|Home | About | Journals | Submit | Contact Us | Français|
To revise the staging system for cutaneous melanoma on the basis of data from an expanded American Joint Committee on Cancer (AJCC) Melanoma Staging Database.
The melanoma staging recommendations were made on the basis of a multivariate analysis of 30,946 patients with stages I, II, and III melanoma and 7,972 patients with stage IV melanoma to revise and clarify TNM classifications and stage grouping criteria.
Findings and new definitions include the following: (1) in patients with localized melanoma, tumor thickness, mitotic rate (histologically defined as mitoses/mm2), and ulceration were the most dominant prognostic factors. (2) Mitotic rate replaces level of invasion as a primary criterion for defining T1b melanomas. (3) Among the 3,307 patients with regional metastases, components that defined the N category were the number of metastatic nodes, tumor burden, and ulceration of the primary melanoma. (4) For staging purposes, all patients with microscopic nodal metastases, regardless of extent of tumor burden, are classified as stage III. Micrometastases detected by immunohistochemistry are specifically included. (5) On the basis of a multivariate analysis of patients with distant metastases, the two dominant components in defining the M category continue to be the site of distant metastases (nonvisceral v lung v all other visceral metastatic sites) and an elevated serum lactate dehydrogenase level.
Using an evidence-based approach, revisions to the AJCC melanoma staging system have been made that reflect our improved understanding of this disease. These revisions will be formally incorporated into the seventh edition (2009) of the AJCC Cancer Staging Manual and implemented by early 2010.
The current melanoma staging system was substantially revised in 2001 for the sixth edition of the Cancer Staging Manual, on the basis of an analysis of 17,600 patients in the American Joint Committee on Cancer (AJCC) Melanoma Staging Database.1,2 For this analysis, we expanded the sample size of the melanoma staging database and added mitotic rate of the primary melanoma as a new covariate because of recent studies demonstrating this to be an important and independent prognostic factor. The database for stage IV patients was expanded five-fold and, for the first time, contained data about the prognostic value of the serum lactate dehydrogenase (LDH) level. During the 7 years since the previous analysis, the sentinel node procedure has become a standard for staging nodal metastases in patients with clinically uninvolved lymph nodes, with the net result that microscopically detected nodal metastases at initial presentation are now detected in many more melanoma patients. It was important, therefore, to verify that the criteria for stage III used in the past, with long-term follow-up, were still valid in this contemporary era of nodal staging. The staging recommendations resulted from an unprecedented collaboration by melanoma centers that contributed the largest data set from melanoma patients ever analyzed.
The AJCC Melanoma Staging Committee used previously published guidelines to determine criteria that should be used in the TNM classification and the stage groupings.1 The evidence-based analysis that led to melanoma staging recommendations for the seventh edition of the Cancer Staging Manual was based on the updated AJCC Melanoma Staging Database (data through 2008) containing prospective data on 30,946 patients with stages I, II, and III melanoma and 7,972 patients with stage IV melanoma. Patients were treated at 17 major medical centers, free-standing cancer centers, or cancer cooperative groups (Appendix, online only). Independent prognostic factors were considered by the AJCC Melanoma Committee for defining the TNM categories and stage groupings on the basis of results published in the literature as well as our prognostic factors analyses of the AJCC Melanoma Staging Database. The statistical approaches and data dictionary definitions are virtually the same as described in our previous publication.1 Mitotic rate was examined for the first time in this analysis. The Melanoma Staging Committee recommended that mitotic rate be determined by the “hot spot” approach and expressed as the number of mitoses per square millimeter of primary tumor.3 Statistical analyses of the AJCC Melanoma Staging Database primarily used methods similar to those for survival analysis. Survival times were calculated from the initial melanoma diagnosis (or first distant metastasis for the stage IV analysis) and considered censored for patients who were alive at last follow-up or who died without evidence of melanoma. Melanoma-specific survival curves were generated according to the Kaplan-Meier product-limit method and were compared using the log-rank test. Multivariate analyses of prognostic factors were based on the Cox proportional hazards model.
The TNM categories for the seventh edition of the AJCC Staging Manual are defined in Table 1, and the stage groupings are defined in Table 2. The updated Melanoma Staging Database was used to calculate survival rates for patients with stages I to IV melanoma. Substages for stages I, II, and III are shown in Figure 1A-D and TNM categories for stage IV in are shown in Figure 2. Changes in the melanoma staging system are summarized in Table 3. These recommendations of the AJCC Melanoma Staging Committee have been approved by both the AJCC Executive Committee and the International Union Against Cancer (UICC) TNM Committee. The final recommendations of the melanoma staging criteria will be formally implemented in January 2010.3
The AJCC Melanoma Staging Database includes prospectively accumulated data on more than 27,000 stage I and II melanoma patients for whom tumor thickness and follow-up information is available. Five-year and 10-year survival rates based on TNM classification range from 97% and 93% for patients with T1aN0M0 melanomas to 53% and 39%, respectively for patients with T4bN0M0 melanomas (P < .0001; Fig 1A). By substage, 10-year survival ranged from 93% for stage IA to 39% for stage IIC melanoma (P < .0001; Fig 1B).
Recommendations for using melanoma thickness in TNM categories and stage groupings in the seventh edition remain unchanged, ie, the T category thresholds of melanoma thickness are defined in even integers (1.0, 2.0, and 4.0 mm). In the 2008 AJCC Melanoma Staging Database, as tumor thickness increased, there was a highly significant decline in 5- and 10-year survival rates (P < .0001). Among the 11,841 patients with T1 melanomas (≤ 1.00 mm thickness), the 10-year survival was 92%, while it was 80% in the 8,046 T2 patients with melanomas 1.01 to 2.00 mm thick, 63% in the 5,291 T3 patients with melanomas 2.01 to 4.00 mm thick, and 50% in the 2,461 T4 patients with melanomas more than 4.00 mm thick (P < .0001).
Recommendations for using ulceration status in defining TNM categories and stage groupings also remain unchanged. Survival rates of patients with an ulcerated melanoma are proportionately lower than those of patients with a nonulcerated melanoma of equivalent T category but are remarkably similar to those of patients with a nonulcerated melanoma of the next highest T category. For example, 5-year survival was 79% for a T3a nonulcerated melanoma and was 82% for a T2b ulcerated melanoma; both are defined as stage IIA. A T4a nonulcerated melanoma has a 5-year survival of 71%, similar to that of a T3b ulcerated melanoma with a 68% rate; both are defined as stage IIB. A T4b ulcerated melanoma has a 5-year survival of 53% and is categorized as stage IIC.
Proliferation of the primary melanoma as defined by the mitotic rate was identified as a powerful and independent predictor of survival. As a result, primary tumor mitotic rate is now a required element for the seventh edition melanoma staging system. Multiple thresholds of mitotic rate were examined statistically, and the most significant correlation with survival was identified at a threshold of at least 1/mm2. Data from the AJCC Melanoma Staging Database demonstrated a highly significant correlation between increasing mitotic rate and declining survival rates (P < .0001). In a multifactorial analysis of 10,233 patients with clinically localized melanoma, mitotic rate was the second most powerful predictor of survival, after tumor thickness (χ2 = 79.1; P < .0001).
Although melanomas 1 mm or less in thickness constitute a good prognosis group, we found that the 10-year survival outcome was variable, ranging from 85% to 99%, depending on the presence of secondary characteristics of mitotic rate and tumor ulceration. In a multivariate analysis of 4,861 T1 melanomas, tumor thickness, mitotic rate, and ulceration were the most powerful predictors of survival outcome for T1 melanoma patients, and the level of invasion was no longer statistically significant when mitotic rate and ulceration were included in the analysis (Table 4). The 10-year survival rate was 95% for nonulcerated T1 melanomas with a mitotic rate of less than 1/mm2 and dropped to 88% if the mitotic rate was at least 1/mm2 (P < .0001). Ulcerated T1 melanomas were associated with a mitotic rate of ≥ 1/mm2 in 78% of patients, but the 10-year survival rate was the same regardless of whether the mitotic rate was less than 1 or ≥ 1/mm2 (85% v 87%; P = .41). Therefore, the Melanoma Staging Committee has recommended that mitotic rate replace Clark level of invasion as a primary criterion for defining T1b melanoma.
Since mitotic rate will replace level of invasion in defining T1 categories, the Melanoma Staging Committee redefined the criteria for T1a and T1b melanomas. T1a melanomas (approximately 60% of T1 patients in the AJCC Melanoma Database) will be restricted to those meeting the following three criteria: ≤ 1.0 mm thick, no ulceration, and mitotic rate of less than 1/mm2. T1b melanomas (approximately 40% of T1 patients) are now defined as those whose tumor thickness is ≤ 1.0 mm and that have at least one mitosis per square millimeter or tumor ulceration present. In contrast to the sixth edition of the AJCC staging system, level of invasion is no longer routinely considered in defining T1 melanomas, except in the rare circumstances when mitotic rate cannot be accurately determined.
The 2008 AJCC Melanoma Staging Database contains 3,307 stage III patients who had information available to define stage, the vast majority of whom presented with micrometastases identified by a sentinel node biopsy and completion lymphadenectomy. A Cox multivariate analysis of the database demonstrated that the number of tumor-bearing nodes, tumor burden at the time of staging (ie, microscopic v macroscopic), presence or absence of primary tumor ulceration, and thickness of the primary melanoma were the most predictive independent factors for survival in these patients (all P values < .001). These characteristics were incorporated into the stage grouping criteria.
Five-year survival rates based on TNM classification ranged from 70% for patients with T1-4N1aM0 melanomas to 39% for patients with T1-4N3M0 melanomas (P < .0001; Fig 1C). In the absence of nodal metastases, patients with intralymphatic metastases (N2c) have 5- and 10-year survival rates of 69% and 52%, respectively (Fig 1C), while those with combined intralymphatic metastases and nodal metastases (N3) have survival rates of 46% and 33%, respectively. Five-year survival within substages of stage III were 78%, 59%, and 40% for patients with stage IIIA, IIIB, and IIIC melanoma, respectively (P < .0001; Fig 1D).
With the current widespread availability of immunohistochemical (IHC) staining, it is possible to consistently detect nodal metastases at a microscopic level consisting of aggregates of only a few cells.4–6 The availability and widespread use of IHC methods to detect melanoma-associated antigens is sufficiently available worldwide that the AJCC Melanoma Staging Committee considers it acceptable to classify nodal metastases solely on the basis of IHC staining of melanoma-associated markers. Although some IHC markers are sensitive but not specific for melanoma cells (eg, S100 protein, tyrosinase), IHC alone will be accepted if the diagnosis is based on at least one melanoma-associated marker (eg, HMB-45, Melan-A/MART 1) and the cells have malignant morphologic features that can be detected in the IHC stained tissue.4
In patients with distant metastases, the site(s) of metastases and elevated serum levels of LDH are used to delineate the M1 stage into three M categories: M1a, M1b, and M1c. One-year survival rates among 7,972 stage IV patients were 62% for M1a, 53% for M1b, and 33% for M1c melanomas (P < .0001; Fig 2A).
Patients with distant metastasis in the skin, subcutaneous tissue, or distant lymph nodes and a normal LDH level are categorized as M1a; they have a relatively better prognosis compared with those patients with metastases located in any other distant anatomic site (Fig 2A). Patients with metastasis to the lung (or with a combination of lung and skin or subcutaneous metastases) and a normal LDH level are categorized as M1b and have an intermediate prognosis. Those patients with metastases to any other visceral sites or at any location with an elevated LDH level are designated as M1c and have the worst prognosis (Fig 2A and and22B).
The updated AJCC Melanoma Staging Database demonstrated that an elevated serum LDH is an independent and highly significant predictor of survival outcome among patients with stage IV disease. Thus 1- and 2-year overall survival rates for those stage IV patients in the 2008 AJCC Melanoma Staging Database with a normal serum LDH were 65% and 40%, respectively, compared with 32% and 18%, respectively, when the serum LDH was elevated at the time of staging (P < .0001; Fig 2B). Therefore, serum LDH should be measured at the time stage IV disease is documented, and if the LDH level is elevated, those patients are assigned to M1c regardless of the site of their distant metastases.
The survival differences among M categories will be useful for clinical trial stratification; however, the overall prognosis of all patients with stage IV melanoma remains poor, even among patients with M1a. For this reason, the Melanoma Staging Committee recommended no stage groupings for stage IV.
Histological features of the primary melanoma—tumor thickness, mitotic rate, and ulceration—are important hallmarks of melanoma prognosis and staging. Most notably, the mitotic rate has emerged in this analysis as a powerful predictive factor of survival.7–10 After 40 years of being an integral component of melanoma staging, the Clark level is no longer recommended as a staging criterion, since it is not an independent prognostic factor when mitotic rate is included in the analysis. The value of these histologic characteristics for microstaging strongly supports that the initial biopsy is a critical component of both diagnosis and staging. An excisional biopsy of the entire clinically apparent lesion, with a narrow 1- to 2-mm margin of adjacent normal-appearing skin, is the biopsy technique of choice when melanoma is suspected, and shave biopsies should be avoided. An incisional biopsy may be acceptable for larger lesions. A deep saucerization biopsy may be satisfactory when the lesion is flat and the suspicion of melanoma is not high.11 These staging criteria of the primary melanoma should be used for all growth patterns of cutaneous melanoma but do not apply to mucosal or ocular melanomas.
The AJCC Melanoma Staging Committee recommends that sentinel lymph node biopsy be performed as a staging procedure in patients for whom the information will be useful in planning subsequent treatments and follow-up regimens. Specifically, the procedure should be discussed with (and recommended for) otherwise healthy patients who have T2, T3, and T4 melanomas and clinically uninvolved regional lymph nodes; the procedure should be recommended selectively for patients with T1b melanomas.12–21 The use of mitotic rate for the purpose of classifying thin melanomas as T1b was based on a survival analysis. The AJCC Melanoma Staging Database did not contain sufficient data to assess risk of occult nodal micrometastases in this population. However, preliminary evidence from several other large studies suggests that T1 melanomas with a mitotic rate of ≥ 1/mm2 and a thickness of ≥ 0.76 mm are associated with an approximately 10% risk of occult metastases in their sentinel lymph nodes (J. Gershenwald, personal communication, March 2009). These data may be helpful when discussing the indications for sentinel lymph node biopsy for staging with individual patients with T1b melanoma. Furthermore, staging with sentinel node technology should be required as an entry criterion for all melanoma patients presenting with clinical stage IB or II disease before entry into clinical trials involving new surgical techniques or adjuvant therapy.
This staging system is the first to contain long-term follow-up of patients staged with sentinel lymph node biopsy. Reflective of a changing demographic in melanoma, most patients with histologically confirmed stage III melanoma at diagnosis now present with clinically uninvolved regional nodes and micrometastasis diagnosed by sentinel lymph node biopsy. Such improved staging translates into more refined (and favorable) survival estimates for patients with stages IB-IIIA melanoma (Fig 1).
Intralymphatic metastases (ie, satellites or in transit metastases) are another criterion in the N category, regardless of the number of lesions.22 For the first time, there are prospective data and survival rates in the 2008 AJCC/UICC melanoma staging database for patients who manifest intralymphatic metastases. The results were somewhat better than those previously reported in the literature and are higher than those in the remaining cohort of stage IIIB patients. Nevertheless, the category of stage IIIB was still the closest fit statistically, and the AJCC Melanoma Committee recommended that the sixth edition staging definition be retained. Microscopic satellites are defined as any discontinuous nest of metastatic cells more than 0.05 mm in diameter that are clearly separated by normal dermis (not fibrosis or inflammation) from the main invasive component of melanoma by a distance of at least 0.3 mm. Data from the literature show that survival outcome are comparable to that of patients with clinically detectable satellite metastases.23–26 Accordingly, the AJCC Melanoma Staging Committee has recommended that this feature of early lymphatic metastases be retained in the category of N2c melanoma.
The updated AJCC Melanoma Staging Database clearly demonstrates that an elevated serum LDH is an independent and highly significant predictor of survival or outcome of stage IV patients, independent of other factors. Furthermore, this factor was among the most predictive independent factors of diminished survival in all published studies when it was analyzed in a multivariate analysis, even after accounting for site and number of metastases.27–30
The mechanisms or sources of elevated LDH isoenzymes are unknown, and generally the there is a nonspecific pattern of elevation among the various LDH isoenzymes. Survival rates are significantly reduced in patients with an elevated serum LDH at the time of initial assignment to stage IV. Therefore, when serum LDH is elevated above the upper limits of normal at the time of staging, those patients who also have distant metastases are assigned to M1c, regardless of the site of their distant metastases.
The number of metastases at distant sites has previously been documented as an important prognostic factor.27,31,32 This was also confirmed by preliminary multivariate analyses using the AJCC Melanoma Staging Database. However, this feature was not incorporated into the staging system because of significant variability in the deployment of diagnostic tests to comprehensively search for distant metastases among institutions that contributed data. Tests range from a simple chest x-ray in some centers to high-resolution double-contrast computed tomography, positron emission tomography/computed tomography, and/or magnetic resonance imaging in others.
In patients who present with metastases and no known primary site, it is difficult to assign a staging category. When patients have an initial presentation of metastases in the lymph nodes, these should be presumed to be regional (stage III instead of stage IV) if an appropriate staging workup does not reveal any other sites of metastases. These patients have a prognosis and natural history that is similar to, if not more favorable than, patients with the same staging characteristics from a known primary cutaneous melanoma.33,34 When there are localized metastases to the skin or subcutaneous tissues, these should also be presumed to be regional (ie, stage III instead of stage IV) if an appropriate staging workup does not reveal any other sites of metastases. In patients with a presumed single skin metastasis from an unknown primary site, pathology review by an experienced melanoma pathologist is appropriate to confirm that the lesion is not a variant of a primary melanoma, particularly a melanoma with a regressed junctional component. All other presentations (ie, metastases to a visceral site and no known primary melanoma) should be categorized as stage IV melanoma, using the M1 classification criteria described above reflecting metastatic site and serum LDH status.
Finally, the prognostic factors included in the melanoma staging system should be the primary stratification criteria and end results reporting criteria of melanoma clinical trials. The use of a consistent set of criteria will facilitate the reporting of melanoma treatment outcomes and comparability of melanoma clinical trials and thereby accelerate the progress of multidisciplinary melanoma treatment approaches.
Supported by a grant from the American Joint Committee on Cancer (AJCC), by Grant No. P30 CA13148 from the National Cancer Institute at University of Alabama at Birmingham, and by a Specialized Program of Research Excellence grant (P50 CA93459) in Melanoma at The University of Texas M. D. Anderson Cancer Center, Houston, TX. Three meetings held by the AJCC Melaoma Staging Committee were partually supported by an urrestricted educational grant from Schering- Plough.
The following were members of the AJCC Melanoma Staging Committee: Charles M. Balch (Chair), Johns Hopkins Medical Institutions, Baltimore, MD; Jeffrey E. Gershenwald (Vice-Chair), The University of Texas M. D. Anderson Cancer Center, Houston, TX; Seng-jaw Soong (Vice-Chair), University of Alabama at Birmingham, Birmingham, AL; Michael B. Atkins, Beth Israel Deaconess Medical Center, Boston, MA, Eastern Cooperative Oncology Group; David R. Byrd, University of Washington, Seattle, WA; Antonio C. Buzaid, Hospital Sirio Libanes, Sao Paulo, Brazil; Natale Cascinelli, Istituto Nazionale Tumori, WHO Melanoma Program, San Pio X Hospital, Milan, Italy; Alistair J. Cochran, David Geffen School of Medicine, University of California at Los Angeles, Los Angeles, CA; Daniel G. Coit, Memorial Sloan-Kettering Cancer Center, New York, NY; Alexander M. Eggermont, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands, Director, International Union Against Cancer (UICC) representative; David Frishberg, Cedars Sinai Medical Center, Los Angeles, CA, College of American Pathologists representative; Keith T. Flaherty, University of Pennsylvania, Philadelphia, PA; Phyllis A. Gimotty, University of Pennsylvania, Philadelphia, PA; Allan C. Halpern, Memorial Sloan-Kettering Cancer Center, New York, NY; Alan N. Houghton Jr, Memorial Sloan-Kettering Cancer Center, New York, NY; Marcella M. Johnson, The University of Texas M. D. Anderson Cancer Center, Houston, TX; John M. Kirkwood, University of Pittsburgh Cancer Institute, University of Pittsburgh Medical Center, Pittsburgh, PA, and Eastern Cooperative Oncology Group; Kelly M. McMasters, University of Louisville Medical Center, Louisville, KY, Sunbelt Melanoma Trial Group; Martin F. Mihm Jr, Massachusetts General Hospital, Harvard Medical School, Boston, MA; Donald L. Morton, John Wayne Cancer Institute at Saint John's Health Center, Santa Monica, CA; Merrick I. Ross, The University of Texas M. D. Anderson Cancer Center, Houston, TX; Arthur J. Sober, Massachusetts General Hospital, Boston, MA; Vernon K. Sondak, H. Lee Moffitt Cancer Center and University of South Florida College of Medicine, Tampa, FL; Kristen Stephens, Roswell Park Cancer Institute, Buffalo, NY; and John F. Thompson, Sydney Melanoma Unit, University of Sydney, Sydney, New South Wales, Australia.
Sydney Melanoma Group, Sydney, Australia, (John F. Thompson, MD); Istituto Nazionale Tumori, Milan, Italy, (Natale Cascinelli, MD); San Pio X Hospital, Milan, Italy (Natale Cascinelli, MD); Memorial Sloan-Kettering Cancer Center, New York, NY (Daniel G. Coit, MD); The University of Texas M. D. Anderson Cancer Center, Houston, TX (Jeffrey E. Gershenwald, MD; Merrick I. Ross, MD; Marcella Johnson); John Wayne Cancer Institute, Santa Monica, CA (Donald L. Morton, MD); Netherlands Cancer Institute, Amsterdam, the Netherlands (Omgo Niewig, MD); University of Pennsylvania Hospital, Philadelphia, PA (Keith Flaherty, MD; Phyllis A. Gimotty, PhD); University of Michigan, Ann Arbor, MI (Timothy Johnson, MD); H. Lee Moffitt Cancer Center, Tampa, FL (Vernon K. Sondak, MD); University of Alabama at Birmingham, Birmingham, AL (Charles M. Balch, MD; Seng-jaw Soong, PhD); Eastern Cooperative Oncology Group (John M. Kirkwood, MD; Michael B. Atkins, MD); Southwest Oncology Group (Vernon K. Sondak, MD); European Organisation for Research and Treatment of Cancer (Alexander M. Eggermont, MD); Sunbelt Melanoma Trial (Kelly M. McMasters, MD); Sentinel Lymph Node Working Group (Stanley Leong, MD); and Intergroup Melanoma Surgical Trial (Charles M. Balch, MD).
Seng-jaw Soong, PhD, Professor, University of Alabama at Birmingham (UAB); Shouluan Ding, PhD, Visiting Scholar, UAB; Matthew Dickerson, Bachelor of Science (BS), Programmer Analyst II, UAB; Rush Elliott, BS, Data Manager, UAB; Connie Pitts, Program Coordinator, UAB; Marcella Johnson, The University of Texas M. D. Anderson Cancer Center; and Carla Warneke, The University of Texas M. D. Anderson Cancer Center.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. 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 or Leadership Position: None Consultant or Advisory Role: Merrick I. Ross, PharmAdura (C); Vernon K. Sondak, Schering-Plough (C) Stock Ownership: None Honoraria: Charles M. Balch, Schering-Plough; Jeffrey E. Gershenwald, Schering-Plough; Alexander M. Eggermont, Schering-Plough; John M. Kirkwood, Schering-Plough; Kelly M. McMasters, Schering-Plough; Merrick I. Ross, Schering-Plough, Genentech, PharmAdura; Vernon K. Sondak, Schering-Plough Research Funding: Jeffrey E. Gershenwald, American Joint Committee on Cancer; Alexander M. Eggermont, Schering-Plough; John M. Kirkwood, Schering-Plough; Kelly M. McMasters, Schering-Plough Expert Testimony: None Other Remuneration: None
Conception and design: Charles M. Balch, Jeffrey E. Gershenwald, Seng-jaw Soong, John F. Thompson, Michael B. Atkins, Alistair J. Cochran, Daniel G. Coit, Alexander M. Eggermont, Keith T. Flaherty, Phyllis A. Gimotty, John M. Kirkwood, Kelly M. McMasters, Martin C. Mihm Jr, Donald L. Morton, Arthur J. Sober, Vernon K. Sondak
Provision of study materials or patients: Charles M. Balch, Jeffrey E. Gershenwald, John F. Thompson, Michael B. Atkins, Daniel G. Coit, Alexander M. Eggermont, Keith T. Flaherty, John M. Kirkwood, Kelly M. McMasters, Merrick I. Ross, Vernon K. Sondak
Collection and assembly of data: Charles M. Balch, Jeffrey E. Gershenwald, John F. Thompson, Michael B. Atkins, Daniel G. Coit, Shouluan Ding, Alexander M. Eggermont, Phyllis A. Gimotty, John M. Kirkwood, Kelly M. McMasters, Donald L. Morton, Vernon K. Sondak
Data analysis and interpretation: Charles M. Balch, Jeffrey E. Gershenwald, John F. Thompson, Michael B. Atkins, Antonio C. Buzaid, Alistair J. Cochran, Daniel G. Coit, Shouluan Ding, Alexander M. Eggermont, Keith T. Flaherty, Phyllis A. Gimotty, John M. Kirkwood, Kelly M. McMasters, Martin C. Mihm Jr, Merrick I. Ross, Arthur J. Sober, Vernon K. Sondak
Manuscript writing: Charles M. Balch, Jeffrey E. Gershenwald, Seng-jaw Soong, John F. Thompson, Michael B. Atkins, Antonio C. Buzaid, Alistair J. Cochran, Daniel G. Coit, Keith T. Flaherty, Phyllis A. Gimotty, John M. Kirkwood, Kelly M. McMasters, Martin C. Mihm Jr, Arthur J. Sober, Vernon K. Sondak
Final approval of manuscript: Charles M. Balch, Jeffrey E. Gershenwald, Seng-jaw Soong, John F. Thompson, Michael B. Atkins, David R. Byrd, Antonio C. Buzaid, Alistair J. Cochran, Daniel G. Coit, Alexander M. Eggermont, Keith T. Flaherty, Phyllis A. Gimotty, John M. Kirkwood, Kelly M. McMasters, Martin C. Mihm Jr, Donald L. Morton, Merrick I. Ross, Arthur J. Sober, Vernon K. Sondak