JAM ACAD DERMATOL VOLUME 76, NUMBER 2. Research Letters 351

Transcription

1 JAM ACAD DERMATOL Research Letters 351 Standard step sectioning of skin biopsy specimens diagnosed as superficial basal cell carcinoma frequently yields deeper and more aggressive subtypes To the Editor: Correct diagnosis of superficial basal cell carcinoma (BCC) is essential because of the increase of nonsurgical treatments for this subtype. Histologic confirmation by punch biopsy specimen for BCC diagnosis and its subtype is recommended. Although studies have shown the discordance between punch biopsy specimens and excision, there is a lack of research on the sampling error within punch biopsy specimens. 1,2 Moreover, a standardized method for histologic sectioning punch biopsy specimens is currently missing. The current histologic examination protocol of a 3-mm punch biopsy specimen, suspicious for BCC, at the Radboud University Medical Center (Nijmegen, The Netherlands) consists of evaluation of 2 to 3 hematoxylin-eosin-stained tissue sections (4 m) obtained from 1 level. If a BCC is detected in these sections, no additional levels will be cut. If not, the punch biopsy specimen is cut at 4 additional levels and evaluated. We compared the accuracy of histologic examination of only 1 level with a more extensive step-section method. In addition, we investigated whether tumor thickness ([0.4 mm), ulceration, and adnexal extension are determinants of treatment failure or recurrence in superficial BCCs. 3,4 This retrospective study was approved by the institutional review board of the Radboud University Medical Center, Nijmegen, The Netherlands. In all, 116 superficial BCC punch biopsy specimens, obtained in 2014 to 2015, were cut in 4 additional levels at an interval of 200 m. After every 200 m, 10 sections (4 m) were cut, of which the middle 2 sections were hematoxylin-eosin stained and evaluated. Finally, 5 levels per punch biopsy specimen were histopathologically evaluated (Fig 1). Tumor thickness was evaluated by measuring from the granular layer of the epidermis down to the deepest point of invasion (Breslow depth). Included patients were treated with imiquimod cream (n ¼ 23), 5-fluorouracil (n ¼ 19), excision (n ¼ 50), or methylaminolevulinate photodynamic therapy (MAL-PDT) (n ¼ 24) with a follow-up time until January 29, 2016 (Supplemental Table I). In 22.4% (n ¼ 26) a more aggressive BCC subtype was found in the additional hematoxylineosin-stained sections (Table I, Supplemental Table II, and Supplemental Fig 1). There were 2 clinical recurrences (5-fluorouracil, n ¼ 1, and MAL-PDT, n ¼ 1) and 2 treatment failures (MAL-PDT). In the treatment failure/recurrence group there was 1 underdiagnosed BCC (25%); this was the only superficial BCC thicker than 0.4 mm in this group, whereas none showed presence of ulceration or adnexal extension Fig 1. Histopathological examination method of a punch biopsy specimen suspicious for basal cell carcinoma. Schematic overview of the histopathological examination process. The dimensions of the punch biopsy specimen might be smaller than depicted in this figure because of shrinkage after formalin fixation and histologic processing, embedding, and mounting. X: These sections are stained with hematoxylin-eosin and used for histopathological examination. In the initial sections it is unknown which slices are stained and examined.

2 352 Research Letters JAM ACAD DERMATOL FEBRUARY 2017 Table I. Histopathological diagnosis and clinical treatment failure or recurrence Histopathological diagnosis of punch biopsy specimen, n (%)* sbcc nbcc ibcc mnbcc n/mnbcc n/ibcc n/mn/ibcc Total More aggressive subtype missed 90 (77.6) 16 (13.8) 1 (0.9) 1 (0.9) 4 (3.4) 1 (0.9) 3 (2.6) 116 (100.1 y ) 26 (22.4) Clinical treatment failure z or recurrence x Patient Treatment failure or recurrence Location Presence of clinical ulceration Presence of adnexal extension sbcc thickness, mm k Initial treatment Histopathological diagnosis of second punch biopsy specimen { Histopathological diagnosis of excision specimen 1 Recurrence Lower No No FU sbcc sbcc sbcc extremities 2 Recurrence Lower No No 0.20 MAL-PDT sbcc sbcc sbcc extremities 3 No Trunk No No 0.33 MAL-PDT Not performed sbcc sbcc response Head No No 0.74 MAL-PDT n/mnbcc n/mnbcc sbcc 4 Partial response Histopathological diagnosis of punch biopsy specimen # FU, Fluorouracil; ibcc, infiltrative basal cell carcinoma; MAL-PDT, methylaminolevulinate photodynamic therapy; nbcc, nodular basal cell carcinoma; n/mnbcc, micronodular basal cell carcinoma; sbcc, superficial basal cell carcinoma. *Based on histopathological evaluation of 5 levels per punch biopsy specimen, all biopsy specimens based on 1 level evaluation were diagnosed as sbcc. y Numbers do not add up to 100 because of percentages round off. z Clinical treatment failure: partial or no response after initial treatment. x Presence of tumor tissue detected during follow-up after previous tumor clearance. k Largest measurement taken from either the initial or additional hematoxylin-eosin-stained sections. { Extra punch biopsy specimen taken after treatment failure/recurrence for diagnosis and determination of excision margin for treatment. # Based on histopathological evaluation of 5 levels per punch biopsy specimen. (Table I). We did not found an association of tumor thickness, ulceration, or adnexal extension with treatment failure/recurrence. The mean follow-up period was of SD 194 days with a median of 301 days. Hoogedoorn et al 5 found that more than 50% of the treatment failures of superficial BCC, after MAL-PDT, were a result of underdiagnosis of the primary punch biopsy specimen and that in approximately 50% of the recurrences a mixed type BCC was present. Their median follow-up was 2 years, in which they reported higher rates of treatment failures and recurrences. Our limited follow-up time might explain the current lack of association between presence of more aggressive BCC subtypes in punch biopsy specimens and treatment failure/recurrence. This study shows that histologic examination of only 1 level from a punch biopsy specimen leads to underdiagnosis of more aggressive BCC subtypes in biopsy specimens diagnosed as superficial BCCs. We recommend step sectioning to reduce this risk and to prevent undertreatment. We would like to thank J. C. M. Hendriks and E. Bronkhorst for their help with the statistical analysis. Kim P. Nguyen, MD, a G. Jimmy Knuiman, MD, b Piet E. J. van Erp, PhD, a WillekeA.Blokx,MD, PhD, b Malou Peppelman, PhD, a and Marie- Jeanne P. Gerritsen, MD, PhD a Departments of Dermatology a and Pathology, b Radboud University Medical Center, Nijmegen, The Netherlands Drs Nguyen and Knuiman contributed equally to this work. Supplemental material is available at jaad.org. Funding sources: None. Conflicts of interest: None declared. Correspondence to: Kim P. Nguyen, MD, Department of Dermatology, Radboud University Medical Center, PO Box 9101, NL 6500 HB Nijmegen, The Netherlands. T.nguyen@radboudumc.nl REFERENCES 1. Wolberink EA, Pasch MC, Zeiler M, van Erp PE, Gerritsen MJ. High discordance between punch biopsy and excision in

3 JAM ACAD DERMATOL Research Letters 353 establishing basal cell carcinoma subtype: analysis of 500 cases. J Eur Acad Dermatol Venereol. 2013;27(8): Kamyab-Hesari K, Seirafi H, Naraghi ZS, et al. Diagnostic accuracy of punch biopsy in subtyping basal cell carcinoma. J Eur Acad Dermatol Venereol. 2014;28(2): Roozeboom MH, van Kleef L, Arits AH, et al. Tumor thickness and adnexal extension of superficial basal cell carcinoma (sbcc) as determinants of treatment failure for methylaminolevulinate (MAL)-photodynamic therapy (PDT), imiquimod, and 5-fluorouracil (FU). J Am Acad Dermatol. 2015;73(1): McKay KM, Sambrano BL, Fox PS, Bassett RL, Chon S, Prieto VG. Thickness of superficial basal cell carcinoma (sbcc) predicts imiquimod efficacy: a proposal for a thickness-based definition of sbcc. Br J Dermatol. 2013;169(3): Hoogedoorn L, Hendriks JC, Knuiman GJ, et al. Treatment failure in superficial basal cell carcinoma following treatment with photodynamic therapy: is this a result of underdiagnosis? J Eur Acad Dermatol Venereol Epub 2016/06/04. Incidence ratio of basal cell carcinoma to squamous cell carcinoma equalizes with age To the Editor: In the United States, the incidence ratio of basal cell carcinomas (BCCs) to cutaneous squamous cell carcinomas (SCCs) is traditionally taught as 4:1. 1,2 However, the origin of this ratio is unclear, as earlier US-based studies ( published before 1989) report an incidence ratio ranging from 1.4:1 to 9:1. 3 Recently, Rogers and colleagues 4 reported a 1:1 incidence ratio of BCC:SCC using procedural data from the predominantly Caucasian, elderly Medicare fee-for-service population. To evaluate this incidence ratio across all age groups, we performed a retrospective review of BCC and SCC incidence in a northern Californian population. Furthermore, to corroborate our findings, we assessed the lifetime prevalence of BCC and SCC in a larger independent cohort. This study was approved by the institutional review board of Stanford University. To calculate incidences, we analyzed all 8032 diagnostic BCC and SCC dermatopathology reports collected at Stanford Healthcare from 2005 to 2015 (set 1). Patients of all ethnicities were included and 86% were Caucasian. To calculate lifetime prevalence, we used self-reported BCC and SCC cases from 23andMe (set 2), totaling 12,945 patients with BCC, 6579 patients with SCC, and over 250,000 control subjects, all of whom were Caucasian. Accuracy of self-reported answers was previously validated via comparison with medical records in a subset of patients. 5 Analysis of set 1 demonstrates an overall BCC:SCC incidence ratio of 1.4:1 (Table I). In the younger age groups, BCCs heavily outnumber SCCs; however, the Table I. Basal and squamous cell carcinoma biopsy counts stratified by age group, 2005 to 2015 Age, y # [60 Combined BCC SCC BCC:SCC 2.0:1* 5.0:1 2.3:1 1.1:1 1.4:1 BCC, Basal cell carcinoma; SCC, squamous cell carcinoma. Data source: Dermatopathology reports for patients seen and biopsied at Stanford Hospital and Clinics, from June 2005 to June Among patients with BCC, roughly 58% were male and 90% were Caucasian; for SCC cases, these values are 62% and 85%, respectively. Cases defined as recurrences were excluded. For more information about this cohort, please contact the corresponding author (ksarin@stanford.edu). *Small sample size limits accuracy. ratio equalizes as age increases, reaching 1.1:1 in the age group older than 60 years. Set 2 reveals a lifetime prevalence of 11% and 6% for BCC and SCC, respectively, in Caucasians older than 60 years (Table II), consistent with prior studies. 6 As expected for low-mortality diseases, the lifetime prevalence of these cancers increases steadily with age. The overall BCC:SCC incidence ratio of 1.4:1 in this predominantly Caucasian population is significantly lower than 4:1 and corroborates the recent finding of a 1:1 ratio among Medicare beneficiaries. 4 Because these cancers have low mortality and can occur multiple times per individual, the overall lifetime prevalence ratio is expected to approximate the overall incidence ratio yet also demonstrate a slight skew away from the cancer with the greater net increase in incidence across age groups (SCC in this case). Hence, our finding of a 2.0:1 overall lifetime prevalence ratio in an independent cohort substantiates our incidence results. Notably, we included both invasive SCC and SCC in situ under the term SCC. If we exclude SCC in situ from set 1, then the BCC:SCC ratio in the group older than 60 years increases to 1.8:1, which is greater than the corresponding value of 1.2:1 found in the Medicare study. 4 This difference is unsurprising considering that SCC in situ is often treated nonsurgically (and is consequently underreported in Medicare claims data). In addition, the high average ultraviolet index in California likely influenced rates of SCC and SCC in situ in our study. Strengths of this study include large sample size for lifetime prevalence estimates and incidences based on dermatopathology reports. Overall, we demonstrate that the BCC:SCC incidence ratio is much closer to 1:1 than 4:1ethus, clinicians should increase their index of suspicion for SCC, particularly when evaluating older patients.

4 J AM ACAD DERMATOL Research Letters 353.e1 Supplemental Fig 1. Detection of various basal cell carcinoma subtypes and adnexal extension in hematoxylin-eosin-stained sections. A, Superficial component in the initial section of punch biopsy specimen 24 (Supplemental Table II). B to E, In the additional sections of the same punch biopsy specimen (micro-) nodular and infiltrative subtypes are detected. F, Some lesions showed an adnexal extension (arrow). (Original magnification: 35.)

5 353.e2 Research Letters JAM ACAD DERMATOL FEBRUARY 2017 Supplemental Table I. Demographics Patient characteristics N (%) Mean 6 SD Gender Male 43 (48.3) Age, y Tumor characteristics N (%) Mean 6 SD Location Head and neck 15 (12.9) Upper extremities 11 (9.5) Trunk 64 (55.2) Lower extremities 26 (22.4) Size \10 mm 88 (75.9) [10 mm 28 (24.1) Presence of clinical ulceration Yes 1 (0.1) Presence of adnexal extension Yes 13 (11.2) Tumor thickness, mm Initial H&E-stained sections Additional H&E-stained sections * Treatment Imiquimod 23 (19.8) 5-FU 19 (16.4) MAL-PDT 24 (20.7) Excision 50 (43.1) Follow-up time, d y z FU, Fluorouracil; H&E, hematoxylin-eosin; MAL-PDT, methylaminolevulinate photodynamic therapy. *P value when compared with the initial H&E-stained sections was.001. y Until February 1, z Median follow-up period.

6 JAM ACAD DERMATOL Research Letters 353.e3 Supplemental Table II. Histopathological evaluation: presence of additional basal cell carcinoma subtypes* H&E-stained sections in which additional BCC subtypes are present* Punch biopsy specimen Initial H&E First additional level Second additional level Third additional level Fourth additional level 1 No No No No Yes 2 No Yes Yes Yes Yes 3 yz No Yes No No No 4 No No No Yes Yes 5 No No Yes Yes Yes 6 No No Yes No Yes 7 No Yes Yes Yes Yes 8 No No Yes Yes Yes 9 No No No Yes Yes 10 No Yes Yes Yes Yes 11 No Yes Yes Yes Yes 12 No No Yes Yes Yes 13 No No No Yes Yes 14 No Yes Yes Yes No 15 No Yes Yes Yes Yes 16 No No No No Yes 17 yx No No No Yes No 18 No Yes No No Yes 19 No No No Yes Yes 20 No Yes Yes Yes Yes 21 No Yes Yes Yes Yes 22 No Yes No No Yes 23 No No Yes Yes Yes 24 No No Yes Yes Yes 25 No No No No Yes 26 No No No Yes Yes Overall presence of 0 11 (42.3) 14 (53.8) 19 (73.1) 23 (88.5) additional BCC subtypes (%) BCC, Basal cell carcinoma; H&E, hematoxylin-eosin. *Other than superficial basal cell carcinoma subtype. y Other BCC subtype(s) missed when evaluating only second and fourth additional level. z Other BCC subtype(s) missed when evaluating only third and fourth additional level. x Other BCC subtype(s) missed when evaluating only first and fourth additional level.