AJCC Melanoma Staging 2018 - Pathology · 75% of all skin cancer deaths are from melanoma. 1 in 50 Americans will develop melanoma in their lifetime. Melanoma is 2nd most common cancer

HARVARDMEDICAL SCHOOL

MASSACHUSETTSGENERAL HOSPITALDERMATOPATHOLOGY

AJCC Melanoma Staging 2018

Lyn McDivitt Duncan, MDProfessor of Pathology, Harvard Medical SchoolChief, MGH Dermatopathology Unit

75% of all skin cancer deaths are from melanoma

1 in 50 Americans will develop melanoma in their lifetime

Melanoma is 2nd most common cancer diagnosis 17-29 y age group

Every hour someone in the U.S. will die from melanoma

The NumbersOver 90,000 new cases of melanoma in the U.S. in 2017Over 9, 000 melanoma deaths in U.S. in 2017

75

50

2

1

Tumor (T), Node (N), and Metastasis (M) categories and stage groupings

• informs prognostic assessment • Informs clinical decision making• facilitates centralized cancer registry reporting • facilitates the design, conduct, and analysis of clinical trials

Lymphatic mapping and Sentinel lymph node (SLN) biopsy

Major new classes of effective systemic therapeutic agents • immunotherapies (eg, checkpoint inhibitors against CTLA‐4 and/or PD‐1)• molecularly targeted antitumor therapies (eg, BRAF inhibitors +/- MEK inhibitors)

Melanoma Staging 2018 – Clinically Relevant Stratification

Histopathological factors in Melanoma

Primary melanoma:tumor thicknessmitosesulcerationmicroscopic satelliteslymphovascular invasiontumor infiltrating lymphocytesregression neural invasionmarginsmelanoma subtype

Sentinel lymph nodes:presence of metastasisnumber of positive nodessize of metastasisdetection techniquelocation of metastasis

Histopathological factors in AJCC stagingPrimary melanoma:

tumor thicknessulcerationmicroscopic satellitesmitoseslymphovascular invasiontumor infiltrating lymphocytesregression neural invasionmarginsmelanoma subtype

Sentinel lymph nodes:presence of metastasisnumber of positive nodessize of metastasisdetection techniquelocation of metastasis

AJCC Melanoma Staging 2018, 8th edition

1) tumor thickness recorded to nearest 0.1 mm, not 0.01 mm

2) T1a and T1b revised (T1a, <0.8 mm no ulceration; T1b, 0.8‐1.0 mm or <0.8 mm with ulceration) mitotic rate no longer a T category criterion

3) pathological (but not clinical) stage IA is revised to include T1b N0 M0 (formerly pathologic stage IB)

4) N category “microscopic” and “macroscopic” redefined “clinically occult” (SLN+) and “clinically apparent”

5) prognostic stage III groupings increased from 3 to 4 subgroups (stages IIIA‐IIID)

6) definitions of N subcategories are revised, microsatellites, satellites, or in‐transit metastases now N1c, N2c, or N3c based on the number of tumor‐involved regional lymph nodes

7) descriptors are added to each M1 subcategory designation for lactate dehydrogenase (LDH) level (LDH elevation no longer upstages to M1c); and

8) new M1d designation for central nervous system metastases

46, 000 patients, 10 centers world wide

AJCC Melanoma Staging 2018, 8th edition

1) tumor thickness recorded to nearest 0.1 mm, not 0.01 mm

2) T1a and T1b revised (T1a, <0.8 mm no ulceration; T1b, 0.8‐1.0 mm or <0.8 mm with ulceration) mitotic rate no longer a T category criterion

3) pathological (but not clinical) stage IA is revised to include T1b N0 M0 (formerly pathologic stage IB)

4) N category “microscopic” and “macroscopic” redefined “clinically occult” (SLN+) and “clinically apparent”

5) prognostic stage III groupings increased from 3 to 4 subgroups (stages IIIA‐IIID)

6) definitions of N subcategories are revised, microsatellites, satellites, or in‐transit metastasesnow N1c, N2c, or N3c based on the number of tumor‐involved regional lymph nodes

7) descriptors are added to each M1 subcategory designation for lactate dehydrogenase (LDH) level (LDH elevation no longer upstages to M1c); and

8) new M1d designation for central nervous system metastases

46, 000 patients, 10 centers world wide

Primary Melanoma Thickness (Breslow 1970)

Gershenwald, JE, Scolyer, RA, Hess, KR. et al. Melanoma of the Skin. In Amin, MB, Edge, SB, Greene, FL, et al. (Eds.)

AJCC Cancer Staging Manual. 8th Ed. New York: Springer 2017

TX: primary tumor thickness cannot be assessed (e.g. diagnosis by curettage)

T0: no evidence of primary tumor (e.g. unknown primary or completely regressed melanoma)

Tis: melanoma in situ

T1: < or = 1.0 mm

T2: 1.1 – 2.0 mm

T3: 2.1 – 4.0 mm

T4: > 4.0 mm

Tumor thickness measurements now are recorded to the nearest 0.1mm, not the nearest 0.01 mm. Melanomas in the range of 0.75 to 0.84 mm are reported as 0.8 mm, hence T1b.



Mitotic Activity

mitotic figures / mm2 8 year survival*

0 95%

< 6 79%

>6 38%

Tumor mitotic rate was removed as a staging criterion but remains an overall important prognostic factor that should be recorded for all patients with T1-T4 primary cutaneous melanoma.

Mitotic Activity

AJCC Stage I and II: Survival curves by mitoses per mm2

Gershenwald JE, et al. Cancer J Clin 2017;67:472-92

Data from AJCC 8th edition database

Ulceration

Lymphovascular Invasion

D240/S100

Lymphovascular invasion, AJCC stage and time to first metastasis

Xu X et al. Clin Cancer Res 2012

IA

IIB/C + LVIIIA + LVI

IIA no LVI

IIB/C no LVI

IB + LVI

IB no LVI

Anatomic level of invasion: Clark

Tumor infiltrating lymphocytes (TILs)Brisk

Non-brisk

“TILs are present throughout the substance of the vertical growth phase or present infiltrating across the entire base of the vertical growth phase.”

“TILs are present in one or more foci of the vertical growth phase”

TILs – “absent”

Overall survival for patients with localized (Stage I and II) melanoma when segregated by lymphoid infiltrates in the primary tumor

TIL’s 10 ys* 5 ys#brisk 55% 77%non-brisk 43% 53%absent 27% 37%

* Clemente CG, Mihm MC,Jr, Bufalino R, et al. Cancer 1996# Clark WH, Jr., Elder DE, Guerry DI, et al. JNCI 1989

Tumor Infiltrating Lymphocytes

Regression

D240/S100

• Thickness measured to nearest 0.1 mm (not 0.01 mm)– 0.75 mm = 0.8 mm

• Mitoses*: reported but not used in staging• Clark’s level: reported but not used in staging• Difference between T1a and T1b now based on thickness*

– T1a: < 0.8 mm without ulceration– T1b: < 0.8 mm with ulceration

0.8 – 1.0 mm with or without ulceration

AJCC Staging (8th Edition) Summary of Changes (T)

*Mitoses represent a significant prognostic factor across all tumor thicknesses



AJCC Stage I and II Melanoma Survival (SLN+ excluded)Gershenwald JE, et al. Cancer J Clin 2017;67:472-92

AJCC Stage

TNM T stage 5 year survival

0 Tis N0 M0 IA T1a N0 M0

T1b < 0.8 mm, no ulcer < 0.8 with ulcer 0.8-1.0 with or without ulcer

97%

IB T2a N0 M0 1-2 mm no ulcer 91% IIA T2b N0 M0

T3a 1-2 mm + ulcer 2-4 mm no ulcer

81%

IIB T3b N0 M0 T4a

2-4 mm + ulcer > 4 mm no ulcer

68%

IIC T4b N0 M0 > 4 mm + ulcer 53%

AJCC Stage and 5 year survival

Microsatellites

A microscopic metastasis located adjacent or deep to a primary melanoma

No minimal size threshold or distance from the primary tumor

Separated from the tumor by normal tissue (not inflamed or fibrotic)

Levels may be needed to distinguish from periadnexal tracking

Upstages to Stage III

Gershenwald, JE, Scolyer, RA, Hess, KR. et al. AJCC Cancer Staging Manual. 8th Ed. New York: Springer 2017

Non-nodal Locoregional Metastases

Result from intralymphatic or possibly angiotropic spread, similar prognosis for any type of locoregional metastasis

Microsatellite: microscopic metastases found adjacent or deep to primary melanoma on pathologic examination

Satellite: grossly visible cutaneous or subcutaneous metastasis within 2 cm of the primary melanoma

In-transit metastasis: clinically evident metastases > 2 cm from the primary melanoma in the region between the primary and the first echelon of regional lymph nodes.


Locoregional metastasis and prognosis in melanomaGershenwald JE, et al. Cancer J Clin 2017;67:472-92

“intransit” = in transit or satellite



Microsatellite, satellite, intransit – no survival difference, staged by number of lymph nodes involved

When non-nodal locoregional metastases are present:

N1c no regional node disease

N2c one clinically occult (SLN+) or clinically detected lymph node

N3c two or more clinically occult (SLN+) or clinically detected nodes or any number of matted nodes



“microscopic” and “macroscopic” have been redefined as

Clinically occult : clinical stage I – II with nodal metastasis determined at sentinel node biopsy

Clinically detected: clinical stage III

Sentinel lymph node tumor burden is considered a regional disease prognostic factor that should be collected for all patients with positive sentinel lymph nodes but is not used to determine N category groupings

AJCC staging 2018

80 percent of patients diagnosed with melanoma present with localized disease (AJCC clinical stage I & II)

Prognosis for this group varies from 53-98%

SLN is offered to those determined to be at highest risk for microscopic LN metastasis

Who is offered sentinel lymph node mapping?

Patients with no clinical evidence of metastatic disease AND

Clinical stage 1b or greater (e.g. T1b or greater)…SLNB not recommended for primary melanomas ≤ 0.8 mm thick unless there is significant

uncertainty about the adequacy of microstaging

Ulceration, high mitotic rate and lymphovascular invasion (LVI) are rare in melanomas ≤ 0.8 mm thick, when present SLNB may be considered

SLN not recommended in patients with comorbidities that reduce life expectancy, or increase morbidity of SLN biopsy, or the biologically elderly

This is a shifting landscape leading to fewer SLN for melanoma < 0.8 mm

NCCN Guidelines V1. 2017

Negative SLN n = 384Negative SLN n = 384

Positive SLN n = 91

Time (in Months) Time (in Months)

80 8040 40

Prog

ress

ion

Free

Sur

viva

l

Ove

rall

Surv

ival

Luo S et al. JAMA Surgery 2015

Positive SLN n = 91

Progression free and overall survival in 475 MGH patients with sentinel lymph node biopsy

Management of Stage III melanoma is shifting, nevertheless the presence of even one melanoma cell indicates metastatic potential of the primary tumor

Technical aspects impact sensitivity of melanoma detection in SLN:Surgical technique

• is a sentinel node only one node? hottest vs. not-hottest

Histopathological technique • levels vs. not• immunohistochemical stains

Histopathological interpretation • nevus vs. melanoma

Detection of one melanoma cell in SLN upstages to AJCC III

Intraoperative Lymphatic Mapping

Isosulfan blue & Tc99m sulfur colloid

Blue staining of node and lymphatic

Hand held gamma counter

Removal of hot nodes (to 10% of #1)

Wide excision of cutaneous lesion

Elias et al Arch Surg. 2004

SLN melanoma metastases and Survival (n=475)

Surv

ival

SLN-

SLN+ (not-hottest)

SLN+ (hottest)

Non-hottest SLN positive: 21%Counts range 26-97% of hottest

Hottest SLN positive:79%

91 patients with positive SLN

Luo et al. JAMA Surg 2015;150:465-72

Sentinel Lymph Node Analysis

12% of positive lymph nodes are missed if levels and immunohistochemical stains are not performed

94 patients with negative SLN on one H&E (235 LN)

deeper levels and immunohistochemical stains revealed melanoma in 12% of patients

capsular melanocytic nevi in 9% of patients

Yu L, et al. Cancer 1999


No standard procedure exists

Analytical platform varies between Cancer Centers– Single H&E– H&E plus one or two immunohistochemical stains*– H&E plus immunostains* at multiple deeper levels

* S100, Mart-1, HMB-45, Melan-A, MITF

15% false negative rate without deeper levels

SLN workup for melanoma at 18 institutions

Dekker, J et al. Arch Pathol 2013

Metastatic Melanoma MART -1

Metastatic Melanoma MART -1

Diagnostic criteria for metastatic melanoma

individual cells or nests of epithelioid or spindle cells foreign to the lymph node

cytological atypia defined as large cells with nuclear pleomorphism, prominent nucleoli, or dusty cytoplasmic melanin granules

positive staining for one or more melanocytic marker (S-100, MART-1)

identification of the cells on H&E (not required)

Metastatic Melanoma S100


Metastatic Melanoma


Nodal Nevus MART-1

Criteria for melanocytic nevus in LN

individual cells in a linear array or nests of epithelioid or spindled cells foreign to the lymph node

no cytological atypia

positive staining for one or more melanocytic marker (S-100, MART-1, MelanA, MITF)

identification of the cells on H&E

cells are usually present in the fibrous capsule or trabeculae

Metastatic Melanoma and Nodal Nevus

MITF

Distribution of melanoma in SLN

Do we need 3 sets of deeper levels?

475 MGH patients, positive SLNs in 91 patients (19%)

Of 61 cases with all protocol slides available:64% melanoma in every protocol slide (9 slides)36% tumor only on selected levels18% without tumor in the first set of levels

Lobo A. et al. Am J Surg Pathol 2012

Distribution of melanoma metastases in sentinel nodes

Lobo et al AJSP 2012 = 11 cases with no tumor in first set of levels (18%)

Distribution of melanoma metastases in sentinel nodes


> 15% false negative rate if protocol does not include levels into the tissue

> 9% false positive rate by PCR (benign nodal melanocytic nevi)

Immunohistochemical stains aid in detecting microscopic foci of melanoma

Size of sentinel node metastasis correlates with risk of progression

SLN Analysis Take Home Points

The presence of one metastatic tumor cell upstages patient to stage IIIFailure to examine deeper levels leads to a 12-15% false negative rateImmunohistochemical stains help to identify microscopic melanoma

The diagnostic pathology report should contain:Number of positive LN/ total number of LNExtracapsular extension present or absentMaximal diameter of the largest metastatic deposit Immunohistochemical stains performed and results

In ambiguous primary tumors, metastasis is not diagnostic of melanoma (AST)

2007: 37 year old woman Atypical Spitz Tumor (AST)upper arm with severe atypia and mitoses

Atypical SpitzTumors

Cases with +SLN/ total

patients

Mean follow-up in months

Cases with mets beyond the SLN/ total

Deaths

Smith et al. 1989 6/32 (19%) 72 0/30 0/30Lohmann et al. 2002 5/10 (50%) 33.7 0/10 0/10Su et al. 2003 8/18 (44%) 9.8 0/18 0/18Gamblin et al. 2006 3/10 (33%) 33.7 0/10 0/10Urso et al. 2006 4/12 (33%) 26.3 0/12 0/12Murali et al. 2008 6/21 (29%) 25.8 0/21 0/21Ludgate et al. 2009 27/57 (47%) 43.8** 0/57 0/57Busam et al. 2009 11 all + SLNB* 61.4 0/11 0/11

Cerroni et al. 2010 25/35 (71%) 83.5 8/35 1/35Tom et al. 2011 2/4 (50%) 25.3 0/4 0/4Raskin et al. 2011 8/15 (53%) Not reported 0/15 0/15Sepehr et al. 2011 1/6 (17%)*** 64.6 0/76 0/76Barnhill et al. 2011 4/5 * 10.5 0/5 0/5

Hung et al. 2013 6/23 (26%) 55.6 0/23 0/23Totals 102/245 (42%) 8/327 (<3.0%) 1/327 <0.5%

In the cohort of node-positive patients with intermediate-thickness primary melanomas (defined in this study as 1.20 to 3.50 mm), the study showed a significant advantage with respect to melanoma-specific survival for those who underwent sentinel-node biopsy, had a positive result, and underwent early regional lymphadenectomy, as compared with those who underwent wide excision but not sentinel-node biopsy, with regional lymphadenectomy performed only after regional disease developed.

Known, accepted, and confirmed by MSTLI :

Pathological status of the SLN is the most important prognostic factor

SLN removal led to fewer recurrences

SLN and CLND a shifting landscape

Intraoperative lymphatic mapping

Sentinel lymph node biopsy

Completion LymphadenectomyObservation

- +

7th International Melanoma Pathology Workshop, UCSF November 17, 2015

Now known and accepted and confirmed by MSTLI – Pathological status of the SLN is the most important prognostic factor– SLN removal led to fewer recurrences

MSLTIIinternationalmulticenter (63 sites)randomizedphase 3 trial

MSLTI vs. MLSTII

Evaluate the usefulness of Completion Dissectionin melanoma patients

with positive SLN

Per-protocol analysis 1755 patientsBreslow 1.2 mm – 3.5 mmRandomized to CLND ( n= 824) or observation with U/S (n = 931)

• Demographic and primary tumor pathological factors (thickness and ulceration) were similar between cohorts

• Mitoses were not evaluated• Sentinel lymph node metastases size (< 0.1, 0.1-1.0, > 1.0mm)• Fewer than 4% in each cohort had 3 or more positive SLN

MSLTII NEJM 2017

MSLTII NEJM 2017

There was no significant difference in melanoma specific survival between the CLND (86%) and Observation (86%) cohorts with a median follow up time of 43 months

MSLTII NEJM 2017

There was a statistically significant increase in Disease Free Survival in the CLND (68%) cohort when compared to the Observation (63%) cohort, p = 0.05

This is attributed to the increased rate of regional lymph node disease control in the CLND (92%) vs. the Observation (77%) cohorts at 3 years

Lymphedema occurred in 24% of CLND vs. 6% of Observed

Completion Lymphadenectomy is associated with:

- increased rate of regional disease control

- prognostic information

- increased rate of lymphedema

- No increase in melanoma specific survival

MSLTII NEJM 2017: Conclusions

Are there patients who may still benefit from completion lymphadenectomy?

• Patients with more than 2 + SLN represent <5% of all patients in MSLTII• the data do not provide insight into the impact of large SLN metastases (largest cutoff is 1 mm)• there is no evaluation of patients with matted lymph nodes or extracapsular extension• primary tumor mitoses are not included in the analysis• the 3 year (43 month) follow up time is relatively short for early stage melanoma.

Number of positive SLN, and extent of tumor burden in the SLN have been associated with increased risk of non-sentinel lymph node metastasis and decreased melanoma-specific survival.

MSLTII NEJM 2017: CLND may no longer be recommended in most patients with SLN metastases

• Completion lymphadenectomy may not offer a survival advantage for most patients with clinically localized primary cutaneous melanoma and positive sentinel lymph nodes, e.g. patients with 2 or fewer positive lymph nodes, with metastases less than 1 mm, and without extracapsular extension or matted lymph nodes

• Completion lymphadenectomy may be considered in patients with any of these characteristics: 3 or more positive sentinel lymph nodes, SLN tumor burden > 1 mm, extracapsular extension or matted lymph nodes

• Completion lymphadenectomy may be considered in patients with poor compliance and who may be non-surveillable,

• Completion lymphadenectomy may be considered for patients desiring to participate in a clinical trial with completion lymphadenectomy as an enrollment criterion

In the broader view of the existing literature:

SLN (Sentinel Lymph Node)• In years past SLN was offered for even very thin mitogenic or ulcerated

primary cutaneous melanoma (0.3-0.7 mm)• Now only rare circumstances lead to SLN for tumors < 0.8 mm

CLND (Completion Lymphadenectomy)• In years past CLND was offered to patients with SLN metastases of any size

(including those with only one tumor cell detected) • Now CLND is rarely recommended (exceptions include requirement for clinical

trial eligibility, or inability to comply with observation exams and imaging)

SLN and CLND in Melanoma: a shifting landscape

AJCC Prognostic Stage Groups: Clinical Staging


AJCC Prognostic Stage Groups: Pathological Staging


N1a: one clinically occult node (SLN+)N1b: one clinically detected nodeN1c: no regional node but non-nodal met*N2a: 2 or 3 clinically occult (SLN+)N2b: 2 or 3 at least one clinically detectedN2c: 1 occult or clinically detected and *N3a: 4 or more nodes or 2 or more with * or any with matted nodesN3b: 4 or more nodes with at least one clinically detected or mattedN3c: 2 or more occult or clinical, or matted and *

* microsatellite, satellite or in-transit





Stage Nodes In transit, satellite or microsatellite

metsN1a 1 clinically occult node met No

N1b 1 clinically detected node met No

N1c No regional lymph node disease Yes

N2a 2 – 3 clinically occult No

N2b 2 – 3, at least one clinically detected No

N2c 1 clinically occult or clinically detected Yes

N3a > 4 clinically occult No

N3b > 4, at least one clinically detected, or presence of matted nodes No

N3c > 2 clinically occult or clinically detected and/or presence of matted nodes Yes

AJCC Staging 8th edition (N)









Gershenwald JE, et al. Cancer J Clin 2017;67:472-92

SLN +

Clinically detected or one locoregional met 4 or more clinically detected, matted or one locoregional met with one clinically detected LNUlcerated > 4mm primary with 4 or more clinically detected, matted or one locoregional met with one clinically detected LN

Melanoma staging: Evidence‐based changes in AJCC 8th ed.Gershenwald JE, et al. Cancer J Clin 2017;67:472-92

AJCC Stage I and II Melanoma Survival (SLN+ excluded)Gershenwald JE, et al. Cancer J Clin 2017;67:472-92

Tumor (T), Node (N), and Metastasis (M) categories and stage groupings

• informs prognostic assessment • informs clinical decision making• facilitates centralized cancer registry reporting • facilitates the design, conduct, and analysis of clinical trials

Lymphatic mapping and Sentinel lymph node (SLN) biopsy

Major new classes of effective systemic therapeutic agents • immunotherapies (eg, checkpoint inhibitors against CTLA‐4 and/or PD‐1)• molecularly targeted antitumor therapies (eg, BRAF inhibitors +/- MEK inhibitors)

Melanoma Staging 2018 – Clinically Relevant Stratification

Documents

AJCC Melanoma Staging 2018 - Pathology · 75% of all skin cancer deaths are from melanoma. 1 in 50 Americans will develop melanoma in their lifetime. Melanoma is 2nd most common cancer