Sentinel lymph node biopsy: an alternate view

Validation of statistical predictive models meant to select melanoma patients for sentinel lymph node biopsy

INTRODUCTION

To identify melanoma patients at sufficiently low risk of nodal metastases who could avoid sentinel lymph node biopsy (SLNB), several statistical models have been proposed based upon patient/tumor characteristics, including logistic regression, classification trees, random forests, and support vector machines. We sought to validate recently published models meant to predict sentinel node status.

METHODS

We queried our comprehensive, prospectively collected melanoma database for consecutive melanoma patients undergoing SLNB. Prediction values were estimated based upon four published models, calculating the same reported metrics: negative predictive value (NPV), rate of negative predictions (RNP), and false-negative rate (FNR).

RESULTS

Logistic regression performed comparably with our data when considering NPV (89.4 versus 93.6%); however, the model's specificity was not high enough to significantly reduce the rate of biopsies (SLN reduction rate of 2.9%). When applied to our data, the classification tree produced NPV and reduction in biopsy rates that were lower (87.7 versus 94.1 and 29.8 versus 14.3, respectively). Two published models could not be applied to our data due to model complexity and the use of proprietary software.

CONCLUSIONS

Published models meant to reduce the SLNB rate among patients with melanoma either underperformed when applied to our larger dataset, or could not be validated. Differences in selection criteria and histopathologic interpretation likely resulted in underperformance. Statistical predictive models must be developed in a clinically applicable manner to allow for both validation and ultimately clinical utility.

Sentinel lymph node biopsy in melanoma: Facts and controversies

Three decades after its introduction in the 1990s, the sentinel lymph node biopsy for patients with localized cutaneous melanoma is still the subject of great debate in dermatology. Many questions remain unanswered, and studies currently in progress may or may not bring us any closer to determining the truth about sentinel lymph node biopsy and melanoma. We discuss the effect of sentinel lymph node biopsy on overall survival, the clinical and therapeutic implications of sentinel lymph node biopsy, and the melanoma patients who might be candidates for sentinel lymph node biopsy.

Prediction of melanoma metastasis by the Shields index based on lymphatic vessel density

Background

Melanoma usually presents as an initial skin lesion without evidence of metastasis. A significant proportion of patients develop subsequent local, regional or distant metastasis, sometimes many years after the initial lesion was removed. The current most effective staging method to identify early regional metastasis is sentinel lymph node biopsy (SLNB), which is invasive, not without morbidity and, while improving staging, may not improve overall survival. Lymphatic density, Breslow's thickness and the presence or absence of lymphatic invasion combined has been proposed to be a prognostic index of metastasis, by Shields et al in a patient group.

Methods

Here we undertook a retrospective analysis of 102 malignant melanomas from patients with more than five years follow-up to evaluate the Shields' index and compare with existing indicators.

Results

The Shields' index accurately predicted outcome in 90% of patients with metastases and 84% without metastases. For these, the Shields index was more predictive than thickness or lymphatic density. Alternate lymphatic measurement (hot spot analysis) was also effective when combined into the Shields index in a cohort of 24 patients.

Conclusions

These results show the Shields index, a non-invasive analysis based on immunohistochemistry of lymphatics surrounding primary lesions that can accurately predict outcome, is a simple, useful prognostic tool in malignant melanoma.