Association of biological aging with prostate cancer: insights from the National Health and Nutrition Examination Survey

Prostate cancer (PCa) has become a rising cause of mortality, accounting for 15% of all male cancers. Global incidence of PCa is projected to rise from 1.4 million annual cases in 2020 to 2.9 million by 2040, with annual deaths anticipated to increase from 375,000 to nearly 700,000, an increase of about 85% [1]. This surge is expected to impose a significant burden on individuals and societies alike. Consequently, a refined classification scheme is urgently required to facilitate early and accurate identification of high-risk PCa, which would aid in developing more effective treatment paradigms.

Aging is a high-risk factor for PCa, with the incidence of the disease escalating with age. The mean age of affected patients ranges from 72 to 74 years, with about 85% diagnosed post 65 years [2]. A European study indicated that the prevalence of incidental PCa in males aged 30–69 was 30%, which rose to 75% in those over 70 years old [3]. However, a study assessing the trends in metastatic PCa in the U.S. population by age and race revealed that the annual growth rate of metastatic PCa in men under 69 years old was faster than in older men, with the highest projected growth rates in males aged 45–54 years old [4]. While chronological age plays a crucial role in predicting prostate cancer risk, its limitations in representing the true physiological and pathological disorders underlying aging cannot be ignored, possibly explaining the increasing risk among younger individuals. Hence, identifying indicators that more accurately reflect true human aging and exploring their association with PCa risk could better identify the residual risk independent of chronological age.

Biological aging, which refers to the aging processes in multiple biological systems [5], is a leading risk factor for most age-related diseases, physical and cognitive impairments, and mortality [6]. Increasing studies have linked biological aging closely with cancer [7–9]. Hallmarks of aging, such as genome instability, cellular senescence, and epigenetic changes, also overlap with the hallmarks of cancer [9]. Recent research has corroborated the association between aging and PCa in humans. However, biological aging captures the heterogeneity among older individuals [10]. Biological age, on the other hand, integrates information from biological markers and may better reflect an individual's physiological state and risks of age-related diseases and death [10]. Biological age calculated using the KDMAge and PhenoAge, based on a set of easily obtainable clinical measurements and blood test markers, provides a comprehensive assessment of a person's biological age [11]. Compared to chronological age, identifying the biological age of individuals can facilitate timely interventions to prevent the onset of diseases [12]. However, it remains to be determined whether measurements of biological age can serve as a surrogate for biological aging in predicting the incidence of PCa.

In this study, we explore the impact of biological age on PCa in a nationally representative population using cross-sectional data from the National Health and Nutrition Examination Surveys (NHANES). We also analyze the relationship between biological age acceleration and elevated levels of prostate-specific antigen (PSA), indicating a high risk in individuals without PCa. This research aims to provide new insights for early identification and intervention of aging-related prostate cancer risks in clinical settings.

Our study utilized nationally representative data from the NHANES survey to explore the relationships between two measures of biological age—KDMAge and PhenoAge—and their acceleration adjusted for chronological age, with the risk of PCa and highly-probable PCa as defined by integrated PSA levels. After accounting for confounding factors influencing the risk of PCa, we found that increased PhenoAge acceleration was associated with a higher risk of PCa in individuals younger than 65 years old. Notably, even among those under 65 years old without PCa, elevated PhenoAge acceleration was significantly associated with an increased risk of highly-probable PCa, indicating that biological aging acceleration signals an early increased risk of PCa. These findings provide new insights into the role of biological aging in cancer development and offer novel methods for identifying individuals at early risk of PCa during the aging process.

The aging population and extended life expectancy are leading to an increase in older men, a demographic at higher irreversible risk for various cancers [14, 15]. The development of PCa is also closely linked with increasing age [2, 16–18], with studies showing that newly diagnosed PCa in older individuals has more than tripled from 1990 to 2013 [19]. While prostate cancer diagnoses and related mortalities are rare in males younger than 50 years old, approximately 85% of PCa are diagnosed after the age of 65 years old [2]. Men older than 65 have a higher likelihood of adverse pathological findings at radical prostatectomy compared to their younger counterparts (adjusted OR, 1.28; 95% CI 1.00–1.62; P = 0.048) [20]. The mean age of patients with PCa ranges from 72–74 years, with most deaths occurring in men older than 75 years [2, 21]. Autopsy studies suggest that most men over 85 years old have histological PCa [2]. A Swedish population-based PCa screening trial demonstrated that age is linked with the risk of clinically significant PCa, with each one-year increase in age increasing the risk of being diagnosed with a higher Gleason score, indicative of more aggressive PCa [22]. Data from the population-based SEER database show that the 10-year risk of PCa increases from 2.3% at age 50 to 7.3% at age 70, with about 60% of diagnosed cases being men over 65 years old [16]. Furthermore, analysis of large prostate biopsy datasets reveals that older men have larger prostate volumes and increased risks of abnormal DRE and higher Gleason scores[16].

However, the clinical pattern relating chronological age with PCa has also changed noticeably over the past few years [23]. The proportion of men diagnosed before the age of 70 and the proportion of moderately differentiated tumors have both increased. The age at death from PCa peaks between 80–84 years old [24]. A study assessing trends in metastatic PCA in the American population showed that the annual growth rate of metastatic PCA in men under 69 was faster than in older men, with the highest projected growth rates in men aged 45–54 [4]. Moreover, national data from the SEER Program revealed that more men were diagnosed with PCa at a younger age and earlier stage in 2004–2005 compared to previous years, with the average age at diagnosis decreasing from 72.2 to 67.2 years [25]. These evidences suggest that chronological age alone cannot fully explain the increasing incidence of PCA in younger men, highlighting the limitations of chronological age in explaining the onset of PCa related to aging. The real physiological and pathological disorders underlying age progression are better explained by factors independent of chronological age, providing a more accurate identification of the residual risk of PCa.

Our study employed a set of common clinical physical examination and blood test markers to calculate two biological age indicators, KDMAge and PhenoAge, and their accelerated versions adjusted for actual age, better reflecting the degree of accelerated aging in individuals of the same age. By analyzing the correlation between biological age acceleration and the risk of PCa as well as early highly-probable PCa defined by PSA levels, we found significant associations in younger individuals under 65 years old, but not in older adults. In contemporary society, the increasing prevalence of metabolic abnormalities such as hypertension, diabetes, and obesity among younger people, coupled with more young individuals experiencing excessive mental stress and adopting poor lifestyle habits like staying up late, smoking, and excessive drinking, makes younger individuals more susceptible to biological aging. This also explains the increasing trend of younger individuals being diagnosed with PCa. More importantly, we found that in younger individuals without PCa, elevated biological aging acceleration was significantly associated with an increased early risk of PCa as defined by PSA levels, which can identify treatable early-stage PCa, significantly reducing disease-specific mortality and improving the detection of asymptomatic, well-differentiated PCa [26, 27].

Our results are strengthened by a nationally representative sample, which applied survey weights to ensure the generalizability of our findings to adult men in the USA, as well as obtaining comprehensive data on numerous essential covariates through the integration of NHANES data. Moreover, the strong correlation between chronological age and PCa was thoroughly considered by adjusting for age during the calculation of biological age acceleration, and a relatively detailed analysis was conducted in different age subgroups. Nevertheless, our study has some limitations. First, as a cross-sectional analysis, it inherently limits the ability to establish a definitive causal relationship. Second, we utilized self-reported physician-diagnosed cases of PCa, which may introduce recall bias due to the limitations of self-reported methods. Third, histories of diabetes and hypertension were identified through self-reported diagnoses instead of direct blood pressure or glucose measurements, which may lead to the omission of currently affected individuals and potential recall bias. Additionally, the lack of data on tumor stage and grade means we could not account for the severity of cancers. Finally, due to the limited number of study participants, we were unable to fully consider the relationship between biological age and PCa across different races. Therefore, caution is advised during the analysis and interpretation of the data. In the future, prospective studies involving tumor grading and different racial subgroups are essential to provide stronger evidence for our findings.

In conclusion, our findings provide new insights into the increasingly younger age of onset of PCa, linking accelerated biological aging with higher risks of PCa in younger men. Importantly, our study offers novel methods for identifying individuals at early risk of PCa during the aging process.

Below is the link to the electronic supplementary material.

Z.Y. and C.X. was responsible for developing the research concept and design. W.Y. analyzed the data and drafted the manuscript. B.S. and C.Y. contributed to data collection and analyses. C.X. revised the manuscript. Z.Y. and J.J. supervised the project, secured funding, managed the project's administration. All authors reviewed and agreed with the manuscript.

The research was supported by Medical Health Science and Technology Project of Zhejiang Province (No. 2020KY251), Zhejiang Province Traditional Chinene Medical Science and Technology Plan Project (No. 2020ZB215), and Ningbo Top Medical and Health Research Program (No. 2022020203).

All NHANES datasets used in this study are publicly available and listed by survey cycle at wwwn.cdc.gov/nchs/nhanes.

Zejun Yan, Email: fyyyanzejun@nbu.edu.cn.

Chengxin Xie, Email: chengxin_xie@163.com.