Genital lichen sclerosus/balanitis xerotica obliterans in men with penile carcinoma: a critical analysis

Genital lichen sclerosus/balanitis xerotica obliterans in men with penile carcinoma: a critical analysis Prodromos Philippou, Majid Shabbir, David J. Ralph, Peter Malone, Raj Nigam, Alex Freeman*, Asif Muneer and Suks Minhas Departments of Urology and *Pathology, University College London Hospital, London, UK What s known on the subject? and What does the study add? The European Association of Urology guidelines identify lichen sclerosus (LS) as a strong risk factor for penile squamous cell carcinoma (pscc). However, this statement is based on the findings of case control studies (Level of Evidence 2a) and a direct causal relationship between LS/balanitis xerotica obliterans (BXO) and pscc remains to be established. Firm guidelines with respect to the appropriate follow-up policy for LS/BXO are lacking, whereas the impact of synchronous LS/BXO on the prognosis of pscc remains to be determined. The presence of histologically-confirmed synchronous LS/BXO in patients diagnosed with pscc is relatively high, although it is not associated with an increased risk of adverse histopathological features. LS/BXO can develop in extragenital skin grafts used for reconstruction after organ-sparing surgery for pscc. Objectives To determine the rate of lichen sclerosus/balanitis xerotica obliterans (LS/BXO) in patients with penile squamous cell carcinoma (pscc) and establish whether the presence of LS/BXO is associated with adverse histopathological features of pscc. To report the phenomenon of LS involving non-genital skin grafts in patients who underwent organ-sparing surgery and split-skin graft (SSG) reconstruction Patients and Methods Between January 2002 and January 2010, 223 men underwent surgical treatment for pscc. A group of 52 patients with histologically-confirmed synchronous LS was identified (group A; overall rate of LS/BXO = 23.3%) and compared with a group of patients without synchronous LS (group B; n = 171; 76.7%). A subgroup of patients who underwent surgical excision and SSG reconstruction was also identified The histology reports of graft biopsies obtained during follow-up were reviewed and the rate of LS involving the graft was also recorded. Results Mean (range) age at diagnosis was 60.9 (34 81) years and 60.7 (28 89) years for groups A and B, respectively (P = 0.958). The mean (range) duration of follow-up was 38.3 (4 92) months for group A and 45.5 (4 107) months for group B(P = 0.162) No statistically significant differences were noted between groups A and B in terms of tumour grade (P = 0.091), stage (P = 0.697), presence of lymphovascular invasion (P = 0.333), histological subtype (P = 0.107), associated carcinoma in situ (P = 0.246) or nodal status at initial diagnosis (P = 0.555). In the subgroup of 188 patients who underwent SSG reconstruction, 41 (21.8%) patients had histologically-confirmed synchronous LS; in this subgroup, 26 (13.8%) patients underwent graft biopsy during follow-up. Genital LS involving the graft was identified in seven specimens, although none of these seven cases had associated recurrent pscc. Conclusions The presence of histologically-confirmed synchronous LS in patients with pscc is relatively high but is not associated with increased rates of adverse histopathological features, including carcinoma in situ. LS can develop in extragenital skin grafts, although its association with the long-term risk for recurrent pscc is not apparent in the present study. 970 2013 BJU International 111, 970 976 doi:10.1111/j.1464-410x.2012.11773.x

Genital lichen sclerosus/balanitis xerotica obliterans Keywords balanitis sclerotica obliterans, lichen sclerosus, penile, reconstruction, squamous cell carcinoma Fig. 1 Histological appearance of lichen sclerosus/balanitis xerotica involving the glans and coronal sulcus. Band-like infiltrate of lymphocytes and plasma cells in the dermis, hyalinization of collagen in the upper dermis and orthokeratotic hyperkeratosis of the epithelium (haematoxylin and eosin stain; magnification 100). Introduction Lichen sclerosus/balanitis xerotica obliterans (LS/BXO) is a chronic inflammatory process of unknown aetiology affecting the glans penis, prepuce and urethra [1]. It may lead to a significant impairment of sexual and urinary function, although it has also been associated with a risk of malignant transformation [2]. The European Association of Urology (EAU) guidelines identify LS as a strong risk factor for penile squamous cell carcinoma (pscc) [3]. However, this statement is based on the findings of case control studies (Level of Evidence [LoE] 2a) and a direct causal relationship between LS/BXO and pscc remains to be established. The rate of malignancy among patients with LS/BXO is estimated to be 2.3 8.4%, whereas histologically-confirmed LS/BXO is identified in 28 44% of patients with pscc [2]. Firm guidelines with respect to the appropriate follow-up policy for LS/BXO are lacking, whereas the impact of synchronous LS/BXO on the prognosis of pscc remains to be determined. The present study aimed to assess the rate of LS/BXO in patients treated surgically for pscc and to establish whether LS/BXO is associated with adverse histopathological features for pscc. We also report the phenomenon of LS/BXO involving non-genital skin grafts in patients with pscc who underwent organ-sparing surgery and split-skin graft (SSG) reconstruction. Patients and Methods Between January 2002 and February 2010, 223 men underwent surgical treatment for pscc in a tertiary referral centre. All patients had biopsy-proven squamous cell carcinoma of the penis and were selected for surgery on the basis of clinical and radiological staging and after appropriate consultation. Preoperative staging included physical examination, as well as imaging with a penile MRI and a CT scan of the chest, abdomen and pelvis. Surgical techniques included circumcision, wide local excision, total or partial glans resurfacing, glansectomy (+/ distal corporectomy) + SSG reconstruction and subtotal/total penile amputation. The technique of glans resurfacing or glansectomy and SSG reconstruction has been described previously [4,5]. After excisional surgery of the penile lesion, patients with histologically proven tumour-positive inguinal nodes (by fine-needle aspiration, tissue core or open biopsy) underwent inguinal lymph node dissection. Patients with impalpable inguinal lymph nodes were stratified into low, intermediate and high risk in accordance with EAU guidelines [3] and those in the intermediate- and high-risk groups were initially offered superficial modified lymph node dissection and, from 2009 onward, dynamic sentinel node biopsy. With regard to initial nodal status, patients were subdivided into three groups: those who underwent surgical staging and had positive (pn+) or negative (pn0) nodes and those managed by surveillance (cn0). A single dedicated uropathologist reviewed the excised specimens. Surgical margins were inked to aid microscopic analysis and histopathological features (including the presence of lymphovascular invasion, tumour subtype, grade and stage) were documented. The presence or absence of synchronous LS/BXO in the surgical specimen was recorded. The diagnosis of LS/BXO was made on the presence of current established criteria (Figs 1,2 and Table 1) [6]. Patients were reviewed in a dedicated penile cancer clinic in accordance with EAU guidelines [3]. If signs of local recurrence were identified during follow-up, a biopsy was performed. Demographic characteristics, histopathological parameters, treatment and follow-up data were recorded in a prospective electronic database. The database was analyzed to identify two groups of patients, based on the presence or absence of histologically confirmed synchronous LS/BXO. Patients with histological types other than squamous cell carcinoma, recurrence after radiotherapy or primary 2013 BJU International 971

Philippou et al. Fig. 2 Histological appearance of lichen sclerosus/balanitis xerotica involving the glans and coronal sulcus. Band-like infiltrate in the dermis (haematoxylin and eosin stain; magnification 100). diagnosis was 60.9 (34 81) years for group A and 60.7 (28 89) years for group B (P = 0.958). The mean (range) duration of follow-up was 38.3 (4 92) months for group A and 45.5 (4 107) months for group B (P = 0.162). There were four patients in the LS/BXO group who had undergone circumcision for histologically-confirmed LS/BXO in the past and developed pscc of the glans with synchronous LS/BXO on final histology. In all four cases, the time interval between circumcision and pscc diagnosis was more than 5 years (range 67 112 months). These were considered as metachronous cases of pscc developing on a background of chronic LS/BXO. Table 1 Histological criteria for the diagnosis of lichen sclerosus/balanitis xerotica. Criteria Thinning and orthokeratotic hyperkeratosis of epithelium Basal cell layer showing degeneration Band-like infiltrate of lymphocytes and plasma cells in the dermis Pigment incontinence Hyalinization of collagen in the upper dermis carcinoma of the urethra were excluded. A group of 52 patients with histologically confirmed synchronous LS/BXO was identified (group A, overall rate of LS/BXO = 23.3%). Group A was compared with a group of patients without histologically-confirmed synchronous LS/BXO (group B; n = 171; 76.7%) with regard to associated histopathological features (i.e. the primary outcome of the study). A subgroup of patients who underwent surgical excision and SSG reconstruction was identified. The histology reports of graft biopsies obtained during follow-up were reviewed and the rate of LS/BXO and tumour involving the graft was recorded (i.e. the secondary outcome of the study). Data were summarized as means for continuous variables and in frequency tables for categorical variables. Statistical analysis was performed using SPSS, version 13.00 (SPSS Inc., Chicago, IL, USA). A chi-squared test and Fisher s exact test were used for comparison of nominal data and the Mann Whitney U-test was used to compare numerical data. P < 0.05 was considered statistically significant. Results Patient baseline demographic, clinical and histopathological data are summarized in Table 2. Mean (range) age at In comparison to group B, patients in group A had a relatively higher percentage of grade 1 tumours and a relatively lower percentage of grade 3 tumours, although this difference was not statistically significant. In terms of nodal status at initial diagnosis, no statistically significant difference was noted between the two groups (P = 0.555). A subgroup of 188 patients who underwent SSG reconstruction was identified. A review of the histopathology reports of the primary surgical specimen showed that 41 (21.8%) patients had histologicallyconfirmed synchronous LS/BXO. In this subgroup, 26 (13.8%) patients underwent delayed biopsy for cosmetic, functional or diagnostic purposes during follow-up. The mean (range) time interval between initial surgery and penile biopsy was 35.3 (13 70) months. The indications for graft biopsy included lesions suspicious for recurrence (21 cases), re-grafting for penile tip necrosis and subsequent contracture (one case), re-grafting for partial loss of graft (three cases) and chronic lymphoedaema (one case). The histopathology reports of graft biopsies were reviewed to determine the rate of LS involving the graft and the presence of recurrent pscc (Table 3). Genital LS was identified in seven of the 26 graft biopsy specimens (26.9%), although none of these cases was associated with recurrent pscc. All seven patients had histologically confirmed LS/BXO in the primary surgical specimen. Discussion The male form of LS/BXP is a chronic, progressive, lymphocyte-mediated skin condition of unknown aetiology. LS/BXO affects primarily the glans penis and prepuce of uncircumcised men and presents as pale, atrophic plaques, which may coalesce and sclerose, causing phimosis and meatal stenosis [6]. The rate of LS/BXO in the general male population has yet to be determined, although it appears to be underestimated. According to historical data [7], the estimated rate of LS/BXO in patients referred to a community dermatology department was 972 2013 BJU International

Genital lichen sclerosus/balanitis xerotica obliterans Table 2 Baseline demographic, clinical and histopathological data. Variable Group A, LS/BXO (+) Group B, LS/BXO ( ) P Patients, n (%) 52 (23.3) 171 (76.7) Age (years), mean (SD) 60.9 (14.4) 60.7 (11.7) 0.958* Carcinoma in situ (N = 15), n (%) 4/15 (26.7) 11/15 (73.3) Invasive squamous cell carcinoma (N = 208), n (%) 48/208 (23.1) 160/208 (76.9) Tumour grade, n (%) 0.091 Grade 1 8 (16.7) 11 (6.9) Grade 2 19 (39.6) 61 (38.1) Grade 3 21 (43.7) 88 (55.0) Tumour stage, n (%) 0.697 T1 19 (39.6) 69 (33.2) T2 18 (37.5) 63 (39.3) T3 11 (22.9) 28 (17.5) Histological subtype, n (%) 0.107 Classic, not otherwise specified 21 (43.8) 108 (67.5) Basaloid 6 (12.5) 11(6.9) Papillary 7 (14.6) 14 (8.8) Verrucous 3 (6.3) 5 (3.1) Mixed 1 (2.1) 5 (3.1) Condylomatous 9 (18.7) 14 (8.8) Sarcomatoid 1 (2.1) 3 (1.9) Lymphovascular invasion, n (%) 0.333 Positive 31 (64.6) 115 (71.9) Negative 17 (35.4) 45 (28.1) Associated carcinoma in situ, n (%) 0.246 Present 19 (39.6) 49 (30.6) Absent 29 (60.4) 111 (69.4) Nodal status at initial diagnosis, n (%) 0.555 cn0 21 (43.8) 63 (39.4) pn0 17 (35.4) 51 (31.9) pn+ 10 (20.8) 46 (28.8) Duration of follow-up (months), mean (SD) 38.3 (28.9) 45.5 (28.4) 0.162* *Mann Whitney U-test. Chi-squared test. Fisher s exact test. Table 3 Histology results in patients managed by split-skin graft (SSG) reconstruction. Variable Value Patients who underwent surgical excision and SSG 188 reconstruction, n LS/BXO in pathological specimen of primary procedure, n (%) Yes 41 (21.8) No 147 (78.2) Patients who underwent graft biopsy during follow-up, n (%) 26 (13.8) Time interval between primary surgery and graft biopsy 35.3 (13 70) (months), mean (range) Histology results of graft biopsy, n (%) Presence of LS/BXO 7/26 (26.9) Presence of recurrent malignancy 15/26 (57.7) Absence of LS/BXO or recurrent malignancy 4/26 (15.4) LS/BXO, lichen sclerosus/balanitis sclerotica obliterans. between 1 : 300 and 1/1000. The rate of histologically-confirmed LS/BXO in patients referred for medical circumcision appears tobe higher both in children (14%) [8] and adults (32%) [9]. The lack of epidemiological evidence with regard to the incidence of LS/BXO in the general population partly explains the difficulty in establishing a clear link between LS/BXO and pscc. pscc is a rare malignancy in the Western world, with an incidence of 0.1 0.9 cases per 100 000 men in Europe [10]. pscc has been associated with a number of established risk factors and associated diseases or conditions, including phimosis with chronic inflammation, human papillomavirus (HPV) infection, poor hygiene and smoking [2]. LS/BXO has been suggested to be the most important non-hpv-related condition associated with pscc, although the evidence for this is not well established. According to the EAU guidelines [3], LS/BXO represents a strong risk factor for pscc, although this statement is based on the findings of case reports and case control studies [11 21]. Observational descriptive studies investigating the relationship between pscc and LS/BXO are based on two methodological approaches (Fig. 3) (LoE 3) [2]. According to these studies, the incidence of malignancy in patients presenting with LS/BXO is in the range 2.3 8.4% [7,15 17]. Nasca et al. [15] noted a 5.8% rate of metachronous pscc in uncircumcised patients with LS/BXO. The mean (range) lag time from the onset of LS/BXO to a diagnosis of malignancy was 17 (10 23) years. They also noted that 2013 BJU International 973

Philippou et al. Studies investigating the relationship between pscc and LS/BXO (Level of Evidence 3) Fig. 3 Methodology and findings of studies investigating the relationship between penile squamous cell carcinoma (pscc) and lichen sclerosus/balanitis xerotica (LS/BXO). In a population of patients diagnosed with LS/BXO Wallace (n = 44): 2 cases of pscc (4.5%) [7] Nasca et al (n = 86): 5 cases of pscc (5.8%) [15] Depasquale et al (n = 522): 12 cases of pscc (2.3%) [16] Barbagli et al (n = 130): 11 cases of pscc (8.4%) [17] In a population of patients diagnosed with pscc Powell et al (n = 20): Histologicallyconfirmed LS/BXO in 8 cases (40%) [18] Pietrzak et al (n = 155): Histologicallyconfirmed LS/BXO in 44 cases (28.4%) [19] Velazquez et al (n = 207): 68 cases of histologically-confirmed LS/BXO (32.9%) [20] Perceau et al (n = 18): 8 cases of histologically-confirmed LS/BXO (44.4%) [21] Present study: 52 cases of histologicallyconfirmed LS/BXO (23.3%) pscc: penile Squamous Cell Carcinoma LS/BXO: Lichen Sclerosus/Balanitis Sclerotica Obliterans epithelial dysplasia may represent an intermediate stage, leading to the development of malignancy in the background of LS/BXO. In the largest LS/BXO series investigated to date [16], the risk of pscc in uncircumcised and circumcised patients was lower (2.3%) compared to the findings of a multicentre study reporting synchronous malignant changes in 8.4% of patients diagnosed with LS/BXO [17]. An alternative approach involves reporting the presence of histologically confirmed LS/BXO in patients treated for pscc [18 21]. Small series, involving patients managed for pscc in dermatology departments, report relatively high rates of synchronous LS/BXO in primary surgical specimens: 40% according to Powell et al. [18] and 44.4% according to Perceau et al. [21]. Pietrzak et al. [19] reported the experience of a tertiary referral centre for pscc (n = 155) where the rate of histologically-confirmed LS/BXO in patients treated for pscc was 28.4%. Velazquez and Cubilla [20] reported a rate of 32.9% for synchronous LS/BXO in 207 excision specimens from patients treated for pscc. In the present study, the rate of synchronous histological LS/BXO was 23.3% and is within the range previously reported. The relatively low incidence of pscc in the UK has led to the recommendation for centralized treatment in a limited number of centres arranged as supra-regional networks [4]. The centralization of services contributes to minimizing sources of selection bias, such as the non-referral of low-risk pscc cases considered cured by local excision or circumcision alone in the context of district general hospitals. Studies derived from prospective data collection and analysis in the context of tertiary referral centres are a reliable source of evidence, although prospective multicentre studies represent a field for future research. The association between vulvar LS and squamous cell carcinoma is well documented and the risk of vulvar malignancy is 4 7% in women with LS [6]. The exact relationship between LS/BXO and penile cancer has yet to be determined (LoE 3) [2]. Based on the findings obtained from observational studies, a direct pathogenetic link cannot be established or excluded because a relatively rare malignancy such as pscc may co-exist with a relatively common condition such as LS/BXO. Previous studies provide indirect histological evidence in favour of a pathogenic continuum between LS/BXO and invasive pscc (with epithelial dysplasia as an intermediate stage) [15,17,22]. However, according to our findings, the rate of associated carcinoma in situ was higher in group A (39.6% vs 30.6% in group B), although the difference was not statistically significant and therefore does not support this hypothesis. In this series, a few pscc cases developed on a background of chronic LS/BXO. The clinical interpretation of this observation is challenging as a result of the long interval between treatment for LS/BXO and a diagnosis of pscc. High-risk HPV infection has been linked with the development of pscc [23], although the relationship between LS/BXO, pscc and HPV infection is a matter of debate [17,20,21]. In the series reported by Perceau et al. [21], none of the patients with LS/BXO-associated pscc had PCR positivity for high-risk HPV. This is in sharp 974 2013 BJU International

Genital lichen sclerosus/balanitis xerotica obliterans contrast to the findings reported by Nasca et al. [15], who noted 80% PCR-positivity for HPV16 in LS/BXO-related pscc. Prowse et al. [24] noted HPV PCR-positivity in 53.8% and 54.5% of pscc cases associated and not associated with LS/BXO, respectively. These findings augment the arising controversy; LS/BXO may represent the initial step in a non-hpv related oncogenic pathway for pscc [21,25] or it may act as a field susceptible to HPV-related dysplastic changes [17]. In the present study, PCR analysis for HPV presence was not determined. The significance of co-existent LS/BXO for the prognosis of pscc is also controversial. Oertell et al. [25] reported a frequent coexistence of LS/BXO and low-grade pscc in a population with a incidence of high penile cancer. In the series investigated in the present study, patients in group A had higher rates of grade 1 and stage T1 pscc, as well as lower rates of lymphovascular invasion. These differences, however, were not statistically significant. Nodal status at initial diagnosis was comparable between the two groups, indicating that the presence of LS/BXO does not indicate a more aggressive tumour biology. A similar absence of statistically significant differences with regard to adverse histopathological features was previously reported by Pietrzak et al. [19]. The phenomenon of LS/BXO involving extragenital skin grafts used for genital reconstruction has been reported previously [26,27]. Recently, Abdelbaky et al. [27] reported their experience in a series of 56 patients who were managed by organ-sparing surgery and SSG reconstruction. In 18 (32.1%) patients, a graft biopsy was performed during follow-up (mean of 42.7 months) for cosmetic or diagnostic reasons. LS/BXO was identified in six out of 18 specimens and, in one case, it was associated with recurrent verrucous carcinoma. In our cohort, seven cases of LS/BXO involving the graft were noted and none of these was associated with recurrent pscc. At present, an assessment of the oncological implications of this observation is challenging. If LS/BXO precedes malignant changes by many years (e.g. a lag time of 10 23 years according to Nasca et al. [15]), a longer follow-up is necessary to clarify the significance of LS involvement of the graft with respect to the recurrence of carcinoma. By contrast, LS/BXO is clinically underdiagnosed [9] and, to achieve an accurate estimation of the patient population with LS involving the graft, histological confirmation would be required in all patients, although routine biopsy is impractical. LS/BXO involvement of extragenital skin grafts may have some importance in understanding the pathogenesis of LS/BXO itself. The aetiology of LS/BXO has yet to be determined and multiple theories have been proposed [1,6]. Previous studies have identified an association between LS/BXO and systematic factors (autoimmune, genetic and/or hormonal) [28,29], as well as local causes (trauma, infection or radiation) [30 32]. LS/BXO involvement of extragenital skin grafts used for genital reconstruction supports the theory that, when systematically predisposed skin is exposed to local factors of the genital region, the development of LS/BXO is induced [6]. The existing evidence does not allow the formulation of clear guidelines with regard to the appropriate follow-up of patients with LS/BXO [2]. Some studies recommend the routine clinical follow-up of patients diagnosed with LS/BXO [16,18,33], although this approach is not supported by the available LoE. Instructing patients to perform regular self-examination may be a practical universal measure. Regular follow-up of circumcised patients with persistent or chronically active LS/BXO and a low threshold for biopsy of suspicious lesions is advisable [19]. The relationship between LS/BXO and pscc, despite being unclear, may justify histopathological analysis of circumcision specimens when the procedure is performed for medical indications. In conclusion, the role of LS/BXO as a premalignant lesion remains unclear. Although the presence of histologically-confirmed synchronous LS/BXO in patients diagnosed with pscc is relatively high, this is not associated with an increased risk of adverse histopathological features. LS/BXO can develop in extragenital skin grafts and its association with the long-term risk for recurrent pscc needs to be determined. Conflict of Interest None declared. References 1 Pugliese JM, Morey AF, Peterson AC. Lichen sclerosus: review of the literature and current recommendations for management. J Urol 2007; 178: 2268 76 2 Minhas S, Manseck A, Watya S et al. Penile cancer prevention and premalignant conditions. Urology 2010; 76 (Suppl. 1): S24 35 3 Algaba F, Horenblas S, Pizzocaro-Luigi Piva G et al. EAU guidelines on penile cancer. Eur Urol 2002; 42: 199 203 4 Hegarty PK, Shabbir M, Hughes B et al. Penile preserving surgery and surgical strategies to maximize penile form and function in penile cancer: recommendations from the United Kingdom experience. World J Urol 2009; 27: 179 87 5 Shabbir M, Muneer A, Kalsi J et al. Glans resurfacing for the treatment of carcinoma in situ of the penis: surgical technique and outcomes. Eur Urol 2011; 59: 142 7 2013 BJU International 975

Philippou et al. 6 Powell JJ, Wojnarowska F. Lichen sclerosus. Lancet 1999; 353: 1777 83 7 Wallace HJ. Lichen sclerosus et atrophicus. Trans St Johns Hosp Dermatol Soc 1971; 57: 9 30 8 Chalmers RJ, Burton PA, Bennett RF, Goring CC, Smith JB. Lichen sclerosus et atrophicus. A common and distinctive cause of phimosis in boys. Arch Dermatol 1984; 120: 1025 7 9 Aynaud O, Piron D, Casanova JM. Incidence of preputial lichen sclerosus in adults: histologic study of circumcision specimens. J Am Acad Dermatol 1999; 41: 923 6 10 Pow-Sang MR, Ferreira U, Pow-Sang JM et al. Epidemiology and natural history of penile cancer. Urology 2010; 76 (Suppl. 1): S2 6 11 Dillner J, von Krogh G, Horenblas S, Meijer CJ. Etiology of squamous cell carcinoma of the penis. Scand J Urol Nephrol Suppl 2000; 205: 189 93 12 Bingham JS. Carcinoma of the penis developed in lichen sclerosus et atrophicus. Br J Vener Dis 1978; 54: 350 1 13 Weigand DA. Lichen sclerosus et atrophicus, multiple dysplastic keratoses, and squamous-cell carcinoma of the glans penis. J Dermatol Surg Oncol 1980; 6: 45 50 14 Doré B, Irani J, Aubert J. Carcinoma of the penis in lichen sclerosus atrophicus. A case report. Eur Urol 1990; 18: 153 5 15 Nasca MR, Innocenzi D, Micali G. Penile cancer among patients with genital lichen sclerosus. JAmAcad Dermatol 1999; 41: 911 4 16 Depasquale I, Park AJ, Bracka A. The treatment of balanitis xerotica obliterans. BJU Int 2000; 86: 459 65 17 Barbagli G, Palminteri E, Mirri F et al. Penile carcinoma in patients with genital lichen sclerosus: a multicenter survey. J Urol 2006; 175: 1359 63 18 Powell J, Robson A, Cranston D, Wojnarowska F, Turner R. High incidence of lichen sclerosus in patients with squamous cell carcinoma of the penis. Br J Dermatol 2001; 145: 85 9 19 Pietrzak P, Hadway P, Corbishley CM, Watkin NA. Is the association between balanitis xerotica obliterans and penile carcinoma underestimated? BJU Int 2006; 98: 74 6 20 Velazquez EF, Cubilla AL. Lichen sclerosus in 68 patients with squamous cell carcinoma of the penis: frequent atypias and correlation with special carcinoma variants suggests a precancerous role. Am J Surg Pathol 2003; 27: 1448 53 21 Perceau G, Derancourt C, Clavel C et al. Lichen sclerosus is frequently present in penile squamous cell carcinomas but is not always associated with oncogenic human papillomavirus. Br J Dermatol 2003; 148: 934 8 22 Cubilla AL, Velazquez EF, Young RH. Epithelial lesions associated with invasive penile squamous cell carcinoma: a pathologic study of 288 cases. IntJSurg Pathol 2004; 12: 351 64 23 Daling JR, Madeleine MM, Johnson LG et al. Penile cancer: importance of circumcision, human papillomavirus and smoking in in situ and invasive disease. IntJCancer2005; 116: 606 16 24 Prowse DM, Ktori EN, Chandrasekaran D, Prapa A, Baithun S. Human papillomavirus-associated increase in p16ink4a expression in penile lichen sclerosus and squamous cell carcinoma. Br J Dermatol 2008; 158: 261 5 25 Oertell J, Caballero C, Iglesias M et al. Differentiated precursor lesions and low-grade variants of squamous cell carcinomas are frequent findings in foreskins of patients from a region of high penile cancer incidence. Histopathology 2011; 58: 925 33 26 Kumar MV, Harris DL. Balanitis xerotica obliterans complicating hypospadias repair. Br J Plast Surg 1999; 52: 69 71 27 Abdelbaky AM, Aluru P, Keegan P, Greene DR. Development of male genital lichen sclerosus in penile reconstruction skin grafts after cancer surgery: an unreported complication. BJU Int 2012; 109: 776 9 28 Purcell KG, Spencer LV, Simpson PM et al. HLA antigens in lichen sclerosus et trophicus. Arch Dermatol 1990; 126: 1043 5 29 Jebakumar SP, Wooley PD. Balanitis xerotica obliterans. Int J STD AIDS 1995; 6: 81 3 30 Pock L. Koebner phenomenon in lichen sclerosus et atrophicus. Dermatologica 1990; 181: 76 7 31 Milligan A, Graham-Brown RA, Burns DA. Lichen sclerosus et atrophicus following sunburn. Clin Exp Dermatol 1988; 13: 36 7 32 Yates VM, King CM, Dave VK. Lichen sclerosus et atrophicus following radiation therapy. Arch Dermatol 1985; 121: 1044 7 33 Liatsikos EN, Perimenis P, Dandinis K, Kaladelfou E, Barbalias G. Lichen sclerosus et atrophicus. Findings after complete circumcision. Scand J Urol Nephrol 1997; 31: 453 6 Correspondence: Prodromos Philippou, University College Hospital, 250 Euston Road, London NW1 2PQ, UK. e-mail: brucemed@hotmail.com Abbreviations: EAU, European Association of Urology; HPV, human papillomavirus; LS/BXO, lichen sclerosus/balanitis xerotica obliterans; LoE, Level of Evidence; pscc, penile squamous cell carcinoma; SSG, split-skin graft. 976 2013 BJU International