This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wadler, S. Articles by Murgo, A. Search for Related Content PubMed PubMed Citation Articles by Wadler, S. Articles by Murgo, A. Social Bookmarking                            What's this?

Phase II/Pharmacodynamic Trial of Dose-Intensive, Weekly Parenteral Hydroxyurea and Fluorouracil Administered With Interferon Alfa-2a in Patients With Refractory Malignancies of the Gastrointestinal Tract

From the Departments of Oncology, Surgery, and Radiology, Montefiore Medical Center, and the Albert Einstein Cancer Center, Bronx, NY; Department of Molecular Biology, University of Medicine and Dentistry, New Jersey, Osteopathic School of Medicine, Stratford, NJ; and the Investigational Drug Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD.

Address reprint requests to Scott Wadler, Department of Oncology, Hofheimer One, Montefiore Medical Center, 111 East 210th St, Bronx, NY 10467; email wadler{at}jimmy.harvard.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: Combined depletion of pyrimidine and purine DNA precursors has resulted in therapeutic synergism in vitro. The aims of the current study were to test this strategy in patients with refractory tumors and to assess its effects on selected nucleotide pools.

PATIENTS AND METHODS: A single-institution phase II trial was initiated in patients with advanced carcinomas of the stomach and pancreas. Patients had measurable disease and had no prior chemotherapy except adjuvant fluorouracil (5FU) or gemcitabine. 5FU was administered by CADD + pump at 2.6 g/m² intravenously by 24-hour infusion on days 1, 8, 15, 22, 29, and 36. Parenteral hydroxyurea (HU) was administered at 4.3 g/m² as a 24-hour infusion concurrently with 5FU. Interferon alfa-2a (IFN-2a) was administered at 9 million units subcutaneously on days 1, 3, and 5 each week. No drug was administered in weeks 7 and 8. Pharmacodynamic studies were performed to assess drug effects on levels of deoxyuridine triphosphate (dUTP) and thymidine triphosphate (TTP) pools in peripheral-blood mononuclear cells (PBMCs) before and 6 hours after treatment using a highly sensitive DNA polymerase assay.

RESULTS: There were 53 patients enrolled onto the study (gastric carcinoma, 31; pancreatic carcinoma, 22). The median age was 61 years, with 22% of patients 70 years old. The predominant grade 3 to 4 toxicities were leukopenia (49%), granulocytopenia (55%), and thrombocytopenia (22%). Severe diarrhea occurred in 12%, mucositis in 0%, and vomiting in 10% of patients. Patients 70 years had no greater incidence of toxicities. Among the 30 assessable patients with gastric carcinoma, there were two (7%) complete responders and 11 (37%) partial responders (median duration, 7 months). Among the 21 assessable patients with pancreatic carcinoma, there was one responder. Median survival among all patients with gastric carcinoma was 10 months and 13 months for patients with pancreatic carcinoma. Twenty-three patients had samples studied for levels of dUTP and TTP. There was no change in the levels of TTP before and after treatment. Furthermore, dUTP was detected in only five of 28 samples after treatment with no increase in the dUTP/TTP ratio.

CONCLUSION: Combination therapy with high-dose, weekly infusional HU and 5FU with IFN-2a modulation was well-tolerated with activity in gastric cancer. Patients 70 years tolerated therapy as well as younger patients. This was the first study to correlate levels of TTP and dUTP after treatment with clinical outcome. In PBMCs used as a surrogate tissue, HU abrogated the 5FU-induced increase in dUTP levels without reversing the overall efficacy of the regimen.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
CARCINOMAS ARISING in derivatives of the caudal foregut, including gastric and pancreatic tumors, are characterized by persistent local and regional spread, endogenous resistance to chemotherapeutic agents, and a poor clinical outcome. Surgery and radiation therapy are often futile, or at best palliative. Systemic therapies have, in general, failed to result in a substantial impact on survival.

For patients with advanced pancreatic cancer, treatment with the fluorinated nucleoside, gemcitabine, offers a therapeutic advantage compared with treatment with fluorouracil (5FU); however, the difference in survival between the two treatments is only approximately 1 month (5.65 v 4.41 months), and the response rate to gemcitabine is less than 10%.¹ Furthermore, no other agents have reproducible activity in patients with pancreatic cancer. For patients with advanced gastric cancer, the situation is somewhat different. There are a number of active single agents and combinations.² Nevertheless, using two of the most commonly used regimens as benchmarks, median survival is still only 6 to 7 months.³ Therefore, better systemic therapies are required for patients with these refractory gastrointestinal (GI) malignancies.

Before United States Food and Drug Administration approval of gemcitabine, 5FU-based therapies had been a mainstay of the treatment of GI malignancies, although with minimal impact. For example, in patients with pancreatic carcinoma, 5FU plus the modulating agent leucovorin, a regimen active against colorectal carcinoma, was inactive.⁴ Likewise, conventional doses of 5FU + interferon⁵ or 5FU + interferon + leucovorin⁶ were not substantially better than 5FU alone. Predictably, combinations of 5FU with conventional cytotoxic agents were no better than 5FU alone.⁷ For patients with advanced gastric cancer, 5FU has remained a component of many combination regimens such as FAMTX (high-dose 5FU and methotrexate with a leucovorin rescue) as well as other commonly used combination regimens.^8-9

Endogenous resistance to 5FU by the tumors is an important clinical problem. In preclinical systems, resistance to 5FU has resulted from accumulation of the endogenous nucleotide deoxyuridine monophosphate (dUMP), which competes with fluorodeoxyuridine (FdUMP) for binding to the 5FU target enzyme, thymidylate synthase.¹⁰ Intracellular sources of dUMP include reduction of cytidine diphosphate by ribonucleotide reductase (RR); thus inhibition of RR, theoretically, can prevent accumulation of dUMP and can diminish competition for thymidylate synthase by FdUMP, diminishing thymidine nucleotide pools required for DNA synthesis.¹¹ The clinical experience with combinations of 5FU and the RR inhibitor hydroxyurea (HU) in patients with colorectal cancer has been summarized.¹² Two early studies used single-dose HU followed by 5FU on either day 2¹³ or day 4.¹⁴ Three subsequent studies¹² used single-dose 5FU followed by a single oral dose of HU on day 4, all with response rates less than 20%. A fourth trial¹² added vincristine to HU, again without improvement in results. Trials conducted in the 1980s used 5FU for 4 days followed by HU either 4 and 10 hours later¹⁵ or on days 4 and 5,¹⁶ with low objective response rates.

One problem with the regimens described above is that whereas HU either orally or by intravenous bolus administration has a comparable pharmacokinetic profile,¹⁷ oral HU has a relatively short half-life. In detailed pharmacokinetic studies,¹⁸ oral administration of HU 800 mg/m² every 4 hours for 18 doses resulted in peak levels of 800 to 2,480 µmol/L after doses 1, 7, and 13; however, trough levels were only 410 µmol/L after the same doses. Based on in vitro studies with cultured cells, these concentrations of drug are unlikely to result in sustained suppression of RR activity.¹⁹ Furthermore, in the absence of sustained therapeutic levels of HU, resistance develops rapidly because of induction and overexpression of RR.²⁰ In contrast, studies of prolonged parenteral HU administration by the same investigators revealed essentially steady-state plasma levels of up to 1,090 µM for 72 hours with doses of HU from 2 to 3 mg/m²/min (2.9 to 4.3 g/m²/d). Thus, the failure of earlier regimens that used oral HU administration may have resulted at least in part from the lack of consistent plasma levels of drug.

HU is an S-phase–specific agent. Therefore, it is more likely to be active when administered in a continuous fashion. In human colon cancer cell lines, even a 10-fold increase in HU concentration for 24 hours resulted in only a 50% increase in cell kill. In contrast, incubation of cells with 1,000 µM HU for 24 hours had essentially no cytotoxic effect, whereas doubling the length of exposure to drug resulted in 98% cell kill, 72-hour exposure resulted in more than 99% cell kill, and 96-hour exposure resulted in 100% cell kill.¹⁹ Thus, these cell lines exhibited an exquisite schedule dependency to the effects of HU. Furthermore, therapeutic synergism resulted when 5FU was added to cell cultures simultaneously with HU rather than sequentially.²¹

To test the hypothesis that simultaneous parenteral administration of inhibitors of thymidylate synthase and ribonucleotide reductase can result in a clinically active regimen, the combination was tested in patients with refractory tumors of the GI tract. Both drugs were administered as simultaneous infusions every week, using doses that would result in serum concentrations determined by in vitro studies to be therapeutic. Interferon alfa-2a (IFN-2a) was used in combination with these agents as a modulating agent because IFN-2a has previously been shown to augment the activity of both 5FU and HU.^19,22 Finally, to assess the effects of the drug combination on pools of critical nucleotides, we used a highly sensitive DNA polymerase assay to measure levels of deoxyuridine triphosphate (dUTP) and thymidine triphosphate (TTP) before and after treatment.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
A single-institution, prospective, phase II clinical trial was initiated in October 1994. This trial was approved by the National Cancer Institute, the Albert Einstein Cancer Center, and the Montefiore Medical Center Institutional Review Board. The aims of this trial were to determine the objective response rates and survival for patients with pancreatic and gastric carcinomas treated with this regimen, to better define the toxicities resulting from treatment with this regimen, and to assess the effects of this treatment on selected nucleotide pools in peripheral-blood mononuclear cells from treated patients.

Eligibility
Patients were required to have a histologically documented carcinoma of the pancreas or stomach that was beyond the scope of surgical resection. Patients were required to have measurable disease and to have received no prior chemotherapy for advanced disease, although prior adjuvant 5FU or gemcitabine administered in combination with radiation therapy was allowed. All patients were fully ambulatory and had adequate bone marrow (leukocytes > 4,000 cells/µL), renal (serum creatinine < 2.0 mg/dL) and hepatic (AST and serum bilirubin less than three times the upper limit of normal) function. No patients had uncontrolled comorbid disease, and all gave informed consent that met federal, state, and institutional guidelines.

Study Design
HU was generously supplied by Bristol-Myers-Squibb Pharmaceuticals (Princeton, NJ) and distributed by the National Cancer Institute. IFN-2a (Roferon) was generously provided by Hoffman-LaRoche Pharmaceuticals (Nutley, NJ). A double-lumen Port-a-Cath (Horizon Medical Products, Manchester, GA) was placed before initiation of therapy and drug was administered by a CADD + pump (Sims/Deltec, St. Paul, MN), as previously described.²³ A single course of therapy was defined as eight weeks. Patients were administered 5FU 2.6 g/m² intravenously over 24 hours on days 1, 8, 15, 22, 29, and 36. Patients were administered HU 4.3 g/m² as a 24-hour infusion concurrently with the infusion of 5FU. IFN-2a was administered at 9 million units subcutaneously immediately before the start of the 5FU on day 1 of each week and then days 3 and 5 of each week. No drug was administered in weeks 7 and 8 of each cycle. Patients who required a dose reduction or suspension of treatment because of myelosuppression were allowed to receive filgrastim 480 µg subcutaneously on days 3, 4, 5, and 6 of each week in which chemotherapy was administered.

Dose reductions were based on toxicities as defined by the Common Toxicity Criteria.²⁴ The dose of IFN-2a was reduced by 50% for grade 3 neurotoxicity or a two-level decrease in performance status lasting more than 7 days. The dose of 5FU was reduced by 50% for grade 2 to 3 diarrhea or stomatitis. No dose modifications were made for grade 2 myelosuppression. For grade 3 myelosuppression, the doses of 5FU and of HU were reduced by 25%, unless the patient was receiving filgrastim. For patients on filgrastim, the dose of 5FU and HU was held, then reduced only if the leukocyte count was less than 1,500 cells/µL, the granulocyte count was less than 750 cells/µL, or the platelet count was less than 35,000/µL. Patients were removed from the study if they experienced grade 4 non-myelosuppressive toxicity or progressive disease. Response to therapy was determined after each cycle using the Eastern Cooperative Oncology Group criteria.²⁵

Nucleotide Studies
Patients enrolled onto the nucleotide studies signed a separate informed consent, which was institutional review board approved, to participate in those studies. In addition to patients enrolled onto the 5FU/HU/IFN trial, there were five patients enrolled from separate trials using 5FU and IFN-2a only, without HU, to assess the contribution of HU. These trials used slightly lower doses of 5FU with the same doses of IFN-2a and were all being conducted contemporaneously with the 5FU/HU/IFN study. All five patients had colorectal cancer but were otherwise comparable to the other patients in age, sites of metastasis, absence of prior treatment, and performance status.

Drugs and Reagents for Nucleotide Studies
Human dUTPase was purified as described.²⁶ DNA polymerase I, large (Klenow) fragment (exo-) was obtained from New England Biolabs (Beverly, MA). Microcon-50 and Microcon-3 microconcentrators were purchased from Amicon (Beverly, MA). Synthesized oligonucleotides were obtained from GIBCO (Grand Island, NY). dUTP and dTTP were obtained from Pharmacia (Uppsala, Sweden). Deoxyadenosine 5'-triphosphate (2,8-³H) tetra-ammonium salt (specific activity, 32 Ci/mmol) was obtained from Moravek Biochemicals (Brea, CA). DE81 anion exchange papers were purchased from Whatman (Hillsboro, OR). Cell culture media and medium supplements were obtained from GIBCO. All other chemicals were from Sigma (St Louis, MO).

Mononuclear Cell Isolation and Extraction
Fifty milliliters of peripheral blood was collected in heparinized tubes from patients at 0 and 6 hours after the beginning of 5FU and HU infusion. Samples were maintained on wet ice and immediately diluted with an equal quantity of Hanks' balanced salt solution; peripheral-blood mononuclear cells were isolated by density centrifugation.²⁷ In initial studies, only the dUTP/TTP ratio was determined; in later studies, absolute levels of TTP and dUTP were determined and adjusted to cell counts, which were performed using trypan blue exclusion to estimate the number of nonviable cells. Cells were extracted with 400 µl of 80% methanol for 45 minutes at -20°C. Insoluble materials were removed by centrifuging at 1,400 rpm for 30 minutes. The clarified supernatant was centrifuged over Microcon-50 microconcentrators (50,000 molecular weight cutoff) followed by centrifugation through Microcon-3 microconcentrators (3,000 molecular weight cutoff) to remove any DNA fragments. The extract was divided into paired aliquots of equal volume (one each for determination of dUTP and TTP levels) and lyophilized to remove the methanol. Samples were stored at -80°C until the assay was performed.

dUTPase Predigestion
On one of the paired samples at time 0 and 6 hours, dUTPase predigestion was performed as described²⁸ with the following modifications. Six nanograms of dUTPase was added in 40 µL of buffer containing 34 mM Tris-HCl (pH 8.3), 10 mM MgCl₂, 0.5 mM EDTA, and 0.25 mg/mL bovine serum albumin. Samples were incubated for 20 minutes at 37°C. Sixty microliters of 100% methanol was added to produce a final concentration of 60%. Precipitated protein was removed by centrifugation, and the clarified supernatant was lyophilized to remove the methanol.

DNA Polymerase Assay
The DNA polymerase assay was performed as described²⁸ with the following modifications. The template sequence used was 5'-TTT ATT TAT TTA TTT ATT TAG GCG GTG GAG GCG G-3'. The primer sequence was 5'-CCG CCT CCA CCG CC-3'. To each sample was added 0.3 units of DNA polymerase I, large (Klenow) fragment (exo-) in 100 µL of 34 mM Tris-HCl pH 8.3. A mixture of 10 mM MgCl₂, 0.2 mM EDTA, 0.25 mg/mL bovine serum albumin, 0.01 µCi/µL 2,8-³H, and 32 nM oligonucleotide (template + primer), incubated at 60°C for 5 minutes and maintained at 42°C, was added to each sample. The final mixture was then incubated at 42°C for 45 minutes, then 30 µL of each sample was blotted onto a DE81 filter (Whatman). The filters were washed six times for 5 minutes with 1% sodium pyrophosphate and 5% trichloroacetic acid and dried, and the radioactivity was counted by liquid scintigraphy.

Data Analysis
Standard curves were obtained for each experiment by assaying serial dilutions of dUTP and TTP in triplicate, with and without dUTPase pretreatment. The correlation coefficient was routinely greater than 0.99. dUTP concentrations were determined by subtracting the counts attributable to TTP (sample with dUTPase predigestion) from the results of the aliquot not treated with dUTPase (which measured TTP + dUTP). The assay was sensitive to 30 fmol/tube, which was two-fold baseline. Comparisons of nucleotide levels before and after treatment were compared by one-sample signed rank test.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Demographic Characteristics
As listed in Table 1, 53 patients were enrolled onto the study (31 patients with gastric carcinoma and 22 patients with pancreatic carcinoma). There were two ineligible patients: one patient with a pancreatic tumor had an islet-cell histology on pathologic review, and one patient with gastric cancer was ineligible because of elevated serum liver enzyme levels. The median age was 63 years, with 11 (22%) of the 51 eligible patients ranging in age from 70 to 85 years. The median age distribution was 61 years for gastric carcinoma patients and 64 years for pancreatic carcinoma patients. As expected, the predominant site of metastatic involvement was the liver (28 [55%] of 51 patients) and regional lymph nodes (19 [37%] of 51). Thirty-six (71%) of the 51 patients had prior surgical treatment.

Toxicities
As listed in Table 2, the predominant toxicities were myelosuppression, with 25 (49%) of 51 patients having grade 3 to 4 leukopenia, 28 (54%) of 51 having grade 3 to 4 granulocytopenia, and 11 (22%) of 51 having grade 3 to 4 thrombocytopenia. Complications of myelosuppression, including severe or life-threatening infection or hemorrhage, occurred in 10 (20%) of 51 and zero of 49 patients, respectively. A total of 25 (49%) of 51 patients eventually received filgrastim on a regular basis so that they could continue to receive treatment on schedule. No deaths resulted from the treatment.

Grade 3 or 4 gastrointestinal toxicities occurred less commonly than hematologic toxicities. Diarrhea occurred in six (12%) of 51 patients, mucositis in zero of 51 patients, and vomiting in five (10%) of 51 patients. Constitutional symptoms occurred commonly, with grade 2 fever in 25 (49%) of 51 patients and moderate or severe fatigue in five (10%) of 51 and four (8%) of 51 patients, respectively. Grade 3 neurotoxicity occurred in 8% of patients and was likely related to the classical IFN-2a–mediated fatigue syndrome.

When analyzed by age, patients 70 years old had no greater incidence of toxicities than younger patients, except for a higher incidence of grade 3 to 4 leukopenia (64% v 45%). Moderate fatigue was observed in 18% of patients younger than 70 versus 18% of those 70 or older. An indirect measure of tolerability in the older patient population was median duration on study: 3.1 months for patients younger than 70 and 5.8 months for patients 70 or older.

Dose reduction or temporary suspension of one or more drugs was required in 35 (69%) of 51 patients. The median time to first dose reduction or suspension was 20 days (range, 3 to 189 days). The median time to dose reduction in patients 70 or older was 21 days.

Clinical Outcome
Clinical response to treatment is listed in Table 3. The overall response rate for all 51 assessable patients was 27% (95% confidence interval [CI], 16% to 42%). Among the 30 assessable patients with gastric carcinoma, there were two (7%) complete responders and 11 (37%) partial responders, for an overall response rate of 43% (95% CI, 25% to 63%). One complete response was pathologically confirmed; the other consisted of complete radiographic normalization of liver metastases with normalization of tumor markers. The median duration of response was 7 months (range, 2 to 21 months). The duration of response for the complete responders was 16 and 19 months. Three patients with gastric carcinoma had complete bowel obstruction before study entry. Two of those patients had complete resolution of their obstruction lasting 21 and 6+ months. Among the 21 assessable patients with pancreatic carcinoma, there was one responder.

Ten of 14 responders had elevated serum levels of carcinoembryonic antigen, cancer antigen (CA)-19-9, and/or CA-125 before study entry (medians before study entry were 21 ng/dL, 309 U/dL, and 87 U/dL, respectively). One patient did not have follow-up tumor markers. After treatment, markers in nine of nine patients who had follow-up measurements decreased, confirming the radiographic impression of clinical response (median at response was 4 ng/dL, 97 U/dL, and 11U/dL, respectively).

As with toxicities, advanced age did not diminish response to therapy. The response rate among gastric cancer patients 70 years old was 50% (three of six), including both patients who had complete responses. The response durations for patients 70 years old were 9, 16, and 19 months. The average duration on the study was 5.8 months for both gastric carcinoma and pancreatic carcinoma patients. With a follow-up period of 10 months, median survival time among patients with gastric carcinoma was 10 months. For pancreatic carcinoma patients with a median follow-up period of 8 months, median survival time was 13 months.

Nucleotide Studies
Peripheral-blood mononuclear cells were isolated from 32 patients before treatment and 6 hours after the start of treatment for nucleotide studies. Four of 32 specimens were technically inadequate; among the remaining 28 paired patient samples, 23 patients were treated with 5FU + HU + IFN-2a (FHI) on this study and five patients were treated with 5FU + IFN-2a (FI) on another study and were included to assess the role of HU.

Using a cutoff of 30 fmol, measurements of TTP were informative in specimens from 18 (64%) of 28 patients, including 14 (61%) of 23 treated with FHI and four (80%) of five treated with FI. Among the 23 patients treated with FHI, 10 had complete cell counts for samples drawn before and after treatment. Levels of TTP before treatment were 32.6 ± 19.9 fmol/10⁶ cells (mean ± SEM). After treatment, measurements of TTP were informative in 14 of 23 patients treated with FHI. Mean levels of TTP were 43.0 ± 21.6 fmol/10⁶ cells (mean ± SEM, P = .99). Among these patients, there were two outliers (TTP levels 3.8- and 6.4-fold above the mean before treatment). If these patients were excluded, there was an increase in TTP levels from 2.8 ± 0.8 fmol/10⁶ cells to 13.0 ± 6.9 fmol/10⁶ cells (mean ± SEM, P = .5469). There were too few informative samples from patients treated with FI to assess whether there was a significant decrease in levels of TTP.

Using 30 fmol as a cutoff, levels of dUTP were detectable after treatment in five (17.9%) of 28 patients, all treated with FHI. After correction for cell counts, levels of dUTP ranged from 0.85 to 30.0 fmol/10⁶ cells. Among the five patients treated with FI, none had an increase in levels of dUTP to detectable levels. Furthermore, among 13 patients in whom TTP was measurable after treatment, 10 (77%) of 13 had unmeasurable levels of dUTP. Among all 28 patients studied, the dUTP/TTP ratio never increased after treatment with either FHI or FI.

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Our earlier phase I trial²³ demonstrated the tolerability of high-dose, weekly infusional HU in combination with high-dose, weekly 5FU and IFN-2a. The current trial, in a larger cohort of 51 assessable patients, confirmed the tolerability of this regimen in patients with refractory GI malignancies. Because patients with refractory malignancies of the GI tract are often in their 70s and 80s, it was of interest that this regimen was well tolerated by the 22% of patients 70 years old, and that response rates and duration ofresponse were somewhat better among this cohort of patients than among younger patients. This may relate in part to biologic differences between GI malignancies in older versus younger patients.²⁹ Furthermore, this may require a re-evaluation of the concept that dose intensive therapy for disseminated or locally recurrent refractory GI malignancies is not warranted in the elderly patient population. Finally, this regimen also demonstrated the tolerability of moderate-dose IFN-2a in combination with high-dose chemotherapy in this elderly patient population; time to first dose reduction was the same, 21 and 20 days, as in the younger patients.

Parenteral HU was used 30 years ago in the treatment of refractory malignancies. Moertel et al³⁰ administered HU 50 to 100 mg/kg for 8 hours daily for 4 to 5 days. Gottlieb et al³¹ administered HU 0.5 to 1.0 g/m² every 8 hours for 5 days. Belt et al ¹⁸ have recommended HU 3 mg/m²/min (4.3 g/m²/d) for 72 hours when used as a single agent and 2.5 mg/m²/min when used in combination with other agents. Recently, single-agent infusional doses of HU up to 27 g/m² have been safely used in patients with chronic leukemia.³² Therefore, prolonged, continuous exposure to HU is tolerable. Of greater importance, however, is the biochemical rationale that suggests that prolonged infusion is preferable. Specifically, preclinical studies in human colon cancer cell lines¹⁹ have shown that duration of exposure to drug is at least as important as drug concentration for the cytotoxic effect.

Our trial is the first to use parenteral HU as a modulator of 5FU in this group of patients. Various sequences have been studied in preclinical tumor model systems. Studies in murine models in which 5FU was administered before HU demonstrated therapeutic synergism for the combination.¹¹ In 9L rat brain tumor cells,^33-34 treatment with low concentrations of 5FU followed by HU enhanced cell kill, presumably because of accumulation of cells in S phase before administration of HU. Muggia and Moran¹² suggested that sequential administration of 5FU and HU may be required to prevent inhibition by HU of 5FU anabolism to FdUMP, and that the interaction of 5FU/HU is sequence dependent, occurs at clinically achievable concentrations of both drugs, and represents a general strategy combining fluoropyrimidines and RR inhibitors. In contrast, HU also enhanced the antitumor activity of 5FU in a dose-dependent fashion against Ehrlich ascites tumor cells³⁵ and SW480 human colon cancer cells²¹ when the two drugs were administered simultaneously.

The mechanism of interaction of 5FU and HU at the cellular level is unclear. Treatment with inhibitors of thymidylate synthase resulted in cellular accumulation of dUTP³⁶ and incorporation of dUTP into cellular DNA.³⁷ Although this may be critical to the cytotoxicity of these agents,^38-39 this phenomenon is cell-type specific.^40-42 For example, treatment with inhibitors of thymidylate synthase in vitro has resulted in rapid increases in intracellular levels of dUTP^27,43 in the absence of a consistent association with apoptotic cell death. Thus, we were interested in whether such an increase in levels of dUTP could account for the clinical antitumor effects observed with FHI treatment.

This was the first study to use measurements of dUTP levels in the clinical setting to assess the effects of drug treatment on nucleotide pools. Among various cell lines, we have observed depletion of pools of TTP by FU with accumulation of dUTP and blunting of these effects with HU, although we noted great disparity depending on the cell lines studied.⁴³ In this clinical study, we sought to determine whether treatment with 5FU resulted in depletion of TTP pools and accumulation of dUTP using the five FU/IFN-2a patients as a positive control, and whether these effects would be blunted with the addition of HU. Using a highly sensitive assay for dUTP, capable of measuring to 30 fmol, an increase in dUTP pools to detectable levels was observed in only five (17%) of 28 patient samples after treatment. Among 14 patients with measurable levels of TTP after treatment, only five (35.7%) had measurable levels of dUTP; thus, this was not a general failure of the assay. Therefore, it is possible that HU abrogated a 5FU-induced increase in pools of dUTP, similar to those observed in vitro after fluoropyrimidine treatment in the absence of HU,^27,43 without diminishing the overall efficacy of this regimen. These conclusions need to be interpreted with caution, however. Peripheral blood mononuclear cells were used as a surrogate for tumor tissue because it was thought that serial biopsies of pancreatic and gastric lesions were impracticable. Because terminally differentiated peripheral-blood mononuclear cells likely have much lower nucleotide pools than actively dividing tumors cells, this may not be an accurate measure of drug effects occurring at the tumor cell level.

Levels of TTP failed to decrease 6 hours after treatment with FHI, consistent with previous observations that depletion of TTP pools is much less thorough in cells treated with FHI than in those treated with FI or 5FU alone.²¹ Although the results in patients receiving FI alone would have been of interest, there were too few patients studied in this clinical trial to reach a definitive conclusion. We have postulated that the combination of 5FU and HU is synergistic by simultaneously inducing aberrations in pools of purine and pyrimidine nucleotides⁴³; however, this remains to be demonstrated in the clinical setting.

The clinical outcome with this regimen was encouraging. The median survival time of 10 months in patients with gastric carcinoma was nearly double that observed with conventional treatment regimens,³ as was the median survival time among patients with pancreatic carcinoma. The efficacy of this regimen in patients with gastric carcinoma is being investigated in the multi-institutional setting by the Eastern Cooperative Oncology Group (EST 6296). The advantages in patients with pancreatic cancer are less clear for the 24-hour schedule. Therefore, we intend to examine the 48-hour schedule, which was piloted in our phase I trial, in patients with these tumors.

	ACKNOWLEDGMENTS

 
Supported in part by FD-R-001009 from the United States Food and Drug Administration; and by R21CA72443-01 and Cancer Center Support grant no. CA 13330 from the National Cancer Institute, National Institutes of Health.

We thank Dr Terry Dugan and Dr Al Fallavollita for their generous assistance, Sora Yoon and John Lussier for excellent technical assistance, and Dr Wynne Aherne for helpful discussions.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Burris HA III, Moore MJ, Andersen J, et al: Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: A randomized trial. J Clin Oncol 15:2403-2413, 1997[Abstract/Free Full Text]

2. Wadler S, Green M, Muggia F: The role of anthracyclines in the treatment of gastric cancer. Cancer Treat Rev 12:105-132, 1985[Medline]

3. Kelsen D, Atiq OT, Saltz L, et al: FAMTX versus etoposide, doxorubicin, and cisplatin: A random assignment trial in gastric cancer. J Clin Oncol 10:541-548, 1992[Abstract/Free Full Text]

4. DeCaprio JA, Mayer RJ, Gonin R, et al: Fluorouracil and high-dose leucovorin in previously untreated patients with advanced adenocarcinoma of the pancreas: Results of a phase II trial. J Clin Oncol 9:2128-2133, 1991[Abstract]

5. Pazdur R, Ajani JJ, Abbruzzese JL, et al: Phase II evaluation of fluorouracil and recombinant?alpha-2a-interferon in previously untreated patients with pancreatic adenocarcinoma. Cancer 70:2073-2076, 1992[Medline]

6. Scheithauer W, Pfeffel F, Kornek G, et al: A phase II trial of 5-fluorouracil, leucovorin, and recombinant alpha-2a-interferon in advanced adenocarcinoma of the pancreas. Cancer 70:1864-1866, 1992[Medline]

7. Cullinan SA, Moertel CG, Fleming TR, et al: A comparison of three chemotherapeutic regimens in the treatment of advanced pancreatic and gastric carcinoma. JAMA 253:2061-2067, 1985[Abstract/Free Full Text]

8. Cascinu S, Labianca R, Graziano F, et al: Intensive weekly chemotherapy for locally advanced gastric cancer using 5-fluorouracil, cisplatin, epidoxorubicin, 6S-leucovorin, glutathione and filgrastim: A report from the Italian Group for the Study of Digestive Tract Cancer (GISCAD). Br J Cancer 78:390-393, 1998[Medline]

9. Taal BG, Teller FGM, ten Bokkel Huinink WW, et al: Etoposide, leucovorin, 5-fluorouracil (ELF) combination chemotherapy for advanced gastric cancer: Experience with two treatment schedules incorporating intravenous or oral etoposide. Ann Oncol 5:90-92, 1994[Abstract/Free Full Text]

10. Spears CP, Hayes A, Shahinian AH, et al: Deoxyuridylate effects on thymidylate synthase-5-fluorodeoxyuridylate-folate ternary complex formation. Biochem Pharmacol 38:2985-2993, 1989[Medline]

11. Moran RG, Danenberg PV, Heidelberger C: Therapeutic response of leukemic mice treated with fluorinated pyrimidines and inhibitors of deoxyuridylate synthesis. Biochem Pharmacol 31:2929-2935, 1982[Medline]

12. Muggia F, Moran RG: Treatment of colon cancer based on biochemical modulation of fluoropyrimidines by hydroxyurea. Semin Oncol 19:90-93, 1992 (suppl 9)

13. Lokich JJ, Pitman SW, Skarin AT: Combined 5-fluorouracil and hydroxyurea therapy for gastrointestinal cancer. Oncology 32:34-37, 1975[Medline]

14. Lerner HJ: Clinical experience using hydroxyurea and 5-fluorouracil in hepatic metastasis from adenocarcinoma of the colon. Proc Am Assoc Cancer Res 15:32, 1974 (abstr)

15. Walsh C, Speyer JL, Wernz JC, et al: Trial based on biochemical modulation of 5-fluorouracil (5FU) by daily sequential hydroxyurea (HU) in colorectal cancer. Proc Am Assoc Cancer Res 26:168, 1985 (abstr)

16. Kao AK, Muggia FM, Dubin N, et al: Evaluation of a sequential 5-FU and hydroxyurea combination in advanced bowel cancer. Cancer Treat Rep 68:1383-1385, 1984[Medline]

17. Rodriguez GI, Kuhn JG, Weiss GR, et al: A bioavailability and pharmacokinetic study of oral and intravenous hydroxyurea. Blood 91:1533-1541, 1998[Abstract/Free Full Text]

18. Belt RJ, Haas CD, Kennedy J, et al: Studies of hydroxyurea administered by continuous infusion: Toxicity, pharmacokinetics, and cell synchronization. Cancer 46:455-462, 1980[Medline]

19. Wadler S, Horowitz R, Rao J, et al: Interferon augments the cytotoxicity of hydroxyurea without enhancing its activity against the M2 subunit of ribonucleotide reductase: Effects in wild-type and resistant human colon cancer cells. Cancer Chemother Pharmacol 38:522-528, 1996[Medline]

20. Wadler S, Zhang HY, Cammer M, et al: Quantification of ribonucleotide reductase expression in wild-type and hydroxyurea-resistant cell lines using in situ reverse transcriptase PCR and a computerized image analysis system. Anal Biochem 267:24-29, 1999[Medline]

21. Wadler S, Horowitz R, Zhang HY, et al: Depletion of deoxyribonucleotide triphosphates fails to inhibit expression of the G1/S transition state enzyme, ribonucleotide reductase. Biochem Pharmacol 55:1353-1360, 1998[Medline]

22. Wadler S, Wersto R, Weinberg V, et al: Interaction of fluorouracil and interferon in human colon cancer cell lines: Cytotoxic and cytokinetic effects. Cancer Res 50:5735-5739, 1990[Abstract/Free Full Text]

23. Wadler S, Haynes H, Schechner R, et al: Phase I trial of high-dose infusional hydroxyurea, high-dose infusional 5-fluorouracil and recombinant interferon-alpha-2a in patients with advanced malignancies. Invest New Drugs 13:315-320 1996

24. National Cancer Institute: Guidelines for Reporting of Adverse Drug Reactions. Bethesda, MD, Division of Cancer Treatment, National Cancer Institute, 1988

25. Oken MM, Creech RH, Tormey DC, et al: Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol 5:649-655, 1982[Medline]

26. Ladner RD, McNulty DE, Carr SA, et al: Characterization of distinct nuclear and mitochondrial forms of human deoxyuridine triphosphate nucleotidehydrolase. J Biol Chem 271:7745-7751, 1996[Abstract/Free Full Text]

27. Boyum A: Isolation of mononuclear cells and granulocytes from human blood: Isolation of mononuclear cells by one centrifugation and sedimentation at 1 g. Scan J Clin Lab Invest 97:77-89, 1968 (suppl)

28. Horowitz RW, Zhang H-Y, Schwartz EL, et al: Measurement of deoxyuridine triphosphate and thymidine triphosphate in the extracts of thymidylate synthase-inhibited cells using a modified DNA polymerase assay. Biochem Pharmacol 54:635-638, 1997[Medline]

29. Maehara Y, Emi Y, Tomisaki S, et al: Age related characteristics of gastric carcinoma in young and elderly patients. Cancer 77:1774-1780, 1996[Medline]

30. Moertel CG, Reitemeier RJ, Hahn RG: Evaluation of hydroxyurea by parenteral infusion. Cancer Chemother Rep 49:27-29, 1965[Medline]

31. Gottlieb JA, Frei E, Luce JV: Dose schedule studies with hydroxyurea in malignant melanoma. Cancer Chemother Rep 55:277-280, 1971[Medline]

32. Gandhi V, Plunkett W, Kantarjian H, et al: Cellular pharmacodynamics and plasma pharmacokinetics of parenterally infused hydroxyurea during a phase I clinical trial in chronic myelogenous leukemia. J Clin Oncol 16:2321-2331, 1998[Abstract]

33. Kobayashi S, Hoshino T: Combined cytotoxic effect of low-dose 5-fluorouracil and hydroxyurea on 9L cells in vitro. Cancer Res 43:5309-5313, 1983[Abstract/Free Full Text]

34. Hoshino T, Nagashima T: Factors modifying cytotoxicity induced by 5-FU and hydroxyurea. Cancer Treat Rep 69:993-997, 1985[Medline]

35. Frankfurt OS: Enhancement of antitumor activity of 5-fluorouracil by drug combinations. Cancer Res 33:1043-1047, 1973[Abstract/Free Full Text]

36. Curtin NJ, Harris AL, et al: Mechanism of cell death following thymidylate synthase inhibition: 2'-Deoxyuridine-5'-triphosphate accumulation, DNA damage, and growth inhibition following exposure to CB3717 and dipyridamole. Cancer Res 51:2346-2352, 1991[Abstract/Free Full Text]

37. Ingraham HA, Tseng BY, Goulian M: Nucleotide levels and incorporation of 5-fluorouracil and uracil into DNA of cells treated with 5-fluorodeoxyuridine. Mol Pharmacol 21:211-216, 1982[Abstract]

38. Canman CE, Radany EH, Parsels LA, et al: Induction of resistance to fluorodeoxyuridine cytotoxicity and DNA damage in human tumor cells by expression of Escherichia coli deoxyuridinetriphosphatase. Cancer Res 54:2296-2298, 1994[Abstract/Free Full Text]

39. Canman CE, Lawrence TS, Shewach DS, et al: Resistance to fluorodeoxyuridine-induced DNA damage and cytotoxicity correlates with an elevation of deoxyuridine triphosphatase activity and failure to accumulate deoxyuridine triphosphate. Cancer Res 53:5219-5224, 1993[Abstract/Free Full Text]

40. Parsels LA, Parsels JD, Wagner LM, et al: Mechanism and pharmacological specificity of dUTPase-mediated protection from DNA damage and cytotoxicity in human tumor cells. Cancer Chemother Pharmacol 42:357-362, 1998[Medline]

41. Maybaum J, Burton EC, Shelton DA, et al: Divergent patterns of incorporation of bromodeoxyuridine and iododeoxyuridine in human colorectal tumor cell lines. Biochem Pharmacol 42:131-137, 1991[Medline]

42. Sundseth R, Singer S, Yates B, et al: Thymineless apoptotic death induced by 1843U89 in human tumor cell lines is independent of dUTP accumulation and misincorporation of dUMP into DNA. Proc Am Assoc Cancer Res 38:3182a, 1997 (abstr)

43. Damle S, Ladner R, Wadler S: Measurement of deoxyuridine triphosphate (dUTP) and deoxythymidine triphosphate (dTTP) using a modified DNA polymerase assay in antimetabolite treated human colon cancer cells and its correlation with cell survival. Proc Am Assoc Cancer Res 40:401, 1999 (abstr)

Submitted December 2, 1998; accepted February 16, 1999.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				S. Wadler, D. Makower, C. Clairmont, P. Lambert, K. Fehn, and M. Sznol Phase I and Pharmacokinetic Study of the Ribonucleotide Reductase Inhibitor, 3-Aminopyridine-2-Carboxaldehyde Thiosemicarbazone, Administered by 96-Hour Intravenous Continuous Infusion J. Clin. Oncol., May 1, 2004; 22(9): 1553 - 1563. [Abstract] [Full Text] [PDF]

				K.-D. Lee, T.-W. Liu, C.-W. Wu, C.-M. Tiu, J. M. Liu, T.-R. Chung, J.-Y. Chang, J. Whang-Peng, and L.-T. Chen Non-surgical treatment for afferent loop syndrome in recurrent gastric cancer complicated by peritoneal carcinomatosis: percutaneous transhepatic duodenal drainage followed by 24-hour infusion of high-dose fluorouracil and leucovorin Ann. Onc., July 1, 2002; 13(7): 1151 - 1155. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wadler, S. Articles by Murgo, A. Search for Related Content PubMed PubMed Citation Articles by Wadler, S. Articles by Murgo, A. Social Bookmarking                            What's this?