Background

Prostate cancer is the most frequently diagnosed cancer among men worldwide and predominantly affects older adults. Prognosis is mainly based on tumor characteristics at diagnosis (TNM stage, Gleason score), but these factors alone do not adequately predict survival, particularly in older patients with competing comorbidities and frailty. Inflammation-related biomarkers, including C-reactive protein, albumin, neutrophil, lymphocytes, platelets and their composite inflammation scores, have been associated with poorer survival in several cancer types, such as lung, colorectal, breast, and gastrointestinal cancers. However, for prostate cancer, evidence remains scarce and inconsistent, especially in older patients. 

Objectives

This study aims to investigate the prognostic role of inflammation-related biomarkers in older patients with prostate cancer for all-cause mortality.

Methods

We used data from the ELCAPA (ELderly CAncer PAtient) cohort, a prospective, multicenter open cohort including patients aged 70 years and older with solid or hematologic tumors referred for a geriatric assessment prior to cancer treatment in Greater Paris area, France. For the present study, men with prostate cancer were included between 2007 and 2023. Baseline data comprised socio-demographic characteristics, body mass index (BMI), comorbidities, frailty indicators and metastatic status. Cellular inflammation indices (platelet-to-lymphocyte ratio [PLR], neutrophil-to-lymphocyte ratio [NLR], and systemic immune-inflammation index [SII]) and molecular markers (C-reactive protein, and the C-reactive protein/albumin ratio [CAR]) were analyzed as continuous variables and standardized (z-scores) to estimate hazard ratios per 1 standard deviation increase. PLR and NLR were additionally analyzed as categorical variables using predefined literature-based cut-offs (PLR: <150, 150–300, >300; NLR: <3, ≥3). The Glasgow prognostic score (GPS), which is a cumulative inflammation-based cancer-prognostic marker composed of serum elevation of CRP and decrease in albumin concentration, was analyzed as a categorical variable (0,1,2). Associations with mortality were assessed using Cox proportional hazards models adjusted for age, BMI, metastatic status, and comorbidity burden.

Results

Among the ELCAPA cohort, 618 patients with prostate cancer were included (median age: 81 years [IQR: 77–86]). At baseline, 238 patients (75.8%) had a Gleason score ≥7 and 406 patients (65.7%) presented with metastatic disease, mainly bone metastases (N = 355; 57.4%). The median follow-up was 7.4 years (IQR: 2.7–11.7), and 5-year all-cause mortality reached 70%. Higher levels of PLR, NLR, SII, and GPS = 2 were independently associated with overall mortality after adjustment (adjusted HRs for 1 SD increase ranging from 1.25 to 1.31; all p<0.05; adjusted HR of 2.16 [1.52–3.06] for GPS = 2 versus 0). For CAR, associations were stronger in non-metastatic patients (adjusted HR = 4.56 [2.07–10.02]) compared with metastatic patients (adjusted HR = 1.15 [1.03–1.28]). Similarly, for CRP, the association was stronger in non-metastatic patients (adjusted HR = 2.48 [1.54–4.00]) than in metastatic patients (adjusted HR = 1.09 [0.99–1.20]).  

Conclusions / Implications

In our cohort, inflammation-related biomarkers were independently associated with mortality in older patients with prostate cancer,after adjustment for age, comorbidity burden, and metastatic status. These markers may improve prognostic stratification beyond conventional clinical characteristics and support more personalized decision-making and follow-up.