American Joint Cancer Committee (AJCC) staging system for primary cutaneous melanoma (8 th Edition) and principles of sentinel lymph node evaluation Emphasis on concise and accurate reporting of primary and metastatic melanoma for effective risk stratification and prognosis Michael T. Tetzlaff MD, PhD Associate Professor Department of Pathology, Section of Dermatopathology Department of Translational and Molecular Pathology The University of Texas MD Anderson Cancer Center Executive Officer Translational Research Program The Alliance for Clinical Trials
Outline 8th Edition of AJCC staging system: N-category Lymph node metastases and satellitosis: criteria for pathologic N-category Evaluating the sentinel lymph node Guidelines and pitfalls
T = Primary tumor Breslow (Tumor) thickness Mitotic rate (7 th Edition) Ulceration N = Regional metastasis Lymph node Microscopic (a) Macroscopic (b) Skin/subcutaneous tissue (c) In-transit metastasis/ satellites/ microsatellites TNM staging M = Distant metastasis Lymph node Skin/subcutaneous tissue Visceral metastasis Pulmonary Non-pulmonary Serum LDH Local Nodal Systemic
Melanoma staging N-category: number, clinical extent and satellites Clinically occult [pnxa] versus clinically evident [pnxb] Presence of a microsatellite, satellite or in-transit metastasis modifies to [pnxc]
N-category determined by sentinel node evaluation in most cases If all nodes node are detected microscopically, then clinically occult and [pnxa] If at least one node is detected clinically and others are detected microscopically, then [pnxb] If satellites present, then [pnxc]
Satellite, Microsatellite and In-transit metastases pnc Microsatellite microscopically identified deposit of tumor separated from the primary lesion by normal tissue (no intervening fibrosis or inflammation) Satellite clinically identified separate nodule of tumor separate from primary tumor < 2 cm. In-transit metastasis clinically identified nodule of tumor > 2 cm from the primary tumor between the primary tumor and regional lymph node.
Pitfalls of satellitosis Intravascular deposit (IHC informative) Lesion contiguous on deeper sections Incidental nevus? Mart-1/K-67 8 th Edition of the AJCC No minimum size threshold or minimum distance from the primary lesion. Microsatellites close to the primary lesion should have additional sections cut to exclude connection.
N-category determined by sentinel node evaluation in most cases Most lymph node metastases are detected by examination of the sentinel lymph node. Regions of skin first drain to an initial node (the sentinel node) Different regions drain to different sentinel nodes Most likely node(s) to contain metastatic disease and Gershenwald
Advantages of a sentinel lymph node? Minimally invasive Provides accurate staging: Identification of the critical SLN Systematic histologic and immunohistochemical assessment maximizes sensitivity More sensitively identify the target population that will benefit from adjuvant therapy Provide more rational basis for selection of therapeutic regional node dissection
N-category determined by sentinel node evaluation in most cases Most lymph node metastases are detected by examination of the sentinel lymph node. Sentinel lymph nodes in melanoma About 20% positive 16% initial H&E 4% additional sections/ihc <5% with extracapsular extension
Indications for SLN Examination Breslow 1.00 mm (+ deep margin) 0.80 mm now generally accepted Ulceration >1 mitotic figure/mm 2 In addition, high risk features (MDACC) Clark level IV Regression (particularly when extensive) Vascular invasion Satellitosis
Surgical Technique Injection of dye: Blue Radioactive Geiger counter (10% or x2 higher than surrounding tissue) One or several nodes and Gershenwald
Lymphatic Mapping: Dye and colloid Initial scan Rescan basin First SLN Additional SLN and Gershenwald
Sentinel lymph node processing Bisection vs. breadloafing No frozen sections: Sub-optimal morphology Loss of subcapsular region Loss of tissue Entirely submitted (minimal disease) Histologic examination (PCR?)
No Frozen sections Loss of subcapsular area
Melanoma cells or macrophages? Morphology is difficult to interpret
Macrophages or melanoma? Morphology is difficult to interpret HMB45
Sentinel lymph node evaluation
Subcapsular cluster of cells in the lymph node
Subcapsular cluster of cells in the lymph node
Original melanoma SLN
Sentinel lymph node evaluation Positive on H&E with extracapsular extension Defined as the presence of nodal metastasis extending through the lymph node capsule into adjacent tissues usually seen as microscopic extension of metastatic melanoma into the perinodal adipose tissue.
Sentinel lymph node evaluation Positive by IHC Pan-Melanoma (HMB-45, tyrosinase, Mart-1) HMB-45
SLN: Clinical significance of histologic examination why do IHC? 479 patients (1389 SLN) from 10/97-1/01 Median Breslow: 2.09 mm Initial slides: + in 14% patients After immuno/additional sections: Additional 6% patients with positive SLN Prieto and Clark. Ann Diagn Pathol, 2002
Sentinel lymph node evaluation Positive by IHC Metastatic melanoma Pan-Melanoma (HMB-45, tyrosinase, Mart-1) Advised to confirm with HMB-45 IHC alone
What is the significance of isolated cells in the SLN? Criteria for positive lymph node Any tumor deposit size Isolated tumor cell on IHC acceptable H&E confirmation no longer required Positive for at least one melanocytic marker (HMB45, Melan-A/Mart-1) Cells have malignant morphology (similar to primary tumor) and in the correct location (subcapsular versus intracapsular) No unequivocal evidence for a minimum size threshold of microscopic tumor burden that is applied to define node positive disease.
Immunohistochemistry: The challenge of focal positivity Pan-Melanoma (HMB-45, tyrosinase, Mart-1) Anti-HMB-45
Immunohistochemistry: The challenge of focal positivity Pan-Melanoma (HMB-45, tyrosinase, Mart-1) Anti-HMB-45
Sentinel lymph node evaluation Pitfalls of IHC Melan-A positive macrophages in SLN Advised to confirm with HMB-45 IHC alone
Sentinel lymph node evaluation Pitfalls of IHC Melan-A positive macrophages in SLN Advised to confirm with HMB-45 IHC alone
Is there a threshold: How many cells for positive SLN? 56-yo man Pigmented nodule right forearm Tattoo (American beer) Pigment bleaching Biopsy and wide local excision
Diagnosis Melanoma Superficial Spreading Type Clark level IV Breslow 11 mm Associated pigment (tattoo)
HMB45
Clinical Follow-Up Sentinel lymph node (1/9) Two isolated cells Completion lymphadenectomy (-) Two years later presented with widespread metastases Died shortly thereafter
Desmoplastic melanoma
SLN
SLN S100
Sentinel lymph node evaluation Pitfalls of IHC: Capsular nevus Benign melanocytes in lymph nodes Up to 20% of lymphadenectomies for melanoma More common in MM associated with nevi Capsule (rarely intraparenchymal) Biddle et al; AJSP 2003
Sentinel lymph node evaluation Negative by IHC: Mimics and pitfalls Capsular nevus
Sentinel lymph node evaluation Negative by IHC: Mimics and pitfalls Pan-Melanoma (HMB-45, tyrosinase, Mart-1) HMB-45 HMB-45
Intraparenchymal Capsular Nevus
Intraparenchymal Capsular Nevus MART1 Pan-Melanoma (HMB-45, tyrosinase, Mart-1)
HMB45 HMB45 Intraparenchymal Capsular Nevus
Ki67 (MIB1) Ki67 (MIB1) Intraparenchymal Capsular Nevus
Algorithm for Capsular Nevi Location Capsular Parenchymal HMB45 Ki67 HMB45 Ki67 Comparison with previous Positive Melanoma Negative Nevus
Prognostic significance of SLN-positivity 580 patients, SLN + in 85 pts (15%) - in 495 pts (85%) SLN status: most significant prognostic factor (disease-free and disease-specific survival), uni- and multivariate analysis Tumor thickness and ulceration in SLNnegative patients Gershenwald et al, J Clin Oncol 1999
Prognostic Factors Influencing Disease-Specific Survival Multiple covariate Prognostic Factor Univariate Hazard Ratio p-value Age NS - NS Sex NS - NS Axial location.03 - NS Tumor thickness <.0001 1.1.04 Clark level > III.001 2.3.01 Ulceration <.0001 3.3 <.0001 Total Positive Nodes <.0001 0 1.0 1 or 2 2.3 3+ 12.0 Gershenwald et al., ASCO, 2000
What other variables can we provide to inform prognosis? 1,417 pts (2,203 SLN), 1990-2002 Breadloafed and examined with routine hematoxylin and eosin Entirely submitted Negative cases were examined with immunohistochemistry
What other variables can we provide to inform prognosis? Total number of lymph nodes Number of lymph nodes involved Size of tumor deposit (s): Largest deposit Two dimensions Addition of all deposits Location of deposit: Subcapsular/ Intraparenchymal Both Extracapsular extension
Measure the largest focus in two dimensions
Measure the largest focus in two dimensions Subcapsular Subcaps/ Intramedullary Multiple foci Extracaps extension
Disease-Specific Survival by SLN Status Proportion Surviving 1.0 Negative (n=1180) 0.8 Positive (n=237) 0.6 0.4 0.2 0.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Years
Extent of SLN burden correlates with DSS Total number of positive nodes Proportion Surviving 1.0 SLN+ patients only 0.8 0.6 0.4 0.2 0 1 (n=231) 2-3 (n=86) 3+ (n=42) Log-rank: 22.7 P<0.0001 0 2 4 6 8 10 12 Years
Proportion Surviving DSS by Tumor Burden: Largest Focus 1.0 0.8 0.6 0.4 0.2 0 <=2mm (n=207) >2mm (n=115) Log-rank: 47 P<0.0001 0 2 4 6 8 10 12 Years
Proportion Surviving Tumor Location in SLN drives DSS 1.0 0.8 Subcapsular 0.6 0.4 Intramedullary Both 0.2 0 Sub-capsular (n=193) Intramedullary (n=31) Both (n=84) Log-rank: 47 P<0.0001 0 2 4 6 8 10 12 Years
Proportion Surviving Ulceration drives DSS even among SLN Positive Patients 1.0 SLN+ patients only 0.8 0.6 0.4 0.2 0 No Ulceration (n=205) Ulceration (n=154) Log-rank: 32.2 P<0.0001 0 2 4 6 8 10 12 Years
Microscopic Tumor Burden in SLN correlates with increased risk of additional nodal disease Variable Median (mean) DSS (p-value) Number SLN metastatic foci 1.0 (2.2) 0.002 SLN tumor size area (mm 2 ) 1.4 (13.3) <0.001 Largest focus in SLN (mm) 1.6 (3.0) <0.001 SLN tumor location <0.001 Subcapsular 65 Intramedullary 11 Both 24 Extracapsular extension 0.012 SLN tumor burden is an important predictor of DSS in patients with microscopic nodal disease Location of metastatic deposit (subcapsular vs. parenchymal) and size of largest focus should be reported when examining SLN from patients with cutaneous melanoma
Proportion Surviving Microscopic Stage III Melanoma Predictive Model SLN + Patients Only (Tumor burden, Ulceration, # positive LNs) 1.0 0.8 0.6 0.4 0.2 0 Low risk (n=126) Intermediate risk (n=105) High risk (n=91) Log-rank: 55.4 P<0.0001 0 2 4 6 8 10 12 Years IIIA IIIB IIIC Low risk SLN focus<= 2 mm No ulceration 1-2 total LNs + Intermediate risk ulceration OR SLN focus > 2 mm High risk 3 total pos. LNs OR ulceration AND SLN focus > 2 mm
Polymerase Chain Reaction (PCR) evaluation of the SLN Detection of mrna associated with melanocytic differentiation: Tyrosinase gp100 (HMB45) MART1 Not testing a melanomaspecific mrna Thus, PCR does not reliably distinguish between capsular nevi and metastatic melanoma
PCR as a prognostic factor Because PCR does not distinguish between capsular nevi and metastatic melanoma, it has limited use in the clinical setting. Romanini et al. Ann Oncol 2005; 16: 1832 Giese et al. JID 2005; 124: 633
Template for reporting lymph nodes Number of lymph nodes (SLN and non-sln) Number containing metastasis Tumor location Subcapsular Intraparenchymal Tumor size Reported in mm Measured gross exam or slide Currently NOT a component of the staging system but envisioned to guide future prognostic models and possibly direct how (neo)adjuvant therapy is deployed Extracapsular extension 5.3 x 1.8 mm
Self-Assessment Cases 43-yo man 2.00 mm invasive ulcerated melanoma of the left shoulder HMB45
Your diagnosis is: 1. Capsular nevus 2. Metastatic melanoma 3. Tattoo 4. Dermatopathic lymphadenopathy
Your diagnosis is: 1. Capsular nevus 2. Metastatic melanoma 3. Tattoo 4. Dermatopathic lymphadenopathy
Self-Assessment Cases 29 F pregnant Intermediate thickness melanoma
MART1 MART1 and Ki67
Your diagnosis is: 1. Capsular nevus 2. Metastatic melanoma 3. Dermatopathic lymphadenopathy MART1 and Ki67
Your diagnosis is: 1. Capsular nevus 2. Metastatic melanoma 3. Dermatopathic lymphadenopathy MART1 and Ki67
Self-Assessment Cases 24 woman with invasive melanoma of the right foot: 1.45 mm, 3 mitotic figures/mm 2
HMB45
Your diagnosis is: 1. Capsular nevus 2. Metastatic melanoma 3. Dermatopathic lymphadenopathy HMB45
Your diagnosis is: 1. Capsular nevus 2. Metastatic melanoma 3. Dermatopathic lymphadenopathy HMB45
Thank you Victor G. Prieto MD PhD Jonathan L. Curry MD Carlos A. Torres-Cabala MD PhD Priya Nagarajan MD PhD Phyu Aung MD PhD Doina Ivan MD Jeff Gershenwald MD