| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
1 Sydney Melanoma Unit and Melanoma and Skin Cancer Research Institute, Sydney Cancer Centre, Gloucester House, Royal Prince Alfred Hospital, Camperdown, New South Wales 2050, Australia
2 Department of Surgery, University of Sydney, Sydney, New South Wales 2006, Australia
Correspondence: Address correspondence and reprint requests to: John F. Thompson, MD; E-mail: thompson{at}smu.org.au
There is general consensus that sentinel node biopsy (SNB) is a valuable staging tool in patients with intermediate-thickness melanomas. The results of the 2000-patient Multicenter Selective Lymphadenectomy Trial (MSLT) suggest that SNB may also provide a survival advantage for patients with a positive sentinel node (SN) who have an immediate completion lymph node dissection (CLND) compared with those who subsequently develop clinically detectable metastatic disease in their regional lymph nodes and have a therapeutic lymph node dissection.1 It cannot necessarily be assumed that all patients who have micrometastatic disease in an SN will ultimately develop clinical disease in their regional lymph nodes if that SN is not removed. However, the finding that the percentage of patients with a positive SN (19%) was virtually identical to the percentage of patients who subsequently developed clinical disease in their regional lymph nodes in the control group (18%), after a median follow-up period of almost 5 years, seems very unlikely to be coincidental. These recent MSLT findings are consistent with the previously reported randomized World Health Organization Melanoma Program study of elective lymph node dissection for truncal melanomas >1.5 mm in thickness, which also suggested that early removal of nodes containing metastatic disease confers a survival benefit.2
Although it thus appears that there may be a survival benefit for some patients who have an SNB and then a CLND if a positive SN is found, the situation for patients with thick melanomas (>4.0 mm) remains in doubt. It seems likely that a substantial proportion of thick melanomas metastasize to systemic sites at the same time as they metastasize to regional lymph nodes, so that control of disease in the regional lymph nodes may be of relatively less importance.3
The situation for patients with thin melanomas (
1.0 mm) is similarly controversial. The matter is of great clinical significance, since thin melanomas account for the great majority of melanomas in most developed countries, and the overall prognosis for patients with thin melanomas is very good. Population-based data from Australia reveal a 10-year survival rate of >92% in these patients.4 Nevertheless, some patients with a thin primary melanoma (TPM) do develop metastatic disease, but often after a protracted period of observation. The key question, therefore, is which, if any, patients with TPMs are likely to benefit from SNB? In their article in this issue of Annals of Surgical Oncology, Wong et al. have thoroughly reviewed the literature and presented their data from the Memorial Sloan-Kettering Cancer Center in an attempt to clarify the situation.5 This prompted us to examine our own Sydney Melanoma Unit (SMU) data on patients with TPMs who had undergone SNB. Our experience has been very similar to that of Wong et al. (Table 1
).
|
DIFFERING CRITERIA FOR SELECTION OF "HIGH RISK" PATIENTS WITH TPMs AND THE SMALL NUMBER OF PATIENTS WITH POSITIVE SNs
It has been difficult to derive any meaningful conclusions on patients with TPMs undergoing SNB until now, because only one study6 had more than nine patients with TPMs and SN metastases, and most studies had fewer than five patients in this category. Those groups performing SNB on patients with TPMs have attempted to select patients on the basis of their perception of "unfavorable characteristics": mostly Clark level IV or V invasion, the presence of ulceration, and/or the presence of regression. However, not all series have found Clark level or ulceration to be predictive factors for SN positivity, and the evidence for regression being predictive is sparse. One fact that does emerge is that SN positivity has been zero in patients with melanomas .75 mm in all series except two,7,8 in which a total of only three melanomas .5 mm in thickness and with no identifiable high-risk factors metastasized to an SN. In relation to the selection of "high-risk" groups of patients with TPMs, a recent analysis of 884 such patients treated in the pre-SNB era provides useful information.9 Only vertical growth phase (VGP), tumor mitotic rate (TMR) >0, the presence of tumor-infiltrating lymphocytes, and the presence of ulceration were found to be independent predictors of survival after a median follow-up period of 200 months. Paradoxically, no survival difference was found in this study for patients with melanomas .75 mm in thickness and those that were .76 to 1.0 mm in thickness or between those with Clark level II/III invasion and those with Clark level IV invasion. These were the only two parameters found to be of value in predicting SN positivity in the study by Wong et al.
In none of the reported series have any additional positive nodes been found on CLND. In some series, patients with TPMs have been selected for SNB on the basis of VGP being present, with some other "high-risk" factors. Kesmodel et al.7 found, for example, that TMR was a highly significant predictor of SN positivity and therefore advocated SNB in patients with VGP and a high TMR. The SMU experience (unpublished data) has also been that TMR is a highly significant predictor of SN positivity. Another putative "high risk" factor that has emerged recently is young patient age,10 variously described as <45 or <40 years of age. Again, the SMU experience supports this proposal, and we have considered SNB in patients <45 years of age with melanomas .75 to 1.0 mm in thickness. In some centers, particularly in North America, a high proportion of patients with TPMs will present requesting that SNB be performed. It seems that in some studies up to half the patients with TPMs undergoing SNB were treated in this way because they requested it. In the SMU experience of SNBs for patients with TPMs, after completion of patient accrual to the MSLT, this was the basis on which SNB was performed in almost one third of the patients.
INADEQUATE DURATION OF FOLLOW-UP
It is well documented that although they recur relatively infrequently, thin melanomas can do so after prolonged periods. The follow-up intervals in almost all studies reporting the results of SNB in patients with TPMs are far too short to make valid estimates of ultimate outcome. The median follow-up intervals in all published series and in our own SMU series are <5 years and mostly <3 years.
One matter that is often overlooked when considering the issue of SNB for patients with TPMs is the cost-effectiveness of this procedure. Examining the situation for patients with melanomas <1.2 mm in thickness, Agnese et al.11 estimated the cost of identifying a single positive SN as between US$696,000 and US$1,051,000, yet only a small number of patients will have disease identified that will alter their prognosis. It is likely that cost would be substantially lower in many other countries, but the expense of an SNB procedure nevertheless needs to be weighed against the possible benefit.
What, then, should be the treatment recommendation for patients with thin melanomas? It seems that follow-up times are generally too short to draw any conclusion about whether SNB is of value in such patients, and the available evidence suggests that SN status may not be a reliable predictor of outcome in this patient group in any case, whereas it does appear to be useful in those with intermediate-thickness melanomas. Larger studies of patients with TPMs with more strictly defined criteria for performing SNB and with much longer periods of follow-up are clearly required. For the moment, then, it appears that there is no place for SNB in patients with tumors <.75 mm in thickness and probably no place for SNB in those with tumors between .75 and 1.0 mm either, unless there are patient or tumor features that indicate a higher risk than is predicted by the tumor thickness alone. However, even if an SNB is performed and a positive SN is found, there is no indication for CLND.
Received for publication August 22, 2005. Accepted for publication August 25, 2005.
REFERENCES
1-mm melanoma. Langenbecks Arch Chir (in press).
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
