Surgical management for localised penile cancer

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To assess the effects of various surgical treatments for localised penile cancer, including oncological efficacy, quality of life and functional impact. (FS: do we need to be more specific than this?)

The comparisons of interest are: 

• one type of surgery versus another; 

• surgical treatment versus non‐surgical treatment.

Background

Penile cancer is a rare malignancy worldwide, representing only 1% of all cancers affecting men (Barnholtz‐Sloan 2007). Its prevalence varies between geographical areas, from 0.5 to 1.5 per 100,000 men across European countries (Curado 2007). Outside Europe the prevalence can be higher, ranging from eight per 100,000 men in South America (Curado 2007), up to a lifetime risk of 1000 per 100,000 men in Africa (Barnholtz‐Sloan 2007). It also varies between races in the same geographical area, such as the USA (Barnholtz‐Sloan 2007). In Western cultures grouped together, penile cancer appeared to be declining in frequency from 0.84 to 0.58 per 100,000 men from the 1970s to 1990s (Barnholtz‐Sloan 2007). However, in England and Scotland the prevalence increased significantly between 1993 and 1995 and from 2008 to 2010 (P < 0.01) but not in Wales or Northern Ireland; the rate across the four countries in the United Kingdom ranged from 1.2 to 1.6 per 100,000 between 1993 and 1995, whereas from 2008 to 2010 it ranged from 1.3 to 2.0 per 100,000 (South West Public Health Observatory 2013). The age distribution of penile cancer peaks in the eighth decade, with increased incidence from above the age of sixty (Dillner 2000; Graafland 2011). However, 20% of men affected are under 50 years old, and penile cancer has even been reported in children (Curado 2007).

Description of the condition

Histologically, 95% of penile cancers are squamous cell carcinomas (SCCs) that behave much like SCCs in other areas of the body. Hence, similar aetiological mechanisms have been proposed and several risk factors have been identified. Chronic inflammation from certain disorders such as lichen sclerosus, balanoposthitis, or phimosis has a 10‐fold risk (Daling 2005). Smoking incurs a 4.5‐fold increase in risk (Muñoz 2006). In addition, a viral aetiology has been postulated with human papilloma virus (HPV) reportedly associated with 70% to 100% of high grade penile intraepithelial neoplasia (PIN) and 40% to 50% of invasive penile cancer (Bezerra 2001). There is marked variability in HPV prevalence in men. For example in Asia, Thailand has a prevalence of 22% of men with HPV and 81% of penile cancers associated with HPV. In comparison, Japan has 5.3% HPV prevalence with only 12% of penile cancer cases in one study exhibiting HPV association (WHO/ICO 2010). Nevertheless, there is no conclusive evidence that HPV directly causes penile SCC, and it is likely that other co‐factors are responsible for its development, such as smoking and multiple sexual partners, although it may be implicated in some men (Daling 2005; Dillner 2000; Heideman 2007). (FS: suggest deleting this in response to referee Inman' s comments) Prevention through infant circumcision has been shown to almost eliminate the risk of penile cancer, but there is no risk reduction with adult circumcision (Dillner 2000). Hygiene is thought to prevent inflammation and hence may reduce the risk of penile SCC, although there is no evidence to suggest that smegma is carcinogenic (Van Howe 2006).

Penile cancer presents as a discrete penile lesion, either single or multiple. The commonest sites of involvement are glans penis (50%), prepuce (20%), both glans and prepuce (10%), coronal sulcus (6%), penile shaft (2%), and urethra (< 1%) (Sufrin 1991). Macroscopically, most of the lesions are either flat, ulcerative, or exophytic. Pathologically, the lesions can be classified as classic SCC, basaloid, verrucous, sarcomatoid, adeno‐squamous, or mixed. Basaloid and sarcomatoid penile cancer types have a poorer prognosis compared to classic SCC (Cubilla 1993). Verrucous and adeno‐squamous carcinoma have low malignant potential and very little reported association with HPV (Cubilla 2010). Unfortunately, many men present late and the disease can extensively replace the normal penile anatomy. At presentation, 20% to 30% of men have palpable inguinal lymphadenopathy (Horenbras and Watkin 2013 [pers comm]), of which more than 90% have inguinal nodal disease (Watkin 2013 [pers comm]), although the percentage may vary depending on the level of healthcare and socioeconomic factors for specific population groups. Men with impalpable nodes at presentation may have up to 20% risk of having occult metastases in the inguinal lymph nodes (Emerson 2001). Penile SCC has four different growth patterns: superficial spreading, vertical growth, verrucous, and multicentric (Cubilla 1993). These have different prognoses due to different dissemination rates. Three histological grades exist, corresponding to degree of differentiation, i.e. well differentiated (Grade 1 or G1), moderately differentiated (Grade 2 or G2), and poorly differentiated (Grade 3 or G3) (Velasquez 2008). In terms of staging, the most widely used staging system is the 2010 TNM staging system (Sobin 2009).

The diagnosis of penile cancer is usually made clinically with examination of the lesion along with determination of any inguinal lymphadenopathy. A biopsy is indicated for histological confirmation. Smaller lesions can be excised completely with excisional biopsy, whereas larger lesions require incisional biopsy followed by definitive treatment of the primary tumour. For men with palpable inguinal nodes, the options include: (1) immediate inguinal node biopsy; (2) bilateral radical inguinal lymphadenectomy if the primary lesion is high grade (i.e. G3) or high stage (i.e. ≥ T2), or both. The previous practice of a period of treatment with antibiotics and observation is no longer routinely recommended (Hakenberg 2014). If inguinal node biopsies confirm nodal disease, bilateral radical inguinal lymphadenectomy is required. For men with fixed inguinal nodes in which biopsies have confirmed cancer, neoadjuvant chemotherapy or radiotherapy can be given followed by radical inguinal lymphadenectomy. In patients without palpable inguinal nodes, inguinal node management depends either on risk stratification based on nomograms, or on grade and stage of the primary lesion or on the results of sentinel node biopsy (Ficarra 2006; Ficarra 2009; Graafland 2010; Ornellas 2008). (FS: editor suggests getting rid of much of this)

This Cochrane Review focuses on the management of primary tumour of localised penile cancer (i.e. carcinoma in situ (CIS), Ta, T1 to T3, N0, M0).

Description of the intervention

Definitive treatment for the primary tumour (i.e. penile) in localised penile cancer is dependent on the disease grade and stage. For Tis (or CIS) or Ta lesions, conservative penile sparing options can be considered. These include topical 5‐fluorouracil, topical imiquimod, CO₂ or Nd:YAG laser, photodynamic therapy, or Mohs micrographic surgery (MMS). These penile‐sparing treatments can give good cosmetic results but often circumcision is recommended along with treatment applied to the entire glans (Bandieramonte 2008). A relatively new technique of glans resurfacing has been employed where epithelial and sub‐epithelial tissue is removed and extra‐genital skin graft applied (Palminteri 2007). For G1 to G2 T1 disease, penile‐preserving treatment can be used following circumcision, and the options include wide local excision, glansectomy (i.e. amputation of entire glans, sometimes combined with reconstruction of a neo‐glans), partial glansectomy, laser, external beam radiotherapy (EBRT), or brachytherapy.

Amongst these options, glansectomy is reported to offer the best cancer outcome (Hadway 2006). Partial glansectomy is appropriate in select patients with > 50% normal glans. For G3 T1 disease, select patients can still undergo penis‐preserving treatments (Bissada 2003). For T2 to T3 and some G3 T1 disease, due to the high risk of progression and local recurrence associated with penile‐preserving techniques, most patients will require either a partial or total penectomy. Partial penectomy involves amputating the penis with a sufficient margin. Historically, a margin of at least 2 cm (centimetres) was thought to be essential, but 5 to 10 mm (millimetre) margins may be adequate (Minhas 2005). Ensuring a good margin can be assisted by intra‐operative frozen section. Reconstruction of the neo‐glans using corporal tissue and split skin grafting can be performed simultaneously to partial penectomy. If the proximal penile shaft is involved, then a total penectomy is required. With total penectomy, a perineal urethrostomy is created to aid voiding. The scrotum and testes are not routinely removed. EBRT and brachytherapy are options for T2 tumours < 4 cm, but both procedures are associated with a higher risk of local recurrence than surgery (Azrif 2006; Rozan 1995; Zouhair 2001), and hence are usually reserved for patients unfit for surgery. 

For the management of inguinal lymph nodes in patients with non‐palpable nodes, the options are dependent on the grade and stage of the primary tumour, and include active monitoring, dynamic sentinel node biopsy (Leijte 2007), and either bilateral modified or radical inguinal lymphadenectomy. If the inguinal nodes are found to be involved with cancer following radical inguinal lymphadenectomy, subsequent pelvic lymph node dissection is usually required, unless there is only one node with intracapsular spread (Lont 2007; Ornellas 2008). In patients with extensive nodal disease (i.e. N2 to N3), radiotherapy or chemotherapy can be considered either on its own, or in combination with surgery in a neoadjuvant or adjuvant fashion.

How the intervention might work

The various interventions under consideration have different effects on penile cancer. Surgical excision of any magnitude aims to eliminate the tumour cells before they can metastasise. Penile‐preserving procedures aim to limit functional loss whilst still achieving clearance of the patient’s cancer.

Cancer cells treated with lasers undergo a photothermal effect. If the energy density is sufficient it can ablate the tissue once over 100 degrees Celsius. Carbon dioxide laser have a much shallower penetration depth than ND:YAG lasers. The shallow penetration depth would theoretically limit their potential to treat tumours of a certain stage.

Mohs micrographic surgery examines surgical excision blocks during the procedure that then guide the surgeon based on margin information.

Glansectomy maintains the copora cavernosa and can be combined with skin grafting techniques to avoid burying of the remaining penis.

Glans resurfacing involves removal of the epithelial and sub‐epithelial tissue of the glans. This can be partial or total excision of the glans and is combined with skin grafting thereafter to maintain function.

Why it is important to do this review

The management of localised penile cancer is complex, and is influenced by a host of factors including site and extent of tumour involvement, tumour grade, local stage, regional lymph node status, presence of distant metastases, surgical fitness, and age. A range of surgical and non‐surgical therapeutic options exist for various stages of the disease. Increasingly, surgeons are tending to recommend penile preservation techniques, which allows for reconstruction of the penis following excision of the cancer. However, there is no ideal procedure. All options have advantages and disadvantages and it is not clear which option is best for different types of patients. This Cochrane Review will compare the clinical benefits and harms of the different options for the surgical management of localised penile cancer in terms of the primary tumour. The rarity of penile cancer means that individual surgeons and centres do not manage large numbers of such patients and hence their ability to clarify best practice is limited. This review aims to clarify current best practice underpinned by the evidence, where available, whilst also highlighting gaps needing primary research.

A penile cancer care pathway developed by an international collaboration of methodologists and clinical content experts with representation from British Association of Urological Surgeons (BAUS) and European Association of Urology (EAU) Penile Cancer Guidelines Panel members informed the scope of this review (Figure 1). The methodology underpinning the development of the pathway is described elsewhere (MacLennan 2011).

Figure 1.

Penile cancer care pathway. Abbreviations: 5‐FU = fluorouracil; ‐ve = negative; +ve = positive; chemo = chemotherapy; CIS = carcinoma in situ; CT = computed tomography; DSNB = dynamic sentinel node biopsy; EBRT = external beam radiation therapy; FNAC = fine needle aspiration cytology; MDT = multidisciplinary team; mets = metastasis; MRI = magnetic resonance imaging; SCC = squamous cell carcinoma; USS = ultrasound scanning; xRT = radiotherapy. Source: MacLennan 2011.

Objectives

To assess the effects of various surgical treatments for localised penile cancer, including oncological efficacy, quality of life and functional impact. (FS: do we need to be more specific than this?)

The comparisons of interest are: 

• one type of surgery versus another; 

• surgical treatment versus non‐surgical treatment.

Methods

Criteria for considering studies for this review

Types of studies

Eligible studies were randomised controlled trials (RCTs) or quasi‐RCTs (alternate allocation), non‐randomised comparative studies and non‐comparative studies (case series). Studies in any language published as full‐text articles and meeting abstracts were eligible.

Types of participants

Eligible participants were men diagnosed with primary penile cancer with CIS (Tis), pTa, or T1 to T3 squamous cell carcinoma, or Jackson stage I and II. We excluded men with T1 to T3, N1 to N3 tumours; T4 tumours; tumours other than squamous cell carcinoma (such as extra‐mammary Paget's disease or melanoma); unresectable inguinal nodes or pelvic nodes; metastases; or recurrent tumours.

During the study selection process we tightened the inclusion criteria to focus on participants with pathological N0 (see Appendix 1 for definition). Due to the small number of studies available, however, clinical N0 was accepted as a proxy, if there was no pathological confirmation of N0. Studies that combined data from patients with N‐ and N+ disease were included, so long as the results for N0 were reported separately or patients with N+ did not constitute >10% of the study population. Only those studies that included 10 or more patients per intervention group were included in the review.

Types of interventions

The following interventions were eligible.

• Laser (excision or vaporisation)

• Mohs micrographic surgery

• Circumision and local excision

• Glans resurfacing

• Glansectomy

• Partial penectomy

• Excision plus primary closure

Valid comparators were one of the included surgical treatments, non‐surgical treatments (including chemotherapy and radiotherapy), or surveillance. We excluded studies assessing the following interventions in their own right: adjuvant or neoadjuvant therapy, brachytherapy, external beam radiation therapy (EBRT), chemotherapy, photodynamic therapy, lymphadenectomy, or surveillance. 

Types of outcome measures

Reporting of any of the specified outcome measures was not a prerequisite for including studies.

Primary outcomes

• Cancer‐specific survival

• Recurrence‐free survival

• Adverse effects (e.g. lymphoedema)

Secondary outcomes

• Overall survival

• Penile preservation rate

• Urinary function

• Sexual function

• Disease‐specific quality of life

• General health‐related quality of life

• Non‐pre‐specified outcomes judged important when performing the review

Main outcomes for ‘Summary of findings’ table

• Cancer‐specific survival at two years

• Cancer‐specific survival at five years

• Recurrence‐free survival at two years

• Recurrence‐free survival at five years

• Adverse effects

• Urinary function

• Sexual function

Search methods for identification of studies

Electronic searches

We carried out extensive electronic searches to identify reports of published and ongoing RCTs, quasi‐RCTs and non‐randomised comparative studies on the surgical management of localised penile cancer. The search strategies were designed to retrieve all studies that assessed the efficacy of the specified surgical interventions. We sought both full‐text papers and conference abstracts. The search strategies were not restricted by year of publication or language. We searched the following databases (last search July 2015): the Cochrane Central Register of Controlled Trials (CENTRAL) on the Cochrane Library; Ovid MEDLINE; Ovid MEDLINE‐in‐Process; Ovid EMBASE; Science Citation Index; Biosis; Scopus in Press; and LILACS (Latin American and Caribbean Health Sciences Literature). We also searched for systematic reviews and other background publications in the Cochrane Database of Systematic Reviews, Health Technology Assessment (HTA) Database, and DARE (Database of Abstracts of Reviews of Effects). The current research registers were checked, including Current Controlled Trials (www.controlled‐trials.com), ClinicalTrials.gov (www.clinicaltrials.gov), and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/en). We searched the conference proceedings of key organisations including: the American Society of Clinical Oncology (ASCO), American Urological Association (AUA), EAU, BAUS, and Société Internationale d'Urologie (SIU). In addition, we searched the websites of key professional organisations, such as the National Cancer Institute and the American Cancer Society as well as grey literature sources.

Searching other resources

We screened reference lists of included studies for relevant citations. Where appropriate, we contacted trial authors and investigators to obtain unpublished data and to seek clarification on published data.

Data collection and analysis

Selection of studies

Following de‐duplication, two review authors independently screened the titles and abstracts of identified records for eligibility. We retrieved the full‐text articles of all potentially eligible records and two review authors independently screened them using a standardised form, linking together multiple records of the same study in the process. We resolved any disagreements by discussion or by consulting a third review author. Using a PRISMA flow diagram, we describe the study selection process (Liberati 2009).

Data extraction and management

Two review authors independently extracted outcome data using a standardised data extraction form. One review author extracted study characteristics and a second review author checked data extractions for accuracy. Any disagreements were resolved by discussion or by consulting a third review author.

For long‐term outcomes (e.g. survival), we extracted both time‐to‐event data (hazard ratio (HR)) and categorical data (survival rate) at one, two, and five years, where available.

Assessment of risk of bias in included studies

Two review authors independently assessed the risk of bias of each included study. Any disagreements were resolved by discussion or by consulting a third review author. In order to have a common standard with randomised controlled trials, the assessment used the approach recommended in the Cochrane Handbook (Higgins 2011; Higgins 2011a; Reeves 2011) for assessing risk of bias on the following seven domains: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective outcome reporting (reporting bias) and ‘other’ sources of bias (other bias).

For the assessment of non‐randomised studies, we have incorporated a draft extended ‘risk of bias’ instrument which was developed by the Cochrane Non‐Randomised Studies Methods Group (Reeves 2010), which, in addition to all the seven domains above, includes an item to assess the risk of findings being explained by confounding. Based on this instrument, we developed a list of potential confounding variables a priori with clinical consensus for specific outcomes.

• Tumour grade

• Tumour stage

• Lymphovascular invasion

• Histological cell type (basaloid/sarcomatoid versus squamous of no specific type)

• Previous CIS

For peri‐operative, functional and quality of life outcomes, we extracted information about the following potential confounding factors:

• Age

• Comorbidity

• Baseline sexual function

• Baseline urinary function

• Size of tumour

• Number and experience of surgeons involved

• Nature of penile reconstruction

• Length of penile stump

• Margins of clearance

Note that, for a number of variables for peri‐operative, functional and quality of life outcomes, such as baseline sexual and urinary functions, few meaningful data were available; these confounding variables therefore could not be assessed and are not presented in this review.

Furthermore, because the recommended Cochrane approach focuses on internal validity of comparative studies and is not always useful to differentiate the methodological quality of case series (with no internal comparison), concern was given to addressing external validity (applicability of results to different people, places or time) by assessing whether study participants were selected consecutively or representative of a wider patient population, an item which is common to various published sources offering guidance on assessing the methodological quality of non‐randomised studies (AHRQ 2012; Cowley 1995; Dalziel 2005; Deeks 2003; Guyatt 2008; Kang 2010; OCEBM 2011; SIGN 2011; Tseng 2009).

The quality of evidence for each of the outcomes was assessed using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach. The GRADE approach classifies the quality of evidence into four categories: high, moderate, low and very low. Evidence from randomised studies begin as ‘high’ quality evidence, whereas evidence from non‐randomised studies, due to their inherent bias, defaults to ‘low’ quality evidence. Evidence quality is rated up or down according to pre‐defined characteristics. Evidence quality is lowered by limitation of study design (risk of bias, or internal validity), inconsistency of results (heterogeneity, or widely differing estimates of the effect), indirectness of evidence, imprecision, and other sources of bias such as publication bias. If quality of evidence was not downgraded for any reasons, quality may be upgraded according to the following characteristics: large magnitude of effect, dose‐response gradient, or all plausible confounding which would reduce the effect or suggest a spurious effect if no effect was observed (Guyatt 2008).

Measures of treatment effect

We planned to report HRs for time‐to‐event data, risk ratio (RR) for categorical outcomes, and mean difference (MD) for continuous outcomes with corresponding 95% confidence intervals (CIs).

Unit of analysis issues

The primary analysis is per man randomised.

Dealing with missing data

We intended to conduct an intention‐to‐treat analysis if sufficient data were available. Instead the data allowed only for an available case analysis. We did not impute any missing data.

Assessment of heterogeneity

We aimed to assess heterogeneity between studies by visual inspection of plots of the data, the Chi² test for heterogeneity, and I² statistic (Higgins 2003). However, the data were not amenable to assess heterogeneity in this way. Instead, we narratively explored the possible reasons for heterogeneity, such as differences in the population studied, the treatment given, or the way in which the outcomes were assessed.

Assessment of reporting biases

We planned to use funnel plots to assess the risk of reporting or publication bias if we identified ten or more studies for a given comparison and outcome however this was precluded by the small amount of literature available.

Data synthesis

Data are tabulated by type of intervention and summarised narratively. Quantitative synthesis (meta‐analysis) was not performed due to clinical heterogeneity inherent in case series.

Subgroup analysis and investigation of heterogeneity

Subgroup analysis was planned to explore potential heterogeneity based on the tumour stage (CIS/Ta vs. T1‐3) but this was not performed due to a lack of data.

Sensitivity analysis

Sensitivity analyses by methodological quality of included studies could also not be performed due to a lack of data.

'Summary of findings' table

We assessed the overall quality of evidence using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach. GRADE classifies the quality of evidence into four categories: high, moderate, low, and very low. Evidence from RCTs begins as 'high' quality evidence, whereas evidence from non‐randomised studies, due to their inherent bias, defaults to 'low' quality evidence. Evidence quality is upgraded or downgraded according to pre‐defined characteristics. Evidence quality is lowered by limitation of study design (risk of bias), inconsistency of results (heterogeneity), indirectness of evidence, imprecision, and other sources of bias such as publication bias. If quality of evidence is not downgraded for any reason, quality may be upgraded according to the following characteristics: large magnitude of effect, dose‐response gradient, or all plausible confounding which would reduce the effect or suggest a spurious effect if no effect was observed (Guyatt 2008).

For long‐term data such as survival and recurrence in the summary of findings tables, we chose two years and five years as the time points most relevant for decision making. However, reported data are not amenable to these (e.g. some studies report recurrence‐free survival probability at two years, while others reported the rate of recurrence at last follow‐up) and therefore the following pragmatic decisions were made for the purpose of the summary of findings tables:

• If probability estimates at two and five years (e.g. survival curve) were reported, these were given priority.

• If probability estimates at two and five years were not reported, but the study reported data (the number of events) collected at last follow‐up, the arbitrary cut‐off point was set at 42 months (halfway between two and five years). That is, data collected at or before a mean or median follow‐up of 42 months were included in the 2‐year data. Data collected at or after a mean or median follow‐up of 43 months were included in the 5‐year data.

• Recurrence‐free survival rates (using censored data) of, say, 95%, were considered equivalent to recurrence rates of 5% (using uncensored data). Similarly, an overall survival rate of 95% (censored data) and an all‐cause mortality rate of 5% (uncensored data) were considered equivalent.

Appendices

Appendix 1. Definitions of clinical and pathological N0 used in this review

• ≤T3 disease with no indication to undergo further diagnostic inguinal biopsy/FNAC (i.e. clinical N0) • ≤T3 disease with indication (e.g. palpable nodes or intermediate/high grade disease) to undergo further diagnostic inguinal biopsy/FNAC, with subsequent diagnostic biopsies confirming no histological evidence of nodal disease (i.e. pathological N0) • ≤T3 disease with indication (e.g. palpable nodes or high grade disease) to modified inguinal lymphadenectomy with the results confirming no histological evidence of nodal disease (i.e. pathological N0) • ≤T3 disease with indication (e.g. palpable nodes or high grade disease) to undergo bilateral radical inguinal lymphadenectomy with the results confirming no histological evidence of nodal disease (i.e. pathological N0)

Appendix 2. Search strategy

CENTRAL

1. MeSH descriptor Penile Neoplasms, this term only

2. (penis or penile) near/4 (cancer* or carcinoma*or tumo*r* or neoplas*)

3. #1 or #2

MEDLINE (Ovid)

1. Penile Neoplasms/su [Surgery]

2. Penile Neoplasms/

3. ((penis or penile) adj4 (cancer$ or carcinoma$ or tumo?r$ or neoplas$)).tw.

4. ((penile or penis) adj2 (squamous cell carcinoma or scc)).tw.

5. or/2‐4

6. exp Fluorouracil/

7. (5‐fu or Efudix or Aldara or Zydara).tw.

8. imiquimod.tw.

9. (laser adj3 (excision or vapo?ri?ation)).tw.

10. ((Mohs adj4 surgery) or mms or Moh's paste).tw.

11. (moh$ and (surgery or technique)).tw.

12. (intraoperative chemotherapy or chemical surgery or chemosurgery).tw.

13. (glans adj3 resurfac$).tw.

14. glansectomy.tw.

15. penectomy.tw.

16. (excision adj3 primary closure).tw.

17. or/7‐16

18. randomized controlled trial.pt.

19. controlled clinical trial.pt.

20. randomi?ed.ab.

21. placebo.ab.

22. drug therapy.fs.

23. randomly.ab.

24. trial.ab.

25. groups.ab.

26. comparative study/

27. follow‐up studies/

28. time factors/

29. (preoperat$ or pre operat$).tw.

30. (chang$ or evaluat$ or reviewed or baseline).tw.

31. (prospective$ or retrospective$).tw.

32. (cohort$ or case series).tw.

33. (compare$ or compara$).tw.

34. case report/ use oemezd

35. case reports.pt.

36. or/18‐35

37. 5 and 17 and 36

38. 1 and 36

39. 37 or 38

40. exp animals/ not humans/

41. 39 not 40

Embase (Ovid)

1. Penile Neoplasms/su [Surgery]

2. exp Penis tumor/su [surgery]

3. or/1‐2

4. exp penis tumor/

5. ((penis or penile) adj4 (cancer$ or carcinoma$ or tumo?r$ or neoplas$)).tw.

6. ((penile or penis) adj2 (squamous cell carcinoma or scc)).tw.

7. or/4‐6

8. exp Fluorouracil/

9. (5‐fu or Efudix or Aldara or Zydara).tw.

10. imiquimod/

11. (laser adj3 (excision or vapo?ri?ation)).tw.

12. chemosurgery/

13. ((Mohs adj4 surgery) or mms or Moh's paste).tw.

14. (moh$ and (surgery or technique)).tw.

15. imiquimod.tw.

16. (intraoperative chemotherapy or chemical surgery or chemosurgery).tw.

17. (glans adj3 resurfac$).tw.

18. glansectomy.tw.

19. penis amputation/

20. penectomy.tw.

21. excision/

22. (excision adj3 primary closure).tw.

23. or/8‐22

24. exp clinical trial/

25. randomization/

26. randomi?ed.ab.

27. placebo.ab.

28. drug therapy.fs.

29. randomly.ab.

30. trial.ab.

31. groups.ab.

32. Treatment outcome/

33. major clinical study/

34. controlled study/

35. clinical trial/

36. (preoperat$ or pre operat$).tw.

37. (chang$ or evaluat$ or reviewed or baseline).tw.

38. (prospective$ or retrospective$).tw.

39. (cohort$ or case series).tw.

40. (compare$ or compara$).tw.

41. case report/

42. or/24‐41

43. 7 and 23 and 42

44. 3 and 42

45. 43 or 44

46. exp animals/ not humans/

47. 45 not 46

Science Citation Index

1. TS=(((penis or penile) SAME (cancer* or carcinoma* or tumor* or tumour* or neoplas* or squamous cell or scc) AND (surgery or surgical)))

2. TS=(Fluorouracil or 5‐fu)

3. TS=(laser SAME (excision or vapo*ri*ation))

4. TS=(Moh* SAME (surgery or technique))

5. TS=(imiquimod or chemosurgery or intraoperative chemotherapy or chemical surgery)

6. TS=(glansectomy OR (glans SAME resurfac*))

7. TS=((amputat* SAME (penis or penile)) OR penectomy)

8. TS=(excision SAME primary closure)

9. #9 OR #8 OR #7 OR #6 OR #5 OR #4 OR #3 or#2

10. TS=(trial* or random* or comparison or compare or comparative)

11. #1 and #9 and #10

BIOSIS (Web of Knowledge)

1. TS=((penis or penile) SAME (cancer* or carcinoma* or tumor* or tumour* or neoplas* or squamous cell or scc))

2. TS=(#1 SAME (surgery or surgical))

3. TS=(Fluorouracil or 5‐fu)

4. TS=(laser SAME (excision or vapo*ri*ation))

5. TS=(Moh* SAME (surgery or technique))

6. TS=((penis or penile) SAME (imiquimod or chemosurgery or intraoperative chemotherapy or chemical surgery))

7. TS=(glansectomy OR (glans SAME resurfac*))

8. TS=((amputat* SAME (penis or penile)) OR penectomy)

9. TS=(excision SAME primary closure)

10. #9 OR #8 OR #7 OR #6 OR #5 OR #4 OR #3

11. #10 AND #1

12. #11 OR #2

13. TS=(#12 SAME (trial* or random* or comparison or compare or comparative))

LILACS (Latin American and Caribbean Health Sciences Literature)

penis or penile or pene or peniano [Words]

and

cancer$ or carcinoma$ or tumor$ or tumour$ or neoplas$ or squamous cell carcinoma or scc [Words]

and

Pt CLINICAL TRIAL or Pt RANDOMIZED CONTROLLED TRIAL or Pt CONTROLLED CLINICAL TRIAL or random$ or trial$ or compara$ or compare$ or cohort$ or retrospective or prospective [Words]

Cochrane Database of Systematic Reviews

1. MeSH descriptor Penile Neoplasms, this term only

2. (penis or penile) near/4 (cancer* or carcinoma*or tumo*r* or neoplas*)

3. #2 or #3

HTA (Health Technology Assessment)

1. MeSH DESCRIPTOR Penile Neoplasms EXPLODE ALL TREES

2. ((penis or penile) adj4 (cancer* or carcinoma* or tumor* or tumour* or neoplas* or squamous cell carcinoma or scc))

3. #1 or #2

DARE (Database of Abstracts of Reviews of Effects)

1. MeSH DESCRIPTOR Penile Neoplasms EXPLODE ALL TREES

2. ((penis or penile) adj4 (cancer* or carcinoma* or tumor* or tumour* or neoplas* or squamous cell carcinoma or scc))

3. #1 or #2

Current Controlled Trials

Basic search: penile cancer

ClinicalTrials.gov

Basic search: penile cancer

WHO ICTRP (World Health Organisation International Clinical Trials Registry Platform

Basic search: penile cancer

ASCO (American Society of Clinical Oncology

Basic search: penile cancer

AUA (American Urological Association)

Basic search: penile cancer

EAU (European Association of Urology

Basic search: penile cancer

BAUS (British Association of Urological Surgeons)

Basic search: penile cancer

SIU (Société Internationale d'Urologie)

Basic search: penile cancer

National Cancer Institute

Basic search: penile cancer

American Cancer Society

Basic search: penile cancer

What's new

Last assessed as up‐to‐date: 11 April 2016.

Date	Event	Description
27 March 2017	Amended	​This title has been withdrawn in agreement with the authors due to a lack of progress and low prioritization.

Contributions of authors

Fiona Stewart: developed and ran search strategies, screened search, data extraction, data analysis, drafted manuscript.

Mari Imamura, Steven MacLennan: wrote the protocol, screened search, data extraction, data analysis, drafted manuscript.

Temitope Adewuyi: screened search, data extraction, data analysis, reviewed manuscript.

Thomas BL Lam: wrote the protocol background, screened search, clinical advice, reviewed analysis, reviewed manuscript.

Ross Vint: wrote the protocol background, data extraction (German language), reviewed manuscript.

Graeme MacLennan: statistical support, reviewed manuscript.

Muhammad Imran Omar: reviewed analysis, reviewed manuscript.

T R Leyshon Griffiths, Duncan J Summerton (both BAUS): developed care pathway, clinical advice, reviewed analysis, reviewed manuscript.

Nick Watkin, Simon Horenblas, Oliver Walter Hakenberg (all EAU) and Sara MacLennan (UCAN urological cancer charity): clinical and content expert advice, reviewed manuscript.

James MO N'Dow: concept, clinical advice, reviewed analysis, reviewed manuscript.

Sources of support

Internal sources

• UCAN (a Scottish Urological Cancer charity), UK.

External sources

• No sources of support supplied

Declarations of interest

Fiona Stewart, Mari Imamura, Steven MacLennan, Temitope Adewuyi, Thomas BL Lam, Ross Vint, Graeme MacLennan, Muhammad Imran Omar, Sara MacLennan, Nick Watkin, Simon Horenblas and James MO N'Dow have no known conflicts of interest.

T R Leyshon Grifffiths declares the following relevant financial activities outside the submitted work: received travel support from GlaxoSmithKline; honoraria for lectures from GlaxoSmithKline, GE Healthcare, and Astellas, for consultancy from Alliance Pharma, and for review preparation from GE Healthcare on topics including lower urinary tract symptoms, bladder cancer, and renal cysts/masses; declares having no financial activities related to the submitted work.

Oliver Walter Hakenberg declares the following relevant financial activities outside the submitted work: employed as Professor of Urology; received honoraria for lectures at meetings for postgraduate medical education; received royalties from book contributions in urology; received support for meeting participation at the German Transplant Society and EAU meetings sponsored by Astellas and GlaxoSmithKline.

Duncan J Summerton declares the following financial activities related to the submitted work: received support for travel and accommodation to attend urological conferences from Lilly UK, GlaxoSmithKline, Coloplast, and American Medical Systems (AMS), including support from Lilly to attend EAU 2012 at which the penile cancer panel meeting occurred.

Notes

This title has been withdrawn in agreement with the authors due to a lack of progress and low prioritization.

Withdrawn from publication for reasons stated in the review