1. Sentinel Lymphadenectomy Identifies Biologically Significant Nodal Metastases In Early-Stage Melanoma: Results Of MSLT-I -- An International Multicenter Trial
Donald L. Morton, MD1, John F. Thompson, MD*2, Robert Elashoff, PhD*1, Richard Essner, MD*1, Nicola Mozzillo, MD*3, Omgo E. Nieweg, MD, PhD4*, Daniel F. Roses, MD*5, Harald J. Hoekstra, MD, PhD6, Constantin Karakqusis, MD7, Douglas S. Reintgen, MD*8, Brendon J. Coventry, MD9, Alistair J. Cochran, MD*1, Multicenter Selective Lymphadenectomy Trial Group, (MSLT)1
1John Wayne Cancer Institute, Santa Monica, CA; 2Sydney Melanoma Unit, Sydney, Australia; 3Istituto Nazionale Dei Tumori de Napoli, Naples, Italy; 4Netherlands Cancer Institute, Amsterdam, The Netherlands; 5New York University, New York, NY; 6Groningen University Hospital, The Netherlands; 7Millard Fillmore Hospital, Buffalo, NY; 8H. Lee Moffitt Cancer Center, Tampa, FL; 9Royal Adelaide Hospital, Adelaide, South Australia
Elective complete lymphadenectomy (CLND) in early-stage melanoma remains controversial because most (80%) patients do not have nodal metastases and cannot benefit. We developed lymphatic mapping and sentinel lymphadenectomy (LM/SL) to identify patients with clinically occult regional metastases. An international trial evaluating the diagnostic and prognostic accuracy of LM/SL is reported.
A 17-center trial randomized 2001 patients with melanoma > 1.0 mm Breslow to two treatment arms: wide excision plus LM/SL, with immediate CLND for sentinel node (SN) micrometastasis (60%); or wide excision plus postoperative nodal observation (WEO), with delayed CLND for nodal recurrence (40%). The intraoperative incidence of SN micrometastases in the LM/SL group was compared to the postoperative incidence of nodal recurrence in the WEO group. Patients were stratified by site and Breslow thickness. Statistical analysis used log rank and Cox regression.
After a median follow-up of 54 months, the 5-year melanoma-related survival of LM/SL patients dropped from 88% to 71% when the SN contained micrometastases (p = .0001). The false-negative rate of LM/SL was 5.2% (25/483) after 50 procedures. The incidence of SN metastases (LM/SL) vs. postoperative nodal recurrence (WEO) was 15% vs. 16.1% when Breslow was 1.2-3.5 mm, and 33.0% vs. 32.8% when Breslow exceeded 3.5 mm.
SN involvement is the most important prognostic factor for disease-free and melanoma-specific survival; SN micrometastases will lead to nodal recurrence if not removed with the primary melanoma. The accuracy of LM/SL increases with surgical volume and reaches 95%. LM/SL should become standard care for primary melanoma.