TY - JOUR
T1 - Predicting survival outcome of localized melanoma
T2 - An electronic prediction tool based on the AJCC melanoma database
AU - Soong, Seng Jaw
AU - Ding, Shouluan
AU - Coit, Daniel
AU - Balch, Charles M.
AU - Gershenwald, Jeffrey E.
AU - Thompson, John F.
AU - Gimotty, Phyllis
N1 - Funding Information:
ACKNOWLEDGMENT The following institutions and study groups generously contributed their patients to the 2008 AJCC Melanoma Database (an asterisk indicates the institutions and study groups that contributed localized melanoma data for this study): *Sydney Melanoma Unit, 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, and 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, and Phyllis A. Gimotty, PhD); *University of Michigan, Ann Arbor, MI (Timothy Johnson, MD); *H. Lee Moffitt Cancer Center, Tampa, FL (Vernon K. Sondak, MD, and Douglas S. Reintgen, MD); *University of Alabama at Birmingham, Birmingham, AL (Charles M. Balch, MD, Seng-jaw Soong, PhD, and Marshall Urist, MD); Eastern Cooperative Oncology Group (John M. Kirkwood, MD, and Michael B. Atkins, MD).; *Sunbelt Melanoma Trial Group (Kelly M. McMasters, MD); *Sentinel Lymph Node Working Group (Stanley Leong, MD); *Intergroup Melanoma Surgical Trial Group (Charles M. Balch, MD, and Seng-jaw Soong, PhD); National Cancer Institute, Naples, Italy (Corrado Caraco, PhD, MD). The planning and development of the AJCC staging system and predictive model has been guided by the AJCC Melanoma Task Force Committee consisting of the following members: Charles M. Balch, MD (chair), Jeffrey E. Gershenwald, MD (vice-chair), Seng-jaw Soong, PhD (vice-chair), Michael B. Atkins, MD, David R. Byrd, MD, Antonio C. Buzaid, MD, Natale Cascinelli, MD, Alistair J. Cochran, MD, Daniel G. Coit, MD, Alexander M. M. Eggermont, MD, David Frishberg, MD, Keith T. Flaherty, MD, Phyllis A. Gimotty, PhD, Allan C. Halpern, MD, Alan N. Houghton, Jr, MD, Marcella M. Johnson, MS, John M. Kirkwood, MD, Kelly M. McMasters, MD, Martin F. Mihm, Jr. MD, Donald L. Morton, MD, Merrick I. Ross, MD, Arthur J. Sober, MD, Vernon K. Sondak, MD, Kristen Stephens, CTR, John F. Thompson, MD. We thank Troy Bland, computer specialist, for programming the electronic prediction tool and for maintaining the Web site, and Connie Pitts for her assistance in manuscript preparation. Supported in part by a grant from the AJCC, by unrestricted grants from Schering Plough to the AJCC, and by research funds (including CA13148 from NCI) from the Comprehensive Cancer Center of the University of Alabama at Birmingham.
PY - 2010/8
Y1 - 2010/8
N2 - Background: We sought to develop a reliable and reproducible statistical model to predict the survival outcome of patients with localized melanoma. Methods: A total of 25,734 patients with localized melanoma from the 2008 American Joint Committee on Cancer (AJCC) Melanoma Database were used for the model development and validation. The predictive model was developed from the model development data set (n = 14,760) contributed by nine major institutions and study groups and was validated on an independent model validation data set (n = 10,974) consisting of patients from a separate melanoma center. Multivariate analyses based on the Cox model were performed for the model development, and the concordance correlation coefficients were calculated to assess the adequacy of the predictive model. Results: Patient characteristics in both data sets were virtually identical, and tumor thickness was the single most important prognostic factor. Other key prognostic factors identified by stratified analyses included ulceration, lesion site, and patient age. Direct comparisons of the predicted 5-and 10-year survival rates calculated from the predictive model and the observed Kaplan-Meier 5-and 10-year survival rates estimated from the validation data set yielded high concordance correlation coefficients of 0.90 and 0.93, respectively. A Web-based electronic prediction tool was also developed (http://www.melanomaprognosis.org/). Conclusions: This is the first predictive model for localized melanoma that was developed based on a very large data set and was successfully validated on an independent data set. The high concordance correlation coefficients demonstrated the accuracy of the predicted model. This predictive model provides a clinically useful tool for making treatment decisions, for assessing patient risk, and for planning and analyzing clinical trials.
AB - Background: We sought to develop a reliable and reproducible statistical model to predict the survival outcome of patients with localized melanoma. Methods: A total of 25,734 patients with localized melanoma from the 2008 American Joint Committee on Cancer (AJCC) Melanoma Database were used for the model development and validation. The predictive model was developed from the model development data set (n = 14,760) contributed by nine major institutions and study groups and was validated on an independent model validation data set (n = 10,974) consisting of patients from a separate melanoma center. Multivariate analyses based on the Cox model were performed for the model development, and the concordance correlation coefficients were calculated to assess the adequacy of the predictive model. Results: Patient characteristics in both data sets were virtually identical, and tumor thickness was the single most important prognostic factor. Other key prognostic factors identified by stratified analyses included ulceration, lesion site, and patient age. Direct comparisons of the predicted 5-and 10-year survival rates calculated from the predictive model and the observed Kaplan-Meier 5-and 10-year survival rates estimated from the validation data set yielded high concordance correlation coefficients of 0.90 and 0.93, respectively. A Web-based electronic prediction tool was also developed (http://www.melanomaprognosis.org/). Conclusions: This is the first predictive model for localized melanoma that was developed based on a very large data set and was successfully validated on an independent data set. The high concordance correlation coefficients demonstrated the accuracy of the predicted model. This predictive model provides a clinically useful tool for making treatment decisions, for assessing patient risk, and for planning and analyzing clinical trials.
UR - http://www.scopus.com/inward/record.url?scp=77954952885&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77954952885&partnerID=8YFLogxK
U2 - 10.1245/s10434-010-1050-z
DO - 10.1245/s10434-010-1050-z
M3 - Article
C2 - 20379784
AN - SCOPUS:77954952885
SN - 1068-9265
VL - 17
SP - 2006
EP - 2014
JO - Annals of surgical oncology
JF - Annals of surgical oncology
IS - 8
ER -