Improving accuracy in NMSC diagnosis

July 14, 2016

The trend toward smaller biopsies may compromise the accuracy of dermatopathologists' diagnoses.

When assessing nonmelanoma skin cancer (NMSC), often the accuracy of a diagnosis depends upon the quality of a biopsy specimen.

No national registries exist for NMSC, says Whitney A. High, M.D., J.D., M.Eng., who spoke at MauiDerm 2016. He is associate professor of dermatology and pathology at the University of Colorado School of Medicine, Denver.

Dr. High says that there are about two to four million cases of basal cell carcinoma (BCC) and 500,000 cases of squamous cell carcinoma (SCC) diagnosed annually in the United States. The risk of BCC and SCC increase with age and UV exposure.

With BCC, Dr. High says that the tumor’s growth pattern drives treatment decisions.

"Superficial BCC can be treated with freezing, burning or even topical medications like imiquimod. Nodular BCCs are usually better managed with electrodesiccation and curettage or excision. Micronodular, infiltrative, desmoplastic and morpheaform BCC, and related aggressive patterns, are often best treated with Mohs micrographic surgery, particularly in high-risk areas or areas where tissue conservation is important," he adds.

Dr. High, who is both a dermatologist and dermatopathologist, expresses concern that a trend toward smaller biopsies may impact the assessment of a growth pattern, and hence, appropriate management. A shallow shave may not encapture enough of the tumor to reveal the predominant growth pattern, he explains, leading to erroneous conclusions. 

A retrospective study of 500 randomly selected skin biopsies showed that as the percentage of shave biopsies rose from 37% to 68% between 1988 and 2005 (perhaps due to patients' aesthetic concerns), the mean size of shave biopsy specimens fell from 65 mm3 to 34 mm3.1

A recent study in the area of melanoma showed that shaves and punches were associated with misdiagnosis and microstaging inaccuracy, at least in comparison to excisions.2 More common than misdiagnosis, say these authors, is the issuance of nonspecific, descriptive diagnoses that dermatopathologists must offer when samplings are insufficient.

To optimize the histologic assessment of neoplasms, Dr. High suggests that contributing dermatology professionals do what he refers to as "biopsies in 3D," to include the following information, to assist in making accurate diagnoses:

  • Description (of the tumor and what type of biopsy was done)

  • Diameter (size or extent)

  • Diagnosis (made by the clinician)

The procurement of an adequate biopsy yields a better, more confident diagnosis from a dermatopathologist, says Dr. High. For example, in a superficial biopsy it may be impossible to distinguish a benign syringoma from a malignant microcystic adnexal carcinoma.3

Next: SCC and staining

You may also like:

Pediatric dermatopathology

 

SCC and staining

Along with age and UV exposure, other risk factors for SCC include chemical exposures, nonhealing wounds and human papillomavirus (HPV) infection.

SCC caused by HPV may overexpress P16, and this can be exploited in the diagnosis, particularly of HPV-related SCC of the genitals ("Bowenoid papulosis"),4 says Dr. High. Immunostaining may also detect aberrant overexpression of P53, and this can also identify SCC.5

Dermatopathologists often classify SCC by its degree of histologic differentiation (well, moderately or poorly differentiated), and these designations also impart some measure of risk, with lesser differentiation associated with a lesser prognosis, Dr. High explains.

High-risk SCC may also include those located on the ear, lip and genitalia, or in chronic wounds (Marjolin's ulcer). Other characteristics of higher-risk lesions include pain, rapid growth, paresthesia and cranial nerve involvement.

Not all perineural invasion is equal, Dr. High says. "The size of the nerves involved matters." Recent articles have shown that involvement of small unnamed nerves is less prognostically significant than is invasion of larger named nerves.6 Also, the presence of other risk factors such as tumor diameter, lymphovascular invasion and invasion beyond subcutaneous fat may be important in predicting less favorable outcomes. 

"Often, the diagnosis of poorly differentiated SCC requires immunostaining," Dr. High says. When malignant spindle cells are encountered, immunostains for S100, SOX10, pancytokeratin, P63, CD10 and procollagen 1 are often utilized to include spindle-cell SCC, and exclude spindled/desmoplastic melanoma and atypical fibroxanthoma (AFX). "More than one immunostain may be needed to optimally include or exclude spindle-cell SCC," he adds.

AFX usually appears as dome-shaped lesions on the head or neck that exhibit low-grade malignant behavior. "Be careful that the sample is large enough to exclude a so-called pleomorphic sarcoma or malignant fibrocystiocytoma (MFH) that has been only barely encaptured in a shave," Dr. High cautions. The current standard treatment for AFX is surgical excision, he says, although Mohs micrographic surgery may be appropriate in locations that require tissue sparing.

"I have heard it explained that since the putative etiologic cell is unknown, AFX is, in large part, a ‘diagnosis by convention.’  It includes all malignant-appearing lesions that do not mark with immunostains in a manner consistent with melanoma or spindle-cell SCC," Dr. High emphasizes.

"There is an old adage in dermatopathology, and in pathology in a larger sense, that states:  ‘garbage in = garbage out.’  If you put poor specimens and incomplete data into the system, then you should not be surprised to get poor diagnoses and lesser opinions out of the system," Dr. High explains. He also laments that every effort should be made to provide representative samplings, and accurate clinical data, such that optimal histologic opinions and better dermatologic care can transpire.  

Disclosures: Dr. High reports no relevant financial interests.

References

1. Fernandez EM, Helm T, Ioffreda M, Helm KF. The vanishing biopsy: the trend toward smaller specimens. Cutis. 2005;76(5):335-9.

2. Ng JCSwain SDowling JPWolfe RSimpson PKelly JW. The impact of partial biopsy on histopathologic diagnosis of cutaneous melanoma: experience of an Australian tertiary referral service. Arch Dermatol. 2010;146(3):234-9.

3. Hoang MP, Dresser KA, Kapur P, et al. Microcystic adnexal carcinoma: an immunohistochemical reappraisal. Mod Pathol. 2008;21:178-185.

4. Wang H, Sun R, Lin H, Hu WH. P16INK4A as a surrogate biomarker for human papillomavirus-associated oral pharyngeal carcinoma: consideration of some aspects. Cancer Sci. 2013;104:1553-1559.

5. Homann NAndl TNees MSchuhmann AHerold-Mende CBosch FX. Significance of aberrant p53 protein in head-neck tumors and its effect on proliferation and differentiation. HNO. 1993;41(5):254-60.

6. Carter JB, Johnson MM, Chua TL, Karia PS, Schmults CD. Outcomes of primary cutaneous squamous cell carcinoma with perineural invasion: and 11-year cohort study. JAMA Dermatol. 2013;149:35-41.