|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 25, No 15 (May 20), 2007: pp. 2057-2062 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.07.7776 Multiple Drug Resistance in Osteogenic Sarcoma: INT0133 From the Children's Oncology Group
From the Hasbro Children's Hospital/Brown Medical School, Providence, RI; Montefiore Medical Center, Bronx; Memorial Sloan Kettering Cancer Center, New York, NY; Children's Hospital of Pittsburgh, Pittsburgh, PA; Keck School of Medicine, University of Southern California, Los Angeles; Children's Oncology Group Operations Center, Arcadia, CA; Dana-Farber Cancer Institute and Children's Hospital, Boston, MA; and Hôpital Sainte-Justine Montreal, Quebec, Canada Address reprint requests to Cindy L. Schwartz, MD, Hasbro Children's Hospital/Brown Medical School, 593 Eddy St MP117, Providence, RI 02903; e-mail: cschwartz1{at}lifespan.org
Purpose Multiple drug resistance due to P-glycoprotein (P-gp) expression has been reported to be a cause of disease recurrence in osteosarcoma. Tumor specimens derived from children and young adults with osteosarcoma enrolled onto a national Intergroup trial (INT0133) were analyzed prospectively to determine the role of multiple drug resistance in osteosarcoma. Patients and Methods From October 15, 1992, to November 25, 1997, 685 patients with localized, high-grade osteosarcoma were enrolled onto INT0133. Paraffin-embedded diagnostic tumor specimens were assayed for P-gp using monoclonal antibodies C-494 (139 patients) and JSB-1 (133 patients). Percent necrosis at the time of definitive surgery (NEC), event-free survival (EFS), and overall survival (OS) were evaluated as outcome measures for patients with P-gp–positive disease and were compared with patients with P-gp–negative disease. Results P-gp expression in the biopsy specimen did not significantly increase the risk for adverse outcomes as measured by EFS, OS, or NEC. EFS for those patients with C-494–positive tumors was 59% at 4 years versus 61% at 4 years for patients with C-494–negative tumors (P = .79), or 58% at 4 years versus 61% at 4 years for patients with JSB-1–positive versus JSB-1–negative tumors (P = .65). OS for patients with C-494–positive tumors was 82% at 4 years versus 82% at 4 years for patients with C-494–negative tumors (P = .61). Conclusion Prospective analysis of the role of multiple drug resistance in localized osteosarcoma did not find that immunohistochemical analysis of P-gp expression predicted outcome for patients treated on INT0133.
Chemotherapy has improved the cure rate of patients with localized osteosarcoma from the 15% to 20% achieved with surgery alone to approximately 70%.1,2 Nonetheless, nearly one third of patients with localized osteosarcoma experience recurrent or progressive disease; most patients whose disease recurs ultimately die as a result of the disease. In the mid-1970s, Rosen et al3 noted that response to initial chemotherapy was predictive of long-term outcome; others have subsequently confirmed this finding.4,5 Although Rosen postulated that resistance to methotrexate played a significant role in disease recurrence,3 substitution of methotrexate by cisplatin in the T-10 regimen (doxorubicin, methotrexate, bleomycin, dactinomycin, cyclophosphamide, with cisplatin for poor response) did not improve the outcome for those with poorly responsive disease.3 Given that patients also received doxorubicin during the induction phase, an alternative hypothesis could be that adverse outcomes are attributable to doxorubicin resistance rather than to methotrexate resistance. A meta-analysis highlights the prognostic significance of doxorubicin dose in sarcomas.6 In the 1960s, Kessel et al7 described a phenomenon of multiple-drug resistance in which a cell line resistant to one chemotherapeutic agent was also resistant to unrelated agents. A transmembrane protein known as P-glycoprotein (P-gp) was found to be an adenosine triphosphate–dependent efflux pump that facilitates the efflux of hydrophobic substances from the cell. Many chemotherapeutic agents (including doxorubicin, vincristine, dactinomycin, etoposide, and others) are hydrophobic and susceptible to P-gp–mediated extrusion from the cell.8,9 Improved survival has been reported for patients with P-gp–negative osteosarcoma.1,10-19 In 1992, the Children's Cancer Group (CCG) and Pediatric Oncology Group (POG) initiated a national Intergroup trial (INT0133) for the treatment of children and young adults with osteosarcoma.10 Patient samples were analyzed prospectively to determine the role of multiple-drug resistance in osteosarcoma. This report describes the results of this investigation.
Patients and Treatment From October 15, 1992, to November 25, 1997, 685 patients with localized, high-grade osteosarcoma at 116 institutions of the CCG and the POG were enrolled onto INT0133. Eligibility criteria included age younger than 30 years and adequate renal, cardiac, and liver function without prior chemotherapy or radiation. Patients with metastatic disease were eligible for INT0133 at institutions of the CCG, but are not included in the current analysis. Written informed consent was obtained according to the institutional review board guidelines from the patient or his/her parent or legal guardian. All patients were treated with high-dose methotrexate, cisplatin, and doxorubicin, and had complete surgical resection of their primary tumor.10 Patients underwent a double randomization to receive or not receive ifosfamide as part of therapy from enrollment, as well as to receive or not receive muramyl tripeptide phosphatidylethanolamine (MTP-PE), an immune modulator, from the time of definitive surgery (Fig 1). Tumor specimens were requested at the time of diagnosis, as well as at the time of definitive surgery. For this study, results are based on specimens obtained at diagnosis.
Assay for P-Glycoprotein There were designated reference laboratories for the POG (L.T.) and for the CCG (R.G.); each performed immunohistochemical analysis for P-gp as noted in the following paragraph. Tumor tissue was sent directly to the reference laboratory in formaldehyde (POG only) or sections of paraffin-embedded material were cut at 4 to 5 µm. If sections were prepared at the local institution, it was recommended that they were prepared on poly-L-lysine slides. Antigen retrieval was performed through digestion with 0.05% trypsin followed by microwave treatment for 10 minutes. Slides were saturated with 0.05% bovine serum albumin then preincubated with normal horse serum (Cappel Research, Durham, NC) at 1:20 dilution in 2% bovine serum albumin/phosphate-buffered saline for 15 minutes at room temperature. Slides were then incubated for 16 hours with JSB-1 (Signet Laboratories, Dedham, MA) at a dilution of 1:20 and C-494 at a dilution of 1:20 (Signet Laboratories) with 2% bovine serum albumin/phosphate-buffered saline to detect P-gp. Slides were rinsed with phosphate-buffered saline for 30 minutes; biotinylated antimouse immunoglobulin G (Vector Laboratories, Burlingame, CA) was applied at 1:500 dilution in 1% bovine serum albumin/phosphate-buffered saline for 60 minutes at room temperature. After rinses with phosphate-buffered saline for 30 minutes, slides were incubated with peroxidase-conjugated streptavidin (Dako Corp, Carpinteria, CA) at a 1:500 dilution in 1% bovine serum albumin/phosphate-buffered saline for 45 minutes, then rinsed in phosphate-buffered saline for 30 minutes. Color was developed by incubating slides in 0.06% diaminobenzidine in phosphate-buffered saline for 15 minutes. Slides were then rinsed with tap water, counterstained with Harris modified hematoxylin (Fisher, Pittsburgh, PA) rinsed for 2 seconds in 1% acid alcohol and 15 seconds in ammonia water, dehydrated, and covered with a coverslip. Normal adrenal gland (CCG) or normal kidney (POG) was used as a positive control for P-gp. Negative controls in which the primary antibody was omitted were included with each run. Each sample was scored by a pathologist blinded to patient identity. For P-gp, any membrane staining on tumor cells above background was considered positive.
Statistical Analysis Marker positivity. All patients' marker status was determined as one of the following three categories: positive, negative, or undetermined. C-494 and JSB-1 were all evaluated separately. Concordance. The concordances of the various markers were cross-tabulated. Discordance rates were calculated as the fraction of the total evaluated population on which one of the selected markers was positive and the other selected marker was negative. Event-free survival. This was taken to be the time from study entry until relapse, diagnosis of a second malignancy, death, or last patient contact, whichever occurred first. Patients who experienced a relapse, a second malignancy, or who died were considered to have experienced an event. Otherwise, the patient was considered censored for event-free survival (EFS) at last contact. This outcome measure was used for the assessment of the prognostic significance of marker status at the time of study enrollment. Overall survival. This was taken to be the time from study entry until death or last patient contact, whichever occurred first. Patients who died were considered to have experienced an event. Otherwise, the patient was considered censored for overall survival (OS) at last contact. Necrosis grading. Patients who underwent definitive resection after induction had evaluation of necrosis by the institutional pathologist. Necrosis was graded according to the method of Huvos, as modified by CCG.1,11 Briefly, the percent necrosis in the definitive surgical specimen was categorized according to the following scheme: grade 1, no effect identified; grade 2, 1% to 95% viable tumor in the specimen; grade 3, scatter foci, but totaling less than 5% viable tumor remaining; and no viable tumor remaining. The equality of the distribution of qualitative characteristics was compared across groups of patients by means of the exact conditional test of proportions.12 The outcome of patients who were evaluated for the relevant marker was compared with patients who were not evaluated for the marker. The relative risk of an EFS event (disease progression, diagnosis of second malignant neoplasm, or death before disease progression or second malignant neoplasm) or death in patients who were not evaluated for the marker was estimated using a proportional hazards regression model.20 Survivor functions for EFS and OS were estimated using the method of Kaplan and Meier.20 Given that those evaluated for the relevant markers were comparable to those who were not evaluated, we analyzed the data set further to determine the association between marker positivity and outcome among the subset of individuals on whom marker status could be determined (analytic cohort). The relative risk of EFS event or death in this group of patients was estimated using a proportional hazards regression model.20 Randomization in this study was stratified by elevated lactate dehydrogenase at enrollment (yes v no), primary site above the knee or elbow (yes v no), and definitive surgery performed before enrollment (yes v no). Therefore, all relative risk estimates and CIs were obtained using a proportional hazards regression model stratified by these two characteristics. Only 12 patients among the analytic cohort had definitive surgery before enrollment, and this characteristic was not considered as a stratification factor.
Patients Specimens were evaluated using C-494 in 139 samples and JSB-1 in 133 samples. Patients who were evaluated for C-494 and JSB-1 were similar to each other and to the larger group of patients not examined for the marker with respect to clinical characteristics and the outcome measures examined (Table 1).
Although 685 patients were enrolled onto this study and their specimens stained with hematoxylin and eosin were sent for pathology review, the numbers with specimens available for P-gp immunohistochemistry were significantly limited. This was because tumor tissue had to be sent directly to the reference laboratory in formaldehyde or the institution had to send either blocks of paraffin material or sections of paraffin-embedded material that were cut at 4 to 5 µm, preferably prepared on a poly-L-lysine slide. Furthermore, several patients were diagnosed at institutions that were not members of POG or CCG. Properly prepared tumor tissue often was not available for these individuals.
Analysis Based on Initial Pretreatment Tumor Specimen Outcome: EFS, OS, and necrosis grading. Five hundred eight patients were alive at last patient contact for this analysis. The median follow-up time for such patients was 7.7 years. Patients with tumors that demonstrated antibody positivity at study entry were not at significantly increased risk for adverse events as defined above. This was true for all markers evaluated (Figs 2A and 2B). The adjusted relative risk for an EFS event for patients with C-494–positive tumors was 1.0 (95% CI, 0.58 to 1.8). The adjusted relative risk for an EFS event for patients with JSB-1–positive tumors was 1.2 (95% CI, 0.69 to 2.1).
Given that cellular efflux of ifosfamide is not mediated by the P-gp efflux pump, the ifosfamide theoretically could have mitigated any adverse impact of P-gp. The risk for an EFS event was similar regardless of whether the patient was randomly assigned to receive ifosfamide. For patients who were not randomly assigned to receive ifosfamide, the relative risk for an EFS event for patients who were C-494 positive was 1.3 (95% CI, 0.53 to 3.0); for those who were JSB-1 positive, the relative risk for an EFS event was 1.4 (95% CI, 0.63 to 3.1). For patients who were randomly assigned to receive ifosfamide, the relative risk for an EFS event for patients who were C-494 positive was 0.84 (95% CI, 0.38 to 1.9); for those who were JSB-1 positive, the relative risk for an EFS event was 0.98 (95% CI, 0.45 to 2.2). A similar pattern was observed for the risk associated with marker positivity within groups defined by randomization to receive MTP-PE. For patients who were randomly assigned not to receive MTP-PE, the relative risk for an EFS event for patients who were C-494 positive was 1.8 (95% CI, 0.78 to 4.2); for those who were JSB-1 positive, the relative risk for an EFS event was 1.8 (95% CI, 0.81 to 4.1). For patients who were randomly assigned to receive MTP-PE, the relative risk for an EFS event for patients who were C-494 positive was 0.67 (95% CI, 0.30 to 1.5); for those who were JSB-1 positive, the relative risk for an EFS event was 1.0 (95% CI, 0.58 to 1.8). Patients with tumors that demonstrated antibody positivity at study entry were not at significantly increased risk for death as defined above (Figs 3A and 3B. The adjusted relative risk for death for patients C-494–positive tumors was 1.0 (95% CI, 0.52 to 1.1). The adjusted relative risk for an EFS event for patients with JSB-1–positive tumors was 1.2 (95% CI, 0.63 to 2.4). Marker positivity was not related to risk of death within groups defined by randomization to receive ifosfamide or MTP-PE (data not shown).
Of the 139 patients evaluated for C-494 and 133 patients evaluated for JSB-1, 122 and 123, respectively, also had information on necrosis grading in the surgical specimen. The odds ratios for a poor grading (grade 1 or II v III or IV) associated with antibody positivity derived from this data set are 1.3 for C-494 (P = .56) and 1.2 for JSB-1 (P = .71).
This prospective analysis of the role of multiple-drug resistance in localized osteosarcoma did not find that immunohistochemical analysis of P-gp expression predicted outcome for patients treated during INT0133. This is in contrast to retrospective results,13-19 in which the expression of P-gp was a significant adverse prognostic factor for patients with osteosarcoma. A meta-analysis by Pakos et al21 has also suggested that P-gp is associated with an increased risk of disease progression. However, the largest prospective study previously published did not find an association of P-gp with outcome.22 The necrosis grading after induction chemotherapy for osteosarcoma has long been known to be the major prognostic factor for outcome of patients with initially localized disease.3-5 Nonetheless, in none of the aforementioned studies has a correlation between percent necrosis after induction chemotherapy and P-gp expression been noted. Our study also evaluated necrosis grading versus P-gp expression and did not find a correlation. One could hypothesize that the impact of P-gp on necrosis was not apparent at the early time point at which necrosis is evaluated because high-dose methotrexate and cisplatin are included in the INT0133 induction regimens. Given that neither of these agents is affected by the P-gp efflux pump, tumor necrosis could be significant despite doxorubicin resistance. Small numbers of residual, resistant cells might still lead to ultimate relapse. Long-term outcome could still be affected by doxorubicin resistance, as determined by P-gp expression. We used C-494 and JSB-1 as antibodies. Although they recognize different epitopes of the P-gp, the results were not statistically different. Different antibodies and methods of staining theoretically could influence the likelihood of finding a significant correlation between P-gp positivity and EFS; the literature nonetheless does not suggest that one antibody is a better predictor of outcome in osteosarcoma.13-19,21-23 Because of the retrospective nature of the study, the mean follow-up for patients on some studies14,15 is longer than for the patients on this study. P-gp has not been associated with a more proliferative or aggressive cellular phenotype23; instead, it may represent resistance in more slowly growing cells that persist despite chemotherapy. Although Scotlandi et al23 did note that metastasis occurred earlier in those with P-gp–negative osteosarcoma compared with P-gp–positive osteosarcoma, patients enrolled onto our trial are already a mean of 7.7 years from the time of diagnosis. Late relapses could not account for the difference in predicted value noted between our study and that of Chan et al,14 who reported a recurrence rate of more than 90% in those with P-gp–expressing tumors versus 10% in those whose tumors were P-gp negative. Although methodology can differ from laboratory to laboratory, both of our reference laboratories independently found similar proportions in patients who were P-gp positive. The results from neither group, CCG nor POG, were associated significantly with prognosis as determined by the occurrence of an analytic event. Our study is unique in that it was prospective; samples were requested at initial diagnosis. No bias in selection for patients with good or adverse outcome would have occurred using this approach. The effect of P-gp could have been mitigated by the chemotherapeutic agents added to the chemotherapy regimen. Half of the patients enrolled onto this study were randomly assigned to receive ifosfamide, an agent for which P-gp is not an effective efflux pump. If this were the dominant explanation, then patients receiving ifosfamide would have been influenced positively by the use of ifosfamide. This was not shown to be the case. Patients were also randomly assigned to MTP-PE, an immune modulator. There is no evidence to support a hypothesis that this might affect drug efflux. In addition, another POG study using cyclosporine to overcome the effect of P-gp in combination with doxorubicin and cisplatin in patients with recurrent osteosarcoma showed a low response rate.24 These results do not provide sufficient evidence to support the modification of treatment regimens for patients whose tumors express P-gp. On the basis of this study, inhibitors of P-gp would not be predicted to affect outcome. In addition, doxorubicin remains a major component of chemotherapy treatment for patients with osteosarcoma. It should not be eliminated from the chemotherapy regimen, even in those patients with P-gp–expressing tumors.
Although all authors completed the disclosure declaration, the following authors or their immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors. Employment: N/A Leadership: N/A Consultant: Richard Gorlick, Oncolytics Inc Stock: N/A Honoraria: N/A Research Funds: N/A Testimony: N/A Other: N/A
Conception and design: Cindy L. Schwartz, Richard Gorlick, Mark Krailo, Holcombe E. Grier, Paul Meyers Administrative support: Cindy L. Schwartz Provision of study materials or patients: Cindy L. Schwartz, Richard Gorlick, Mark Krailo, Allen Goorin, Paul Meyers Collection and assembly of data: Cindy L. Schwartz, Richard Gorlick, Zhengjia Chen, Holcombe E. Grier, Paul Meyers Data analysis and interpretation: Cindy L. Schwartz, Richard Gorlick, Lisa Teot, Mark Krailo, Zhengjia Chen, Holcombe E. Grier, Paul Meyers Manuscript writing: Cindy L. Schwartz, Richard Gorlick, Lisa Teot, Mark Krailo, Holcombe E. Grier, Mark L. Bernstein, Paul Meyers Final approval of manuscript: Cindy L. Schwartz, Richard Gorlick, Lisa Teot, Mark Krailo, Zhengjia Chen, Holcombe E. Grier, Mark L. Bernstein, Paul Meyers
Supported by Grants No. U10 CA 98543, CA 30969, and CA 13539 from the National Institutes of Health/National Cancer Institute. A complete listing of grant support for research conducted by Children's Cancer Group and Pediatric Oncology Group before initiation of the Children's Oncology Group grant in 2003 is available online at: http://www.childrensoncologygroup.org/admin/grantinfo.htm. Presented in part at the 36th Annual Meeting of the American Society of Clinical Oncology, May 20-23, 2000, New Orleans, LA. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Provisor AJ, Ettinger LJ, Nachman JB, et al: Treatment of nonmetastatic osteosarcoma of the extremity with preoperative and postoperative chemotherapy: A report from the Children's Cancer Group. J Clin Oncol 15: 76-84, 1997 2. Goorin AM, Schwartzentruber DJ, Devidas M, et al: Presurgical chemotherapy compared with immediate surgery and adjuvant chemotherapy for nonmetastatic osteosarcoma: Pediatric Oncology Group Study POG-8651. J Clin Oncol 21: 1574-1580, 2003 3. Rosen G, Caparros B, Groshen S, et al: Primary osteogenic sarcoma of the femur: A model for the use of preoperative chemotherapy in high risk malignant tumors. Cancer Invest 2: 181-192, 1984[Medline] 4. Bacci G, Avella M, Brach Del Prevert A, et al: Neoadjuvant chemotherapy for osteosarcoma of the extremities: Good response of the primary tumor after preoperative chemotherapy with high-dose methotrexate followed by cisplatinum and adriamycin—Preliminary results. Chemioterapia 7: 138-142, 1988[Medline] 5. Winkler K, Beron G, Delling G, et al: Neoadjuvant chemotherapy of osteosarcoma: Results of a randomized cooperative tri (COSS-82) with salvage chemotherapy based on histological tumor response. J Clin Oncol 6: 329-337, 1988[Abstract] 6. Smith MA, Ungerleider RS, Horowitz ME, et al: Influence of doxorubicin dose intensity on response and outcome for patients with osteogenic sarcoma and Ewing's sarcoma. J Natl Cancer Inst 83: 1460-1470, 1991 7. Kessel D, Hall TC, Roberts D: Uptake and retention of daunomycin by mouse leukemic cells as factors in drug response. Cancer Res 28, 1968 8. Riordan JR, Ling V: Purification of P-glycoprotein from plasma membrane vesicles of Chinese hamster ovary cell mutants with reduced colchicine permeability. J Biol Chem 254: 12701-12705, 1979 9. Gottesman MM, Pastan I: The multidrug transporter, a double-edged sword. J Biol Chem 263: 12163-12166, 1988 10. Meyers PA, Schwartz CL, Krailo M, et al: Osteosarcoma: A randomized, prospective trial of the addition of ifosfamide and/or muramyl tripeptide to cisplatin, doxorubicin, and high-dose methotrexate. J Clin Oncol 23: 2004-2011, 2005 11. Meyers PA, Heller G, Healey J, et al: Chemotherapy for nonmetastatic osteogenic sarcoma: The Memorial Sloan-Kettering experience. J Clin Oncol 10: 5-15, 1992[Medline] 12. Armitage P, Berry G: Statistical Methods in Medical Research. London, United Kingdom, Blackwell Scientific, 1971 13. Schwartz CL, Rosier R, Willis J, et al: P-Glycoprotein (P-GP) expression in osteosarcoma (O.S.) and Clinical Outcome. Proc Am Soc Clin Oncol 11: 285, 1992 14. Chan HS, Grogan TM, Haddad G, et al: P-glycoprotein expression: Critical determinant in the response to osteosarcoma chemotherapy. J Natl Cancer Inst 89: 1706-1715, 1997 15. Baldini N, Scotlandi K, Barbanti-Brodano G, et al: Expression of P-glycoprotein in high-grade osteosarcomas in relation to clinical outcome. N Engl J Med 333: 1380-1385, 1995 16. Serra M, Scotlandi K, Reverter-Branchat G, et al: Value of P-glycoprotein and clinicopathologic factors as the basis for new treatment strategies in high-grade osteosarcoma of the extremities. J Clin Oncol 21: 536-542, 2003 17. Hornicek FJ, Gebhardt MC, Wolfe MW, et al: P-glycoprotein levels predict poor outcome in patients with osteosarcoma. Clin Orthop Relat Res 373: 11-17, 2000[CrossRef][Medline] 18. Park YB, Kim HS, Oh JH, et al: The co-expression of p53 protein and P-glycoprotein is correlated to a poor prognosis in osteosarcoma. Int Orthop 24: 307-310, 2001[CrossRef][Medline] 19. Yamamoto O, Wada T, Takahashi M, et al: Prognostic value of P-glycoprotein expression in bone and soft-tissue sarcoma. Int J Clin Oncol 5: 164-170, 2000[CrossRef] 20. Kalbfleisch JD, Prentice RL: The Statistical Analysis of Failure Time Data. New York, NY, John Wiley and Sons, 1980 21. Pakos EE, Ioannidis JP: The association of P-glycoprotein with response to chemotherapy and clinical outcome in patients with osteosarcoma: A meta-analysis. Cancer 98: 581-589, 2003[CrossRef][Medline] 22. Wunder JS, Bull SB, Aneliunas V, et al: MDR1 gene expression and outcome in osteosarcoma: A prospective, multicenter study. J Clin Oncol 18: 2685-2694, 2000 23. Scotlandi K, Serra M, Nicoletti G: Multidrug resistance and malignancy in human osteosarcoma. Cancer Res 56: 2434-2439, 1996 24. Schwartz CL, Bernstein M, Devitas M, et al: Cyclosporin A to overcome multiple drug resistance in osteogenic sarcoma. Med Pediatr Oncol 37: 177, 2001 Submitted June 5, 2006; accepted February 16, 2007.
Related Correspondence
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
