Paschalis:

You are right to be suspicious: the MSLT-1 (Multicenter Selective Lymphadenectomy) study [1] exhibits several serious methodological compromises that leave the question still not settled dispositively:

First, the MSLT-1 trial's real importance lies in staging and as a prognostic indicator, but true overall survival is not benefited, and for the observed improvement in disease-free survival (DFS) - but not overall survival - what remains true is that it is impossible to know which small fraction of intermediate-thickness melanoma patients would actually be the beneficiaries from the procedure, as we lack any validated identification criteria. [Note: there was a survival benefit accruing from SLN biopsy but only in the subset of patients who had intermediate-thickness melanomas AND were lymph node positive, and note further that this survival analysis was not the primary outcome of the trial].

Second, benefit/harm tradeoffs are not discussed despite the fact that "morbidity of procedures" was explicitly specified and required as a "Secondary Outcome Measures" in the authorizing protocol of the clinical trial approved by NCI [2] but omitted by the investigators despite the protocol requirement, and with a 10% complication rate acknowledged by the authors elsewhere [3,4], the tradeoff is especially pertinent given the absence of overall survival benefit.

Third, although the authors concluded - correctly - that (the procedure "enhances regional disease control" (RDC) for those patients with intermediate and thick melanomas, nonetheless the importance of this finding is in fact arguable since regional disease control may be obtainable subsequently when a node becomes palpable, and this deferred removal of lymph nodes still does not compromise patient survival time.

Fourth, even the conclusion of a survival benefit in the subset of node-positive intermediate-thickness melanoma patients is methodologically illicit since it derives from use of a post-randomization selected subgroup analysis, yet for this to be valid, the patients concerned must be identified and stratified BEFORE randomization, which was manifestly not done in this study.

Finally, the investigators rely extensively on the highly questionable practice of performing multiple post-hoc subgroup analyses, selectively reporting their findings, such selective reporting and post-hoc subgroup analyses being something they were already previously criticized for when they reported the interim results.

Therefore, we still await more robust prospective RCT data to help settle this long-standing issue (which MAY be reported in MSLT-2 when it completes, but since the same investigators are conducting that trial, we are not sanguine that the same, or other, methodological problems beset that trial too).

REFERENCES

1. Morton DL, Thompson JF, Cochran AJ, … MSLT Group. Final trial report of sentinel-node biopsy versus nodal observation in melanoma. N Engl J Med 2014 Feb 13; 370(7):599-609.

2. Multicenter Selective Lymphadenectomy Trial (MSLT). NCT00275496. ClinicalTrials.Gov. At: http://clinicaltrials.gov/show/NCT00275496.

3. Morton DL, Cochran AJ, Thompson JF, Elashoff R, Essner R, Glass EC, et al. Sentinel node biopsy for early-stage melanoma: accuracy and morbidity in MSLT-I, an international multicenter trial. Ann Surg 2005;242:302-11; discussion 11-3.

4. Scoggins CR, Martin RC, Ross MI, Edwards MJ, Reintgen DS, Urist MM, et al. Factors associated with false-negative sentinel lymph node biopsy in melanoma patients. Ann Surg Oncol 2010;17:709-17.

The issue of sentinel node biopsy (SLNB) has intensively been discussed during the past years. The MSLT-1 trial has some shortcomings but remains a landmark study regarding the design, number of patients, and follow-up. Unfortunately, the recent paper again focuses on a speculative survival benefit based on not accordingly powered subgroup analyses.

Most data suggest that the survival benefit, if any, is at best rather small. Whether or not the picture might differ between certain subgroups or patient populations will probably remain an unanswered question, just as it was the case for elective lymph node dissection (ELND) in melanoma for decades.

The most important factor is the powerful dissection of groups at high risk for relapse, and therefore the identification of patients eligible for adjuvant treatment. This might become even more important as new treatment options for melanoma emerge.

Counterwise, in practice it is very reassuring for patients to learn that their melanoma has not spread to the nodes, improving their survival perspective as compared to the estimates with unknown sentinel status.

One comment to the question of local control. I would not necessarily accept that regional tumor control is as favorable with clinically detected nodes, in particular if a thorough follow-up with experienced lymph node ultrasound is not performed.

One may add yet another aspect to the quite sensible response that Dr. Kaniklides gave.

The hypothesis of "false-positive" SLNB has been brought up by some authors, stating that patients might undergo unnecessary surgery, adjuvant treatment etc. with small positive sentinels that would never become "true" metastasis.

The results on MSLT-1 in NEJM 2014 gives valuable information, that after sufficient the rate of regional becomes virtually the same in both groups. This is clear indication that a "false-positive" sentinel, if existent at all, is a very rare event. After all, MSLT-1 provides good evidence for the validity of the method of SLNB in melanoma patients.