Metastatic castration-resistant prostate cancer (MCRPC) represents the final stage of prostate cancer progression, resulting in limited prognosis and often serious clinical symptoms such as bone pain, anemia, cachexia, and dyspnea (1-6). Palliative chemotherapy with docetaxel has long been a mainstay of life-prolonging systemic therapy for MCRPC (7). In recent years, additional options have emerged, including chemotherapy-free alternatives such as abiraterone acetate, enzalutamide, Ra-223, other radioligands, and PARP inhibitors (depending on genetic alterations) (8-12). Due to changes in preceding treatment during the less advanced, hormone-sensitive phase, which may involve upfront administration of docetaxel, abiraterone acetate, enzalutamide, and other drugs (13-17), MCRPC therapy has become less standardized. Docetaxel continues to represent a reasonable option in chemotherapy-eligible and -sensitive patients (18).

Survival after initiation of first-line systemic therapy for MCRPC is highly variable and is influenced by several baseline factors such as performance status (PS), disease extent, and organ function (19). Blood test results may reflect aspects of disease extent and organ function, while also reflecting levels of inflammation and nutritional status (20,21). An increasing body of evidence suggests that body mass index (BMI) is one of the clinically relevant, readily available prognostic factors that may aid in the prediction of survival (22-28). Some of these studies have focused specifically on docetaxel-treated patients with MCRPC (22,24). In general, patients with lower BMI had shorter survival than their peers with higher BMI. However, a complex interplay may exist between BMI, the presence of comorbidities and blood test results. It is possible that patients with a low BMI as a result of cachexia and the presence of visceral metastases should be distinguished from those with less advanced disease and different causes of low BMI. The present study was performed to analyze the impact of BMI and numerous covariates on survival in a homogeneously treated cohort of men who received first-line docetaxel for MCRPC.

Patients. This study utilized our institutional quality-of-care monitoring database, which has been described before (12,18,29). All patients had MCRPC and were managed with first-line docetaxel in various dosing regimens. The latter included the most efficacious dosing every three weeks (75 mg/m²) together with prednisolone and androgen deprivation therapy (ADT). For patients with anticipated tolerance issues, two lower-dose alternatives were available: weekly or biweekly dosing. The weekly regimen was preferred in frail patients with Eastern Cooperative Oncology Group (ECOG) PS 2, very elderly patients, or those with comorbidity-related concerns. Only patients with ECOG PS 0-2 and appropriate bone marrow function received docetaxel. The number of cycles was adjusted to tolerance, effect, and patient preference. Chemotherapy holidays were provided as needed. After docetaxel, further systemic treatment options included abiraterone acetate, enzalutamide, cabazitaxel or Ra-223. Newer targeted radioligands and PARP inhibitors were not accessible in the time period studied here (2009-2023). We used traditional staging methods, such as radionuclide bone scan, computed tomography or magnetic resonance imaging, rather than routine positron emission tomography. A total of 112 patients, all Caucasian and covered by the public Norwegian healthcare system, had received first-line docetaxel for MCRPC and were included in the study.

Methods. Data were analyzed retrospectively. BMI (kg/m²) was calculated from the chemotherapy dosing notes in the hospital’s electronic health record, which contained body weight and height before administration of the first docetaxel cycle. At the same timepoint, routine blood tests were performed. All patients had complete information available, i.e., no missing values. Actuarial survival from the day of docetaxel initiation was calculated with the Kaplan–Meier method and compared between subgroups with the log-rank test. A single patient was alive at the time of analysis in October 2024 and censored at 25 months of follow-up. Date of death was recorded in all other patients. Associations between different variables of interest were assessed with the chi-square or Fisher’s exact probability test (two-tailed). The impact of continuous variables such as BMI and blood test results on survival was examined in univariate Cox analyses. A multivariate forward conditional Cox analysis of prognostic factors for survival was then performed. A p-value <0.05 was considered statistically significant. Statistical analyses were performed with IBM SPSS Statistics 29 (IBM Corp., Armonk, NY, USA).

Baseline characteristics. The majority of patients (54%) belonged to the overweight category (BMI 25.0-29.9), followed by healthy weight (29%) and obesity (17%). The median age was 70 years and typical patients were free from serious comorbidities, had bone-only metastases, and metachronous development of metastases after an initial diagnosis of non-metastatic prostate cancer (Table I). Serum albumin below the institutional lower limit of normal was very rare (2%), while other abnormalities were common [elevated C-reactive protein (CRP), lactate dehydrogenase (LDH), and alkaline phosphatase (ALP), anemia]. The median prostate-specific antigen (PSA) level was 119 µg/l, indicating a relatively high burden of metastases.

Parameters associated with lower BMI. Given that none of the patients were underweight (BMI <18.5), comparisons were made between those with healthy weight (about the first quartile, in exact terms 29%) and all remaining patients (overweight/obese). All baseline characteristics displayed in Table I were compared. A limited number of statistically significant differences emerged. Healthy-weight patients were less often statin users (13 versus 32%, p=0.04), on anticoagulants and/or cardiovascular prevention drugs (19 versus 44%, p=0.02), and their hemoglobin values were lower (median 11.9 versus 13.0 g/dl, p=0.001). However, neither the pattern of metastatic spread nor the comorbidity burden was significantly different. We did not collect information on conditions such as hypercholesterolemia and arterial hypertension. The differences in medication use likely reflect the higher prevalence of these conditions in overweight/obese patients. We also compared median BMI and proportion of men in the two different BMI categories analyzed here in patients with a different pattern of spread such as visceral metastases or bone metastases alone. All these comparisons revealed similar distributions and p-values >0.2.

Survival. Median survival was 16.1 months. At 5 years, 8% were still alive. All baseline characteristics displayed in Table I were included in univariate analysis of prognostic factors for survival. With a p-value of 0.19, BMI was not significantly associated with survival in the Cox analysis. When displayed as Kaplan–Meier curves (Figure 1), healthy-weight patients had shorter survival than overweight/obese patients (median 11.8 versus 19.4 months, p=0.15). Factors with p-value <0.1 are shown in Table II. These included all blood test results, pattern of metastases, presentation setting (de novo versus metachronous metastases), and chemotherapy dosing regimen. In the multivariate forward conditional Cox analysis of prognostic factors for survival, only three baseline parameters retained statistical significance: LDH (p=0.03), hemoglobin (p=0.007), and non-bone metastases (p=0.004). Replacing the continuous variable hemoglobin with the dichotomized variable anemia (local laboratory cut-off <13.4 g/dl) did not change any of the results.

This study was performed as an addendum to several recent publications, which have suggested that BMI is a statistically significant prognostic factor for survival in MCRPC settings including but not limited to docetaxel-treated patients (22-28). In the first docetaxel study (published in 2016, n=63) several body composition parameters (skeletal muscle mass, muscle attenuation, visceral, and subcutaneous fat tissue) were measured at spinal level L3 in addition to BMI (22). Sarcopenia was present in 47% of patients. Measurements of adiposity were not predictive of dose-limiting chemotherapy toxicity. Neither sarcopenia nor sarcopenic obesity was associated with survival. In multivariate analysis of prognostic factors, BMI ≥25 kg/m² was a significant predictor of longer survival and both visceral fat index ≥ median and anemia [hazard ratio (HR)=2.8, 95% confidence interval (CI)=1.3-6.1, p=0.009] were significant predictors of shorter overall survival. The second docetaxel study (published in 2021, n=170) assessed the geriatric nutritional risk index (GNRI), which was calculated using serum albumin and BMI, with a poor nutritional status defined as GNRI <92 (24). In the study, 45 patients were of poor nutritional status. The median survival was 30.4 months in the good nutritional status group and 11.1 months in the poor nutritional status group (p<0.001). In multivariate analysis, poor nutritional status was an independent prognostic factor, together with a high metastatic volume and cumulative docetaxel dose. In the third docetaxel study published in 2022, Verma et al. studied 466 patients without data on serum hemoglobin and found a median survival of 16.6, 20.1, and 21.4 months (p=0.002) for lean, overweight, and obese patients, respectively (27). After adjusting for baseline and tumor characteristics, overweight (p=0.006) and obese (p=0.003) patients had significantly better survival compared with lean patients. Martini et al. studied 1,577 men with MCRPC treated with docetaxel (25). Their analyses were adjusted for age, PSA, ECOG PS, number of metastases, and prior treatment. Regarding survival, BMI emerged as a protective factor both as a continuous variable and as a categorical variable (obesity: HR=0.71, 95%CI=0.53-0.96; p=0.027, relatively to normal weight). No interaction was detected between the BMI categories and the docetaxel dose at any level. In other words, the protective effect of BMI was not related to receiving higher chemotherapy doses.

Our own study did not include GNRI, because only two patients had abnormally low serum albumin. Imaging-based body composition data were not available in our database. Although not statistically significant, lower BMI was associated with shorter survival. There was however an interaction with serum hemoglobin, which was significantly lower in patients with lower BMI. Given that serum hemoglobin was a strong prognostic factor for survival, the impact of BMI remained non-significant in all our analyses. In essence, all five docetaxel studies had a variable extent of available baseline parameters, and eventually different multivariate prognostic model outcomes.

In addition to the study’s size and its retrospective design, limitations of the present work include the lack of information on patient-reported factors such as nausea, appetite, and recent weight loss, which may also predict a shorter survival. It would be interesting to stratify a larger patient population into different categories of low BMI, such as cancer-related, comorbidity-related, and genuine body composition. Strengths of our study include the homogeneous population (Caucasian men with MCRPC receiving first-line therapy), availability of comorbidity and medication data, and complete blood test information. Interestingly, we found no statistically significant impact of comorbidity and medications on survival. BMI did not significantly correlate with the presence of visceral metastases, i.e., more aggressive disease. Analyses of other efficacy endpoints or chemotherapy toxicity were not performed.

Finally, the remaining studies on BMI will be discussed. The Govindan et al. study (n=5,231) is unique, because veterans treated with abiraterone acetate or enzalutamide within the Veterans Health Administration were assessed (28). The Charlson comorbidity index was included. BMI was associated with survival and a longest median survival of 29.8 months in BMI ≥30, 23.9 months in BMI 25-30, 15.9 months in BMI 18.5-25, and 9.2 months in BMI <18.5 (p<0.001). In the multivariate analysis, compared to normal BMI, increased mortality was observed in BMI <18.5 and a decreased mortality in BMI 25-30 and BMI >30. Modonutti et al. studied patients enrolled in the control arm of five randomized trials (first-line MCRPC therapy), randomly split between training (n=1,636) and validation cohorts (n=700) (26). These authors constructed a novel nomogram. At multivariate analysis, several independent predictors of survival emerged: sites of metastasis (visceral vs. bone metastasis), PSA, aspartate transaminase, ALP, BMI, and hemoglobin (≥13 g/dl versus <11 g/dl). A nomogram based on these variables was developed, which showed favorable discrimination and calibration characteristics on external validation. However, all patients had ECOG PS 0-1 and LDH was not part of the studied parameters. Despite these drawbacks, given the data source, methods, and study size, this work lends credit to the overall impression from the recent literature, suggesting that lower BMI is prognostically unfavorable in men starting treatment for MCRPC. Instead of BMI, sarcopenia assessed on computed tomography images has also been studied, with inconsistent results (30).

It is also relevant to note that recent studies have highlighted several trends in MCRPC treatment, e.g., treatment initiation at lower PSA level and often with androgen receptor axis-targeted agents, which allow for treatment of elderly patients unfit for docetaxel (31). These developments impact on second-line systemic therapy, which also has life-prolonging potential. Recent data on second-line treatment suggested that time from first-line androgen receptor axis-targeted agents to progression under 12 months was associated with shorter survival in the nest line consisting of either docetaxel or androgen receptor axis-targeted agents (32). A subgroup of patients may benefit from docetaxel rechallenge during the course of MCRPC, depending on initial response and toxicity profile (33).

An interaction between BMI and hemoglobin was present, explaining the observed survival difference between lean and overweight/obese patients. Comorbidity and medications had no significant impact on survival in this population with limited prognosis (median survival 16.1 months).

The Authors declare that they have no conflicts of interest in relation to this study.

CN participated in the design of the study and performed the statistical analysis. CN, LS, AD, and ECH conceived the study and drafted the article. All Authors read and approved the final article.