Non‐Cancer Causes of Death in Patients With Metastatic Hormone‐Sensitive Prostate Cancer: A Real‐World Analysis Authors: Marta Garcia de Herreros, Clara Rodrigo, Laura Ferrer‐Mileo, Laia Fernández‐Mañas, Joel Mases, Nely Diaz, Antoni Vilaseca, Mariana Altamirano, Roser Gibanel, Maria José Ribal, Natalia Jimenez, Oscar Reig, Mariona Figols, Miguel Ángel Climent, Albert Font, Begoña Mellado Published: 2026 ABSTRACT ABSTRACTBackgroundAs overall survival improves in metastatic hormone‐sensitive prostate cancer (mHSPC), non‐cancer causes of death are increasingly relevant.MethodsWe conducted a real‐world multicenter study of patients diagnosed with mHSPC treated with ADT alone or in combination. Causes of death before progression to castration‐resistant prostate cancer (CRPC) were classified as cancer‐ or non‐cancer‐related. Deaths were classified as non‐cancer‐related if they occurred in the absence of cancer progression.ResultsAmong 565 patients, 35 (6.2%) died from non‐cancer‐related causes without disease progression, mainly infections (34%) and cardiovascular events (20%).ConclusionsNon‐cancer mortality represents a meaningful proportion of deaths in mHSPC, supporting systematic cardiovascular risk evaluation and infection‐prevention strategies. Introduction As therapeutic intensification with docetaxel and androgen receptor pathway inhibitors (ARPIs) improves metastatic hormone‐sensitive prostate cancer (mHSPC) outcomes [1, 2], a growing proportion of deaths in this population arise from non‐cancer causes [3, 4]. Men with mHSPC usually present at a median age of around 70 years and frequently have comorbidities such as cardiovascular disease, chronic lung disease, metabolic syndromes, or chronic infections [5]. Androgen deprivation therapy (ADT) and ARPIs may exacerbate several risk factors (weight gain, dyslipidemia, hypertension, and insulin resistance) potentially increasing cardiovascular risk [6]. Besides, they induce bone loss and reduced muscle mass, which may also worsen frailty, reduce mobility, and contribute to vulnerability to infectious complications [7]. Population‐based studies suggest that 20%–40% of deaths among men with advanced prostate cancer are unrelated to malignancy, yet contemporary real‐world data specifically describing non‐cancer mortality in the mHSPC setting remain scarce. Understanding these patterns is increasingly important as more patients receive prolonged, intensified systemic therapy, underscoring the need for proactive prevention strategies. We conducted an ambispective study characterizing non‐cancer deaths in mHSPC, aiming to identify their frequency, timing, and major contributing factors. Methods All patients diagnosed with mHSPC between January 2010 and January 2025 were included in an ongoing multicenter biomarker study. Here, we report the mortality data. Eligible patients received one of the following first‐line treatments: ADT alone, ADT plus docetaxel, ADT plus an ARPI, or triplet therapy (ADT plus docetaxel and an ARPI). Data extracted from electronic medical records included demographic information, comorbidities, functional status (ECOG performance status and Barthel Index), treatment details, and date and cause of death. Functional status was retrospectively collected from patients' electronic medical records, from the most recent clinical visit prior to initiation of first‐line treatment, when available. Comorbidities were categorized into six predefined groups: cardiovascular, respiratory, neuropsychiatric, rheumatologic, renal, and others. Causes of death were categorized as prostate cancer‐related or non‐cancer‐related. Non‐cancer‐related deaths were defined as those occurring during treatment for mHSPC without progression to castration‐resistant prostate cancer (CRPC) and were subclassified as infectious, cardiovascular, unknown, or miscellaneous. Overall survival (OS) was calculated from the date of mHSPC diagnosis to death, and CRPC‐free survival (CRPC‐FS) was defined as the time from mHSPC diagnosis to the development of CRPC, based on either radiological or PSA progression, whichever occurred first. Kaplan–Meier methodology was used for OS and CRPC‐FS estimates. Univariate and multivariate analyses of variables of interest were performed by a Cox regression analysis. Kruskal–Wallis tests compared continuous variables across cause‐of‐death categories; Fisher's exact test compared categorical variables. Results Of 565 patients with mHSPC, 93 (16.5%) received ADT alone, 125 (22.1%) ADT + docetaxel, 321 (56.8%) ADT + ARPI, and 26 (4.6%) triplet therapy. Median follow‐up was 44.6 months (range 4.5–228.9). Median CRPC‐FS and OS for the global cohort were 26.2 (95% CI 20.8–30.2) and 51.8 months (95% CI 44.9–61.3), respectively. Overall, 381 patients (67.4%) died. Of these, 35 deaths were attributed to non‐prostate cancer, accounting for 6.2% of all patients and 9.2% of all deaths, and all occurred before progression to CRPC (Figure S1). Among these patients, the median age was 77 years (range 51–92). Notably, 68,6% had metachronous disease, 65.7% low volume disease, and 8.6% had visceral metastases. Over half of the patients (51.4%) had ≥ 3 comorbidities, predominantly cardiovascular (91.4%) (Table S1). Most of the patients received ADT + ARPI (29/35, 82.9%), whereas fewer were treated with ADT + docetaxel (3/35, 8.6%), triplet therapy (1/35, 2.9%), or ADT alone (1/35, 2.9%). Median OS in the non‐cancer death group was 16 months (95% CI 12.6–43.8). The main causes of death were infectious (12/35, 34.3%), cardiovascular (7/35, 20%), noninfectious respiratory (3/35, 8.6%), and other causes detailed in Table 1. No deaths were associated with febrile neutropenia or were chemotherapy‐related. Most patients (25/35, 71%) were receiving active systemic therapy at the time of death, most often ADT + ARPI (12/25, 48%). Among these patients, 13.5% had previously required dose reductions. No significant association was observed between treatment type and cause of death (data not shown). Of note, 30 patients (85.7%) had undetectable PSA, defined as < 0.04 ng/mL, at the time of death. Table 1Non‐prostate cancer‐related causes of death.Causes of death N = 35 (%)Infections12 (34.3) –COVID‐19 2 (5.7) –Urinary 5 (14.3) –Non‐COVID pneumonia 3 (8.6) –Other infections 2 (5.7)Cardiovascular7 (20) –Ischemic heart disease 3 (8.6) –Heart failure 2 (5.7) –Aneurysm 1 (2.9) –Stroke 1 (2.9)Chronic obstructive pulmonary disease3 (8.6)Unknown2 (5.7)Miscellaneous –Delirium 2 (5.7) –Second neoplasm 2 (5.7) –Volvulus 1 (2.9) –Euthanasia 1 (2.9) –Fall with cranial trauma 1 (2.9) –Pulmonary thromboembolism 1 (2.9) –Sudden death 3 (8.6)John Wiley & Sons, Ltd. Non‐prostate cancer‐related causes of death. –COVID‐19 –COVID‐19 COVID‐19 –Urinary –Urinary Urinary –Non‐COVID pneumonia –Non‐COVID pneumonia Non‐COVID pneumonia –Other infections –Other infections Other infections –Ischemic heart disease –Ischemic heart disease Ischemic heart disease –Heart failure –Heart failure Heart failure –Aneurysm –Aneurysm Aneurysm –Stroke –Stroke Stroke –Delirium –Delirium Delirium –Second neoplasm –Second neoplasm Second neoplasm –Volvulus –Volvulus Volvulus –Euthanasia –Euthanasia Euthanasia –Fall with cranial trauma –Fall with cranial trauma Fall with cranial trauma –Pulmonary thromboembolism –Pulmonary thromboembolism Pulmonary thromboembolism –Sudden death –Sudden death Sudden death Patients who died from infectious causes tended to be older and have poorer functional status than those who died from cardiovascular causes (median age 84 vs. 72 years, p = 0.05; median Barthel index 60 vs. 100, p = 0.03) (Table S2). All patients who died from cardiovascular causes had established cardiovascular risk factors (88.6% hypertension, 77.1% dyslipidemia, and 28.6% diabetes). Respiratory deaths occurred in patients with the highest comorbidity burden (median 5). Interestingly, patients with 3 or more comorbidities had a higher risk of early non‐cancer‐related death (univariate Cox HR 2.1; 95% CI 1.2–3.7, p = 0.039). Discussion In this large real‐world mHSPC cohort, non‐cancer causes accounted for a relevant proportion of early mortality, preceding CRPC progression. Infections and cardiovascular events were the predominant contributors, in line with previous evidence [3, 4, 8]. The absence of deaths related to febrile neutropenia supports the safety of docetaxel‐based intensification in appropriately selected patients. Cardiovascular deaths occurred in younger patients with better functional status, consistent with concerns that ADT and ARPIs may accelerate cardiometabolic toxicity in predisposed individuals [6, 9]. Importantly, all patients in this subgroup had pre‐existing cardiovascular risk factors. These observations underscore the need for structured cardiovascular evaluation at baseline, including screening for hypertension, diabetes, dyslipidemia, and prior ischemic disease. Early multidisciplinary management, ideally incorporating a cardio‐oncologist, may reduce preventable mortality [9]. Moreover, promoting healthy lifestyles and exercise programs may also be beneficial [10]. Infectious deaths were more frequent among older and frailer patients with higher comorbidity burden. This pattern highlights the importance of vaccination (influenza, pneumococcal, COVID‐19), comprehensive functional assessment, smoking‐cessation support, and early interventions to address frailty and sarcopenia. Recent studies have proposed treatment de‐escalation or intermittent strategies in patients achieving profound PSA responses [11]. In our cohort, nearly 85% of patients had undetectable PSA at the time of death. These findings provide additional support for considering treatment intermittence in individuals with substantial comorbidity burden who achieve deep biochemical responses and had low risk/volume disease [11]. This study is limited by its retrospective design, potential heterogeneity in clinical practice across centers, and the need for better characterization of risk factors associated with non‐cancer mortality in mHSPC, their impact in different patient populations, and the duration and type of oncologic treatments involved. Overall, these real‐world data reinforce the importance of assessing comorbidities and life expectancy when initiating systemic therapy in mHSPC. Integrating cardiovascular risk assessment, infection‐prevention measures, lifestyle education, and tailored follow‐up may help reduce non‐cancer mortality in this growing patient population. Author Contributions Conceptualization: Marta Garcia de Herreros, Clara Rodrigo, and Begoña Mellado. Methodology: Marta Garcia de Herreros, Clara Rodrigo, and Begoña Mellado. Data collection and curation: Marta Garcia de Herreros and Clara Rodrigo. Formal analysis: Marta Garcia de Herreros and Clara Rodrigo. Writing: Marta Garcia de Herreros, Clara Rodrigo, and Begoña Mellado. Review and editing: all authors. All authors have read and agreed to the published version of the manuscript. Ethics Statement This study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the institutional ethics committees of all participating centers. Consent All patients provided signed informed consent before their inclusion in the study. Conflicts of Interest Begoña Mellado certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript, are the following: Marta Garcia de Herreros reports speaker honoraria from Ipsen and travel and accommodation expenses from Novartis and Pfizer. Laura Ferrer‐Mileo reports speaker honoraria and travel and accommodation expenses from Pfizer and Kyowa Kirin, and research funding from Roche. Òscar Reig reports a consulting or advisory role for BMS, EISAI, and Ipsen, and travel and accommodation expenses from Ipsen and Pfizer. Begoña Mellado reports research funding from Janssen, Roche, Bayer, and Pfizer; speaker bureau participation for Roche, Sanofi, Janssen, Astellas, Pfizer, Novartis, and Bristol‐Myers Squibb; and travel and accommodation expenses from Janssen and Pfizer. M Jose Ribal reports speaker honoraria from Bayer and Roche. Albert Font reports research funding from AstraZeneca; advisory for Janssen, Astellas, and Bayer; and travel and accommodation expenses from Janssen. Antoni Vilaseca has received honoraria from Janssen for advisory and travel expenses. Mariona Figols reports speaker bureau participation for MSD, Ipsen, and Astellas, and travel and accommodation expenses from Merck, Janssen, Ipsen, Bayer, and Roche. Miguel Ángel Climent reports a consulting or advisory role for BMS, MSD, Bayer, EUNSA, Pfizer, Roche, Janssen, Pierre Fabre, and Ipsen; and travel and accommodation expenses from Janssen, Astellas, Roche, Ipsen, and MSD The remaining authors have nothing to disclose. Supporting information Supplementary Figure 1: Study flowchart of the patients included for the analysis. Supplementary Table 1: Baseline characteristics of patients who died from non–cancer‐related causes. Supplementary Table 2: Cause of death by age and comorbidities.