) supplemented with 15% heat-inactivated horse serum, 2% fetal bovine serum, 0.3 mM ascorbic acid, 100 µg/ml penicillin, 100 µg/ml streptomycin, 4 mM glutamine, 2 µg/ml insulin, 3 µg/ml transferrin, 4.8 ng/ml hydrocortisone, 1.2% MEM, nonessential amino acids, 3.75 µg/ml catalase, and 0.022% sodium pyruvate (Sigma Chemical Co.). Cells were plated in 35-mm Petri dishes in a top layer of soft agar over an underlayer of agar to prevent growth of fibroblasts. Three plates were prepared for each data point. Hydroxyurea was tested at concentrations ranging from 30 to 3000 µM. Hydroxyurea-treated and control plates were placed in a 7% CO2 incubator at 37°C. Colonies (>50 cells) usually developed by day 14 of culture. The number of colonies in each dish was then counted under a microscope at x30. To ensure the presence of an excellent single-cell suspension, a positive control consisting of the cell toxin orthosodium vanadate at a concentration of 200 µg/ml was used. For an experiment to be considered valid, the orthosodium vanadate had to produce less than 30% survival of colony-forming units. When survival was >=30%, then the single-cell suspension on day 0 was poor, and the tumor sample test was considered nonevaluable. The use of a positive control has been shown to greatly increase the reproducibility of the human tumor cloning assay.

Methods for Chromosomal Analysis of dmins.
As described previously (20) , cytogenetic methods were used to determine the number of metaphase spreads with dmins and the average number and the range of dmins per metaphase spread. In the laboratory, 3–5 x 106 cells were placed into tissue culture flasks in the same medium. One culture was set up for a direct harvest in trypsin/hypotonic/Colcemid (THC), and placed in a 37°C water-jacketed CO2 incubator. The remaining cultures were set up and immediately exposed to Colcemid (0.1 µg/ml; Life Technologies, Inc.) for 2–6 h, harvested with trypsin 0.25% (Life Technologies, Inc.), placed in a hypotonic solution (0.075 M KCl) for 20 min, and fixed in methanol:glacial acetic acid (3:1). Chromosomal preparations were made and stained for dmins with 0.25% Wright-Giemsa stain for 2–4 min. Banding procedures were avoided because these methods generally cause dmins to stain only faintly. An attempt was made to score at least 50 mitoses for each ovarian tumor sample. dmins were reported only when they were clearly discernible, i.e., as paired acentric chromatin bodies approximately the width of a metaphase chromosome with similar staining characteristics and with refractivity similar to that of chromosomes. A variation of more than 50% in the percentage of spreads with dmins and micronuclei was considered significant.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Patient Characteristics.
A total of 16 patients (median age, 61; range, 44–88) with documented progression of advanced ovarian carcinoma entered the study (Table 1)Citation . All but one patient were previously exposed to cisplatin-based chemotherapy and were considered to have primary and secondary resistance to cisplatin in eight and seven cases, respectively. The median number of chemotherapy regimens prior to the study was three (range, one to four), including paclitaxel in nine cases. The median time interval between the last course of chemotherapy and the inclusion in the study was 2 months (range 1–11 months). Biological and cytogenetic parameters were fully evaluable in 14 patients (2 patients presented with early progression and were not fully evaluable for cytogenetic studies). The mean initial CA125 value was 450 units/ml (range, 16–10,800 units/ml). Pathological examination revealed that papillary, serous, poorly differentiated, and undifferentiated carcinomas were present in 10, 2, 2, and 2 cases, respectively. Cytological grade 3 carcinomas were observed in 62.5% of samples. Cytological examination of ascites showed that initial abdominal paracenteses contained an average of 90% cancer cells (range, 50–99%).


View this table:
[in this window]
[in a new window]
  		Table 1 Patient characteristics

 
Pharmacokinetics of Hydroxyurea.
The median number of courses (a course defined as a 6-week period of treatment) of hydroxyurea was one (range, one to nine). After oral administration, we observed considerable interpatient variability in the plasma pharmacokinetics of hydroxyurea (Fig. 1)Citation . However, in most patients, peak plasma concentrations were observed 3 h after oral administration. The peak plasma concentrations ranged between 723 and 1830 µmol/l (mean, 1349 ± 414 µmol/liter).



View larger version (34K):
[in this window]
[in a new window]
  		Fig. 1. Pharmacokinetics of oral low-dose hydroxyurea in plasma and ascites of patients with advanced ovarian cancer. Hydroxyurea was given p.o. every 3 days (80 mg/kg). Variations (mean ± SDs) of plasma ({circ}) and ascites (•) concentrations. The accumulation of hydroxyurea increased in ascites within the first 4 h (insert, Ratio A/P, ascites:plasma concentrations).

 
In ascites, the concentration of hydroxyurea increased progressively from 113.6 ± 102.3 to 610.3 ± 76.3 µmol/l after 1 and 4 h of oral administration, respectively (Fig. 1)Citation . No modification of pharmacokinetic parameters was detectable in two patients who had repeated pharmacokinetic analyses.

Antitumor Activity.
Biological and clinical analyses showed that the dose of hydroxyurea that we used in this study (i.e., the concentration of hydroxyurea in ascites) had no cytotoxic activity against ovarian cancers in our patients who were previously exposed to cisplatin.

In vitro, the antiproliferative effects of 30, 300, and 3000 µM hydroxyurea were initially tested on freshly expanded cancer cells taken directly from patients’ abdominal paracentesis using a human tumor cloning assay. All but two samples (87.5%) showed adequate growth in diluent controls. In vitro response, defined as a 50% reduction in the number of colonies as compared with control, was observed in only one specimen (overall in vitro response, 7%; 95% confidence interval, 0–34%) with 3000 µM hydroxyurea. Therefore, it was felt that any antitumor effect noted in this study were not attributable to the cytotoxicity of hydroxyurea.

Clinically, no complete or partial objective responses were observed in this population of patients heavily pretreated with cisplatin-based chemotherapy. Only one patient presented a transient 30% decrease in number of tumor cells in ascites at the completion of treatment. No grade 3–4 hydroxyurea-related toxicity was observed.

Effects of Hydroxyurea on dmins.
As can be seen in Table 2Citation , the cytogenetic analysis was performed in some patients at the time of the screening visit, at the time of the treatment initiation (day 1), and repeated several times during the treatment period (as long as paracenthesis was required for the patient’s benefit). Table 2Citation shows that there was a large interpatient variability in terms of numbers of spreads with dmins, total dmins, and percentages of spreads with dmins both at the time of screening and/or at the time of treatment initiation. In addition, in several patients, minor fluctuations in the numbers of spreads with dmins led us to consider that only a 50% variation above or below baseline values was significant. The initial value at baseline (at the screening visit or at the time of treatment initiation) was taken into account, each patient being her own control. The number and percentage of spreads with dmin in this population of patients at baseline was comparable with that of a control group of four patients who were not entered in this study and did not received hydroxyurea (Table 3)Citation .


View this table:
[in this window]
[in a new window]
  		Table 2 Variation of dmins before and during treatment with hydroxyurea in patients with advanced ovarian carcinoma

 

View this table:
[in this window]
[in a new window]
  		Table 3 Variations of dmins in a control group of patients with advanced ovarian cancer who were not entered into this study

 
Eleven (78.6%) of 14 evaluable spreads from patient specimens contained dmins before therapy. At the completion of the first course of chemotherapy, the percentage of spreads with dmins was stable or had increased in six specimens (55%), and was reduced by more than 50% in five specimens (45%) including the two specimens from patients with stable diseases (Fig. 2A)Citation . Early tumor progressions occurred in 6 patients. Therefore, the number of spreads with dmins was evaluable in only five patients after the second course of hydroxyurea: three of them had a significant reduction (>50%), one had an increase (>50%), and one had a stable number of spreads with dmins. Attempts to correlate hydroxyurea concentrations in ascites with variations of dmins were limited by the small number of patients in this study. However, most of the concentrations of hydroxyurea in ascites below 290 µmol/l were not associated with a reduction in dmins (Fig. 3)Citation .



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2. Variation of dmins (A) and micronuclei (B) in metaphase spreads from ovarian cancer cells in patients treated with oral low-dose hydroxyurea. C1, first course; C2, second course of oral hydroxyurea.

 


View larger version (43K):
[in this window]
[in a new window]
  		Fig. 3. Correlation between each individual patient’s concentration of hydroxyurea in ascites and the percentage of metaphase spreads with dmins.

 
Micronuclei were present in 13 of 14 evaluable specimens (92.8%). About 27.5% of spreads contained micronuclei before treatment. After the first course of hydroxyurea, the number of micronuclei declined significantly in three patients and increased in one patient (Fig. 2B)Citation .

Two patients (patients 1 and 7) with stage III cisplatin-resistant ovarian cancer experienced stabilization of their disease lasting 6 and 10 months, respectively. Hydroxyurea concentrations in ascites were 661.5 and 60.6 µM/l, for patients 1 and 7, respectively. Those concentrations induced no significant cytotoxic effects against tumor colony-forming units in vitro. Similarly, in those 2 cases, no reduction in the number of tumor cells in ascites was observed. In patient 1 (Fig. 4A)Citation , initial CA125 was within normal ranges. We also observed a progressive decrease in the percentage of spreads with dmins in this patient after the first course of hydroxyurea. No significant variation in the number of micronuclei was observed. In patient 7 (Fig. 4B)Citation , CA125 was initially elevated, then progressively declined after the first course of hydroxyurea. The percentage of spreads with dmins closely paralleled the variation of CA125 levels. A transient increase in the number of micronuclei was detected after the first course of treatment, followed by a return to basal level after the fourth course of treatment. In the two patients with stable disease, the percentage of spreads with dmins did not increase at the time of clinical progression. Analysis of molecular events associated with gene amplification was performed in those 2 patients. In patient 1, samples were probed with c-myc (dot blot analysis) and showed 19 to 21 gene copies before treatment and then a 52% reduction of the c-myc gene copy numbers after treatment with hydroxyurea (data not shown). However, most of the c-myc copies were not extrachromosomal. In patient 7, molecular events associated with gene amplification are displayed in Fig. 5Citation . This allowed us to demonstrate that the marker chromosome was chromosome 15 with a translocated segment of chromosome 8 (Fig. 6)Citation .



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 4. Variation of dmins (•) and micronuclei ({circ}) in spreads from patients 1 (A) and 7 (B), who experienced a prolonged stabilization under treatment with low-dose hydroxyurea. The follow-up was performed by monitoring variations of CA125 in the plasma ({square}). Samples were assessed pretreatment, then after each course (C) of chemotherapy.

 


View larger version (23K):
[in this window]
[in a new window]
  		Fig. 5. Analysis of molecular events associated with gene amplification in patient 7. Patient 7 dmins were microdissected. Metaphase spread from patient 7, hybridized with whole-chromosome paints, demonstrated that the marker chromosome is human chromosome 15 with a translocated segment of human chromosome 8 (data not shown). The patient’s chromosomes hybridized with the dmins O2 probe biotin-labeled (top panel), and the dmins stained with DNA stain (bottom panel), propidium iodide, to demonstrate that staining material in top panel was not artifactual.

 


View larger version (12K):
[in this window]
[in a new window]
  		Fig. 6. Marker chromosome analysis in patient 7. A, composite of normal chromosome 15 (left) and G-banded appearance of the marker chromosome 15 of patient 7 (right). B, localization of dmin-microdissected from O1 localizes to human chromosome 15q22 (biotin-labeled probe against propidium iodide chromosome counterstain). The marker chromosome localization of the O1 dmin probe is just centromeric to its normal locus. In addition, the dmin probe hybridizes to a hyperstaining region on the marker chromosome 15 (counterstained with 4',6-diamidino-2-phenylindole). C, normal peripheral blood lymphocyte metaphase chromosome (left) hybridized to commercially available marker (Oncor); the Prader-Willi (PW) marker, which localizes to 15q11-Q13 (D15S11), and the promyelocytic leukemia (PML) locus at 15q22 describe this region. The marker chromosome 15 has deleted sequences between PW and PML loci (right). D, normal peripheral blood lymphocyte chromosome hybridized to demonstrate that the O1 dmin-derived sequence is just centromeric to the PML locus. E, demonstration that the marker chromosome is composed predominantly of chromosome 15 with a terminal segment derived from chromosome 8.

 

   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
dmins have been shown to correlate with tumor progression (20) . Moreover, previous studies demonstrated that genetic instability in untreated ovarian cancer increases with time (30) . In this study, we used hydroxyurea in an attempt to increase the loss of dmins in patients with ovarian cancers. Other techniques using antisense RNA technology have been proposed but have been difficult to implement in clinical trials (31) . In addition, other drugs (topoisomerase-I inhibitors, etoposide) have been shown to reduce the number of dmins in cancer cells (18) . Hydroxyurea was selected because it appears to cause a reduction of extrachromosomal DNA regulating the genes contained within dmins, and it could be implemented with minimal delay in clinical trials. Several preclinical studies indicated that loss of dmins can be achieved with concentrations of hydroxyurea below 100 µM (19 , 20 , 22) . In this study, we used an oral dose of 80 mg/kg hydroxyurea every 3 days. As expected, this low dose was not associated with any severe clinical toxicity. Pharmacokinetic data showed important interpatient variability in both the plasma and ascites concentrations that might be attributable to variations in the oral bioavailability of hydroxyurea. In some samples, the drug was still detectable in ascites 30 h after oral administration at concentrations below 100 µM. However, because late time points were lacking for a number of patients, it remains possible that, for some patients, the absence of activity resulted from insufficient concentrations of hydroxyurea in ascites.

In this study, we documented the background fluctuation of dmins in the population of patients with advanced ovarian carcinoma by using the initial values of dmins before treatment with hydroxyurea. Indeed, data from a control group without treatment with hydroxyurea would have documented the background fluctuations in dmins and would have strengthened this study. However, conducting a large placebo-controlled randomized trial would have required a large number of patients and might have been difficult to realize for ethical reasons in patients with documented tumor progressions. In this study, the number of dmins from each patient before treatment with hydroxyurea gave us a rough idea of the background fluctuation of dmins in a population of patients. Data obtained in our population of patients were comparable with that of a comparable population of patients with ovarian cancer who were not included in this study and who were analyzed as a control group (Table 3)Citation . As can be seen in Table 2Citation , there was a large interpatient variability. In addition, there was intrapatient variations detected in six patients (patients 1, 5, 6, 7, 10, and 15) who had several samplings before receiving hydroxyurea. In this study, the number of dmins before treatment was, therefore, taken as the control value, each patient being her own control. To minimize the risk of misinterpretations of the results because of spontaneous minor fluctuations in the numbers of spreads with dmins, we decided to consider that only a 50% variation above or below baseline values was significant. Although this study provides some evidence supporting the hypothesis that hydroxyurea can reduce the percentage of spreads with dmins in patients, the study design does not allow a definitive conclusion, and the proof of concept shall be obtained from a larger randomized trial.

Our results indicate that a single oral administration of low-dose hydroxyurea can decrease the number of dmins in ovarian cancer cells present in ascites in 45% of patients after the first course of treatment. Moreover, a prolonged loss of dmins was observed in two patients who experienced long-term stabilization of their disease. Despite the absence of an untreated control group of patients in this pilot study, the evidence presented suggests that variations of dmins observed in patients are attributable to hydroxyurea treatment and are unlikely to occur spontaneously. Hydroxyurea has been shown to decrease the number of dmins, thereby accelerating the loss of extrachromosomally amplified oncogenes and reducing the tumorigenicity in several human cancer models in vitro and in vivo (19 , 20 , 22) . The mechanisms by which hydroxyurea affects the loss of dmins involve the preferential capture of dmins within micronuclei (20) . In this study, we observed a transient increase in the number of micronuclei in one patient who experienced long-term stabilization, whereas no variation in the number of micronuclei was observed in another. Therefore, our data do not allow further speculation on the putative mechanism by which hydroxyurea induces loss of dmins.

In this study, we observed no complete or partial responses with this dose of hydroxyurea in patients with advanced ovarian carcinoma. After the first course of chemotherapy, only two of five patients who had a decrease in the number of dmins experienced stabilization. We documented that long-term stabilization was not related to the cytotoxic effects of hydroxyurea. The hydroxyurea concentrations in ascites were far below concentrations that are supposed to have cytotoxic effects against ovarian cancer cells. Moreover, we demonstrated, using primary culture in soft agar, that ovarian tumor cells taken from our patients were not sensitive to hydroxyurea in vitro. We assumed, therefore, that long-term stabilization in our two patients, who had progressive disease under previous treatment with cisplatin, was not likely derived from hydroxyurea cytotoxicity. In those two patients, the important loss of dmins in cancer cells was striking and closely followed the variations of CA125 in one of the patients. This suggests that noncytotoxic doses of hydroxyurea designed to eliminate dmins could delay ovarian cancer progression. Interestingly, in patients who experienced prolonged stabilization, the loss of dmins was maintained at the time of tumor progression, which suggests that factors other than dmins can control tumor progression in ovarian cancer cells. A previous in vitro study showed that clonal cell selection and immortalization can modify the cytogenetic patterns and favors the expression of HSRs (4) . Moreover, recent studies have demonstrated the importance of p53 mutations in ovarian cancer progression (1 , 32) and have shown that this event is independent of dmin overexpressed genes (33) . This suggests that other additional genetic events independent of dmins may ultimately be responsible for tumor progression in patients with ovarian cancer treated with hydroxyurea.

Our study along with other previously published clinical trials (34) suggest that single-agent hydroxyurea is ineffective for inducing a response in patients with ovarian cancer refractory to cisplatin. However, considering the absence of toxicity with this dosage of hydroxyurea, the combination of hydroxyurea with cytotoxic drugs like cisplatin and paclitaxel that are active in the treatment of ovarian cancer has been proposed. Laboratory evidence showed that dmins contain amplified genes involved in resistance against anticancer drugs (10 , 11 , 13 , 35) ; therefore, the elimination of dmins may improve the results with conventional chemotherapy. A recent study showed that mdm2 expression in glioblastoma cells transfected with a human mdm2 expression vector conferred resistance to cisplatin-induced apoptosis (31) . In addition, treatment with antisense oligonucleotides that are targeted against mdm2 mRNA corrected the susceptibility of those cells to apoptosis, which suggests that mdm2 proteins act as a negative regulator of cisplatin-induced apoptosis and might play a role in the development of resistance to cisplatin in human tumors. A recently published clinical trial tested paclitaxel and a fixed oral dose of 500 mg hydroxyurea as second-line therapy in 30 patients with non-small cell lung cancer previously treated with cisplatin (36) . Treatment was well tolerated; 1 patient had a partial remission, and 15 had stable disease (including 6 with minor responses). Unfortunately, no pharmacokinetic or cytogenetic analyses were performed to determine whether the effects of low doses of hydroxyurea were related to the loss of dmins.

In summary, despite a limited number of patients, this initial study suggests that a noncytotoxic dose of hydroxyurea can increase the loss of dmins in tumor cells in some patients with advanced ovarian carcinoma. On the basis of our results, a larger controlled randomized trial of hydroxyurea versus no hydroxyurea has been implemented to determine whether hydroxyurea can delay cancer progression in ovarian and other types of tumors that frequently express dmins.


   ACKNOWLEDGMENTS
 
We thank Dr. Geoffrey Wahl for his helpful comments and the critical reading of the manuscript, Gregory Hannibal for his assistance in preparing the manuscript for publication, and Judy Turner and Peggy Durack for their excellent assistance in collecting the data.


   FOOTNOTES
 
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 in part by NIH Grant U01CA48405 from the Department of Health and Human Service, The San Antonio Cancer Center Support Grant P30CA54194, and the Cancer Therapy and Research Foundation of South Texas. E. R. received a grant from the Association pour la Recherche contre le Cancer (ARC), France. Back

2 To whom requests for reprints should be addressed, at Department of Medicine, Gustave Roussy Institute, 39 Rue Camille Desmoulins, 94805 Villejuif CEDEX, France. Phone: 011-33-1-4211-4289; Fax: 011-33-1-4211-5217; E-mail: raymond{at}igr.fr Back

3 The abbreviations used are: dmin, double-minute chromosome; HSR, homogeneously staining region; ULN, upper limit of normal. Back

4 Present address: Arizona Cancer Center, Office of Director, 1515 N. Campbell Avenue, Tucson, AZ 85724. Back

Received 7/ 6/00; revised 12/28/00; accepted 2/ 7/01.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Liu E., Nuzum C. Molecular sleuthing: tracking ovarian cancer progression. J. Natl. Cancer Inst. (Bethesda), 87: 1099-1101, 1995.[Free Full Text]
  2. Wahl G. M. The importance of circular DNA in mammalian gene amplification. Cancer Res., 49: 1333-1340, 1989.[Free Full Text]
  3. Hahn P. J. Molecular biology of double-minute chromosomes. Bioassays, 15: 477-484, 1993.[CrossRef][Medline]
  4. Benner S. E., Wahl G. M., Von Hoff D. D. Double-minute chromosomes and homogeneously staining regions in tumors taken directly from patients versus in human tumor cell lines. Anticancer Drugs, 2: 11-25, 1991.[Medline]
  5. McGill J. R., Beitzel B. F., Nielsen J. L., Walsh J. T., Drabek S. M., Meador R. J., Von Hoff D. D. Double minutes are frequently found in ovarian carcinomas. Cancer Genet. Cytogenet., 71: 125-131, 1993.[CrossRef][Medline]
  6. Carrol S. M., DeRose M. L., Gaudray P., Moore C. M., Needham-Vandevanter D. R., Von Hoff D. D., Wahl G. M. Double minutes chromosomes can be produced from precursors derived from chromosomal deletion. Mol. Cell. Biol., 8: 1525-1533, 1988.[Abstract/Free Full Text]
  7. Von Hoff D. D., Forseth B., Clare C. N., Hansen K. L., VanDevanter D. Double minutes arise from circular extrachromosomal DNA intermediates which integrate into chromosomal sites in human HL-60 leukemia cells. J. Clin. Investig., 85: 1887-1895, 1990.
  8. Windle B., Draper B. W., Yin Y., O’Gorman S., Wahl G. M. A central role for chromosome breakage in gene amplification, deletion formation, and amplicon integration. Genes Dev., 5: 160-174, 1991.[Abstract/Free Full Text]
  9. Fakharzadeh S. S., Rosenblum-Vos L., Murphy M., Hoffman E. K., George D. L. Structure and organization of amplified DNA on double minutes containing the mdm2 oncogene. Genomics, 15: 283-290, 1993.[CrossRef][Medline]
  10. Kaufman R. J., Brown P. C., Schimke R. T. Amplified dihydrofolate reductase genes in unstable methotrexate-resistant cells are associated with double minute chromosomes. Proc. Natl. Acad. Sci. USA, 76: 5669-5673, 1979.[Abstract/Free Full Text]
  11. Haber D. A., Schimke R. T. Unstable amplification of an altered dihydrofolate reductase gene associated with double-minute chromosomes. Cell, 26: 355-362, 1981.[CrossRef][Medline]
  12. Ruiz J. C., Choi K., Von Hoff D. D., Roninson I. B., Wahl G. M. Autonomously replicating episomes contains mdr1 genes in a multidrug-resistant human cell line. Mol. Cell. Biol., 9: 109-115, 1989.[Abstract/Free Full Text]
  13. Eijdems E. W., De Haas M., Coco-Martin J. M., Ottenheim C. P., Zaman G. J., Dauwerse H. G., Breuning M. H., Twentyman P. R., Borst P., Baas F. Mechanisms of MRP overexpression in four human lung-cancer cell lines and analysis of the MRP amplicon. Int. J. Cancer, 60: 676-684, 1995.[Medline]
  14. Von Hoff D. D., Needham-Van Devanter D. R., Yucel J., Windle B. E., Wahl G. M. Amplified human MYC oncogenes localized to replicating submicroscopic circular DNA molecules. Proc. Natl. Acad. Sci. USA, 85: 4804-4808, 1988.[Abstract/Free Full Text]
  15. Finlay C. A. The mdm-2 oncogene can overcome wild-type p53 suppression of transformed cell growth. Mol. Cell. Biol., 13: 301-306, 1993.[Abstract/Free Full Text]
  16. Reifenberger G., Liu L., Ichimura K., Schmidt E. E., Collins V. P. Amplification and overexpression of the MDM2 gene in a subset of human malignant gliomas without p53 mutations. Cancer Res., 53: 2736-2739, 1993.[Abstract/Free Full Text]
  17. McGill J., Walsh J., Durak M., Van den Berg C., Segura L., Von Hoff D. Unstable gene amplification found in breast cancers taken directly from patients. Proc. Am. Assoc. Cancer Res., 34: 129 1993.
  18. Wani M. A., Snapka R. M. Drug-induced loss of unstably amplified genes. Cancer Investig., 239: 465-468, 1990.
  19. Von Hoff D. D., Waddelow T., Forseth B., Davidson K., Scott J., Wahl G. Hydroxyurea accelerates loss of extrachromosomally amplified genes from tumor cells. Cancer Res., 51: 6273-6279, 1991.
  20. Von Hoff D. D., McGill J. R., Forseth B. J., Davidson K. K., Bradley T. P., Van Devanter D. R., Wahl G. M. Elimination of extrachromosomally amplified MYC genes from human tumor cells reduces their tumorigenicity. Proc. Natl. Acad. Sci. USA, 89: 8165-8169, 1992.[Abstract/Free Full Text]
  21. Van den Berg C., Von Hoff D. D. Use of hydroxyurea to alter drug resistance of human tumor cells. Cancer Treat. Res., 78: 95-114, 1995.[Medline]
  22. Canute G. W., Longo S. L., Longo J. A., Winfield J. A., Nevaldine B. H., Hahn P. J. Hydroxyurea accelerates the loss of epidermal growth factor receptor genes amplified as double-minute chromosomes in human glioblastoma multiform. Neurosurgery, 39: 976-983, 1996.[Medline]
  23. Markman M., Hoskins W. Responses to salvage chemotherapy in ovarian cancer: a critical need for precise definition of the treated population. J. Clin. Oncol., 10: 513-514, 1992.[Free Full Text]
  24. Rosner F., Rubin H., Parise F. Studies of the absorption, distribution, and excretion of hydroxyurea (NSC-32065). Cancer Chemother. Rep., 55: 167-171, 1971.[Medline]
  25. Blumenreich M. S., Kellihan M. J., Joseph U. G., Lalley K. A., Sherrill E. J., Sullivan D. M., Hamm J. T., Gentile P. S., Sheth S. P., Seeger J., Woodcock T. M. Long-term intravenous hydroxyurea infusion in patients with advanced cancer. Cancer (Phila.), 71: 2828-2832, 1993.[CrossRef][Medline]
  26. Koda R., Parimoo D., Spears C. P., Israel V., Jeffers S., Rogers M., Leichman L., Pujari M., Muggia F. Hydroxyurea (HU) iv 72-hour infusion during intraperitoneal (IP) floxuridine (FUDR) and leucovorin (LV): pharmacologic/clinical correlates. Proc. Am. Assoc. Cancer Res., 37: 1181 1996.
  27. Van den Berg C. L., McGill J. R., Kuhn J. G., Walsh J. T., De La Cruz P. S., Davidson K. K., Wahl G. M., Von Hoff D. D. Pharmacokinetics of hydroxyurea in nude mice. Anticancer Drugs, 5: 573-578, 1994.[CrossRef][Medline]
  28. Fabricius F., Rajewsky M. F. Determination of hydroxyurea in mammalian tissues and blood. Rev. Eur. Etud. Clin. Biol., 16: 675-686, 1991.
  29. Von Hoff D. D., Kronmal R., Salmon S. E., Turner J., Green J. B., Bonorris J. S., Moorhead E. L., Hynes H. E., Pugh R. E., Belt R. J., Alberts D. S. A Southwest Oncology Group study of the use of human tumor cloning assay for predicting response in patients with ovarian cancer. Cancer (Phila.), 67: 70-77, 1991.
  30. Thompson F. H., Liu Y., Emerson J., Weinsten R., Makar R., Trent J. M., Taetle R., Alberts D. S. Simple numeric abnormalities as primary karyotype changes in ovarian carcinoma. Genes Chromosomes Cancer, 10: 262-266, 1994.[Medline]
  31. Kondo S., Barnett G. H., Hara H., Morimura T., Takeuchi J. MDM2 protein confers the resistance of a human glioblastoma cell line to cisplatin-induced apoptosis. Oncogene, 10: 2001-2006, 1995.[Medline]
  32. Zheng J., Benedict W. F., Xu H. J., Hu S. X., Kim T. M., Velicescu M., Wan M., Cofer K. F., Debeau L. Genetic disparity between morphologically benign cyst contiguous to ovarian carcinomas and solitary cystadenomas. J. Natl. Cancer Inst. (Bethesda), 87: 1146-1153, 1995.[Abstract/Free Full Text]
  33. Foulkes W. D., Stamp G. W., Afzal S., Lalani N., McFarlane C. P., Trowsdale J., Campbell I. G. MDM2 overexpression is rare in ovarian carcinoma irrespective of TP53 mutation status. Br. J. Cancer, 72: 883-888, 1995.[Medline]
  34. Kavanagh J. J., Kudelka A. P., Edwards C. L., Freedman R. S., Gibbs H., Gonzalez de Leon C., Canetta R., Harper K. J., Kopplin S., Mante R., Krakoff I. H. A randomized crossover trial of parenteral hydroxyurea vs. high-dose Taxol in cisplatin/carboplatin-resistant epithelial ovarian cancer. Proc. Am. Soc. Clin. Oncol., 12: 1259 1993.
  35. Stark G. R. DNA amplification in drug resistant cells and tumors. Cancer Surv., 5: 1-23, 1986.[Medline]
  36. Stewart D. J., Tomiak E. M., Goss G., Gertler S. Z., Logan D., Huan S., Yau J., Dulude H., Evans W. K. Paclitaxel plus hydroxyurea as second-line therapy for non-small cell lung cancer. Lung Cancer, 15: 115-123, 1996.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Raymond, E.
Right arrow Articles by Von Hoff, D. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Raymond, E.
Right arrow Articles by Von Hoff, D. D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online