Pathology was rereviewed for selected cases, including 2 cases with recurrence and matched primary, 2 cases with recurrence (for which the matched primary was unavailable for review), and 5 representative primary cases that were not complicated by recurrence. Pathology review was conducted in a blinded manner by one of the authors (P.W.H) who is a board-certified dermatopathologist for approximate depth of invasion from the granular layer, perineural invasion, bone invasion, infiltrative growth, presence of conventional squamous cell carcinoma, and margin status.
Statistical analysis was performed when appropriate using an N1 χ2 test or Student t test.
Results
Demographics and Comorbidities—The median age of the patients at the time of diagnosis was 55 years (range, 34–77 years). There were 12 males and 7 females (Table 1). Two patients were Black and 17 were White. Almost all patients had additional comorbidities including tobacco use (68%), alcohol use (47%), and diabetes (47%). Only 1 patient had an autoimmune disease and was on chronic steroids. No significant difference was found between the demographics of patients with recurrent lesions and those without recurrence.
Tumor Location and Clinical Presentation—The most common clinical presentation included a nonhealing ulceration with warty edges, pain, bleeding, and lowered mobility. In most cases, there was history of prior treatment over a duration ranging from 1 to 8 years, with a median of 5 years prior to biopsy-based diagnosis (Table 1). Six patients had a history of osteomyelitis, diagnosed by imaging or biopsy, within a year before tumor diagnosis. The size of the primary tumor ranged from 2.4 to 6 cm, with a mean of 4 cm (P=.20). The clinical presentation, time before diagnosis, and size of the tumors did not differ significantly between recurrent and nonrecurrent cases.
The tumor location for the recurrent cases differed significantly compared to nonrecurrent cases. All 5 of the patients with a recurrence presented with a tumor on the nonglabrous part of the foot. Four patients (80%) had lesions on the dorsal or lateral aspect of the great toe (P=.002), and 1 patient (20%) had a lesion on the low ankle (P=.09)(Table 1). Of the nonrecurrent cases, 1 patient (7%) presented with a tumor on the plantar surface of the great toe (P=.002), 13 patients (93%) presented with tumors on the distal plantar surface of the foot (P=.0002), and 1 patient with a plantar foot tumor (Figure 1) also had verrucous carcinoma on the thumb (Table 1 and Figure 2).
FIGURE 1. Typical clinical features of nonrecurrent verrucous carcinoma.
Histopathology—Available pathology slides for recurrent cases of verrucous carcinoma were reviewed alongside representative cases of verrucous carcinomas that did not progress to recurrence. The diagnosis of verrucous carcinoma was confirmed in all cases, with no evidence of conventional squamous cell carcinoma, perineural invasion, extension beyond the dermis, or bone invasion in any case. The median size of the tumors was 4.2 cm and 4 cm for nonrecurrent and recurrent specimens, respectively. Recurrences displayed a trend toward increased depth compared to primary tumors without recurrence (average depth, 5.5 mm vs 3.7 mm); however, this did not reach statistical significance (P=.24). Primary tumors that progressed to recurrence (n=2) displayed similar findings to the other cases, with invasive depths of 3.5 and 5.5 mm, and there was no evidence of conventional squamous cell carcinoma, perineural invasion, or extension beyond the dermis.
FIGURE 2. Verrucous carcinoma of the thumb.
Treatment of Nonrecurrent Cases—Of the 16 total cases treated at the University of Michigan, surgery was the primary mode of therapy in every case (Tables 2 and 3). Of the 11 nonrecurrent cases, 7 patients had wide local excision with a dermal regeneration template, and delayed split-thickness graft reconstruction. Three cases had wide local excision with metatarsal resection, dermal regeneration template, and delayed skin grafting. One case had a great toe amputation. Surgical margins were not reported in all the cases but ranged from 0.5 to 2 cm (8/11 [73%] reported). Three cases had positive margins at the time of primary resection; 2 were treated with further resection, and 1 had a below the knee amputation (BKA). Follow-up on average was 12 months, with a range of 3 to 36 months.
