lower genital tract squamous lesions - ucsf departments...

Selected Dilemmas in Lower Genital Tract Pathology

Christopher P. Crum, MD

Topic 1: Unusual and difficult variants of VIN

Topic 2: Invasive vulvar carcinoma, yes or no?

Topic 3: CIN or immature metaplasia?

Topic 4: Where can I get into trouble and not even know it?


Most VINs are classified into two categories; 1) classic or usual VINs, which exhibit full or near full-thickness atypia, are HPV positive and closely resemble high grade SIL of the cervix, and 2) differentiated VINs, which are classified by moderate basal atypia and abnormal keratinocyte differentiation, are HPV negative, and frequently harbor p53 mutations. A number of other variations on this theme can be encountered and include the following:

3) Classic VINs with superimposed lichen simplex chronicus. These lesions appear relatively normal in the upper third of the epithelium and may be confused with differentiated VIN. However, they typically exhibit more prominent atypia and are diffusely p16 positive.

4) Subtle classic VINs. These are similar to descriptions of bowenoid dysplasia.i ii iii The atypia is full-thickness but is subtle. Thus the differential diagnosis includes lower grade lesions such as VIN1 etc.

5) Pagetoid VIN. This unusual variant is characterized by a poorly differentiated intraepithelial neoplasm that permeates the normal squamous epithelium analogous to more classical paget’s disease.iv

6) VIN with columnar differentiation. This is a rare variant.v

7) Metastatic carcinomas growing on the vulvar mucosa. This can occur rarely with serous carcinomas of the upper genital tract and with urothelial carcinomas extending from the urinary tract.

8) VINs that are not really VINs. These include mucosal proliferations that display minimal atypia but increase risk of malignancy, such as LSC/LSA, VAAD and verruciform LSC. These are approached with caution and a recommendation for followup.vi vii


Excluding invasive carcinoma of the vulva is a paramount issue when evaluating vulvar biopsies. Pitfalls include the following:

1) Tangential sectioning of VIN. VINs are notoriously thick, tend to displace the underlying stroma with pushing margins and extend into appendages. It is the uniformity of the nesting pattern and preservation of cell polarity that aids in distinguishing this process from invasion.

2) Inflammation at the epithelial stromal interface in VIN. This can produce small discrete nests with vague appearing boundaries. The key to excluding cancer is the uniformity of the nests and the preservation of cell polarity.

3) Pseudo-epitheliomatous hyperplasia. This process takes two different forms. The most obvious is the fine interlaced strands of epithelium that are associated with inflammation. The second is the presence of discrete nests with mild alterations in squamous differentiation. The latter can be confused with invasive cancer.

4) Confusing entities such as keratoacanthoma. These lesions are problematic due to their uncertain risk of metastasis. We make this diagnosis rarely, to say the least.

Topic 3. CIN or immature metaplasia?

It is helpful to approach this problem with the following knowledge:

1) The association of a lesion with HPV16 increases as a function of inter-observer agreement for a diagnosis of CIN2 or higher. viii ix

2) The association between HPV16 (and cross-reacting HPVs) and high grade disease is such that nearly 85% of all CIN2-3 lesions are expected to disappear once the vaccines are established. x

3) While maturation is an important parameter in determining lesion grade, the emergence of HPV positive immature metaplasias with mild atypia requires attention to the fact that immaturity alone is not sufficient to warrant classification as HSIL.

4) Over 40% of CIN2s (by consensus) in women under age 25 will regress within 6 months.

5) The pathologist must be aware of five metaplastic patterns.

a) Papillary immature metaplasia (PIM) or immature condyloma. xi

b) Flat immature metaplasias with mild atypia.xii

c) Flat reserve cell atypias with columnar differentiation and mild atypia.

d) Flat immature metplastic HSILs. xiii

e) Stratified mucin producing intraepithelial lesions. xiv

Lesions in categories a-c can be followed with repeat cytology; categories d&e require cone biopsy. All but category a will be strongly positive for p16; categories a-c will generally have a lower Mib-1 index than d&e.

f) Biomarkers are helpful for ascertaining if a lesion is present, but not the grade of the lesion, with exception of SILs showing negative or patchy staining (eg exophytic condyloma). These latter examples will usually be associated with lower risk HPVs. Otherwise the distinction between a flat LSIL and a flat HSIL is principally morphologic.xv


The following are situations in which the pathologist can be fooled when evaluating lower genital tract disease.

a) Giant condylomas of the cervix can be misclassified as malignancies in reproductive age women.

b) Multiple fragments of extraneous tissue (floater). These can be a problem in the setting of excluding ectopic pregnancy (mature villi), and endometrial carcinoma (floaters in curettings). They have a high medical legal import.

c) Biopsy artifacts in some settings will be mistaken as invasive cancer when juxtaposed to a co-existing squamous intraepithelial lesions.

d) Endometrial stromal tumors can mimic aglandular functionalis

e) Exceedingly subtle placental site trophoblastic tumors.

f) Spindle cell vaginal neoplasms: spindle cell epithelioma

g) Spindle cell or poorly differentiated vulvo/vaginal neoplasms: malignant melanoma

g) Clerical errors in reports. Most errors are picked up by one of the following individuals: the resident (in a training program), secretary, pathology attending and clinician. In rare instances all of these persons could miss an error. For this reason, virtually all climactically significant mistakes are the product of two or more detection failures.

i Medeiros F, Nascimento AF, Crum CP. Early vulvar squamous neoplasia: advances in classification, diagnosis, and differential diagnosis. .Adv Anat Pathol. 2005 Jan;12(1):20-6.

ii Hart WR. Vulvar intraepithelial neoplasia: historical aspects and current status. Int J Gynecol Pathol. 2001 Jan;20(1):16-30.

iii Ulbright TM, Stehman FB, Roth LM, Ehrlich CE, Ransburg RC.Cancer. Bowenoid dysplasia of the vulva.1982 Dec 15;50(12):2910-9.

iv Raju RR, Goldblum JR, Hart WR. Pagetoid squamous cell carcinoma in situ (pagetoid Bowen's disease) of the external genitalia Int J Gynecol Pathol. 2003 Apr;22(2):127-35.

v McCluggage WG, Jamison J, Boyde A, Ganesan R Vulval intraepithelial neoplasia with mucinous differentiation: report of 2 cases of a hitherto undescribed phenomenon. Am J Surg Pathol. 2009 Jun;33(6):945-9.

vi Nascimento AF, Granter SR, Cviko A, Yuan L, Hecht JL, Crum CP. Vulvar acanthosis with altered differentiation: a precursor to verrucous carcinoma? Am J Surg Pathol. 2004 May;28(5):638-43.

vii Jones RW, Sadler L, Grant S, Whineray J, Exeter M, Rowan D. Clinically identifying women with vulvar lichen sclerosus at increased risk of squamous cell carcinoma: a case-control study. J Reprod Med. 2004 Oct;49(10):808-11

viii Galgano MT, Castle PE, Stoler MH, Solomon D, Schiffman M. Can HPV-16 genotyping provide a benchmark for cervical biopsy specimen interpretation? Am J Clin Pathol. 2008 Jul;130(1):65-70.

ixCrum CP. Laboratory management of CIN 2: the consensus is consensus. Am J Clin Pathol. 2008 Aug;130(2):162-4.

x Paavonen J, Naud P, Salmerón J, Wheeler CM, et al. Efficacy of human papillomavirus (HPV)-16/18 AS04-adjuvanted vaccine against cervical infection and precancer caused by oncogenic HPV types (PATRICIA): final analysis of a double-blind, randomised study in young women. Lancet. 2009;374:301-14.

xi Trivijitsilp P, Mosher R, Sheets EE, Sun D, Crum CP.Papillary immature metaplasia (immature condyloma) of the cervix: a clinicopathologic analysis and comparison with papillary squamous carcinoma. Hum Pathol. 1998 Jun;29(6):641-8

xii Ma L, Fisk JM, Zhang RR, Ulukus EC, Crum CP, Zheng W. Eosinophilic dysplasia of the cervix: a newly recognized variant of cervical squamous intraepithelial neoplasia. Am J Surg Pathol. 2004 Nov;28(11):1474-84

xiii Iaconis L, Hyjek E, Ellenson LH, Pirog EC. p16 and Ki-67 immunostaining in atypical immature squamous metaplasia of the uterine cervix: correlation with human papillomavirus detection. Arch Pathol Lab Med. 2007 Sep;131(9):1343-9.

xiv Park JJ, Sun D, Quade BJ, Flynn C, Sheets EE, Yang A, McKeon F, Crum CP. Stratified mucin-producing intraepithelial lesions of the cervix: adenosquamous or columnar cell neoplasia? Am J Surg Pathol. 2000 Oct;24(10):1414-9.

xvNucci MR, Crum CP. Redefining early cervical neoplasia: recent progress. Adv Anat Pathol. 2007 Jan;14(1):1-10.

Selected Dilemmas in Lower Genital Tract Pathology

Christopher P. Crum, MDDivision of Women’s and Perinatal Pathology

Brigham and Women’s HospitalBoston, MA


Christopher P. Crum, MD

Division of Women’s and Perinatal Pathology

Brigham and Women’s Hospital and Harvard Medical School

VULVAR CARCINOMA MODEL

40%35-65STDHPV

Classic VINModerate to Poorly Diff SCC

60%55-85

Lichen Sclerosus and LSCp53 mutations

Differentiated VINKeratinizing SCC

Crum CP. Carcinoma of the vulva: Epidemiology and pathogenesis. Obstet Gynecol 79:448-454, 1992

HPV-Positive Usual (Classic) VIN

Characterized by full or near full-thickness atypia

Nuclear enlargement, multinucleation, abnormal mitoses

HPV 16

Classic VIN (CIS)

Classic VIN with Lichen Simplex Chronicus-Like Change

Subtle Classic HIVIL (bowenoid dysplasia)

Pagetoid HIVIL HIVIL with Columnar Differentiation

Biomarker Staining for Classic VIN

• P16 – relatively specific for classic VIN

• Caveats– Some cases demonstrate predominately

cytoplasmic staining

– Heterogeneity

Riethdorf et alMedeiros et al

Classic/Usual VIN• Full or near full-thickness

atypia• Diffuse horizontal p16ink4

positive• P16 staining parallels the

level of differentiation

Case 12

p16

Classic VIN – p16 Utility of p16 Staining

• Will be weak or negative in all variants of exophytic Low grade lesions

• May be strong in flat condylomata (VIN 1)

• Is occasionally helpful in evaluating margins with subtle atypias.

• Resolving problematic atypias.

Subtle Lesions Margins

Differentiated VIN (HPV negative)

Less well defined category

Exhibits one or more of the following features

Atypia confined to the first 2-3 cell layers (basal atypia)

Discrete basal/parabasal cellularity/hyperchromasia

Acantholysis

Abnormal cell maturation with abnormal keratinization

Case 5

Differentiated VINHPV negative VSCC

Neither precursor nor invasive cancer may exhibit surface atypia

p53 MUTATIONS IN DVIN AND VULVAR SCC • Is not well understood

due to molecular heterogeneity in both cancers and precursors

• Is implied in some cases by strong co-staining with p53 but the role of p53 mutations is unknown

• Has not been established by p53 mutation analysis

Lin et al 1998; Pinto et al 1999,2000; Yang and Hart 2000

Differentiated VIN – p53

Results

Differentiated VINs contain p53 mutations

Laser capture microdissected DNAsfrom both lower and upper epithelial layers contain the mutations

•P53 staining is uniquely basal or immature cell-specific

dVINdVINdVIN

743G>GA743G>GA743G>GA

R248RQR248RQR248RQ

p53 p16

A

B

C

A) 746G>GA, 415A>ATB) 746G>GA, 415A>ATC) 746G>GA

T) 746G>GA

T

Case 5

p53

p53

p53p53

P53 is normally expressed in the basal cells of squamous epithelium and decays with maturation

This pattern of expression is identical to p63, which is a marker of keratinocyte basal cells

In both benign and neoplastic squamous epithelium, p53 and p63 are concordant in their distribution

Thus, in squamous epithelium (unlike glandular neoplasms) mutant p53 accumulates in the basal (type) cell only and is degraded as function of maturation

Nl

dVIN

VSCC

p53 p63

There is a link between p53 mutations and a subset of p53 immunopositive dVINs described by Yang and Hart.

However, p53 staining will not identify all dVINs with p53 mutations.

Because of the frequency (and sometimes multiplicity) of p53 mutations in the vulvar epithelium, the predictive value of a given p53 mutation for a SCC outcome is unclear.

Other Alterations that may Confer Increased Risk

• LSA with superimposed LSC

• Verruciform lichen simplex chronicus (non-specific)

• Verruciform acanthosis with altered differentiation

Verruciform Lichen Simplex Chronicus

Vulvar Acanthosis with Altered Differentiation

• Often seen adjacent to verrucous carcinoma

• Five components– Verruciform architecture (variable)– Non-invasive– Minimal nuclear atypia– Multilayered-parakeratosis– Superficial epithelial cell pallor

Nascimento et al

VAAD

VAAD VAAD

Multi-layeredparakeratotic sheets

Cytoplasmic “pallor” withloss of keratohyaline granules

VAAD is distinct from differentiated (simplex) VIN

Basal layeratypia

Abnormal keratinization

Differentiated VIN VAADNascimento et al

VAAD is distinct from classic VIN

Full-thicknessatypia

Classic VIN VAAD Nascimento et al

VAAD is distinct from verrucous carcinoma

Penetrating edges

Verrucous carcinoma VAADNascimento et al

Vulvar epithelium

HPV 16 infection

Classic VIN

Invasive Squamous Cell Ca

Inflammatorydermatoses

VAAD

Verrucouscarcinoma

Differentiated(simplex) VIN

Inflammatorydermatoses

Pathogenesis of Vulvar Carcinoma


Patterns of Invasion

• PD infiltrative

• PD nested

• WD infiltrative

• WD nested

• WD blunt

• VC blunt

Mimics

• PD nested or infiltrative– Tangentially sectioned epithelium

– Inflammatory artifacts in VIN

• WD infiltrative– Pseudoepitheliomatous hyperplasia

– Isolated foci of dysmature epithelium

• WD nested– Severe reactive inflammatory

Mimics• PD nested or infiltrative

– Tangentially sectioned epithelium



Non-invasive Invasive


– Inflammatory artifacts in VIN

Mimics• WD nested or infiltrative


Mimics• WD nested or

infiltrative– Tangentially

sectioned epithelium

Mimics

• WD infiltrative– Pseudoepitheliomatous hyperplasia

PEH PEH with a differentiating nest WD SCC

Mimics• Other problems

– Are we there yet?NO Definitely

Topic 3: CIN or immature metaplasia?

Issues

• The spectrum of CIN

• Accuracy of grading

• Dynamics of HPV infection over time

• How frequent is “true” progression from LSIL to HSIL?

Understanding Early Cervical Neoplasia

• Our job is to:– Determine what is and what is not a precursor

lesion

– Grade the lesion to guide management• HSIL = LEEP

• LSIL = Follow

– Avoid the over-diagnosis of HSIL

– Manage the more recently described “metaplastic spectrum” of CIN

p63 Localizes to HPV Target Cells

Transformation Zone Ectocervix

Indifferent Squamous Columnar Squamous

Basal/Reserve Cells in the Cervix

Important Points

• CIN has been re-defined over the years, largely as a function of therapeutics

• Classic diagrams of CIN are a simplification

• The transformation zone has an important impact on the presentation and interpretation of CIN

Classification Systems and Management

1960s

1970-80s

1990s

Mild Moderate Severe CIS

Condyloma/CIN 1 CIN 2 CIN 3

LSIL HSIL HSIL

Follow Cone/Hysterectomy

Cryotherapy Laser/Cone

Follow LEEP

2007+LSIL HSIL HSIL

Follow LEEP

?

Mature Immature Reserve

Normal

Low

High

Classification

• LSIL– Corresponds to those lesions with milder

forms of atypia and can be followed

• HSIL– Corresponds to those lesions with specific

patterns of atypia that reflect the biologic effects of viral oncogenes


Normal

Low

High

Categories of LSIL

Cytopathic effectVariations in Size/StainingCorrespond to LSIL in the cytologic smear

Mild atypia in the lower third of the epitheliumIndicates that the parabasal cells have not undergone a significant morphologic transformation


Normal

Low

High

HSILParabasal nuclear enlargement, differences in size and staining with coarse chromatin, corresponding to higher grade cells on smearReflects fundamental change in the biology of the replicating cell population

What do all SILs have (generally) in common?

• Nuclear atypia – variations in nuclear size and staining

• Increased nuclear density in the upper epithelial cell layers – very helpful.

Non-classic SILs

• SILs highlighting the metaplastic-columnar transition– Immature metaplastic LSIL – Immature condyloma

– Immature metaplastic HSIL

– Partially mature metaplastic SIL (“Eosinophilic dysplasia”)

– Microglandular SIL

– Stratified mucin-producing intraepithelial lesions (SMILE)

Immature Condyloma

• Imagine infecting immature epithelium with a low risk HPV

• Resembles condyloma with papillary architecture

• Koilocytosis is not obvious because the cells cannot mature

• Regular nuclear spacing with nucleoli• Low Ki-67 index


Normal

Low

High

Papillary Immature Metaplasia (LSIL)


Normal

Low

High

Mild atypias (SIL) in metaplastic epithelium (eosinophilic dysplasia)

Zheng et al, 2004

SIL in Reserve Cells (LSIL)

SIL in Microglandular Change (Reserve cell SIL)


Normal

Low

High

Immature metaplastic Phenotype (HSIL)


Normal

Low

High

SIL with Columnar Differentiation

SIL with Columnar Differentiation (SMILE)

Immature columnar cells

Reserve Cells

Immature metaplastic cells

Mature metaplastic cells

HPV can transform epithelial cellsat any point in this spectrum of differentiation. For this reason, you can expect a wide range of histologic patterns. The distinction of low from high grade lesions is based on distribution and severity of atypia

Biomarker Staining

• P16 – Particularly useful for immature epithelia in reproductive age women

• MiB-1 – Atrophic background

• We use neither when the differential diagnosis is LSIL vs Normal

• P16 immunostaining will not discriminate LSIL from HSIL.

Atrophy +HSILMiB1 p16

MiB1 p16

Reactive

MiB1 p16

HSIL (immature met phenotype)

LSIL (immature met phenotype)Eosinophilic Dysplasia

Zheng et al, 2004

Eosinophilic Dysplasia

Zheng et al, 2004

Immature columnar cells

Reserve Cells

Immature metaplastic cells

Mature metaplastic cells

Ascertaining Outcome Risk

• Most high risk HPVs will not result in an HSIL (CIN3) outcome (Kahn)

• 40% or more of confirmed CIN2 biopsies will be followed by regression in women under age 25 (Crum, unpublished)

• The risk of HSIL in women with mild abnormalities and negative colpo or a biopsy of CIN1 is 11% (Cox)

Prospective Risk of ≥CIN3 (1)

Follow-up time (years)

1 2 3 4 5 6 7 8 9 10 11

Cu

mu

lati

ve i

nci

den

ce r

ate

(%)

0

5

10

15

20

HPV16+

HPV18+

HC2+

HR HPV-

Portland Khan et al., JNCI

Progression

• Must be defined in the context of– The transition in question:

• LSIL- HSIL

• HSIL-Malignancy

– How LSIL or CIN1 is defined

– How many HPVs are involved

– The reliability of the pathologic interpretation

Defining CIN1

1970 2000

Possible Outcomes

• Lesion disappears following biopsy

• Lesion transiently persists then disappears

• Lesion persists until cone biopsy

• More than one lesion is present, with persistence/regression of one or more

• New infections develop during follow-up and may or may not contribute to pathology

HPV Status and Outcome

• Viral parameter HSIL LSIL Nl

• Persistence of one 62 22 27

• Additional types 27 78 34

• Replacement types 6 39 7

• Clearing of all 3 0 50

Summary

• Persistence of a single HPV type significantly influenced risk of HSIL– 62% of HSILs had a single HPV type

throughout the study

• The presence of new or replacement HPV types correlates significantly with a CIN 1 outcome– 78% of CIN1 outcomes were associated with

one or more additional HPVs during the study

Frequency of True Progression

• 12% of LSILs are followed by an HSIL at 2 years (Cox et al).

• 1% of LSILs (CIN1) progress to carcinomas (Östor’s review)

• What percent are true progressions from LSIL to HSIL (CIN2)?

Assessing Progression

• Ascertain percentage of biopsy proven LSIL resulting in an HSIL outcome

• Rate of HSIL outcome verification by biopsy review

• Compare to rate of HSIL outcome verification in a random sample

Chen E, and Crum CP, unpublished

Results

• 29/264 LSIL biopsies followed by HSIL outcome on report (11%).

• 22/24 reviewed confirmed initial LSIL

• 5/17 with outcome review (30%) confirmed the diagnosis of CIN2 or higher

• 42/50 (84%) randomly reviewed cases confirmed the original diagnosis

Chen E, and Crum CP, unpublished

CONCLUSION1. The most likely explanations for “progression” from LSIL to HSIL, based on record and histologic review, are, in decreasing frequency:

a) Over-diagnosis of HSIL on subsequent biopsy/cone (especially when the outcome diagnosis is CIN2).

b) Change in HPV type over time (based on p16 stain discrepancy)

c) Under-diagnosis of HSIL on initial biopsy

2. Studies that use progression from LSIL to HSIL as an endpoint must take the above into account and any study that claims differences in progression rates must be viewed critically with the above possibilities in mind.

CONCLUSION3. A diagnosis of HSIL on a follow-up biopsy following an initial biopsy diagnosis of LSIL is more likely to represent a misclassification than a routine diagnosis of HSIL.

i.e. Such a diagnosis should be subject to quality assurance review.

Agreement by two or more observers on a diagnosis of CIN2 is recommended prior to proceeding to LEEP.


Trouble Spots We have Seen

• a) Giant condylomas of the cervix misclassified as malignancies.

• b) Extraneous tissues (floater).

• c) Biopsy artifacts.

• d) Endometrial stromal tumors vs. aglandular functionalis

• e) Exceedingly subtle PSTTs

• f) Poorly diff vaginal tumor-r/o melanoma

• g) Clerical errors in reports.


a) Giant condy-lomas of the cervix mis-classified as malignancies.


b) Extraneous tissues (floater).


c) Biopsy artifacts.


d) Endometrial stromal tumors vs. aglandular functionalis

Trouble Spots We have Seene) Exceedingly subtle PSTTs

Normal early gestation PSTT

Trouble Spots We have Seenf) Vaginal spindle cell tumors: exclude

spindle cell epithelioma (benign mixed tumor)

Trouble Spots We have Seeng) Vulvo-vaginal spindle cell tumors: exclude

malignant melanoma


g) Clerical errors in reports.

lower genital tract squamous lesions - ucsf departments...

Documents