Abstract
Background/Aim: Melanoma, an aggressive skin cancer, poses a significant threat to patients’ lives, with lower limbs and hips being among the most affected regions. Epidemiology and survival outcomes of patients with melanoma in the lower extremities were investigated and compared to other sites to better understand tumoral behavior and identify predictors of decreased survival. Patients and Methods: The Surveillance, Epidemiology, and End Results (SEER) database was used to search for all skin melanoma cases between 2000 and 2019. Demographic, pathological, and therapeutic factors were compared between affected regions. Overall and disease specific survival were calculated and compared among subgroups. A multivariable analysis was conducted to identify independent prognostic factors. Results: A total of 50,109 patients were diagnosed with melanoma in lower limbs and hips, while 224,121 patients had melanomas in other areas. More women (70.8%) and younger people (mean 55.2 years, SD 16.5) were affected with lower extremities melanoma, with better survival rates than other skin regions. Factors associated with better survival included female sex, younger age, horizontal growth pattern melanomas, and surgery with <1 cm margins. Conclusion: Melanoma affecting lower extremities is commonly diagnosed in young females. Prognosis depends on age, stage at diagnosis, and histologic subtype, but remains better compared to other locations.
With nearly 100,000 new cases projected in the USA in 2023, skin melanoma carries a significant disease burden (1). Lower extremities are considered high-risk sites for melanoma due to high sun exposure (2). Incidence of melanoma in these areas is particularly high among women and young adults (3). Skin melanoma is one of the most common forms of skin cancer in this location (4).
Early detection and adequate treatment are crucial for improving patient outcomes. However, diagnosing lower limbs and hip melanomas can be challenging due to difficult-to-visualize locations (5-8).
Treatment choice depends on tumor stage, with security margin resection being standard for localized melanoma, while sentinel lymph node biopsy is used to detect early regional invasion. For regional or distant disease, adjuvant therapies are often used, with targeted therapy and immunotherapy emerging as promising new treatments (9). Multidisciplinary tumor panels are recommended to improve patient care.
Despite the high prevalence of lower extremity skin melanomas, few studies assessed its epidemiology and survival outcomes. The Surveillance, Epidemiology, and End Results (SEER), a program of the U.S. National Cancer Institute (NCI) that provides annual cancer reports on 26.5% of the U.S. population, was used to analyze the epidemiology and survival of lower extremity and hip melanomas. We aimed to provide comprehensive data to inform healthcare professionals, improving primary prevention strategies allowing early detection, and potentially enhancing patients outcomes (10).
Patients and Methods
Patient selection. Seventeen SEER registries were selected using SEER*Stat software (v.8.4.0.2) (8). Skin melanoma cases were obtained using ICD-O3 diagnostic codes 8720-8790 and skin primary site codes (C44.0-C44.9). Patients were categorized into lower limb and hip skin (C44.7) and other skin locations (C44.0-C44.6, C44.8-C44.9). In situ melanomas were excluded. No institutional review board approval was needed as SEER provides de-identified public data.
Variable selection. Demographic variables (sex, race, age at diagnosis, year of diagnosis), clinicopathological variables (staging, histological subtypes, tumor size, Breslow thickness, mitotic rate, lymph node involvement), treatment patterns (radiotherapy, surgery, chemotherapy), and survival outcomes (survival in months, cause of death) were extracted.
SEER’s summary stage classification, a simplified AJCC derivative, was used because of its consistency over time. Localized cancer (AJCC stages I-II) is limited to the dermis; regional cancer (AJCC stage III) involves adjacent tissues or regional lymph nodes; distant cancer (AJCC stage IV) involves distant organs or lymph nodes invasion.
Surgery was classified into six subgroups, including excision with different margins (<1 cm; 1-2 cm; >2 cm), Mohs surgery, local excision/destruction, no surgery, and unknown.
Breslow Thickness data was documented from 2010 until the cut-off date, with values reported as “less or equal to 0.1 mm” recoded as 0.1 mm and those reported as “9.7 mm and more” recoded as 9.7 mm.
SEER registries recorded chemotherapy and radiotherapy as binary data (“yes”, “no/unknown”). Data on immunotherapy and targeted therapy were unavailable. The study’s cut-off date was December 2019 and survival time is reported in months. For consistency purpose, patients were divided into four age groups identical to the ones established in our previous study: ≤50 years, 51-61 years, 62-71 years, and ≥72 years (11).
Statistical analysis. IBM SPSS version 28 (I.B.M., Armonk, NY, U.S.A.) was used for statistical analysis. Demographic, clinicopathological, and treatment characteristics were compared between melanoma locations (lower extremities and other) using t-tests for continuous variables and chi-square tests for categorical variables after excluding “unknown” values. Overall and melanoma-specific survival analyses for lower limb and hip skin melanoma were modeled with the Kaplan–Meier estimator. The log-rank test was used for univariable analysis of survival outcomes. A multivariable Cox regression model analyzed the relative effect of sex, age at diagnosis, histologic subtypes, stage, treatment sequence, Breslow thickness, ulceration, and mitosis rate on survival. For clarity purposes, only the five most prevalent histologic subtypes were included. Statistical significance was set at a p-value less than 0.005.
Results
Patient characteristics. A total of 274,230 melanoma cases were recorded between 2000 and 2019: 50,109 in lower limbs and hip, and 224,121 in other skin locations. Table I presents the cohort’s demographic and clinicopathologic characteristics.
Demographic, clinicopathological, and treatment characteristics.
Lower limb and hip melanomas were predominantly found in women (35,490; 70.8%) compared to men (14,619; 29.2%) (p<0.005). Melanomas in other body locations were more frequent in men (62.2%). The mean age for lower limb and hip melanoma was 55.2 years (SD 16.5), significantly younger than patients with melanoma in other areas (59.6 years, SD 16.2) (p<0.005). Most lower limb and hip tumors were diagnosed in white patients (92.4%). The number of cases diagnosed increased progressively during the study period.
The mean Breslow thickness of lower limb and hip melanoma was 1.33 mm (SD 1.8), thicker than other skin areas (1.28 mm, SD 1.8) (p<0.005). Ulceration status was reported in 25,507 cases, with ulceration in 3,985 patients (8.0%). Out of 19,391 biopsied lymph nodes, 4,511 (23.3%) were found to be positive. The most represented histologic subtypes were superficial spreading melanomas (17,909; 35.7%), nodular melanomas (3,326; 6.6%), acral lentiginous melanomas (2,177; 4.3%), and lentigo maligna melanomas (741; 1.5%); 49.2% (24,635) were classified as “not otherwise specified” (NOS). Compared to other skin locations, lower limb and hip melanomas had higher proportions of superficial spreading melanoma (35.7% vs. 31.1%) and acral lentiginous melanoma (4.3% vs. 0.3%), but lower proportions of lentigo maligna melanoma (1.5% vs. 7.3%), and desmoplastic melanoma (0.3% vs. 1.2%).
Most cases were diagnosed at localized stage (83.7%), followed by regional stage (10.6%), and distant disease (1.8%). Compared to other skin locations, distant disease was less frequent, and regional disease more frequent. The most common surgical approach was excision with less than one-centimeter margins (42.1%), followed by resection with one to two-centimeters margins (35.1%). Wide resection remained rare (4.2%). Mohs surgery was performed in 1.4% of cases. Local excision/destruction was used in 5,844 patients (11.7%), while 2,417 patients (4.8%) had no surgery. The main reason for not performing surgery was “surgery not recommended” (1,825). Surgery was recommended but not performed in 447 cases.
Overall survival (OS). Mean overall survival (mOS) in patients with melanoma of the lower limbs and hips was 193.9 months (95%CI=193.1-194.8) significantly better than melanoma of other skin locations (173 months; 95%CI=172.6-173.5; p<0.005). Survival rates of lower limb melanoma were 97.4% at 1 year, 91.6% at 3 years, 87.5% at 5 years, and 80.1% at 10 years (Figure 1).
Disease-specific survival according to localization.
Significant differences in mOS were found across age groups for lower limb and hip melanomas (p<0.005), with the highest survival rates in patients aged 50 or less (mOS 224.6 months; 95%CI=223.8-225.4) and the lowest in patients aged 72 or older (mOS 104.6 months; 95%CI=102.5-106.6) (Figure 2).
Disease-specific survival according to age.
Survival was better in female patients (mOS 200.0 months; 95%CI=199.1-201.0) than males (mOS 178.8 months; 95%CI=177.1-180.5; p<0.005) (Supplementary Figure 1).
Superficial spreading melanoma had the best mOS (211.0 months; 95%CI=209.8-212.2), while nodular (mOS 136.4 months; 95%CI=132.5-140.4) and acral lentiginous melanoma (mOS 140.1 months; 95%CI=135.2-145) had the worst (Figure 3).
Disease-specific survival according to histological subtype.
Patients with localized melanoma had better mOS (205.6 months; 95%CI=204.8-206.4) compared to regional stage (127.5 months; 95%CI=124.4-130.6) and distant stage (56.1 months; 95%CI=49.9-62.2; p<0.005) (Supplementary Figure 2).
Surgical resection with <1 cm margins was associated with the best survival (mOS 204.3 months; 95%CI=202.8-205.8), while resection with >2cm margins had the poorest survival (mOS of 153.6 months; 95%CI=149.0-158.3). Mohs surgery had a similar impact on OS, when compared to 1-2 cm margins resection (mOS 174.9 months; 95%CI=168.7-181.1; p=0.527), and local excision/destruction (mOS 189.8 months; 95%CI=187.4-192.1; p=0.092). Patients who had no surgery had an mOS of 168.6 months (95%CI=163.6-173.6) (Supplementary Figure 3).
The use of radiotherapy (mOS 60.4 months; 95%CI=51.8-69) or chemotherapy (mOS 99.2 months; 95%CI=91.5-106.9) was associated with lower OS compared to no/unknown radiotherapy (mOS 195.1 months; 95%CI= 194.3-196) or chemotherapy (mOS 195.4 months; 95%CI= 194.6-196.3; p<0.005) (Supplementary Figure 4 and Supplementary Figure 5).
Disease specific survival. Mean disease specific survival (mDSS) in patients with lower limb and hip melanoma was 217.6 months (95%CI=217-218.2), better than melanoma in other skin locations (211 months; 95%CI=210.6-211.3; p<0.005) (Figure 1B).
Significant differences were found in mDSS across age groups, with the highest survival rates in patients aged 50 or younger (mDSS 228.5 months; 95%CI=227.8-229.2) and the lowest in patients aged 72 or older (mDSS 183.7 months; 95%CI=181.2-186.2; p<0.005) (Supplementary Figure 6). Mean DSS was better in females (mDSS 222.5 months; 95%CI=221.9-223.2) than males (mDSS 205.4 months; 95%CI=204-206.8; p<0.005) (Supplementary Figure 1).
Superficial spreading melanoma (mDSS 230.4 months; 95%CI=229.7-231.1) and lentigo maligna melanoma (mDSS 232.4 months; 95%CI=229.2-235.6) had significantly better mDSS than other subtypes, but mDSS did not differ significantly between both (p=0.259). Patients with nodular melanoma (mDSS 168.2 months; 95%CI=164.3-172.2) and acral lentiginous melanoma (mDSS 177.9 months; 95%CI=173.1-182.8) had the worst DSS (Supplementary Figure 7).
Localized melanoma patients had better mDSS (228.8 months; 95%CI=228.4-229.3) compared to regional stage (151.8 months; 95%CI=148.6-155.1) and distant stage (68.8 months; 95%CI=61.8-75.9; p<0.005) (Supplementary Figure 2).
Surgical resection with a margin <1 cm was associated with the best survival (mDSS 225.3 months; 95%CI=224.3-226.3), while resection with >2 cm margins had the poorest survival (mDSS 179.4 months; 95%CI=175.3-183.6; p<0.005). Mohs surgery impacted OS in a similar fashion than <1 cm margins resection (mDSS 198 months; 95%CI=194-202; p=0.642) and local excision/destruction (mDSS 217.1 months; 95%CI=215.3-218.9; p=0.005). Patients who had no surgery had a mDSS of 197.5 months (95%CI=193.5-201.6) (Supplementary Figure 3).
DSS was lower in patients who received radiotherapy (mDSS 75.6 months; 95%CI=65.7-85.6) and chemotherapy (mDSS 108.4 months; 95%CI=100.3-116.5) compared to no/unknown radiotherapy (mDSS 218.9 months; 95%CI=218.3-219.5) and chemotherapy (mDSS 219.4 months; 95%CI=218.8-220) (Supplementary Figure 4 and Supplementary Figure 5).
Multivariable cox regression of overall survival. Age at diagnosis, sex, histologic subtype, disease stage, type of surgery, adjuvant therapies, Breslow thickness, ulceration, and mitotic rate, were identified as significant prognostic factors in the multivariable analysis (p<0.005; Table II).
Multivariable analysis.
After adjusting for confounding factors, the analysis showed that patients aged 72 years or older had a significantly higher risk mortality risk (HR=13.6; 95%CI=12.7-14.5) compared to those under 50 years. Males had a higher risk of dying from melanoma (HR=1.6; 95%CI=1.5-1.6) compared to females.
Nodular melanoma patients had higher mortality (HR=2.4; 95%CI=2.2-2.5), while superficial spreading melanoma and lentigo maligna melanoma patients were at lower risk (HR=0.67; 95%CI=0.6-0.7 and 0.67; 95%CI=0.6-0.8 respectively).
Excision with <1 cm margins was associated with a lower hazard of death (HR=0.5; 95%CI=0.4-0.5) wide resection (HR=0.6; 95%CI=0.6-0.9) compared to no surgery. Patients receiving radiotherapy or chemotherapy had higher mortality risk (HR=1.7; 95%CI=1.5-1.9 and HR=2.0, 95%CI=1.8-2.3, respectively).
After adjusting for confounding factors, increased Breslow thickness was associated with higher mortality risk. A Breslow thickness of >4 mm had a HR of 8 compared to a Breslow thickness of <0.8 mm (95%CI=7.1-9.0).
Discussion
This study is, to our knowledge, the largest population-based assessment of lower limb and hip melanoma survival outcomes.
Our results show that melanoma in these areas tends to occur at a younger age, predominantly affecting females. A recent report by Shakeel et al. highlighted sex-specific differences based on lower limb subsites. The authors found no significant difference in foot melanomas distribution by sex, but a higher frequency of lower limb melanomas in females, especially in the thighs of young females compared to males (12, 13). This trend could be due to the popularity of tanning and showcasing bare legs on social media, contributing to increased leg melanomas incidence (14-16).
Overall survival and disease-specific survival for lower limb and hip melanomas was better compared to other skin sites, potentially due to their visibility, allowing for earlier detection and treatment (17). In this cohort, 83.7% of the tumors were detected at a localized stage, those are less likely to recur than those on the head, neck, and trunk which is a major factor in determining the prognosis since they can recur as metastases at distant sites (18-20). The distal localization of melanomas is associated with a lower risk of brain metastasis (21). Superficial spreading melanoma, tend to have a less aggressive horizontal growth pattern and were proportionally more frequent in the lower limbs compared to other skin areas, while nodular melanomas were proportionally less common in the lower extremities (22-24).
Despite better survival rates, the mean Breslow thickness was higher in the lower limb cohort. This result should be interpreted cautiously as Breslow thickness was not documented for 49% of the study population.
Females had significantly better OS and DSS than males, potentially due to sex-related biological differences in melanoma behavior (25-29).
Age affected melanoma DSS, and older patients were more susceptible to advanced disease due to delayed diagnosis, and immunosenescence (30).
Superficial spreading melanoma was the most common subtype in lower extremities, with a favorable OS. Nodular and acral lentiginous melanoma had the poorest survival rates, potentially due to vertical growth pattern (17, 31-34). Lentigo maligna melanoma and SSM had similar outcomes in terms of DSS. However, lentigo maligna melanoma cases had a lower OS, possibly due to its prevalence in older individuals with chronic and higher UV exposure, with increased non-melanoma mortality risk.
The preferred surgical approach for melanoma involves removing the tumor and a surrounding zone of healthy tissue, with margin width whose extent is determined by Breslow’s thickness. Adhering to the National Comprehensive Cancer Network’s guidelines improves melanoma-specific survival (35). In this cohort, the proportion of wide margin resections increased with Breslow thickness (Supplementary Table I). The category “No surgery and no margins assessment” was mainly limited to Breslow thickness of less than 0.8 mm or more than 4 mm. However, wide resection surgery was associated with lower OS rates, even after adjusting for various factors than those who had less extensive surgeries. Resection with >2 cm margins, typically reserved for tumors with Breslow thickness >2 mm, is associated with a poorer prognosis. Indeed, while the surgery offers local control it does not impact regional or distant disease if present (36).
Additionally, the use of radiation therapy or chemotherapy was associated with poorer survival rates, consistent with studies suggesting their poor efficacy (37, 38). Patients diagnosed with melanoma at an advanced stage are often the ones who receive such treatments. This could explain the relatively lower survival rates (39, 40).
Our study has limitations, including the lack of detailed therapy data in the SEER database, such as targeted therapies and immunotherapies. Theses therapies play an increasingly important role in managing advanced melanoma and significantly impact patient outcomes (5, 9, 40). Our findings are limited due to lack of data on these treatments, emphasizing the need for future studies with more detailed data. The absence of data related to Breslow’s thickness, ulceration, and mitotic rate before 2010 limits the analysis, as it restricts the application of multivariable and Breslow analysis to only half of the study participants. Moreover, Breslow values over 9.7 mm were reported as 9.7 mm, underestimating the mean Breslow thickness. Another limitation is the unavailability of precise anatomical locations, particularly relevant since foot melanomas are associated with poorer prognosis (41, 42).
Conclusion
Melanoma of the lower limbs and hips generally affects younger patients and predominantly females. Prognosis is primarily influenced by patient age, stage at diagnosis, and histological subtype. However, overall survival and disease-specific survival are better compared to melanomas in other skin locations.
Footnotes
Authors’ Contributions
Conceptualization, S.N.W., and C.M.O.; methodology, S.N.W., M.S, J.M., and C.M.O.; validation, D.F.K. and C.M.O.; formal analysis; S.N.W., M.S, J.M., S.M., and C.M.O.; investigation, S.N.W.; data curation, S.N.W., M.S, and J.M.; writing – original draft preparation; S.N.W.; writing – review and editing, J.M., M.S., D.F.K., S.M., C.M.O.; supervision, D.F.K. and C.M.O. All Authors have read and agreed to the published version of the manuscript.
Supplementary Material
Available at: https://data.mendeley.com/datasets/23746xpmmr/2
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
- Received June 30, 2023.
- Revision received July 26, 2023.
- Accepted July 27, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.