The Relationship Between Melanoma Subtype and Organ-Specific Metastases: An Observational Study

Melanoma rates in the U.S. have more than doubled over the last 30 years, now making up 5% of all new cancers1. The 4 main subtypes are superficial spreading melanoma (SSM), nodular melanoma (NM), lentigo maligna melanoma (LMM), and acral lentiginous melanoma (ALM), each with different behaviors. Five-year relative survival rates vary: SSM at 99.5%, NM at 74.6%, LMM at 100%, and ALM at 81%2.

Many factors affect risk of metastasis including Breslow thickness, ulceration, and mitotic rate. Sentinel lymph node status is a crucial prognostic assessment. Despite the many factors identified to aid in risk stratification, data remain limited regarding whether cutaneous melanoma subtype is associated with specific sites of metastasis. This data has been elucidated for ocular melanoma, with the liver being the predominant site of metastasis3.

Accordingly, our retrospective study provides a descriptive analysis of the distribution of metastasis sites across cutaneous melanoma subtypes and organ specific metastases in a single institution, with consideration of Breslow depth and ulceration status. These findings are exploratory and intended to generate hypotheses rather than establish predictive associations.

A retrospective observational study was conducted using data from the University of Arizona’s (UA) melanoma database from January 2019 to August 2022. Approved by the University of Arizona Institutional Review Board (Study #00001667), data was collected via REDCap, focusing on melanoma staging, subtype, thickness, ulceration, mitoses, sentinel lymph node metastases, and distant metastases. Eligibility criteria included patients 18 years or older, diagnosed with melanoma, who developed metastases within the study period and received treatment at the UA Cutaneous Oncology Program Clinic in Tucson, Arizona. Exclusions were patients under 18, diagnosed with melanoma in situ or mucosal/ocular melanoma, or having multiple primary invasive melanomas.

Statistical analysis was performed using R Core Team (2022) software (Version 4.2.2; R Foundation for Statistical Computing, Vienna, Austria), and the data was visualized using the ggplot2 package4. Each patient was considered the unit of analysis. If a patient developed multiple metastatic sites, all locations were recorded. For visualization, each metastatic site is displayed, but proportions in the text are reported at the per-patient level.

Of 1,000 patients screened, 32 met inclusion criteria: 56.3% male, 43.8% female, ages 23–87 (median 63.5). Most (97%) were White. The cohort included 32 patients, each serving as the unit of analysis. Six of these patients each had a primary melanoma with multiple, distinct sites of metastasis. Among these, 2 patients had 4 metastatic sites, 2 patients had 3, and 3 patients had 2. The remaining 25 patients had 1 metastatic site each, yielding a total of 45 metastatic sites presented in Figs. 1B and 2. Of the 32 patients, their primary melanoma subtypes included 11 SSM, 17 NM, 3 ALM, and 1 LMM. Out of all the other subtypes, NM patients were more frequently observed with metastases beyond lymph nodes, with 7 of 17 patients (41.2%) having visceral involvement, most commonly in the lung (4 patients, 23.5%) and liver (4 patients, 23.5%). Nodal metastases were observed in patients across all subtypes, with only one distant non-nodal metastasis outside the NM group, occurring in an ALM patient with right groin involvement.

Fig. 1
Melanoma subtypes and location of metastasis. (A) Summary of metastatic patterns across melanoma subtypes. Summary of patient counts, total metastatic sites, and most common metastatic sites by melanoma subtype in 32 patients. NM had the highest metastatic burden, frequently involving lymph nodes, lung, liver, and brain. Of the 6 patients with metastases to multiple sites—2 patients had 4 metastatic sites, 2 patients had 3, and 3 patients had 2. The remaining 25 patients had 1 metastatic site each, yielding a total of 45 metastatic sites reported in the figures below. (B) The relationship between melanoma subtype and metastasis location, stratified by Breslow depth only. Heatmap showing the relationship between melanoma subtype and metastasis location, stratified by Breslow depth. The color intensity represents the number of cases, with darker shading indicating higher counts. NM demonstrated the broadest distribution of metastasis sites, while SSM was most frequently associated with lymph node metastases.
SSM: superficial spreading melanoma, NM: nodular melanoma, LMM: lentigo maligna melanoma, ALM: acral lentiginous melanoma.

• Click for larger image
• Download as PowerPoint slide

Fig. 2
The relationship between melanoma subtype and metastasis location, stratified by Breslow depth and ulceration status. Bubble plot depicting the relationship between melanoma subtype, metastasis location, Breslow depth, and ulceration status. Each point represents a metastasis site for a given subtype. Circle markers indicate ulcerated tumors and triangle markers indicate non-ulcerated tumors. The size of each marker corresponds to the number of cases, and the color gradient represents the mean Breslow depth (mm), with darker shading indicating greater depth.

• Click for larger image
• Download as PowerPoint slide

Variation in metastatic patterns was noted among subtypes. All patients with multiple distinct metastatic sites had primary melanomas classified as NM. SSM and ALM lesions ≤0.9 mm Breslow depth were observed only in lymph nodes, whereas NMs of similar depth were observed in both lymph nodes and visceral organs. The liver and lungs were the most frequently observed sites of distant metastasis in NM, and ulcerated lesions were observed with higher rates of distant spread (Fig. 1). Only one distant metastasis occurred in non-nodular subtypes, limiting conclusions about organ-specific patterns for these groups. Although several NM cases involved lung and liver metastases, the small sample size means these findings are descriptive and hypothesis-generating rather than definitive.

Overall, this exploratory study suggests a possible tendency for NM to metastasize to the liver and lung, but the limited dataset and small sample size preclude firm conclusions. Studies with larger sample size and extended follow-up are needed to further elucidate these preliminary observations and to better assess organ-specific metastases across all primary cutaneous melanoma subtypes. Future investigations in this area would aid in refining surveillance strategies and guiding patient care.