Purpose:

In patients with cutaneous melanoma, metastasis in a nonsentinel lymph node (non-SLN) is a strong independent adverse prognostic factor. However, patients with a tumor-involved SLN no longer routinely undergo completion lymph node dissection (CLND). We hypothesized that SLN tumor burden may predict non-SLN tumor burden.

Experimental Design:

We compared tumor burden parameters between SLN and non-SLN in patients with cutaneous melanoma who underwent SLN biopsy with a positive SLN during 2003 to 2008 at The University of Texas MD Anderson Cancer Center.

Results:

We identified 336 eligible patients with a positive SLN. Of these, 308 (92%) underwent CLND, and 35 (10%) had non-SLN metastasis. The median follow-up time was 6.0 years. For patients with maximum diameter of tumor in the SLN ≤2.0 mm, >2.0–5.0 mm, and >5.0 mm, non-SLN metastasis was detected in 5 of 200 patients (3%), 10 of 63 patients (16%), and 20 of 57 patients (35%), and the mean maximum diameters of the non-SLN tumor deposits were 0.09, 1.56, and 2.71 mm, respectively (P < 0.0001). The percentage of patients with both subcapsular and intraparenchymal non-SLN tumor was higher for patients with SLN tumor in both locations than for patients with SLN tumor in only one location (P < 0.0001). Extranodal extension in a non-SLN was more common in patients with extranodal extension in an SLN (P = 0.003).

Conclusions:

In patients with cutaneous melanoma who undergo CLND, SLN tumor burden predicts non-SLN tumor burden. SLN tumor burden parameters provide accurate prognostic stratification independent of non-SLN status and should be considered for incorporation into future staging systems and integrated risk models.

Translational Relevance

Due to the findings from two recent clinical trials, patients with cutaneous melanoma with microscopic sentinel lymph node (SLN) metastasis may no longer be routinely offered CLND. However, in the current melanoma staging system of the American Joint Committee on Cancer, the total number of tumor-involved regional nodes is one of the major criteria that determines the N category, and the total number of tumor-involved regional nodes cannot be fully determined without performing CLND. The prognosis of patients with stage III disease varies widely depending on the subcategory; 5-year melanoma-specific survival (MSS) rates in patients with N1a, stage IIIA, N3c, and stage IIID melanoma were 84%, 93%, 52%, and 32%, respectively. If fewer CLNDs are performed, other parameters will be needed for prognostic stratification. In this study, we compared tumor burden parameters between SLN and non-SLN in 336 patients with cutaneous melanoma who underwent SLN biopsy with a positive SLN and subsequent CLND at a single institution. We found that in patients with cutaneous melanoma who underwent CLND, higher SLN tumor burden predicted higher non-SLN tumor burden. Our results suggest that SLN tumor burden parameters provide accurate prognostic stratification independent of non-SLN status and should be considered for incorporation into future staging systems and integrated risk models.

Two recent clinical trials, the DeCOG-SLT trial (1, 2) and the MSLT-II trial (3), have demonstrated that completion lymph node dissection (CLND) was not associated with increased melanoma-specific survival in patients with cutaneous melanoma with microscopic metastasis in sentinel lymph nodes (SLN; refs. 1–3). In patients with cutaneous melanoma, the presence of disease in nonsentinel lymph nodes (non-SLN) has been found to be a strong independent adverse prognostic factor for survival (3). However, in light of the above-mentioned trials, patients with cutaneous melanoma with microscopic SLN metastasis no longer routinely undergo CLND (4, 5).

In the 8th edition American Joint Committee on Cancer (AJCC) melanoma staging system, the total number of tumor-involved regional nodes continues to be one of the criteria that determines the N category. However, the total number of tumor-involved regional nodes cannot be definitively determined without performing CLND. The European Organization for Research and Treatment of Cancer (EORTC) Melanoma Group showed that CLND led to higher N category in 19% of patients and higher AJCC stage group in 5% to 6% of patients in a retrospective cohort of 1,015 patients with positive SLNs (6). Other N category criteria include the extent of regional nodal involvement (clinically occult or clinically detected) and the presence or absence of nonnodal locoregional metastases, such as microsatellite, satellite, and in-transit metastases; together, these define N subcategories N1a thru N3c. The 8th edition AJCC melanoma staging system has also recommended tumor burden measurement (7–9). On the basis of status of regional nodes and nonnodal locoregional metastases, together with T category criteria (tumor thickness and ulceration), four stage III groups were defined (IIIA–IIID; refs. 7, 8).

The prognosis of patients with stage III disease varies widely depending on the subcategory and substage (10, 11); 5-year melanoma-specific survival (MSS) rates in patients with N1a, stage IIIA, N3c, and stage IIID melanoma were 84%, 93%, 52%, and 32%, respectively (7). Significant risk is also demonstrated in the 5-year MSS rates in patients with N3a (4 or more positive regional lymph nodes without in-transit or satellite disease) and IIIC disease, which were 60% and 69%, respectively. Thus, although SLN status was found to be the strongest predictor of death from cutaneous melanoma in several studies (12–14), the mere determination of positive or negative SLN status may not be sufficient to stratify patients with localized cutaneous melanoma in terms of prognosis.

Because significantly fewer CLNDs are now performed, integration of other parameters to improve prognostic stratification is needed (9). Tumor burden in tumor-involved SLNs has been well studied, and a substantial number of studies have demonstrated correlations between SLN tumor burden and prognosis and between SLN tumor burden and the presence or absence of metastasis in non-SLNs (15–22). Indeed, in the 8th edition AJCC melanoma staging system, it is recommended that assessment of SLN tumor burden be performed (7, 8). We previously reported that SLN tumor burden predicts whether or not disease is present in non-SLNs (17). We tested the hypothesis that SLN tumor burden also predicts tumor burden in non-SLNs and assessed whether these pathologic features may potentially improve risk stratification and contribute to decision-making in an era in which most patients no longer undergo routine CLND.

Patients

In this retrospective study, we reviewed pathology and clinical databases to identify patients with cutaneous melanoma with positive SLN biopsy and subsequent CLND at The University of Texas MD Anderson Cancer Center. CLND was performed from January 1, 2003, through December 31, 2008. Wide excision of the primary melanoma was performed with margins appropriate for the tumor thickness. Lymphatic mapping and SLN biopsy were performed as described previously (14, 17). Patients with a positive SLN were offered CLND during a discussion of their treatment plan. Permission to perform the present study and a waiver for informed consent were obtained from the MD Anderson Institutional Review Board. All aspects of our research were performed in accordance with the Declaration of Helsinki.

Clinicopathologic factors

Pathologic features obtained from patients' records included maximum diameter of nodal tumor deposits, microanatomic locations of nodal tumor deposits (subcapsular, intraparenchymal, or both locations), and the presence or absence of extranodal extension for both SLNs and non-SLNs. Tumor burden in the SLN was quantified similarly to prior studies (11, 13). These SLN tumor burden parameters are consistent with the recommended data elements for pathology reporting according to the College of American Pathologists protocol (23).

Pathologic evaluation of the SLN was performed as described previously (14, 24, 25). Briefly, SLNs were bread-loafed, and formalin-fixed, paraffin-embedded sections were examined using routine hematoxylin and eosin (H&E) staining. If the initial H&E-stained slide was negative for metastasis, three serial sections were cut deeper in the block; one was stained with H&E and another was labeled with a pan-melanocytic cocktail (HMB45, anti-MART1, and antityrosinase; refs. 24, 25). Non-SLNs were bread-loafed, and a single section was examined using H&E staining. In non-SLNs with morphologic features suggestive of melanocytic cells, IHC was performed to either confirm or rule out metastatic melanoma.

Statistical methods

Cutoff Finder software, which evaluates every possible cutoff point and determines the value that indicates the most significant difference between the two groups (26), selected 5.8 mm as the best cutoff point to stratify SLN-positive patients into two prognostic categories for MSS. For ease of potential clinical use, we rounded the cutoff point to 5.0 mm. The cutoff of 2.0 mm was chosen because it was associated with higher risk of non-SLN positivity as reported in our previous study (17). The associations between clinicopathologic factors and recurrence-free survival (RFS) and MSS were assessed using univariate Cox proportional hazard regression models. In addition, a final multivariable model was produced for each survival outcome using backward selection with an entrance criterion of P < 0.10 and exit criterion of P ≥ 0.05. RFS was computed from the date of the SLN biopsy to the date of disease recurrence after definitive surgery or death. Positive non-SLN was not considered to be an event for RFS. In addition, of the 28 patients who had a positive SLN and did not undergo CLND, 3 (11%) had in-transit metastatic melanoma, 4 (14%) had metastatic melanoma in a regional LN, and 5 (18%) had distant metastasis. All of those patients were included as having an event in the RFS assessment. Patients who were alive and had no recurrence at the last follow-up date were censored. MSS was computed from the date of the SLN biopsy to the date of death due to melanoma. Patients who died from other causes as well as patients who were alive at the last follow-up were censored. Differences in and associations between tumor burden parameters in SLNs and non-SLNs were evaluated. Differences between groups were assessed by Kruskal–Wallis test, while associations were determined using generalized Fisher exact test. All the statistical analyses were performed using SAS 9.4 for Windows (SAS Institute Inc.). A significance level of 5% was used for all the statistical tests; no adjustments were made for multiple testing.

Patients

We identified 336 eligible patients with at least one positive SLN; of these, 35 (10%) had a non-SLN metastasis detected on CLND (Table 1). The median primary tumor thickness was 2.6 mm (range, 0.3–21.0 mm). Ulceration was present in the primary tumors of 107 patients (32%; Table 1). CLND was performed in 308 patients (92%). Adjuvant systemic therapy was administered to 136 patients (40%). The most frequently used agent, administered to 100 patients (30%), was interferon-alfa. The median follow-up time was 6.0 years (range, 0.1–11.4 years), and the median follow-up time for the 200 patients who were alive at their last follow-up was 7.3 years (range, 0.1–11.2 years).

Table 1.

Demographic and clinicopathologic factors (n = 336)

FactorPatients, n (%)
Sex 
 Female 123 (37) 
 Male 213 (63) 
Age, years 
 <40 65 (19) 
 40–60 151 (45) 
 >60 120 (36) 
Primary tumor site 
 Trunk 138 (41) 
 Head and neck 65 (19) 
 Upper limb 42 (13) 
 Lower limb 91 (27) 
Subtype 
 SSM 155 (46) 
 NM 89 (26) 
 LMM 7 (2) 
 ALM 32 (10) 
 Unclassified 50 (15) 
 Unknown 3 (1) 
Tumor category 
 T1 35 (10) 
 T2 98 (29) 
 T3 126 (38) 
 T4 77 (23) 
Ulceration 
 Absent 228 (68) 
 Present 107 (32) 
 Unknown 1 (0.3) 
Mitotic rate, mitoses per mm2 
 <1 32 (10) 
 1 52 (15) 
 2–5 135 (40) 
 >5 103 (31) 
 Unknown 14 (4) 
Number of positive SLNs 
 1 226 (67) 
 2–3 86 (26) 
 ≥4 24 (7) 
 Maximum diameter of tumor deposit in SLNs, mm 
 ≤2.0 200 (60) 
 >2.0–5.0 63 (19) 
 >5.0 57 (17) 
 Unknown 16 (5) 
Microanatomic location of tumor deposit in SLN 
 Subcapsular 175 (52) 
 Intraparenchymal 27 (8) 
 Both 125 (37) 
 Unknown 9 (3) 
Extranodal extension in SLN 
 Absent 290 (86) 
 Present 42 (13) 
 Unknown 4 (1) 
CLND performed 
 No 19 (6) 
 Yes 308 (92) 
 Unknown 9 (3) 
Number of positive non-SLNs 
 0 282 (84) 
 1 14 (4) 
 ≥2 21 (6) 
 Unknown 19 (6) 
Maximum diameter of tumor deposit in non-SLNs, mm 
 No metastasis 282 (84) 
 ≤2.0 10 (3) 
 >2.0–5.0 9 (3) 
 >5.0 16 (5) 
 Unknown 19 (6) 
Microanatomic location of tumor deposit in non-SLNs 
 No metastasis/not both locations 290 (86) 
 Both subcapsular and intraparenchymal 27 (8) 
 Unknown 19 (6) 
Extranodal extension in non-SLNs 
 No metastasis 282 (84) 
 Absent 25 (7) 
 Present 10 (3) 
 Unknown 19 (6) 
Adjuvant systemic therapy 
 None 195 (58) 
 Administered 136 (40) 
 Unknown 5 (1) 
FactorPatients, n (%)
Sex 
 Female 123 (37) 
 Male 213 (63) 
Age, years 
 <40 65 (19) 
 40–60 151 (45) 
 >60 120 (36) 
Primary tumor site 
 Trunk 138 (41) 
 Head and neck 65 (19) 
 Upper limb 42 (13) 
 Lower limb 91 (27) 
Subtype 
 SSM 155 (46) 
 NM 89 (26) 
 LMM 7 (2) 
 ALM 32 (10) 
 Unclassified 50 (15) 
 Unknown 3 (1) 
Tumor category 
 T1 35 (10) 
 T2 98 (29) 
 T3 126 (38) 
 T4 77 (23) 
Ulceration 
 Absent 228 (68) 
 Present 107 (32) 
 Unknown 1 (0.3) 
Mitotic rate, mitoses per mm2 
 <1 32 (10) 
 1 52 (15) 
 2–5 135 (40) 
 >5 103 (31) 
 Unknown 14 (4) 
Number of positive SLNs 
 1 226 (67) 
 2–3 86 (26) 
 ≥4 24 (7) 
 Maximum diameter of tumor deposit in SLNs, mm 
 ≤2.0 200 (60) 
 >2.0–5.0 63 (19) 
 >5.0 57 (17) 
 Unknown 16 (5) 
Microanatomic location of tumor deposit in SLN 
 Subcapsular 175 (52) 
 Intraparenchymal 27 (8) 
 Both 125 (37) 
 Unknown 9 (3) 
Extranodal extension in SLN 
 Absent 290 (86) 
 Present 42 (13) 
 Unknown 4 (1) 
CLND performed 
 No 19 (6) 
 Yes 308 (92) 
 Unknown 9 (3) 
Number of positive non-SLNs 
 0 282 (84) 
 1 14 (4) 
 ≥2 21 (6) 
 Unknown 19 (6) 
Maximum diameter of tumor deposit in non-SLNs, mm 
 No metastasis 282 (84) 
 ≤2.0 10 (3) 
 >2.0–5.0 9 (3) 
 >5.0 16 (5) 
 Unknown 19 (6) 
Microanatomic location of tumor deposit in non-SLNs 
 No metastasis/not both locations 290 (86) 
 Both subcapsular and intraparenchymal 27 (8) 
 Unknown 19 (6) 
Extranodal extension in non-SLNs 
 No metastasis 282 (84) 
 Absent 25 (7) 
 Present 10 (3) 
 Unknown 19 (6) 
Adjuvant systemic therapy 
 None 195 (58) 
 Administered 136 (40) 
 Unknown 5 (1) 

Abbreviations: ALM, acral lentiginous melanoma; CLND, completion lymph node dissection; LMM, lentigo maligna melanoma; NM, nodular melanoma; SLN, sentinel lymph node; SSM, superficial spreading melanoma.

SLN tumor burden parameters

The maximum diameter of tumor deposit in SLNs was ≤2.0 mm in 200 patients (60%), >2.0–5.0 mm in 63 patients (19%), and >5.0 mm in 57 patients (17%). Microanatomic location of tumor deposit in SLNs was subcapsular in 175 patients (52%), intraparenchymal in 27 (8%), and both locations in 125 (37%). Extranodal extension in the SLN was absent in 290 patients (86%), and present in 42 (13%).

The median maximum diameter of tumor deposit in the SLN was 0.7 mm (range, 0.01–10.0 mm) when disease was located only in the subcapsular region, 1.0 mm (range, 0.08–40.0 mm) when disease was located only in the intraparenchymal region, and 3.1 mm (range, 0.02–37.0 mm) when disease was located in both the subcapsular and intraparenchymal regions (P < 0.0001). Similarly, the median maximum diameter of tumor deposit in the SLN was 1.0 mm (range, 0.01–22.0 mm) for the lesions without extranodal extension and 6.5 mm (range, 0.8–40.0 mm) for those with extranodal extension (P < 0.0001).

Relationships between SLN tumor burden parameters, non-SLN tumor burden parameters, and survival

Analysis of relationships between SLN and non-SLN tumor burden parameters and survival showed that larger maximum diameter of tumor deposit in SLN and in non-SLN, both subcapsular and intraparenchymal tumor deposit in SLN and in non-SLN (compared with only subcapsular), and the presence of extranodal extension in SLN and in non-SLN were associated with poor RFS (Table 2) and MSS (Table 3) in univariate models. In multivariable models, both subcapsular and intraparenchymal tumor deposit in SLN (compared with only subcapsular) and the presence of extranodal extension in SLN were independently associated with both RFS (Table 2) and MSS (Table 3).

Table 2.

Factors associated with RFS in patients with a positive SLN

UnivariateMultivariable
FactorPatients, nHR (95% CI)P valuePatients, nHR (95% CI)P value
Sex 
 Female (ref) 123 — —  NI  
 Male 205 1.18 (0.86–1.63) 0.31    
Age, years 
 <40 (ref) 64 — — 55 — — 
 40–60 149 2.10 (1.22–3.62) 0.008 122 1.53 (0.86–2.73) 0.15 
 >60 115 3.89 (2.27–6.68) <0.0001 84 3.68 (2.06–6.59) <0.0001 
Primary tumor site 
 Trunk (ref) 135 — —  NI  
 Head and neck 61 1.48 (0.99–2.21) 0.055    
 Upper limb 42 0.65 (0.37–1.16) 0.15    
 Lower limb 90 1.03 (0.70–1.51) 0.88    
Subtype 
 SSM (ref) 151 — —  NI  
 NM 88 1.02 (0.69–1.51) 0.92    
 LMM 1.82 (0.57–5.79) 0.31    
 ALM 32 2.65 (1.66–4.22) <0.0001    
 Unclassified 48 1.27 (0.81–2.00) 0.29    
Tumor category 
 T1 (ref) 35 — —  NI  
 T2 97 0.81 (0.43–1.51) 0.50    
 T3 123 1.68 (0.95–2.98) 0.08    
 T4 73 2.39 (1.31–4.35) 0.004    
Ulceration 
 Absent (ref) 225 — —  NI  
 Present 102 1.72 (1.25–2.36) <0.001    
Mitotic rate, mitoses per mm2 
 <1 32 — —  NI  
 1 51 0.68 (0.32–1.45) 0.32    
 2–5 134 1.15 (0.62–2.14) 0.65    
 >5 97 2.22 (1.20–4.11) 0.012    
Number of positive SLNs 
 1 (ref) 238 — —  NI  
 2–3 85 1.34 (0.95–1.88) 0.10    
 ≥4 1.26 (0.31–5.12) 0.74    
Maximum diameter of tumor deposit in SLNs, mm 
 ≤2.0 (ref) 195 — —  NI  
 >2.0–5.0 62 1.49 (1.00–2.23) 0.050    
 >5.0 55 2.60 (1.76–3.83) <0.0001    
Microanatomic location of tumor deposit in SLN 
 Subcapsular (ref) 172 — — 139 — — 
 Intraparenchymal 27 1.40 (0.78–2.54) 0.26 23 1.35 (0.71–2.56) 0.36 
 Both 121 1.99 (1.43–2.76) <0.0001 99 1.90 (1.25–2.90) 0.003 
Extranodal extension in SLN 
 Absent (ref) 284 — — 227 — — 
 Present 40 2.57 (1.72–3.85) <0.0001 34 2.11 (1.31–3.39) 0.002 
Number of positive non-SLNs 
 0 (ref) 275 — — 229 — — 
 1 14 2.31 (1.21–4.40) 0.011 13 1.67 (0.83–3.35) 0.15 
 ≥2 21 2.94 (1.74–4.96) <0.0001 19 2.28 (1.26–4.11) 0.006 
Maximum diameter of tumor deposit in non-SLNs, mm 
 No metastasis (ref) 273 — —  NI  
 ≤2.0 10 1.78 (0.78–4.04) 0.17    
 >2.0–5.0 1.85 (0.76–4.54) 0.18    
 >5.0 16 4.06 (2.37–6.96) <0.0001    
Microanatomic location of tumor deposit in non-SLNs 
 No metastasis/not both (ref) 283 — —  NI  
 Both subcapsular and intraparenchymal 27 2.87 (1.80–4.56) <0.0001    
Extranodal extension in non-SLNs 
 No metastasis (ref) 275 — —  NI  
 Absent 25 1.84 (1.09–3.10) 0.022    
 Present 10 10.78 (5.44–21.37) <0.0001    
CLND performed 
 No (ref) 18 — —  NI  
 Yes 302 0.64 (0.34–1.22) 0.18    
Adjuvant systemic therapy 
 None (ref) 190 — —  NI  
 Administered 135 0.59 (0.43–0.82) 0.002    
UnivariateMultivariable
FactorPatients, nHR (95% CI)P valuePatients, nHR (95% CI)P value
Sex 
 Female (ref) 123 — —  NI  
 Male 205 1.18 (0.86–1.63) 0.31    
Age, years 
 <40 (ref) 64 — — 55 — — 
 40–60 149 2.10 (1.22–3.62) 0.008 122 1.53 (0.86–2.73) 0.15 
 >60 115 3.89 (2.27–6.68) <0.0001 84 3.68 (2.06–6.59) <0.0001 
Primary tumor site 
 Trunk (ref) 135 — —  NI  
 Head and neck 61 1.48 (0.99–2.21) 0.055    
 Upper limb 42 0.65 (0.37–1.16) 0.15    
 Lower limb 90 1.03 (0.70–1.51) 0.88    
Subtype 
 SSM (ref) 151 — —  NI  
 NM 88 1.02 (0.69–1.51) 0.92    
 LMM 1.82 (0.57–5.79) 0.31    
 ALM 32 2.65 (1.66–4.22) <0.0001    
 Unclassified 48 1.27 (0.81–2.00) 0.29    
Tumor category 
 T1 (ref) 35 — —  NI  
 T2 97 0.81 (0.43–1.51) 0.50    
 T3 123 1.68 (0.95–2.98) 0.08    
 T4 73 2.39 (1.31–4.35) 0.004    
Ulceration 
 Absent (ref) 225 — —  NI  
 Present 102 1.72 (1.25–2.36) <0.001    
Mitotic rate, mitoses per mm2 
 <1 32 — —  NI  
 1 51 0.68 (0.32–1.45) 0.32    
 2–5 134 1.15 (0.62–2.14) 0.65    
 >5 97 2.22 (1.20–4.11) 0.012    
Number of positive SLNs 
 1 (ref) 238 — —  NI  
 2–3 85 1.34 (0.95–1.88) 0.10    
 ≥4 1.26 (0.31–5.12) 0.74    
Maximum diameter of tumor deposit in SLNs, mm 
 ≤2.0 (ref) 195 — —  NI  
 >2.0–5.0 62 1.49 (1.00–2.23) 0.050    
 >5.0 55 2.60 (1.76–3.83) <0.0001    
Microanatomic location of tumor deposit in SLN 
 Subcapsular (ref) 172 — — 139 — — 
 Intraparenchymal 27 1.40 (0.78–2.54) 0.26 23 1.35 (0.71–2.56) 0.36 
 Both 121 1.99 (1.43–2.76) <0.0001 99 1.90 (1.25–2.90) 0.003 
Extranodal extension in SLN 
 Absent (ref) 284 — — 227 — — 
 Present 40 2.57 (1.72–3.85) <0.0001 34 2.11 (1.31–3.39) 0.002 
Number of positive non-SLNs 
 0 (ref) 275 — — 229 — — 
 1 14 2.31 (1.21–4.40) 0.011 13 1.67 (0.83–3.35) 0.15 
 ≥2 21 2.94 (1.74–4.96) <0.0001 19 2.28 (1.26–4.11) 0.006 
Maximum diameter of tumor deposit in non-SLNs, mm 
 No metastasis (ref) 273 — —  NI  
 ≤2.0 10 1.78 (0.78–4.04) 0.17    
 >2.0–5.0 1.85 (0.76–4.54) 0.18    
 >5.0 16 4.06 (2.37–6.96) <0.0001    
Microanatomic location of tumor deposit in non-SLNs 
 No metastasis/not both (ref) 283 — —  NI  
 Both subcapsular and intraparenchymal 27 2.87 (1.80–4.56) <0.0001    
Extranodal extension in non-SLNs 
 No metastasis (ref) 275 — —  NI  
 Absent 25 1.84 (1.09–3.10) 0.022    
 Present 10 10.78 (5.44–21.37) <0.0001    
CLND performed 
 No (ref) 18 — —  NI  
 Yes 302 0.64 (0.34–1.22) 0.18    
Adjuvant systemic therapy 
 None (ref) 190 — —  NI  
 Administered 135 0.59 (0.43–0.82) 0.002    

Abbreviations: ALM, acral lentiginous melanoma; CLND, completion lymph node dissection; HR, hazard ratio; LMM, lentigo maligna melanoma; NI, not included in final model; NM, nodular melanoma; ref, reference; SSM, superficial spreading melanoma. The numbers in some groups do not total 336 because data were not available for some patients. Values in bold are statistically significant.

Table 3.

Factors associated with MSS in patients with a positive SLN

UnivariateMultivariable
FactorPatients, nHR (95% CI)P valuePatients, nHR (95% CI)P value
Sex 
 Female (ref) 120 — —  NI  
 Male 208 1.20 (0.81–1.77) 0.37    
Age, years 
 <40 (ref) 64 — — 55 — — 
 40–60 149 2.94 (1.45–5.95) 0.003 121 1.90 (0.88–4.13) 0.10 
 >60 115 4.36 (2.14–8.88) <0.0001 84 3.57 (1.63–7.82) 0.002 
Primary tumor site 
 Trunk (ref) 135 — — 108 — — 
 Head and neck 63 1.13 (0.69–1.85) 0.64 43 0.96 (0.52–1.78) 0.90 
 Upper limb 41 0.51 (0.24–1.07) 0.08 35 0.63 (0.28–1.42) 0.26 
 Lower limb 89 0.94 (0.61–1.47) 0.80 74 0.38 (0.18–0.78) 0.008 
Subtype 
 SSM (ref) 152 — — 117 — — 
 NM 86 0.80 (0.49–1.31) 0.37 78 0.73 (0.42–1.26) 0.26 
 LMM 0.38 (0.02–6.39) 0.50 3.94 (0.15–102.1) 0.41 
 ALM 32 2.67 (1.59–4.49) <0.001 22 3.47 (1.48–8.16) 0.004 
 Unclassified 49 1.14 (0.67–1.96) 0.62 41 0.68 (0.35–1.35) 0.27 
Tumor category 
 T1 (ref) 35 — —  NI  
 T2 96 0.59 (0.28–1.24) 0.16    
 T3 123 1.37 (0.71–2.64) 0.35    
 T4 74 2.25 (1.15–4.42) 0.018    
Ulceration 
 Absent (ref) 222 — —  NI  
 Present 105 1.91 (1.31–2.80) <0.001    
Mitotic rate, mitoses per mm2 
 <1 32 0.48 (0.23–0.97) 0.042 26 0.83 (0.37–1.86) 0.66 
 1 50 0.21 (0.10–0.47) <0.001 41 0.33 (0.14–0.80) 0.014 
 2–5 130 0.51 (0.34–0.78) 0.002 107 0.58 (0.35–0.96) 0.035 
 >5 (ref) 102 — — 86 — — 
Number of positive SLNs 
 1 (ref) 236 — —  NI  
 2–3 87 1.58 (1.06–2.35) 0.025    
 ≥4 2.48 (0.61–10.13) 0.20    
Maximum diameter of tumor deposit in SLNs, mm 
 ≤2.0 (ref) 194 — —  NI  
 >2.0–5.0 62 1.47 (0.89–2.42) 0.13    
 >5.0 56 3.44 (2.22–5.34) <0.0001    
Microanatomic location of tumor deposit in SLN 
 Subcapsular (ref) 170 — — 138 — — 
 Intraparenchymal 27 1.71 (0.86–3.42) 0.13 23 1.21 (0.55–2.70) 0.63 
 Both 122 2.35 (1.56–3.52) <0.0001 99 2.27 (1.34–3.84) 0.002 
Extranodal extension in SLN 
 Absent (ref) 282 — — 225 — — 
 Present 42 2.73 (1.75–4.27) <0.0001 35 1.86 (1.08–3.19) 0.025 
Number of positive non-SLNs 
 0 (ref) 278 — — 230 — — 
 1 13 2.68 (1.24–5.79) 0.012 12 2.28 (0.95–5.49) 0.07 
 ≥2 20 3.84 (2.17–6.78) <0.0001 18 2.53 (1.28–4.99) 0.007 
Maximum diameter of tumor deposit in non-SLNs, mm 
 No metastasis (ref) 278 — —  NI  
 ≤2.0 2.93 (1.19–7.22) 0.020    
 >2.0–5.0 2.53 (1.02–6.23) 0.044    
 >5.0 15 4.26 (2.27–8.01) <0.0001    
Microanatomic location of tumor deposit in non-SLNs 
 No metastasis/not both (ref) 286 — —  NI  
 Both subcapsular and intraparenchymal 25 3.55 (2.08–6.06) <0.0001    
Extranodal extension in non-SLNs 
 No metastasis (ref) 278 — —  NI  
 Absent 23 2.34 (1.28–4.28) 0.006    
 Present 10 8.14 (4.04–16.39) <0.0001    
CLND performed 
 No (ref) 17 — —  NI  
 Yes 303 0.86 (0.35–2.12) 0.75    
Adjuvant systemic therapy 
 None (ref) 188 — —  NI  
 Administered 135 0.68 (0.46–1.00) 0.053    
UnivariateMultivariable
FactorPatients, nHR (95% CI)P valuePatients, nHR (95% CI)P value
Sex 
 Female (ref) 120 — —  NI  
 Male 208 1.20 (0.81–1.77) 0.37    
Age, years 
 <40 (ref) 64 — — 55 — — 
 40–60 149 2.94 (1.45–5.95) 0.003 121 1.90 (0.88–4.13) 0.10 
 >60 115 4.36 (2.14–8.88) <0.0001 84 3.57 (1.63–7.82) 0.002 
Primary tumor site 
 Trunk (ref) 135 — — 108 — — 
 Head and neck 63 1.13 (0.69–1.85) 0.64 43 0.96 (0.52–1.78) 0.90 
 Upper limb 41 0.51 (0.24–1.07) 0.08 35 0.63 (0.28–1.42) 0.26 
 Lower limb 89 0.94 (0.61–1.47) 0.80 74 0.38 (0.18–0.78) 0.008 
Subtype 
 SSM (ref) 152 — — 117 — — 
 NM 86 0.80 (0.49–1.31) 0.37 78 0.73 (0.42–1.26) 0.26 
 LMM 0.38 (0.02–6.39) 0.50 3.94 (0.15–102.1) 0.41 
 ALM 32 2.67 (1.59–4.49) <0.001 22 3.47 (1.48–8.16) 0.004 
 Unclassified 49 1.14 (0.67–1.96) 0.62 41 0.68 (0.35–1.35) 0.27 
Tumor category 
 T1 (ref) 35 — —  NI  
 T2 96 0.59 (0.28–1.24) 0.16    
 T3 123 1.37 (0.71–2.64) 0.35    
 T4 74 2.25 (1.15–4.42) 0.018    
Ulceration 
 Absent (ref) 222 — —  NI  
 Present 105 1.91 (1.31–2.80) <0.001    
Mitotic rate, mitoses per mm2 
 <1 32 0.48 (0.23–0.97) 0.042 26 0.83 (0.37–1.86) 0.66 
 1 50 0.21 (0.10–0.47) <0.001 41 0.33 (0.14–0.80) 0.014 
 2–5 130 0.51 (0.34–0.78) 0.002 107 0.58 (0.35–0.96) 0.035 
 >5 (ref) 102 — — 86 — — 
Number of positive SLNs 
 1 (ref) 236 — —  NI  
 2–3 87 1.58 (1.06–2.35) 0.025    
 ≥4 2.48 (0.61–10.13) 0.20    
Maximum diameter of tumor deposit in SLNs, mm 
 ≤2.0 (ref) 194 — —  NI  
 >2.0–5.0 62 1.47 (0.89–2.42) 0.13    
 >5.0 56 3.44 (2.22–5.34) <0.0001    
Microanatomic location of tumor deposit in SLN 
 Subcapsular (ref) 170 — — 138 — — 
 Intraparenchymal 27 1.71 (0.86–3.42) 0.13 23 1.21 (0.55–2.70) 0.63 
 Both 122 2.35 (1.56–3.52) <0.0001 99 2.27 (1.34–3.84) 0.002 
Extranodal extension in SLN 
 Absent (ref) 282 — — 225 — — 
 Present 42 2.73 (1.75–4.27) <0.0001 35 1.86 (1.08–3.19) 0.025 
Number of positive non-SLNs 
 0 (ref) 278 — — 230 — — 
 1 13 2.68 (1.24–5.79) 0.012 12 2.28 (0.95–5.49) 0.07 
 ≥2 20 3.84 (2.17–6.78) <0.0001 18 2.53 (1.28–4.99) 0.007 
Maximum diameter of tumor deposit in non-SLNs, mm 
 No metastasis (ref) 278 — —  NI  
 ≤2.0 2.93 (1.19–7.22) 0.020    
 >2.0–5.0 2.53 (1.02–6.23) 0.044    
 >5.0 15 4.26 (2.27–8.01) <0.0001    
Microanatomic location of tumor deposit in non-SLNs 
 No metastasis/not both (ref) 286 — —  NI  
 Both subcapsular and intraparenchymal 25 3.55 (2.08–6.06) <0.0001    
Extranodal extension in non-SLNs 
 No metastasis (ref) 278 — —  NI  
 Absent 23 2.34 (1.28–4.28) 0.006    
 Present 10 8.14 (4.04–16.39) <0.0001    
CLND performed 
 No (ref) 17 — —  NI  
 Yes 303 0.86 (0.35–2.12) 0.75    
Adjuvant systemic therapy 
 None (ref) 188 — —  NI  
 Administered 135 0.68 (0.46–1.00) 0.053    

Abbreviations: ALM, acral lentiginous melanoma; CLND, completion lymph node dissection; HR, hazard ratio; LMM, lentigo maligna melanoma; NI, not included in the final model; NM, nodular melanoma; ref, reference; SSM, superficial spreading melanoma. The numbers in some groups do not total 336 because data were not available for some patients. Values in bold are statistically significant.

We also performed analyses using cutoff values of < 0.1 mm, 0.1–1.0 mm, and > 1.0 mm, as these strata have been used in several previously reported studies (16, 18, 19, 22). Patients with maximum diameter of tumor deposit in the SLN > 1.0 mm had significantly worse RFS than patients with maximum diameter of tumor deposit in the SLN < 1.0 mm (P = 0.018), and the association with MSS approached significance (P = 0.08; Supplementary Table S1A). On average, patients with maximum diameter of tumor deposit in the SLN > 1.0 mm had a significantly larger maximum diameter of tumor deposit in the non-SLN than patients with maximum diameter of tumor deposit in the SLN < 1.0 mm (P < 0.0001; Supplementary Table S1B).

Correlation of tumor burden parameters between SLNs and non-SLNs

Among patients with maximum diameter of tumor deposit in the SLN ≤ 2.0 mm, >2.0–5.0 mm, and >5.0 mm, non-SLN metastases were detected in 5 of 200 patients (3%), 10 of 63 patients (16%), and 20 of 57 patients (35%), respectively, and the mean maximum diameter of tumor deposit in the non-SLN was 0.09 mm [standard deviation (SD), 0.78 mm], 1.56 mm (SD, 5.07 mm), and 2.71 mm (SD, 8.11 mm), respectively (P < 0.0001; Table 4; Fig. 1A). A significant association was observed between microanatomic location of tumor deposit in SLN and microanatomic location of tumor deposit in non-SLN; the percentage of patients with non-SLN tumors in both subcapsular and intraparenchymal locations was higher for patients with SLN tumors in both subcapsular and intraparenchymal locations than for patients with SLN tumors in only one location (P < 0.0001; Table 4). Extranodal extension in non-SLNs was more common in patients with extranodal extension in the SLN than in patients without extranodal extension in the SLN (P = 0.003; Table 4). Among patients with a positive non-SLN (n = 35; Table 5; Fig. 1B), the proportion of patients with extranodal extension in non-SLNs trended toward higher in patients with extranodal extension (vs. without) in the SLN, although this difference was not statistically significant (P = 0.08; Table 5).

Table 4.

Relationship between SLN and non-SLN tumor burden parameters

SLN parametersNon-SLN parameters
Maximum diameter of tumor deposit, mm 
 n Mean (SD) Median (range) P value 
≤2.0 187 0.09 (0.78) 0 (0–10) <0.0001 
>2.0–5.0 60 1.56 (5.07) 0 (0–33)  
>5.0 54 2.71 (8.11) 0 (0–57)  
Microanatomic location 
 n No metastasis/not both locations Both locations P value 
Subcapsular only 171 168 (98%) 3 (2%) <0.0001 
Intraparenchymal only 25 23 (92%) 2 (8%)  
Both locations 111 91 (82%) 20 (18%)  
Extranodal extension 
 n No metastasis Absent Present P value 
Absent 275 249 (91%) 21 (8%) 5 (2%) 0.003 
Present 38 29 (76%) 4 (11%) 5 (13%)  
SLN parametersNon-SLN parameters
Maximum diameter of tumor deposit, mm 
 n Mean (SD) Median (range) P value 
≤2.0 187 0.09 (0.78) 0 (0–10) <0.0001 
>2.0–5.0 60 1.56 (5.07) 0 (0–33)  
>5.0 54 2.71 (8.11) 0 (0–57)  
Microanatomic location 
 n No metastasis/not both locations Both locations P value 
Subcapsular only 171 168 (98%) 3 (2%) <0.0001 
Intraparenchymal only 25 23 (92%) 2 (8%)  
Both locations 111 91 (82%) 20 (18%)  
Extranodal extension 
 n No metastasis Absent Present P value 
Absent 275 249 (91%) 21 (8%) 5 (2%) 0.003 
Present 38 29 (76%) 4 (11%) 5 (13%)  

NOTE: Values in bold are statistically significant.

Figure 1.

A, Correlation between maximum diameter of tumor deposit in SLN and non-SLN. B, Correlation between maximum diameter of tumor deposit in SLN and non-SLN in patients with positive non-SLN (n = 35).

Figure 1.

A, Correlation between maximum diameter of tumor deposit in SLN and non-SLN. B, Correlation between maximum diameter of tumor deposit in SLN and non-SLN in patients with positive non-SLN (n = 35).

Close modal
Table 5.

Relationship between SLN and non-SLN tumor burden parameters only in the patients with a positive non-SLN (n = 35)

SLN parametersNon-SLN parameters
Maximum diameter of tumor deposit, mm 
 n Mean (SD) Median (range) P value 
≤2.0 3.30 (3.88) 1.80 (0.7–10) 0.18 
>2.0–5.0 10 9.38 (9.31) 7.10 (0.4–33)  
>5.0 20 7.32 (12.15) 3.30 (0.8–57)  
Microanatomic location 
 n Not both locations, n (%) Both locations, n (%) P value 
Subcapsular only 6 (86) 1 (14) 0.33 
Intraparenchymal only 2 (100)  
Both locations 24 15 (63) 9 (38)  
Extranodal extension 
 n Absent Present P value 
Absent 26 21 (81%) 5 (19%) 0.08 
Present 4 (44%) 5 (56%)  
SLN parametersNon-SLN parameters
Maximum diameter of tumor deposit, mm 
 n Mean (SD) Median (range) P value 
≤2.0 3.30 (3.88) 1.80 (0.7–10) 0.18 
>2.0–5.0 10 9.38 (9.31) 7.10 (0.4–33)  
>5.0 20 7.32 (12.15) 3.30 (0.8–57)  
Microanatomic location 
 n Not both locations, n (%) Both locations, n (%) P value 
Subcapsular only 6 (86) 1 (14) 0.33 
Intraparenchymal only 2 (100)  
Both locations 24 15 (63) 9 (38)  
Extranodal extension 
 n Absent Present P value 
Absent 26 21 (81%) 5 (19%) 0.08 
Present 4 (44%) 5 (56%)  

In light of the results of the MSLT-II and DeCOG-SLT trials, patients with cutaneous melanoma with positive SLNs no longer routinely undergo CLND (1–5). In the DeCOG-SLT trial, the 3-year distant metastasis-free survival rate was 77.0% [90% confidence interval (CI), 71.9%–82.1%] in 233 patients assigned to the observation group and 74.9% (90% CI, 69.5%–80.3%) in 240 patients assigned to the CLND group [hazard ratio (HR), 1.03; 90% CI, 0.71–1.50; P = 0.87; refs. 1, 2]. The DeCOG-SLT trial excluded patients with melanoma of the head and neck, and the sample size was smaller than planned owing to slow participant accrual. The MSLT-II trial included a larger number of patients and included patients with melanoma of the head and neck. Results were similar to those of the DeCOG-SLT trial: the 3-year MSS rate was 86% ± 1.2% in the observation group and 86% ± 1.3% in the CLND group (P = 0.42; ref. 3). Physicians need to pay attention to the comparability of the settings between these trials and clinical practice. In these trials, 66% of patients had SLN metastases measuring 1 mm or smaller, and nodal ultrasonographic assessment of the SLN basin was performed more intensively, every 3 to 4 months. Therefore, from a survival standpoint, active surveillance after SLN biopsy is justified for most patients.

Recently, agents such as anti–PD-1 and BRAF/MEK inhibitors have been formally introduced as FDA-approved regimens in the adjuvant setting following randomized clinical trials demonstrating efficacy (27–30). Although patients with a positive SLN had to undergo CLND in these adjuvant therapy trials, the recent availability of such therapies has already entered the multidisciplinary management of patients with stage III disease who do not undergo CLND. But it is the survival analysis from MSLT2 and DeCOG-SLT that currently drives most decision-making; the impact of adjuvant therapy in this new era continues to be explored, and lessons will continue to be learned going forward. As such, SLN tumor burden may become a more practical prognostic factor than non-SLN positivity. The association between SLN tumor burden and non-SLN positivity was demonstrated in our prior study (17) and in those of others (reviewed in ref. 31). The present study shows that SLN tumor burden is also associated with non-SLN tumor burden and that SLN tumor burden is associated with prognosis. Our findings indicate that SLN tumor burden could be used as a clinically relevant prognostic for patients who do not undergo CLND; these observations are conceptually consistent with reports from the EORTC Melanoma Group that demonstrated prognostic stratification of patients with positive SLNs based on the SLN tumor burden (≤ 1.0 mm or > 1.0 mm) and the presence or absence of ulceration (6), as well as recent analyses from the International Melanoma Database and Discovery Platform that informed recent updates to the AJCC 8th edition melanoma staging system (7, 8).

SLN tumor burden has previously been studied as a potential predictor of non-SLN positivity and survival (15–22). Although previous studies defined SLN tumor burden in different ways, all of them classified SLN tumor burden on the basis of measurements of the size or the location of the tumor deposit in SLNs (31, 32). Our finding that larger maximum diameter of tumor deposit in SLN and non-SLN in univariate analysis was associated with survival (RFS and MSS) is concordant with previous studies (18–22). However, in our study, after adjustment for other factors, larger maximum diameter of tumor deposit in SLN or non-SLN was not an independent prognostic indicator. Determining the optimal cutoff points for the maximum diameter of SLN tumor deposits is one of the major challenges in this field. In the present study, cutoff points of 2.0 mm and 5.0 mm were selected. The cutoff point of 2.0 mm was chosen because the maximum diameter of SLN tumor deposit larger than 2.0 mm was associated with higher risk of non-SLN positivity in our previous study (17). In an international multicenter study (N = 1,085) patients with positive SLNs, maximum diameter of SLN tumor deposit larger than 1.0 mm was an important independent predictor of poor disease-free survival and MSS (22). Moreover, in the recent phase III trials of adjuvant therapy for patients at higher risk of recurrence, enrollment was limited to patients with stage IIIA disease with SLN tumor deposits larger than 1.0 mm in diameter (27, 28, 30, 33, 34). Therefore, a cutoff point of 1.0 mm might also be considered in future studies. With regard to the use of a larger cutoff point of 5.0 mm, Cutoff Finder software (26) identified 5.8 mm as the best cutoff point to stratify SLN-positive patients into 2 prognostic categories for MSS, and we rounded the cutoff point to 5.0 mm for ease of potential clinical use. In the current study, in patients with SLN tumor deposit more than 5.0 mm, 5-year MSS was 42%, and is similar to the 5-year overall survival rate of 43% for patients with cutaneous melanoma and clinically detected nodal involvement in a previous large study of patients with stage III disease from the 7th edition AJCC melanoma database (7). The prognosis of patients with SLN tumor burden of more than 5.0 mm in diameter was similar to those of patients with clinically detected nodal involvement.

Regarding the SLN tumor burden parameters other than maximum diameter, as expected, we found that the presence of both subcapsular and intraparenchymal disease and extranodal extension were associated with poor RFS and MSS. The maximum diameter of SLN tumor deposit was larger in patients with both subcapsular and intraparenchymal disease than in patients with disease limited to one location; it was larger in patients with extranodal extension than in patients without extranodal extension. However, poor survival outcome in patients with tumor deposit in both locations or with extranodal extension might not be due to just a difference in the tumor burden because those parameters were independently associated with RFS and MSS in multivariable models (Tables 2 and 3). Therefore, even though determination of the maximum diameter is relatively straightforward and widely used, microanatomic locations and extranodal extension should also be measured as SLN tumor burden parameters predictive of prognosis, particularly as integrated risk models continue to be refined (9).

A potential limitation of this study is the use of different sectioning and staining procedures for SLNs and non-SLNs. The more intense examination of SLNs could explain why metastatic tumor cells were more likely to be found in SLNs than in non-SLNs. However, the study reflects the current standard operating procedures for examining SLN and non-SLN specimens. Previous studies have reported that evaluation of non-SLNs by ancillary studies such as reverse transcriptase-polymerase chain reaction (RT-PCR) analysis may increase the rate of positivity by up to 15% (35, 36). Although this increased sensitivity for detecting small volume or single-cell metastasis may not affect patient outcome in terms of MSS or overall survival, whether presence of such small volume metastasis in non-SLN may be associated with increased rates of future regional recurrence remains to be established.

In conclusion, in patients with cutaneous melanoma who undergo CLND, tumor burden in SLNs is associated with tumor burden in non-SLNs and is associated with survival. Tumor burden in SLNs may be used to stratify SLN-positive patients who do not undergo CLND in terms of their likelihood of nodal recurrence or expected benefit from adjuvant systemic therapies. SLN tumor burden parameters, which provide accurate prognostic information independent of non-SLN status, should be considered for incorporation into future staging systems and/or integrated risk models to improve clinical decision-making.

M.T. Tetzlaff is an advisory board member/unpaid consultant for Myriad Genetics, Seattle Genetics, Novartis LLC, and NanoString. M. Ross is an employee/paid consultant for Amgen and Merck, and reports receiving commercial research grants from Amgen and speakers bureau honoraria from Merck, Amgen, Novartis, and CastleBiosciences. J.E. Gershenwald is an employee/paid consultant for Merck, Novartis, and Bristol-Myers Squibb. V.G. Prieto is an employee/paid consultant for Myriad. No potential conflicts of interest were disclosed by the other authors.

Conception and design: K. Namikawa, P.P. Aung, M. Ross, V.G. Prieto

Development of methodology: K. Namikawa, P.P. Aung

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): K. Namikawa, P.P. Aung, M.T. Tetzlaff, C.A. Torres-Cabala, J.L. Curry, D. Ivan, J.E. Gershenwald, V.G. Prieto

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P.P. Aung, D.R. Milton, M.T. Tetzlaff, J.E. Gershenwald, V.G. Prieto

Writing, review, and/or revision of the manuscript: K. Namikawa, P.P. Aung, D.R. Milton, C.A. Torres-Cabala, P. Nagarajan, D. Ivan, J.E. Gershenwald, V.G. Prieto

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): C.A. Torres-Cabala

Study supervision: P.P. Aung, V.G. Prieto

P.P. Aung was supported by an Institutional Start-up Funding from The University of Texas MD Anderson Cancer Center (MD Anderson), an Institutional Research Grant from The University of Texas MD Anderson Cancer Center (MD Anderson), and Young Investigator award from Melanoma Research Alliance (MRA; PI: P.P. Aung). The authors thank Ms. Stephanie Deming, ELS, Senior Scientific Editor, Research Medical Library, The University of Texas MD Anderson Cancer Center, for assistance with editing of this manuscript.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Leiter
U
,
Stadler
R
,
Mauch
C
,
Hohenberger
W
,
Brockmeyer
N
,
Berking
C
, et al
Complete lymph node dissection versus no dissection in patients with sentinel lymph node biopsy positive melanoma (DeCOG-SLT): a multicentre, randomised, phase 3 trial
.
Lancet Oncol
2016
;
17
:
757
67
.
2.
Leiter
UM
,
Stadler
R
,
Mauch
C
,
Hohenberger
W
,
Brockmeyer
N
,
Berking
C
, et al
Final analysis of DECOG-SLT trial: survival outcomes of complete lymph node dissection in melanoma patients with positive sentinel node
.
J Clin Oncol
2018
;
36
(
15_suppl
):
9501
-.
3.
Faries
MB
,
Thompson
JF
,
Cochran
AJ
,
Andtbacka
RH
,
Mozzillo
N
,
Zager
JS
, et al
Completion dissection or observation for sentinel-node metastasis in melanoma
.
N Engl J Med
2017
;
376
:
2211
22
.
4.
Wong
SL
,
Faries
MB
,
Kennedy
EB
,
Agarwala
SS
,
Akhurst
TJ
,
Ariyan
C
, et al
Sentinel lymph node biopsy and management of regional lymph nodes in melanoma: American Society of Clinical Oncology and Society of Surgical Oncology Clinical Practice Guideline Update
.
J Clin Oncol
2018
;
36
:
399
413
.
5.
Wong
SL
,
Faries
MB
,
Kennedy
EB
,
Agarwala
SS
,
Akhurst
TJ
,
Ariyan
C
, et al
Sentinel lymph node biopsy and management of regional lymph nodes in melanoma: American Society of Clinical Oncology and Society of Surgical Oncology Clinical Practice Guideline Update
.
Ann Surg Oncol
2018
;
25
:
356
77
.
6.
Verver
D
,
van Klaveren
D
,
van Akkooi
ACJ
,
Rutkowski
P
,
Powell
B
,
Robert
C
, et al
Risk stratification of sentinel node-positive melanoma patients defines surgical management and adjuvant therapy treatment considerations
.
Eur J Cancer
2018
;
96
:
25
33
.
7.
Gershenwald
JE
,
Scolyer
RA
,
Hess
KR
,
Sondak
VK
,
Long
GV
,
Ross
MI
, et al
Melanoma staging: evidence-based changes in the American Joint Committee on Cancer eighth edition cancer staging manual
.
CA Cancer J Clin
2017
;
67
:
472
92
.
8.
Gershenwald
JE
,
Scolyer
RA
,
Hess
KR
. 
Melanoma of the skin
.
In:
Amin
MB
,
editor
.
AJCC cancer staging manual
. 8th ed.
Switzerland
:
Springer
; 
2017
.
p
563
85
.
9.
Gershenwald
JE
,
Scolyer
RA
. 
Melanoma staging: American Joint Committee on Cancer (AJCC) 8th Edition and Beyond
.
Ann Surg Oncol
2018
;
25
:
2105
10
.
10.
Balch
CM
,
Gershenwald
JE
,
Soong
SJ
,
Thompson
JF
,
Ding
S
,
Byrd
DR
, et al
Multivariate analysis of prognostic factors among 2,313 patients with stage III melanoma: comparison of nodal micrometastases versus macrometastases
.
J Clin Oncol
2010
;
28
:
2452
9
.
11.
Egger
ME
,
Callender
GG
,
McMasters
KM
,
Ross
MI
,
Martin
RC
 2nd
,
Edwards
MJ
, et al
Diversity of stage III melanoma in the era of sentinel lymph node biopsy
.
Ann Surg Oncol
2013
;
20
:
956
63
.
12.
Morton
DL
,
Thompson
JF
,
Cochran
AJ
,
Mozzillo
N
,
Elashoff
R
,
Essner
R
, et al
Sentinel-node biopsy or nodal observation in melanoma
.
N Engl J Med
2006
;
355
:
1307
17
.
13.
Morton
DL
,
Thompson
JF
,
Cochran
AJ
,
Mozzillo
N
,
Nieweg
OE
,
Roses
DF
, et al
Final trial report of sentinel-node biopsy versus nodal observation in melanoma
.
N Engl J Med
2014
;
370
:
599
609
.
14.
Namikawa
K
,
Aung
PP
,
Gershenwald
JE
,
Milton
DR
,
Prieto
VG
. 
Clinical impact of ulceration width, lymphovascular invasion, microscopic satellitosis, perineural invasion, and mitotic rate in patients undergoing sentinel lymph node biopsy for cutaneous melanoma: a retrospective observational study at a comprehensive cancer center
.
Cancer Med
2018
;
7
:
583
93
.
15.
Dewar
DJ
,
Newell
B
,
Green
MA
,
Topping
AP
,
Powell
BW
,
Cook
MG
. 
The microanatomic location of metastatic melanoma in sentinel lymph nodes predicts nonsentinel lymph node involvement
.
J Clin Oncol
2004
;
22
:
3345
9
.
16.
van Akkooi
AC
,
de Wilt
JH
,
Verhoef
C
,
Schmitz
PI
,
van Geel
AN
,
Eggermont
AM
, et al
Clinical relevance of melanoma micrometastases (<0.1 mm) in sentinel nodes: are these nodes to be considered negative?
Ann Oncol
2006
;
17
:
1578
85
.
17.
Gershenwald
JE
,
Andtbacka
RH
,
Prieto
VG
,
Johnson
MM
,
Diwan
AH
,
Lee
JE
, et al
Microscopic tumor burden in sentinel lymph nodes predicts synchronous nonsentinel lymph node involvement in patients with melanoma
.
J Clin Oncol
2008
;
26
:
4296
303
.
18.
van Akkooi
AC
,
Nowecki
ZI
,
Voit
C
,
Schafer-Hesterberg
G
,
Michej
W
,
de Wilt
JH
, et al
Sentinel node tumor burden according to the Rotterdam criteria is the most important prognostic factor for survival in melanoma patients: a multicenter study in 388 patients with positive sentinel nodes
.
Ann Surg
2008
;
248
:
949
55
.
19.
van der Ploeg
AP
,
van Akkooi
AC
,
Schmitz
PI
,
Koljenovic
S
,
Verhoef
C
,
Eggermont
AM
. 
EORTC Melanoma Group sentinel node protocol identifies high rate of submicrometastases according to Rotterdam Criteria
.
Eur J Cancer
2010
;
46
:
2414
21
.
20.
Namikawa
K
,
Yamazaki
N
,
Nakai
Y
,
Ihn
H
,
Tomita
Y
,
Uhara
H
, et al
Prediction of additional lymph node positivity and clinical outcome of micrometastases in sentinel lymph nodes in cutaneous melanoma: a multi-institutional study of 450 patients in Japan
.
J Dermatol
2012
;
39
:
130
7
.
21.
Nagaraja
V
,
Eslick
GD
. 
Is complete lymph node dissection after a positive sentinel lymph node biopsy for cutaneous melanoma always necessary? A meta-analysis
.
Eur J Surg Oncol
2013
;
39
:
669
80
.
22.
van der Ploeg
AP
,
van Akkooi
AC
,
Haydu
LE
,
Scolyer
RA
,
Murali
R
,
Verhoef
C
, et al
The prognostic significance of sentinel node tumour burden in melanoma patients: an international, multicenter study of 1539 sentinel node-positive melanoma patients
.
Eur J Cancer
2014
;
50
:
111
20
.
23.
Scolyer
RA
,
Judge
MJ
,
Evans
A
,
Frishberg
DP
,
Prieto
VG
,
Thompson
JF
, et al
Data set for pathology reporting of cutaneous invasive melanoma: recommendations from the international collaboration on cancer reporting (ICCR)
.
Am J Surg Pathol
2013
;
37
:
1797
814
.
24.
Prieto
VG
. 
Sentinel lymph nodes in cutaneous melanoma
.
Clin Lab Med
2011
;
31
:
301
10
.
25.
Prieto
VG
. 
Sentinel lymph nodes in cutaneous melanoma: handling, examination, and clinical repercussion
.
Arch Pathol Lab Med
2010
;
134
:
1764
9
.
26.
Budczies
J
,
Klauschen
F
,
Sinn
BV
,
Gyorffy
B
,
Schmitt
WD
,
Darb-Esfahani
S
, et al
Cutoff Finder: a comprehensive and straightforward Web application enabling rapid biomarker cutoff optimization
.
PLoS One
2012
;
7
:
e51862
.
27.
Long
GV
,
Hauschild
A
,
Santinami
M
,
Atkinson
V
,
Mandala
M
,
Chiarion-Sileni
V
, et al
Adjuvant dabrafenib plus trametinib in stage III BRAF-mutated melanoma
.
N Engl J Med
2017
;
377
:
1813
23
.
28.
Hauschild
A
,
Dummer
R
,
Schadendorf
D
,
Santinami
M
,
Atkinson
V
,
Mandala
M
, et al
Longer follow-up confirms relapse-free survival benefit with adjuvant dabrafenib plus trametinib in patients with resected BRAF V600-mutant stage III melanoma
.
J Clin Oncol
2018
:
JCO1801219
.
29.
Weber
J
,
Mandala
M
,
Del Vecchio
M
,
Gogas
HJ
,
Arance
AM
,
Cowey
CL
, et al
Adjuvant nivolumab versus ipilimumab in resected stage III or IV melanoma
.
N Engl J Med
2017
;
377
:
1824
35
.
30.
Eggermont
AMM
,
Blank
CU
,
Mandala
M
,
Long
GV
,
Atkinson
V
,
Dalle
S
, et al
Adjuvant pembrolizumab versus placebo in resected stage III melanoma
.
N Engl J Med
2018
;
378
:
1789
801
.
31.
Madu
MF
,
Wouters
MW
,
van Akkooi
AC
. 
Sentinel node biopsy in melanoma: current controversies addressed
.
Eur J Surg Oncol
2017
;
43
:
517
33
.
32.
Prieto
VG
. 
Sentinel lymph nodes in cutaneous melanoma
.
Clin Lab Med
2017
;
37
:
417
30
.
33.
Eggermont
AM
,
Chiarion-Sileni
V
,
Grob
JJ
,
Dummer
R
,
Wolchok
JD
,
Schmidt
H
, et al
Adjuvant ipilimumab versus placebo after complete resection of high-risk stage III melanoma (EORTC 18071): a randomised, double-blind, phase 3 trial
.
Lancet Oncol
2015
;
16
:
522
30
.
34.
Eggermont
AM
,
Chiarion-Sileni
V
,
Grob
JJ
,
Dummer
R
,
Wolchok
JD
,
Schmidt
H
, et al
Prolonged survival in stage III melanoma with ipilimumab adjuvant therapy
.
N Engl J Med
2016
;
375
:
1845
55
.
35.
Wrightson
WR
,
Wong
SL
,
Edwards
MJ
,
Chao
C
,
Conrad
AJ
,
Albrecht
J
,
Viar
V
,
McMasters
KM
. 
Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of nonsentinel nodes following completion lymphadenectomy for melanoma
.
J Surg Res
2001
;
98
:
47
51
.
36.
Holm-Rasmussen
EV
,
Jensen
MB
,
Balslev
E
,
Kroman
N
,
Tvedskov
TF
. 
Non–sentinel lymph node metastases associated with isolated breast cancer cells in the sentinel node
.
Breast Cancer Res Treat
2019
;
175
:
713
9
.