Prostate cancer is generally viewed as a disease of older men, however recent observations suggest that prostate cancer is increasing among older adolescents and young adults. Even as young as 17, and about 10% of new diagnoses occur in men age 55 or younger, and these early-onset cancers often have a worse prognosis.
Incidence, mortality, and survival data were obtained from the US National Cancer Institute Surveillance, Epidemiology, and End Results program and the Institute for Health Metrics and Evaluation Global Burden of Disease database.
Results
Worldwide, the incidence of prostate cancer has increased in all groups between ages 15 and 40 years and increased globally at a steady rate averaging 2% per year since 1990.
In the United States, this age group was 6 times more likely than older men to have distant disease at diagnosis. Stage for stage, their survival rate improved less than in older men. Whereas the overall 5-year relative survival rate in the United States for men diagnosed between ages 40 and 80 years was between 95% and 100%, it was 30% in those aged 15 to 24 years, 50% in those aged 20 to 29 years, and 80% in those aged 25 to 34 years.
Conclusions
Prostate cancer in older adolescent and young adult men has increased in most countries. There is some evidence that this may be caused in part by underdiagnosis, prostate-specific antigen screening, and overdiagnosis. It also may be caused by trends in obesity, physical inactivity, HPV infection, substance exposure, environmental carcinogens, and/or referral patterns. How the biology of these cancers differs from that in older men and how the etiologies vary from country to country remain to be determined.
Introduction
By using data from the US Surveillance, Epidemiology, and End Result (SEER) 17 regions, Salinas and colleagues reported an increasing incidence of prostate cancer in men ages 20 to 49 years, especially since 1991.
As we reported previously, and as recognized by investigators at the US Centers for Disease Control and Prevention (CDC),prostate cancer has increased in older adolescents and young adults (AYAs) (defined as ages 15-39 years, inclusive) to a level that merits evaluation, particularly of its epidemiology and biology and potentially different diagnostic and treatment strategies.
Materials and Methods
Incidence, stage, grade, mortality, and survival data were obtained from the Surveillance, Epidemiology, and End Results (SEER) program of the US National Cancer Institute using SEER*Stat version 8.3.5.
Incidence and mortality information for other countries and the world was obtained from the Institute for Health Metrics and Evaluation (IHME) Global Burden of Disease (GBD) public website. The average annual percent change (APC) in incidence was either obtained through SEER*Stat or calculated from incidence rates. Prostate carcinoma was defined as a prostate site and with International Classification of Diseases for Oncology.
Results: Incidence and Prevalence
From 2000 to 2015 in the United States, the youngest and next youngest ages of patients with prostrate carcinoma were 13 and 16 years, respectively, and the incidence rose rapidly among men aged 30 years. One-half of those diagnosed before age 25 years were unstaged, and 9% of those diagnosed between ages 25 and 40 years had distant metastases at the time of diagnosis compared with 4% in those diagnosed between ages 40 and 80 years. The proportion of AYA men who were not staged was inversely proportional to their age, from 50% in those aged 15 to 24 years to 10% in those aged 25 to 29 years, and to 6% in those aged 35 to 39 years, compared with 2% in men aged 40 to 80 years. It was distinctly most common among African Americans and was least common among Asians/Pacific Islanders.
Incidence and Mortality Trends
The incidence and mortality trends from 1975 through 2015 for the 5 most rapidly increasing invasive cancers in American men ages 25 to 39 years.
There was an 11-fold increase in the incidence of prostate carcinoma among American men ages 25 to 39 years during 1975 through 2010 and a decrease thereafter to a level in 2015 that was 6-fold higher than that in 1975.
Survival
Between 2000 and 2015, American AYAs diagnosed with invasive prostate cancer had a worse 5-year relative survival than older males, and their survival rate was directly proportional to age (Fig. 5A).12 Whereas the overall 5-year relative survival rate in the United States for men diagnosed between ages of 40 and 80 years was between 95% and 100%, it was 30% in men aged 15 to 24 years, 50% in those aged 20 to 29 years, and 80% in those aged 25 to 34 years.
Survival was worse in the United States for AYAs than for older men in all evaluable stages (stage I-II, localized/regional stage, and stages II-IV). The higher death rate among AYA men than among older men was because a greater proportion of AYAs had distant disease at diagnosis, and survival also was worse for those who had early stage disease at diagnosis.
The only races/ethnicities with a sufficient number of patients in the youngest age groups to allow an assessment of the 5-year mortality rate in AYAs were the non-Hispanic white and African American populations. Survival in both non-Hispanic whites and African Americans was also directly proportional to age, with a suggestion that the African American AYA population has fared worse than non-Hispanic AYAs.
Survival Trends
In terms of survival trends in the United States, men ages 30 to 39 years with prostate cancer have had progress in survival similar to that in older men. In males ages 15 to 29 years, however, there has been no evidence for survival improvement since the 1970s. The death rate from prostate cancer paralleled the incidence as a function of age until 80 years. The proportion of deaths relative to new cases was higher, up to 50% in men ages 20 to 24 years, than in any group between ages 40 and 80 years. The fewest deaths relative to new cases occurred among men aged 40 to 60 years.
Discussion
Our analyses of SEER and IHME GBD data present evidence that prostate cancer incidence has increased in AYA men in the United States, Canada, Europe, Australasia, and globally. A prior study found that the proportion of men who were ages 35 to 55 years at diagnosis increased from 2.3% between 1988 and 1991 to 9% between 2000 and 2003, and the median age at diagnosis decreased from 72 years in 1988 to 68 years in 2003. An incidence trend similar to ours among men diagnosed with prostate cancer between ages 25 and 40 years has been observed by the RARECARENet group of 42 registries in Europe. The review of autopsy studies is also informative, indicating that a progressive increase in the prevalence of prostate cancer was significant, but only among men of European descent, amounting to 0.3% per year since the 1930s.
We have found that, the younger AYA men are when diagnosed with prostatic carcinoma, the less likely they are to have been adequately staged. Those who have been adequately staged have a higher likelihood of presenting with distant metastases at diagnosis, more than twice that of older men. In the United States, 5-year relative survival after a diagnosis of prostate cancer has been inversely proportional to age up to age 65 years since 1985. In those diagnosed before age 30 years, survival is abysmal, with 5-year rates of 30% to 40% versus 90% to 98% in older men, and there has been little to no improvement since 1975. The results in Europe differ in that, among 34,185 men aged 15 to 44 years who were diagnosed with prostate cancer during 2000 through 2007 and followed through 2008 in 29 countries, the 5-year age-standardized relative survival rate was 81%, which was 9%, 9%, and 13% higher than in men aged 45 to 54, 55 to 64, and 65 to 74 years, respectively.15 How those aged 30 years have fared has not been reported. In contrast to the inverse relationship of more advanced stage with younger age at diagnosis, the increase in incidence among AYAs was almost entirely because of localized/regional disease.
Cause(s) of Increasing Incidence
We have considered several possible explanations for the increasing incidence, including improved recognition (underdiagnosis), screening practices, overdiagnosis, racial/ethnic shifts, societal factors, exogenous carcinogenic exposures, and increasing obesity among AYAs.
Improved recognition (underdiagnosis)
Because prostate cancer originally was rarely considered in a young male, it may not have been recognized previously and thus was underdiagnosed. The increased incidence could have resulted from the increased implementation of diagnostic methodologies, including those for genomic predisposition syndromes. Figure 1C suggests a degree of underdiagnosis in that the youngest men have the higher proportion of metastatic disease at diagnosis. There may have been delays because of underdiagnosis that allowed the cancer ultimately to present with signs and/or symptoms of metastases. That these patients also had the highest proportion of unstaged disease may have been because of underdiagnosis related to less comprehensive workup and understaging. The increasing capability to detect and diagnose prostate cancer in young men is consistent with the observation that nearly all of the increase was of earlier stage disease rather than late-stage disease with distant metastases.
Prostate-specific antigen screening
Screening for prostate cancer with elevated serum levels of prostate-specific antigen (PSA) has been used in men aged 50 years since at least 1992.16 PSA screening has not been recommended by any national medical organization for men in their 30s or 40s. It is worth noting that, in 2012, the US Preventive Service Task Force (USPSTF) on Prostate Cancer recommended that PSA screening be discontinued at all ages.17 There was a 28% decrease in incident diagnoses of prostate cancer in the year after the USPSTF draft recommendation against PSA screening. Since the 2012 recommendation, rates of PSA screening decreased by 3% to 10% in all age groups. The extent to which prostate cancer incidence declined among AYA men in the United States after 2012 suggests that PSA screening might have played a role in their earlier increase. This trend also supports the increase as real and not primarily because of underdiagnosis.
Recently, Shahangian and colleagues at the CDC reported that, from 2000 through 2015 in the United States, 2% of men aged 30 to 39 and 5% to 6% of those aged 40 to 49 years who had health insurance were “screened” with PSA, “contrary to all existing practice guidelines” and “despite all medical organizations recommending against PSA screening of men younger than 50 and older than 69.”20 “Screening” is not an accurate description of this small percentage because they likely had PSA levels obtained because of a family history of prostate cancer or were otherwise clinically suspicious and were not part of population screening. The CDC investigators found no significant temporal trends of PSA testing among men ages 30 to 39 years and a significant downward trend among men aged 40 to 49 years; whereas, in all older age groups, the rate of PSA testing declined proportional to increasing age (Fig. 7B).12, 20 The decline in older men is likely because of a USPSTF recommendation in 2008 against screening men aged 75 years, a recommendation endorsed by the American Academy of Family Physicians and the American College of Preventive Medicine. If this practice recommendation influenced PSA testing in younger men, it should have reduced testing in them too because it was never recommended for them, and the guidelines rendered this practice even more questionable. Also, effective screening increases the proportion of patients diagnosed with early disease, which is the opposite of the pattern by age in young men. Conversely, nearly all of the increase was in earlier stage disease rather than late-stage disease with distant metastases, which is consistent with a PSA screening effect.
Experience in other countries provides valuable comparisons. The Prostate Testing for Cancer and Treatment (ProtecT) study in the United Kingdom enrolled almost 95,000 men, of whom 10% had raised PSA levels. In that group, 12% of men had locally advanced or metastatic disease, and the remainder had clinically localized disease.23 Compared with incident cases in men of the same age (50-69 years), those in the ProtecT study had a younger age distribution, lower PSA levels, and cancers of lower stage and grade. Another study in the United Kingdom examined PSA testing by general practitioners (family physicians) in a retrospective cohort of 450,000 men over a period of 10 years.24 Greater than 25% of men aged 40 to 49 years had undergone testing. Those with an elevated PSA level were much less likely to have had a biopsy or a diagnosis of prostate cancer than men in the experimental arm of the ProtecT screening study. Moreover, a high proportion of the tests appear to have been performed for investigation of lower urinary tract symptoms and not as screening for prostate cancer.
In a randomized, population-based study in Gothenburg, Sweden, the mortality rate from prostate cancer was reduced substantially in the 10,000 men who underwent biennial PSA screening over 18 years, but there was no reduction among the 10,000 men in a control group who had “opportunistic” (random) PSA testing. Those authors acknowledged that the benefits of organized screening were offset somewhat by overdiagnosis of prostate cancer.
Because the proportion of men with distant metastases at diagnosis is inversely proportional to age at diagnosis , screening cannot explain all of the incidence trends in AYAs. Also, the magnitude of the decline is not correlated with the magnitude of the PSA screening decrease, with AYA men having had the least change in PSA screening and the greatest reduction in incidence.
Overdiagnosis
Another concern about prostate cancer diagnosis in AYAs is the possibility of overdiagnosis, as suggested by minimal evidence that prostate cancer mortality in young men in the United States increased during 1975 through 2015 despite the increase in incidence. There was some mortality rate increase during 1988 through 1998, however, when the increase was the greatest, indicating that the excess was real cancer. With the magnitude of the increase, however, a greater corresponding increase in the death rate would be expected, especially if the excess incidence included cancers with a worse survival, as indicated in young men by the survival data. In terms of total numbers of new cases and deaths, the GBD database recorded 2 deaths from prostate cancer in men before age 40 years in 2017 compared with 1005 new cases that year. The overdiagnosis explanation is also supported by the similarity of the incidence-versus-mortality trends of prostate carcinoma, thyroid cancer, and kidney cancer during 1990 through 2015, with the latter two both known to have been affected by overdiagnosis during those years.
Human papillomavirus
HPV infection has been associated with anogenital cancers in males and females and, as HPV prevalence has increased in young men, this vector conceivably also could affect the prostate.
Racial, ethnic, and familial factors
Racial, ethnic, familial, and genetic factors are thought to play a role in the early onset of prostate cancer and have implications for screening. The higher incidence of prostate cancer, notably of a more aggressive behavior among African American AYAs, is suggested also in prior published data. These age-incidence patterns suggest that, if routine population screening for prostate cancer is to be effective, it should not be done in AYAs and not until age 50 years in African Americans, age 55 years in non-Hispanic whites and Hispanics, and age 60 years in Asians and North American Natives.
Environmental carcinogens
Polychlorinate biphenyl and bisphonal A have been implicated as environmental carcinogens for prostate cancer, but it is unclear why younger men would have had a greater increasing exposure to them than middle-aged and older men. Nonetheless, the exposures of young men have changed over the era of the increased incidence, such as exposure to marijuana, which has been linked to testicular cancer, so that environmental carcinogens need to continue to be suspected as contributing factors. That race/ethnicity and cannabis have been linked to testicular cancer indicates how combinations of factors are likely involved once they are individually determined to be causative.
Obesity
Obesity has also been strongly associated with ovarian and breast cancer and with other types of cancer. For prostate cancer, obesity appears to increase the risk of more aggressive disease, which may be because of lower androgen levels, while simultaneously increasing the risk of more indolent disease. That young Americans have become increasingly obese over the same years as their prostate cancer increase suggests another contributing factor, although the decreases in both since 2010 through 2012 and the increasing incidence in countries that have not had an obesity epidemic counter this explanation.
Exercise
Physical activity in men with prostate cancer has increasingly been shown to improve survival and may prevent recurrence. Because youths in general have become less active, decreased exercise may also have contributed to the increased incidence and worse survival.
Societal factors
The increased prostate cancer incidence also could have been caused by greater referrals to urologic oncologists and other specialists, which also may have shifted during this period, allowing patients to seek these experts more easily and earlier.
Limitations
The limitations of our study include not knowing how much of the increased incidence was because of PSA screening. This metric was not available in the data sources for our analyses. We did not determine the extent to which the increasing Hispanic population in the United States may have contributed to the country’s overall increase, albeit this is unlikely because, in AYAs, prostate cancer incidence is lower among Hispanics than among African Americans and non-Hispanic whites. Another limitation is that we did not assess the type of therapy as a possible contributor to the lower survival rate in the AYA population. We doubt that this is a major factor because AYAs are generally treated more aggressively, given their longer life expectancy, and because of a suspicion that prostate cancer detected at a young age might behave more aggressively. In addition, AYAs have a greater tolerance of therapy than older patients. Our correlations made on cancer stage and grade should be taken with caution because data on these factors are usually less precise than data from cancer registries. Limitations of data on the cases we accessed also preclude understanding completely how these patients presented. We also do not know the extent to which men may have been misdiagnosed, which may be a greater problem in young men than in the classic situation of older men.
Prognosis and Biology
It is unclear how early age at diagnosis adversely influences the outcome in young patients with prostate cancer. Although most men diagnosed with prostate cancer before age 40 years have localized disease at diagnosis, they have a lower proportion of high-grade tumors and yet are more likely than older men to be diagnosed with metastatic disease.The tumors in young men are usually undifferentiated, metastasize early, have lytic rather than sclerotic bone metastases, and respond poorly to hormonal therapies.
The biology of early age prostate cancer is not well understood. A genomic analysis of 11 early-onset cases revealed a key role of the androgen-androgen receptor axis, suggesting a specific pathogenesis for young-age prostate cancers distinct from classical elderly onset tumors. Young-age prostate cancer is associated with an increased risk of the disease in family members.
Several susceptibility genes, such as BRCA2 and HOXB13, have been identified.
Longer survival for patients with cancer predisposition syndromes who have children increases the incidence of prostate cancer in young men. Whatever the biologic revelations, these are likely quite different in AYAs, as suggested by clear differences in (and, in some instances, reverses of) incidence, incidence trends, stage, survival, and survival trends in AYAs compared with older men.
Treatment
In young patients with clinically organ-confined disease, radical prostatectomy is offered increasingly over the alternatives based on radiotherapy (radical external-beam radiotherapy and brachytherapy). In view of the comments above, surgery may be preferred to radiotherapy in the very young patient being treated with curative intent. Nerve-sparing techniques to preserve continence and sexual function are possible without sacrificing outcome if the tumor is very small. Robotic surgery may need more consideration in AYAs, although they tolerate open surgical procedures better than older men. Conversely, absence from school or work could be shortened with robotic intervention.
Conclusion:
Men as young as 17 years are experiencing an increasing incidence of carcinoma of the prostate in much of the world. They frequently are staged less adequately, present with more advanced cancer, and have worse survival than middle-aged and older men. The cause of the increasing incidence is unknown. Potential contributing factors that should be investigated are underdiagnosis, PSA utilization, overdiagnosis, obesity, decreased physical activity, HPV infection, substance or environmental exposure trends, and referral patterns. How the biology of these cancers in older AYA men differs from that in older men also remains to be determined.
Here’s a link to the full article with charts and graphs:
Authors: Dr. Archie Bleyer, Dr. Filippo Spreafico, Dr. Ronald Barr
First published: 25 September 2019 https://doi.org/10.1002/cncr.32498Citations: 81