For decades, really, when patients with melanoma were found to have a positive sentinel node they routinely were advised to go on to complete lymph node dissection. Interestingly, in practice this only happened about half of the time and we’re not quite sure why. There was not really any evidence to support this but it was clinical practice. Recently two prospective randomised clinical trials, one an international trial and the other one a German trial, concluded that virtually all of the clinically important endpoints were similar whether a patient with a positive sentinel node had a complete lymph node dissection or not. Given the fact that there are significant side effects associated with complete lymph node dissection this leads one to recommend that the preferred pathway for patients with a positive sentinel node is observation.
A certain percentage of those patients, somewhere between 5-15%, will in follow-up develop lymph node recurrence only and those patients we do recommend a delayed therapeutic lymph node dissection for. The remainder of the patients actually recur initially outside of the lymph node basin and in those patients it’s pretty clear that immediate complete lymph node dissection would have made no difference.
These two trials, and they are very consistent results in terms of no difference in either distant metastasis free or melanoma specific survival associated with more surgery, should be construed as practice changing. I say that with full awareness of how difficult it is to change practice. I heard discussed at a meeting last month a poll of practice in the UK where nearly 50% of people said that the results of this trial would not change their practice. This speaks to the real challenge of disseminating information and changing clinical practice in medicine and particularly in surgery, a field that tends to be pretty conservative in terms of change. But at the end of the day the absolute consistency of these two prospective randomised clinical trials leaves very little doubt that immediate complete lymph node dissection for melanoma patients with a positive sentinel node is no longer the preferred pathway to pursue.
Is it possible to stratify those with a lymph occurrence earlier in order to make a decision sooner?
Both trials examined the whole issue of whether there were predictors of who should and should not have an immediate complete lymph node dissection. Interestingly, both trials were again remarkably consistent in demonstrating that there were no predictors of who might and might not benefit, or particularly who might benefit, from immediate complete lymph node dissection. We’ve long used a criteria recommending lymph node dissection as significant tumour in the sentinel node, either large size of tumour or multiple lymph nodes involved. Neither of these prospective studies suggested that those patients derived any benefit from immediate complete lymph node dissection.
Anything else to add?
The one other point I did want to make is that one of the concerns about omitting complete lymph node dissection is the fact that there is some prognostic information that can be derived from the status of the non-sentinel lymph nodes. Without complete lymph node dissection it becomes a little bit more challenging to risk stratify those patients into those at high risk for recurrence who might be eligible for adjuvant therapy. There’s a general movement right now across the melanoma community to develop effective risk stratification criteria that will be functional and informative without the information provided by the non-sentinel lymph node status. So we’re going to move past the prognostic value of complete lymph node dissection and recognise that where it has little therapeutic value, little if any therapeutic value, it’s just too expensive a piece of information to insist on for prognostic information. We’re going to find alternative ways to get that information.