Development of squamous cell carcinoma in venous ulcers

Abstract

Apart from numerous case reports, studies on the epidemiology and the outcome of malignant transformation in venous leg ulcers are scarce. Even the aetiology is unsettled, and the genetic effects of the chronic ulcer state only partly studied. These aspects of the phenomenon of squamous cell carcinoma (SCC) in venous ulcers were investigated in the present study. Data on a cohort of 1, 170 patients with a diagnosis of venous ulcers from the Department of Dermatology, Karolinska Hospital, Stockholm, were matched with the Swedish Cancer Registry. Four cases of squamous cell carcinoma in venous ulcers were found, and the relative risk (RR) for these patients of having a squamous cell cancer on their lower limb was significantly increased (RR = 5.89; 95% Cl: 1.6 - 15.1). The same methods on a cohort consisting of data from the Swedish In-patient Registry on 10,913 patients with the diagnosis venous ulcer resulted in 23 cases of SCC, of which 17 were considered certainly secondary to venous ulcers and six probably secondary. Calculations done on the 17 cases revealed significantly increased RR (RR = 5.8; 95% Cl 3.1 - 9.3). The use of incidence data from the Cancer Registry might lead to a false low estimate of the RR. The conclusion was that SCC is indeed a complication of venous leg ulcers and the risk is approximately fivefold compared to the general population of Sweden.

The case histories of 25 patients with SCC in venous ulcers were compiled and histological slides re-examined. To estimate survival rates these data were compared with data from the Cancer Registry of 433 patients with squamous cell carcinoma on the lower limb. Twenty-three of the patients were dead. The mean age at cancer diagnosis was 78,5 years, and the median duration of the ulcer before diagnosis of SCC was 25 years. All patients who had a poorly differentiated SCC died within a year from diagnosis. Survival was significantly shorter (P = 0.008) than in the control group with squamous cell carcinoma on the lower limb. Thus, this complication is lethal in many cases, and there should be no hesitation to biopsy unusual or prolonged cases of venous ulcers. When diagnosed, it merits a thorough investigation of the spread and degree of differentiation. Treatment is by wide excision or amputation; radiation treatment is not feasible.

Material collected included paraffin imbedded blocks of SCC in venous ulcers from 23 patients and blocks from the non-cancerous ulcers of 35 patients. These samples were studied for the presence of human papillomavirus (HPV) by PCR with GP05+/06+ (mucosal) and with nested PCR with CP65/70 and CP66169 (EV-associated) primers. A subsequent non-radioactive Southern blot hybridisation was used to confirm the specificity of the PCR procedure. Ten samples were HPV-positive, all of which were from the non-cancerous ulcers and all mucosal type. Thus, HPV infection is probably not the carcinogen responsible for the malignant transformation of venous leg ulcers. The difference in positivity between the ulcers and the SCCs was statistically significant (p = 0.01), raising the question of a "hit and run" infection of the cancers. The samples specified above were analysed with immunohistochemistry for expression of p21WAF1/CIP1 (p21), p53, bc1-2 and Ki-67. There was no expression of p21, p53 or bc1-2 in the venous ulcer samples and no expression of bc1-2 in any of the samples. 22 SCC samples were positive for p21 and 14 for p53. The absence of p53 expression in the ulcers and adjacent to the SCCs probably reflects the non-UV carcinogenic mechanism of these cancers.