|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2005.03.6285 on February 21 2006 © 2006 American Society of Clinical Oncology. Prognostic Factors in Primary Cutaneous B-Cell Lymphoma: The Italian Study Group for Cutaneous Lymphomas
From the Institute of Hematology and Oncology "Seràgnoli," University of Bologna, Bologna; Department of Biomedical Sciences and Human Oncology, Section of Dermatology; Department of Genetics, Biology and Medical Chemistry, Section of Medical Statistics, University of Turin, Turin; Department of Dermatological Sciences and Department of Human Pathology and Oncology, University of Florence, Florence; Department of Dermatology IRCCS Ospedale Maggiore of Milan and University of Milan-Bicocca, Milan; Institute of Dermatology "Immacolata"; "Tor Vergata" University; Dipartimento Biotecnologie Cellulari ed Ematologia, University "La Sapienza," Rome; University of Ancona, Ancona; Unit of Dermatology, University of Padua, Padua; Anatomic Pathology Section, Department of Human Pathology, University of Pavia and IRCCS Policlinico S. Matteo Pavia, Pavia; Department of Hematology, Ospedale Maggiore and University of Milano, Milano, Italy Address reprint requests to Pier Luigi Zinzani, MD, Istituto di Ematologia e Oncologia "L. e A. Seràgnoli", Policlinico S. Orsola, Via Massarenti 9, 40138 Bologna, Italy; e-mail: plzinzo{at}med.unibo.it
PURPOSE: Primary cutaneous B-cell lymphomas (PCBCLs) are a distinct group of primary cutaneous lymphomas with few and conflicting data on their prognostic factors. PATIENTS AND METHODS: The study group included 467 patients with PCBCL who were referred, treated, and observed in 11 Italian centers (the Italian Study Group for Cutaneous Lymphomas) during a 24-year period (1980 to 2003). All of the patients were reclassified according to the WHO–European Organisation for Research and Treatment of Cancer (EORTC) classification. RESULTS: Follicle center lymphoma (FCL) accounted for 56.7% of occurrences, followed by marginal-zone B-cell lymphoma (MZL; 31.4%); diffuse large B-cell lymphoma (DLBCL), leg type, was reported in 10.9% of patients. Radiotherapy was the first-line treatment in 52.5% of patients and chemotherapy was the first-line treatment in 24.8% of patients. The complete response rate was 91.9% and the relapse rate was 46.7%. The 5- and 10-year overall survival (OS) rates were 94% and 85%, respectively. Compared with FCL/MZL, DLBCL, leg type, was characterized by statistically significant lower complete response rates, higher incidence of multiple cutaneous relapses and extracutaneous spreading, shorter time to progression, and shorter OS rates. The only variable with independent prognostic significance on the OS was the clinicopathologic diagnosis according to the WHO-EORTC classification (DLBCL, leg-type, showed a significantly worse prognosis v FCL and MZL; P < .001), whereas the only variable with independent prognostic significance on disease-free survival was the presence of a single cutaneous lesion (P = .001). CONCLUSION: Our study identifies a possible PCBCL subclassification and the extent of cutaneous involvement as the two most relevant prognostic factors in PCBCL. These data can be considered reasonably as the clinical background for an appropriate management strategy.
The term primary cutaneous B-cell lymphoma (PCBCL) was introduced in the early 1990s to identify a heterogeneous group of lymphoproliferative disorders with distinctive clinical features characterized by a clonal proliferation of B lymphocytes primarily involving the skin.1-4 Controversy developed about the classification system to be used for clinical purposes. In fact, both the Revised European-American Lymphoma5 and the WHO6 classifications for non-Hodgkin's lymphomas do not deal specifically with cutaneous lymphoma, even if they can be adapted to include most of the entities primarily involving the skin. Conversely, the European Organisation for Research and Treatment of Cancer (EORTC)7 classification, which has been tailored specifically to cutaneous lymphoma, identifies three major groups of PCBCL: follicle center cell lymphoma, immunocytoma/marginal zone lymphoma, and the large B-cell lymphoma (LBCL) of the leg. The proposed WHO-EORTC classification8 recognizes four main types: (1) primary cutaneous marginal-zone B-cell lymphoma (MZL; primary cutaneous plasmacytoma, which is exceedingly rare and overlaps with primary cutaneous MZL, is included in this spectrum); (2) primary cutaneous follicle center lymphoma (FCL); (3) primary cutaneous diffuse large B-cell lymphoma (DLBCL), leg type; and (4) primary cutaneous diffuse large B-cell lymphoma, other. It is well known that PCBCL generally is associated with a favorable prognosis, with good response rates to radiotherapy and chemotherapy. However, the relapse rate varies widely between 25% to 68% according to the different studies.9-14 Herein, we report a large retrospective Italian multicenter cohort study of 467 PCBCL patients, collected from the groups participating in the Italian Study Group for Cutaneous Lymphomas (GILC) and followed up over 24 years. The main objective of the present study was to identify, by means of multivariate analyses, the main clinicopathologic parameters with independent values relating to disease-free survival (DFS) and overall survival (OS).
Patient Selection A retrospective review of clinical data on PCBCL patients seen during a period of 24 years in 11 Italian centers (GILC) from 1980 to 2003 was performed. PCLBL was defined as a B-cell lymphoma presenting in the skin with no evidence of extracutaneous spread within 6 months after diagnosis. Staging procedures at first diagnosis included brain, chest, and abdomen computed tomography scan, upper digestive tract and nasopharynx endoscopies, and bone marrow biopsy. Lymph node biopsy was performed in the presence of clinically detectable adenopathies. The medical records were reviewed for presentation, staging evaluation, treatment, recurrence, and status at last follow-up. A total of 543 patients were identified, among whom 76 patients were excluded. The 467 patients included in this study had complete and carefully compiled clinical information together with a pathologic diagnosis including immunohistochemistry. The follow-up time ranged from 1 to 24 years (median, 12.5 years).
Histopathology Specifically, the Kiel categories immunocytoma, plasmacytoma, monocytoid B-cell lymphoma/marginal zone cell lymphoma/low-grade malignant B-cell lymphoma of mucosa-associated lymphoid tissue type, and the EORTC categories marginal zone lymphoma/immunocytoma and plasmacytoma were included in the WHO-EORTC group MZL. The patients included in the EORTC category follicle center cell lymphoma were considered to represent bona fide FCL according to the WHO-EORTC scheme only if the pathologic diagnosis according to the Kiel classification was centroblastic-centrocytic lymphoma (including follicular, follicular and diffuse, and diffuse subtypes) or centroblastic lymphoma, multilobated subtype. Conversely, if the Kiel category was centroblastic lymphoma (excluding the multilobated subtype) or immunoblastic lymphoma, these patients were included in the WHO-EORTC group DLBCL, leg type. Finally, patients were included in the WHO-EORTC group DLBCL, leg type, if they were diagnosed with LBCL of the leg (according to EORTC classification 1997), or centroblastic lymphoma—excluding the multilobated subtype—or immunoblastic lymphoma (according to the Kiel classification). Those diagnosed with centroblastic lymphoma, multilobated subtype (according to the Kiel classification), were otherwise included in the WHO-EORTC group FCL. Finally, three patients originally classified as T-cell–rich B-cell lymphoma, plasmablastic lymphoma, and intravascular LBCL were included in the WHO-EORTC group DLBCL, other.
Response Criteria and Statistical Analysis The hazard functions were used to evaluate the incidence of relapses. The hazard of relapse was defined as the probability per time unit that a patient, who was disease free at the beginning of the respective interval, will experience relapse in that interval. It was computed as the number of recurrences per time units in the respective interval, divided by the number of disease-free patients entering the same time interval. The time intervals selected were yearly up to 15 years from the date of complete clinical response. The primary end points for both univariate and multivariate analyses were DFS and OS. DFS was calculated as the time elapsed from complete clinical response (CCR) to the first-line therapy to the date of recurrence or last disease-free follow-up visit; only patients who achieved a CCR following the first treatment were included. The disease recurrence in patients with CCR was defined as the development of new cutaneous and/or extracutaneous sites of involvement confirmed histologic analysis. OS was established from diagnosis to time of death or last known date alive; all deaths were counted as events. Time to progression (TTP) was calculated as the time elapsed from diagnosis to the date of relapse (for responding patients), progression (for nonresponding patients), or last disease-free follow-up visit. The parameters included in the univariate analysis model were patient sex and age at first diagnosis, extent of cutaneous involvement (single v regional v disseminated), site of cutaneous lesions, classification according WHO-EORTC,8 and first-line therapy (surgery v radiotherapy v chemotherapy). Life-table estimates of survival were derived by the Kaplan-Meier method17 and compared statistically using the stratified log-rank test of Mantel.18 The multivariate DFS and OS analyses, stratified for patient age at first diagnosis, were performed using the Cox proportional hazards regression model, with a stepwise selection of the significant variables. The two assumptions of the Cox proportional hazards model19 were tested. All of the parameters introduced in the multivariate analysis (patient sex, site and extent of cutaneous lesions, WHO-EORTC diagnosis, and first-line treatment) were categoric.
Clinicopathologic Features at Diagnosis The study population included 467 CBCL patients diagnosed from 1980 to 2003. There was a sharp increase in the incidence of new patients diagnosed: from 109 in the decade 1980 to 1990 up to 291 in the decade 1990 to 2000. Patients' clinicopathologic characteristics at diagnosis are listed in Table 1 according to the different histologic subsets. Patients with DLBCL, leg type, shared different clinical features compared with the other subtypes, including a significantly older age at diagnosis (median age, 70 v 55 years in MZL and 70 v 51 years in FCL patients; P < .001, Kruskal-Wallis test) and a higher incidence of regional and/or disseminated cutaneous involvement (only 33.3% of patients with a single cutaneous lesion; P < .001).
Therapeutic Approaches, Response to Treatment, and Relapse Rate Radiotherapy was performed as a rule in the presence of single or regional cutaneous lesions, whereas chemotherapy was mostly reserved for patients with DLBCL, leg type, and/or disseminated cutaneous involvement. The total dose of radiotherapy administered per patient varied between 35 and 45 Gy; more than one radiotherapy field was required in 94 (38.4%) of 245 patients because of the presence of multifocal lesions. Conversely, chemotherapy (usually a standard cyclophosphamide, doxorubicin, vincristine, and prednisone, or similar regimen) was the treatment of choice in 39.2% of DLBCL, leg type, versus only 23.1% of FCL/MZL patients. In 106 patients (22.7%) who showed a single and small cutaneous lesion, only a surgical excisional biopsy was performed; surgery alone was performed in 24.7% of FCL/MZL versus only 5.9% of patients with DLBCL, leg type (P < .001). Tables 1 and 2 summarize the treatment results according to the different therapeutic approaches and clinicopathologic subtypes. Globally, 429 patients (91.9%) achieved a CR after the first treatment. After a median follow-up of 12.5 years, a disease relapse occurred in 46.7% of the CR patients, with 23% experiencing two or more relapses. The trend in the hazard of relapse was characterized by high values in the first year (15.2%), followed by a progressive decrease down to 5.7% in the fourth year. Patients disease-free after 4 years from the date of first response maintained thereafter a risk of relapse between 1.4% and 6.4% beyond even a 10-year follow-up period (Fig 1).
Disease relapse was confined to the skin in the vast majority of patients (90.2%). An extracutaneous involvement was observed in 42 patients (mainly to lymph nodes, bone marrow, spleen, lung, and GI tract); in 28 of these 42 patients, extracutaneous involvement was associated with cutaneous lesions. Only 14 (2.9%) of 467 patients experienced an exclusively extracutaneous disease progression. The mean time until extracutaneous dissemination was 1.6 years (range, 7 months to 13.9 years). According to the clinicopathologic subtypes identified in the WHO-EORTC classification, MZL and FCL patients shared a similar disease course in terms of CCR rate (91% to 95%), relapse rate (44% to 46%), and extracutaneous spreading (6% to 11%; Table 1). Conversely, patients with DLBCL, leg type, were characterized by statistically significant lower CCR rates (82.3%; P = .02), higher incidence of multiple relapses (more than 40% of patients with two or more; P = .013), shorter TTP rates (5-year TTP, 46% v 57%; P = .039; Fig 2A) and a 1.85-fold higher incidence of extracutaneous spreading (16.7% v 9% in MZL/FCL). Regarding the overall relapse rate, LBCL, leg type, showed higher percentage values (54.8%) than MZL/FCL (44.4% and 46.5%, respectively), but the difference was not statistically significant. A higher relapse rate in DLBCL, leg type, was confirmed both for patients with single or regional lesions treated with radiotherapy and for patients with disseminated cutaneous involvement undergoing chemotherapy as first-line treatment (Table 2). Interestingly, the relapse rate of MZL and FCL patients treated with surgery alone was similar to that of patients treated with radiotherapy. No clinical benefits were demonstrated in patients undergoing multimodality treatments with respect to patients treated with chemotherapy alone.
Univariate and Multivariate Analysis of DFS and OS The 5- and 10-year OS rates were 94% and 85% for all patients, respectively (Fig 2B), whereas the 5- and 10-year DFS rates were 55% and 43%, respectively (Fig 2C). Notably, despite the low aggressiveness of the disease and the favorable prognosis, a plateau in both OS and DFS curves was not reached before the 15th year. The results of the univariate analyses of OS and DFS are summarized in Table 3. The clinicopathologic diagnosis was significantly associated to the OS: patients with MZL or FCL showed a statistically significant higher OS (5-year OS, 96.6% and 96.1%, respectively) as compared with DLBCL, leg type (5-year, 73.1%; P < .0001; Fig 2D). Other variables statistically associated with an unfavorable prognostic significance on OS were age older than 55 years, presence of regional or disseminated cutaneous lesions, lower limb localization, and first-line chemotherapy treatment.
The DFS of CR patients after the first-line treatment was influenced only by the extent of cutaneous involvement at the first diagnosis (Fig 2E). Patients with a single cutaneous lesion showed a significantly longer DFS than patients with regional or disseminated cutaneous lesions (5-year DFS, 64% v 43%). No difference in the DFS was found according to the WHO-EORTC classification or according to first-line treatment. The OS differences according to the age at first diagnosis led us to stratify patients for this variable in the multivariate analysis (Table 4). The only variable with independent prognostic significance on the OS was the clinicopathologic diagnosis according to the WHO-EORTC classification (patients with DLBCL, leg type, showed a significantly worse prognosis than those patients with MZL/FCL; P < .001). The only variable with an independent prognostic relevance on DFS was the presence of a single cutaneous lesion (P = .001). According to both the WHO-EORTC classification and extent of cutaneous involvement, the DFS of MZL/FCL patients with a single lesion was significantly higher than that of patients with regional/disseminated lesions and the same histology (5-year DFS, 62% v 44%; P = .0002); conversely, no DFS difference were found according to the extent of cutaneous involvement in patients with DLBCL, leg type (5-year, 55% for single v 44% for regional/disseminated lesions).
The OS of MZL/FCL patients with a single lesion was significantly higher than that of patients with regional/disseminated lesions and the same histology (5-year OS, 97% v 85%; P = .0026); conversely, the difference between single and regional/disseminated cutaneous involvement in patients with DLBCL, leg type (5-year OS, 79% for single v 67% for regional/disseminated lesions), was only of borderline significance (P = .06).
To our knowledge, this article presents the largest series of PCBCL patients ever published in literature, with data collected during a 24-year period. All of the patients were reclassified according to the WHO-EORTC classification.8 From a clinical point of view, the main findings of our study are as follows. First, our results confirm the good overall prognosis of PCBCL, with a 10-year OS rate of 85%, a 92% CCR rate after the first treatment, and a low tendency to extracutaneous spread (9.8%). However, nearly half of the patients experienced relapse after the first treatment (46.7%), and 23.8% experienced two or more relapses during the follow-up. The disease relapse was nearly always confined to the skin. Few patients developed an extracutaneous involvement; in most cases, this associated with a concomitant cutaneous relapse. The DFS was 55% at 5 years and 43% at 10 years. Although the higher incidence of relapses was observed in the first 4 years, a constant 2% to 6% risk of relapse was observed even beyond 10 years, and the DFS curve shows a plateau only after 15 years from the date of response. Second, our results confirm the clinical usefulness of the WHO-EORTC classification, which basically divides PCBCL patients into two main groups, characterized by significant clinicopathologic and prognostic differences: the MZL/FCL group and DLBCL, leg type, group. Willemze et al2 reported a uniquely poor prognosis associated with DLBCL localized to the lower limbs. The previous largest series, reported by Grange et al,12 analyzed 145 patients and the multivariate analysis found the EORTC classification and generalized versus localized disease to correlate with disease-specific survival. Others also identified an elevated lactate dehydrogenase,20 disseminated cutaneous disease,21,22 and extracutaneous progression23 as adverse prognostic factors. All of these studies analyzed fewer than 60 patients. Among biologic prognostic factors, bcl-2 expression was significantly related to death from lymphoma in univariate and multivariate analysis.14 The results of our study clearly show that the clinicopathologic diagnosis is the most relevant prognostic factor in PCBCL patients. In fact, the multivariate analysis (stratified for age to avoid the biases linked to the lymphoma-unrelated deaths in elderly patients) identified DLBCL, leg type, as the only unfavorable prognostic factor with independent value on OS. The worse prognosis of DLBCL, leg type, is dependent on a series of factors, namely lower CCR rates, higher incidence of multiple relapses, shorter TTP, and more frequent extracutaneous spreading. On the other hand, the overall relapse rate of DLBCL, leg type, was only slightly higher than that of MZL/FCL, and no significant DFS differences were found according to the WHO-EORTC diagnosis. The only variable with independent value on DFS was the extent of cutaneous involvement. Therefore, the risk of relapse after the achievement of CCR is independent from the treatment used to obtain the response and significantly lower in patients with a single cutaneous lesion than in patients with either regional or disseminated cutaneous involvement. Regarding therapy, polychemotherapy was usually carried out as first-line treatment in the presence of multiple cutaneous lesions untreatable by radiotherapy, or in patients with unfavorable clinicopathologic diagnosis; this explains the lower CCR rates and OS of patients treated with polychemotherapy as compared with those treated with radiotherapy. To confirm the favorable biologic behavior of PCBCLs, it is notable that the relapse rate of patients treated with surgery alone was similar to that of radiotherapy: if confirmed, this finding would mean that there is no need to perform a radiation treatment after a macroscopically complete surgical excision. Conversely, the similar DFS and OS of patients with regional and disseminated cutaneous involvement suggests the potential need of controlled studies comparing radiotherapy versus chemotherapy in these patients. Finally, the worse prognosis of DLBCL, leg type, and the high relapse rate of patients not only with regional/disseminated but also single cutaneous lesions, clearly suggests that multimodality treatments and new therapeutic strategies have to be planned. In this view, the role of rituximab in PCBCL management remains to be clarified. In this series, rituximab alone or rituximab in association with chemotherapy was administered to fewer than 20 patients, predominantly in the presence of relapsed/refractory disease after standard radiotherapy or chemotherapy. Despite the favorable results obtained (data not shown), the retrospective and nonrandomized setting of this study, together with the low number of patients treated, does not allow us to ascertain its potential clinical activity. In conclusion, our study indicates and confirms, on a large series of patients, that the WHO-EORTC classification and the extent of cutaneous involvement at the time of diagnosis are the two more relevant and statistically significant prognostic factors in PCBCL. These data can be taken reasonably as the clinical background for an appropriate management (treatment and follow-up) strategy.
The authors indicated no potential conflicts of interest.
Supported in part by BolognAIL. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Berti E, Alessi E, Caputo R, et al: Reticulohistiocytoma of the dorsum. J Am Acad Dermatol 19 : 259 -272, 1988[Medline] 2. Willemze R, Meijer CJLM, Sentis HJ, et al: Primary cutaneous large cell lymphomas of follicular center cell origin. J Am Acad Dermatol 16 : 518 -526, 1987[Medline] 3. Pimpinelli N, Santucci M, Carli P, et al: Primary cutaneous follicular center cell lymphoma: Clinical and histological aspects. Curr Probl Dermatol 19 : 203 -220, 1990[Medline] 4. Bernengo MG, Aloi F, Doveil GC, et al: Cutaneous lymphomas localized on the back: A clinical and immunohistological study of 15 cases—VI International Dermatopathology Colloquium. Firenze 1985. Italian General Review of Dermatology 23 : 31 , 1986 (abstr) 5. Harris NL, Jaffe ES, Stein H, et al: A revised European-American classification of lymphoid neoplasms: A proposal from the International Lymphoma Study Group. Blood 84
: 1361
-1392, 1994 6. Jaffe ES, Harris NL, Diebold J, et al: World Health Organization Classification of neoplastic diseases of the hematopoietic and lymphoid tissues. Am J Clin Pathol 111 : S8 -S12, 1999 (suppl 1)[Medline] 7. Willemze R, Kerl H, Sterry W, et al: EORTC classification for primary cutaneous lymphomas: A proposal from the cutaneous lymphoma study group of the European Organization for Research and Treatment of Cancer. Blood 90
: 354
-371, 1997 8. Willemze R, Jaffe E, Burg G, et al: WHO-EORTC classification of cutaneous lymphomas. Blood 105
: 3768
-3785, 2005 9. Garcia CF, Weiss LM, Warnke RA, et al: Cutaneous follicular lymphoma. Am J Surg Pathol 10 : 454 -463, 1986[CrossRef][Medline] 10. Pimpinelli M, Santucci M, Bosi A, et al: Primary cutaneous follicular centre-cell lymphoma: A lymphoproliferative disease with favourable prognosis. Clin Exper Dermatol 14 : 12 -19, 1989[CrossRef][Medline] 11. Santucci M, Pimpinelli N, Arganini L: Primary cutaneous B-cell lymphoma: A unique type of low-grade lymphoma. Cancer 67 : 2311 -2326, 1991[CrossRef][Medline] 12. Grange F, Hedelin G, Joly P, et al: Prognostic factors in primary cutaneous lymphomas other than mycosis fungoides and the Sézary syndrome: The French Study Group on Cutaneous Lymphomas. Blood 93
: 3637
-3642, 1999 13. Vermeer MH, Geelen FAMJ, van Haselen CW, et al: Primary cutaneous large B-cell lymphomas of the legs: A distinct type of cutaneous B-cell lymphoma with an intermediate prognosis. Arch Dermatol 132
: 1304
-1308, 1996 14. Grange F, Petrella T, Beylot-Barry M, et al: Bcl-2 protein expression is the strongest independent prognostic factor of survival in primary cutaneous large B-cell lymphomas. Blood 103
: 3662
-3668, 2004 15. Stansfeld AG, Diebold J, Kapanci Y, et al: Updated Kiel classification for lymphomas. Lancet 1 : 292 -293, 1988[Medline] 16. Cheson BD, Horning SJ, Coiffier B, et al: Report of an international workshop to standardize response criteria for non-Hodgkin's lymphomas. NCI Sponsored International Working Group. J Clin Oncol 17
: 1244
-1253, 1999 17. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53 : 157 -181, 1958 18. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother 34 : 163 -170, 1966 19. Cox DR: Regression models and life tables. J R Stat Soc B 34 : 187 -202, 1972 20. Joly P, Vasseur E, Esteve E, et al: Primary cutaneous medium and large cell lymphomas other than mycosis fungoides: An immunohistological and follow-up study on 54 cases—French Study Group for Cutaneous Lymphomas. Br J Dermatol 132 : 506 -512, 1995[Medline] 21. Kurtin PJ, DiCaudo DJ, Habermann TM, et al: Primary cutaneous large cell lymphomas: Morphologic, immunophenotypic, and clinical features of 20 cases. Am J Surg Pathol 18 : 1183 -1191, 1994[Medline] 22. Brice P, Cazals D, Mounier N, et al: Primary cutaneous large-cell lymphoma: Analysis of 49 patients included in the LNH87 prospective trial of polychemotherapy for high grade lymphomas—Groupe d'Etude des Lymphomes de l'Adulte. Leukemia 12 : 213 -219, 1998[CrossRef][Medline] 23. Joly P, Charlotte F, Leibowitch M, et al: Cutaneous lymphomas other than mycosis fungoides: Follow-up study of 52 patients. J Clin Oncol 9
: 1994
-2001, 1991 Submitted July 28, 2005; accepted January 11, 2006.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
