1. Home>
2. Specialties>
3. Dermatology>
4. Summary and Comment

The Prognosis for Sentinel Lymph Node Biopsies

From "Are we there yet?" to "Where are we now?"

For some, the status of sentinel lymph node biopsy, or lymphatic mapping/sentinel lymphadenectomy (LM/SL), as standard surgical care is highly dependent on the results of the Multicenter Selective Lymphadenectomy Trial-I (MLST-I). For others, the technique has become so ingrained that its well-established prognostic power justifies its continued use, regardless of the trial outcomes. Now that MLST-I has been published, the short answer is that survival is not altered by LM/SL.

Patients with primary melanomas of intermediate thickness (1.2–3.5 mm) were randomized to wide resection and observation (OBS; 500 patients) or to wide resection and sentinel-node biopsy (SN; 764 patients). Patients with nodal involvement underwent subsequent completion lymphadenectomy. There was no difference in melanoma-specific survival between the SN and OBS groups (5-year survival, 87.1% vs. 86.6%). As in many previous studies, patients with tumor-negative sentinel nodes (SN–) did much better than the patients with tumor-positive (SN+) sentinel nodes (5-year survival, 90.2% vs. 72.3%; P<0.001), demonstrating the prognostic role of sentinel-node biopsy in a prospective setting. The disease-free survival data are difficult to interpret, as most recurrences in the OBS group occurred in the regional basin, an event largely eliminated by the LM/SL procedure. It is worth mentioning, however, that with LM/SL, bulky nodal recurrences are dramatically diminished because the diseased node is removed earlier.

Seventy-eight (15.6%) of the OBS group had clinical relapse (the CN+ subgroup), whereas 122 SN patients (16%) were SN+. Some might argue that this near equivalence in ultimate nodal involvement suggests that all SN+ patients would eventually have metastasis to regional nodes, although the true rate of SN+ patients in the OBS arm was obviously not discoverable. Put another way, this argument proposes that microscopic nodal disease is a direct precursor to macroscopic nodal disease, rather than a biologically distinct class of disease. The alternative hypothesis would be that a systematic bias led to an overrepresentation of microscopic disease in the OBS arm and that a fraction of these individuals eventually relapsed clinically.

Perhaps the most interesting (and possibly the most controversial) point was that the 5-year survival rate was significantly greater in the SN+ patients than in the CN+ patients (72.3% vs. 52.4%; P=0.004). The authors conclude that early lymphadenectomy (i.e., LM/SL) conferred a survival advantage over delayed lymphadenectomy (essentially reducing the trial to the 122/764 SN+ patients and the 78/500 CN+ patients). Supporters of LM/SL would argue that the presence of a large population of healthy node-negative patients in both arms diluted the detectable survival benefit and statistically masked it when the trial was examined in total. A similar argument was used in defense of earlier elective lymph-node dissection when trials showed a like absence of survival benefit.

More CN+ patients than SN+ patients had nodal stage 2 and nodal stage 3 disease, which again has obvious prognostic implications given the better survival rate in SN+ patients; this finding also strongly suggests that there is a stepwise increase in involvement from one node to multiple nodes.

Comment: Several major take-home messages emerge from this largest prospective LM/SL study to date. First, the prognostic significance of LM/SL has been reaffirmed, which is not surprising, given the extant data. Second, the near absence of clinical nodal relapse after LM/SL (4%) is worth noting, as some patients who do not undergo LM/SL will relapse with bulky regional disease requiring extensive surgery. So the heart of the argument is which to consider: the lack of an overall survival benefit in the SN group over the OBS group or the presence of a survival benefit in the SN+ group over the CN+ group. There is no obvious right answer.

The authors argue that because LM/SL improves survival in node-positive patients, this procedure should be preferred over observation. However, only about 15 of every 100 patients receiving LM/SL will benefit from earlier node removal. Therefore, the more precise interpretation is that LM/SL should be performed on patients known to have nodal involvement at the time of wide resection. This is currently impossible, although advances in imaging might render it workable in the future.

Proponents of LM/SL will cite as benefits prognostic precision, identification of patients eligible for interferon, and reduction in regional morbidity in a small subset of patients. Critics will cite the lack of an overall survival benefit and the absence of evidence that interferon has any effect on microscopic nodal disease. So what should we do with our patients? A discussion with the patient of pros and cons, including the data from this trial, is a good idea. Such a discussion is clearly spelled out in the recommendations of the National Comprehensive Cancer Network. LM/SL is unlikely to go away in the near future based on these results. The long-term utility of LM/SL requires further study, and sequels to MLST-I have already begun.

— Hensin Tsao, MD, PhD

Published in Journal Watch Dermatology September 29, 2006

Citation(s):

Morton DL et al. Sentinel-node biopsy or nodal observation in melanoma. N Engl J Med 2006 Sep 28; 355:1307-17.

• Original article (Subscription may be required)
• Medline abstract (Free)

Balch CM and Cascinelli N. Sentinel-node biopsy in melanoma. N Engl J Med 2006 Sep 28; 355:1370-1.

• Original article (Subscription may be required)
• Medline abstract (Free)