MSLT-I results should impact future clinical trial design in melanoma

Issue: May 25, 2015

NEW YORK — Results from the Multicenter Selective Lymphadenectomy Trial-I, or MSLT-I, should influence eligibility criteria and patient stratification within future adjuvant trials in stage III melanoma, according to a presentation at the HemOnc Today Melanoma and Cutaneous Malignancies meeting.

“As we move into an era with new and effective drugs in stage IV melanoma, and as we move into adjuvant trials for stage III with more agents — which have a different risk and toxicity profile — we are going to have to do an even more precise job of calibrating the patients we recommend to go on to clinical trials,” Charles M. Balch, MD, professor of surgery at the University of Texas Southwestern Medical Center, said during his presentation.

Every patient with melanoma enrolled on an adjuvant therapy clinical trial must have pathological staging of their lymph nodes, and stratification criteria should include primary tumor thickness with a threshold of 4 mm, tumor ulceration, nodal tumor burden, the number of metastatic nodes and patient age. Further, until the results of the MSLT-II trial are published, all patients with node-positive disease determined by sentinel lymph node biopsy (SLNB) should undergo completion lymph node dissection (CLND).

MSLT-I

Data from MSLT-I indicated sentinel lymph node biopsy (SLNB)-determined sentinel node negativity or positivity are significantly prognostic for the risks for disease recurrence and death from melanoma (P ˂ .001 for both). The most powerful predictor of survival among patients with tumor thickness greater than 1.2 mm in the trial was the presence or absence of sentinel node metastases, Balch said.

“Regardless of whether there is a survival benefit, the staging value is compelling,” Balch said.

Results from MSLT-1 demonstrated that SLNB was significantly associated with improved DFS for patients with intermediate thickness melanomas (1.2 mm-3.5 mm; P = .01) and thick melanomas (≥ 3.5 mm; P = .05). A final report of 290 patients with thick melanomas indicated SLNB offered prognostic information but did not significantly improve survival outcomes.

Among node-positive patients, early lymphadenectomy was significantly associated with improved melanoma-specific survival for melanomas of all Breslow thickness (HR = 0.66; 95% CI, 0.47-0.91) and intermediate-thickness melanomas (HR = 0.56; 95% CI, 0.37-0.84) compared with regional lymphadenectomy delayed until the development of regional disease.

Long-term follow-up results from MSLT-I — published in 2014 in The New England Journal of Medicine — demonstrated that the benefits associated with SLNB among patients with melanomas of intermediate thickness persisted in terms of melanoma-specific survival (P ˂ .05), distant DFS (P ˂ .05) and DFS (P ˂ .0001).

These data and other studies led to the development of joint ASCO and Society of Surgical Oncology guidelines. According to the guidelines, CLND is recommended for all patients with a positive SLNB and should be performed for these patients until there is evidence that the procedure does not improve regional disease control or survival.

“This should be our standard of care, especially in patients entered into clinical trials, because we don’t know whether that can influence the outcome,” Balch said.

MSLT-II

The goal of the MSLT-II trial is to determine whether CLND is necessary for all patients with sentinel lymph node-positive disease. The trial has reached its accrual target and is composed of 1,939 patients at 64 sites.

Balch presented early MSLT-II data on baseline nodal ultrasound analyses. Previous data have suggested that ultrasound could detect micrometastases.

However, results from MSLT-II indicated that a high proportion of patients with positive ultrasounds had negative pathology, and patients with negative ultrasounds had positive pathology. Sensitivity of preoperative ultrasound was 8.1%, specificity was 97.5% and the negative predictive value was 80.5%.

Thus, preoperative ultrasound is not sufficiently accurate to detect occult metastases, Balch said.

Although overall trial results will not be available for 5 to 10 years, researchers should be cautious when interpreting the results, Balch said.

“My concern about [MSLT-II] results, regardless of how they turn out, is that there is a significant referral bias,” Balch said. “Patients entered into this trial were selected by the surgeon’s judgement that their probability of sentinel node metastases was small because they had very little micrometastases in their sentinel node. So this trial in fact may be negative because of the dilutional effect of patients without sentinel node metastasis, which will be the majority of patients entered into the trial, and the judgement of surgeons excluding those patients … who had a higher risk of sentinel node metastases.”

Subset analyses should be conducted to evaluate the association between tumor volume and the incidence of non-sentinel node metastases, Balch said. — by Alexandra Todak

References:

Balch CM. Results of MSLT I/II melanoma surgical trials: How they can be used in clinical decisions. Presented at: HemOnc Today Melanoma and Cutaneous Malignancies Meeting; April 10-11, 2015; New York.

Morton DL, et al. N Engl J Med. 2014;doi:10.1056/NEJMoa1310460.

Wong SL, et al. J Clin Oncol. 2012;doi:10.1200/JCO.2011.40.3519.

Disclosure: Balch reports financial relationships with Amgen and Merck.