Experts debate role of completion lymphadenectomy for node-positive melanoma

Issue: April 25, 2016

NEW YORK — Two speakers debated whether completion lymphadenectomy is still standard of care for patients with sentinel lymph-node positive melanoma at HemOnc Today Melanoma and Cutaneous Malignancies.

The ASCO/Society of Surgical Oncology joint guidelines on sentinel lymph node biopsy (SLNB) recommend completion lymph node dissection (CLND) for any positive node. Further, evidenced-based guidelines from the National Comprehensive Cancer Network state CLND should be discussed and offered for patients with stage III, lymph node-positive disease.

“These guidelines often offer precious little in the way of actual guidance, in part because they include a use of words that don’t really help us understand what we are supposed to do,” Vernon K. Sondak, MD, chair of the department of cutaneous oncology at Moffitt Cancer Center and a HemOnc Today Editorial Board member, said during his presentation. “Do we ‘discuss,’ ‘offer,’ ‘recommend?’ Is it ‘indicated,’ ‘necessary,’ ‘beneficial?’ What are we really trying to say?”

Vernon K. Sondak

Standard of care

The final analysis of the Multicenter Selective Lymphadenectomy Trial-I, or MSLT-I, provides justification for the continued use of CLND as the standard of care, Sondak said. In that trial, researchers randomly assigned node-positive patients to immediate CLND or observation.

Results showed patients with intermediate-thickness melanomas who received immediate CLND had significantly superior 10-year melanoma-specific survival (HR = 0.55; 95% CI, 0.37-0.82).

However, separation of the survival curves did not occur until after 2 years, Sondak said.

“It wasn’t until well after leaving the tumor-containing node in place for 2 or 3 years that the impact of initial treatment became manifest,” he said. “The curves are widening, not narrowing, with time. So we have to be very careful about early conclusions about this important topic.”

Sondak also noted that SLNB identified an average of 1.4 positive nodes, whereas observation patients who recurred in the nodes had an average of 3.2 positive nodes (P ˂ .001).

“Observation of the sentinel node-positive basin is associated with a substantial risk of in-basin failure, seemingly well in excess of what would be expected from the number of positive non-sentinel nodes,” Sondak said. “Therefore, it is likely that subclinical involvement, rather than secondary metastases, explain multiple nodal metastases seen in regional recurrences.”

The timing of CLND also impacts the morbidity of the dissection, Sondak said. Early CLND is associated with lower rates of wound separation, seroma/hematoma, hemorrhage and infection. Further, significantly fewer patients who undergo early CLND experience lymphedema (12.4% vs. 20.4%; P = .043).

“My recommendation is that lymph node dissection is indicated for all patients with clinically positive nodes, and CLND is the standard of care recommendation for all patients with a positive SLN biopsy,” Sondak said. “For every patient who is a long-term survivor without regional or distant recurrence after nodal basin observation, there will be another who is disadvantaged by regionally recurrent disease that is more difficult to treat and/or more likely to kill the patient.”

Lack of OS benefit

There are other ways of looking at the data from the MSLT-I trial that may not support the use of CLND as standard of care, Axel Hauschild, MD, professor of dermatology at University Hospital in Kiel, Germany, said during his presentation.

Axel Hauschild

Although subgroup analyses may have shown benefit, the trial was designed to improve OS, which did not differ between the biopsy and observation groups according to the first trial results published in 2006, Hauschild said.

However, DFS outcomes did significantly differ between SLNB-positive and –negative patients (72.3% vs. 90.2%).

“This is not surprising,” Hauschild said. “We know that patients have inferior disease-free survival when they have a positive node vs. a negative node.”

Results of the final analysis published in 2014 showed no difference in melanoma-specific survival, and only in DFS, among patients with intermediate-thickness and thick melanomas.

The cumulative incidence of nodal metastasis did not greatly differ among patients with a false-negative SLNB vs. those who underwent observation for intermediate-thickness (21.9% vs. 19.5%) and thick melanomas (42% vs. 41.4%).

These similar rates of metastases were achieved in the observation arm, despite that most patients did not receive treatment.

“Don’t forget that 80% of the patients in the observation arm never needed completion lymphadenectomy, so therefore they could not suffer from lymphedema, infection and other complications,” Hauschild said. “The vast majority of patients were untreated, and their regional lymph node was not touched because there was no relapse.”

Hauschild then reviewed data from the DeCOG trial, which evaluated observation vs. CLND in patients with a positive sentinel node.

Results showed no difference in distant metastases-free survival, which was the study’s primary endpoint, at 3 years (HR = 1.02; 95% CI, 0.67-1.56). Although Hauschild agreed 3 years is still early to assess survival, he said the data do not suggest there will be a later difference.

CLND also did not show a benefit for melanoma-specific survival (HR = 1.01; 95% CI, 0.64-1.59), RFS (HR = 0.89; 95% CI, 0.63-1.27) and cumulative incidence of recurrence (HR = 0.86; 95% CI, 0.6-1.24).

Based on results from this trial, CLND is no longer standard of care for sentinel-node positive patients with less than 1 mm-thick lymph node metastases according to German guidelines that will go into effect April 1, Hauschild said. In patients with larger metastases, CLND should be discussed with the patient, but it is no longer the standard of care.

“If the future lies in precision medicine initiatives, we might not even touch this issue of surgical issues for long, because we have adjuvant modalities that might be more effective than in the past,” Hauschild said. “I believe for the future, completion lymphadenectomy for low-risk patients with low tumor burden in the sentinel node is no longer standard of care.” – by Alexandra Todak

Reference:

Hauschild A and Sondak VK. Great debate: Lymph node dissection for the sentinel node-positive patient. Presented at: HemOnc Today Melanoma and Cutaneous Malignancies; March 18-19, 2016; New York.

Disclosure: Hauschild reports no relevant financial disclosures. Sondak reports consultant roles with Amgen, Bristol-Myers Squibb, GlaxoSmithKline, Merck, Navidea, Novartis and Provectus.