A 10-year retrospective audit of penile cancer management in the UK

Authors


Miss Tina Mistry, Department of Urology, University Hospital Coventry & Warwickshire, Clifford Bridge Road, Coventry CV2 2DX, UK.
e-mail: mistrytina@hotmail.com

Abstract

OBJECTIVE

To audit the penile cancer workload, management and outcome within a regional cancer network serving a population of ≈ 1 million in the West Midlands (UK), comparing these data to that published by the British Association of Urological Surgeons National Cancer Registry, the UK National Institute of Clinical Excellence and the European Associations of Urology guidelines.

PATIENTS AND METHODS

Patients diagnosed with or treated for penile cancer within the Arden Cancer Network over a 10-year period were identified retrospectively, and data relating to histology, local treatment, lymph node management, outcome and survival were recorded.

RESULTS

Data were obtained for 65 patients; 61 (94%) had histologically confirmed squamous cell carcinoma (SCC) of the penis, equating to ≈ 0.6 cases per 100 000 population per year. Their mean age at diagnosis was 63 years. Of SCCs, 86% were located on the glans and/or foreskin. Thirty-six patients had conservative primary local therapy, mostly for T0 or T1 disease. The 5-year relapse-free survival after radiotherapy was 63%, although survival after salvage penectomy was 75% at 4 years. Forty-seven patients had lymph node surveillance; 11 developed lymph node disease and had lymph node dissection (LND) with or with no radiotherapy, but survival was poor. Primary inguinal LND with or without radiotherapy was used in eight patients, and was associated with a good survival, although three were found to have negative histology after LND. Survival was strongly influenced by T and N stage at presentation and the 5-year survival for the whole group was 71%.

CONCLUSION

The workload, incidence and overall mortality from penile cancer within the Arden Cancer Network are in line with those in the rest of the UK. Rates of conservative therapy were good in this group and associated with good survival. Survival could be improved by identifying and aggressively treating those patients at high risk of lymph node disease.

Abbreviations
SCC

squamous cell carcinoma

EAU

European Association of Urology

NICE

National Institute for Clinical Excellence

RT

radiotherapy

LND

lymph node dissection.

INTRODUCTION

In Western populations carcinoma of the penis is an uncommon disease, accounting for <1% of all male cancers in the UK. Of penile cancers, > 95% are squamous cell carcinomas (SCC), and in the UK ≈ 360 new cases are recorded each year [1], giving an incidence of 0.6 per 100 000 men; hence each UK urologist expects to diagnose fewer than one new patient with the disease each year. Due to the lack of large randomized trials and the rarity of the disease, there remains considerable debate about the management of penile cancer, in particular the treatment of early local disease and lymph node metastases [2–4]. The European Association of Urology (EAU) recently updated their guidelines on the management of penile cancer [5], and in 2002 the UK National Institute for Clinical Excellence (NICE) Guidance on UK Cancer Services recommended that specialized penile cancer multidisciplinary teams should be established jointly between neighbouring networks, each serving a population base of ≥ 4 million and expecting to manage ≥ 25 new patients each year [6].

The aim of the present study was to audit the penile cancer workload, management and outcome within the Arden Cancer Network, which serves a population of ≈ 1 million in the West Midlands (UK); we compared these data to published BAUS National Cancer Registry data and to the EAU guidelines [1,5].

PATIENTS AND METHODS

The West Midlands Cancer Intelligence Unit was used to retrospectively identify patients diagnosed with penile cancer at hospitals within the Arden Cancer Network between 1993 and 2003. Case notes were reviewed and data relating to histology, local and lymph node management, outcome and survival were recorded. Additional information was obtained from pathology and oncology databases.

In all, 65 patients were diagnosed with penile cancer between 5 July 1993 and 15 May 2003, and notes were available for all; 61 patients (94%) from a population of ≈ 1 million were diagnosed with primary penile SCC within a 10-year period, an incidence of ≈ 0.6 per 100 000 per year within the Arden Network. Four patients were found to have metastatic penile adenocarcinoma from bladder (two), colon (one) and an unknown primary (one).

The mean (range) age at diagnosis for patients with primary SCC was 63 (27–94) years, and the mean tumour size at diagnosis was 2.94 (0.5–6) cm. Most primary SCCs were on the glans (56%), foreskin (20%) or both (10%). Where specified, other sites of primary disease were the corona (7%), shaft (5%) and distal urethra (3%). The growth pattern was verrucous in 11%, superficial in 8%, nodular in 1%, basaloid in 1% and unknown in 79%. The presence or absence of lymphatic or vascular invasion was poorly documented in both the notes and histology reports, and was not noted in 79% and 62% of patients, respectively.

The distribution of the pathological grade and TNM stage of the tumours according to the TNM system (2002) [7] is shown in Table 1. Most tumours (43%) were well-differentiated, grade 1 tumours. Carcinoma in situ (Tis) and noninvasive verrucous carcinoma (Ta) accounted for 16% of tumours.

Table 1.  The distribution of pathological grade and stage (TNM system) of disease in 61 patients with primary penile SCC within the Arden Cancer Network
Grade/stagen (%)
Grade at presentation:
 126 (43)
 215 (24)
 311 (18)
 Unknown 9 (15)
Disease stage (TNM):
 T0 0
 Tis 7 (11)
 Ta 3 (5)
 T129 (48)
 T213 (21)
 T3 1 (2)
 T4 0
 TX 7 (11)
 Unknown 1 (2)
 N0 (or X)47 (77)
 N1 3 (5)
 N2 5 (8)
 Unknown 6 (10)
 M0 (or X)60 (98)
 M1 1 (2)

An oncologist was involved in the care of 48 (74%) patients and a urologist in 62 (95%) patients; 85% of patients were seen and diagnosed by a urological surgeon, 8% by a general surgeon and 3% by a plastic surgeon.

RESULTS

Table 2 shows the primary local treatment undertaken, according to stage of the primary tumour; 36 patients (59%) had conservative primary local therapy for their SCC, which included local excision, radiotherapy (RT), topical chemotherapy and glansectomy.

Table 2. 
Primary local therapy in patients with SCC according to the disease stage of the primary tumour. Values are n or n (%) for larger totals
TherapyTX/0Tis/TaT1T2T3T not known
  1. LE, local excision; TC, topical chemotherapy.

Conservative4 (57) 9 (90)18 (62) 4 (31)1
 LE 3 5 – 
 LE + RT 2 5 – 
 RT + TC 1 – – 
 RT only4 3 6 31 
 glansectomy – 2 1 
Penectomy:
 partial3 (43) 1 (10) 9 (31) 4 (31)1
 total – 2 (7) 4 (31)
None/refused – – 1
Not known – – –
Total710291311

RT was given to 18 patients as primary radical treatment, including locally advanced and unstaged cases. The technique used was standard for disease localized to the penis: immobilization of the penis using a wax block, with a field arrangement of two opposing lateral photon fields. Doses used varied from 55 Gy in 16 fractions to 50 Gy in 20 fractions (dose per fraction 3.4–2.5 Gy). The equivalent dose in 2 Gy fractions was 52–62 Gy, but there was no correlation with recurrence. One patient did not respond to RT and required salvage surgery within 2 months. Complications associated with local RT were necrosis (two patients) and one each with phimosis, erectile dysfunction and urethral stricture.

Eight patients relapsed after RT and all went on to have either total or partial salvage penectomy. Six of the patients were alive and disease-free at a median of 4 years follow-up; one patient died from penile cancer but also had lymph node recurrence; no survival data were available for one patient. In all, 18 patients had primary local excision including three glansectomies; seven also had postoperative RT.

Five patients had conservative therapy for tumours of stage ≥ T2; these patients also had G2 or G3 tumours. One of these patients had declined surgery and opted to undergo RT, and one had N2 disease at presentation and had palliative RT; they both subsequently died from penile cancer. One patient had a primary glansectomy and reconstruction. The remaining two patients had RT only, and although they both developed local recurrence of their disease, they proceeded to have penile amputations, and are both still alive and disease-free at mean of 6 years of follow-up.

In all, 24 patients (39%) had a penectomy (18 partial and six total), done by 11 different surgeons. One patient having a partial amputation developed local recurrence and progressed to a total penectomy, but eventually died from the disease. Two of the six patients having a primary total penectomy developed local recurrence; they both had T2G3 disease at presentation, and subsequently had salvage RT. One died from penile cancer and one is alive and disease-free at 33 months; the latter had salvage RT to the penis alone, using the technique previously described, with a dose of 52.5 Gy in 20 fractions. There were no recurrences among patients undergoing partial penectomy. Complications associated with primary penectomy were wound infection (one patient) and meatal stenosis (one).

The Kaplan-Meier curve for recurrence-free survival by treatment is shown in Fig. 1a; the 5-year recurrence-free survival for all cases after RT was 63%.

Figure 1.

a, Cumulative penile cancer survival for all patients according to primary treatment; b, T stage (T1 vs > T1); c, nodal disease; and d, overall survival for patients with early, T1N0 penile cancer according to primary treatment.

Table 3 shows the types of primary lymph node therapy for patients within the Arden Network, and subsequent outcome; 47 patients (77%) had lymph node surveillance, consisting of regular clinical examination in the outpatient department, by either an oncologist or urologist. Forty-three patients undergoing surveillance had no clinical lymphadenopathy, three had negative histology after excision biopsy for palpable lymph nodes, and one had a palpable inguinal lymph node but no evidence of lymphadenopathy on CT. Two unilateral and one bilateral biopsies were taken. Eleven (23%) of the patients undergoing surveillance developed lymph node disease and subsequently had salvage RT with or with no lymph node dissection (LND). Salvage RT to the groin was given with a direct anterior field to a dose of 40 Gy in 15 fractions. In one patient the pelvis was treated to 45 Gy in 25 fractions. All but three of these patients developing lymph node disease had moderate- to high-grade primary penile disease with a tumour stage of ≥ T1 at diagnosis. Eight of the 11 patients developing lymph node recurrence while under surveillance died from penile cancer, with a mean time to death of 27.8 months from diagnosis.

Table 3.  Primary lymph node management and the related outcome and recurrence rates
 SurveillanceLND onlyLND + RTRT onlyNone/unknown (U)
n (%)47 (77)7 (11)1 (2)4 (7)2 (3)
Primary SCC
 G1193021
 G211400
 G3 8011
 Unknown 711
LN recurrence, n(%)11/47 (23)3/70/12/4U
Alive
 no recurrence36511U
 recurrence 3000 
Dead
 total 8203U
 penile cancer 82 3 
Mean disease-free survival, years 9.96.43.53.8U

In all, eight patients had primary inguinal LND; there were no pelvic LNDs. One patient also received adjuvant RT (groin field only, with a dose of 40 Gy in 15 fractions) for high-grade and histologically confirmed nodal disease. There were five unilateral and three bilateral LNDs; four were synchronous and four were delayed. Half of the LNDs were done using a modified modern approach and half used older techniques. Six patients with clinically palpable lymph nodes at presentation had positive histology after LND. Two patients undergoing prophylactic primary LND for low- to moderate-grade T1 disease were found to have negative histology. One patient with clinically palpable nodes also had negative histology after LND. Two of the patients who had primary LND died from penile cancer at a mean of 21.5 months from diagnosis. There were more complications associated with LND in those patients who had older methods of LND, i.e. one each with lymphoedema/lymphocele, seroma and infected lymphocele; two patients developed lymphoedema after a modified LND approach.

Four patients had RT alone for palliation of clinically palpable nodal disease; one of these also had metastatic disease at presentation. The doses in these patients varied, i.e. 40 Gy in 15 fractions, 35 Gy in 10 fractions, 20 Gy in five fractions and 12 Gy in two fractions. Three of these patients died from penile cancer at a mean of 9 months from diagnosis.

Systemic chemotherapy was used in four patients; one had nodal disease at presentation, two had developed lymph node recurrence after either primary or salvage LND, and one had lymph node recurrence after RT for nodal disease. The chemotherapy regimen used was fluorouracil 1 mg/m2 and mitomycin 12 mg/m2 three-weekly for two to four cycles. One patient did not respond to this regimen and was switched to gemcitabine 1000 mg/m2 on days 1, 8 and 15, and carboplatin (area under curve) three-weekly. Three patients died from penile cancer at a mean of 15 months after diagnosis; one was alive at 78 months of follow-up.

The survival of these 61 patients was significantly influenced by stage and lymph node status at presentation (Fig. 1b,c). The 5-year survival for patients with T1 disease at presentation was almost 80%, whereas that for > T1 was about half. For nodal disease, N1/2 disease was associated with a 25% 3-year survival, while for N0 disease it was 85%. The overall survival for early T1N0 patients according to treatment is shown in Fig. 1d; from the Kaplan-Meier curve, the 5-year overall survival for the whole group was 71%.

DISCUSSION

In the present study the incidence and disease-specific mortality from penile cancer was in accordance with the national incidence of the disease as reported in the BAUS Cancer Registry data [1]. This was also true of the distribution of cases among urologists, with each consultant seeing about 0.5 cases per year. However, 11% of patients were seen and/or diagnosed by other than urologists, and this generalized management of the disease is an area that the NICE guidelines aim to rectify. As most of the present patients were seen before the introduction of the NICE guidelines in 2002, this reinforces the value of the implementation of these guidelines to allow more appropriate and specialized management of this rare disease.

Most penile cancers detected in the present group were small, distal tumours. There was a high rate of conservative primary treatment, with almost 60% of patients having primary penis-conserving treatment, which in most consisted of RT either with or without local excision. The current trend for conservative surgery in patients with low-grade Ta/T1 tumours was not reflected in this audit [5], although a prospective audit would determine if practice has changed towards conservative therapy in such patients in recent years. Recent studies showed a low rate of local recurrence after penis-sparing, conservative surgery for selected cases, with Bissada et al.[8] reporting long-term follow-up data. Local brachytherapy and laser ablation, although not used as treatments in this group of patients, are also associated with low local recurrence rates [9–11]. Micrographic surgery has also been used to treat penile carcinomas, with good recurrence rates and cosmetic results [12].

Although the recurrence rate after RT was 37% at 5 years, in accordance with previous studies [3,13], all patients went on to have salvage penectomy, with good survival. The organ-preservation rate with RT or laser therapy is 55–84%, thus justifying the use of a penis-preserving strategy as first-line therapy for penile cancer [5]. The EAU guidelines also recommend further conservative treatment for local recurrences that are small and with no corpora cavernosal invasion [11,14], with larger and/or infiltrating recurrences benefiting from partial or total penectomy. Azrif et al.[15] report that RT is suitable for patients with T1-2 penile cancers, provided it is associated with scrupulous surveillance to detect local recurrences, and early salvage surgery, without jeopardizing overall survival. Local recurrence after conservative treatment for penile carcinoma was previously shown not to compromise survival [16], and this is supported by the good outcomes after salvage penectomy in the present series.

Lymph node status is the primary determinant of survival in penile carcinoma [2,17,18]. Several risk factors for lymph node metastases have been identified, particularly high grade and T2 disease [19,20]. Primary inguinal LND is an effective therapy in patients with palpable and/or histologically positive lymph nodes [18,21,22]. A 6-week course of broad-spectrum antibiotics after excision of the primary tumour can be a useful method to discern inflammatory lymphadenopathy, as nodes become clinically insignificant in half of patients who receive such therapy; these patients can then be followed with regular surveillance [23]. Cure rates of up to 75% with LND have been reported in patients presenting with one or two involved nodes, and even in patients with pelvic node involvement cure rates of almost 20% have been achieved [22]. However, LND is associated with significant morbidity which precludes its use as a prophylactic treatment in patients at low risk of lymph node metastases, although the modern modified approaches for LND might reduce complication rates [20,24,25]. In the present study, complication rates were lower in those patients who had a modified LND, although few patients had LND overall. All the present patients with evidence of nodal disease at presentation had either primary inguinal LND, RT or both. The survival benefit of early LND in men with positive nodal disease was also shown in a recent prospective study by Hegarty et al.[23].

Most patients who were clinically or histologically node-negative at presentation had regular lymph node surveillance in the present study. However, the development of nodal disease while under surveillance was generally predictable according to the histological grade of the primary penile tumour. More than half the patients who developed nodal disease while under surveillance had T2G3 disease at presentation, a group of patients recognized to be at high risk of nodal involvement [19,20]; in retrospect, these patients might have benefited from early LND. The incidence of occult nodal metastases within this high-risk group of patients is 68–73%[24,25]. Three of the patients undergoing surveillance who developed nodal disease had T2 and/or G2 disease. These patients are at intermediate risk of developing nodal metastases, and vascular and lymphatic invasion of the primary tumour is an important prognostic indicator of the likelihood of nodal disease [26,27]. The presence or absence of vascular or lymphatic invasion was not consistently reported in this series, which might hinder the ability of risk stratification to identify patients at intermediate risk of developing node metastases.

In conclusion, the workload and incidence of penile cancer within the Arden Cancer Network is in line with the rest of the UK. Most cases were small, distal tumours which were effectively treated conservatively. Although primary local RT was associated with a significant risk of local relapse, survival was good after salvage surgery. Primary LND was associated with a high cure rate, although just over a third of patients were subsequently found to have benign histology. Moreover, patients with initially poor prognostic factors did not receive primary LND and subsequently developed lymph node disease which was associated with a poor survival rate; prophylactic LND could perhaps have been justified in these patients. The establishment of super-specialist multidisciplinary teams to treat penile cancer, as recommended by NICE, should thus improve the rates of conservative management for early penile cancer, while also ensuring that those patients with high risk of lymph node disease and locally advanced tumours are managed appropriately.

ACKNOWLEDGEMENTS

We thank the following people who assisted with the collection of data: Val Lawrence (Cancer Services, George Eliot Hospital, Nuneaton), Indira Lal (Arden Cancer Centre, University Hospital, Coventry), Nichola Salmons (Audit Department, Warwick Hospital), Stacey Croft (West Midlands Cancer Intelligence Unit), Prof N. James (Queen Elizabeth Hospital, Birmingham), and all Consultants within the Arden Cancer Network. We also thank Mr Kieran Jefferson for his valuable comments on this manuscript.

CONFLICT OF INTEREST

None declared.

Ancillary