Journal Pre-proof
Dermatopathologic Features of Cutaneous Squamous Cell Carcinoma and Actinic
Keratosis: Consensus Criteria and Proposed Reporting Guidelines
Rachel E. Christensen, BS, Dirk M. Elston, MD, Brandon Worley, MD, MSc,
McKenzie A. Dirr, BA, BS, Noor Anvery, BA, Bianca Y. Kang, MD, Soon Bahrami,
MD, Robert T. Brodell, MD, Lorenzo Cerroni, MD, Carly Elston, MD, Tammie
Ferringer, MD, M. Yadira Hurley, MD, Kyle Garton, MD, PhD, Joyce Siong See Lee,
MRCP, Yeqiang Liu, PhD, John C. Maize, MD, Jennifer M. McNiff, MD, Ronald P.
Rapini, MD, Omar P. Sangueza, MD, Christopher R. Shea, MD, Cheng Zhou, MD,
Murad Alam, MD, MSCI, MBA
PII: S0190-9622(23)00271-2
DOI: https://doi.org/10.1016/j.jaad.2022.12.057
Reference: YMJD 17455
To appear in: Journal of the American Academy of Dermatology
Received Date: 30 June 2022
Revised Date: 2 December 2022
Accepted Date: 19 December 2022
Please cite this article as: Christensen RE, Elston DM, Worley B, Dirr MA, Anvery N, Kang BY, Bahrami
S, Brodell RT, Cerroni L, Elston C, Ferringer T, Hurley MY, Garton K, Lee JSS, Liu Y, Maize JC, McNiff
JM, Rapini RP, Sangueza OP, Shea CR, Zhou C, Alam M, Dermatopathologic Features of Cutaneous
Squamous Cell Carcinoma and Actinic Keratosis: Consensus Criteria and Proposed Reporting
Guidelines, Journal of the American Academy of Dermatology (2023), doi: https://doi.org/10.1016/
j.jaad.2022.12.057.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
? 2023 Published by Elsevier Inc. on behalf of the American Academy of Dermatology, Inc.
1
Article Type: Dermatopathology 1
2
Title: Dermatopathologic Features of Cutaneous Squamous Cell Carcinoma and Actinic Keratosis: 3
Consensus Criteria and Proposed Reporting Guidelines 4
5
Rachel E. Christensen, BS1; Dirk M. Elston, MD2; Brandon Worley, MD, MSc1; McKenzie A. Dirr, BA, 6
BS1; Noor Anvery, BA1; Bianca Y. Kang, MD1; Soon Bahrami, MD3; Robert T. Brodell, MD4,5; Lorenzo 7
Cerroni, MD6; Carly Elston, MD7; Tammie Ferringer, MD8,9; M. Yadira Hurley, MD10; Kyle Garton, MD, 8
PhD11; Joyce Siong See Lee, MRCP12; Yeqiang Liu, PhD13; John C. Maize, MD2; Jennifer M. McNiff, 9
MD14,15; Ronald P. Rapini, MD16; Omar P. Sangueza, MD17,18; Christopher R. Shea, MD19; Cheng Zhou, 10
MD20; Murad Alam, MD, MSCI, MBA1,21,22 11
12
1Department of Dermatology, Northwestern University Feinberg School of Medicine, Chicago, IL, USA 13
2Department of Dermatology and Dermatologic Surgery, Medical University of South Carolina, 14
Charleston, SC, USA 15
3Department of Medicine, Division of Dermatology, University of Louisville School of Medicine, 16
Louisville, KY, USA 17
4Departments of Dermatology and Pathology, University of Mississippi Medical Center, Jackson, MS, 18
USA 19
5Staff Physician, Sonny Montgomery Veterans Administration Hospital, Jackson, MS, USA 20
6Research Unit of Dermatopathology, Department of Dermatology, Medical University of Graz, Austria 21
7Department of Dermatology, University of Alabama, Birmingham, Alabama, USA 22
8Department of Pathology, Geisinger Medical Center, Danville, PA, USA 23
9Department of Dermatology, Geisinger Medical Center, Danville, PA, USA 24
10Department of Dermatology, Saint Louis University School of Medicine, St. Louis, MI, USA 25
11Division of Dermatology, Department of Medicine, University of Washington, Seattle, WA, USA 26
12National Skin Centre, Singapore 27
13Department of Pathology, Shanghai Skin Disease Hospital, Tongji University School of Medicine, 28
Shanghai, China 29
14Department of Dermatology, Yale School of Medicine, New Haven, CT, USA 30
15Department of Pathology, Yale School of Medicine, New Haven, CT, USA 31
16Department of Dermatology, University of Texas McGovern Medical School at Houston, TX, USA 32
17Department of Pathology, Wake Forest School of Medicine, Winston-Salem, NC, USA 33
18Department of Dermatology, Wake Forest School of Medicine, Winston-Salem, NC, USA 34
19Department of Medicine, Section of Dermatology, University of Chicago, Chicago, IL, USA 35
20Department of Dermatology, Peking University People’s Hospital, Beijing, China 36
21Department of Otolaryngology, Northwestern University Feinberg School of Medicine, Chicago, IL, 37
USA 38
22Department of Surgery, Northwestern University Feinberg School of Medicine, Chicago, IL, USA 39
40
Corresponding Author: 41
Murad Alam, MD 42
676 N St Clair St, Ste 1600 43
Chicago, IL 60611 44
(312) 695-6647 45
(312) 695-4541 fax 46
m-alam@northwestern.edu 47
48
Funding Sources: None. 49
50
Conflicts of Interest: None declared. 51
Journal Pre-proof
2
52
Patient consent: Not applicable. 53
54
55
Manuscript Word Count: 2400/2500 Abstract: 159/200 Capsule Summary: 43/50 56
References: 7 57
Figures: 0 Tables: 3 Mendeley Supplemental Tables: 1 58
59
Keywords: International; consensus; squamous cell carcinoma; actinic keratosis; pathology; diagnosis; 60
immunohistochemistry 61
Journal Pre-proof
3
ABSTRACT 62
Background: There is considerable variation in the literature regarding the dermatopathologic diagnostic 63
features of, and reporting guidelines for, actinic keratosis (AK) and cutaneous squamous cell carcinoma 64
(cSCC). 65
Objective: To develop consensus recommendations regarding diagnostic criteria, nomenclature, and 66
reporting of AK and cSCC. 67
Methods: Literature review and cross-sectional multi-round Delphi process including an international 68
group of expert dermatopathologists followed by a consensus meeting. 69
Results: Consensus was achieved regarding the key dermatopathologic features necessary for diagnosing 70
cSCC, AK, and associated variants; grading of degree of cellular differentiation in cSCC; utility of 71
immunohistochemistry for diagnosis of cSCC; and pathologic features that should be reported for cSCC 72
and AK. 73
Limitations: Consensus was not achieved on all questions considered. 74
Conclusion: Despite the lack of clarity in the literature, there is consensus among expert 75
dermatopathologists regarding diagnostic criteria and appropriate reporting of AK and cSCC. Widespread 76
implementation of these consensus recommendations may improve communication between 77
dermatopathologists and clinicians, facilitating appropriate treatment of AK and cSCC. 78
Journal Pre-proof
4
Capsule Summary 79
? Consensus regarding the key dermatopathologic features necessary for diagnosing and reporting 80
cutaneous squamous cell carcinoma, actinic keratosis, and associated variants was achieved, reducing 81
inconsistencies and simplifying definitions. 82
? Implementation of these consensus recommendations may improve communications among 83
dermatopathologists and clinicians to facilitate appropriate treatment. 84
Journal Pre-proof
5
INTRODUCTION 85
Actinic keratoses (AK) are premalignant lesions in sun-exposed areas in light-skinned individuals 86
that affect 11-26% of Americans.1 Over time, 10% of AKs may progress to cutaneous squamous cell 87
carcinoma (cSCC).2 Cutaneous SCC can be further classified as in situ or invasivea, with the latter 88
presenting a risk of metastasis.3,4 Despite the importance uniform definitions and diagnostic criteria for 89
AK, cSCC in situ, and invasive cSCC, there is considerable variation in descriptions in the literature.5 90
Moreover, there is a lack of consensus regarding key histopathologic features and their relationship to 91
disease prognosis. This potentially complicates clinicians’ ability to understand diagnostic nuance and 92
select appropriate treatment.6,7 93
Histopathologic identification of AK and cSCC may be improved with more precise definitions 94
for each. Such uniformity may also make it easier to interpret the results of clinical trials evaluating 95
treatments. 96
To develop consensus criteria for AK and cSCC, a comprehensive literature search was 97
conducted, and an international group of dermatopathologists was convened. Consensus regarding the 98
key histopathological features associated with cSCC and AK was achieved through a formal Delphi 99
consensus method. 100
The purpose of this exercise was to clarify areas in which there was widespread agreement 101
among pathologists, as well as grey areas in which agreement was absent. The purpose was not to change 102
existing diagnostic categories or create new ones, but rather to find areas of pre-existing agreement so that 103
these could be highlighted. To the extent that dermatopathologists concurred on certain definitions, visual 104
and histopathologic features, prognosis, and other aspects of AK and SCC, publicly declaring these areas 105
as points of consensus would increase the utility of these terms for both dermatopathologists and treating 106
clinicians. Pathologists and clinicians would now know they were speaking about the same phenomena, 107
aInvasion refers to the tumor extending below the epidermis, whereas thickness is the overall dimension
of involvement from the top of the stratum corneum to the deepest layer of the tumor. In other words, a
thin tumor that is deeply invasive is “worse” than a thick tumor that is minimally invasive.
Journal Pre-proof
6
thereby reducing the risk of misunderstandings. Conversely, when consensus could not be reached 108
regarding certain terms, publicly acknowledging that experts differed on the meanings and relevance of 109
these diagnostic categories would motivate clinicians and dermatopathologists to reduce use of the 110
controversial terms and categories, or ask for clarification as to a particular speaker’s specific meaning. In 111
short, the purpose of this study was to allow pathologists’ views regarding AK and SCC diagnostic 112
categories to see the light of day and be explicitly, publicly aired. Finally, the need for further research 113
would be identified in areas in which consensus had not yet emerged. 114
115
METHODS 116
A comprehensive literature search on PubMed was performed to identify articles discussing key 117
histopathologic features of AK and cSCC. Pathology textbooks were also consulted. A steering 118
committee (BW, DE, MA) then developed definitions for cSCC, AK, and keratoacanthoma. Additional 119
statements relating to degree of pathologic differentiation, perineural invasion, immunohistochemistry 120
findings, and pathologic reporting standards were similarly generated. 121
An international working group of 15 dermatopathologists was convened. Members were 122
identified as experts through relevant publication history, international recognition, relevant clinical 123
expertise, and/or peer nomination. 124
A draft list of declarative statements was refined through two rounds of Delphi surveys. In the 125
first round, participants rated the accuracy and completeness of statements relevant to the pathological 126
interpretation of cSCC and AK (1=strongly disagree; 9=strongly agree). Participants also provided open-127
ended feedback. In the second round, participants considered their and others’ responses from the first 128
round before rating the revised statements as well as any new statements suggested by the participants. 129
Consensus entailed at least 70% of participants strongly agreeing (ratings: 7-9), and fewer than 15% of 130
participants strongly disagreeing (ratings: 1-3). At a subsequent virtual consensus meeting, members of 131
the steering committee (DE, MA, BW, RC) facilitated discussion of statements which did not meet 132
Journal Pre-proof
7
consensus during the Delphi process. All statements which met consensus were recorded (Table I-III; 133
Supplemental Table 1 via Mendeley https://data.mendeley.com/datasets/2m5kbdrwsx). 134
135
RESULTS 136
I. Actinic Keratosis 137
Typical AK presents clinically as a rough, ill-defined, scaly patch or papule usually <1 mm in height 138
or with a cutaneous horn but without induration at the base. In contrast, hyperkeratotic actinic keratosis 139
presents as a rough, ill-defined spiny papule >1 mm in height. Histologically, both AK and hyperkeratotic 140
AK may demonstrate atypia of the keratinocytes without pagetoid scatter, clonal nesting, or broad, full-141
thickness atypia of the epidermis and follicles. In a transected specimen (e.g., the lesion base cannot be 142
fully viewed due to biopsy transection or sampling within the tissue block), full-thickness epidermal 143
involvement raises concern that the lesion may be more likely to progress without additional treatment, or 144
that invasive cSCC is already present. 145
146
II. Actinic Keratosis with Focal Squamous Cell Carcinoma 147
In the case of AK with cSCC, whether the cSCC is in situ or invasive should be made clear. AK with 148
focal invasive cSCC should be considered cSCC and treated as such. AK with focal in situ cSCC may 149
require more aggressive treatment than typical AK, while in situ cSCC with adjacent AK is treated as in 150
situ cSCC. 151
152
III. Cutaneous Squamous Cell Carcinoma 153
Primary cSCC in situ was defined as full-thickness atypia of keratinocytes and disordered 154
maturation, without dermal invasion. Histological features may include altered maturation and mitoses 155
above the basal layer, pagetoid scatter, clonal nesting, or full-thickness atypia of follicles. Primary 156
invasive cSCC was defined as an invasive tumor arising from partial- to full-thickness keratinocyte 157
Journal Pre-proof
8
epidermal atypia. Recurrent cSCC was defined as an in situ or invasive keratinocytic malignancy arising 158
within or adjacent to a scar identified at a prior surgical site 159
160
IV. Keratoacanthoma 161
Keratoacanthomas are eruptive, crateriform keratinocyte neoplasms with thick keratin cores. 162
Histologically, keratoacanthomas demonstrate glassy atypia of keratinocytes and an absence of 163
acantholysis with or without the presence of neutrophil microabscesses, elastic trapping, or eosinophils 164
within the surrounding stroma. Clinical history and tissue samples suspicious for keratoacanthoma often 165
are inadequate for definitive diagnosis or to predict the subsequent behavior of the lesion. 166
167
V. Pathologic Differentiation 168
Well differentiated cSCC was defined as a malignant keratinocytic neoplasm with abundant 169
cytoplasm and glassy atypia with invasive islands of atypical keratinocytes resembling the epidermis, 170
often with evidence of keratinization. In-situ or tumor only involving the epidermis has identical features 171
but lacks invasive cells or islands. Moderately differentiated cSCC was defined as a disordered malignant 172
keratinocytic neoplasm with evidence of keratinization but without anaplastic or spindled (sarcomatoid) 173
characteristics. Finally, poorly differentiated cSCC was defined as a malignant keratinocytic neoplasm 174
with little to no evidence of keratinization and anaplastic or spindled (sarcomatoid) characteristics. For 175
poorly differentiated cSCC, immunohistochemistry is often required to identify the neoplasm as 176
keratinocytic. 177
178
VI. Perineural Invasion 179
Perineural invasion should be reported when invasion involves a nerve that measures >0.1 mm in 180
diameter or a named nerve, or when there is clinical evidence of frank neurologic motor or sensory 181
findings. Perineural invasion outside the tumor mass or within the subcutaneous tissue is more critical to 182
prognosis than that within the tumor bulk. 183
Journal Pre-proof
9
184
VII. Immunohistochemistry 185
It is appropriate to perform CK5, AE1/3, p63, or p40 immunohistochemistry to confirm the diagnosis 186
of poorly differentiated cSCC. In-office immunostaining with a high molecular weight cytokeratin at the 187
time of Mohs surgery or complete circumferential peripheral and deep margin assessment (CCPDMA) is 188
sometimes an appropriate approach to confirm negative margins. However, tumor characteristics should 189
guide its use, such as for suspected risk of single cell spread. 190
191
VIII. Reporting of Pathologic Findings 192
Recommended reproducible categories for reporting included: 1) AK; 2) cSCC in situ; 3) well to 193
moderately differentiated cSCC; and 4) poorly differentiated cSCC. 194
For excised cSCC (i.e., specimens in which the whole lesion is represented), the following features 195
should always be reported to the clinician: tumor site; peripheral margins; deep margins; grade of 196
differentiation (well/moderate or poor); perineural invasion and, if applicable, the diameter and depth of 197
the nerve involved; lymphovascular invasion (e.g., number of lymph nodes examined, number of lymph 198
nodes involved, and extranodal extension); acantholysis; other histological characteristics that predict 199
worse outcomes (e.g., spindle cell sarcomatoid and poorly differentiated); and pathologic subtypes (e.g., 200
verrucous subtype and adenosquamous subtype). Moreover, when possible, the type of biopsy procedure 201
that was performed (i.e., shave/tangential, punch, or incisional/excisional) should be reported by the 202
clinician. When the type of tissue sampling procedure that was performed could not be determined at the 203
time of histological evaluation, there may be value in noting the depth of the skin structures traversed by 204
the procedure (e.g., the sample extends to the level of the papillary dermis and tumor involves the deep 205
margin). Notably, consensus was not reached regarding reporting of depth of invasion of SCC, with a 206
majority of panelists unable to agree on the need to report such depth or how to measure it, whether it be 207
from the uninvolved dermoepidermal junction, from the dermoepidermal junction with in-situ cSCC, 208
Journal Pre-proof
10
from the top of the granular cell layer of the epidermis overlying the tumor (that is, similar to Breslow 209
thickness), or from the top of the granular layer of uninvolved skin. 210
For AK, the anatomic site of biopsy should be reported. Whether the lesion is recurrent at the same 211
site (e.g., within a scar) cannot be determined by pathological diagnosis alone and requires 212
clinicopathological correlation. Reporting margins on AK may not be clinically helpful. 213
214
DISCUSSION 215
Based on a literature review and formal Delphi process, consensus was achieved regarding the key 216
dermatopathologic features necessary for diagnosing cSCC, AK, and associated variants, including those 217
that are low risk and those that may be higher risk (i.e., with perineural invasion); grading of degree of 218
cellular differentiation in cSCC, and, in particular, distinguishing well differentiated, moderately 219
differentiated, and poorly differentiated cSCC; and using immunohistochemistry for diagnosis of cSCC; 220
and pathologic features that should be reported for cSCC and AK. 221
Importantly, there were several controversial topics for which consensus was not achieved. The 222
expert panel could not reach agreement on the definition of ‘Bowenoid’ AK, including whether such are 223
cSCC in situ, whether they have greater potential for progression to cSCC than other AKs, and whether 224
they require more aggressive treatment. Expert panelists held varying views on whether basosquamous 225
carcinoma is more accurately defined as cSCC with focal areas of basal cell morphology or BCC with 226
squamous differentiation (beyond simple keratin pearls). Whether basosquamous carcinoma represents a 227
more aggressive variant than either BCC or cSCC also remained without consensus. Although experts 228
agreed on several statements related to cSCC immunohistochemistry, they did not reach consensus 229
regarding whether in-office cytokeratin pan monoclonal antibody (MNF 116) immunostaining at the time 230
of Mohs surgery or CCPDMA to confirm negative margins is appropriate. While some panel members 231
preferred that cSCC differentiation should be grouped into two categories rather than three, consensus 232
was not achieved on which two categories should be collapsed. 233
Journal Pre-proof
11
Upon querying panelists to better understand why they preferred certain immunohistochemical stains 234
for identifying poorly differentiated SCC, we found that there was consensus that IHC staining with CK5, 235
AE1/3, p63, and p40 were best supported, widely accessible, and best able to differentiate SCC in the 236
context of diagnostic ambiguity or spindle cell morphology. Half of respondents asserted that IHC 237
staining with p40 was the preferred marker for identifying poorly differentiated SCC due to its labelling 238
specific p63 isoforms, but on this consensus was not reached. When we queried panelists regarding 239
synoptic reporting, there was consensus that this was recommended to guide management for poorly 240
differentiated SCC, with reporting ideally indicating tumor type, depth beyond the granular layer in 241
millimeters, deepest tissue invaded (e.g., fascia), presence of perineural invasion, and perivascular or 242
perilymphatic invasion. 243
Panelists disagreed as to whether keratoacanthoma was a variant of well-differentiated cutaneous 244
SCC, with a majority (64%) strongly agreeing but some (21%) strongly disagreeing. Clearly this is a 245
diagnostic category for which consensus is elusive, and further specification of lesions along the KA-SCC 246
spectrum may be necessary to resolve the disparate opinions. 247
The intention of this investigation was not to eliminate the use of controversial diagnostic categories, 248
for which there is not widespread consensus among pathologists regarding definitions, features, or 249
prognosis. Instead, the purpose was to highlight that certain categories are controversial so that those who 250
refer to these categories would know to clarify exactly what they mean. For instance, that many believe 251
tissue samples suspicious for keratoacanthoma are inadequate to predict subsequent behavior does not 252
mean that all pathologists would now avoid diagnosing KA, but it may mean they would add an 253
addendum clarifying their estimate of the likely prognosis, or perhaps their inability to exclude aggressive 254
behavior. 255
This investigation may spur further refinement of controversial categories or creation of new terms. 256
Improvement in nomenclature, combined with future research into the pathophysiology and progression 257
of AK and SCC, may eventually lead to increased consensus regarding diagnostic categories. A few years 258
Journal Pre-proof
12
from now, the categories deemed controversial at present may be replaced by newer, evidence-based 259
categories for which there is wider consensus. 260
The current study is limited in that consensus was not reached on all issues and questions that were 261
discussed. Given that definitions and diagnostic categories are inherently subjective and cannot be elicited 262
from empirical data alone, we believe the expert-based consensus process, albeit not perfect, was the most 263
appropriate methodology. The majority of our panelists were dermatology-trained pathologists, which 264
may bias the results. To the extent that this consensus process was focused on diagnostic categories, there 265
was limited guidance and consensus regarding therapeutic interventions that may be appropriate for 266
particular lesion types. For instance, while the panelists suggested that AK with focal in situ cSCC may 267
require more aggressive treatment than typical AK, exactly what such aggressive treatment may entail, 268
whether more invasive modalities (e.g., electrodessication and curettage versus liquid nitrogen), several 269
concurrent interventions (e.g., imiquimod plus topical 5-fluorouracil versus 5-fluorouracil alone), or more 270
focal treatment (e.g., curettage versus photodynamic therapy) remains unspecified. 271
272
CONCLUSION 273
Through a literature review and formal Delphi consensus process involving expert 274
dermatopathologists, consensus was met for the following diagnostic categories related to cSCC and AK: 275
the clinical presentation of cSCC, AKs, and keratoacanthomas; the corresponding histologic criteria and 276
definitions; the distinction between well differentiated, moderately differentiated, and poorly 277
differentiated cSCC; the utility of immunohistochemistry in poorly differentiated cSCC; and the reporting 278
of pathologic features for cSCC and AK. Consensus regarding the clinical and histopathologic features 279
and reporting of AK and cSCC may improve communication among pathologists as well as between 280
pathologists and clinicians, facilitating future research regarding diagnosis and management. 281
Journal Pre-proof
13
ABBREVIATIONS 282
AK: Actinic keratoses 283
cSCC: Cutaneous squamous cell carcinoma 284
CCPDMA: Complete circumferential peripheral and deep margin assessment 285
Journal Pre-proof
14
ACKNOWLEDGEMENTS 286
Author Contributions: Dr. Murad Alam had full access to the data in the study and take responsibility 287
for the integrity of the data and the accuracy of the data analysis. Study concept and design: Alam, 288
Worley. Acquisition, analysis, and interpretation of data: Christensen, D. Elston, Worley, Kang, Dirr, 289
Anvery, Bahrami, Brodell, Cerroni, C. Elston, Ferringer, Hurley, Garton, Lee, Liu, Maize, McNiff, 290
Rapini, Sangueza, Shea, Zhou, Alam. Drafting of the manuscript: Christensen, Dirr, Anvery. Critical 291
revision of the manuscript for important intellectual content: Christensen, D. Elston, Worley, Dirr, 292
Anvery, Kang, Bahrami, Brodell, Cerroni, C. Elston, Ferringer, Hurley, Garton, Lee, Liu, Maize, McNiff, 293
Rapini, Sangueza, Shea, Zhou, Alam. Statistical analysis: Christensen. Obtained funding: N/A. 294
Administrative, technical, or material support: Christensen. Study supervision: Alam. 295
Journal Pre-proof
15
References 296
1. de Oliveira ECV, da Motta VRV, Pantoja PC, et al. Actinic keratosis - review for clinical 297
practice. Int J Dermatol. 2019;58(4):400-407. 298
2. Salasche SJ. Epidemiology of actinic keratoses and squamous cell carcinoma. J Am Acad 299
Dermatol. 2000;42(1 Pt 2):4-7. 300
3. Burton KA, Ashack KA, Khachemoune A. Cutaneous squamous cell carcinoma: A review of 301
high-risk and metastatic disease. Am J Clin Dermatol. 2016;17(5):491-508. 302
4. Tokez S, Wakkee M, Louwman M, Noels E, Nijsten T, Hollestein L. Assessment of cutaneous 303
squamous cell carcinoma (cSCC) in situ incidence and the risk of developing invasive cSCC in 304
patients with prior cSCC in situ vs the general population in the Netherlands, 1989-2017. JAMA 305
Dermatol. 2020;156(9):973-981. 306
5. Jagdeo J, Weinstock MA, Piepkorn M, Bingham SF; Department of Veteran Affairs Topical 307
Tretinoin Chemoprevention Trial Group. Reliability of the histopathologic diagnosis of 308
keratinocyte carcinomas. J Am Acad Dermatol. 2007;57(2):279-284. 309
6. Khanna M, Fortier-Riberdy G, Dinehart SM, Smoller B. Histopathologic evaluation of cutaneous 310
squamous cell carcinoma: results of a survey among dermatopathologists. J Am Acad Dermatol. 311
2003;48(5):721-726. 312
7. Carr RA, Houghton JP. Histopathologists'' approach to keratoacanthoma: a multisite survey of 313
regional variation in Great Britain and Ireland. J Clin Pathol. 2014;67(7):637-638. 314
Journal Pre-proof
16
Table I. Statements related to cutaneous squamous cell carcinoma, actinic keratosis, and 315
keratoacanthoma meeting consensus. 316
317
Statement Category Statement
Respondents (%)
Agree
(7-9)
Neutral
(4-6)
Disagree
(1-3)
Cutaneous Squamous
Cell Carcinoma
Primary cSCC in situ: Full-thickness atypia of
keratinocytes and disordered maturation, without dermal
invasion. Features may include altered maturation and
mitoses above the basal layer, pagetoid scatter, clonal
nesting, or full-thickness atypia of follicles.
93% 7% 0%
Primary invasive cSCC: Invasive tumor arising from
partial- to full-thickness keratinocyte epidermal atypia. 93% 7% 0%
Recurrent cSCC: In situ or invasive keratinocytic
malignancy arising within or adjacent to a scar identified as a
prior surgical site.
100% 0% 0%
Actinic Keratosis
Actinic keratosis: Clinically, a rough, ill-defined, scaly
patch or papule usually <1 mm in height or with cutaneous
horn but without induration at the base. Histologically, atypia
of the keratinocytes without pagetoid scatter, clonal nesting,
or broad, full-thickness atypia of epidermis and follicles.
93% 7% 0%
Hyperkeratotic actinic keratosis: Clinically, a rough, ill-
defined, spiny papule > 1 mm in height with thick keratotic
cap or cutaneous horn. Histologically, atypia of the
keratinocytes without pagetoid scatter, clonal nesting, or
broad, full-thickness atypia of epidermis and follicles.
79% 14% 7%
Actinic keratosis, additional statement: In a transected
specimen (e.g., the lesion base cannot be fully viewed due to
biopsy transection or sampling within the tissue block), full
thickness epidermal involvement raises concern that the
lesion may be more likely to progress, or that invasive SCC
is already present.
71% 15% 14%
Keratoacanthoma
Keratoacanthoma: Eruptive crateriform keratinocyte
neoplasm with thick keratin core. Histologically, crateriform
with glassy atypia of keratinocytes and absence of
acantholysis +/- neutrophil microabscesses, elastic trapping,
or eosinophils within surrounding stroma.
86% 7% 7%
The clinical history and tissue samples suspicious for
keratoacanthoma often are inadequate for definitive
diagnosis or to predict the subsequent behavior of the lesion.
79% 7% 14%
Histology alone is inadequate to provide an understanding of
clinical behavior of KA vs SCC. Due to the potential for
partial sampling, variability in biology with tumor subtypes
and individual patient factors, expert consensus recommends
not solely basing treatment decisions on the diagnosis of KA
or SCC.
80% 10% 10%
318
319
Journal Pre-proof
17
Table II. Statements related to clinical pathologic features meeting consensus. 320
321
Statement Category Statement
Respondents (%)
Agree
(7-9)
Neutral
(4-6)
Disagree
(1-3)
Pathologic
Differentiation
Well differentiated cSCC (in situ or invasive): Malignant
keratinocytic neoplasm with abundant cytoplasm and glassy
atypia with or without invasive islands of atypical
keratinocytes, often with evidence of keratinization.
86% 14% 0%
Moderately differentiated cSCC: Malignant keratinocytic
neoplasm with evidence of keratinization but without
anaplastic or spindled (sarcomatoid) characteristics.
86% 14% 0%
Poorly differentiated cSCC: Malignant keratinocytic
neoplasm with little to no evidence of keratinization and
anaplastic or spindled (sarcomatoid) characteristics.
Immunohistochemistry may be required to identify the
neoplasm as keratinocytic.
93% 7% 0%
Immunohistochemistry is often required for diagnosis of
poorly differentiated cSCC. 86% 7% 7%
Perineural Invasion
Perineural invasion should be reported when it involves a
nerve that measures > 0.1 mm, a named nerve, or has frank
neurologic motor or sensory findings.
93% 0% 7%
Perineural invasion outside the tumor mass or within the
subcutaneous tissue is more critical to prognosis than those
that are within the tumor bulk.
93% 7% 0%
Immunohistochemistry
Poorly differentiated squamous cell carcinoma is an
appropriate histology to submit for final CK5
immunohistochemistry.
78% 15% 7%
Poorly differentiated squamous cell carcinoma is an
appropriate histology to submit for final AE1/3
immunohistochemistry.
79% 7% 14%
Poorly differentiated squamous cell carcinoma is an
appropriate histology to submit for final p63
immunohistochemistry.
86% 7% 7%
Poorly differentiated squamous cell carcinoma is an
appropriate histology to submit for final p40
immunohistochemistry.
93% 0% 7%
In-office immunostaining with a high molecular weight
cytokeratin at the time of Mohs surgery or CCPDMA is
sometimes an appropriate approach to confirm negative
margins.
71% 29% 0%
IHC staining with CK5, AE1/3, p63, and p40 was felt to be
best supported, widely accessible, and best able to
differentiate SCC where there were diagnostic ambiguity or
spindle cell morphology.
80% 20% 0%
322
323
Journal Pre-proof
18
Table III. Statements related to the reporting of pathologic findings meeting consensus. 324
Statement
Category Statement
Respondents (%)
Agree
(7-9)
Neutral
(4-6)
Disagree
(1-3)
Reporting of
Pathologic
Findings to
Standardize
Diagnosis
Use reproducible categories, such as: AK; cSCC in situ; Well to moderately
differentiated cSCC; Poorly differentiated cSCC 100% 0% 0%
The diagnosis of actinic keratosis with focal cSCC typically entails an area of
invasive cSCC. 86% 7% 7%
Actinic keratosis with focal cSCC may need to be treated more aggressively than
typical actinic keratosis. 86% 7% 7%
Actinic keratosis with focal invasive cSCC is cSCC and should be treated as such. 79% 14% 7%
When actinic keratosis with cSCC is reported, it should be made clear whether the
cSCC is in situ or invasive. 93% 7% 0%
For poorly differentiated tumors, it is recommended that a formal synoptic report
indicating tumor type, depth beyond the granular layer in millimeters, deepest tissue
invaded (eg. fascia), perineural invasion, and perivascular or perilymphatic invasion
be issued to convey important prognostic information that can influence further
management.
70% 30% 0%
For poorly differentiated tumors, a formal synoptic report indicating tumor type,
depth beyond the granular layer in millimeters, deepest tissue invaded (eg. fascia),
perineural invasion, and perivascular or perilymphatic invasion may be helpful to
convey important prognostic information that can influence further management.
70% 20% 10%
Features to be
Reported to the
Clinician for
cSCC
Tumor site 86% 14% 0%
Peripheral margins 71% 29% 0%
Deep margins 71% 22% 7%
Grade of differentiation (well/moderate or poor) 86% 7% 7%
Perineural invasion 93% 0% 7%
If perineural invasion is present, the diameter and depth of the nerve involved 79% 7% 14%
Lymphovascular invasion (e.g., number of lymph nodes examined, number of lymph
nodes involved, and extranodal extension) 79% 7% 14%
Acantholysis 71% 15% 14%
Presence of adenosquamous or sarcomatoid differentiation 93% 7% 0%
Type of biopsy procedure that was performed (i.e., shave/tangential, punch,
incisional/excisional) should be reported by the clinician for cSCC when possible.
When the type of tissue sampling procedure that was performed cannot be
determined at the time of histological evaluation, there may be value in noting the
depth of the skin structures traversed by the procedure (e.g., the sample extends to
the level of the papillary dermis and tumor involves the deep margin).
79% 14% 7%
Invasive squamous cell carcinoma may benefit from reporting depth of invasion from
the granular layer and anatomic depth. However, there is no consensus regarding the
significance of overall thickness of the tumor as it is not clear this provides a clear
understanding of the tumor biology or aggressiveness.
70% 20% 10%
Reporting of
cSCC Pathologic
Subtypes
The pathologic subtype of cSCC should be reported. 71% 22% 7%
Verrucous subtype should be reported. 71% 22% 7%
Adenosquamous cSCC subtype should be reported. 71% 15% 14%
Histological characteristics that predict worse outcomes (e.g., spindle cell
sarcomatoid, poorly differentiated, acantholytic) should be reported. 100% 0% 0%
Features to be
Reported to the
Clinician for AK
Site of biopsy 93% 0% 7%
Length/width of lesion 14% 15% 71%
Thickness of lesion 14% 7% 79%
Reporting margins on actinic keratosis may not be clinically helpful. 93% 0% 7%
Whether the lesion is recurrent at the same site (e.g., within a scar) cannot be
determined by pathological diagnosis alone without clinicopathological correlation. 79% 7% 14%
325
Journal Pre-proof
Capsule Summary 1
? Consensus regarding the key dermatopathologic features necessary for diagnosing and reporting 2
cutaneous squamous cell carcinoma, actinic keratosis, and associated variants was achieved, reducing 3
inconsistencies and simplifying definitions. 4
? Implementation of these consensus recommendations may improve communications among 5
dermatopathologists and clinicians to facilitate appropriate treatment. 6
Journal Pre-proof
|
|
