OUP user menu

Clinical Characteristics and Survival Data of Melanoma Patients With Nevus Cell Aggregates Within Sentinel Lymph Nodes

Thilo Gambichler MD, Lisa Scholl MD, Markus Stücker MD, Falk G. Bechara MD, Klaus Hoffmann MD, Peter Altmeyer MD, Nick Othlinghaus MD
DOI: http://dx.doi.org/10.1309/AJCPG83CMAVFBWLC 566-573 First published online: 1 May 2013

Abstract

Histopathologic differentiation of nevus cell aggregates and metastatic melanoma in lymph nodes is challenging. Patients with melanoma who had undergone sentinel lymph node (SLN) biopsy were evaluated using univariate and multivariate analyses as well as Kaplan-Meier statistics. Of the 651 patients, 50 (7.7%) had a nodal nevus in the SLN. In the logistic regression model, primary melanoma on the lower extremities proved to be the strongest independent negative predictor of nodal nevi with an odds ratio of 0.11 (95% confidence interval, 0.034–0.36; P = .0002). Overall 5-year survival (P = .17) and 5-year disease-free survival (P = .45) of patients with nodal nevi did not significantly differ from that of patients with negative SLNs. The frequency and anatomic localization of nodal nevi observed in the present study are in line with previous studies. Our 5-year survival data clearly demonstrate that nevus cell aggregates in lymph nodes have to be considered a benign condition even though it occurs in patients with melanoma. This study provides an indirect proof of validity and accuracy of current histopathologic methods for differentiation between nodal nevi and melanoma metastasis.

Key Words:
  • Malignant melanoma
  • Nodal nevus
  • Nevus cell aggregates
  • Sentinel lymph node

Nevus cell aggregates (nodal nevi) in lymph nodes usually present as intracapsular or trabecular collections of small, uniform melanocytes that may resemble those seen in intradermal melanocytic nevi. Pathologists must be aware of nodal nevi because they can mimic micrometastasis of malignancies. Although not uncommon, the frequency with which nodal nevi occur is controversial. In patients with malignant melanoma, nodal nevi have been found in 0.12% to 0.54% of nodes from full lymph node dissections,14 1.2% of nodes from selective lymph node dissections,1 and 3.9% to 13% of sentinel lymph nodes (SLNs).1,57 Nodal nevus frequency is less in lymphadenectomy specimens for other malignancies—in breast carcinoma, for example, nodal nevi have been found in 0.017% to 0.1% of lymph nodes in this setting.1,2,6,8 By contrast, for example, Biddle et al9 observed nodal nevi in up to 7.3% of patients without melanoma. Nodal nevi have been found most frequently in axillary lymph nodes, but they also have been identified in cervical and inguinal regions; involvement of visceral nodes and other sites is rare.1,3,10,11 In most cases, the cellular aggregates have the appearance of conventional nevi. However, nodal nevi have also been seen with blue nevus, cellular blue nevus, and atypical spitzoid tumors.2,1215

The concept of how nevus cells arrive in or around lymph nodes can be divided into 2 broad categories: (1) arrested migration of neural crest progenitor cells during embryonic development and (2) transport of cells from cutaneous melanocytic lesions to lymph nodes, also termed mechanical transport or benign metastasis.3 Nevus cell aggregates in melanoma-draining lymph nodes can be mistaken for melanoma metastases, even though immunohistochemistry and other methods have improved the differential diagnosis between nodal nevi and metastatic melanoma.1620 The biological meaning of nodal nevi is still controversial.1,2,3,9,21,22 So far, the validity of current histopathology guidelines for the differentiation between nodal nevi and metastatic melanoma has not been proved formally.1 We therefore studied for the first time clinical characteristics and survival data of a reasonably large cohort of patients with melanoma who had nodal nevi and had previously undergone SLN biopsy.

Materials and Methods

Patients

This monocenter study was performed at the Skin Cancer Center Ruhr-University (Bochum, North-Rhine-Westphalia, Germany). The study was approved by the ethics review board of the Ruhr-University Bochum and conducted according to the principles of the Declaration of Helsinki. Between the end of 1999 and the beginning of 2011, SLN biopsy had been performed in 768 consecutive patients with malignant melanoma. All tumors were diagnosed by primary excision. Predominant indication for SLN biopsy was a Breslow tumor thickness of 1 mm or more. Moreover, upgrading in tumors less than 1 mm was considered in the presence of a Clark level of IV or higher, ulceration and lesions with signs of regression, or subungual localization of the primary melanoma. Prior to SLN biopsy, evidence of macrometastatic disease in regional lymph nodes or distant sites was ruled out by physical examination and individual staging procedures such as ultrasound, chest x-ray, and computed tomography. Patients with metastatic SLNs were subjected to complete lymphadenectomy. All patients with a primary melanoma thickness of 1.5 mm or more were considered for adjuvant low-dose interferon alfa-2b (Roferon; Roche Pharma AG, Grenzach-Wyhlen, Germany) therapy, and patients with melanoma-positive lymph nodes were considered for adjuvant high-dose interferon (Intron; MSD, Munich, Germany) therapy. Patients with coincident nodal nevi and lymph node melanoma metastasis were excluded from further evaluation. Clinical parameters such as age, sex, preexisting lesion of the primary melanoma (eg, congenital nevus, acquired melanocytic nevus), anatomic site of the primary melanoma, melanoma subtype, Clark level, and tumor regression, ulceration, and thickness (Breslow) were analyzed. Follow-up data were collected using chart review and contacting patients, relatives, and resident practitioners and dermatologists.

SLN Biopsy Procedure

Patients underwent lymphatic mapping and SLN biopsy as previously described in more detail by Wong et al.23 In brief, lymphoscintigraphy was performed by intradermal injection of technetium-99m sulfur colloid adjacent to the tumor or biopsy site to identify draining lymphatic basins by γ imaging. Intradermal injection of methylene blue dye was similarly performed during surgery. SLN biopsies were predominantly carried out under general anesthesia. Blue-stained and/or radioactive (>10% of the ex vivo counts) lymph nodes were removed and considered SLNs.

Histology and Immunohistochemistry

Preparation, macroscopic examination, sampling, and microscopic examination were performed in line with the recommendations for pathologic examination of SLNs from patients with melanoma as proposed by Scolyer et al.24,25 All SLNs were serially sectioned and stained with H&E and had immunohistochemical staining with S100 and Melan-A/MART-1 (DAKO, Hamburg, Germany) Image 1, Image 2, and Image 3. All SLNs were assessed by 2 senior dermatohistopathologists (M.S. and P.A.). In ambiguous cases, reevaluation of sections was performed with additional immunohistology for HMB45 and Ki-67. Morphologic features useful in distinguishing between nodal nevi and metastatic melanoma were applied in accordance with Scolyer and coworkers Table 1.24,25

Statistics

Data analysis was performed using the statistical package MedCalc Software (Mariakerke, Belgium). Distribution of data was assessed by the D’Agostino-Pearson test. Non-normally distributed data were expressed as medians and range. Data were analyzed using the χ2 test, Mann-Whitney test, Kendall’s τ procedure, and a forward logistic regression model including odds ratios (ORs) and 95% confidence intervals (CIs), the Hosmer-Lemeshow test, and receiver operating characteristic curve analysis. Overall 5-year survival (OS) and 5-year disease-free survival (DFS) were examined by using the Kaplan-Meier method, and differences between the curves were assessed by the log-rank test, including hazard ratios (HRs) and 95% CIs. Survival curves were calculated from the time of diagnosis of the primary melanoma and considered censored for non–melanoma-related deaths and unavailable data. P values less than .05 were considered significant.

Results

We included 651 patients with melanoma (303 males and 348 females; median age, 57 years [range, 15–85 years]) for descriptive and analytical statistics. The overall median tumor thickness was 1.5 mm, ranging from 0.15 to 84 mm (mean ± SD, 2.3 ± 3.8 mm). Fifty (7.7%) patients with a nodal nevus in the SLN were observed. Furthermore, we found 7 nodal nevi in non-SLNs on complete lymphadenectomy that were not included in the statistical analysis. Most nodal nevi (39/50; 78%) were exclusively found within the lymph node capsule, others were located in the trabeculae or capsule and trabeculae, and only 1 nodal nevus was observed within the lymph node parenchyma. Median age of patients with nodal nevi did not significantly differ from that of patients without nodal nevi (54 vs 57 years; P = .55). However, sex distribution significantly differed between patients with and without nodal nevi (P = .033). Patients without nodal nevi had a median Breslow tumor thickness comparable to that of patients with nodal nevi (1.5 mm [range, 0.35–6.6 mm] vs 1.4 mm [0.15–84 mm]; P = .70). Nodal nevi were significantly associated with superficial spreading melanoma, nodular melanoma, and unclassified melanoma; nodal nevi were not found in patients with lentigo maligna melanoma, acral melanoma, or others (P = .0077). Tumor regression, ulceration, Clark level, and the presence of preexisting lesions of the primary melanoma, such as congenital nevi and common melanocytic nevi, did not significantly differ between patients with and without nodal nevi (P = .12, P = .36, P = .24, and P = .11, respectively). The occurrence of the primary melanoma on the lower extremities was significantly more frequently observed in patients without nodal nevi (P < .0001). Patients’ data are also detailed in Table 2. In the forward logistic regression model, we included significant parameters (anatomic site of the primary melanoma, melanoma subtypes, and sex) with the presence or absence of nodal nevi as the dependent variable. Male sex proved to be an independent predictor for nodal nevi with an OR of 2 and a 95% CI of 1.1 to 3.6 (coefficient of regression, 0.68; standard error, 0.31; P = .029). However, after inclusion of the anatomic site of the primary melanoma variable, sex did not retain in the logistic regression model. Primary melanoma on the lower extremities proved to be the strongest independent negative predictor of nodal nevi with an OR of 0.11 and a 95% CI of 0.034 to 0.36 (coefficient of regression, −2.2; standard error, 0.60; P = .0002). The overall model fit had a significance level of less than .0001.

Image 1

A, Sentinel lymph node showing intracapsular nevus cell aggregates (arrow) that appear to surround a lymphatic vessel (H&E, magnification ×200). Neither significant nuclear or cell polymorphy nor subcapsular infiltration of the lymph node parenchyma is seen. B, Melanocytic origin of the intracapsular nevus cell aggregates (arrow) (Melan-A immunohistochemistry, magnification ×100).

Image 2

A, Sentinel lymph node showing metastatic melanoma (arrow) infiltrating deep into the lymph node parenchyma vessel. Polymorphous nuclei and cells as well as mitotic figures are observed (H&E, ×100). B, Melanocytic origin of the tumor infiltrates (arrow) (Melan-A immunohistochemistry, ×50).

Image 3

A, Monomorphic nevus cell aggregates within the lymph node capsule. The melanocytes are partly located around a blood and/or lymph vessel without evidence of vascular infiltration (H&E, ×400). B, Subcapsular-located melanoma metastasis with infiltrates of polymorphic cytoplasm-rich melanocytes, including nuclear changes and sporadic mitotic figures. Lymphatic tissue appears to be completely displaced (H&E, ×400).

View this table:
Table 1

Melanoma metastases of SLNs were observed in 150 (23%) patients, of whom 125 (83.3%) received complete lymphadenectomy and 116 (77.3%) high-dose interferon. Of the patients with lymph node metastases, 123 (82%) were diagnosed with stage IIIA melanoma. Positive SLN status was not significantly associated with age (P = .14), presence of a preexisting lesion (P = .59), and tumor regression (P = .19). However, Clark level (P < .0001), sex (P = .0006), anatomic site of the primary melanoma (P = .0023), melanoma subtype (P = .0014), tumor ulceration (P < .0001), and tumor thickness (P < .0001) proved to be significantly associated with positive SLN status. The logistic regression model using the absence or presence of SLN metastasis included Clark level, sex, anatomic site of the primary melanoma, melanoma subtype, tumor ulceration, and tumor thickness. Clark level, sex, and tumor ulceration did not retain in the model. Hence, the strongest independent positive predictor of SLN metastasis was tumor thickness with an OR of 3.6 and a 95% CI of 2.2 to 5.3 (coefficient of regression, 1.27; standard error, 0.26; P < .0001). Primary superficial spreading melanoma (coefficient of regression, −0.74; standard error, 0.27; P = .0065; OR, 0.48; 95% CI, 0.28–0.81) and primary melanoma on the upper extremities (coefficient of regression, −0.86; standard error, 0.37; P = .020; OR, 0.42; 95% CI, 0.21–0.87) proved to be independent predictors for negative SLN status. Goodness of fit for the logistic regression model was demonstrated by the Hosmer-Lemeshow test (P = .76). As shown in Figure 1, the OS (P = .0088; HR, 0.0) and DFS (P = .0068; HR, 0.23; 95% CI, 0.11–0.46) of patients with nodal nevi significantly differed from that of patients with positive SLNs, indicating that the prognosis of patients with melanoma who have nodal nevi is much more favorable than that of patients with melanoma-positive SLNs. Post hoc power analysis for comparing the OS and DFS of patients with nodal nevi and positive SLNs revealed a statistical power of 98% and 86%, respectively. Moreover, the OS (P = .17; HR, 0.0) and DFS (P = .45; HR, 0.64; 95% CI, 0.24–1.7) of patients with nodal nevi did not significantly differ from that of patients with negative SLNs. As expected, the OS (P < .0001; HR, 3.7; 95% CI, 1.9–7.4) and DFS (P < .0001; HR, 2.8; 95% CI, 1.7–4.6) of patients with positive SLNs were much worse when compared with patients with negative SLNs (Figure 1). In Table 3, the relapse and death rates of patients with melanoma are shown in detail.

View this table:
Table 2

Discussion

In the present study, the occurrence of metastatic SLNs (23%) was relatively high, which can be explained by the relatively high tumor thickness observed. Studies with lower proportions of thin primary melanomas or higher mean thickness values (1.4–2.7 mm) reported SLN positivity rates of 14.2% to 23.1%.26,27 As expected, tumor thickness was the strongest predictor of metastatic SLNs. By contrast, the presence of superficial spreading melanoma and primary melanoma on the upper extremities proved to be predictors for negative SLNs. Callender et al28 recently showed that anatomic location of the primary melanoma is an important independent predictor of SLN status and prognosis. Patients with primary melanomas of the head/neck and trunk have a worse prognosis than those with primary melanomas in other anatomic locations. Ito et al29 and Homsi et al30 also demonstrated that metastatic SLNs were significantly correlated with higher Breslow thickness, tumor subtype, and anatomic site of the primary melanoma. The 5-year DFS rates of patients with stage III melanoma have been reported to be 65% for stage IIIA, 37% for IIIB, and 26% for stage IIIC. Moreover, the 5-year survival within substages of stage III is 78%, 59%, and 40% for patients with stage IIIA, IIIB, and IIIC melanoma, respectively. As demonstrated in Figure 1, patients with stage III melanoma had remarkably favorable 5-year DFS and OS rates when compared with other patients with stage III melanoma reported in the literature.31,32 This may have occurred because our study population with positive SLNs consisted of a very high proportion of stage IIIA patients who had undergone lymphadenectomy and high-dose interferon treatment.

Figure 1

Kaplan-Meier curves show that overall survival (OS) and disease-free survival (DFS) of patients with nodal nevus (NN) significantly differ from OS (A, P = .0088) and DFS (B, P = .0068) of patients with a positive sentinel lymph node (SLN). Moreover, OS and DFS of patients with NN do not significantly differ from OS (C, P = .17) and DFS (D, P = .45) of patients with negative SLN. It is confirmed that OS and DFS of patients with negative SLN significantly differ from OS (E, P < .0001) and DFS (F, P < .0001) of patients with positive SLN.

The prevalence (7.7%) of nodal nevi observed in the present study is in line with the literature on patients with melanoma who showed a nodal nevus after SLN biopsy.1,57 We detected most nodal nevi in the SLNs, and only a small number of nodal nevi were seen in non-SLNs, favoring the mechanical transport theory.1,3,5,7 Our data also confirmed data obtained from previous studies indicating that nodal nevi are predominantly found within the capsule of the SLN.1,3,57,24,25 The usual capsular location with sparing of sinuses may rather favor the theory of arrested migration of neural crest progenitor cells during embryogenesis.3 In contrast to research by Holt et al,6 who studied 72 melanoma patients including 8 patients with nodal nevi, we did not observe a significant association between the occurrence of nodal nevi and high tumor thickness. This observation appears to speak against the mechanical transport theory of nodal nevi.3 On univariate analysis, we found that nodal nevi were significantly associated with male sex as well as superficial spreading melanoma, primary nodular melanoma, and unclassified melanomas. Nevertheless, multivariate analysis using logistic regression did not demonstrate male sex and specific melanoma subtypes as significant independent predictors of nodal nevi. Notably, primary melanoma on the lower extremities proved to be the strongest independent negative predictor of nodal nevi. This is in line with data from previous studies indicating that nodal nevi are predominantly found in axillary lymph nodes.1,3,10,11 All in all, the clinical characteristics of patients with nodal nevi observed in the present study do not favor consistently one of the aforementioned theories of how nevus cell aggregates may develop in lymph nodes. Hence, large prospective trials on this issue are warranted. In the present study, however, we have demonstrated for the first time that the OS and DFS of patients with nodal nevi are much more favorable when compared with that of patients with positive SLNs. We also showed that patients with nodal nevi have the same OS and DFS rates as patients with negative SLNs. Our study, however, is limited by the follow-up analysis end point of 5 years only. A 10-year survival analysis was not possible because of the limited number of patients with such a long follow-up. Nonetheless, we think that our data can serve as an indirect proof of validity of current histopathologic criteria and methods for the differentiation between nodal nevi and “true” melanoma metastasis, as recently proposed by Scolyer and colleagues.24,25 Histopathologic differentiation between nodal nevi and metastatic lymph nodes has a profound prognostic meaning.24,25,3032 The present data provide the definitive clinical implication that aggressive treatments such as complete lymphadenectomy and high-dose interferon are not indicated in patients with melanoma who have nodal nevi but do not show evidence for “true” metastatic lymph node disease.

View this table:
Table 3

In conclusion, this was the largest published study on nodal nevi in patients with melanoma, including survival analysis. The occurrence of primary melanoma on the lower extremities appears to be the strongest independent predictor for the absence of nodal nevi. OS and DFS rates of patients with melanoma who have nodal nevi in the SLN are favorable and do not differ from patients with melanoma who do not have “true” metastatic SLNs. This was the first study on nodal nevi in patients with melanoma, providing indirect proof of the validity and accuracy of current histopathologic methods for differentiation between nodal nevi and melanoma metastasis.

CME/SAM

Upon completion of this activity you will be able to:

  • list histologic features of cell groups in lymph nodes that favor nevus cell aggregates rather than melanoma metastases.

  • apply immunohistochemical staining in differentiation of nevus cells from melanoma metastases.

The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module.

The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.

Questions appear on p 690. Exam is located at www.ascp.org/ajcpcme.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
View Abstract