58.19 Utility of Brain MRI for Patients with Malignant Melanoma and a Positive Sentinel Node Biopsy

C. S. Boutros1, H. Kakish1, A. W. Loftus1, L. D. Rothermel1, R. S. Hoehn1  1University Hospitals Cleveland Medical Center, Surgery, Cleveland, OH, USA

Introduction: Due to inconsistent data to guide metastatic work up, the NCCN considers baseline staging (whole body PET scan with or without brain MRI) for asymptomatic patients found to have a positive sentinel lymph node biopsy (SLNB). We aim to investigate the rates of metastases, specifically brain metastases, in this patient population, as well as explore positive predictors of metastases to guide necessity of further imaging.

Methods:  We created a cohort using the National Cancer Database (2012-2020) to mimic a common clinical scenario: patients with invasive malignant melanoma who were clinically node negative, underwent excision with SLNB and had a positive sentinel lymph node(s). Patients were categorized into three groups: no metastases, any metastases, or brain metastases. Demographic and clinical characteristics were compared between the groups using univariable analysis. Multivariable regression was used to assess variables associated with the presence of any metastasis.

Results: Of 10,170 patients who met the inclusion criteria 132 (1.30%) had any metastases. Of patients with metastatic disease, 39 (29.5%) patients had brain metastases and 26 (19.6%) had brain-only metastases. Higher rates of overall distant metastasis were seen in head and neck tumors (p-value= 0.002), nodular and lentigo maligna subtypes (p <0.001), higher Breslow depth (p-value <0.001), ulceration (p <0.001), and lymphovascular invasion (p <0.001). Lentigo Maligna subtype (OR: 3.29, 95% CI: 1.25 – 8.63), Breslow depth > 4 mm (3.39, 1.29 – 8.90), and lymphovascular invasion (2.19,1.45 – 3.30) were independently associated with metastasis on multivariable logistic regression. Table 1.  

Conclusion: This study showed that distant metastasis is rare on baseline clinical workup in clinically node-negative melanoma patients who are found to have a positive SLNB, occurring at a rate of only 1.3%. As such, costly and intrusive metastatic workups may not be appropriate for all patients. Factors such as tumor subtype, Breslow depth, and lymphovascular invasion may be considered to guide more selective use of whole-body PET and brain MRI for these patients.