Advanced stage at diagnosis and aggressive histology dominate prostate cancer presentation in Kenya: a registry analysis

Highlights

• 85% of men in this Kenyan tertiary-care cohort presented with locally advanced or metastatic prostate cancer at diagnosis.
• PSA levels over 50 ng/mL were strongly associated with advanced-stage disease, with 88% in this category presenting with advanced disease.
• High-grade tumours (ISUP grade group 4 and 5) constituted the majority of advanced-stage cases at 54.7%.
• These findings provide an empirical baseline for developing risk-stratified diagnostic and referral pathways in Kenya.

1 Introduction

Prostate cancer is the most frequently diagnosed malignancy among men in Kenya and represents a major contributor to cancer related morbidity and mortality across sub-Saharan Africa (1–3). Although global advances in early detection and treatment have led to improved outcomes in many high-income settings, men in African countries continue to experience disproportionately poor prostate cancer outcomes, largely driven by late-stage presentation and limited access to diagnostic and treatment services (3, 4). Across sub-Saharan Africa, the majority of men with prostate cancer present with locally advanced or metastatic disease at diagnosis. This pattern reflects a combination of delayed symptom recognition, low uptake of prostate cancer screening, and structural barriers within health systems, including limited access to urological care, specialist pathology services, and timely diagnostic imaging (4–6). In Kenya, these challenges are particularly pronounced, with prior registry based and hospital studies consistently reporting a high burden of advanced disease at first presentation (2, 3). As a result, opportunities for curative treatment are frequently missed, contributing to persistently high prostate cancer mortality in the region.

Beyond health system constraints, emerging evidence suggests that biological and genetic factors may also contribute to the aggressive clinicopathological profile of prostate cancer observed among men of African ancestry. Studies examining genetic susceptibility and tumour biology have reported ancestry-associated differences in tumour behavior, disease aggressiveness, and PSA-related risk stratification (7, 8). These findings underscore the likelihood that the advanced disease patterns observed in African populations reflect not only delayed diagnosis but also underlying biological heterogeneity.

Clinical decision making in prostate cancer relies heavily on three interrelated diagnostic markers: PSA level, tumour stage, and ISUP grade goup. PSA remains widely used for prostate cancer detection and risk stratification, and although its performance is imperfect, markedly elevated PSA levels are frequently associated with advanced or aggressive disease in African settings (9–11). ISUP grade group s, the principal histopathological grading system for prostate cancer, is a well-established predictor of prognosis, treatment response, and survival, with higher scores reflecting more aggressive tumour behavior (12). Studies from African populations have consistently reported a high prevalence of poorly differentiated tumours, including ISUP grade groups of 4 and 5, at the time of diagnosis, reinforcing the substantial burden of aggressive disease in the region (13).

Despite the central role of PSA and ISUP grade groups in prostate cancer management, limited data are available describing how these markers relate to tumour stage at diagnosis among Kenyan men. In resource limited settings, where advanced disease remains common and access to comprehensive staging investigations may be constrained, understanding the relationship between biochemical, histological, and clinical indicators of disease extent is particularly important. Such information may help inform pragmatic diagnostic pathways and improve risk stratification at the point of presentation.

This study therefore aimed to describe the distribution of PSA categories and ISUP grade groups at diagnosis and to examine their association with tumour stage among Kenyan men with prostate cancer enrolled in two prospective clinical registries. By characterizing these relationships in a contemporary cohort, we sought to provide context specific evidence to inform clinical practice and highlight opportunities for earlier detection and improved diagnostic pathways in Kenya.

2 Materials and methods

2.1 Study design and setting

We conducted a cross-sectional analysis using baseline data from two prospective registries: IRONMAN and ACGR. These registries collect harmonized clinical, demographic, and pathological data on men with prostate cancer across participating centers. For the present analysis, we included men recruited from two tertiary hospitals in Nairobi: Kenyatta National Hospital, a national public referral center, and MP Shah Hospital, a private tertiary facility.

2.2 Study population

Eligible participants were men aged 40 years or older with histologically confirmed prostate adenocarcinoma enrolled in the IRONMAN and ACGR registries. The sample size was calculated using the Cochran's formula the minimum required number of patients was 162 based on a 95% confidence interval (z = 1.96), a 5% margin of error and an estimated population proportion of 0.88, representing the patients presenting with advanced disease in Kenya. The IRONMAN registry constituted patients with locally advanced and metastatic prostate cancer, while the ACGR registry enrolled all patients with prostate cancer irrespective of stage. Patients included in this study from both registries were enrolled once to avoid potential overlap and duplication.

2.3 Variables

Tumour stage was derived from clinical records and categorized as localized or advanced disease. Patients with T1-2, N0 M0 were classified as localized while T3-4 irrespective of the N and M stage were classified as advanced. The data on the stages and histology of the prostate cancer was collected using the same clinical data format therefore this ensured harmonization of the staging information. PSA level at diagnosis was grouped into clinically meaningful categories based on observed distributions and was data driven: <=10, 11–20, 21–50, >50. Gleason scores were obtained from diagnostic pathology reports and summarized according to individual WHO/ISUP Grade Groups for descriptive analysis.

2.4 Statistical analysis

PSA values were available for 152 participants, ISUP grade groups grades for 168 participants and stage data were available for all 170 men and included in the stage-stratified analysis. Descriptive analyses and cross-tabulations were used to characterize the distribution and associations of these markers across localized vs. advanced disease. PSA categories and WHO/ISUP group grades were cross tabulated against tumour stage to characterize their distribution at diagnosis. The association between PSA level at diagnosis and disease stage was assessed using the chi-square test, with statistical significance defined as a two-sided p-value <0.05All analyses were conducted using Stata version 15.1 (Stata Corp, College Station, TX, USA).

2.5 Ethical considerations

Ethical approval was obtained from the Kenyatta National Hospital-University of Nairobi Ethics and Research Committee, with regulatory clearance from the National Commission for Science, Technology and Innovation (NACOSTI). All participants provided written informed consent at enrolment.

3 Results

3.1 Demographic characteristics

Majority of the patients were 70 years and older and were recruited from the public hospital at 85.9% (n = 146). 85% (n = 144) were diagnosed with advanced disease while majority had PSA >50 ng/mL at 54% (n = 92). WHO/ISUP grade group 5 dominated the histological patterns at 49.4% Table 1.

Table 1. Demographic characteristics of participants.

3.2 PSA levels and tumour stage

PSA levels at diagnosis varied widely, and a clear, progressive gradient was evident across tumour stages. Men with localized disease were predominantly concentrated within the lower PSA ranges, particularly below 20 ng/mL. As PSA levels increased, the distribution shifted progressively toward advanced-stage presentation. Men with PSA values between 21 and 50 ng/mL were already more likely to present with advanced disease than with localized disease. This trend became even more pronounced at higher PSA thresholds. Among men with PSA values over 50 ng/mL, 81 out of 92 (88%) had advanced disease at diagnosis (Figure 1). After adjustment for age, higher PSA categories were associated with greater odds of advanced-stage disease. Compared with men with PSA ≤10 ng/mL, those with PSA levels of 11–20 ng/mL had significantly higher odds of advanced-stage disease (AOR=9.61, 95% CI: 1.61–57.26; p = 0.013). Similarly, PSA levels >50 ng/mL were associated with significantly higher odds of advanced-stage disease (AOR=9.26, 95% CI: 2.12–40.34; p = 0.003). Although men with PSA levels of 21–50 ng/mL also had higher odds of advanced-stage disease, the association did not reach statistical significance (AOR=3.99, 95% CI: 0.79–19.97; p = 0.093). In a separate analysis treating PSA category as an ordinal variable, each increase in PSA category was associated with 59% higher odds of advanced-stage disease after adjustment for age (AOR=1.59, 95% CI: 1.05–2.41; p = 0.029). Age was not independently associated with advanced-stage disease (AOR=1.00, 95% CI: 0.94–1.06; p = 0.956). The wide confidence intervals around the category-specific estimates indicate considerable uncertainty and warrant cautious interpretationOverall, these findings support an association between higher PSA levels and greater likelihood of advanced-stage disease at diagnosis, although the category-specific estimates did not demonstrate a consistent monotonic increase (Figure 2 and Table 2).

Table 2. Multivariable logistic regression analysis of PSA category and age in relation to advanced-stage prostate cancer.

OR, odds ratio; CI, confidence interval. Neither PSA category 21–50 ng/mL nor age was significantly associated with advanced-stage prostate cancer after mutual adjustment.

3.3 ISUP grade group patterns and tumour stage and PSA

ISUP grade groups demonstrated a strong association with tumour stage at diagnosis. Among the 168 men with available grade groups and stage data, 25 (15%) had localized disease, while 143 (85 %) had advanced disease. Lower-grade tumours were more frequently observed among men with localized disease, whereas higher-grade tumours predominated among men with advanced-stage cancer. Notably, the ISUP grade 3 pattern was strongly associated with advanced disease, with 28 of 34 cases (82.4%) occurring among men with advanced-stage tumours. High-grade tumours (WHO/ISUP grade group 5) accounted for a substantial proportion of advanced-stage presentations, underscoring the close relationship between tumour grade and disease extent at diagnosis in this cohort (Figure 1). Advanced disease predominated across all grade groups, ranging from 82.4% among patients with WHO/ISUP grade 3 disease to 86.9% among those with Grade 5 disease. There was no significant association between ISUP grade group and disease stage at diagnosis [Pearson χ²(1) = 0.39, p = 0.53]. In the multivariable logistic regression analysis, advanced-stage disease was associated with 31% higher adjusted odds of having a higher grade group; however, this association was not statistically significant (OR= 1.31; 95% CI: 0.55–3.14; p = 0.543). After adjustment for age, each one-category increase in PSA (≤ 10, →11–20, →21→50,→ ≥ 50) was associated with 44% higher odds of having a higher ISUP grade (OR=1.44; 95% CI: 1.02–2.04; p = 0.040). Compared with men with PSA ≤ 10 ng/mL, PSA 11–20 ng/mL had 1.73 times the odds of high ISUP grade group (OR=1.73, 95% CI:0.37–7.99, p = 0.483). PSA 21–50 ng/mL had 1.6 times the odds of high-grade group (OR=1.64, 95% CI: 0.35–7.63, p = 0.531). PSA > 50 ng/mL had 2.3 times the odds of high-grade group (OR=3.23, 95% CI: 0.79–13.09, p = 0.101). Age was not significantly associated with the grade group category (AOR 1.01; 95% CI 0.97–1.06; p = 0.682) Table 3.

Table 3. Multivariable logistic regression of factors associated with higher wHO/ISUP grade group.

OR, odds ratio; CI, confidence interval. PSA category was treated as an ordinal variable. Increasing values of the outcome represented higher ISUP grades. Neither PSA category nor age was significantly associated with higher ISUP grade after mutual adjustment.

3.4 Overall pattern

Taken together, the PSA and ISUP grades distributions reveal a consistent clinicopathological profile in this cohort. Men diagnosed with localized disease represented a minority and were largely characterized by lower PSA values and lower-grade tumours. In contrast, advanced-stage disease predominated and was associated with markedly elevated PSA levels and a higher-grade histological pattern. These parallel gradients underscore the close relationship between tumour biology and disease extent at diagnosis.

4 Discussion

Prostate cancer in sub-Saharan Africa is frequently characterized by late presentation, markedly high PSA levels, and aggressive histological features, and the findings of this study align closely with these regional patterns. In this Kenyan cohort, both PSA level and ISUP grade groups demonstrated strong and parallel associations with tumour stage at diagnosis, highlighting the combined influence of delayed presentation and adverse tumour biology. In this tertiary-care cohort, only a small minority of men were diagnosed with localized, potentially curable disease, reflecting the substantial burden of advanced-stage presentation among men who ultimately reach specialist cancer care. Similar patterns have been reported across African settings. Population-based data from Nairobi demonstrate a rising incidence of prostate cancer with substantial late-stage burden (2), while regional reviews have consistently described advanced presentation as the dominant clinical phenotype in sub-Saharan Africa (3). Systematic estimates across the continent further suggest that prostate cancer incidence is increasing in parallel with limited early detection infrastructure (6). Studies from East and West Africa also attribute late diagnosis to low screening uptake, limited awareness, and prolonged referral pathways (4), alongside emerging evidence of genetic susceptibility and aggressive tumour biology among men of African descent (5). Together, these regional data may suggest that the advanced-stage predominance observed in this cohort reflects both structural health system constraints and underlying biological vulnerability within African populations. An additional observation from this cohort was that advanced-stage disease predominated across all age categories, including men younger than 60 years. Although prostate cancer was most frequently diagnosed among men aged 70 years and older, the consistent predominance of advanced disease across age groups suggests that delayed diagnosis is not restricted to a particular demographic. Rather, it likely reflects broader challenges within the prostate cancer care pathway, including delayed health-seeking behavior, limited access to early diagnostic evaluation, and barriers to timely referral and specialist care which should be a subject of further research in this Kenyan population.

The predominance of advanced-stage disease observed in this cohort is broadly consistent with reports from other African settings, although the proportion identified in our study appears particularly high. Multi-country analyses have documented advanced-stage presentation among approximately 78%–85% of men with prostate cancer across sub-Saharan Africa, while recent Kenyan data reported advanced disease in 84.7% of patients at diagnosis (13, 14). The higher burden observed in this tertiary-care cohort may reflect referral bias, prolonged diagnostic pathways, differences in health-seeking behavior, or underlying biological heterogeneity. Notably, our finding that all patients with ISUP Grade Group 4 disease presented with advanced-stage cancer aligns with reports from other African populations describing a predominance of high-grade tumours among men diagnosed at later stages of disease (10, 13).

In Kenya, fear of diagnosis, stigma, misattribution of symptoms, cultural and religious beliefs, prolonged referral pathways, limited access to urological services, and out-of-pocket diagnostic costs likely compound these challenges, contributing to the high proportion of men presenting with advanced disease. This is supported by a recent qualitative study involving the same cohort of patients that revealed the reasons for late presentation and diagnosis were multifaceted The findings of the qualitative study suggested that late presentation may not simply reflect patient-level delays but may also indicate inefficiencies within diagnostic and referral systems (14).

A central, clinically actionable finding of this analysis is that the strong association observed between PSA level and tumour stage supports the continued clinical utility of PSA as clinical marker in resource-limited settings. Similar relationships between elevated PSA levels and advanced or high-grade disease have been reported across African populations, where men frequently present with substantially higher PSA values than those observed in high-income settings (9–11). In this context, PSA may provide important clinical information to support timely referral, staging, and treatment decision-making, particularly for men presenting with symptomatic disease (15). The observed association between markedly elevated PSA levels and advanced-stage disease suggests that PSA values, particularly those above 50–100 ng/mL, may have potential utility as a pragmatic indicator for prioritizing expedited urological referral and staging at primary and secondary care levels, especially where specialist evaluation is not immediately accessible. However, given the cross-sectional nature of this study, these findings do not establish a referral threshold, and prospective studies would be required to validate the clinical utility of such PSA-based triage.

The aggressive disease profile observed in this cohort may reflect the combined effects of delayed diagnosis and underlying tumour biology. Studies among men of African ancestry have identified ancestry-associated differences in tumour biology, PSA-related risk stratification, and genetic susceptibility that may contribute to higher-grade disease at diagnosis (7, 8). This relationship cannot be concluded from this study due to the a absence of molecular and genetic data. Consistent with multi-country African reports, high-grade tumours, including WHO/ISUP grade groups 4 and 5 disease, were common in our cohort and predominated among men with advanced-stage cancer, while lower-grade tumours were more frequent in localized disease. The predominance of WHO/ISUP grade 3 over group 2 disease among men with advanced cancer is also clinically relevant, given its association with less favourable outcomes. Together, these findings suggest that delayed detection and aggressive tumour biology may jointly contribute to the adverse clinicopathological presentation of prostate cancer in this population (13).

Overall, the combination of markedly elevated PSA, high-grade histology, and advanced-stage presentation highlights the need for earlier diagnosis and more efficient referral pathways in Kenya. PSA provides an accessible means of identifying men who require expedited evaluation, while the high burden of aggressive disease underscores the need for timely staging and access to appropriate treatment. Strengthening early detection, public awareness, and context-appropriate referral pathways may therefore help reduce advanced-stage presentation and improve prostate cancer outcomes.

Strengths of this study include the use of harmonized registry data from two prospective prostate cancer registries and the inclusion of patients from both public and private tertiary institutions, providing insight into a diverse population of Kenyan men with prostate cancer. The study also offers one of the few contemporary descriptions of the relationship between PSA level, ISUP grade group, and tumour stage in a Kenyan setting, addressing an important evidence gap in sub-Saharan Africa.

The study has two main limitations. First, recruitment from the ACGR and IRONMAN registries and other hospital patients was primarily conducted at tertiary referral hospitals, which may preferentially capture patients with more advanced or complex disease and limit the generalizability of the findings to the broader Kenyan population. Second, the cross-sectional study design limits assessment of temporal relationships and precludes causal inference.

Overall, the combination of markedly elevated PSA, high-grade histology, and advanced-stage presentation highlights the need for earlier diagnosis and more efficient referral pathways in Kenya. PSA provides an accessible means of identifying men who require expedited evaluation, while the high burden of aggressive disease underscores the need for timely staging and access to appropriate treatment. Strengthening early detection, public awareness, and context-appropriate referral pathways may therefore help reduce advanced-stage presentation and improve prostate cancer outcomes. Future research should explore the development and validation of pragmatic, data-driven approaches to prostate cancer risk assessment, referral, and early detection, and evaluate their feasibility and effectiveness within Kenyan and other sub-Saharan African healthcare settings.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Edvīns Miklaševičs, Riga Stradiņš University, Latvia

Reviewed by: Ibe Michael Usman, University of Rwanda, Rwanda

Rispah Torrorey-Sawe, Moi University, Kenya

Abbreviations ACGR, African cancer genomic registry; IRONMAN, international registry for men with advanced prostate cancer; ISUP, international society of urological pathology; KNH, Kenyatta national hospital; NACOSTI, national commission for science, technology and innovation; NIH, national institutes of health; PCCTC, prostate cancer clinical trial consortium; PSA, prostate specific antigen; TNM, tumour, node, metastasis.