From Wiki Journal Club
Jump to: navigation, search
Morton DL, et al. "Multicenter Selective Lymphadenectomy Trial". The New England Journal of Medicine. 2014. 370(7):599-609.

Clinical Question

Among patients undergoing a wide excision for melanoma, does sentinel node biopsy with reflex completion lymphadenectomy improve melanoma specific survival compared with nodal observation?

Bottom Line

Among all patients undergoing a wide excision for melanoma, sentinel node biopsy improves disease free survival compared to nodal observation. Further, in intermediate-thickness melanomas (1.2-3.5cm), sentinel node biopsy provides useful prognostic information, and may prevent local nodal recurrence.

Major Points

Sentinel Lymph Node Biopsy (SLNB) is a minimally invasive staging procedure performed at initial wide excision of melanoma, often using dual methylene blue dye and lymphoscintigraphy with intraoperative gamma probe. SLNB provides crucial prognostic information to stratify patients with melanoma in the absence of clinically apparent nodal disease. Positive SLNB evokes a Completion Lymph Node Dissection (CLND) with the goal of improved lymph node basin control, in addition to the potential for adjuvant systemic therapy. However, CLND is associated with a substantial risk of morbidity, including chronic problems such as lymphedema and neuropathy.

The Multicenter Selective Lymphadenectomy Trial [1] sought to investigate the benefit of SLNB with CLND compared with nodal observation in a cohort with majority intermediate (1.2-3.5mm) and thick melanomas (>3.5mm), which are more likely to have subclinical nodal involvement than thin melanomas (<1.2 mm). Although SLNB did not demonstrate an effect on the primary endpoint of melanoma specific survival at 10-year follow up, MSLT-I showed that SLNB does have an effect on disease free survival, predominantly by preventing nodal recurrence in melanomas > 1.2 mm. The trial also confirmed the value of the prognostic information of SLNB. Finally, MSLT-I purported a survival advantage in terms of microscopic versus macroscopic detection of nodal metastases in intermediate thickness melanomas, (10 year Disease Specific Survival, 62% Biopsy with nodal metastasis versus 41.5% Observation with nodal metastasis, P = 0.006). The results of this latter result were further investigated by the subsequent MSLT-II trial [2], which failed to demonstrate melanoma-specific survival among patients with melanoma with sentinel-node metastases and completion lymph-node dissection compared to nodal observation.


NCCN guidelines (Version 1.2018)

  • Recommend SLNB be discussed and offered for patients with higher-risk stage 1B (>1mm thick or 0.76-1.00mm thickness with ulceration, mitotic rate ≥ 1 mm^2), and can be considered in stage II.

Society for Surgical Oncology (SSO) and American Society of Clinical Oncology (ASCO) Clinical Practice Guidelines, Published December 13, 2017

  • SLNB recommended for melanomas with Breslow thickness 1.0-4mm (T2, T3)
  • SLNB may be offered for melanomas with Breslow thickness >4mm (T4)


  • Phase III, multicenter, randomized controlled trial
  • N=2,001 patients (1,661 randomized)
    • 340 thin melanomas (<1.20 mm)
    • 1347 intermediate thickness melanomas (1.2-3.5mm)
      • 770/814 (94.6%) underwent Biopsy
      • 500/533 (93.8%) underwent Observation
    • 314 thick melanomas
      • 173/186 (93.0%) underwent Biopsy
      • 117/128 (91.4%) underwent Observation
  • Setting: Multiple clinical centers in the United States and Internationally
  • Enrollment: 1994-2002
  • Mean follow-up: 10 years
  • Analysis:
    • Parallel Per-protocol and intention-to-treat analysis
    • Kaplan-Meier method to estimate 5-year and 10-year rates of melanoma-specific and disease-free survival
    • Multivariate analysis for disease recurrence and death among patients with Intermediate-Thickness Melanoma who underwent SLNB (cofactors: SNLB status, Breslow thickness, ulceration, site of melanoma, sex, age, Clark level)
    • Latent subgroup analysis to account for ascertainment bias known only in biopsy group
  • Primary endpoint: Melanoma specific survival at 10 year follow-up
  • Secondary endpoints: Disease free survival, survival with tumor-positive or tumor-negative sentinel nodes, and the incidence of sentinel node metastases, as compared with the incidence of clinically detected nodal metastasis


Inclusion Criteria

Described in a previous publication.[3]

  • Primary melanomas of the head and neck, extremities, trunk, sole of foot, palm of hand, or subungal
  • age 18 to 75
  • Breslow thickness ≥ 1.00mm AND Clark level III
  • Any Breslow thickness AND Clark level IV or V

Exclusion Criteria

  • Primary melanomas of the ear
  • Entered trial no more than 10 weeks after skin biopsy
  • Lymphatic mapping/sentinal lymph node biopsy performed within 12 weeks of diagnosis
  • History of prior operative procedure disruptive of lymphatic drainage from primary site, including prior wide excision of the primary with shortest margin >=1.5cm margins
  • History of melanoma or other invasive malignancy within 5 ears of diagnosis
  • Life expectancy < 10 years
  • Primary or secondary immune deficiency
  • Pregnancy

Baseline Characteristics

  • Please see supplementary appendix for full patient characteristics, including characteristics of patients with and without nodal metastases

Intermediate Thickness Melanoma, Biopsy Group

  • N=770
  • Male sex 58%
  • Age (mean±SD) 52±13.7 years
  • Primary melanoma location
    • Arm or leg 46.5%
    • Other site 53.5%
  • Breslow thickness (mean±SD): 1.98±0.63mm
  • Clark level
    • III 43.5%
    • IV 55.4%
    • V 1.0%
  • Ulceration
    • Present 26.8%
    • Absent 65.3%
    • Unknown 7.9%
  • Nodal metastasis- 16.0% (122/765)
  • Site of first recurrence
    • Nodal 38(4.9%)
    • Distant 107 (13.9%)
    • Local/intratransit 54(7.0%)
  • No recurrence 571 (74.2%)

Intermediate Thickness Melanoma, Observation Group

  • N=500
  • Male sex 55.0%
  • Age (mean±SD) 52.1±13.9 years
  • Primary melanoma location
    • Arm or leg 42.6%
    • Other site 57.4%
  • Breslow thickness (mean±SD): 2.01±0.65mm
  • Clark level
    • III 41.0%
    • IV 57.4%
    • V 1.6%
    • Ulceration
    • Present 29.6%
    • Absent 63.8%
    • Unknown 6.6%
  • Nodal metastasis- 17.4% (87/500)
  • Site of first recurrence
    • Nodal 73 (14.6%)
    • Distant 56(11.2%)
    • Local/intratransit 32(6.4%)
  • No recurrence 339(67.8%)

Thick melanoma, Biopsy Group

  • N =173
  • Male sex 63.6%
  • Age (mean±SD) 53.4±13.9 years
  • Primary melanoma location
    • Arm or leg 42.8%
    • Other site 57.2%
  • Breslow thickness (mean±SD): 5.51±2.17mm
  • Clark level
    • III 22.0%
    • IV 63.0%
    • V 15.0%
  • Ulceration
    • Present 52.0%
    • Absent 41.0%
    • Unknown 6.9%
  • Nodal metastasis- 32.9% (57/173)
  • Site of first recurrence
    • Nodal 15(8.7%)
    • Distant 43(24.9%)
    • Local/intratransit 22(12.7%)
  • No recurrence 93(53.8%)

Thick Melanoma, Observation Group

  • N= 117
  • Male sex 59.0%
  • Age (mean±SD) 56.3±13.3 years
  • Primary melanoma location
    • Arm or leg 41.9%
    • Other site 58.1%
  • Breslow thickness (mean±SD): 5.86±4.16mm
  • Clark level
    • III 26.5%
    • IV 57.3%
    • V 16.2%
    • Ulceration
    • Present 47.8%
    • Absent 46.2%
  • Unknown 6.0%
  • Nodal metastasis- 37.6%(44/117)
  • Site of first recurrence
    • Nodal 40(34.2%)
    • Distant 19(16.2%)
    • Local/intratransit 9(7.7%)
  • No recurrence 49(41.9%)


  • Wide excision margins recommended as 2-3cms for intermediate (1.2 - 3.5mm) and thick melanomas (> 3.5mm)
  • Biopsy Group Wide excision of the primary melanoma with sentinel-node biopsy
    • Immediate completion lymphadenectomy if nodal metastasis discovered on SLNB, defined as gross lymphatic invasion or positive immunohistochemistry
  • Observation Group Wide excision fo the primary melanoma with post-operative nodal surveillance
    • Delayed lymphadenectomy if nodal metastases developed during observation
  • Postoperative surveillance: clinical examination, lab surveillance, chest radiography
    • Years 0-2: q3 months
    • Year 3: q4 months
    • Years 4-5: q6 months
    • Years 6-10: Annual
  • Site specific PET scans, nodal ultrasound, melanoma markers (S-100, LDH)


Comparisons are biopsy group vs. observation group

Primary Outcomes

Melanoma specific survival,
Intermediate thickness melanoma: 81.4±1.5% vs. 78.3±2.0%(HR 0.84 ; 95% 0.64-1.09; P=0.18)
Thick melanoma: 58.9±4.1% vs. 64.4±4.6%(HR 1.12; 95% 0.76-1.67; P=0.56)

Secondary Outcomes

Disease Free Survival
Intermediate thickness melanoma: 71.3±1.8% vs. 64.7±2.3%(HR 0.76 ; 95% 0.62-0.94; P=0.01)
Thick melanoma: 50.7±4.0% vs. 40.5±4.7%(HR 0.70 ; 95% 0.50-0.96; P=0.03)

Comparisons are Biopsy group SLNB positive vs. Biopsy group SLNB negative

Melanoma specific-survival rate
Intermediate thickness melanoma: 62.1±4.8% vs. 85.5±1.5%(HR 3.09 ; 95% 2.12-4.49; P<0.001)
Thick melanoma: 48.0±7.0% vs. 64.6±4.9%(HR 1.75 ; 95% 1.07-2.87; P=0.03)

Incidence of Sentinel Node metastases

Observation Group Surveillance Nodal Metastases
Intermediate thickness melanomas 17.4% (87/500), estimated 10-year cumulative incidence = 19.5% (median 19.2 months, 95% CI 13.6-24.1)
Thick melanoma 37.6% (44/117), estimated 10-year cumulative incidence = 41.4% (median 9.2 months, 95% CI 6.4-12.2)
Biopsy Group Sentinal-Node Metastases
Intermediate-thickness melanoma 16.0% (122/765)
Thick melanoma 32.9%(57/173)
Biopsy Group Surveillance Nodal Metastases Among those Initially Tumor Free
Intermediate-thickness melanoma 4.8% (31/643)
Thick Melanoma 10.3%(12/116)
Estimated 10-year cumulative incidence of nodal metastases in Biopsy Group
Intermediate-thickness melanoma undergoing 21.9%, true incidence = 20.0% (153/765)
Thick melanoma 42.0%, true incidence = 39.9%(69/173)

Survival Comparisons

'Comparisons are Biopsy vs. Observation Group

Melanoma specific survival rate, with Nodal Metastases
Intermediate thickness melanoma: 62.1±4.8% vs. 41.5±5.6% (HR death from melanoma 0.56; 95% CI 0.37-0.84, P=0.006)
Thick melanoma: 48.0±7.0% vs. 45.8±7.8% (HR death from melanoma 0.92; 95% CI 0.53-1.60, P=0.78)
Melanoma specific survival rate, without Nodal Metastases
Intermediate Thickness Melanoma: 88.0±1.4% vs. 86.6±1.8% (HR death from melanoma 0.89; 95% CI 0.61-1.29,P=0.54)
Thick Melanoma: 69.8±5.0% vs. 76.1±5.2% (HR death from melanoma 1.18; 95% CI 0.63-2.18,P=0.61)

Comparisons are Immediate (Biopsy group, Sentinel-Node positive) vs. Delayed (Observation group, nodal recurrence) Lymphadenectomy

Distant disease-free survival
Intermediate Thickness Melanoma: 54.8±4.8% vs. 35.6±5.4%(HR distant metastasis 0.62, 95% CI 0.42-0.91, P=0.0.02)
Thick Melanoma: 45.3±6.9% vs. 43.8±7.7%(HR distant metastasis 0.96, 95% CI 0.56-1.64, P=0.88)

Subgroup Analysis

Latent-subgroup statistical methods to estimate the treatment effect of sentinal lymph node biopsy with immediate lymphadenectomy in the subgroup of patients with nodal metastasis, for intermediate thickness melanoma

Disease free survival
Estimated treatment effect = 1.17 (P<0.001)
Distant disease free survival
Estimate treatment effect = 0.73 (P=0.04)
Melanoma specific survival
Estimated treatment effect = 0.68 (P=0.05)

Adverse Events

Comparisons are Biopsy Group (Sentinel Lymph Node Biopsy, Completion Lymph Node Dissection) vs. Observation Group[4]

Wound separation
3.3%(1.2% SLNB, 3.0% CLND) vs. 3.5%
4.4% (5.5% SLNB, 23.1% CLND) vs. 2.8%
8.3%(4.6% SLNB, 15.8% CLND) vs. 8.4%
Skin Graft Failure
2.2% vs. 1.8%
13.8% (10.1% SLNB, 37.2% CLND) vs. 13.9%
  • Regional complications
    • Leg edema 8
    • Thrombophlebitis 7
  • Systemic Complications
    • Pulmonary 7
    • Urinary 3
    • Other 4
  • Methylene urticaria 0.17% (2/1173)
  • Intermediate thickness cohort increased to 1200 patients because of lower than anticipated event rates


  • Overall concerns of overstating purported therapeutic benefit of completion lymph node dissection, especially in light of subsequent MSLT-II trial failing to demonstrate melanoma-specific survival among patients with sentinel-node metastases with completion lymph-node dissection compared to nodal observation
  • The benefits of improved regional control could in theory be gained by lymphadenectomy at recurrence without costing the patient any survival time, and perhaps delaying or avoiding consequences of immediate lymphadenectomy
  • Post-randomization subgroup analysis between the Biopsy and Observation nodal metastatic groups, reported as a statistically significant difference in intermediate thickness melanomas with disease specific survival, may be methodologically unsound due to confounding factors (non-random comparison groups)
  • Lymph-node involvement definition inconsistently applied between the two arms of the study


  • National Cancer Institute Grant 29605 to Dr. Don Morton
  • Australian and New Zealand Melanoma Trials Groups contributed funding
  • Dr. Thompson discloses payment from advisory board positions and/or travel fees from GlaxoSmithKline, Roche and Provectus
  • Dr. Kashani-Sabet discloses fees for service on advisory boards from Merck and Myriad Genetics, honoraria and grant support from Merck, holding stock in Melanoma Diagnostics, and holding a patent on the molecular classification of melanoma (patent no. 8492102), which has been licensed to Myriad Genetics.

Further Reading

  1. Morton DL et al. Final trial report of sentinel-node biopsy versus nodal observation in melanoma. N. Engl. J. Med. 2014. 370:599-609.
  2. Faries MB et al. Completion Dissection or Observation for Sentinel-Node Metastasis in Melanoma. N. Engl. J. Med. 2017. 376:2211-2222.
  3. Morton DL et al. Sentinel node biopsy for early-stage melanoma: accuracy and morbidity in MSLT-I, an international multicenter trial. Ann. Surg. 2005. 242:302-11; discussion 311-3.
  4. Morton DL et al. Sentinel node biopsy for early-stage melanoma: accuracy and morbidity in MSLT-I, an international multicenter trial. Ann. Surg. 2005. 242:302-11; discussion 311-3.