Penile cancer is an uncommon malignancy in developed countries, with an estimated 1290 new cases of invasive penile cancer and 290 deaths among men in the United States in 2009, but is much more common in the developing countries of Asia, Africa, and South America. This disease can result in loss of function, disfigurement, and death. Thus, recognizing penile cancer early in the clinical setting and accurately diagnosing the patients is critical. Because the management and prognosis varies by the extent of local disease, lymph node status, and other factors, accurate staging of penile cancer is of utmost importance. This article focuses on the presentation, diagnosis, and staging of invasive squamous cell carcinoma of the penis. The authors highlight the recent changes to the American Joint Committee on Cancer’s staging system for penile carcinoma and discuss other prognostic factors and predictive models.
Penile cancer is an uncommon malignancy in developed countries, with an estimated 1290 new cases of invasive penile cancer and 290 deaths among men in the United States in 2009, but is much more common in the developing countries of Asia, Africa, and South America. Between 1998 and 2001, invasive squamous cell penile cancer accounted for 0.1% of all invasive tumors diagnosed among men in the United States, whereas this figure is reportedly was high as 10% to 20% of cancers among men in some countries. This disease can result in loss of function, disfigurement, and death. The 5-year survival for all patients is approximately 50%; it ranges from 66% for patients with negative lymph nodes to 25% to 30% for those with positive inguinal lymph nodes and approaches 0% for those with pelvic lymph node metastases. Thus, recognizing penile cancer early in the clinical setting and accurately diagnosing the patients is critical. Because the management and prognosis varies by the extent of local disease, lymph node status, and other factors, accurate staging of penile cancer is of utmost importance.
Other articles in this issue have described the premalignant and early carcinomas of the penis as well as imaging for penile cancer and management of the lymph nodes. This article focuses on the presentation, diagnosis, and staging of invasive squamous cell carcinoma of the penis. The authors highlight the recent changes to the staging system of the American Joint Committee on Cancer (AJCC) for penile carcinoma, and discuss other prognostic factors and predictive models.
Risk factors
Penile cancer can occur in any male patient, but certain risk factors have been identified. Although up to 15% of cases occur in men younger than 50 years, penile cancer is a disease of older men, with a median age at diagnosis in the United States of 68 years and with an increased risk as the age advances beyond 50 years. Penile cancer is far more common among uncircumcised men than those who were circumcised early in life. Phimosis is strongly associated with the risk for penile cancer and hampers surveillance of the glans, inner preputial layer, and coronal sulcus, which are the areas of high incidence. Human papilloma virus (HPV) infection is identified in about half of the men with penile cancer and some HPV subtypes (such as HPV-16 and HPV-18) have been associated with malignant transformation of condyloma acuminata. Sexual behaviors such as high number of lifetime partners also confer additional risk of penile cancer. Human immunodeficiency virus (HIV) is associated with an 8-fold increased risk of penile cancer, but this may be mediated to an extent by the higher incidence of HPV among men with HIV. Cigarette smokers are 3 to 4.5 times more likely to develop penile carcinoma than nonsmokers, and users of other tobacco products are also at increased risk. Psoralen-UV-A photochemotherapy for psoriasis has also been shown to be a risk factor for penile cancer. Lichen sclerosus (balanitis xerotica obliterans) is associated with a 3% to 9% risk of development of penile cancer over long-term follow-up, and other premalignant lesions are also, by definition, risk factors.
Presentation
Approximately two-thirds of men with penile carcinoma present with localized disease. Penile carcinoma most often presents with a visible or palpable lesion on the penis. The lesion may arise de novo as invasive carcinoma, or may have progressed from a premalignant lesion or carcinoma in situ. The appearance varies from an ulceration with heaped-up edges to areas of induration or erythema to a warty, exophytic appearance. In a review of more than 3500 published cases from 1908 to 1984, Huben and Sufrin found that the most common lesion was described as a mass, lump, or nodule in 47% of cases, a sore or ulcer in 35%, an inflammatory lesion in 17%, and an incidental diagnosis on evaluation of a circumcision specimen in 0.7%. The lesion may be hidden by the foreskin, particularly in patients with phimosis, which has been reported to be as high as 60% among patients with penile cancer in Brazil. The lesion may also be obscured by associated inflammatory conditions such as balanitis or lichen sclerosis.
Penile carcinoma can be associated with penile pain, discharge, bleeding, or foul odor. These secondary signs may be particularly important components of the presentation in patients with phimosis.
The most common locations for penile cancer are the glans penis, prepuce, and coronal sulcus. A recent study of almost 5000 cases of invasive carcinoma of the penis among men in the United States showed that the primary site of disease was the glans penis in 34.5% of cases, prepuce in 13.2%, shaft in 5.3%, overlapping in 4.5%, and unspecified in 42.5%. Thus, of those cases with a specified site, only 11.2% had the disease confined to the shaft. These results corroborate earlier data compiled by Huben and Sufrin, who found that the lesions were confined to the glans in 48% of patients, only on the prepuce in 21%, glans or prepuce with extension to the shaft in 14%, both glans and prepuce in 9%, coronal sulcus in 6%, and isolated carcinomas on the shaft in less than 2%.
The patients may also have noticed masses in their inguinal regions, which can be a sequela of local inflammation and infection or evidence of lymph node metastases. Patients with advanced disease may complain of fatigue, weight loss, or bone pain.
There is often a delay in presentation with penile cancer, likely owing to the social stigma and denial. Between 15% and 50% of patients delayed seeking treatment for at least 1 year after first noticing symptoms. For example, Narayana and colleagues showed that of 176 patients with documentation of the interval between onset of symptoms and seeking medical attention, 85 (48.3%) sought treatment within 6 months, 37 (21.0%) waited between 6 and 12 months, and 54 (30.7%) waited for more than 1 year. Penile cancer is lethal once it has spread to regional lymph nodes and beyond, making any delay in diagnosis, whether because the patient delays seeking treatment or the physician delays a necessary diagnostic procedure, unacceptably perilous.
In addition to the generic histopathologic designation of squamous cell carcinoma there are 4 subtypes recognized by AJCC (verrucous, papillary squamous, warty, and basaloid); some investigators have also reported an adenosquamous subtype and a sarcomatoid variant. A review of 61 cases in the United States by Cubilla and colleagues found that 59% of patients had conventional squamous cell carcinoma, 15% papillary, 10% basaloid, 10% warty, 3% verrucous, and 3% sarcomatoid. A more recent study of 333 cases in Brazil by Guimarães and colleagues showed a similar distribution: 65% of patients had conventional squamous cell carcinoma, 5% papillary, 4% basaloid, 7% warty, 7% verrucous, 1% sarcomatoid, 1% adenosquamous, and 10% mixed. In both studies, the basaloid and sarcomatoid variants carried the worst prognosis. Each subtype has a unique histopathologic appearance and natural history and may present in different ways. For example, verrucous carcinomas, which rarely metastasize, are typically large, exophytic, fungating masses without other sites of involvement. On the other hand, basaloid carcinomas often present at higher stages and have a higher likelihood of lymph node metastases than conventional squamous cell carcinomas. Sarcomatoid carcinomas are also very aggressive. In a series of 15 cases, most tumors were large (2.5–7cm) polypoid lesions, with surface ulceration affecting the glans in 93% cases and invading deeply into the corpora cavernosa in 80%. Eight out of 9 patients who underwent lymphadenectomy had positive lymph nodes.
Presentation
Approximately two-thirds of men with penile carcinoma present with localized disease. Penile carcinoma most often presents with a visible or palpable lesion on the penis. The lesion may arise de novo as invasive carcinoma, or may have progressed from a premalignant lesion or carcinoma in situ. The appearance varies from an ulceration with heaped-up edges to areas of induration or erythema to a warty, exophytic appearance. In a review of more than 3500 published cases from 1908 to 1984, Huben and Sufrin found that the most common lesion was described as a mass, lump, or nodule in 47% of cases, a sore or ulcer in 35%, an inflammatory lesion in 17%, and an incidental diagnosis on evaluation of a circumcision specimen in 0.7%. The lesion may be hidden by the foreskin, particularly in patients with phimosis, which has been reported to be as high as 60% among patients with penile cancer in Brazil. The lesion may also be obscured by associated inflammatory conditions such as balanitis or lichen sclerosis.
Penile carcinoma can be associated with penile pain, discharge, bleeding, or foul odor. These secondary signs may be particularly important components of the presentation in patients with phimosis.
The most common locations for penile cancer are the glans penis, prepuce, and coronal sulcus. A recent study of almost 5000 cases of invasive carcinoma of the penis among men in the United States showed that the primary site of disease was the glans penis in 34.5% of cases, prepuce in 13.2%, shaft in 5.3%, overlapping in 4.5%, and unspecified in 42.5%. Thus, of those cases with a specified site, only 11.2% had the disease confined to the shaft. These results corroborate earlier data compiled by Huben and Sufrin, who found that the lesions were confined to the glans in 48% of patients, only on the prepuce in 21%, glans or prepuce with extension to the shaft in 14%, both glans and prepuce in 9%, coronal sulcus in 6%, and isolated carcinomas on the shaft in less than 2%.
The patients may also have noticed masses in their inguinal regions, which can be a sequela of local inflammation and infection or evidence of lymph node metastases. Patients with advanced disease may complain of fatigue, weight loss, or bone pain.
There is often a delay in presentation with penile cancer, likely owing to the social stigma and denial. Between 15% and 50% of patients delayed seeking treatment for at least 1 year after first noticing symptoms. For example, Narayana and colleagues showed that of 176 patients with documentation of the interval between onset of symptoms and seeking medical attention, 85 (48.3%) sought treatment within 6 months, 37 (21.0%) waited between 6 and 12 months, and 54 (30.7%) waited for more than 1 year. Penile cancer is lethal once it has spread to regional lymph nodes and beyond, making any delay in diagnosis, whether because the patient delays seeking treatment or the physician delays a necessary diagnostic procedure, unacceptably perilous.
In addition to the generic histopathologic designation of squamous cell carcinoma there are 4 subtypes recognized by AJCC (verrucous, papillary squamous, warty, and basaloid); some investigators have also reported an adenosquamous subtype and a sarcomatoid variant. A review of 61 cases in the United States by Cubilla and colleagues found that 59% of patients had conventional squamous cell carcinoma, 15% papillary, 10% basaloid, 10% warty, 3% verrucous, and 3% sarcomatoid. A more recent study of 333 cases in Brazil by Guimarães and colleagues showed a similar distribution: 65% of patients had conventional squamous cell carcinoma, 5% papillary, 4% basaloid, 7% warty, 7% verrucous, 1% sarcomatoid, 1% adenosquamous, and 10% mixed. In both studies, the basaloid and sarcomatoid variants carried the worst prognosis. Each subtype has a unique histopathologic appearance and natural history and may present in different ways. For example, verrucous carcinomas, which rarely metastasize, are typically large, exophytic, fungating masses without other sites of involvement. On the other hand, basaloid carcinomas often present at higher stages and have a higher likelihood of lymph node metastases than conventional squamous cell carcinomas. Sarcomatoid carcinomas are also very aggressive. In a series of 15 cases, most tumors were large (2.5–7cm) polypoid lesions, with surface ulceration affecting the glans in 93% cases and invading deeply into the corpora cavernosa in 80%. Eight out of 9 patients who underwent lymphadenectomy had positive lymph nodes.
Diagnosis
Clinicians must maintain a high index of suspicion for penile carcinoma in men with cutaneous lesions of the penis, particularly in men with risk factors for penile carcinoma. A careful history must be taken including the length of time the lesion has been present, the evolution of the lesion, and previous treatments for it. A complete review of systems is appropriate to identify symptoms of locally advanced disease (such as urinary obstruction), regional spread (such as lower extremity edema), and distant or advanced disease (such as constitutional symptoms).
The aims of the physical examination are to determine the need for biopsy, the local extent of disease, and the extent of spread. The physician should document the site of the lesion and describe its characteristics, including size, configuration (ulcerative, flat, warty, or exophytic), and a clinical assessment of the involvement of adjacent or deep structures (such as the scrotum, corporal bodies, urethra, pubic bone, and prostate).
A short trial of topical treatment for superficial, inflammatory-appearing lesions is acceptable, but prompt follow-up and further evaluation for incompletely treated lesions is imperative.
Ultimately, as stated in the European Association of Urology (EAU) guidelines, the AJCC, and the International Union Against Cancer (UICC), punch, excisional, or incisional biopsy is required to make a pathologic diagnosis. Obtaining a deep biopsy specimen is essential for accurate staging, and thus superficial or shave biopsies are not appropriate. Biopsy is followed by definitive local treatment, either after frozen-section diagnosis or as a separate procedure. In either case, the physician must obtain informed consent from the patient for the full extent of the expected procedure.
Clinical staging
Primary Tumor
Local extent of disease is determined clinically by physical examination. As described in the previous section, the aim is to document the location and estimate the extent and depth of local involvement. Specifically, the physician should determine whether there is involvement of the corporal bodies, urethra, scrotum, prostate, or pubic bone, because these factors impact the stage and, in turn, the likelihood of lymph node disease and prognosis.
There has been debate regarding the use of imaging for determining the clinical stage of the primary tumor. This topic is covered in detail elsewhere in this issue by Inman. In brief, Lont and colleagues compared physical examination with ultrasonography and magnetic resonance imaging (MRI) for clinical staging of penile cancer. Thirty-three patients underwent physical examination and both imaging modalities before surgery. The investigators found that physical examination was highly accurate for detecting corpora cavernosa invasion, with a positive predictive value of 100%, sensitivity of 86%, and specificity of 100%. The respective figures were 75%, 100%, and 91% for MRI and a disappointing 67%, 57%, and 91% for ultrasonography. Other investigators have compared MRI with pharmacologically induced erection to physical examination and have found the imaging to add to clinical staging accuracy. For example, Petralia and colleagues found that MRI and pharmacologically induced erection accurately staged 12 of 13 patients, whereas physical examination understaged 3 of 12 and overstaged 2 of 12 patients, for an overall accuracy of 8 of 13. Further studies are required to determine the best method for clinical staging of penile cancer.
Regional Lymph Nodes
Lymph node involvement is a critical component of treatment planning and prognosis. Up to 58% of patients have palpable inguinal lymph nodes at the time of presentation, whereas less than half of these patients have positive lymph nodes at final pathologic examination; other enlarged lymph nodes are caused by inflammatory conditions or infection at the primary site of disease. On the other hand, 15% to 20% of patients with nonpalpable lymph nodes turn out to have disease. Status of the inguinal lymph nodes is ordinarily determined by physical examination augmented by computed tomography (CT) in obese patients or those with a previous inguinal surgery.
Because of the inadequacies of clinical staging of the inguinal lymph nodes, some clinicians have turned to the use of imaging and predictive models to guide management. The roles of ultrasonography, CT, MRI (with and without nanoparticle enhancement), and positron emission tomography (PET) have been investigated. Other researchers have investigated fine-needle aspiration (FNA), but so far FNA has a limited role in evaluating lymphadenopathy in men with low-risk disease who otherwise lack indications for lymphadenectomy. Dynamic sentinel lymph node mapping is another option for staging the lymph nodes that is taking hold in some centers of excellence. Each of these alternatives to traditional clinical staging is discussed by other investigators in this issue.
Many investigators have focused on identifying factors that predict positive lymph nodes in men with invasive squamous cell carcinoma of the penis. The aim of such efforts is to identify subclinical lymph node metastases and, perhaps, to avoid the morbidity of lymphadenectomy in men who are unlikely to harbor disease in the lymph nodes. For example, in 2006 Ficarra and colleagues combined several of these factors into a nomogram designed to predict the presence of lymph node metastases based on clinical and tumor characteristics. The nomogram was based on 175 Italian men from 11 centers who underwent partial or total penectomy. Variables predictive of pathologically positive lymph nodes were tumor thickness greater than 5 mm, superficial growth pattern (compared with vertical), high-grade disease, lymphatic tumor embolization, corpora cavernosa invasion, corpus spongiosum invasion, urethral invasion, and clinically positive lymph nodes. The area under the receiver operating characteristic (ROC) curve was 0.876, suggesting excellent accuracy, but this result is yet to be validated in other cohorts. A more recently published nomogram, relying on age, clinical status of the lymph nodes, tumor stage, and grade has an area under the curve of 0.74. Another model, based on 193 cases from Brazil and the United States, combines histologic grade, level of infiltration, and presence of perineural invasion to develop a prognostic index, which correlates with the likelihood of lymph node involvement. Although this model is simpler in requiring fewer features to arrive at the risk group, the investigators did not rely on a standard staging system to categorize tumors with regard to depth of invasion. Furthermore, they assigned values to each feature (eg, 1 for lamina propria invasion, 2 for corpus spongiosum, urethra, or dartos, and 3 for corpus cavernosum or preputial skin) without the use of models to determine their relative contribution to the likelihood of lymph node metastases.
A slightly different approach was taken by Solsona and colleagues in 2001, who developed a risk stratification system based on tumor stage and grade. These investigators used a retrospective series of 66 patients to develop the risk strata and a prospective series of 37 cases to test it. Patients were categorized as low risk (Tis or T1, G1), intermediate risk (T1 G2 or G3, or T2 G1), or high risk (T2 G2, or T2–3 G3). Including both the retrospective and prospective groups, positive lymph nodes were identified in 0 of 32 low-risk patients; 12 of 34 (35%) intermediate-risk patients; and 30 of 37 (81%) high-risk patients. The EAU guidelines adopted a similar risk stratification scheme but defined low risk as Tis, Ta G1–2 or T1 G1, intermediate risk as T1 G2, and high risk as pT2 or higher or G3 tumors. The systems are similar with respect to the categorization of low-risk patients, but the approach adopted by Solsona and colleagues categorizes more patients as intermediate-risk and fewer as high-risk. These 2 systems were compared for their accuracy in predicting the presence of lymph node metastases in a large cohort of Italian patients with penile cancer. The investigators found that each of the systems was an independent predictor of lymph node metastases when controlling for clinical stage of lymph nodes, tumor thickness, and lymphovascular invasion. However, the area under the ROC curve was only 0.632 (95% confidence interval [CI], 0.548–0.715) and 0.697 (95% CI 0.618–0.777) for the risk groups defined by the EAU guidelines and Solsona and colleagues, respectively, suggesting that both schemes are about equivalent but neither of them have high accuracy in predicting lymph node involvement.
In addition to grade and stage, individual factors that have been shown to be associated with the likelihood of lymph node involvement include lymphovascular invasion (embolization), perineural invasion, depth of tumor invasion, structures involved (ie, difference between urethral, corpus spongiosum, and corpora cavernosa invasion), size of the tumor, anatomic location, pattern of growth (superficial spreading vs vertical, and so forth), front of invasion (infiltrative vs pushing), histologic subtype, percentage of poorly differentiated tumor, positive surgical margin, and palpable inguinal lymph nodes. Genetic and molecular markers are also being investigated for this purpose. Further studies are needed to determine the pathologic, genetic, and molecular markers that would be most valuable in predicting lymph node involvement. It should be emphasized that the use of predictive factors and prognostic models should be based on the final pathologic tumor specimen rather than on the result of a biopsy, because the biopsy can underestimate the depth of invasion and miss other key features of the tumor.
Distant Metastases
Distant spread should be ruled out with chest radiograph and CT of the abdomen and pelvis in patients with pathologic risk factors for lymph node involvement or palpable lymph nodes. The EAU guidelines support the use of cross-sectional imaging only in patients with palpable lymph nodes. However, cross-sectional imaging also has a role in patients in whom an inguinal examination is compromised (obese patients and those with previous inguinal surgery). It also may be important in patients with nonpalpable lymph nodes who have indications for lymphadenectomy. Although it could be argued that these patients would be staged surgically at the time of lymphadenectomy, preoperative evaluation of the pelvic lymph nodes with CT could guide the use of multimodal therapy, just as in patients with palpable lymph nodes. Staging should also include blood tests such as a complete blood count and comprehensive metabolic panel, and coagulation studies should be performed before any biopsy or intervention. Bone scan is performed in selected situations, such as bone pain, elevated alkaline phosphatase or serum calcium levels, or advanced disease.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
