Demographic and Pathologic Differences in the Incidence of Invasive Penile Cancer in the United States, 1995-2003

Penile cancer accounts for <1% of malignancies among men in developed countries (1). The risk of penile cancer shows marked international variation, with rates 10- to 20-fold higher in some regions in Central and South America (Goiania, Brazil), Asia (Chiang Mai, Thailand), and Africa (Kyadondo, Uganda) than in other parts of the world (1-3). Approximately 1,280 men in the United States will be diagnosed with penis and other genital cancers and 290 men will die from these malignancies in 2007 (4). African Americans have been reported to have twice the risk of penile cancer compared with European Americans, but this study was small and the statistics need to be updated (5).

Penile cancer is uncommon, so there have been few etiologic studies conducted of this disease (6-10). Factors associated with the risk of penile cancer include a history of smoking, poor hygiene (especially phimosis in uncircumcised men), multiple sexual partners, and sexually transmitted infections. Recent evidence is compelling for a role of human papillomavirus (HPV) in the etiology of squamous cell carcinoma of the penis, with high-risk HPV-16 detected in 29% of tumors in a recent Dutch study (11).

Descriptive investigations of penile cancer in the United States are presently lacking because of the low incidence of this malignancy. A review of the literature revealed only one recent descriptive analysis of penile cancer using Surveillance Epidemiology End Results (SEER) data, although this report on squamous cell carcinomas focused on survival (12). The investigators found that African American men presented with penile cancer at a younger age and more advanced stage of disease at diagnosis than did European Americans, but these results were significant only among men with advanced, nonmetastatic disease. Based on the limited knowledge of the causes and distribution of penile cancer, we examined racial and ethnic variation in this malignancy by age at diagnosis, stage, histology, geographic region, and calendar year at diagnosis.

The North American Association of Central Cancer Registries provided two data sets that were used in this analysis. The first file, Cancer Incidence in North America Deluxe, was used in the analysis of race (White, Black, American Indian, Asian-Pacific Islander; ref. 13). The second file, Cancer in U.S. Hispanics/Latinos, was used in the analysis of Hispanic ethnicity (14). These two files included data for persons diagnosed with cancer from 1995 to 2003 that were obtained from registries meeting the standards for data quality defined by Ellison et al. (13). States that participated in this analysis through written consent were California, Colorado, Connecticut, Delaware, Florida (excluding Hispanics), Georgia, Hawaii, Idaho, Indiana, Iowa, Kentucky, Louisiana, Maine, Massachusetts, Michigan, Minnesota, Montana, Nebraska, Nevada, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Texas, Utah, Washington, West Virginia, and Wisconsin. These States represented ∼68% of the U.S. population during the 9-year time period, including 68% of the White population, 59% of the Black population, 48% of the Alaska Native-American Indian population, 83% of the Asian-Pacific Islander population, and 75% of the U.S. Hispanic population. Although there were too few Alaska Native cases to carry out some of the subgroup analyses (N = 20), they were included as part of the “total”.

All penile cancer cases (N = 6,539 men) meeting the North American Association of Central Cancer Registries standards with International Classification of Diseases for Oncology (15) topography codes C60.0 to C60.9 were included in the analytic file. A total of 98% of the cases were microscopically confirmed. Hispanic identification was available from 5,841 (89%) of the men included in the analysis.

A summary stage variable, developed by the SEER Program, was used to group cases into one of four categories: localized, regional, distant, or unknown (16). Localized tumors were those confined entirely to the organ of origin; regional tumors were those that extended into surrounding organs and tissues (or regional lymph nodes); and distant tumors were those that had spread to remote organs or lymph nodes. Tumor grade was classified as well-differentiated (grade 1), moderately differentiated (grade 2), poorly differentiated and undifferentiated (grades 3 and 4), or unknown grade.

To examine time trends in penile cancer incidence, we used SEER data from public use files including data from nine cancer registries (Atlanta, Connecticut, Detroit, Hawaii, Iowa, New Mexico, San Francisco Bay Area, Seattle-Puget Sound, and Utah) between 1973 and 2003. A total of 2,388 penile cancers were identified through this source, representing ∼26% of the U.S. population.

Population counts, used as denominators for calculating the cancer rates included in this analysis, were derived from the 2000 U.S. Census (17). The Census Bureau provided estimates of the resident populations of the U.S. counties by 5-year age groups and newborns (0, 1-4, 5-9,…, 80-84, and 85 years), gender (male or female), race (White, Black, American Indian, Asian-Pacific Islander), and Hispanic origin (Hispanic, non-Hispanic).

SEER*Stat (version 6.2.4) was used to calculate both annual and average annual age-adjusted incidence rates expressed per million population (18). All rates were age adjusted by 5-year age groups to compare racial and ethnic groups by demographic, pathologic, and clinical variables of interest. Confidence intervals for the rates were calculated by the method of Fey and Feuer (19). The 2000 U.S. population was used for age standardization. For all analyses, rates were suppressed when the category had fewer than 20 cases although counts were always provided. To control for possible confounding of the difference in the age distribution among specific racial and ethnic groups on the comparison of, for example, stage or histology, a relative percentage of the column (C%) age-adjusted incidence rate was calculated instead of a percentage based on case counts. This quantity was calculated as the age-adjusted incidence rate for a particular category divided by the age-adjusted incidence rate for the total group in the column. For example, the relative percentage of localized cancers for Whites was the age-adjusted localized rate divided by the overall age-adjusted incidence rate for Whites. Additionally, to control for the differences in race distributions among geographic regions on the comparison of stage, the relative column percentages were race standardized using the 2000 U.S. population as the standard. We examined differences in the log age-specific incidence rates separately by race and Hispanic ethnicity using the Poisson model as follows:

• log (cases/population) = β₀ + β₁ Black + β₂ API + β₃ Age

• log (cases/population) = β₀ + β₁ Hispanic + β₂ Age

To assess a trend in incidence rates, the annual percent change in rates between 1973 and 2003 was calculated. The statistical comparison of trends was based on the t test of the trend regression parameter. Three-year moving averages were used in the trend plots to reduce the effect of random variation (“noise”) in the plots.

Squamous cell carcinomas were the most common type of penile cancer accounting for 93% of all reported malignancies (Table 1). Verrucous carcinomas were the most frequently specified subtype of squamous cell carcinoma, followed by large cell, microinvasive, and spindle cell types. Other epithelial cancers included basal cell and transitional cell carcinomas. Rare histologic types included adenocarcinoma, melanoma, and Kaposi sarcoma.

Hispanic men experienced the highest age-adjusted rate of penile cancer among all racial and ethnic groups with rates that were 64% higher than in Black men, 69% higher than in White men, 134% higher than in American Indian men, and 174% higher than in Asian or Pacific Islander men (Table 2). The relative risk of penile cancer among Hispanic men was 1.80 [95% confidence interval (95% CI), 1.66-1.96] compared with non-Hispanic men.

Distinct age differences in the incidence of penile cancer were observed. The mean age at diagnosis was 67 years among White men and 63 years among Black and Asian-Pacific Islander men. The incidence of invasive penile cancer increased monotonically with increasing age beginning at age 20 to 24 years (Fig. 1). Age-specific rates among Asian-Pacific Islanders were lower than rates for Whites and Blacks beginning in the 35-39 year age group. The global difference between White and Asian-Pacific Islander curves for men 35 years and older was highly significant (P = 0.0002), whereas there was no difference in the curves for White and Blacks (P = 0.77). The rate for Black men 85 years and older (N = 56 cases) was 36.2 per million (95% CI, 27.3-47.0) compared with 23.1 per million (95% CI, 21.2-25.1) among White men in this age group (N = 545 cases).

Age-specific rates of penile cancer among Hispanic men (mean age at diagnosis, 58 years) were higher than among non-Hispanic men (mean age at diagnosis, 67 years) beginning in early adulthood (Fig. 2). In contrast to non-Hispanic men whose penile cancer rates increased monotonically throughout life, penile cancer rates among Hispanic men increased substantially after the age of 69 years with rates among Hispanics (46.9 per million; 95% CI, 33.9-63.1) more than twice that among non-Hispanics (21.1 per million; 95% CI, 19.1-23.2) after the age of 85 years. The difference in slopes for the age-specific rates for men 35 years and older was significant for these two groups (P < 0.0001).

The majority of penile cancers (61%) were diagnosed at a localized stage (Table 3). White men were diagnosed at earlier stages than were Black men: the age-adjusted rate of localized penile cancer was higher among White men than among Black men, whereas the rates of more advanced and unstaged cancers among Blacks exceeded those of Whites. Hispanics were also more likely to be diagnosed with advanced or unstaged disease compared with non-Hispanics.

The majority of penile cancers were diagnosed at well-differentiated (25%) or moderately differentiated grade (32%), followed by poorly or undifferentiated (17%; data not shown). Unknown grade accounted for 26% of all tumors. Although there was some variation in tumor grade by racial and ethnic group, these differences were small.

The incidence of penile cancer among all races combined was significantly higher in the South (4.42 per million) and significantly lower in the West (3.28 per million) than in the other regions of the United States (Fig. 3). The high rates of penile cancer in the South and the low rates in the West were consistent for all racial groups (Table 4). White men had the highest rates of penile cancer in the Northeast and West, in contrast to Black men who had the highest rates in the South and Midwest. The highest rate of penile cancer was found among Black men in the South (4.77 per million) and the lowest rate among Asian-Pacific Islander men in the West (1.84 per million).

Hispanic men in the West had significantly lower (45%) rates of penile cancer than Hispanic men in the Midwest. Although penile cancer rates were consistently higher among Hispanic than among non-Hispanic men, the ratio (relative risk) of Hispanic-to-non-Hispanic rates was highest in the Midwest (2.62; 95% CI, 1.63-4.11) and lowest in the West (1.50; 95% CI, 1.28-1.75).

Men were more likely to be diagnosed with penile cancer at a localized stage in the Midwest and more likely to be diagnosed at a regional or distant stage in the West than in other regions of the United States.

An analysis of time trends in the risk of penile cancer showed a significant (P = 0.0002) decrease of −1.2% (95% CI, −0.6% to −1.8%) in the average annual incidence rate between 1973 and 2003 (Fig. 4). The decline in rates was more dramatic for Black men than White men: the average percent change in incidence was −1.9% (95% CI, −0.7% to −3.2%) for Black men (P = 0.004) and −1.1% (95% CI, −0.5% to −1.7%) for Whites (P = 0.0009) during the 31-year time period. In 1973, the penile cancer rate was 8.2 (95% CI, 3.7-21.4) for Black men and 4.8 (95% CI, 3.6-6.4) for White men compared with rates of 1.9 (95% CI, 0.6-4.9) for Black men and 3.2 (95% CI, 2.5-4.1) for White men in 2003.

This analysis supports distinct racial and ethnic differences in the incidence of penile cancer in the United States. A unique finding of this analysis was the high rate of penile cancer among Hispanic men in the United States compared with other racial and ethnic groups. This is consistent with the high incidence of penile cancer reported in Goiania, Brazil (27 per million) and Puerto Rico, United States (26 per million), rates which were only exceeded by Kyadondo County, Uganda (40 per million; ref. 1). Rates among White and Black men were similar but significantly higher than the incidence of penile cancer among Asian-Pacific Islanders.

Contrary to reports from other countries (20), penile cancer was extremely rare in children—only two cases were reported among 15- to 19-year-olds. A comparison of age-specific rates of penile cancer suggests that the high rate of penile cancer among Hispanic men and the low rate among Asian-Pacific Islander men were not limited to a single age group: the risk profile for this malignancy seems to begin in early adulthood for all racial and ethnic groups. We are unable to explain the sharp increase in the incidence of penile cancer among very old (≥85 years) Hispanic and Black men. The possibility of cohort effect–associated reduced rates of circumcision or increasingly poor genital hygiene among older men cannot be excluded (21, 22).

Squamous cell carcinomas represented the predominant histologic type of penile cancer with verrucous (“warty”) carcinoma accounting for 6% of all incident cancers. Kaposi sarcoma, a rare histologic type strongly associated with HIV infection (23), was the only penile cancer subtype that was more common among Black men than among White men. Up to 18% of men with AIDS-related Kaposi sarcoma are reported to have penile involvement (24).

SEER data show that the rate of penile cancer in the United States has been declining among White and Black men during the past three decades and rates for these two groups seem to have converged. Etiologic studies of penile cancer have identified several risk factors including HPV infection, lack of circumcision or phimosis (the inability to retract the foreskin over the glans), and cigarette smoking (2, 3, 5, 6, 9, 10, 25).

Presently, no population-based data are available about the prevalence of HPV infection of the penis among men in the United States. In a recent systematic review of the literature, Dunne et al. (26) found a wide range (1-73%) of genitourinary HPV prevalence among men worldwide, with most studies reporting a prevalence of ≥20%, which is similar to the HPV prevalence found among women (∼27%; ref. 27). We note that the differences in the rates of penile cancer by racial/ethnic group parallel those of cervical cancer, another HPV-related malignancy, with higher rates among Hispanic women and lower rates among Asian-Pacific Islander women (1). Indeed, a correlation between penile and cervical cancer incidence rates has been noted by several investigators (28-30). The sharp peak in penile cancer incidence among very old Hispanic and Black men may indicate differences in age at initial infection and/or viral persistence compared with other racial/ethnic groups. The seropositivity to HPV-16, HPV-18, or HPV-45, the most common oncogenic types of HPV, was 46% among penile cancer cases and 12% among controls (odds ratio, 5.0; 95% CI, 1.4-17.2) in a case-control study in Uganda (25). Positive HPV-16 serology was found among 24% of cases and 12% of controls in a North American case-control study (odds ratio, 1.9; 95% CI, 1.2-3.2), and 80% of penile cancer tissue specimens were positive for HPV-DNA (9). Variability in the prevalence of HPV in penile cancer tissue from published reports (21-83%) may be explained by geographic variation in the distribution of HPV genotypes and differences in HPV detection methods (31).

The lower rate of penile cancer among Asian-Pacific Islanders compared with other racial and ethnic groups suggests that oncogenic HPV infection may be less prevalent among Asian-Pacific Islander men because screening variation is not relevant to penile cancer (32). Because genital HPV infection is spread through sexual contact, behavioral differences between racial and ethnic groups may be a partial explanation for the variation in rates. Indeed, Asian-Pacific Islander high school students are significantly less likely than White, Black, or Hispanic students to have ever engaged in sexual intercourse (33).

Data from the National Health and Social Life Survey showed substantial variation in the prevalence of circumcision by birth cohort, racial and ethnic group, and geographic region of the United States that parallels some of the variability in penile cancer rates (21, 22). Laumann et al. (22) observed a steady increase in the prevalence of circumcision from a low of 31% in 1932, peaking in 1965 at 85%, and declining to 77% in 1973, the last year of the survey. The prevalence of circumcision among racial and ethnic groups noted in this survey was inversely associated with rates of penile cancer found in our analysis: the level of circumcision among Whites (81%) was much higher than the prevalence among Blacks (65%) or Hispanics (54%). The convergence of circumcision rates among younger men of all ethnic and racial groups in the United States may explain the steeper decline in penile cancer incidence observed in the SEER data between 1973 and 1999 and the much higher rates of penile cancer among older Black men. Circumcision rates among boys born in the 1940s and early 1950s in the South were much lower than in other parts of the United States (21), paralleling the higher rate of penile cancer found in this geographic region.

The majority of penile cancers were diagnosed at a localized stage, in accord with other observations (34), but stage at diagnosis varied by race and ethnicity. Black and Hispanic men were diagnosed with a higher relative percentage of advanced-stage and unstaged penile cancer than were White and non-Hispanic men, suggesting ethnic and racial disparities in health care. This observation may reflect interregional variation in diagnostic patterns because few Black or Hispanic men with penile cancer resided in the Midwest, the region with the highest percentage of early-stage cancer. Alternatively, socioeconomic status may influence stage at diagnosis, and at least one study reported an inverse gradient between penile cancer and social class (35).

Social class and racial/ethnic differences in the prevalence of tobacco smoking (36) may explain reports of an association of smoking with the risk for penile cancer in some studies (9, 10), but not others (6).

In summary, this analysis revealed substantial racial/ethnic and geographic variation in the incidence of penile cancer in the United States. The high rate of penile cancer among Hispanic and Southern Black men suggests differences in risk factors for this malignancy that may have changed during the past three decades or by birth cohort. This descriptive study of penile cancer may reasonably contribute to the development of novel etiologic hypotheses about this malignancy and provide the basis for future collaborative population-based case-control studies.