LAUSR

DEAR EDITOR, Multiple surgical and nonsurgical treatments for cutaneous squamous cell carcinoma in situ (SCCIS), sometimes called Bowen disease, have been described. The best treatment option has not yet been determined. Efficacy comparison studies are rare and difficult to execute due to the variety of treatment protocols and follow-up periods. In addition, choice of treatment depends on several factors, including body site, lesion size and thickness, age, immune status, comorbidities, patient preference and physician expertise. Given that SCCIS occurs more often in elderly patients and is frequently located on body sites with poor wound healing, noninvasive or topical therapies are often preferred. A recently published guideline from the British Association of Dermatologists gave strength of recommendation ‘A’ to photodynamic therapy, ‘B’ to nonsurgical treatments including cryotherapy, 5-fluorouracil and imiquimod, while surgical treatments were given grades ‘C’ and ‘D’. One argument against nonsurgical treatment is that it may treat without pathological confirmation of clearance, with a possibility that invasive SCC is harboured within a lesion. Our current study evaluated the frequency of invasive SCC in patients with initial biopsy-proven SCCIS undergoing surgical re-excision. This single-institution retrospective study (approved by the Boston University institutional review board) searched the Skin Pathology Laboratory database from January 2008 to December 2012 to identify all cases of initial biopsy-proven SCCIS that underwent subsequent re-excision. Re-excision specimen slides were reviewed by a board-certified dermatopathologist to evaluate the final histological diagnosis. The primary outcome measure was the frequency of invasive SCC, SCCIS, actinic keratosis (AK), scar or other diagnoses in re-excision specimens. Secondary outcome measures included patient demographics, including age, sex, location and size of lesion, type of referring physician (academic vs. nonacademic) and recurrence rate. Subgroup analyses compared clinical characteristics in groups harbouring invasive SCC compared with those that did not. Statistical analysis was performed using SAS version 9.4. (SAS Institute Inc., Cary, NC, USA). Exclusion criteria included initial biopsy showing additional primary findings and those with initial biopsy not in the Skin Pathology Laboratory database. We identified 726 cases of biopsy-proven SCCIS that underwent re-excision (detailed summary of clinical characteristics are available on request). Briefly, the cases were comprised of 56% males and 44% females with an average age at biopsy of 72 6 years. The majority (84%) of re-excision cases were from private physicians. Of the 726 total cases that underwent re-excision, 702 cases (96 7%) had a histological diagnosis of SCCIS, AK or scar, with the majority showing SCCIS (n = 429, 59 1%) and scar (n = 235, 32 4%). Of the 726 total cases, only 24 cases (3 3%) had an ‘upgraded’ diagnosis, with the majority showing well-differentiated SCC (n = 15, 62 5%). In subgroup analyses comparing the group with unchanged diagnoses from the group with upgraded diagnoses, there was no statistically significant difference in age of onset, sex, type of institution or lesion size (Table 1). A higher percentage of upgraded diagnoses were on the head and neck region, but differences in location of body were not statistically significant. The overall percentage of reported recurrences was low (n = 5, 0 69%), with no statistically significant difference between groups. This study demonstrates a very low frequency of invasive SCC in re-excisions of initial biopsy-proven SCCIS (3 3%) and suggests that surgery may not be necessary for the treatment of SCCIS. Only one prior study has evaluated the final histological diagnosis in biopsy-proven SCCIS following surgical intervention. Chuang et al. published a prospective study of 29 patients with biopsy-proven SCCIS who underwent Mohs excision, with n = 9 (31%) found to harbour invasive SCC on final histology. Invasive SCC was more likely found in lesions demonstrating clinical signs of residual tumour or those larger than 1 4 cm in diameter. The authors of this study argued that surgical treatment should be strongly considered for SCCIS. However, this study had several limitations, including small sample size, data from a single institution and inclusion of patients referred only for Mohs excision, suggesting possible selection bias. Our study failed to distinguish any clinical characteristics, including size of initial lesion, which might predict the harbouring of invasive SCC in SCCIS biopsy samples. In addition, cases with upgraded diagnoses did not have a higher recurrence rate compared with those with SCCIS. The advantages of this study include a large sample size and a broad dermatopathology database that includes both academic and private institutions. In addition, all excisions, rather than just Mohs excision, were included in the study sample, which may limit selection bias. There were some limitations of the study.