|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 24, No 9 (March 20), 2006: pp. 1421-1427 © 2006 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.04.6052 Postradiotherapy Neck Dissection for Lymph Node–Positive Head and Neck Cancer: The Use of Computed Tomography to Manage the Neck
From the Departments of Radiation Oncology, Radiology, and Otolaryngology, University of Florida College of Medicine, Gainesville, FL Address reprint requests to William M. Mendenhall, MD, Department of Radiation Oncology, University of Florida Health Science Center, P.O. Box 100385, Gainesville, FL 32610-0385; e-mail: mendewil{at}shands.ufl.edu
PURPOSE: To determine how to use node response on computed tomography (CT) to indicate the need for neck dissection. PATIENTS AND METHODS: Five hundred fifty patients with lymph node–positive head and neck cancer were treated between 1990 and 2002 with radiotherapy (RT) at a median dose of 74.4 Gy; 24% of these patients (n = 133) were treated with chemotherapy. Three hundred forty-one patients (62%) underwent planned post-RT neck dissection. Physical examination and contrast-enhanced CT were performed 30 days after completion of RT. CT images were reviewed in 211 patients for lymph node size (largest axial dimension) and presence of a focal abnormality (lucency, enhancement, or calcification). By correlating post-RT CT to neck dissection pathology, criteria associated with a low likelihood of residual disease were identified. A subset of patients who fit these criteria of radiographic response who did not undergo post-RT neck dissection was observed for recurrence. RESULTS: Radiographic complete response (rCR) was defined as the absence of any large (> 1.5 cm) or focally abnormal lymph node. Correlation of response with neck dissection pathology indicated a negative predictive value of 77% for complete clinical response and 94% for rCR. In 32 patients (median follow-up time, 3.2 years) with rCR who did not undergo post-RT neck dissection, the 5-year ultimate neck control rate (100%) and cause-specific survival rate (72%) were not significantly different from the rates of patients with a negative post-RT neck dissection. CONCLUSION: Patients with rCR 4 weeks after RT can be spared from a post-RT neck dissection regardless of initial node stage.
Approximately 40,000 patients in the United States are diagnosed each year with head and neck cancer.1 Many present with locally advanced disease that is amenable to organ-preserving radiation therapy (RT), in which neck dissection is often an important additional component of treatment.2,3 Traditionally, after definitive RT to the head and neck, patients with extensive neck disease are felt to be at higher risk for locoregional failure, and thus, neck dissection is recommended even after complete response to initial therapy.3,4 However, over the last decade, treatment for head and neck cancer has become increasingly successful, in part as a result of the incorporation of adjuvant chemotherapy and altered fractionation. Optimal management of the neck after RT remains undefined; unfortunately, no data from randomized controlled trials adequately address this issue. In the 1990s, institutions began to report acceptable neck control in patients treated with definitive RT who had a complete response in the neck by physical examination.5-8 Shortly thereafter, at the University of Florida, computed tomography (CT) was used to assess post-treatment response 4 weeks after RT. The first analysis of the data9 found that radiographic complete response (rCR; no lymph nodes > 1.5 cm and no focal lucency or extracapsular extension) was associated with a negative post-RT neck dissection specimen 97% of the time. Patients with an rCR have since been observed clinically, regardless of initial nodal staging. The primary purpose of this study is to update the correlation between complete response and neck dissection pathology using redefined CT criteria and to describe the outcome of patients who are spared neck dissection based on post-treatment response. A secondary goal is to report the outcome of lymph node–positive patients who undergo definitive RT with or without chemotherapy and to identify factors that may influence response to and success of treatment.
The records of 1,001 consecutive patients with head and neck cancer who were treated with definitive RT at the University of Florida between January 1990 and November 2002 were retrospectively reviewed under institutional review board approval. This study included 550 patients who underwent RT with curative intent for lymph node–positive squamous cell carcinoma of the oropharynx, hypopharynx, larynx, or an unknown head and neck primary site and who had no prior neck dissection or history of RT. Patient characteristics are listed in Table 1. RT usually consisted of a parallel-opposed lateral technique, as defined by CT planning, using 6-MV photons or cobalt-60, with a median dose of 74.4 Gy (range, 55.0 to 81.75 Gy) predominantly administered at 1.2 Gy/fraction (77%) twice daily to the primary field and upper neck. An off-cord reduction was made after 40 to 45 Gy, and 8 to 10 MeV electrons were used to supplement the dose to the tissues overlying the spinal cord. For oropharyngeal, hypopharyngeal, and laryngeal primary tumors, a second mucosal reduction was performed after 60 Gy. The low neck was treated with an anterior field prescribed to Dmax (the depth at which the dose is maximum) and typically received 70 Gy at 2 Gy/fraction once daily if the patient was lymph node positive and 50 Gy if the patient was clinically negative. Doses to treat an unknown head and neck primary site tended to be lower (median dose, 59.5 Gy administered at 1.8 Gy/fraction once daily), and the low neck doses were typically 60 to 70 Gy at 2 Gy/fraction once daily to the involved neck. The median total treatment time for all patients was 46 days (range, 26 to 77 days). Chemotherapy was administered in 133 patients; it was administered neoadjuvantly (67%), concurrently (32%), or both (1%) and was usually cisplatin based (78%). Neoadjuvant chemotherapy usually consisted of fluorouracil 600 to 1,000 mg/m2 over 4 to 5 days and cisplatin 80 to 100 mg/m2 bolus, with a median of three cycles administered. When cisplatin was administered concurrently, it was administered as 6 mg/m2 daily, 30 mg/m2 weekly, or bolus 80 to 100 mg/m2 for two to three cycles. Fifteen patients (10%) received intra-arterial cisplatin for four cycles, as outlined by an institutional protocol. Twenty-one patients (16%) received concurrent carboplatin and paclitaxel, which was most commonly administered in doses of 100 and 45 mg/m2, respectively, for 5 to 7 weekly cycles.
Post-treatment response of the neck was assessed by physical examination and contrast-enhanced CT (n = 211) at a median of 29 days (range, –3 to 125 days) after completion of RT. Only four patients had CT scans within 2 weeks of the completion of RT, and the cause was related to patient-specific scheduling conflicts. Complete response by physical examination (clinical complete response [cCR]) was defined as resolution of all adenopathy and induration in the initially involved areas. Complete response by CT (rCR) was defined as absence of lymph nodes of more than 1.5 cm and lymph nodes with any focal lucency, focal enhancement, or focal calcification. Two hundred eleven sets of contrast-enhanced CT images between 1995 and 2002 were available for blinded rereview on soft copy (ie, digital film that could be manipulated on a workstation). CT images were rereviewed for patients known to have neck dissection data available for correlation or, in later years, if post-RT CT was used to manage the neck. Maximum and minimum lymph node diameter (greatest axial dimension of lymph nodes in levels I to V) and number of abnormal lymph nodes were recorded for each hemi-neck. Data was collected only for those hemi-necks initially involved with disease. Lymph node abnormalities were graded jointly by one neuroradiologist and one radiation oncologist on the following 5-point scale: 0, definitely normal; 1, probably normal; 2, indeterminate; 3, probably abnormal; and 4, definitely abnormal. The radiation oncologist was able to learn the system quickly and consistently report findings in agreement with the neuroradiologist for this study. At the time of CT scan before this study, images were initially interpreted by the neuroradiologist and later agreed on by the radiation oncologist at tumor board. Examples of post-treatment CTs are included in Figures 1 through 6.
Three hundred forty-one patients (62%) underwent post-RT planned neck dissection, which was usually a modified radical neck dissection including levels I to V, at a median of 47 days (range, 23 to 175 days) after completion of RT. Two hundred eighty-six patients had a unilateral neck dissection, and 55 patients had a bilateral neck dissection. Standard pathologic processing of the neck dissection specimen involved sectioning at 4-µm slices before microscopic examination. After the first review of the data indicated high neck control rates without planned neck dissection, patients with a complete response on post-treatment CT did not routinely undergo neck dissection. Forty-eight patients did not receive planned post-RT neck dissection because of unresectable neck disease (n = 19), medical contraindications (n = 11), development of distant metastases (n = 8), refusal (n = 5), uncontrolled primary disease (n = 3), intercurrent death (n = 1), and unresectable neck disease with distant metastases (n = 1). For this study, the neck dissection specimen was correlated to post-RT CT characteristics for all patients for whom time between imaging and surgery was  60 days. This time interval was chosen because most neck dissections are performed no more than 60 days from post-treatment scan, and any greater interval could confound study results as a result of a change in burden of disease from time of CT scan (although no discrepancy was seen in the seven excluded patients in this study with a time interval > 60 days). Return follow-up appointments occurred on a 4- to 6-week basis for the first year and were gradually extended to yearly after 5 years. Patients unavailable for clinical follow-up were contacted by phone. Minimum potential follow-up was 2 years, whereas median observed follow-up was 3.3 years (range, 0.1 to 14.7 years); 23 patients (4%) were lost to follow-up and censored at time of last follow-up at a median of 5.0 years (range, 0.5 to 9.4 years) after completion of RT. Follow-up on living patients ranged from 0.5 to 14.7 years (median, 5.2 years). All data were analyzed using SAS statistical software (SAS OnlineDoc, Version 8; SAS Institute Inc, Cary, NC). Estimates of freedom from selected time-dependent end points were calculated using the Kaplan-Meier product-limit method.10 The log-rank test statistic was used to detect statistically significant differences in freedom from these end points between strata of selected explanatory variables. Multiple regression of these end points on a select group of explanatory variables was accomplished with Cox proportional hazards regression.11 Nominal logistic regression was used for the analysis of CT response.
Post-RT Neck Dissection In 550 patients, there were 704 hemi-necks initially involved with disease. Three hundred ninety-six post-RT neck dissections were performed in 341 patients. Neck dissection pathology was available in 383 hemi-necks (325 patients) and was negative in 256 hemi-necks (67% of dissections). The specimen contained a single positive node in 19%, two to three positive nodes in 9%, and four or more positive nodes in 5% of dissections. Of 121 patients with residual disease in the neck, 59 patients (of 86 with available data, 69%) had extracapsular spread (62 hemi-neck specimens).
Correlation of Post-RT Physical Examination to Neck Dissection Pathology
Correlation of Post-Treatment CT to Neck Dissection Pathology The negative predictive values of other post-RT CT criteria to predict for a negative neck dissection specimen are listed in Table 2. The presence of any lymph node with a focal abnormality or large size was associated with a positive predictive value as follows: focal lucency, 36%; focal enhancement, 46%; calcification, 31%; and size more than 1.5 cm, 32%.
Outcomes for Patients With Complete Response by CT (rCR) Sixty-one patients met criteria for an rCR. Thirty patients (49%) were stage T3-4, 37 patients (61%) were stage N2, and four patients (7%) were stage N3. With a median follow-up of 3.2 years after RT (range, 0.3 to 6.3 years), the 32 patients who did not undergo neck dissection had the same ultimate neck control at 5 years as the 29 patients who did undergo neck dissection (100%). Outcomes for these patients in relation to other patients who did and did not undergo neck dissection are listed in Table 3. Patients with a complete response by post-treatment CT who were observed did not have a significantly inferior outcome compared with patients with a negative post-RT neck dissection (5-year neck control rate, 97% v 98%, respectively; P = .4747; and cause-specific survival rate, 72% v 85%, respectively; P = .2225). Only one of 32 patients who did not undergo neck dissection because of rCR experienced a neck failure, at 0.5 years from the end of RT. He was treated with RT alone for a T2N1 soft palate cancer. Ipsilateral neck failure was suspected clinically at his 4-month follow-up examination, and CT demonstrated multiple abnormal nodes. He underwent immediate salvage neck dissection, which showed five of 21 positive nodes in the specimen. Without any further adjuvant treatment, he has no evidence of disease 6 years after salvage neck dissection.
Univariate analysis indicated a higher likelihood of rCR for patients with N1 disease (67% rCR rate for N1 patients v 21% for > N1 patients; P = .0009). There were no significant differences in the rates of rCR with regard to chemotherapy (P = .4148), primary site (P = .1282), T stage (P = .3923), differentiation (P = .5568), initially fixed lymphadenopathy (P = .6848), or fractionation (P = .2071). Multivariate analysis also indicated that only earlier nodal stage was associated with rCR (Table 4).
Neck and Survival Outcomes for All Patients The 5-year outcome rates were as follows: neck control, 90%; local control, 84%; cause-specific survival, 69%; freedom from recurrence, 66%; and overall survival, 52%. At 5 years, patients who underwent neck dissection, compared with patients who did not, were more likely to have a better neck control rate (94% v 85%, respectively; P = .0051) and cause-specific survival rate (72% v 63%, respectively; P = .0042). However, after exclusion of patients who did not receive neck dissection as recommended because of unresectable neck disease, progression of disease, refusal, or comorbidity, these differences disappeared. Patients with negative neck specimens fared better than those with positive specimens, with 5-year neck control rates of 98% v 86%, respectively (P < .0001), and cause-specific survival rates of 85% v 56%, respectively (P < .0001). For patients with any residual disease in the neck dissection specimen, a shorter time to neck dissection (< a median of 47 days from the end of RT) was associated with a more favorable 5-year neck control rate compared with patients with no residual disease (91% v 76%, respectively; P = .0030), but this did not translate into a significant benefit in cause-specific survival rate at 5 years (61% v 51%, respectively; P = .2685). Multivariate analysis was performed to identify potential factors associated with neck control and cause-specific survival (Table 4). Complete response by rCR was not associated with improved neck control or cause-specific survival. None of the patient parameters that were analyzed were associated with improved neck control. However, tumor and node stage were significantly associated with cause-specific survival (Table 4).
The role of planned neck dissection after definitive RT is not well defined. The goal to achieve regional disease control at the outset is critically important, given that the reduction of regional failures may improve survival12-14 and that salvage neck dissection is noted by some to be rarely, if ever, successful.15,16 However, because the post-RT neck dissection is often negative, some patients undergo surgery with its associated risks for little benefit. Physical examination is a relatively unreliable method to assess disease status of the post-RT neck; the negative predictive value of a cCR 4 to 6 weeks after therapy for negative neck dissection was 77%, which is similar to other reported series.12,17-21 However, whether residual disease in the post-RT neck specimen consistently represents viable disease that is at risk for progression is uncertain because not all residual disease seen on hematoxylin and eosin staining shows proliferative capacity by Ki-67 staining.22 The rate of post-RT pathologic failure does not seem to correlate well with the rate of clinical neck failure in patients observed without neck dissection, and timing between RT and neck dissection may play a factor in this discrepancy. At present, clinical outcomes do not provide any consensus opinion. In the last decade, the concept of a planned neck dissection after RT has been challenged by several authors using aggressive definitive RT (ie, altered fractionation or concurrent chemotherapy) who report high rates of complete response and low rates of isolated neck recurrence without adjuvant neck dissection.5-8,18,19,23 Other recent reports continue to support planned neck dissection for patients with N2 or N3 disease12,13,17,19 even after complete clinical response because of inferior neck control and disease-free survival13 and/or overall survival12 with observation. Our data indicate that CT is a more reliable method than physical examination to assess post-RT response. The absence of lymph nodes larger than 1.5 cm and lack of any focal lymph node defect on post-RT CT 4 weeks after therapy had a negative predictive value of 94%. Similar results were reported previously,9 but in an attempt to reduce the subjectivity of interpretation, post-RT CT was graded based on the presence of focal abnormality and size and not on the presence of extracapsular spread. Although patient numbers are still relatively small, patients who have an rCR have a high rate of neck control. Only one (3%) of 32 observed patients with rCR experienced treatment failure in the neck. Included in these 32 patients who were observed after rCR were 19 patients with N2-3 disease and three patients with fixed lymphadenopathy, indicating that all patients may be candidates for observation if complete response criteria by post-RT CT are met. Neck control and cause-specific survival were not different compared with patients who had a negative adjuvant neck dissection. This updated study seems to be the only reported series in which CT is used to manage the post-RT neck. A few reports have attempted to identify the utility of positron emission tomography (PET) in this setting, but data are limited, and the timing and interpretation of post-RT PET are not standardized. Fluorodeoxyglucose-18 PET (FDG-PET) 1 month after RT is unreliable; seven of 35 patients who were observed after negative FDG-PET scan eventually experienced failure in the neck (negative predictive value, 72%),24 whereas six of seven patients who underwent planned neck dissection had residual disease in the neck specimen (negative predictive value, 14%).25 FDG-PET has been shown to have a higher negative predictive value for neck control at 3 to 5 months (negative predictive value, 97% to 100%),26-28 but assessment of the neck at this time frame may be too late for a planned post-RT neck dissection. The optimal timing of adjuvant neck dissection is between 4 and 8 weeks after RT,29 to allow for resolution of acute inflammatory effects while minimizing the chance for metastatic development and preceding development of late fibrosis, which could complicate neck dissection. Our institutional preference is to rely on post-RT CT 4 weeks after RT to manage the neck, particularly given these timing limitations. Lymph nodes characterized by post-RT CT 4 weeks after RT are highly likely to be associated with negative neck dissection pathology if they are less than 1.5 cm in size and without focal abnormality. Patients meeting these criteria who do not undergo planned neck dissection after RT can have an ultimate neck control rate that approximates 100%. Patients who do not undergo neck dissection should undergo surveillance CT every 3 to 4 months for approximately 1 year after therapy to maximize the chance for salvage neck dissection.
The authors indicated no potential conflicts of interest.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Ries LAG, Eisner MP, Kosary CL, et al: (eds): SEER Cancer Statistics Review, 1975-2002. Bethesda, MD, National Cancer Institute, 2005 2. Kutler DI, Patel SG, Shah JP: The role of neck dissection following definitive chemoradiation. Oncology (Williston Park) 18 : 993 -998, 2004[Medline] 3. Mendenhall WM, Villaret DB, Amdur RJ, et al: Planned neck dissection after definitive radiotherapy for squamous cell carcinoma of the head and neck. Head Neck 24 : 1012 -1018, 2002[CrossRef][Medline] 4. Mendenhall WM, Million RR, Cassisi NJ: Squamous cell carcinoma of the head and neck treated with radiation therapy: The role of neck dissection for clinically positive neck nodes. Int J Radiat Oncol Biol Phys 12 : 733 -740, 1986[Medline] 5. Peters LJ, Weber RS, Morrison WH, et al: Neck surgery in patients with primary oropharyngeal cancer treated by radiotherapy. Head Neck 18 : 552 -559, 1996[CrossRef][Medline] 6. Johnson CR, Silverman LN, Clay LB, et al: Radiotherapeutic management of bulky cervical lymphadenopathy in squamous cell carcinoma of the head and neck: Is postradiotherapy neck dissection necessary? Radiat Oncol Investig 6 : 52 -57, 1998[CrossRef][Medline] 7. Corry J, Rischin D, Smith JG, et al: Radiation with concurrent late chemotherapy intensification ('chemoboost') for locally advanced head and neck cancer. Radiother Oncol 54 : 123 -127, 2000[CrossRef][Medline] 8. Chan AW, Ancukiewicz M, Carballo N, et al: The role of postradiotherapy neck dissection in supraglottic carcinoma. Int J Radiat Oncol Biol Phys 50 : 367 -375, 2001[CrossRef][Medline] 9. Ojiri H, Mendenhall WM, Stringer SP, et al: Post-RT CT results as a predictive model for the necessity of planned post-RT neck dissection in patients with cervical metastatic disease from squamous cell carcinoma. Int J Radiat Oncol Biol Phys 52 : 420 -428, 2002[CrossRef][Medline] 10. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53 : 457 -481, 1958[CrossRef] 11. Cox DR: Regression models and life tables. J R Stat Soc B 34 : 187 -220, 1972 12. Brizel DM, Prosnitz RG, Hunter S, et al: Necessity for adjuvant neck dissection is setting of concurrent chemoradiation for advanced head-and-neck cancer. Int J Radiat Oncol Biol Phys 58 : 1418 -1423, 2004[CrossRef][Medline] 13. Lavertu P, Adelstein DJ, Saxton JP, et al: Management of the neck in a randomized trial comparing concurrent chemotherapy and radiotherapy with radiotherapy alone in resectable stage III and IV squamous cell head and neck cancer. Head Neck 19 : 559 -566, 1997[CrossRef][Medline] 14. Ellis ER, Mendenhall WM, Rao PV, et al: Incision or excisional neck-node biopsy before definitive radiotherapy, alone or followed by neck dissection. Head Neck 13 : 177 -183, 1991[Medline] 15. Bernier J, Bataini JP: Regional outcome in oropharyngeal and pharyngolaryngeal cancer treated with high dose per fraction radiotherapy: Analysis of neck disease response in 1646 cases. Radiother Oncol 6 : 87 -103, 1986[Medline] 16. Mabanta SR, Mendenhall WM, Stringer SP, et al: Salvage treatment for neck recurrence after irradiation alone for head and neck squamous cell carcinoma with clinically positive neck nodes. Head Neck 21 : 591 -594, 1999[CrossRef][Medline] 17. McHam SA, Adelstein DJ, Rybicki LA, et al: Who merits a neck dissection after definitive chemoradiotherapy for N2–N3 squamous cell head and neck cancer? Head Neck 25 : 791 -798, 2003[CrossRef][Medline] 18. Grabenbauer GG, Rodel C, Ernst-Stecken A, et al: Neck dissection following radiochemotherapy of advanced head and neck cancer: For selected cases only? Radiother Oncol 66 : 57 -63, 2003[CrossRef][Medline] 19. Argiris A, Stenson KM, Brockstein BE, et al: Neck dissection in the combined-modality therapy of patients with locoregionally advanced head and neck cancer. Head Neck 26 : 447 -455, 2004[CrossRef][Medline] 20. Stenson KM, Haraf DJ, Pelzer H, et al: The role of cervical lymphadenectomy after aggressive concomitant chemoradiotherapy: The feasibility of selective neck dissection. Arch Otolaryngol Head Neck Surg 126
: 950
-956, 2000 21. Wanebo H, Chougule P, Ready N, et al: Surgical resection is necessary to maximize tumor control in function-preserving, aggressive chemoradiation protocols for advanced squamous cancer of the head and neck (stage III and IV). Ann Surg Oncol 8 : 644 -650, 2001[CrossRef][Medline] 22. Strasser MD, Gleich LL, Miller MA, et al: Management implications of evaluating the N2 and N3 neck after organ preservation therapy. Laryngoscope 109 : 1776 -1780, 1999[CrossRef][Medline] 23. Garden AS, Glisson BS, Ang KK, et al: Phase I/II trial of radiation with chemotherapy "boost" for advanced squamous cell carcinomas of the head and neck: Toxicities and responses. J Clin Oncol 17
: 2390
-2395, 1999 24. Greven KM, Williams DW III, McGuirt WF Sr, et al: Serial positron emission tomography scans following radiation therapy of patients with head and neck cancer. Head Neck 23 : 942 -946, 2001[CrossRef][Medline] 25. Rogers JW, Greven KM, McGuirt WF, et al: Can post-RT neck dissection be omitted for patients with head-and-neck cancer who have a negative PET scan after definitive radiation therapy? Int J Radiat Oncol Biol Phys 58 : 694 -697, 2004[CrossRef][Medline] 26. Yao M, Graham MM, Smith RB, et al: Value of FDG PET in assessment of treatment response and surveillance in head-and-neck cancer patients after intensity modulated radiation treatment: A preliminary report. Int J Radiat Oncol Biol Phys 60 : 1410 -1418, 2004[CrossRef][Medline] 27. Yao M, Buatti JM, Dornfeld KJ, et al: Can post-RT FDG PET accurately predict the pathologic status in neck dissection after radiation for locally advanced head and neck cancer? In regard to Rogers, et al (Int J Radiat Oncol Biol Phys 2004;58:694-697). Int J Radiat Oncol Biol Phys 61 : 306 -307, 2005[Medline] 28. Porceddu SV, Jarmolowski E, Hicks RJ, et al: Utility of positron emission tomography for the detection of disease in residual neck nodes after (chemo)radiotherapy in head and neck cancer. Head Neck 27 : 175 -181, 2005[CrossRef][Medline] 29. Million RA, Cassisi NJ: Management of Head and Neck Cancer: A Multidisciplinary Approach (ed 2). Philadelphia, PA, J.B. Lippincott Company, 1994 Submitted October 25, 2005; accepted January 13, 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
