SEARCH

SEARCH BY CITATION

Keywords:

  • uracil and ftorafur (UFT);
  • breast cancer;
  • metastatic;
  • leucovorin

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

BACKGROUND:

A combination of uracil and ftorafur (UFT) was developed to combine the cytotoxic effects of 5-fluorouracil (5-FU) with convenient oral dosing. Leucovorin was combined with UFT to further potentiate the effect of 5-FU on tumor cells. Orally administered UFT and leucovorin provided higher peak plasma concentrations of 5-FU and prolonged therapeutic 5-FU concentrations compared with continuous infusion of 5-FU.

METHODS:

Ninety-one patients with metastatic breast cancer who had been previously treated with anthracyclines and/or taxanes were treated with UFT and leucovorin, given orally, for the first 28 days of a 35-day cycle. The total daily dose of UFT was 300 mg/m2, administered in 3 doses of 100 mg/m2 each every 8 hours. The primary endpoint was time to disease progression (TTP). Secondary objectives included overall tumor response rate (overall response equals complete response plus partial response) and overall survival.

RESULTS:

Of the 91 patients enrolled, 70 were evaluable for efficacy. Although no complete responses were observed, 7 patients had partial responses, for an overall response rate of 10% in the evaluable population. The median TTP for the evaluable population was 10 weeks, and the proportion of patients who were free of disease progression at 6 months was 23%. The median overall survival was 59.4 weeks for all patients enrolled. Common, drug-related ≥ grade 3 adverse events (graded according to National Cancer Institute Common Toxicity Criteria, version 2) included diarrhea, vomiting, abdominal pain, and nausea.

CONCLUSIONS:

The combination of UFT and leucovorin administered orally in a 3-times-daily regimen was found to have modest activity. Grade 3 toxicities were manageable with appropriate dose adjustments in patients with metastatic breast cancer previously treated with anthracyclines and/or taxanes. Cancer 2010. © 2010 American Cancer Society.

The optimal regimen for treating metastatic breast cancer has not yet been determined. Thus, novel compounds and various combinations of both investigational and approved therapeutic agents are being tested in clinical studies. Among the new therapies being explored is an oral combination therapy of uracil and ftorafur (UFT; Taiho Pharmaceutical Ltd., Tokyo, Japan; BMS-200604, Bristol-Myers Squibb; Princeton, NJ) and leucovorin.

5-fluorouracil (5-FU) is frequently used to treat metastatic breast cancer and colorectal cancer. 5-FU is rapidly metabolized by the liver. Its mean plasma half-life when given as an intravenous (iv) infusion in humans is 16 minutes.1 Results of a meta-analysis of randomized advanced colorectal cancer clinical studies that compared 5-FU bolus with 5-FU continuous infusion indicated that continuous infusion was superior in terms of tumor response (14% for bolus vs 22% for continuous infusion) and median survival time (11.3 months for bolus vs 12.1 months for continuous infusion).2 The primary side effects observed with 5-FU therapy were gastrointestinal, myelotoxic, or dermatologic (hand-foot syndrome).3 When 5-FU was administered iv, the most frequently reported gastrointestinal side effects were oral mucositis and potentially severe diarrhea.4 Attempts to reduce side effects while maintaining the antitumor effects of 5-FU led to the development of ftorafur and, ultimately, of UFT.

Ftorafur (FT), also known as tegafur, is a fluorinated pyrimidine that has efficacy similar to that of 5-FU.5 Development of an injectable formulation of FT in the United States was discontinued in the 1980s due to its neurotoxicity when given at doses of 1 to 3 g. FT was subsequently developed as an oral formulation in Japan, where it is used as adjuvant therapy for gastrointestinal, head and neck, breast, and lung cancers. The 5-FU released when FT is metabolized is itself metabolized by the same pathway as iv 5-FU, and thus has the same cytotoxic mechanism.

UFT, a combination of uracil and FT in a 4:1 M ratio, was developed in Japan to combine the cytotoxic effects of 5-FU with the convenience of oral dosing. Uracil is also metabolized by dihydropyrimidine dehydrogenase (DPD) and competes with 5-FU for the enzyme when the 2 are co-administered,6 resulting in a significantly prolonged half-life for 5-FU.

Leucovorin (folinic acid), is a tetrahydrofolic acid derivative that can enhance the therapeutic and toxic effects of fluoropyrimidines such as 5-FU.1 It is combined with UFT to further potentiate the effect of 5-FU on tumor cells.

Preclinical studies using combination therapy of orally administered UFT and leucovorin revealed that UFT provides higher peak plasma concentrations of 5-FU and prolonged therapeutic 5-FU concentrations compared with continuous infusion of 5-FU. In addition, the combination may reduce the toxic effects on normal tissue seen with the administration of 5-FU or FT alone.

The oral combination therapy of UFT and leucovorin was initially developed in the United States for treating metastatic colorectal cancer. The efficacy and safety profiles of the combination have been examined in several phase 2 and phase 3 studies, which either used UFT and leucovorin alone or compared the combination with 5-FU and leucovorin.7-15 The phase 2 studies focused primarily on response rates and toxicity evaluation in metastatic colorectal cancer. Overall response rates ranged from 16.9% to 42%. Toxicity was generally less severe than that noted when 5-FU was used singly or in combination with leucovorin; the absence of hand-foot syndrome was most notable.7-11

The phase 3 studies compared the oral UFT and leucovorin combination with iv 5-FU and leucovorin.13-15 A dose of 300 mg/m2/day of UFT with 25 to 30 mg of leucovorin 3 times daily (repeated every 35 days), was the most common regimen for these metastatic colorectal cancer studies. Results from 1 study indicated that patients who received the oral combination experienced less diarrhea and mucositis, with minimal need for dose interruption or reduction, compared with patients who received iv 5-FU and leucovorin.13

5-FU has been used with leucovorin as second-line therapy in treating metastatic breast cancer. FT, with and without leucovorin, has been explored as an oral, less toxic alternative to 5-FU. In a phase 2 study of patients with previously treated metastatic breast cancer, 25 patients received 750 mg/m2/day of FT with 45 mg/day of leucovorin for 21 days, repeated every 28 days.16 An overall response rate of 32% was observed, and 48% of patients experienced disease stabilization (median duration, 6 months). The median time to disease progression (TTP) was 9 months. The results of this study were striking considering the patients' advanced conditions and number of previous chemotherapy regimens. The most common grade 3 toxicities observed were mucositis (24% of patients) and diarrhea (12%).

In the current study, UFT was divided into 3 daily doses, each administered with 30 mg of leucovorin during the first 28 days of a 35-day cycle. The effectiveness of combining UFT with leucovorin in a 3 times a day regimen was examined in patients with metastatic breast cancer.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Patient Eligibility

Eligible patients had histologically confirmed breast cancer with metastases and had received prior chemotherapy with anthracyclines and/or taxanes as adjuvant treatment, neoadjuvant treatment, or as treatment for metastatic disease were eligible. Treatment with at least 1, but not more than 2, regimens of prior chemotherapy for metastatic disease was required. Eastern Cooperative Oncology Group (ECOG) performance status had to be ≤2. Patients who demonstrated no evidence of disease progression since their last treatment regimen were excluded, as were patients who had received prior treatment with UFT, capecitabine, ethynyl uracil, or low-dose continuous infusion 5-FU. The patient must have had adequate hematologic, renal, and liver function. Before the initiation of the study, the protocol was approved by an Institutional Review Board and, for each patient, written informed consent was obtained and documented by the appropriate signatures.

Treatment

Patients were given 300 mg/m2/day of UFT, in 3 doses of 100 mg/m2 each, given orally every 8 hours. A 30-mg dose of leucovorin was given with each dose of UFT. Patients received study medication for the first 28 days of each 35-day cycle. If toxicity occurred, the UFT dose was decreased in increments of 50 mg/m2/day. Once the UFT dose level was decreased, it was not to be increased. If a patient required a dose reduction that brought her below 200 mg/m2/day, she was removed from the study.

The dose of leucovorin was not modified unless the UFT dose was actually withheld, in which case the leucovorin was also withheld. If the UFT dose was withheld due to toxicity, the patient resumed treatment once the adverse effects resolved. If the patient did not fulfill re-treatment criteria within 2 weeks, she was removed from the study. For patients remaining in the study, treatment was continued until disease progression occurred or intolerable toxicities developed.

Note that, if study medication was withheld due to toxicity, the days it was withheld were counted as treatment days. Therefore, administration of study medication did not extend beyond Day 28.

Endpoints

The primary endpoint was TTP from the on-study date. Patients who had no disease progression were censored at their off-study date. Planned secondary endpoints were overall tumor response rate (OR), overall survival (OS), and toxicity.

Disease was categorized as measurable, evaluable, or nonevaluable. Measurable disease response criteria were defined as follows. A complete response (CR) was defined as the complete disappearance of all tumor lesions for at least 4 weeks from the date of documentation of complete response. A partial response (PR) was defined as a decrease by >50% in the sum of the products of the largest perpendicular dimensions of all measurable lesions, as determined by 2 consecutive observations at least 4 weeks apart. No lesions, measurable or not, should have progressed, and no new lesions should have appeared. Stable disease (SD) was considered when there was a failure to observe remissions as defined above, in the absence of any progressive or new lesions, as determined by 2 consecutive observations at least 4 weeks apart. Progressive disease (PD) was defined as an increase of ≥25% in the size of any measurable or evaluable lesion, and/or the appearance of any new lesions or the occurrence of malignant pleural effusion or ascites.

Evaluable disease response criteria were defined as follows. A CR was the complete disappearance of all known disease for at least 4 weeks. For bone lesions, complete recalcification of lytic lesions or normalization of a bone scan for blastic lesions. Normalization of tumor markers, if previously abnormal. A PR was defined as an estimated decrease in tumor size of ≥50% for at least 4 weeks. For bone lesions, PR was defined as substantial recalcification of lytic lesions. SD was used when there was no significant change in disease status for at least 4 weeks, including unchanged disease, estimated decrease of <50%, and lesions with estimated increase of <25%. The appearance of any new lesion not previously identified, or an estimated increase of ≥25% in existing lesions was considered as PD. Death secondary to malignant disease was documented as PD.

Response Evaluations

Tumor response was evaluated every 2 cycles, and the evaluations were continued until disease progression. Patients who were removed from the study before the first tumor response assessment were not considered evaluable for response unless there was clear evidence of clinical disease progression. All suspected responses were documented with a repeat measurement at 4 weeks.

Statistical Analyses

A total of 81 eligible patients were needed to demonstrate that a 6-month progression-free survival proportion of 30% had a 95% confidence interval (95% CI) between 20% and 40%. A total of 90 patients were planned. The 95% CIs were calculated for rates of 6-month progression-free survival and overall response. The Kaplan-Meier estimator was used to examine overall TTP and OS. Cox proportional hazards regression models were used to relate TTP with patient age, ECOG performance status, estrogen/progesterone receptor (ER/PR) status, disease-free interval, presence of visceral metastases, presence of bone metastases, number of prior chemotherapy regimens, and prior hormone therapy.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Between April 1999 and September 2001, 91 patients, all women, were enrolled at 17 sites. The median age of the patients was 54 years. Of these 91 patients, 2 withdrew consent before receiving any study medication, leaving 89 patients evaluable for safety. Patient baseline characteristics, including ECOG performance status, prior therapy, and sites of metastases are summarized in Table 1. Sixty-five (71.4%) patients received prior hormonal therapy, 57 (62.6%) received prior adjuvant chemotherapy, and 86 (94.5%) had received prior chemotherapy for metastatic breast cancer (55 patients with 1 prior regimen, 30 patients with 2 prior regimens, and 1 patient with 3 prior regimens). Five patients did not receive prior chemotherapy for metastatic disease and were therefore considered ineligible.

Table 1. Patient Demographics and Baseline Characteristics: All Patients (N=91)
CharacteristicValue
  • ECOG indicates Eastern Cooperative Oncology Group; 5-FU, 5-fluorouracil.

  • a

    Most frequently reported sites of metastases (>10% of patients enrolled).

  • b

    Based on 89 evaluable patients.

Sex, no. (%) 
Female91 (100)
Male0
Race, no. (%) 
 White71 (78)
 Hispanic10 (11)
 Black9 (9.9)
 Other1 (1.1)
Age, y 
 Median54
 Range34-87
ECOG performance status, no. (%) 
 033 (36.3)
 150 (54.9)
 28 (8.8)
Prior therapy, no. (%) 
 Chemotherapy91 (100)
 Prior adjuvant chemotherapy57 (62.6)
 Prior metastatic chemotherapy86 (94.5)
 Prior 5-FU bolus regimen52 (57.1)
 Radiotherapy66 (72.5)
 Hormone therapy65 (71.4)
 Immunotherapy2 (2.2)
 Bone marrow transplantation8 (8.8)
Sites of metastases, no. (%)a,b 
 Bone58 (65.9)
 Lung48 (54.5)
 Liver40 (45.5)
 Lymph33 (37.5)
 Skin9 (10.2)

Seventy-two patients (79.1%) had visceral metastases; 58 (63.7%) had bone metastases. Fifty-five patients (60.4%) had ER-positive tumors, 29 (31.9%) had ER-negative tumors, and 7 (7.7%) had unknown ER status.

The metastatic disease-free interval (calculated from initial diagnosis to recognition of metastatic disease) was ≥12 months for 74 patients (81.3%), between 6 and 12 months for 2 patients (2.2%), and <6 months for 15 patients (16.5%).

Two populations were analyzed for efficacy: the eligible population and the efficacy-evaluable population. The eligible population included all patients who received at least 1 dose of study medication and was used for the safety and tolerability analysis. The eligible population excluded 8 ineligible patients: 5 had received prior adjuvant therapy only and had not received prior chemotherapy for metastatic disease, 1 patient continued to receive tamoxifen after beginning the study, 1 patient's prestudy blood chemistries were not obtained, and 1 patient received pamidronate within 4 months of study initiation and had bone-only disease. The evaluable population excluded 7 of the 8 ineligible patients because, although 1 patient was deemed ineligible because prestudy blood chemistries were not obtained, the patient completed 2 cycles of therapy and received tumor assessment at the end of cycle 2. In addition, 14 patients were excluded from the evaluable population because they did not have a tumor assessment after receiving at least 2 cycles of therapy. Only the evaluable population will be discussed in this document.

Summary of Drug Administration

A total of 294 cycles of therapy was administered; the mean was 3.3 cycles. The average number of treatment days per cycle was 25, and the average dose administered per day was 281.25 mg/m2.

Drug administration was summarized separately for patients with prior bone marrow transplant (8 patients) and those without (81 patients). For the 8 patients with prior bone marrow transplant, 20 cycles were administered, with an average of 2.5 cycles (range, 1‒5 cycles). The average number of treatment days per cycle was 25.1, and the average dose per administration was 294.7 mg/m2. For the patients without prior bone marrow transplant; 274 cycles were administered, with an average of 3.4 cycles (range, 1‒25 cycles). The average number of treatment days per cycle was 25.5, and the average dose per administration was 280.28 mg/m2.

Dose Modifications

Dose reductions were required for 18 patients (20.2%), most frequently for grade 2 or grade 3 diarrhea. Of these patients, only 1 required a second dose reduction; both reductions were for grade 3 diarrhea. Dose interruptions due to noncompliance occurred in 10 patients (11.2%). The reason for dose interruptions in an additional 19 patients (21.3%) was unknown. Four patients (4.5%) were removed from the study due to toxicity; 2 of these patients had already undergone dose reduction.

The percentage of patients who experienced dose modification or interruption has been summarized for all 89 patients who received study therapy. Twenty-five patients (28.0%) had drug administration interruptions for a maximum of less than 5 days, 21 patients (24.0%) had drug administration interruptions for a maximum of 5 to 10 days, and 8 patients (9.0%) had drug administration interruptions for >10 days. Patients with drug administration interruptions due to toxicity were as follows: 22 patients (25.0%) had drug administration interruptions for a maximum of less than 5 days, 18 patients (20.0%) had drug administration interruptions for a maximum of 5 to 10 days, and 6 patients (7.0%) had drug administration interruptions for >10 days.

Efficacy

Table 2 summarizes tumor response for the 70 patients who were evaluable for efficacy. None of the patients had a complete response. Seven patients had a partial response, and the overall response rate (CR + PR) was 7 patients (10%) (95% CI, 4.5-19.6). Twenty-five patients (35.7%) had SD (95% CI, 24.9-48.0), and 38 patients (54.3%) had PD (95% CI, 42.1-66.1).

Table 2. Tumor Response: Efficacy-Evaluable Patientsa (N=70)b
ResponseNo. (%)95% CI
  • 95% CI indicates 95% confidence interval; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease.

  • a

    The efficacy-evaluable population included patients who completed a tumor response assessment after receiving at least 2 cycles of therapy.

  • b

    Ineligible and unevaluable patients were excluded.

CR + PR7 (10)4.5-19.6
CR0 
PR7 (10)4.5-19.6
SD25 (35.7)24.9-48.0
PD38 (54.3)42.1-66.1

The median TTP for the efficacy-evaluable population was 10 weeks (95% CI, 9.5-14.7). The proportion of patients who were free of disease progression at 6 months was 23% (95% CI, 13-33). A Cox proportional hazards regression model was used to relate TTP with ECOG performance status, patient age, ER/PR status, the interval between diagnosis and on-study, the presence of visceral metastases, the presence of bone metastases, the number of prior metastatic chemotherapy regimens (1 or 2), and prior hormone therapy.

Age and ER/PR status (positive vs negative) were found to be significant (P<.05) predictors of TTP. The median TTP for patients aged >65 years was 28.3 weeks, compared with approximately 10 weeks for those aged <65 years. The median TTP for patients who were ER/PR-negative was 5.3 weeks, compared with 10.6 weeks for patients who were ER and/or PR-positive.

The median overall survival (Kaplan-Meier estimate) was 59.4 weeks (95% CI, 41.1-79.4). The 6-month survival rate was 77% (95% CI, 68-86), and the 12-month survival rate was 53% (95% CI, 42-63).

Safety and Tolerability

All 89 patients experienced at least 1 adverse event. The most commonly reported drug-related adverse events included nausea (59.6%), diarrhea (53.9%), asthenia (42.7%), and vomiting (36%).

Side effects were graded using the National Cancer Institute Common Toxicity Criteria version 2 (NCI CTC v.2). Twenty-eight patients (31.5%) experienced at least 1 adverse event that was grade 3 in severity and was considered related to study medication. The most commonly reported of these were diarrhea (19.1%), vomiting (6.7%), abdominal pain (5.6%), nausea (5.6%), and asthenia (4.5%). Diarrhea was the only treatment-related adverse event that occurred with a grade 3 intensity in >10% of patients. Severe treatment-related stomatitis was noted in only 1 patient.

Twelve patients (13.5%) experienced at least 1 grade 3 or greater serious adverse event (SAE) that was considered related to study medication. Study drug related SAEs included nausea, diarrhea, vomiting, intestinal obstruction, abdominal pain, dehydration, stomatitis, and choreoathetosis.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

The combination of UFT and leucovorin given orally was designed to be a potentially less toxic and more convenient alternative therapy to the combination of 5-FU and leucovorin given by iv infusion. This multicenter phase 2 study evaluated the safety and efficacy of the oral combination given 3 times daily for the first 28 days of a 35-day cycle in patients with recurrent metastatic breast cancer who had been previously treated with anthracyclines and/or taxanes.

Of the 70 patients who were evaluable for efficacy, 7 had partial responses, for an overall response rate of 10%. The median TTP was approximately 10 weeks; the median overall survival was 59.4 weeks. Approximately 25% of patients were free of disease progression at 6 months. The most commonly reported drug-related adverse events included nausea, diarrhea, asthenia, and vomiting. Approximately one-third of patients experienced at least 1 grade 3 or higher adverse event.

These findings suggest that the oral combination therapy of UFT and leucovorin, given in a 3-times-daily regimen, produces objective responses. Grade 3 toxicities were manageable with appropriate dose adjustments in patients with metastatic breast cancer previously treated with anthracyclines and/or taxanes.

CONFLICT OF INTEREST DISCLOSURES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Supported by Bristol Myers Squibb Co., Princeton, New Jersey.

Dr. Hortobagyi has received honoraria for consultancy or advisory roles from Bristol-Myers Squibb.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES
  • 1
    Physician's Desk Reference. 52nd ed. New York: Thomson Healthcare; 1998: 2463-2464.
  • 2
    Rougier P, Buyse M, Ryan L, et al. Meta-analysis of all studies comparing intravenous bolus administration to continuous infusion of 5-fluorouracil in patients with advanced colorectal cancer [abstract]. Proc Am Soc Clin Oncol. 1997; 16: 267a. Abstract 946.
  • 3
    Schilsky RL. Antimetabolites. In: PerryMC, ed. The Chemotherapy Source Book. Baltimore, Md: Williams and Wilkins; 1992: 301-317.
  • 4
    Fata F, Ron IG, Kemeny N, O'Reilly E, Klimstra D, Kelsen DP. 5-Fluorouracil-induced small bowel toxicity in patients with colorectal carcinoma. Cancer. 1999; 86: 1129-1134.
  • 5
    Taguchi T. Experience with UFT in Japan. Oncology. 1997; 11( suppl 10): S30-S34.
  • 6
    Maehara Y, Kakeji Y, Ohno S, Baba H, Sugimachi K. Scientific basis for the combination of tegafur with uracil. Oncology. 1997; 11( suppl 10): S14-S21.
  • 7
    Pazdur R, Lassere Y, Rhodes V, et al. Phase II study of uracil and tegafur plus oral leucovorin: an effective oral regimen in the treatment of metastatic colorectal carcinoma. J Clin Oncol. 1994; 12: 2296-2300.
  • 8
    Saltz LB, Leichman CG, Young CW, et al. A fixed-ratio combination of uracil and ftorafur (UFT) with low dose leucovorin. Cancer. 1995; 75: 782-785.
  • 9
    Gonzalez-Baron M, Feliu J, de la Gandara I, et al. Efficacy of oral tegafur modulation by uracil and leucovorin in advanced colorectal cancer. A phase II study. Eur J Cancer. 1995; 31A: 2215-2219.
  • 10
    Abad A, Navarro M, Sastre J, et al. UFT plus oral folinic acid as therapy for metastatic colorectal cancer in older patients. Oncology. 1997; 11( suppl 10): S53-S57.
  • 11
    Feliu J, Gonzalez-Baron M, Espinosa E, et al. Uracil and tegafur modulated with leucovorin. Cancer. 1997; 79: 1884-1889.
  • 12
    Ron IG, Lotan A, Inbar MJ, Chaitchik S. Advanced colorectal carcinoma: redefining the role of oral tegafur. Anticancer Drugs. 1996; 7: 649-654.
  • 13
    Skillings JR. 5-FU or UFT combine with leucovorin for previously untreated metastatic colorectal cancer. Oncology. 1997; 11( suppl 10): S48-S49.
  • 14
    Carmichael J. UFT plus leucovorin versus 5-FU plus leucovorin for metastatic colorectal cancer. Oncology. 1997; 11( suppl 10): S50-S52.
  • 15
    Mamounas E, Wieand HS, Jones J, Wickerham DL, Wolmark N. Future directions in the adjuvant treatment of colon cancer. Oncology. 1997; 11( suppl 10): S44-S47.
  • 16
    Sole LA, Albanell J, Bellmunt J, Ribas A, Gallego O, Carulla J. Phase II study of an all-oral regimen of tegafur and folinic acid in patients with previously treated metastatic breast cancer. Cancer. 1995; 75: 831-835.