|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 24, No 18 (June 20), 2006: pp. 2858-2865 © 2006 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.05.6176 Utility of Computed Tomography and Magnetic Resonance Imaging Staging Before Completion Lymphadenectomy in Patients With Sentinel Lymph Node–Positive Melanoma
From the Departments of Surgical Oncology, Biostatistics and Applied Mathematics, and Diagnostic Radiology, The University of Texas M.D. Anderson Cancer Center, Houston, TX Address reprint requests to Paul Mansfield, MD, Department of Surgical Oncology, M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Unit 444, Houston, TX 77030-4009; e-mail: pmansfie{at}mdanderson.org
PURPOSE: Although melanoma patients with regional nodal metastases are frequently imaged with computed tomography (CT) and magnetic resonance imaging (MRI) scans, the efficacy of routine radiologic staging in asymptomatic patients with microscopic nodal involvement has not been established. To determine the utility of this approach, we analyzed the incidence of synchronous distant metastases (SDM) detected by CT or MRI of the head, chest, and abdomen in a large group of patients with sentinel lymph node (SLN) –positive melanoma. PATIENTS AND METHODS: Positive SLNs were identified in 314 (16.2%) of the 1,934 melanoma patients who underwent sentinel lymphadenectomy at our institution from 1996 to 2003. Within 3 months of sentinel lymphadenectomy, 270 (86.0%) of the 314 SLN-positive patients were radiologically staged. To determine which prognostic factors were associated with SDM, associations between final staging outcomes and clinicopathologic variables, including SLN tumor burden, were analyzed. RESULTS: CT and/or MRI scans identified lesions that were suspicious for SDM in 23 (8.6%) of the 270 patients who underwent staging. In eight of these patients, further diagnostic studies determined that these abnormalities were benign. The remaining 15 suspicious lesions were percutaneously biopsied (10 negative and five positive), yielding a radiologically detectable SDM rate of 1.9%. Detection of SDM was associated with primary tumor thickness (P = .011), ulceration (P = .018), and SLN tumor burden (P = .018). CONCLUSION: These data suggest that the vast majority of asymptomatic patients with a new diagnosis of microscopic SLN-positive melanoma do not harbor radiologically detectable SDM and can proceed to completion lymph node dissection without immediate CT or MRI staging.
In 2005, it was estimated that 59,580 people would be diagnosed with cutaneous melanoma in the United States.1 Although 80% of these patients will present with clinical stage I or II disease1 (ie, clinically negative regional lymph nodes), on further evaluation, approximately 20% will be found to have occult regional nodal metastases.2-8 Patients with regional nodal disease are at increased risk for the development of distant metastases,2-4,6,8 suggesting that some patients may harbor occult distant disease at diagnosis. Detection of synchronous distant metastases (SDM) is important because this finding significantly alters treatment planning. For example, patients who are diagnosed with stage IV disease can avoid unnecessary regional procedures and their associated morbidity and, instead, receive stage-appropriate systemic therapy. A number of radiologic modalities, including liver sonography, nuclear liver and bone scans, computed tomography (CT), and magnetic resonance imaging (MRI), have been assessed for their ability to accurately detect occult metastases in patients with clinically localized, locally advanced, and regionally metastatic melanoma.9-15 Collectively, these studies have reported an incidence of SDM in asymptomatic melanoma patients of less than 5% for patients with clinically localized disease and only 7% for patients with demonstrated regional nodal disease. In addition, these modalities yielded false-positive results in up to 33% of patients imaged, prompting the performance of an unacceptably high number of additional diagnostic studies and unnecessary invasive procedures. From these data, the authors of these studies concluded that routine radiologic evaluation of asymptomatic melanoma patients was not useful. Subsequent to the publication of these reports, the prognostic significance of sentinel lymph node (SLN) involvement (microscopic stage III melanoma) has been validated in several large studies.2-8 These data indicate that microscopic SLN involvement is the strongest predictor of melanoma-specific survival in clinical stage I and II patients, even outweighing the prognostic value of tumor thickness and ulceration. These data may also indicate that SLN status could be used to identify a subset of asymptomatic melanoma patients who are at high risk for occult SDM. Largely on the basis of this hypothesis, interest in radiologic staging of patients with a new diagnosis of melanoma has been renewed. Despite a lack of supportive data, many asymptomatic patients with stage I to III melanoma continue to be routinely imaged with a variety of radiologic modalities including CT, MRI, and, more recently, fluorodeoxyglucose-positron emission tomography (FDG-PET) scans. On the basis of the availability of a novel prognostic indicator as powerful as SLN status and recent advances in imaging techniques, it is reasonable to re-examine the utility of radiologic staging for asymptomatic patients with SLN-positive melanoma before completion lymphadenectomy. To address this issue, we queried a contemporary melanoma surgery database containing patients with microscopic stage III melanoma diagnosed via SLN biopsy (SLNB) who underwent detailed radiologic staging using contemporary chest CT, abdominopelvic CT, and head CT or brain MRI techniques.
After The University of Texas M.D. Anderson Cancer Center institutional review board approval, we searched our prospectively collected melanoma surgery database and identified 314 patients (16.2%) with SLN-positive melanoma among 1,934 patients who underwent SLNB from January 1996 to June 2003. This cohort consisted of patients whose SLNs were involved with metastatic melanoma as determined by routine histology and/or immunohistochemical staining. For statistical analysis, basic demographic and pathologic data were compiled from the melanoma surgery database. An additional melanoma pathology database was consulted to collect detailed data describing the SLN tumor burden. Retrospective review was used to compile data on all radiologic studies performed within 3 months of SLNB, including the types of imaging studies obtained, their timing in relation to lymphadenectomy, and their result. Complete radiologic staging was defined as CT and/or MRI evaluation for brain, lung, and intra-abdominal metastasis. Partial radiologic staging was defined as CT and/or MRI of one or two of these three sites. In patients with lower extremity or truncal primary lesions with potential drainage to inguinal nodal basins, both abdominal CT scan and pelvic CT scan were obtained. Each radiologic study was graded on a 4-point scale that is summarized in Table 1. Patients with benign or indeterminate findings too small to further characterize on initial imaging underwent repeat imaging every 3 months for a total of at least 6 months to confirm stability or resolution of the imaging abnormality.
Patients who present to our institution with primary cutaneous invasive melanoma are routinely screened with a chest radiograph and a serum lactate dehydrogenase level. Patients with an abnormal chest radiograph and/or elevated lactate dehydrogenase level undergo additional preoperative radiologic staging. Only patients with normal findings on initial screening and patients with abnormal findings on initial screening who have no evidence of distant disease on more detailed evaluation proceed to SLNB. In the study group, six of the 314 patients were found to have abnormalities on pre-SLNB chest radiographs (five pulmonary nodules and one mediastinal abnormality) that prompted further preoperative radiologic evaluation. To assess these abnormalities, all six patients underwent chest CT scan, and in four of the six patients, the scan was either normal or contained benign findings. In two patients, the chest CT found that the lesion on the plain chest radiograph was suspicious for malignancy. One patient underwent a video-assisted thoracoscopic biopsy, and the other patient was percutaneously biopsied, with benign pathology obtained in both patients. Both patients subsequently proceeded to SLNB. Because we do not routinely include FDG-PET imaging in the staging for melanoma patients, results for FDG-PET imaging were available for only a small minority of patients with positive SLNs. Therefore, PET data was not formally analyzed as part of this investigation.
Statistical Considerations
Extent of Radiologic Imaging A total of 718 radiologic staging studies was performed in 270 (86.0%) of the 314 SLN-positive patients, and other than a plain chest radiograph, 44 patients (14.0%) had no radiologic staging. Of the 270 staged patients, 203 (75.2%) were completely staged, and the remaining 67 patients (24.8%) were partially staged. To determine whether any clinical or pathologic study variables were associated with the extent of radiologic staging, the distribution of these variables was statistically compared (Table 2). This analysis revealed that each of the study variables was similarly distributed between the groups with two exceptions. The ratio of male to female patients in both the complete staging (1.5:1) and the no staging (3.4:1) groups was higher than in the partial staging group (1.1:1; P = .029). In addition, the median tumor thickness was not statistically different between the groups. Median primary melanoma thickness in the no staging group (2.0 mm) was less than the thickness in patients in either the partial staging group (2.7 mm) or the complete staging group (2.5 mm; P = .036).
Assessment of Radiologic Findings For the 270 staged patients, the overall interpretation of each radiologic work-up, the results of any subsequent additional tests, and the final diagnostic outcome are described in Figure 1. Sixty-nine (25.6%) of these 270 patients had at least one abnormality detected. In 36 patients (13.3%), the radiologically detected abnormality was interpreted as benign or indeterminate and too small to further characterize (grade 2). Subsequent imaging over a 6-month period confirmed that the behavior of each of these lesions was consistent with a benign entity. In addition, radiologic staging identified abnormal lesions suggestive of occult second primary tumors (grade 4) in 10 patients (3.7%). In all 10 patients, these diagnoses were subsequently confirmed and addressed.
In 23 patients (8.6%), the initial radiologic studies detected lesions that were interpreted as suspicious for melanoma metastasis (grade 3). All of the patients in this group underwent additional diagnostic testing. In eight patients, alternate noninvasive studies revealed that the lesion initially interpreted as suspicious actually represented a benign finding. The most common clinical scenario in these patients was a liver lesion read as suspicious on abdominal CT but found to be benign on subsequent liver ultrasound, dedicated liver protocol CT, or liver MRI. The other 15 patients with suspicious findings on staging evaluation underwent percutaneous biopsy of the index lesion. No immediate complications from these biopsies were reported. In 10 of these 15 patients, biopsy analysis determined that the lesion was benign. In the remaining five patients, the biopsy was positive for metastatic melanoma, resulting in a true-positive yield from staging radiologic evaluation of 1.9% (five of 270 patients). The results of each radiologic study type, categorized by our 4-point grading system, are reported in Table 3. The most commonly ordered study was abdominal or abdominopelvic CT scan (n = 265). In 10 (3.8%) of these 265 patients, CT identified an intra-abdominal mass suspicious for melanoma metastasis (grade 3); however, subsequent evaluation revealed that this was a true-positive finding in only one patient (0.4%). Five patients (1.9%) were found on abdominal CT to have a second neoplasm (three patients with renal cell carcinoma and two patients with benign adrenal cortical neoplasms). The second most commonly ordered study was chest CT scan (n = 243). Twelve (4.9%) of these 243 studies contained suspicious lesions, and in four of these patients (1.6%), this was confirmed to be a true-positive finding. Four additional patients (1.6%) were found to have a second malignancy in the chest (two patients with non–small-cell lung cancer and one patient each with mediastinal parathyroid adenoma and intrathoracic lymphoma). In total, 25 head CT scans and 185 brain MRI scans were performed to detect brain metastasis. Only one head CT was suspicious for melanoma metastasis, and on subsequent imaging, this finding was determined to represent a benign cyst. In subsequent evaluation of the three patients with MRI results suspicious for melanoma metastasis, two patients (1.1%) were determined to have metastatic melanoma. One additional patient (0.5%) was incidentally found on MRI staging to have a meningioma (Fig 2).
Clinical and Pathologic Characteristics of Patients With Positive Radiologic Staging The clinical and pathologic characteristics of the five patients with positive radiologic staging, including the sites of SDM for each, are listed in Table 4. Although the small number of patients with positive radiologic staging limited the power of statistical comparisons, these data suggested that patients with positive staging had more advanced primary lesions than their counterparts who were determined to be free of metastasis by radiologic staging (Table 5). The median primary tumor thickness in the positive staging group was 5.0 mm (four of five patients with primary tumor thickness > 3.4 mm), and in all five patients, the primary lesion was ulcerated. In contrast, the median tumor thickness in the negative staging group was 2.5 mm (P = .011), and 44% of the primary lesions in this group were ulcerated (P = .018).
Data analysis also suggested that the patients who were found to have distant metastases at radiologic evaluation had a greater SLN tumor burden. As described in Table 5, all recorded measures of SLN tumor burden seemed to be increased in the positive radiologic staging group compared with patients with negative staging, including the median size of the largest metastatic focus (8.0 v 1.3 mm, respectively; P = .021) and the median square area of metastasis (75.6 v 1.0 mm2, respectively; P = .018). In addition, the degree of nodal involvement seemed to be greater in patients with positive radiologic staging. Patients with positive radiologic staging were three times more likely to have nodal metastases involving both the subcapsular and intramedullary spaces compared with patients with negative staging (80% v 25%, respectively; P = .024), and 60% of the nodal metastases in patients with positive staging extended through the node capsule, whereas only 14% of patients without evidence of distant metastases were found to have nodal metastases with extracapsular extension (P = .026).
Although early identification of melanoma patients who harbor occult distant metastases at diagnosis would allow for optimal treatment planning, achievement of this goal has been elusive. Previous studies investigating the ability of multiple radiologic staging modalities to detect occult stage IV disease in asymptomatic patients have determined that none of the various modalities studied were helpful.9-14 Despite these data and other issues related to radiologic staging, including financial costs and dye-related complications, many clinicians continue to order multiple radiologic studies in an attempt to detect SDM. A renewed interest in radiologic staging for patients with melanoma stems from two main sources. First, there is a perception that recent improvements in the sensitivity and specificity of staging modalities, including CT and MRI, will improve the yield of these studies. Second, clinicians who treat melanoma can now use information from SLNB to identify a group of melanoma patients who are at higher risk for developing distant metastases2-5,7,16 and may, therefore, be at higher risk for SDM. Since the introduction of the SLNB technique, the incidence of SDM in a population of patients with microscopic (SLN-positive) stage III melanoma has been examined in only one other study.17 For their analysis, Miranda et al17 reviewed radiology records for 185 SLN-positive melanoma patients and found that radiologic staging with chest radiograph, CT, and MRI detected occult distant metastases in 0.5% of patients. Our analysis, which included CT and MRI study results from 270 SLN-positive patients, found a true-positive detection rate of occult distant metastasis of 1.9%. These two contemporary studies indicate that the incidence of SDM in asymptomatic patients with microscopic stage III melanoma is low and has not been significantly impacted by recent technical advances in imaging techniques and interpretation. However, our data demonstrate that advances in imaging technology have improved the accuracy of CT and MRI not only for the detection of melanoma metastases, but also for the detection of second primary lesions, which were identified in 10 (3.7%) of the 270 patients who were imaged. Regarding the detection of melanoma metastases, current imaging equipment and techniques resulted in a lowering of the number of false-positive results from as high as 33% in previous studies12 to 12% in our study. Although the rate of false-positive results has declined, the incidence of these findings remains unacceptably high. This factor and the low incidence of true-positive results lead us to conclude that the vast majority of asymptomatic patients with microscopic stage III melanoma do not benefit from initial comprehensive radiologic staging with CT and/or MRI. Some clinicians believe that FDG-PET is more accurate than CT in the detection of distant metastases in melanoma patients.18 This may be true when FDG-PET is used to stage patients who have clinically evident regional or distant metastasis, where PET has been found to be superior to CT and MRI in the detection of lesions that were mainly outside of the conventional imaging field (in transit, distant soft tissue) and in the detection of some intra-abdominal (small bowel) metastases.19-27 However, these data do not seem to be transferable to melanoma patients with subclinical regional nodal involvement, in whom distant metastases are likely below the resolution limits of PET scan.28 The only study reporting superior results for PET compared with anatomic imaging in patients without clinically evident regional or distant disease was a retrospective analysis that included only 52 patients.19 The findings of this study have not been confirmed despite multiple subsequent attempts,29-33 and no studies prospectively comparing PET with CT in this population have been reported to definitively answer this question. In addition, when applied to patients with other malignancies, PET has repeatedly been found to be inferior to anatomic cross-sectional imaging for the detection of pulmonary and brain metastases, which are two frequent metastasis sites in patients with melanoma.26,27 The data on CT and MRI staging from our group and others suggest that the vast majority of newly diagnosed asymptomatic melanoma patients do not harbor radiologically detectable SDM. Given this finding, no staging modality that aims to detect macroscopic distant metastases, no matter how technologically advanced or expensive, is likely to increase the detection rate of metastatic disease in this patient population. Therefore, we do not advocate the use of FDG-PET imaging for asymptomatic patients diagnosed with SLN-positive melanoma outside of a clinical trial comparing the accuracy of this modality with modern CT and MRI. Our data indicate that, in SLN-positive patients, CT and MRI staging rarely detect abnormalities that alter the indication for completion lymphadenectomy. Furthermore, false-positive results generated by these studies may prompt additional diagnostic tests, unnecessarily delaying the completion lymphadenectomy procedure. However, early radiologic staging may provide valuable baseline data for patients entering adjuvant clinical trials. To reconcile these issues, we propose that CT and MRI before completion lymphadenectomy be performed only for SLN-positive patients with poor prognostic factors who seem to be at the highest risk of having detectable SDM, including patients with thick (T4) and/or ulcerated primary lesions and patients with large SLN tumor burden (Fig 3). 34,35 Otherwise, patients with a positive SLNB who intend to enter adjuvant clinical trials can be staged after completion lymphadenectomy within a time interval proximate to study initiation (usually within 30 days). This policy ensures that the imaging results are not outdated and required to be repeated, as often happens when patients are radiologically staged before SLNB.
Although the proposed staging algorithm is based on the best available data, our analysis and those of other investigators indicate that a need exists for prospective evaluation of staging algorithms in melanoma that aim to define the appropriate use of CT, MRI, and, possibly, PET scan from both patient-centered and cost-centered viewpoints.
The authors indicated no potential conflicts of interest.
We thank the M.D. Anderson Cancer Center’s Department of Scientific Publications for their editorial assistance with this manuscript.
Supported in part by Grant No. P50 CA93459 from The University of Texas M.D. Anderson Cancer Center Specialized Programs of Research Excellence in Melanoma (J.E.G., J.E.L., M.M.J., and C.W.S.). Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Jemal A, Murray T, Ward E, et al: Cancer statistics, 2005. CA Cancer J Clin 55:10-30, 2005 2. Gershenwald J, Thompson W, Mansfield P, et al: Multi-institutional melanoma lymphatic mapping experience: The prognostic value of sentinel lymph node status in 612 stage I or II melanoma patients. J Clin Oncol 17:976-983, 1999 3. Balch C, Soong S, Gershenwald J, et al: Prognostic factor analysis of 17,600 melanoma patients: Validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol 19:3622-3634, 2001 4. Clary B, Brady M, Lewis J, et al: Sentinel lymph node biopsy in the management of patients with primary cutaneous melanoma: Review of a large single-institutional experience with an emphasis on recurrence. Ann Surg 233:250-258, 2001[CrossRef][Medline] 5. Dessureault S, Soong S, Ross M, et al: Improved staging of node-negative patients with intermediate to thick melanomas (>1 mm) with the use of lymphatic mapping and sentinel lymph node biopsy. Ann Surg Oncol 8:766-770, 2001 6. Thompson J: The Sydney Melanoma Unit experience of sentinel lymphadenectomy for melanoma. Ann Surg Oncol 8:44S-47S, 2001 (suppl 9)[Medline] 7. Morton DL, Wen DR, Wong JH, et al: Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 127:392-399, 1992[Abstract] 8. Gershenwald JE, Colome MI, Lee JE, et al: Patterns of recurrence following a negative sentinel lymph node biopsy in 243 patients with stage I or II melanoma. J Clin Oncol 16:2253-2260, 1998[Abstract] 9. Zartman G, Thomas M, Robinson W: Metastatic disease in patients with newly diagnosed malignant melanoma. J Surg Oncol 35:163-164, 1987[Medline] 10. Ardizzoni A, Grimaldi A, Repetto L, et al: Stage I-II melanoma: The value of metastatic work-up. Oncology 44:87-89, 1987[Medline] 11. Buzaid A, Sandler A, Mani S, et al: Role of computed tomography in the staging of primary melanoma. J Clin Oncol 11:638-643, 1993[Abstract] 12. Huang C, Provost N, Marghoob A, et al: Laboratory tests and imaging studies in patients with cutaneous malignant melanoma. J Am Acad Dermatol 39:451-463, 1998[CrossRef][Medline] 13. Kersey P, Iscoe N, Gapski J, et al: The value of staging and serial follow-up investigations in patients with completely resected, primary, cutaneous malignant melanoma. Br J Surg 72:614-617, 1985[Medline] 14. Terhune M, Swanson N, Johnson T: Use of chest radiography in the initial evaluation of patients with localized melanoma. Arch Dermatol 134:569-572, 1998 15. Buzaid AC, Tinoco L, Ross MI, et al: Role of computed tomography in the staging of patients with local-regional metastases of melanoma. J Clin Oncol 13:2104-2108, 1995 16. Balch C, Soong S, Ross M, et al: Long-term results of a multi-institutional randomized trial comparing prognostic factors and surgical results for intermediate thickness melanomas (1.0 to 4.0 mm): Intergroup Melanoma Surgical Trial. Ann Surg Oncol 7:87-97, 2000[Abstract] 17. Miranda EP, Gertner M, Wall J, et al: Routine imaging of asymptomatic melanoma patients with metastasis to sentinel lymph nodes rarely identifies systemic disease. Arch Surg 139:831-836, 2004 18. Friedman K, Wahl R: Clinical use of positron emission tomography in the management of cutaneous melanoma. Semin Nucl Med 34:242-253, 2004[CrossRef][Medline] 19. Rinne D, Baum RP, Hor G, et al: Primary staging and follow-up of high risk melanoma patients with whole-body 18F-fluorodeoxyglucose positron emission tomography: Results of a prospective study of 100 patients. Cancer 82:1664-1671, 1998[CrossRef][Medline] 20. Gritters LS, Francis IR, Zasadny KR, et al: Initial assessment of positron emission tomography using 2-fluorine-18-fluoro-2-deoxy-D-glucose in the imaging of malignant melanoma. J Nucl Med 34:1420-1427, 1993 21. Steinert HC, Huch Boni RA, Buck A, et al: Malignant melanoma: Staging with whole-body positron emission tomography and 2-[F-18]-fluoro-2-deoxy-D-glucose. Radiology 195:705-709, 1995 22. Swetter SM, Carroll LA, Johnson DL, et al: Positron emission tomography is superior to computed tomography for metastatic detection in melanoma patients. Ann Surg Oncol 9:646-653, 2002 23. Gulec SA, Faries MB, Lee CC, et al: The role of fluorine-18 deoxyglucose positron emission tomography in the management of patients with metastatic melanoma: Impact on surgical decision making. Clin Nucl Med 28:961-965, 2003[CrossRef][Medline] 24. Dietlein M, Krug B, Groth W, et al: Positron emission tomography using 18F-fluorodeoxyglucose in advanced stages of malignant melanoma: A comparison of ultrasonographic and radiological methods of diagnosis. Nucl Med Commun 20:255-261, 1999[Medline] 25. Tyler DS, Onaitis M, Kherani A, et al: Positron emission tomography scanning in malignant melanoma. Cancer 89:1019-1025, 2000[CrossRef][Medline] 26. Holder WD Jr, White RL Jr, Zuger JH, et al: Effectiveness of positron emission tomography for the detection of melanoma metastases. Ann Surg 227:764-769, 1998[CrossRef][Medline] 27. Eigtved A, Andersson AP, Dahlstrom K, et al: Use of fluorine-18 fluorodeoxyglucose positron emission tomography in the detection of silent metastases from malignant melanoma. Eur J Nucl Med 27:70-75, 2000[CrossRef][Medline] 28. Wagner JD, Schauwecker DS, Davidson D, et al: FDG-PET sensitivity for melanoma lymph node metastases is dependent on tumor volume. J Surg Oncol 77:237-242, 2001[CrossRef][Medline] 29. Schafer A, Herbst R, Beiteke U: Sentinel lymph node excision (SLNE) and positron emission tomography in the staging of stage I-II melanoma patients. Hautarzt 54:440-447, 2003[Medline] 30. Mijnhout G, Hoekstra O, van Lingen A: How morphometric analysis of metastatic load predicts the (un)usefulness of PET scanning: The case of lymph node staging in melanoma. J Clin Pathol 56:283-286, 2003 31. Longo M, Lazaro P, Bueno C, et al: Fluorodeoxyglucose-positron emission tomography imaging versus sentinel node biopsy in the primary staging of melanoma patients. Dermatol Surg 29:245-248, 2003[CrossRef][Medline] 32. Hafner J, Schmid MH, Kempf W, et al: Baseline staging in cutaneous malignant melanoma. Br J Dermatol 150:677-686, 2004[CrossRef][Medline] 33. Havenga K, Cobben DC, Oyen WJ, et al: Fluorodeoxyglucose-positron emission tomography and sentinel lymph node biopsy in staging primary cutaneous melanoma. Eur J Surg Oncol 29:662-664, 2003[CrossRef][Medline] 34. Gershenwald J, Prieto V, Johnson M, et al: Microscopic tumor burden in sentinel lymph nodes best predicts nonsentinel lymph node involvement in patients with melanoma. 57th Annual Meeting of the Society of Surgical Oncology, Denver, CO, March 14-17, 2002 35. Gershenwald J, Prieto V, Johnson M, et al: Heterogeneity of microscopic stage III melanoma in the SLN era: Implications for AJCC/UICC staging and future clinical trial design. 6th World Congress on Melanoma, Vancouver, Canada, September 1-5, 2005 Submitted January 11, 2006; accepted March 14, 2006. Related Correspondence
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
2 test or Fisher’s exact test, as appropriate. Continuous data were compared between groups using the Mann-Whitney U test or Kruskal-Wallis test, as appropriate. All analyses were performed using SAS software version 8.2 (SAS Institute, Cary, NC). 
