Sentinel lymph node biopsy for melanoma: a plea to let the data speak.

Abstract

Twenty-two years ago, Morton and colleagues reintroduced the world to a deceptively simple concept: that the initial regional node to which a melanoma would metastasize (if it was going to do so) could reliably be identified, removed, and examined to provide important prognostic information and help guide treatment decisions for patients with this diagnosis. At a time when routine elective lymph node dissection had been largely abandoned as excessively morbid with no impact on survival, the robust and reproducible technique of sentinel lymph node biopsy (SLNB) was rapidly adopted in the initial management of patients presenting with localized invasive melanoma; further, clinical relevance was soon found in guiding the treatment of many more patients with breast cancer. In the interim, while others have either analyzed retrospective series or criticized the procedure, no other single physician devoted more time and energy than Dr. Morton to rigorously and critically evaluate the role of SLNB in melanoma, conceiving and completing two seminal international prospective randomized surgical trials, each accruing nearly 2,000 patients, to define the value of initial management with SLNB (MSLT-1) and, among patients with a positive SLNB finding, the role of completion lymph node dissection (MSLT-2). The final report of findings from MSLT-1, published in the New England Journal of Medicine, has provided clinicians with invaluable, highquality information to interpret and use in the management of their melanoma patients. I write to express both my enormous appreciation for the opportunity to examine these data—but also my concern about how the authors have presented this information. The final report of MSLT-1 has largely confirmed the principal findings of the initial report published in 2006 on patients with intermediate-thickness melanomas (1.2– 3.5 mm) and provides new information on patients with thick melanoma ([3.5 mm). Initial management with wide excision (WE) and SLNB is not associated with any improvement in melanoma-specific survival, the primary end point of the trial, compared to initial management with WE and nodal basin observation (OBS). This is a negative trial. This result should not surprise us, because the trial design of MSLT-1 was essentially identical to that of the three other major randomized trials of elective lymph node dissection with negative findings, with the principal difference being that the patients in MSLT-1 with negative sentinel nodes at presentation were spared the morbidity of elective lymphadenectomy. The current report confirmed an improvement in recurrence-free survival after SLNB, a secondary end point of the trial, for patients with both intermediate-thickness and thick melanomas. As clearly shown in the article’s supplementary appendix, this difference was primarily due to an expected increased incidence of initial nodal failure in the OBS group. Although this is listed as one of the end points that ‘‘validate’’ the use of SLNB, the authors fail to provide information on the ultimate clinical relevance of improved nodal basin control after SLNB. The authors might cite the potential for incrementally higher complication rates after therapeutic LND for a clinically positive node compared to completion LND for a positive sentinel node, and an incrementally higher rate of nodal basin relapse after therapeutic LND than after completion LND. However, these events will affect only a small subset of the subset of patients with positive nodes. How many more actual patients in the OBS group experienced more complications following therapeutic LND, or ultimate loss of Society of Surgical Oncology 2014