Disparities in Penile Cancer Pranav Sharma, MD, Kamran Zargar-Shoshtari, MD, Curtis A

Differences in racial/ethnic, sociodemo-

graphical, and economic disparities

in penile cancer can influence cancer

control and survival-related outcomes.

Birds Nest. X-Ray image courtesy of Bryan Whitney. www.x-rayphotography.com.

Disparities in Penile Cancer Pranav Sharma, MD, Kamran Zargar-Shoshtari, MD, Curtis A. Pettaway, MD, Matthew B. Schabath, PhD, Anna R. Giuliano, PhD, and Philippe E. Spiess, MD

Background: Although penile cancer is a rare malignancy in developed nations, racial and socioeconomic differences exist in the incidence of the disease and its associated survival-related outcomes. Methods: A search of the literature was performed for research published between the years 1990 and 2015. Case reports and non–English-language articles were excluded, instead focusing specifically on large, population-based studies. Results: The incidence of penile cancer is higher in Hispanic and African American men compared with whites and Asians. Men with penile cancer also appear to have a distinct epidemiological profile, including lower educational and income levels, a history of multiple sexual partners and sexually transmitted infections, and lack of circumcision with the presence of phimosis. African American men presented at a younger age with a higher stage of disease and worse survival rates when compared with white men. Rates of cancer-specific mortality increased with age, single marital status, and among those living in regions of lower socioeconomic status. Conclusions: An understanding of sociodemographical differences in the incidence and survival rates of patients with penile cancer can help advance health care policy changes designed to improve access and minimize disparities in cancer care for all men alike.

Introduction However, its incidence is higher in some countries of Penile cancer is rare in the United States, with an Asia, Africa, and South America, where it represents estimated 2,030 new cases in the United States and 10% of all malignancies diagnosed among men.2 340 related deaths estimated to occur this year.1 Patients with penile cancer appear to have a distinct epidemiological profile consisting of lower income and education levels, poor hygiene, a history of From the Departments of Genitourinary Oncology (PS, KZ-S, PES) and Cancer Epidemiology (MBS, ARG), H. Lee Moffitt Cancer Center sexually transmitted infections, phimosis, and multiple & Research Institute, Tampa, Florida, and the Department of sexual partners.3 Risk factors for penile cancer include Urology (CAP), University of Texas MD Anderson Cancer Center, lack of circumcision, presence of phimosis or paraphi- Houston, Texas. mosis, poor hygiene, human papillomavirus (HPV) in- Submitted January 21, 2016; accepted February 26, 2016. fection, presence of genital warts, chronic inflamma- Address correspondence to Philippe E. Spiess, MD, Department of Genitourinary Oncology, Moffitt Cancer Center, 12902 Magnolia Drive, tory conditions of the foreskin, glans penis, or penile WCB-GU PROG, Tampa, FL 33612. E-mail: [email protected] shaft (eg, balanitis xerotica obliterans, lichen sclerosis), Dr Sharma is now affiliated with the Department of Urology at Texas presence of smegma (a whitish substance that accumu- Tech University Health Sciences Center in Lubbock, Texas. lates underneath the foreskin), tobacco abuse, or histo- No significant relationships exist between the authors and the com- 4,5 panies/organizations whose products or services may be referenced ry of psoriasis treatment with ultraviolet light. Tumor in this article. grade and stage, basaloid or sarcomatoid histological

October 2016, Vol. 23, No. 4 Cancer Control 409 features, and the presence and extent of lymph-node (SEER) program of the National Cancer Institute. The (LN) involvement in the groin or pelvis have all been annual, average age-adjusted incidence rate of pe- identified as important prognostic factors for penile nile SCC was 0.81 cases per 100,000 men, and rates cancer in determining long-term, cancer-related surviv- steadily increased with age.8 The incidence of penile al outcomes.6 In addition to disease-specific character- SCC was comparable in whites and African Americans, istics, significant demographical, racial/ethnic, social, but it was approximately twofold lower in Asians and and economic disparities exist with regard to both the Pacific Islanders.8 Rates among Hispanics were 72% incidence and survival rates of penile cancer.7 This is higher compared with non-Hispanics, which may be especially important considering the diverse patient attributable to a decreased prevalence of circumci- population in which this disease presents.2 These find- sion in this ethnic group.11 From 1993 to 2002, white ings also highlight the importance of considering these Hispanics had the highest incidence rates of penile differences when developing health care policies to cancer (1.01 per 100,000 men) followed by Alaska Na- help improve access to standard medical care and re- tives/American Indians (0.77 per 100,000 men) and Af- duce the treatment gap in cancer management. rican Americans (0.62 per 100,000 men).11 Incidence of penile SCC was also elevated in the southern United Methods States and in regions of lower socioeconomic status. A search of the medical literature was performed for However, the study had limitations because the educa- research published between January 1990 and Decem- tion level of participants was not analyzed. ber 2015. Case reports and non–English-language ar- Colón-López et al10 compared the incidence of pe- ticles were excluded, focusing instead on large, pop- nile cancer in Puerto Rico with other racial and ethnic ulation-based studies from national or international groups in the United States. Puerto Rican men had a datasets. A total of 54 studies were initially identified, higher incidence of penile cancer than non-Hispanic 30 of which are included in this review. white men (standardized rate ratio [SRR] 3.33), non- Hispanic African American men (SRR 3.04), and oth- Incidence er US Hispanic men (SRR 2.59).10 This difference was The incidence of penile cancer varies within spe- more pronounced for Puerto Rican men younger than cific racial/ethnic and socioeconomic groups, and a 60 years of age when compared with similarly aged US summary of the reviewed literature analyzing dispari- counterparts.10 Puerto Rican men with the lowest so- ties in penile cancer incidence based on large popula- cioeconomic status had a 70% higher incidence rate of tion-based data sets is shown in Table 1.8-10 penile cancer than Puerto Rican men with the highest Hernandez et al8 examined the burden of inva- socioeconomic status (SRR 1.70).10 Less education sive squamous cell carcinoma (SCC) of the penis in (< 12 years of education) was also statistically associat- the United States from 1998 to 2003 based on data ed with a higher incidence rate of penile cancer among from the Surveillance, Epidemiology, and End Results Puerto Rican men (SRR 2.18).10 However, the overall rate of penile cancer in Puerto Rico declined over time. In other studies, Hispanic and African American Table 1. — Overview of the Disparities in Incidence of Penile men have been reported to have the highest age-ad- Cancer in Select Population-Based Studies justed rates of penile cancer compared with whites, Study Data Set Select Findings Asians, or American Indians.9,12 The highest rates of Colón-López10 Puerto Rico Puerto Rican men had higher penile cancer were observed in Hispanic and African Cancer Registry rates of penile cancer than other American men older than 85 years of age, but penile SEER program groups cancer was rare among males younger than 20 years of Lower socioeconomic status and educational level associated age. Analyses to assess temporal trends from 1995 to with higher incidence rates in 2003 revealed a statistically significant decline in the Puerto Rican men incidence of penile cancer for African Americans and Goodman9 US population- Hispanic and African American men whites. Approximately 60% of penile cancers were di- based registries had highest rates of penile cancer agnosed at a localized stage (≤ pT2) among all racial African American men diagnosed and ethnic groups, although Hispanic and African at advanced stage with more node- positive disease American men tended to be diagnosed at an advanced Highest incidence in the southern stage (≥ pT3) with a higher incidence of node-positive United States disease than whites.9 Racial and ethnic differences in Hernandez8 SEER program Penile cancer rates higher in tumor grade were not found. The incidence of penile Hispanics and those living in regions cancer was highest in the southern United States and of lower socioeconomic status or in the southern United States lowest in the western states. The highest age-adjusted incidence rate of penile SCC was seen in African Amer- SEER = Surveillance, Epidemiology, and End Results. ican men living in the southern United States, and the

410 Cancer Control October 2016, Vol. 23, No. 4 lowest rate was found among Asian-Pacific Islanders tumors when compared with patients who had fewer living in western states.9 The incidence of primary than 6 lifetime female sexual partners.3 However, this malignant penile cancer has decreased significant- trend was not significant when the number of sexual ly over time from 1973 to 2002 based on data from partners was adjusted for age at first coitus and past SEER; however, the incidence of nodal disease has in- history of sexually transmitted infections. HPV-posi- creased over time.13 tive tumors were found in 36% of cases studied, and The overall incidence of penile cancer in the they were characterized by warty or basaloid morphol- United Kingdom (England and Wales) is higher than ogy in addition to a high histological grade in the ma- in Australia and the United States, although the inci- jority of cases.3 dence of penile cancer in all 3 countries was stable over time.14 Although cancer-specific mortality rates Treatment were higher in England and Wales compared with The management of penile carcinoma has evolved Australia and the United States, the incidence ra- in recent years, resulting in a shift in the treatment of tios were similar for all 3 countries.14 In Finland, the the disease.24 Options for the treatment of localized age-adjusted incidence rate of penile cancer has de- disease have expanded to organ-sparing therapy, and creased from the 1960s onward.15 Penile cancer was treatment of advanced disease has developed into a primarily a disease of elderly men during this time multidisciplinary, team-based approach using multi- in Finland, and no systematic geographical variation modal therapies.24 Demographical, racial/ethnic, and or predilection in terms of incidence was observed.15 socioeconomic disparities can influence the treatments The most common predisposing risk factors in Finland offered for patients with penile cancer in different pop- were phimosis (44% of cases) and condylomatous ulations (Table 2).25,26 genital lesions (20% of cases).15 Zhu et al26 used data from SEER to identify indi- viduals diagnosed with penile SCC from 1998 to 2009 Human Papillomavirus Infection and treated with either local tumor excision or par- An estimated 40% to 80% of invasive penile cancers tial or total penectomy for their primary tumor. Of the have been linked to HPV infection.16-18 Warty and ba- 1,292 eligible patients, 24.2% underwent local tumor saloid penile carcinomas have been associated with excision for surgical treatment of penile cancer.26 HPV infection, particularly HPV types 16 and 18, more For stage T1 disease, the rates of local tumor exci- so than typical SCC or verrucous carcinoma of the pe- sion increased from 29% to 40% within 10 years.26 nis.19,20 No significant association between HPV status Following a multivariable analyses, age younger than and pathological tumor stage, grade, or LN status was 65 years, African American descent, tumor size of found in this study.19 less than 3 cm, and pathological stage disease of no The impact of HPV infection on the prognosis of higher than T1 were reported predictors for the use patients with penile cancer is still controversial. Lont of local tumor excision as surgical treatment of the et al21 reported an association between high-risk HPV primary penile tumor as part of a penile-sparing ap- infection and reduced disease-specific mortality from proach.26 The authors concluded that penile-sparing penile cancer (hazard ratio [HR] 0.14), whereas Her- surgery is underutilized in the general population, nandez et al16 observed no association between HPV and significant age and racial/ethnic disparities exist status and rate of survival in patients with penile can- in its use.26 cer. HPV-associated penile malignancies, similar to Other studies have noted that the majority of pa- other nonviral penile cancers, have been associated with lower education level (incidence density ratio 1.32; P = .0002) and income status (incidence density Table 2. — Overview of the Disparities in Treatment for ratio 1.59; P < .0001).22 Current and former smokers Penile Cancer in Select Population-Based Studies were significantly more likely to have an α-HPV infec- Study Data Set Select Findings 23 tion than any other type. Zhu25 SEER pro- Men < 50 y of age more likely to receive In a study by Chaux et al3 of 103 Paraguayans with gram extensive LN dissection (≥ 8 LNs removed) penile cancer, patients with penile cancer were gener- during inguinal LN dissection for high-risk penile cancer ally found to dwell in rural or suburban areas (82%), Zhu26 SEER pro- More likely to receive penile-sparing approach live below the poverty line (75%), have a lower level of gram for surgical treatment of primary tumor: education (91%), and have a significant smoking his- Age < 65 y tory (76%). Phimosis (57%), moderate or poor hygienic African American decent habits (90%), and a history of sexually transmitted in- Tumor size < 3 cm 3 fections (74%) were also observed. Patients with more Pathological stage ≤ T1 than 10 lifetime female sexual partners were nearly LN = lymph node, SEER = Surveillance, Epidemiology, and End Results. 4 times as likely to present with HPV-positive penile

October 2016, Vol. 23, No. 4 Cancer Control 411 tients identified with penile SCC in the SEER data set Table 3. — Overview of the Disparities in Survival Rates of were treated with surgical therapy, and a small per- Penile Cancer in Select Population-Based Studies centage of patients received radiotherapy alone or as adjuvant therapy.27,28 No statistically significant so- Study Data Findings ciodemographic or racial/ethnic differences were re- Set ported with regard to radiotherapy for penile cancer Hernandez8 SEER Penile SCC mortality elevated in: program due to small sample size.28 African American men Zhu et al25 also assessed the impact of patient Hispanic men characteristics on the extent of inguinal LN dissec- Elderly (age > 80 y) tion when it is clinically indicated for patients with Southern US residents high-risk penile SCC. Using SEER data, the authors Low-income areas identified 393 patients who underwent regional LN dis- Less-educated populations section for penile cancer; this study cohort was then Rippentrop27 SEER African American race independent stratified into 2 groups, limited LN dissection (< 8 LNs program predictor of worse cancer-specific survival removed) and extensive LN dissection (≥ 8 LNs re- Sharma31 NCDB African American race independent predictor of worse OS moved).25 The median number of dissected LNs in this Private insurance and higher median series was 15, and 28% of patients underwent limited income independent predictors of better OS 25 LN dissection. The prevalence of extensive LN dissec- Thuret32 SEER Single marital status independent predictor of: tion decreased in tandem with increasing age (81% in program Locally advanced primary tumor 25 men aged < 50 years to 65% in men aged ≥ 70 years). Higher-grade disease On multivariable analysis, age alone was the inde- Increased overall mortality pendent predictor of extensive LN dissection (odds Tyson30 SEER Tumors on penile shaft and overlapping 25 ratio 0.98; P = .01). Inadequate LN dissection with few program lesions increased cancer-specific mortal- LNs removed was observed in some patients with pe- ity vs preputial lesions nile cancer, particularly among elderly men, even de- Verhoeven3 SEER 5-year cancer-specific survival of penile spite evidence that inguinal lymphadenectomy has a program cancer in United States decreased from similar morbidity rate in this age group compared with 1998 to 2011 men younger than 65 years of age.25,26 NCDB = National Cancer Database, OS = overall survival, SEER = Surveillance, Epidemiology, and End Results. Survival Factors affecting health insurance coverage and ac- had a 1.5-fold higher risk of having locally advanced cess to early detection and treatment could influence (pT3/T4, N+, M+) or higher grade (grade 3/4) disease the cancer-specific mortality rate of penile cancers. A at the time of surgery (P < .001) and a 1.3-fold higher summary of the literature analyzing disparities in pe- risk of overall mortality (P = .001); however, marital nile cancer survival rates based on select, large popula- status had no effect on cancer-specific mortality.32 Age, tion-based data sets is shown in Table 3.3,8,27,30-32 socioeconomic status (median family income above Rippentrop et al27 first described the racial and or below the poverty line within the patient’s county socioeconomic disparity in survival from penile car- of residence), and race (white vs African American vs cinoma based on data from SEER. They reported that other) did not correlate with rates of locally advanced African American men presented at a younger age and or higher-grade disease and cancer-specific mortality.32 with a higher stage of disease compared with other However, the study was limited by its small sample size racial groups. The disease-specific risk of death for of African Americans (n = 183). African American men was also nearly twice as high In a multivariable competing-risk model, Thuret than for whites (P = .0023), and marriage was associ- et al33 studied a population of patients with cN0T1 pe- ated with improved disease-specific rates of survival in nile SCC after primary tumor excision and before ini- the localized (P = .0002) and regional stages (P = .001) tial LN dissection and found no statistically significant of disease.27 Multivariable analysis demonstrated that a association between race and age (≥ 70 vs ≤ 69 years) higher stage at diagnosis, age older than 65 years, Af- for cancer-specific mortality after accounting for rican American race, and positive LNs were all signifi- tumor grade. Tyson et al30 also found no significant independent predictors of cancer-specific survival cant association between age and penile SCC-spe- from penile carcinoma.27 cific mortality rate after adjusting for tumor grade Thuret et al32 also used SEER data to evaluate the based on SEER data. Penile tumors on the body of association between marital status, tumor stage and the penis (HR 1.61) and overlapping lesions (HR grade, overall mortality, and cancer-specific mortal- 1.79; P = .01) had increased cancer-specific mortality ity in patients with penile SCC. Multivariable logis- compared with preputial lesions.30 Furthermore, the tic regression models revealed that unmarried men disease-specific 10-year survival rate of those with

412 Cancer Control October 2016, Vol. 23, No. 4 preputial tumors was 89.4% compared with 78.7% for cally significant higher mortality rate when compared tumors at other locations combined (P < .0001).30 with non-Hispanic white men (HR 3.32), non-Hispanic Hernandez et al8 analyzed more than 4,900 cases African American men (HR 2.51), and other US Hispan- of histologically confirmed invasive penile SCC in the ic men (HR 1.89).10 A higher rate of mortality related United States using SEER data, the Centers for Disease to penile cancer was also observed among Puerto Ri- Control and Prevention’s National Program for Cancer can men in the lowest socioeconomic status (HR 1.61) Registries, and the National Center for Health Statis- when compared with Puerto Rican men in the highest tics. The authors found that the rate of mortality due socioeconomic status, although this difference was not to penile SCC increased with age (mortality rate at age statistically significant.10 An increasing trend in mortal- < 50 years: 0.02 vs age > 80 years: 1.80).8 African Amer- ity due to penile cancer was observed among men in ican men were also diagnosed at a significantly young- Puerto Rico due to a possible convergence of early and er age than white or Asian men (median age, 62 vs 68 high levels of sexual activity, low circumcision rates, vs 68 years, respectively) and had a lower incidence-to- and high prevalence rates of sexually transmitted in- mortality ratio than white or Asian men (3.44 vs 5.18 fections in Puerto Rico — particularly among the medi- vs 5.63, respectively; P < .01 for all).8 This was also true cally underserved. for Hispanic men compared with non-Hispanic men Paiva et al34 reported an increased cancer-specif- (median age, 58 vs 69 years; incidence-to-mortality ra- ic mortality rate in patients younger than 40 years of tio: 4.61 vs 5.25; P < .01).8 Rate of mortality due to pe- age at a mean follow-up period of 22.4 months (19%) nile SCC was also increased among those men living in compared with patients aged between 40 and 60 years southern states, in regions of lower socioeconomic sta- (11%) and those older than 60 years of age (13%; tus, and among those living in areas with lower levels P < .05). Younger patients also had an increased like- of educated people.8 lihood of perineural invasion (26% vs 13%) and infil- Sharma et al31 used the National Cancer Data- trative growth pattern (70% vs 50%) on pathological base (NCDB) to study patients with a diagnosis of review, and they had a higher risk of recurrence penile SCC from 1998 to 2011, and they found that (45% vs 25%; P < .05 for all).34 Study findings from African American and Hispanic men were diag- Sharma et al31 similarly showed an increased preva- nosed at a younger age (< 55 years: 38.3% vs 25.9% lence of node-positive disease in patients younger than and 51.0% vs 24.5%; P < .01), and African Americans 55 years of age compared with those older than presented with a higher stage of disease (pT3/T4: 75 years (21.5% vs 13.1%; P < .01) but a decreased prev- 16.6% vs 13.2%; P = .027) and had worse median over- alence of pT3 to T4 disease (12.0% vs 15.4%; P = .007). all survival (OS) rates (68.6 vs 93.7 months; P < .01). Others have also shown younger men (< 50 years) were The estimated median OS rate in the entire cohort more likely to receive more extensive LN dissection was 91.9 months at a median follow-up period of (> 8 LNs removed) than older men (> 70 years; 44.7 months, and survival did not change over the 81% vs 65%; P = .01) with better rates of cancer-spe- study period.31 Hispanic ethnicity was not associated cific survival P( = .006).25 These findings remained with all-cause mortality in the penile cancer popula- true even in the subgroup of study patients with tion.31 Location and education level also did not have node-positive penile cancer (P = .01).25 Older age was a significant relationship with OS. Patients with pri- an independent predictor of OS in the population- vate health insurance and a median income of at least based study using the NCDB on multivariable analysis $63,000 based on zip code presented with a lower (P < .01), but its relationship with cancer-specific mor- stage of disease (pT3/T4: 11.6% vs 14.7%, P = .002; tality is unknown.31 12.0% vs 14.0%, P = .042) and had better median rates Large population-based data have also been used of OS (163.2 vs 70.8 months, P < .01; 105.3 vs 86.4 to demonstrate penile cancer survival trends over time. months, P = .001).31 Multivariable analysis revealed that The NCDB study did not show a correlation between African American race (HR 1.39; 95% confidence inter- the year of diagnosis and OS during the period from val [CI]: 1.21–1.58; P < .01) was independently associat- 1998 to 2011.31 However, Verhoeven et al35 reported ed with worse OS, whereas private insurance (HR 0.79; that the 5-year cancer-specific survival rate of pa- 95% CI: 0.63–0.98; P = .028) and higher median income tients with penile cancer in the United States has sig- of at least $63,000 (HR 0.82; 95% CI: 0.72–0.93; P = .001) nificantly decreased from 72% to 63% based on SEER were independently associated with better OS.31 data, and they also confirmed a significantly increas- Colón-López et al10 compared the mortality rates ing relative risk of death with increasing age (relative of penile cancer in Puerto Rican men with other ra- increase risk of death = 1.15; P < .01). Other groups cial and ethnic groups in the United States and evalu- have also observed an increase in the incidence of ated the influence of socioeconomics and educational regional-stage disease (pN+) over time in US men that level on the rates of mortality observed in Puerto Rico. also paralleled rising age.36 Furthermore, Arya et al37 From 2000 to 2004, Puerto Rican men had a statisti- reported on more than 9,500 men diagnosed with pri-

October 2016, Vol. 23, No. 4 Cancer Control 413 pathologic differences in the incidence of invasive penile cancer in mary penile cancer in England from 1979 to 2009 and the United States, 1995-2003. Cancer Epidemiol Biomarkers Prev. noted that the mortality rates fell by 20% after 1994 2007;16(9):1833-1839. 10. Colón-López V, Ortiz AP, Soto-Salgado M, et al. Penile cancer dis- (from 0.39 to 0.31 per 100,000 men). The 5-year sur- parities in Puerto Rican men as compared to the United States population. vival rate they observed also increased from 61.4% to Int Braz J Urol. 2012;38(6):728-738. 11. Laumann EO, Masi CM, Zuckerman EW. Circumcision in the United 70.2%, and the 5-year survival rate for men diagnosed States. Prevalence, prophylactic effects, and sexual practice. JAMA. between 2006 and 2010 was 77% for men younger than 1997;277(13):1052-1057. 12. Slopnick EA, Kim SP, Kiechle JE, et al. Racial disparities differ for 60 years of age and 53% for men older than 80 years African Americans and hispanics in the diagnosis and treatment of penile of age.37 The 8% difference in 5-year survival rate (66% cancer. Urology. 2016;96:22-28. 13. Barnholtz-Sloan JS, Maldonado JL, Pow-sang J, et al. Incidence vs 74%) between men living in the most affluent vs the trends in primary malignant penile cancer. Urol Oncol. 2007;25(5):361-367. most deprived areas during this time was not statisti- 14. Sewell J, Ranasinghe W, De Silva D, et al. Trends in penile cancer: 37 a comparative study between Australia, England and Wales, and the US. cally significant. The authors propose that the 21% Springerplus. 2015;4:420. increase in incidence of penile cancer since the 1970s 15. Maiche AG. Epidemiological aspects of cancer of the penis in Finland. Eur J Cancer Prev. 1992;1(2):153-158. in England could be explained by changes in sexual 16. Hernandez BY, Goodman MT, Unger ER, et al. Human papillomavirus habits, greater exposure to sexually transmitted infec- genotype prevalence in invasive penile cancers from a registry-based United States population. Front Oncol. 2014;4:9. tions, such as HPV infection, and the decreasing rates 17. Kirrander P, Kolaric A, Helenius G, et al. Human papillomavirus of childhood and neonatal circumcision.37 They also prevalence, distribution and correlation to histopathological param- eters in a large Swedish cohort of men with penile carcinoma. BJU Int. suggest that the improvement in survival may be due 2011;108(3):355-359. to advances in medical care, including diagnostic, stag- 18. Miralles-Guri C, Bruni L, Cubilla AL, et al. Human papillomavi- 37 rus prevalence and type distribution in penile carcinoma. J Clin Pathol. ing, and surgical techniques. 2009;62(10):870-878. 19. Gregoire L, Cubilla AL, Reuter VE, et al. Preferential association of human papillomavirus with high-grade histologic variants of penile-invasive Conclusions squamous cell carcinoma. J Natl Cancer Inst. 1995;87(22):1705-1709. Disparities exist in the incidence, treatment, and out- 20. Rubin MA, Kleter B, Zhou M, et al. Detection and typing of human papillomavirus DNA in penile carcinoma: evidence for multiple independent comes of penile carcinoma. Younger patients are often pathways of penile carcinogenesis. Am J Pathol. 2001;159(4):1211-1218. offered more penile-sparing approaches for primary 21. Lont AP, Kroon BK, Horenblas S, et al. Presence of high-risk human papillomavirus DNA in penile carcinoma predicts favorable outcome in survival. tumors but more aggressive treatment such as lymph- Int J Cancer. 2006;119(5):1078-1081. adenectomy for nodal disease. African American men, 22. Benard VB, Johnson CJ, Thompson TD, et al. Examining the association between socioeconomic status and potential human papillomavirus- Hispanics, the uninsured, those who are less educat- associated cancers. Cancer. 2008;113(10 suppl):2910-2918. ed, and those with lower socioeconomic status appear 23. Sichero L, Pierce Campbell CM, Fulp W, et al. High genital prevalence of cutaneous human papillomavirus DNA on male genital skin: the to have lower rates of survival when compared with HPV Infection in Men Study. BMC Infect Dis. 2014;14:677. their insured, educated, white, non-Hispanic counter- 24. Sharma P, Zargar-Shoshtari K, Spiess PE. Current surgical management of penile cancer. Curr Probl Cancer. 2015;39(3):147-157. parts. Marital status, number of sexual partners, and 25. Zhu Y, Gu CY, Ye DW. Population-based assessment of the number human papillomavirus status also seem to play a role of lymph nodes removed in the treatment of penile squamous cell carcinoma. Urol Int. 2014;92(2):186-193. in prognosis. Although the treatment options and rates 26. Zhu Y, Gu WJ, Wang HK, et al. Surgical treatment of primary disease of survival from penile squamous cell carcinoma have for penile squamous cell carcinoma: a Surveillance, Epidemiology, and End Results database analysis. Oncol Lett. 2015;10(1):85-92. improved during the past several decades, further un- 27. Rippentrop JM, Joslyn SA, Konety BR. Squamous cell carcinoma derstanding of the racial/ethnic, socioeconomic, and of the penis: evaluation of data from the surveillance, epidemiology, and end results program. Cancer. 2004;101(6):1357-1363. age-related differences seen in penile malignancies 28. Burt LM, Shrieve DC, Tward JD. Stage presentation, care patterns, may help minimize future disparities in cancer care. and treatment outcomes for squamous cell carcinoma of the penis. Int J Radiat Oncol Biol Phys. 2014;88(1):94-100. 29. Gopman JM, Djajadiningrat RS, Baumgarten AS, et al. Predicting References postoperative complications of inguinal lymph node dissection for penile cancer in an international multicentre cohort. BJU Int. 2015;116(2):196-201. 1. American Cancer Society. Cancer Facts and Figures 2016. http:// 30. Tyson MD, Etzioni DA, Wisenbaugh ES, et al. Anatomic site-specific www.cancer.org/acs/groups/content/@research/documents/document/ disparities in survival outcomes for penile squamous cell carcinoma. Urology. acspc-047079.pdf. Accessed October 4, 2016. 2012;79(4):804-808. 2. Bleeker MC, Heideman DA, Snijders PJ, et al. Penile cancer: epide- 31. Sharma P, Ashouri K, Zargar-Shoshtari K, et al. Racial and economic miology, pathogenesis and prevention. World J Urol. 2009;27(2):141-150. disparities in the treatment of penile squamous cell carcinoma: results from 3. Chaux A, Netto GJ, Rodriguez IM, et al. Epidemiologic profile, sexual the National Cancer Database. Urol Onc. 2016;34(3):122. history, pathologic features, and human papillomavirus status of 103 patients 32. Thuret R, Sun M, Budaus L, et al. A population-based analysis of with penile carcinoma. World J Urol. 2013;31(4):861-867. the effect of marital status on overall and cancer-specific mortality in patients 4. Stern RS, Bagheri S, Nichols K, et al. The persistent risk of genital with squamous cell carcinoma of the penis. Cancer Causes Control. tumors among men treated with psoralen plus ultraviolet A (PUVA) for psoriasis. 2013;24(1):71-79. J Am Acad Dermatol. 2002;47(1):33-39. 33. Thuret R, Sun M, Abdollah F, et al. Competing-risks analysis in patients 5. Tsen HF, Morgenstern H, Mack T, et al. Risk factors for penile with T1 squamous cell carcinoma of the penis. BJU Int. 2013;111(4 pt B):E174-E179. cancer: results of a population-based case-control study in Los Angeles 34. Paiva GR, de Oliveira Araújo IB, Athanazio DA, et al. Penile cancer: County (United States). Cancer Causes Control. 2001;12(3):267-277. impact of age at diagnosis on morphology and prognosis. Int Urol Nephrol. 6. Jayaratna IS, Mitra AP, Schwartz RL, et al. Clinicopathologic char- 2015;47(2):295-299. acteristics and outcomes of penile cancer treated at tertiary care centers 35. Verhoeven RH, Janssen-Heijnen ML, Saum KU, et al. Population- in the Western United States. Clin Genitourin Cancer. 2014;12(2):138-142. based survival of penile cancer patients in Europe and the United States of 7. Pow-Sang MR, Ferreira U, Pow-Sang JM, et al. Epidemiology and America: no improvement since 1990. Eur J Cancer. 2013;49(6):1414-1421. natural history of penile cancer. Urology. 2010;76(2 suppl 1):S2-S6. 36. Barnholtz-Sloan JS, Maldonado JL, Pow-Sang J, et al. Incidence 8. Hernandez BY, Barnholtz-Sloan J, German RR, et al. Burden of inva- trends in primary malignant penile cancer. Urol Oncol. 2007;25(5):361-367. sive squamous cell carcinoma of the penis in the United States, 1998-2003. 37. Arya M, Li R, Pegler K, et al. Long-term trends in incidence, survival Cancer. 2008;113(10 suppl):2883-2891. and mortality of primary penile cancer in England. Cancer Causes Control. 9. Goodman MT, Hernandez BY, Shvetsov YB. Demographic and 2013;24(12):2169-2176.

414 Cancer Control October 2016, Vol. 23, No. 4