PDF
Citation
Print
INTRODUCTION
Prostate cancer (PCa) is one of the most prevalent cancers among males, and its incidence is expected to increase with an aging population. Although the increased incidence of PCa in older males is a global phenomenon, its extent differs between the West and Asia.1 Two-thirds of patients with PCa detected in Japan are over 75 years old.2 In South Korea, one-third of patients are older than 75 years.3 In contrast, in the United States (US), where the male life expectancy is around 5 years shorter than that in Japan and South Korea, the proportion of patients aged 75 years or older is less than one-fifth.4 In contrast to the average age of 66 years at the time of diagnosis in the US, the average age has remained constant at 71 years in South Korea for the past 20 years.
One of the unique characteristics of PCa found in older males is that it is a more aggressive type of cancer than that found in younger males. Existing evidence suggests that older patients have a higher risk of biochemical recurrence, distant metastasis, and disease-specific mortality after diagnosis of PCa.5 At the time of diagnosis, more tumors with an advanced stage, a higher grade, and a larger volume have also been reported in patients aged over 70 years compared with their younger counterparts.67 Given that most cases of PCa are detected in the non-metastatic stage without specific symptoms, the serum prostate-specific antigen (PSA) level plays a decisive role in the decision to biopsy for diagnosis. Therefore, age may be a significant factor that should be considered to affect the aggressiveness of PCa, even when the PSA level is constant. In this study, we aimed to investigate the clinicopathologic pattern of PCa in elderly patients compared with their younger counterparts within similar PCa risk, especially those with low-risk disease (defined as PSA < 10 ng/mL, the level at which prostate biopsy typically starts to be recommended). If high-risk disease is more prevalent in older males, the cut-off for biopsy based on the PSA level should be lower considering that current guidelines recommend active surveillance in low-risk PCa.
METHODS
Selection of patients and data acquisition
A nationwide database of PCa at the time of biopsy from 2010 to 2020 was constructed from 39 hospitals across South Korea. Patients in the study met the inclusion criterion of having histologically confirmed PCa regardless of the biopsy method, including transrectal, transperineal, magnetic resonance imaging (MRI) fusion, and saturation biopsy, which were suggested based on the institutional policy during the study period. The number of cores taken at the time of the prostate biopsy and the number of positive cores were also investigated. To stratify the PCa risk, the serum PSA level before biopsy, clinical stage, and Gleason grade group (GGG) were collected. To confirm the clinical stage, the patients underwent radiological examinations, including computed tomography, MRI, and bone scan, all covered by the national health insurance after registering as a PCa patient. Given the reported differences in the frequency of PSA testing according to residence and the real-world differences in PCa screening rates, the type of residence (rural vs. urban) was also examined.8
Study design and statistical analysis
Patients were categorized into age groups of < 64 years, 65–69 years, 70–74 years, and ≥ 75 years, taking into account established guidelines that recommend PSA testing only for those aged 55–69 years during the study period, the average age of Korean PCa registrants of 70.3 years (2010–2019), and the average life expectancy of Korean males of 80.3 years (2020).
PCa risk was determined based on current guidelines (National Comprehensive Cancer Network, 2023 version 1, released on Jul 18, 20228), classifying patients into three categories: high-risk (PSA > 20 ng/mL, or GGG ≥ 4, or clinical stage ≥ T3a), intermediate-risk, and low-risk (PSA < 10 ng/mL, GGG = 1, and clinical stage ≤ T2a). Student’s t-test was used to compare continuous variables, and the χ2 test was used to compare binary and categorical variables. Trends in PCa risk change with age were analyzed by linear-by-linear association. Normality for the distribution of patients according to age was tested using skewness (between −2 and 2) and kurtosis (between −0.33 and 0.37) values. Univariate analysis was performed using a logistic regression model, which was also used to calibrate the multivariate variables. The analysis was performed using SPSS Statistics ver. 27.0 (IBM Corp., Armonk, NY, USA). Statistical significance was defined as P < 0.001.
RESULTS
Characteristics of study subjects
The characteristics of patients with PCa at the time of biopsy according to the age group are summarized in Table 1. The mean age ± standard deviation and median age (interquartile range) were 70.3 ± 8.2 years and 71.0 (65–76) years, respectively. Based on kurtosis (= 0.095), the age of the patients was normally distributed (Fig. 1A). The quantile-quantile plot also demonstrated normality as most of the data were distributed on a straight line (Fig. 1B). When the patients were divided into four age groups, the proportion in each group was as follows: 23.3% (< 64), 20.7% (65–69), 23.9% (70–74), and 32.0% (≥ 75).
Table 1
Clinicopathological characteristics of patients according to age group
Differences in PCa characteristics across age groups
Among 14,548 PCa subjects, 54.1% (n = 7,865), 39.5% (n = 5,747), and 6.4% (n = 936) of them had high-risk disease, intermediate-risk disease, and low-risk disease, respectively. The oldest age group had a higher risk of cancer across all three risk parameters, including PSA, clinical stage, and GGG (linear combination, P < 0.001, Fig. 2A-C). In risk classification based on a combination of the three parameters, the proportion of patients with a high risk was significantly increased in the oldest age group (Fig. 2D). Even when stratified according to PSA alone, an increase in the proportion of patients with high-risk cancer was observed in the oldest age group with < 10 and 10–20 ng/mL PSA (Fig. 3). Furthermore, the same pattern was observed among patients with serum PSA levels of 4–5 ng/mL (P < 0.001), 5–6 ng/mL (P < 0.001), 6–7 ng/mL (P < 0.001), 7–8 ng/mL (P < 0.001), 8–9 ng/mL (P = 0.009), and 9–10 ng/mL (P < 0.001), showing a significant increase in high-risk PCa in the oldest age group (Fig. 4). Even in the lowest tier (4–5 ng/mL), 35.2% of those aged over 75 years had high-risk PCa.
Fig. 2
Distribution of cancer risk according to risk parameters. (A) Risk stratification according to serum PSA. (B) Risk stratification according to cStage. (C) Risk stratification according to GGG. (D) Risk stratification according to overall risk parameters.
PSA = prostate-specific antigen, cStage = clinical stage, GGG = Gleason grade group.
Fig. 3
Risk stratification according to serum PSA. (A) Low risk (< 10 ng/mL). (B) Intermediate risk (10–20 ng/mL).
PSA = prostate-specific antigen.
In multivariate analysis adjusted for type of residence and baseline PSA level, there was a statistically significant increase in high-risk PCa in older males compared with males younger than 64 years (Table 2). In all PSA tiers above 4 ng/mL and below 10 ng/mL, there was a significantly higher risk of developing high-risk PCa in males aged 75 years or older compared with those in the youngest age group, even after accounting for the residential area (Table 3).
Table 2
Multivariate analysis of high-risk prostate cancer according to age and residential area
Table 3
Multivariate analysis of high-risk prostate cancer according to age, residential area, and PSA tier
DISCUSSION
In determining a treatment strategy for PCa, the aggressiveness of the disease is the most crucial factor to consider along with life expectancy. In most international guidelines proposed in 2023, active surveillance is recommended as the preferred strategy for low-risk cancer, and surgery or radiation therapy is recommended as the preferred strategy for intermediate-risk cancer with at least 10 years of life expectancy or higher-risk cancer with at least 5 years of life expectancy.91011 PSA level plays a key role in deciding whether to perform biopsy for most cases of PCa, which mainly involve localized lesions without metastases at the time of diagnosis. The recently updated American Urological Association (AUA) guidelines recommend full engagement in shared decision-making about the pros and cons of screening with PSA for PCa.12 The shift in the aggressiveness of PCa in the US from high-risk to low-risk in the early 2000s,13 coupled with an increase in PSA testing rates to nearly half of all males aged over 50 years, has prompted the new AUA guidelines to clarify that active surveillance should be prioritized over active treatment for low-risk cancer with a high risk of death before biopsy.
However, these guidelines are not directly applicable in South Korea, where life expectancy is longer than that in the US and more aggressive PCa is common.14 Indeed, in a national survey from 2010 to 2020, high-risk PCa accounted for half of all investigated cases, whereas low-risk PCa decreased from 9.2% in 2010 to 4.0% in 2020. Specifically, considering the unique demographic situation where life expectancy has continued to increase rather than decrease over the past 40 years in South Korea, the number of patients who can receive aggressive treatment may be high despite the detection of PCa at an average age of 71 years for the last two decades. Against this background, we investigated the clinicopathologic nature of cancer at the time of biopsy according to age, which has recently emerged as another factor influencing PCa aggressiveness in the literature.
This study investigated approximately 15,000 patients normally distributed according to age, allowing us to reliably determine whether PCa aggressiveness differed in each age group with similar PSA levels, and two interesting observations were made. First, as shown in Fig. 2, a significant age group-related increase in high-risk disease was observed for all PCa aggressiveness markers, including PSA. At the time of diagnosis by biopsy, 42% of high-risk cancers were diagnosed in the youngest age group (≤ 64 years); however, this percentage was increased to 68% in the oldest age group (≥ 75 years). In this age subgroup, the combined proportion of high-risk and intermediate-risk cancers was 96.9%. The relationship between age at diagnosis of PCa and the aggressiveness of the disease has not been consistently reported in previous studies. Older age at diagnosis has been demonstrated across multiple cohorts to be associated with higher risk of tumor upstaging and upgrading following prostatectomy, suggesting a tendency toward more aggressiveness.1516 Accumulated evidence indicates that males aged ≥ 70 years have a higher disease grade and stage and larger tumors.56 In a cohort of 4,561 males who underwent radical prostatectomy, a higher proportion of high-stage, aggressive-grade, and larger tumors was detected in those aged more than 70 years compared with younger age groups (younger than 60 and 60–70 years). Based on a nationwide database of PCa involving 12,081 males, 30% of subjects older than 70 years had a statistically significant higher clinical stage, biopsy grade, and prediagnosis PSA velocity (P < 0.0001). In addition, there have been reports of differences in gene expression associated with the development of PCa as a molecular biological mechanism in older people.17 However, contrary to these findings, there have been reports indicating that age is not a significant factor in the outcome of PCa.18 Moreover, some studies suggest that more aggressive cancers occur in younger males1920; however, these are relatively rare. Nevertheless, the increased incidence of PCa in older males is a consistent pattern observed worldwide; thus, the results of this study showing a marked increase in high-risk PCa with age may have clinical value.
This study also showed that the presence of high-risk cancer was positively associated with age, particularly in cases of PCa detected below the low-risk threshold of 10 ng/mL, a value that plays a crucial role in the decision to perform prostate biopsy. We divided PSA levels below 10 ng/mL into 4–5 ng/mL, 5–6 ng/mL, 6–7 ng/mL, 7–8 ng/mL, 8–9 ng/mL, and 9–10 ng/mL and investigated whether there were any differences across age groups within each tier. Even with PSA levels of 4–5 ng/mL (the lowest tier), 35% of males aged more than 75 years were found to have high-risk PCa. Moreover, after adjusting for PSA and type of residence, compared with an age younger than 64 years, an age of 75 years or older was an independent predictor of high-risk cancer. These observations suggest a clinical need for more aggressive biopsies in older males, especially those aged 75 years or older, as high-risk cancers are likely to be detected even with low PSA levels. Clinical guidelines for PCa may need to be fine-tuned for this population, which is expected to continue to grow in the future.
There are some limitations in this study. This study is based on an observational study in South Korea through the 2020s, where public awareness of PSA is low. As a result, PSA screening rates are estimated to be less than 10% among males aged 50 years or older, which may not be the same as the rates in other countries, where the epidemiology of PCa is thought to be different. The average age of detection is known to vary by around 10 years in other countries21; thus, the country's life expectancy should be considered when developing a diagnostic strategy for PCa. In addition, technical factors that may affect screening rates for PCa could not be investigated in this study given the large number of participating sites (39 nationwide) and the significant advances in the methodology of prostate biopsy over the 10-year study period. Despite the large number of patients accounting for 20.3% of enrolled patients in the national cancer registry during the study period and the use of statistical methods to divide patients into evenly sized groups with age adjustment, it is not clear whether older patients tend to develop high-risk PCa or whether older patients develop high-risk PCa due to delayed cancer detection. The diagnosis of PCa is a one-time event in an individual's life, which makes it challenging to study. Prospective studies across age groups with similar opportunities for PSA testing would be an appropriate approach for future research. Nevertheless, the information from this national, multi-institutional study could contribute to shaping strategies for the diagnosis of PCa in older males.
The older the patient, the more aggressive the PCa. Moreover, there was an increase in high-risk PCa in older males compared with younger males even with a PSA level below the low-risk threshold of 10 ng/mL, suggesting the need to strengthen cancer screening policies in the older population.
XML Download