|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology Randomized, Double-Blind, Placebo-Controlled Trial to Evaluate the Hematopoietic Growth Factor PIXY321 After Moderate-Dose Fluorouracil, Doxorubicin, and Cyclophosphamide in Stage II and III Breast CancerFrom the Texas Oncology, PA, Physician Reliance Network, and Sammons Cancer Center of Baylor University Medical Center, Dallas, TX; Hematology Clinic, Portland, OR; North Shore University Hospital, Manhasset, NY; Michigan State University, East Lansing, MI; and Immunex Corp, Seattle, WA. Address reprint requests to Stephen E. Jones, MD, Sammons Cancer Center, Baylor University Medical Center, 3535 Worth St, Collins Building, Floor 5, Dallas, TX 75246.
PURPOSE: To measure the effect of PIXY321 (granulocyte-macrophage colony-stimulating factor/interleukin-3 S. cerevisiae fusion protein) on the incidence, duration, and complications of neutropenia and thrombocytopenia after moderate-dose fluorouracil 600 mg/m2, doxorubicin 60 mg/m2, and cyclophosphamide 750 mg/m2 (FAC) chemotherapy in patients with stage II and III breast cancer. PATIENTS AND METHODS: In this multicenter, randomized, double-blind placebo-controlled trial, 71 women were to receive four 21-day cycles of treatment with moderate-dose FAC chemotherapy by short intravenous infusion on day 1, followed by either placebo or PIXY321 (375 µg/m2 subcutaneously twice a day) on days 3 to 15. All patients were to receive prophylactic oral ciprofloxacin when the absolute neutrophil count was less than 1,000/µL. RESULTS: PIXY321 significantly reduced the incidence and duration of grade 3 and grade 4 neutropenia in cycles 1 and 2 and the duration of grade 3 neutropenia in cycles 1 through 4. In cycles 3 and 4, grade 3 thrombocytopenia was significantly more common with PIXY321 (P < .05). Two patients, both in the PIXY321 group, required platelet transfusions. Fever and hospitalization for intravenous antibiotics were significantly more common in the PIXY321 group during cycle 1 only. More patients in the PIXY321 group achieved hematologic recovery by day 22 in cycles 1 through 3, and time to recovery was significantly shorter with PIXY321 in all cycles. FAC dose intensity was roughly 2% higher in the PIXY321 group (P = NS). Nonhematologic events of any intensity occurring with significantly greater overall frequency in the PIXY321 group included injection-site reactions, fever, chills, abdominal pain, and arthralgia. No patient died on study or within 30 days of her last dose of study drug. CONCLUSION: PIXY321 decreased the incidence and duration of FAC-induced grade 3 and 4 neutropenia in cycles 1 and 2 and significantly shortened the time to hematologic recovery in all cycles. However, it produced more systemic toxicity as well as thrombocytopenia in cycles 3 and 4.
PIXY321 IS A FUSION protein consisting of recombinant human granulocyte-macrophage colony-stimulating factor (rhuGM-CSF) and recombinant human interleukin-3 (rhuIL-3) joined by a flexible amino acid linker sequence, which allows the binding domains to fold into their native configurations and interact with their independent receptor binding sites. GM-CSF helps regulate the intermediate stages of hematopoiesis, promoting neutrophil and monocyte proliferation, but shows no consistent effect on erythroid or megakaryocytic lineages. IL-3 regulates the early and intermediate stages of hematopoiesis, promoting proliferation and differentiation of multipotent, as well as committed, progenitor cells of the myeloid, erythroid, and megakaryocytic cell lineages. Therapeutic administration of GM-CSF rapidly increases neutrophil count, whereas IL-3 has a more gradual effect, over a greater number of hematopoietic functions, that often persists beyond cessation of therapy. In preclinical studies, the sequential administration of GM-CSF and IL-3 resulted in enhanced circulatory levels of neutrophils and platelets. In some studies, simultaneous administration proved superior to sequential administration.1 PIXY321 was created to combine and enhance the properties of its component cytokines. In vitro, PIXY321 exhibits a five- to 10-fold greater affinity for binding to the IL-3 receptors on KG-1 cells than does IL-3 alone. PIXY321 is 10- to 20-fold more potent than GM-CSF, IL-3, or combined GM-CSF/IL-3 in stimulating the formation of granulocytic/monocytic colony-forming units, erythroid burst-forming units, and granulocytic/erythroid/ monocytic/megakaryocytic colony-forming units. The stimulation of these types of colonies suggests PIXY321 promotes the proliferation of hematopoietic progenitor cells responsible for the development of mature human granulocytes, macrophages, platelets, and RBCs. The purpose of this multicenter phase III study was to test the hypothesis that PIXY321 can significantly decrease granulocytopenia and thrombocytopenia associated with moderate-dose fluorouracil, doxorubicin, and cyclophosphamide (FAC) chemotherapy administered in an adjuvant or neoadjuvant setting to patients with stage II and III breast cancer and, thereby, potentially reduce the complications associated with chemotherapy. If successful, this might also permit increased dose-intensity.2-5
Patient Selection This randomized, double-blind, placebo-controlled trial was open to women 18 to 70 years of age with histologically confirmed stage II or stage III adenocarcinoma of the breast. Patients diagnosed with operable disease were required to have undergone complete excision of the primary tumor within the 10 weeks before enrollment. Patients diagnosed with locally advanced disease (inflammatory or noninflammatory) were eligible if they were judged good candidates for neoadjuvant (preoperative) chemotherapy with moderate-dose FAC. Other selection criteria included Karnofsky performance status (KPS)  80% and adequate renal, hematologic, and hepatic function as evidenced by the following: serum creatinine less than 2.0 mg/dL; WBC count greater than 3,000 cells/µL; absolute neutrophil count (ANC) 1,400 cells/µL; hemoglobin 10.0 g/dL; platelet count (PLTC) 100,000/µL; AST less than 3.0 times the institutional upper limit; and bilirubin 2.0 mg/dL. No patient had evidence of metastatic disease. Eligibility criteria based on the patient's medical history and present state of health included the following: no history of malignancy in the previous 4 years, except squamous or basal cell skin carcinoma or stage I cervical carcinoma in situ; no history of human immunodeficiency virus positivity; no history of myocardial infarction or congestive heart failure; no clinically significant arrhythmia, uncontrolled angina pectoris, uncontrolled severe high blood pressure, or hemodynamically significant valvular disease at the time of enrollment; and no serious intercurrent illness or serious active infection (pneumonia, peritonitis, wound abscess, and so on). Pregnant and lactating women were excluded, and women at risk for pregnancy were required to be practicing a medically approved form of contraception. Concomitant endocrine therapy, radiotherapy, immunotherapy, or chemotherapy were not allowed. Also, chemotherapy or radiotherapy were not allowed in the 4-year period preceding the diagnosis of breast cancer. Patients could not have been treated with IL-3 or GM-CSF at any time in the past or treated with another cytokine in the 4 weeks preceding the study. Patients had to be mentally competent to understand the potential risks and benefits of participation in the study, and each gave written informed consent in accord with institutional and regulatory guidelines.
Treatment Plan
A new cycle of chemotherapy could not be started unless the patient had achieved hematologic recovery, defined as ANC Standard antiemetic therapy was allowed, but dexamethasone could be given only on the day of FAC dosing. During episodes of grade 3 neutropenia (ANC < 1,000/mm2), patients were to receive a prophylactic oral antibiotic. Ciprofloxacin (500 mg orally bid) was preferred, but norfloxacin (400 mg bid) or trimethoprim/sulfamethoxazole (one double-strength tablet bid) were acceptable. Platelet transfusion was recommended for patients with severe thrombocytopenia (PLTC < 20,000/µL). Patients with PLTCs greater than 20,000/µL were not to receive platelet transfusions, except for significant bleeding. In a patient with grade 3 to 4 toxicity that was attributable to the study drug but did not affect a major organ, continued treatment was at the discretion of the principal investigator. Grade 3 to 4 toxicity that was attributable to the study drug and affected a major organ required the study drug be discontinued for the remainder of the cycle. If the toxicity failed to resolve by day 35, the patient was removed from the study.
Patient Evaluations and Dosage Adjustments
Throughout the study, patients kept a diary in which they recorded body temperature, time and date of temperature readings, time and date of study drug administration, and adverse events. Temperature was taken in the morning three times weekly (Monday, Wednesday, and Friday) and immediately before study drug administration. A patient experiencing fever (temperature > 38.2°C) and severe neutropenia (ANC < 500 cells/µL) was to be evaluated by a physician, with the option of either continuing outpatient management with prophylactic oral antibiotics or hospitalization for IV antibiotics. A patient hospitalized to receive IV antibiotics, for fever during a period of severe neutropenia lasting Within 21 days after day 1 of the last cycle, all patients were to undergo final clinical and laboratory evaluations. If hematopoietic recovery was not achieved by day 21 of cycle 4, monitoring of the patient's hematologic status three times a week was to continue until recovery was complete. Follow-up evaluations to gather data on disease status and survival were scheduled at 6 months and 12 months after study and yearly, thereafter, for the next 4 years.
Clinical End Points and Statistical Considerations
Patient Characteristics Between March 1994 and April 1995, 71 women enrolled onto this study at five United States sites. Thirty-seven patients were randomized to receive PIXY321, and 34 were to receive placebo. The treatment groups were well matched, with no significant differences detected in demographic variables or baseline disease assessments (Table 1). Mean age was 47 years (range, 23 to 67 years); mean weight was 76 kg (range, 46 to 136 kg); mean BSA was 1.7 m2 (range, 1.4 to 2.2 m2); and mean KPS was 98% (range, 90% to 100%). Ninety percent of patients had node-positive breast cancer. Seventy-two percent had stage II disease. All patients completed at least one cycle of chemotherapy and were assessable for safety and efficacy.
Drug Delivery
Hematologic Effects, Recovery, and Dose Intensity
In the first treatment cycle, despite a significantly lower incidence of grade 4 neutropenia, more patients in the PIXY321 group were hospitalized for IV antibiotics (seven patients v zero placebo patients (P = .008). The difference can probably be attributed to the higher incidence of fever (temperature > 38.2°C) in cycle 1, affecting 10 patients (28%) in the PIXY321 group but only one patient (3%) in the placebo group (P = .01). Overall, the incidence of hospitalization for fever and neutropenia was 22% for the PIXY321 group and 6% for the placebo group (P = .087).
The incidence and duration of thrombocytopenia, grade 3 (PLTC < 50,000/µL) and grade 4 (PLTC < 25,000/µL), in each cycle are listed in Tables 5 and 6. In general, thrombocytopenia was of brief duration. Five (15%) of 34 patients receiving PIXY321 and one (3%) of 34 patients receiving placebo experienced grade 4 thrombocytopenia in cycles 3 and/or 4 (P = .197). Grade 3 thrombocytopenia was significantly more common and of longer duration in the PIXY321 group during cycles 3 and 4. Mean platelet nadirs in cycle 3 were 81,000/µL among patients in the PIXY321 group and 128,000/µL in the placebo group (P < .001); mean platelet nadirs in cycle 4 were 70,000/µL and 127,000/µL, respectively (P < .001). No patient in the placebo group required a platelet transfusion, although one had a PLTC less than 20,000/µL in cycle 3. Two patients in the PIXY321 group experienced platelet nadirs less than 20,000/µL and received a total of three platelet transfusions, two patients in cycle 3 and one patient in cycle 4.
Six patients in the PIXY321 group received a total of 15 units of packed RBCs, and one patient in the placebo group received 2 units of RBCs during cycle 1. In the PIXY321 group, one patient received 2 units in cycle 1; in cycle 3, three patients received 2 units each, and one received three units; and in cycle 4, two patients received 2 units each.
In the PIXY321 group, the proportion of patients who achieved hematologic recovery (ANC
Among patients completing four cycles of chemotherapy, the mean dose intensity of each FAC component was roughly 2% higher in the PIXY321 group, but the difference was not statistically significant (P = .1) (Table 8).
Tolerability
Grade 3 and 4 nonhematologic laboratory toxicities were uncommon. In the PIXY321 group, one patient experienced grade 3 elevation of serum glutamic-oxaloacetic transaminase in cycle 3; one had grade 3 elevation of total bilirubin in cycle 4; and one had grade 4 low albumin in cycle 3. In the placebo group, one patient had grade 4 low albumin in cycle 1.
Tumor Response and Survival
PIXY321 was developed to combine the distinctive but complementary hematopoietic functions of GM-CSF and IL-3 in a single molecule, with the hope that its therapeutic administration would enhance both neutrophil and platelet production in cytopenic patients. In uncontrolled preliminary clinical evaluations, PIXY321 seemed to reduce the number of days to neutrophil and platelet engraftment in lymphoma patients undergoing high-dose chemotherapy and autologous bone marrow transplantation6; enhance neutrophil and platelet recovery in patients receiving carboplatin and doxorubicin for advanced gastrointestinal malignancy7; increase neutrophil and platelet nadirs in patients with ovarian cancer treated with cyclophosphamide and carboplatin (without increasing dose intensity)8; decrease the duration of neutropenia and thrombocytopenia in pediatric patients receiving ifosfamide, carboplatin, and etoposide (ICE) for solid tumors9; improve neutrophil and platelet recovery in children treated with ICE chemotherapy for recurrent solid tumors10; and reduce the duration and degree of neutropenia and prevent cumulative thrombocytopenia in sarcoma patients receiving cyclophosphamide, doxorubicin, and dacarbazine.11 To further assess its utility in cancer chemotherapy, we compared PIXY321 with placebo in 71 women receiving four 21-day courses of moderate-dose FAC for stage II and III breast cancer. The primary end point of our study was duration of ANC less than 500 cells/µL in cycle 1. Secondary end points included duration of ANC less than 500 cells/µL in cycles 2 to 4; duration of ANC less than 1,000 cells/µL in cycles 1 to 4; incidence of grade 3 and 4 thrombocytopenia in cycles 1 to 4; duration of hospitalization for neutropenia with fever; and FAC dose intensity. We chose a placebo control over GM-CSF or granulocyte colony-stimulating factor (G-CSF) to more accurately characterize the efficacy and tolerability of PIXY321. However, our study design (target population, FAC dosing regimen, ANC threshold for antibacterial prophylaxis, and so on) was identical with one used in an earlier placebo-controlled trial of once-daily GM-CSF (sargramostim 250 µg/kg/d),12 the results of which provide a useful clinical context for comparison with the present study. In the earlier trial,12 GM-CSF significantly reduced the duration of grade 3 to 4 neutropenia in all four cycles of treatment. Overall, the median duration of ANC less than 1,000/µL was 6.0 days with GM-CSF versus 9.1 days with placebo (P < .001); median duration of ANC less than 500/µL was 2.8 days versus 6.8 days, respectively (P < .001). Because significantly fewer GM-CSF patients had ANC nadirs less than 1,000/µL, fewer required antibacterial prophylaxis. Six patients in the GM-CSF group and eight patients in the placebo group were hospitalized for febrile neutropenia. GM-CSF also significantly increased mean FAC dose intensity among patients completing two or more cycles of therapy (P < .001). PIXY321 produced results similar to those seen with GM-CSF in the earlier trial,12 but with a seemingly less impressive effect on neutrophils in the later cycles (cycles 3 and 4). The smaller sample size of our study, 72 patients versus 142 patients, could have obscured the effect of PIXY321 on granulocyte recovery in cycles 3 and 4. Positive effect on granulocyte production in the PIXY321 group was seen in all cycles; the proportion of PIXY321 patients achieving hematologic recovery by day 22 was significantly greater in cycles 1 to 3, and time to hematologic recovery was significantly shortened in cycles 1 to 4 (Table 7). On the whole, we found the use of PIXY321 complicated by mild systemic toxicity. Most nonhematologic events were of grade 1 and 2 intensity. Injection-site reaction, fever, chills, and arthralgia were significantly more common among patients receiving PIXY321 (P < .05), and a significantly higher incidence of fever with PIXY321 resulted in more frequent hospital admissions (22% v 5% with placebo, P = .087) (Table 4). Because the hope for the PIXY321 cytokine was to promote thrombopoiesis, as well as granulocytopoiesis, we were surprised that PIXY321 was associated with a significant degree of late-cycle thrombocytopenia. The incidence of PLTC less than 50,000/µL in cycle 3 was 27% with PIXY321 and 3% with placebo (P < .008) and, in cycle 4, 40% with PIXY321 and 3% with placebo (P < .001). The incidence of late-cycle grade 4 thrombocytopenia (PLTC < 25,000/µL) was greater in the PIXY321 group, but the difference was not significant. Two patients in the PIXY321 group experienced PLTCs less than 20,000/µL and required a total of three platelet transfusions in cycles 3 and 4. No placebo patient required a platelet transfusion. Late-cycle thrombocytopenia was seen in our previous GM-CSF trial; six GM-CSF patients and one placebo patient had PLTCs less than 20,000/µL in cycles 3 and 4, but the duration of those episodes exceeded 48 hours in only two cases, both in the GM-CSF group.12 Late-cycle thrombocytopenia has also been observed in other studies with GM-CSF13-16 and G-CSF.17 It is our impression that PIXY321 was not as tolerated as GM-CSF in the earlier trial,12 although no direct comparison was made. In that study, only 3% of chemotherapy cycles in either treatment group were complicated by hospital admission for febrile neutropenia, which compares with the 3% incidence observed among placebo-treated patients in the present study. In part, the low incidence of febrile neutropenia reflects our policy, in both studies, of using prophylactic antibiotics while the ANC is < 1,000/µL. Despite the use of prophylactic antibiotics in the current study, 19% of patients treated with PIXY321 were admitted for IV antibiotics in cycle 1 because of fever and neutropenia, compared with none who received placebo (P = .008). However, there was an apparent "learning curve" for using PIXY321 and assessing fever in patients on this trial. Apparently, after the first cycle, our investigators suspected that fever was sometimes a side effect of PIXY321 (even though the study was blinded and placebo-controlled). For example, fever and chills, associated with injection-site reactions, were clinically suspicious for drug reactions to PIXY321 even in a blinded study, and the incidence of admission for febrile neutropenia fell in the PIXY321 treatment group for subsequent cycles based on individual clinical judgment in each instance. Lastly, although we observed no obvious problems with compliance in our study, it should be noted that a once-daily subcutaneous injection schedule with GM-CSF or G-CSF is more convenient for the patient than twice-daily dosing with PIXY321. The once daily schedule also reduces the potential for injection-site reaction. PIXY321 and GM-CSF have been directly compared in a prospective, randomized trial of five 21-day courses of fluorouracil, leucovorin, doxorubicin, and cyclophosphamide (FLAC) chemotherapy in patients with advanced breast cancer (42% stage II/III; 58% stage IV).18 Fifty-three women received either PIXY321 375 µg/m2 bid or GM-CSF 250 µg/m2 once daily from day 4 until hematologic recovery. More patients in the PIXY321 group had chills and local reactions, and more stopped PIXY321 therapy because of poor tolerability. The duration of ANC less than 1,000 cells/µL for all cycles was significantly shorter in the GM-CSF group, and no intergroup differences were seen in platelet nadirs, duration of thrombocytopenia, or need for platelet transfusions. The investigators concluded that PIXY321 was not superior to GM-CSF in ameliorating cumulative FLAC-induced thrombocytopenia and was not as tolerated as GM-CSF. Our experience with PIXY321 in the present study and with GM-CSF in a previous study12 supports their conclusions. In summary, PIXY321 seems to be less well tolerated than currently approved cytokines (GM-CSF and G-CSF) because of a higher incidence of systemic toxicity. In addition, administration of PIXY321 in this trial failed to improve thrombopoiesis. Indeed, PIXY321 seemed to produce late-cycle thrombocytopenia in some patients. Based on this trial, we do not recommend further evaluation of PIXY321 in the adjuvant treatment of breast cancer.
Supported by Immunex Corporation, Seattle, WA.
1. Farese AM, Williams DE, Seiler FR, et al: Combination protocols of cytokine therapy with interleukin-3 and granulocyte-macrophage colony-stimulating factors in a primate model of radiation induced marrow aplasia. Blood82:3012-3016, 1993 2. Ang PT, Buzdar AU, Smith T, et al: Analysis of dose intensity in doxorubicin-containing adjuvant chemotherapy in stage II and III breast carcinoma. J Clin Oncol7:1677-1684, 1989[Abstract] 3. Hryniuk WM: Average relative dose intensity and the impact on design of clinical trials. Semin Oncol14:65-74, 1987[Medline] 4. Neidhart JA: Dose-intensive treatment of breast cancer supported by granulocyte-macrophage colony-stimulating factor (GM-CSF). Breast Cancer Res Treat20:15-23, 1991 5. Dhingra K, Esparza L, Valero V, et al: Neoadjuvant chemotherapy (Chemo) of breast cancer: Impact of G-CSF on received dose intensity (DI). Proc Am Soc Clin Oncol12:95, 1993 (abstr 182)
6.
Vose JM, Anderson JE, Bierman PJ, et al: Phase I/II trial of PIXY321 to enhance engraftment following autologous bone marrow transplantation for lymphoid malignancy. J Clin Oncol14:520-526, 1996 7. Taylor C, Modiano M, Garrison L, et al: PIXY plus carboplatin (C) and Adriamycin (A) in patients with advanced gastrointestinal (GI) malignancy. Proc Am Soc Clin Oncol12:195, 1993 (abstr 569) 8. Runowicz CD, Mandeli J, Speyer J, et al: Phase I/II study of PIXY321 in combination with cyclophosphamide (CTX) and carboplatin (CP) in the treatment of patients (PTS) with ovarian cancer. Am J Obstet Gynecol174:1151-1160, 1996[Medline] 9. Furman WL, Marina N, Luo X, et al: A pediatric phase I/II trial of subcutaneous (SQ) PIXY321 administered after ifosfamide/carboplatin/etoposide (ICE) chemotherapy. Proc Am Soc Clin Oncol13:418, 1994 (abstr 1429) 10. Cairo MS, Weinthal JA, Garrison L, et al: Results of a phase I trial of PIXY321 following ifosfamide, carboplatin, and etoposide (ICE) chemotherapy in children with recurrent solid tumors (RST): Improved multilineage hematopoietic reconstitution. Proc Am Soc Clin Oncol14:255, 1995 (abstr 693) 11. Vadhan-Raj S, Papadopoulous NE, Burgess MA, et al: Effects of PIXY321, a granulocyte-macrophage colony-stimulating factor/interleukin-3 fusion protein, on chemotherapy-induced multilineage myelosuppression in patients with sarcoma. J Clin Oncol12:715-724, 1994[Abstract] 12. Jones SE, Schottstaedt MW, Duncan LA, et al: A randomized double-blind prospective trial to evaluate the effects of rhu GM-CSF (sargramostim) versus placebo in a moderate-dose FAC adjuvant chemotherapy program for stage II and III breast cancer. J Clin Oncol14:2976-2983, 1996[Abstract]
13.
Neidhart JA, Mangalik A, Stidley CA, et al: Dosing regimen of granulocyte-macrophage colony-stimulating factor to support dose-intensive chemotherapy. J Clin Oncol10:1460-1469, 1992
14.
Gerhartz HH, Englehard M, Meusers P, et al: Randomized, double-blind, placebo-controlled, phase III study of recombinant human granulocyte-macrophage colony-stimulating factor as adjunct to induction treatment of high-grade malignant non-Hodgkin's lymphoma. Blood82:2329-2339, 1993 15. Ardizzoni A, Venturini M, Sertoli MR, et al: Granulocyte–macrophage colony-stimulating factor (GM-CSF) allows acceleration and dose intensity increase of CEF chemotherapy: A randomised study in patients with advanced breast cancer. Br J Cancer69:385-391, 1994[Medline]
16.
Hamm J, Schiller JH, Cuffie C, et al: Dose-ranging study of recombinant human granulocyte-macrophage colony-stimulating factor in small-cell lung carcinoma. J Clin Oncol12:2667-2676, 1994 17. Trillet-Lenoir V, Green J, Manegold C, et al: Recombinant granulocyte colony stimulating factor reduces the infectious complications of cytotoxic chemotherapy. Eur J Cancer 29A:319-324, 1993
18.
O'Shaughnessy JA, Tolcher A, Riseberg D, et al: Prospective, randomized trial of 5-fluorouracil, leucovorin, doxorubicin, and cyclophosphamide chemotherapy in combination with the interleukin-3/granulocyte-macrophage colony-stimulating factor (GM-CSF) fusion protein (PIXY321) versus GM-CSF in patients with advanced breast cancer. Blood87:2205-2211, 1996 Submitted May 4, 1998; accepted June 30, 1999.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 1999 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
