SEARCH

SEARCH BY CITATION

Keywords:

  • melanoma;
  • sentinel lymph node biopsy;
  • pediatric;
  • young adult;
  • diagnostic

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

BACKGROUND:

Data on sentinel lymph node (SLN) biopsy in children with melanoma are limited. In this study, the authors compared the factors associated with SLN biopsy use and metastases in pediatric and young adult patients with melanoma.

METHODS:

The 2008 Surveillance, Epidemiology, and End Results (SEER) databases were used to examine melanoma cases from 2003 to 2008. Data extracted include age, sex, race, stage, tumor thickness, ulceration, lymph node status, surgical treatment, and survival. Logistic regression models were used for adjusted analyses.

RESULTS:

In total, 717 children (age <20 years) and 1368 young adults (age 20-24 years) were identified who were diagnosed with melanoma. Factors that were associated with SLN biopsy use included tumor ulceration (odds ratio [OR], 2.4; 95% confidence interval [CI], 1.4-4.3) and greater thickness (OR, 17; 95% CI, 12-24 for >1 mm vs ≤1 mm), but not younger age (OR, 1.3; 95% CI, 0.94-1.8) in adjusted analyses. SLN metastasis was correlated with ulceration (OR, 3.0; 95% CI, 1.6-5.8), increased thickness (OR, 6.8; 95% CI, 3.1-15 for 2.01-4.0 mm vs ≤1 mm), and for the interaction between age <20 years and thickness 1.01 to 2.00 mm (OR, 6.5; 95% CI, 1.7-25) in adjusted analyses. Children with nonulcerated melanomas that measured 1.01 to 2.00 mm in thickness were significantly more likely to have SLN metastases than young adults (24% vs 4%; P < .001).

CONCLUSIONS:

Thickness and ulceration were strong predictors of both the use of SLN biopsy and positive SLN biopsy results in children and young adults with melanoma. Compared with young adults, children were more likely to have SLN metastases despite having similar rates of SLN biopsy use. Cancer 2011;. © 2011 American Cancer Society.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

Only 2% to 4% of melanoma cases are diagnosed in individuals aged <20 years.1-3 In 2006, there were 416 recorded cases of melanoma in pediatric patients (aged <20 years) in the United States.4 Because pediatric cases are relatively rare and have benign mimics, it is estimated that 40% have delays in diagnosis and treatment.5-7

Since 1998, sentinel lymph node (SLN) biopsy has been supported increasingly as the standard for detecting regional lymph node metastasis and predicting prognosis in patients with melanoma ≥1 mm in thickness.8-10 SLN metastasis warrants consideration of completion lymph node dissection (CLND) and systemic adjuvant therapy.11-14

The National Comprehensive Cancer Network recommends SLN biopsy for clinically lymph node-negative patients who have melanomas that measure ≥1 mm in thickness or who have high-risk pathologic characteristics like ulceration, Clark level IV invasion, or a high mitotic rate.9 These guidelines were established based on adult data and have not been evaluated for children. Pediatric and adult melanomas can have different presentations, biology, and prognoses.5, 15 Most studies that have evaluated SLN biopsy in children with melanoma have been limited to small series or single-institution studies.16-22 A report from the National Cancer Data Base indicated that 70% of pediatric patients with melanoma underwent SLN biopsy, and its use was higher among patients who had insurance and those at Children's Oncology Group hospitals.23 To date, no studies have compared SLN biopsy use and results in pediatric cases with use and results in adult cases on a population level, an important comparison to understand age-related differences in melanoma characteristics.

For the current study, we analyzed the use of SLN biopsy in children and young adults with melanoma using data from the Surveillance, Epidemiology, and End Results (SEER) databases. Young adults between ages 20 and 24 years were chosen for the comparison group, because they are most similar to children but present with more adult melanoma characteristics. It is known that SLN use and results vary with age, and choosing this subset of young adults limits the heterogeneity of the control group.24 Factors associated with 1) undergoing SLN biopsy and 2) having SLN metastases were examined in pediatric and young adult patients.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

Database

The SEER databases sponsored by the National Cancer Institute contain cancer prevalence, incidence, and survival data from 17 US population-based tumor registries.25 The databases include an estimated 26% of the United States population with registries in the states of Connecticut, Hawaii, Iowa, Kentucky, Louisiana, New Jersey, New Mexico, Utah, Georgia (Atlanta and rural Georgia), Michigan, Washington, California (San Francisco-Oakland, San Jose-Monterey, Greater California, Los Angeles), and Alaska. We used the SEER 17 databases from 2003 to 2008, the years in which coding for SLN biopsy was most recently standardized.

Case Identification

We used SEER*Stat Version 6.6.2 (National Cancer Institute, Bethesda, Md) to identify children and adolescents who were diagnosed with invasive melanoma before age 20 years recorded according to International Classification of Diseases for Oncology, Third Edition melanoma codes (M8720-8790).26 Young adults diagnosed with invasive melanoma between ages 20 and 24 years were selected as a comparison group.27 The cancer registrars only code lesions as melanoma when melanoma is indicated as the designated diagnosis by the pathologist; Melanocytic tumors of uncertain malignant potential, atypical Spitz nevi, and severely atypical melanocytic tumors are not coded as melanoma in the SEER databases and, thus, are not part of this study.

Data extracted on each patient included age, sex, year of diagnosis, SEER registry, SEER stage (localized, regional, or distant), tumor thickness, ulceration, lymph node status, first primary, location of primary, histology, race (white, black, other, or unknown), survival in months, vital status, scope of regional lymph node surgery, and county attributes (median family income, percentage with less than a high school education, and percentage below the poverty line).

Lymph Node Surgery

The scope of regional lymph node surgery code (2003+) records procedures performed during initial evaluation of the melanoma. SLN biopsies fell under 3 headings: SLN biopsy only, SLN biopsy and CLND at the same time or at an unspecified time, and SLN biopsy and CLND at different times. Lymph node surgery results were coded under the regional lymph node-positive variable (1988+). Data from 1998 to 2002 were excluded from our analysis, because CLNDs replaced records of SLN biopsies.

Analysis

Extracted patients and associated data were exported to Intercooled Stata 11.0 for Mac (Stata Corp, College Station, Tex) for all further analysis. We limited our analysis of lymph node surgery to patients who had clinically localized melanoma at diagnosis by excluding those who had clinically palpable lymph nodes or metastatic disease. Baseline characteristics were compared by using chi-square tests or Fisher exact test for categorical variables, and Student t tests or Wilcoxon rank-sum tests were used for continuous variables. We evaluated associations between undergoing lymph node surgery and demographic and tumor characteristics in unadjusted and adjusted logistic regression models. Similar analyses were performed to identify associations with a positive SLN biopsy result.

We used Kaplan-Meier survival estimates to evaluate the association between SLN biopsy result and prognosis by age group. All causes of death were included in the analysis, because few competing comorbidities exist in this young population. The survival functions were compared with the log-rank test for equality of survival functions. All tests were 2-tailed, and P values < .05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

There were 717 cases of melanoma in patients aged <20 years at diagnosis and 1368 cases in patients ages 20 to 24 years at diagnosis recorded in the SEER databases from 2003 to 2008. The median follow-up for all patients was 34 months.

Characteristics at Diagnosis

Significant differences in baseline characteristics were observed among young children (aged <10 years), adolescents (ages 10 to 19 years) and young adults (ages 20 to 24 years) (Table 1). Young children with melanoma more often were male (47%) and nonwhite (9%) compared with adolescents (41% male, 2% nonwhite) and young adults (31% male, 1.4% nonwhite). Melanoma in young children more frequently presented with distant metastases (6%), nodular morphology (10%), and thickness >1 mm (40%) compared with tumors in adolescents (1% metastatic, 5% nodular, and 28% >1 mm thick) and young adults (2% metastatic, 6% nodular, and 22% ≥1 mm thick). Notably, the proportion with ulceration did not differ significantly among these age groups.

Table 1. Characteristics at Diagnosis of Young Children (Aged <10 years), Adolescents (Ages 10-19 years), and Young Adults (Ages 20-24 years) With Invasive Melanoma: 2003 to 2008
 No. of Patients (%) 
CharacteristicAged <10 y, n = 79Ages 10-19 y, n = 638Ages 20-24 y, n = 1368Pa
  • Abbreviation: IQR, interquartile range.

  • a

    Groups were compared by using chi-square tests except for continuous variables.

  • b

    P < .001 in a comparison with adolescents (ages 10-19 years) using a 2-sample Wilcoxon rank-sum test.

  • cP < .001 in a comparison with adults (ages 20-24 years) using a 2-sample Wilcoxon rank-sum test.

Extent of disease   < .001
 Localized49 (62)512 (80)1186 (87) 
 Regional19 (24)93 (15)113 (8) 
 Distant5 (6)8 (1)23 (2) 
 Unstaged6 (8)25 (4)46 (3) 
Sex   < .001
 Male37 (47)260 (41)426 (31) 
 Female42 (53)378 (59)942 (69) 
Race/ethnicity   < .001
 White69 (87)592 (93)1249 (91) 
 Black1 (1)6 (1)5 (0.4) 
 Other6 (8)9 (1)18 (1) 
 Unknown3 (4)31 (5)96 (7) 
Histology   .002
 Superficial spreading10 (13)222 (35)512 (37) 
 Nodular8 (10)30 (5)75 (6) 
 Other13 (16)53 (8)88 (6) 
 Not specified48 (61)333 (52)693 (51) 
Thickness, mm   < .001
 Median [IQR]1.4 [0.40-4.8]a0.6 [0.40-1.3]b0.5 [0.35-1.0] 
 <1.0127 (34)390 (61)956 (70) 
 1.01-2.09 (11)91 (14)178 (13) 
 2.01-4.07 (9)57 (9)80 (6) 
 >4.016 (20)34 (5)37 (3) 
 Unknown20 (25)66 (10)117 (8) 
Ulceration   .27
 Absent45 (57)434 (68)920 (67) 
 Present9 (12)40 (6)100 (7) 
 Unknown25 (31)164 (26)348 (25) 
Primary site   < .001
 Extremities42 (53)240 (38)599 (44) 
 Torso12 (15)258 (40)542 (40) 
 Face, head, and neck22 (28)131 (21)188 (14) 
Other3 (4)9 (1)39 (3) 

Factors Associated With Undergoing Sentinel Lymph Node Biopsy

More children (40%) than young adults (31%) who had localized disease on examination underwent SLN biopsy (Table 2). Of those who met current recommendations for SLN biopsy (>1 mm in thickness or ulceration), 74% of children and 70% of young adults underwent SLN biopsy. In unadjusted analyses for both pediatric and young adult patients, ulceration, histology, thickness, sex, and primary site were correlated significantly with undergoing SLN biopsy. Adjusted analyses also revealed that ulceration and thickness were correlated significantly with undergoing SLN biopsy; whereas age, histology, sex, and primary site were not (Table 3). In adjusted analyses, patients who had melanoma >1 mm in thickness had an odds ratio (OR) of 17 (95% confidence interval [CI], 12-24) of undergoing SLN biopsy compared with those who had melanoma ≤1 mm in thickness.

Table 2. Factors Associated With Undergoing Sentinel Lymph Node Biopsy in Children and Young Adults Diagnosed With Clinically Localized, Malignant Melanoma: 2003 to 2008a
 No. of Patients (%) 
VariableAged <20 y, n = 629Ages 20-24 y, n = 1230Pb
  • Abbreviations: IQR, interquartile range; SLNBs, sentinel lymph node biopsies.

  • a

    Excludes patients who had distant metastases (stage IV) or clinically positive lymph nodes and those who underwent an unknown lymph node procedure or completion lymph node dissection.

  • b

    Groups were compared using chi-square tests except for continuous variables.

  • c

    P < .001 compared with adults (ages 20-24 years) in a 2-sample Wilcoxon rank-sum test.

No. of SLNBs (% of total)251 (40)380 (31) 
Ulceration  < .001
 Absent170 (39)249 (29) 
 Present28 (76)52 (70) 
 Unknown53 (36)79 (26) 
Histology  < .001
 Superficial spreading63 (29)128 (26) 
 Nodular24 (75)39 (68) 
 Other35 (64)33 (45) 
 Not specified129 (40)180 (29) 
Thickness, mm  .001
Median [IQR]1.3 [0.88-2.5]c1.0 [0.70-1.7] 
 <1.0184 (22)179 (20) 
 1.01-2.073 (88)115 (76) 
 2.01-4.046 (94)47 (85) 
 >4.030 (83)20 (74) 
 Unknown18 (25)19 (21) 
Sex  .005
 Male113 (44)134 (37) 
 Female138 (37)246 (28) 
Primary site  .003
 Extremities112 (45)186 (34) 
 Torso80 (33)142 (28) 
 Face, head, neck59 (44)51 (31) 
 Other0 (0)1 (5) 
Table 3. Odds Ratio of Undergoing a Sentinel Lymph Node Biopsy in Patients Diagnosed With Malignant Melanoma: 2003 to 2008
 Univariate Analysis Multivariate Analysis 
VariableOR95% CIPOR95% CIPa
  • Abbreviations: CI, confidence interval; OR, odds ratio.

  • a

    The multivariate model included age, ulceration, thickness, and sex. Race, primary site, and registry were not significantly correlated.

  • b

    The OR was 17.0 (95% CI, 12.0-24.0; P < .001) for thickness ≤1 mm vs >1 mm.

Age <20 y1.51.2-1.8< .0011.20.94-1.6.15
Ulceration5.43.5-8.2< .0012.41.4-4.2.002
Thickness, mmb      
 1.01-2.016.011.0-23.0< .00116.011.0-23.0< .001
 2.01-4.033.018.0-63.0< .00118.09.4-36.0< .001
 >4.015.015.0-8.0< .00124.09.4-63.0 
Males1.51.2-1.8< .0011.20.9-1.6.16

Factors Associated With a Positive Sentinel Lymph Node Biopsy

Of the patients who underwent SLN biopsy, 25% of children and 14% of young adults had SLN metastases (Table 4). Pediatric characteristics that were associated with SLN metastases included increasing thickness, ulceration (62%), and nodular histology (43%). It is noteworthy that the children who had melanoma thickness between 1.01 mm and 2.00 mm had SLN metastases more often than young adults (24% vs 4%) in univariate analysis. Overall, children were more likely than young adults to have SLN metastases for all analyzed characteristics (ulceration, histology, thickness, sex, and primary site). Adjusted analysis indicated that ulceration, thickness, and an interaction between age <20 years and thickness from 1.01 mm to 2.00 mm independently contributed to the probability of having SLN metastases; whereas histology, sex, and primary site did not (Table 5). Patients with melanomas that were >1 mm thick had an OR of 4.9 (95% CI, 2.5-9.6) of having SLN metastases compared with those who had melanomas that were ≤1 mm thick in adjusted analyses. There was no significant interaction between age and ulceration (OR, 0.85; 95% CI, 0.24-2.9). Socioeconomic status variables, including percentage below the poverty line, median income, and percentage graduating high school by census region, were not associated with SLN biopsy or metastases.

Table 4. Factors Associated with a Positive Sentinel Lymph Node Biopsy in Patients Diagnosed With Malignant Melanoma: 2003 to 2008
 No. of Patients (%) 
VariableAged <20 y, n = 244Ages 20-24 y, n = 376Pa
  • Abbreviations: IQR, interquartile range; NOS, not otherwise specified; SLNs, sentinel lymph nodes.

  • a

    Groups were compared by using chi-square tests except for continuous variables.

  • b

    P < .001 in a comparison with adults (ages 20-24 years) using a 2-sample Wilcoxon rank-sum test.

No. of positive SLNs (% of total)62 (25)54 (14) 
Ulceration  < .001
 Absent36 (22)22 (9) 
 Present16 (62)21 (40) 
 Unknown10 (19)11 (14) 
Histology  < .001
 Superficial spreading9 (15)12 (10) 
 Nodular10 (43)15 (38) 
 Other9 (27)8 (24) 
 NOS34 (26)19 (11) 
Thickness, mm  < .001
 Median [IQR]2.5 [1.7-4.8]b2.5 [1.0-4.1] 
 <1.014 (5)13 (7) 
 1.01-2.016 (24)7 (4) 
 2.01-4.020 (43)18 (38) 
 >4.017 (59)13 (65) 
 Unknown5 (28)3 (16) 
Sex  .37
 Male33 (25)32 (13) 
 Female29 (26)22 (17) 
Primary site  .06
 Extremities31 (28)29 (16) 
 Torso16 (21)16 (11) 
 Face, head, neck15 (26)8 (16) 
Table 5. Odds Ratio of a Positive Sentinel Lymph Node Biopsy in Patients Diagnosed With Malignant Melanoma: 2003 to 2008
 Univariate Analysis Multivariate Analysisa 
VariableOR95% CIPOR95% CIP
  • Abbreviations: CI, confidence interval; OR, odds ratio.

  • a

    The multivariate model included age, ulceration, and thickness and an interaction term (age <20 years × thickness 1.01-2 mm). Sex, race, primary site, and registry were not significantly correlated.

  • b

    The OR was 4.9 (95% CI, 2.5-9.6; P < .0001) for thickness ≤1 mm vs >1 mm in the multivariate model.

Age <20 y2.01.4-3.1.0011.10.6-2.3.69
Ulceration5.53.3-9.3< .0013.01.6-5.8.001
Thickness, mmb      
 1.01-2.02.01.0-3.9.040.70.2-2.4.60
 2.01-4.09.75.1-18< .0016.83.1-15.0< .001
 <4.02210-47< .001208.1-48.0< .001
Interaction term: Age <20 y × thickness 1.01-2.0 mm6.51.7-25.0.006

Thickness Subset Analyses

To explore the differences observed in SLN metastases between young adults and children, we evaluated subsets of patients according to thickness and ulceration. Of the patients who had tumor thickness <1.01 mm and no ulceration, 21% of pediatric patients and 20% of young adult patients underwent SLN biopsy (P = .8). Among these, 5.1% of children (3 of 59 patients) and 5.3% of young adults (7 of 132 patients) had SLN metastases (P = .9).

Of the patients who had nonulcerated melanoma 1.01 mm to 2.00 mm in thickness, 88% of pediatric patients and 76% of young adult patients underwent SLN biopsy (P = .12). It is noteworthy that 24% of children (14 of 58 patients) and 4% of young adults (3 of 79 patients) had SLN metastases (P < .001).

Survival Analysis

Kaplan-Meier estimates of survival differed significantly between those with and without SLN metastases in both pediatric and young adult melanoma patients who underwent SLN biopsy (P < .001) (Fig. 1). Survival was not statistically significantly different (P = .07) between lymph node-positive adults and lymph node-positive children (5-year survival rate 87% [95% CI, 68%-95%] vs 89% [95% CI, 74%-96%], respectively). A positive SLN biopsy was associated with poorer prognosis in children (P = .003) and young adults (P = .001). Of those patients who had SLN metastases, 4 of 64 children and 4 of 54 young adults died. Of those patients who had negative SLN biopsies, 1 of 182 children and 3 of 322 young adults died.

thumbnail image

Figure 1. Mortality is illustrated according to sentinel lymph node biopsy (SLNB) results in children and young adults who were diagnosed with melanoma diagnosed 2003 to 2008. Neg indicates negative; pos, positive.

Download figure to PowerPoint

Sentinel Lymph Node Biopsy and Completion Lymph Node Dissection

Among the patients who had negative SLN biopsies, 16% of pediatric patients and 12% of adult patients went on to undergo CLND. Meanwhile, 78% of pediatric patients and 73% of young adult patients who had positive SLN biopsies underwent subsequent CLND.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

SLN biopsies have been used for nearly 2 decades to detect lymph node metastases from melanoma and have been in widespread use for over a decade. The objective of the current study was to determine differences in the use and results of SLN biopsies between children and young adults and reveal the age-related differences in melanoma characteristics. We identified tumor thickness and ulceration as strong predictors of both the receipt of SLN biopsy and the detection of metastatic disease in children and young adults with melanoma. An interaction between younger age and thickness of 1.01 mm to 2.00 mm was associated independently with a greater likelihood of SLN metastases but not with the receipt of SLN biopsy in multivariable analysis. In addition, survival analyses indicate that a positive SLN biopsy is a poor prognostic factor in pediatric and young adult patients with melanoma. We observed that ulceration was associated significantly with SLN metastases in children and adults. Of the patients who had melanoma <1 mm thick without ulceration, only 5% of both children and young adults had a positive SLN biopsy. Our findings support the National Comprehensive Cancer Network recommendation for SLN biopsy in children and young adults who have melanoma with a thickness ≥1 mm or with adverse histologic features.

It is worth noting that younger age was not associated independently with greater SLN biopsy use in our multivariable analysis. The more frequent use of SLN biopsy in younger patients observed in the univariate analysis may be related to greater median tumor thickness in pediatric melanoma. A previous report from the SEER database also indicated that children present with more advanced melanomas, but that report did not address predictors of SLN use.27

After adjusting for ulceration, children who had melanomas with tumor thickness from 1.01 mm to 2.00 mm were more than 6 times as likely to have SLN metastases as young adults. Similar trends have been reported for melanoma overall in children from single institutions and in a meta-analysis of SLN in children with melanoma.28, 29 However, the rate of SLN metastases did not differ significantly between children and young adults with low-risk melanoma (<1 mm with no ulceration); approximately 5% of both subsets had SLN metastases. The greater percentage of SLN metastases in the children who had an SLN appears to be largely because of patients with nonulcerated melanomas 1 mm to 2 mm in thickness; among these patients, children were much more likely to have lymph node metastases than young adults (24% vs 4%). This finding supports the idea that some melanomas in children may differ biologically from melanomas in adults and should encourage future efforts to investigate the biologic heterogeneity of melanoma.

The results from our study indicated that pathologically positive SLN biopsies are associated with decreased survival in children. The 5-year survival rate of 87% in this study for those with SLN metastasis was higher than the 79% for pediatric patients with SLN metastases reported from the Sydney Melanoma Unit.29 However, the small numbers of deaths from 2003 to 2008 and the short average follow-up of 35 months limits the precision of our survival estimates at 5 years, with 95% confidence limits as low as 68%.

CLND has been and remains part of guideline recommendations for patients who have positive SLN biopsy results.30 In the patients analyzed here, the records indicated that 22% of children and 27% of young adults with SLN metastases did not undergo CLND; the reasons for this are unknown. These numbers may be because of patient or physician preference or because of coding errors, but SEER remains the gold standard for population-based cancer registries. The records also indicated that 16% of children and 12% of young adults in our study who had negative SLN biopsy results did undergo CLND. According to the staging system that was in place during the years that we studied, SLN biopsies that were negative according to hematoxylin and eosin staining but positive according immunohistochemistry were coded as “N0.” It is possible that some of the cases of apparent over treatment were cases in which a positive SLN identified by immunohistochemistry alone was considered clinically significant. In addition, misclassification, variation in practice, and unknown factors also may have affected negative SLN biopsies that were followed by CLND.

In comparing the use of SLN biopsy in children and young adults, the SEER databases provide uniformity of collection and population-based data representative of melanoma cases across the United States. However, limitations related to the use of the SEER databases from 2003 to 2008 include lack of a centralized review of diagnostic specimens with a large proportion of cases classified as other or not otherwise specified histology, absence of information regarding mitotic rate (because it was not part of the staging system in those years), and lack of information about patient-specific socioeconomic factors. Socioeconomic factors recorded in the database describe county attributes, which may be subject to ecologic bias. Despite these drawbacks, the SEER databases offer a wealth of accurate and representative data to evaluate clinical factors that may be associated with SLN biopsy and metastases.

In summary, thickness and ulceration were strong predictors of both SLN biopsy use and positive SLN biopsy results in children and young adults with melanoma. Children overall were more likely to have SLN metastases, a difference that was attributed to cases of nonulcerated melanomas measuring 1 to 2 mm in thickness. This finding warrants further investigation into age-related differences in melanoma biology.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES