Abstract
Background/Aim: Sarcopenia is a syndrome characterized by the progressive and generalized loss of skeletal muscle mass and has been reported to be a poor prognostic factor for taxane-treated castration-resistant prostate cancer (CRPC). However, whether sarcopenia affects androgen receptor axis-targeted therapies (ARATs) remains unknown. In the present study, we investigated the association between sarcopenia in CRPC and treatment outcomes of ARATs. Patients and Methods: From January 2015 to September 2022, 127 patients who received ARATs as 1st-line treatment for CRPC at our two hospitals were included in the study. We retrospectively evaluated sarcopenia using computed tomography images and investigated whether sarcopenia affects the progression-free survival (PFS) and overall survival (OS) of patients with CRPC treated with ARATs. Results: Out of 127 patients, 99 were diagnosed with sarcopenia. The PFS of the sarcopenic group administered ARATs was significantly better than that of the non-sarcopenic group. Furthermore, in the multivariate analysis of PFS, sarcopenia was an independent favourable prognostic factor. However, there was no significant difference in the OS between the sarcopenic and non-sarcopenia groups. Conclusion: ARATs could more effectively treat patients with CRPC and sarcopenia than patients with CRPC without sarcopenia. Sarcopenia may positively influence the therapeutic effects of ARATs.
- Androgen receptor axis targeted therapies
- therapeutic effects
- castration-resistant prostate cancer
- sarcopenia
Huggins et al. reported androgen deprivation therapy (ADT) for advanced prostate cancer over 80 years ago, and ADT has been widely performed since then (1, 2). While ADT is effective, most cases acquire drug resistance within approximately 12-20 months and progress to castration-resistant prostate cancer (CRPC) (2, 3). Previously docetaxel (DOC) was the only effective treatment for CRPC for a long time (2); however, androgen receptor-axis targeted drugs (ARATs) such as abiraterone acetate (ABI) (4, 5), enzalutamide (ENZ) (6, 7), and apalutamide (8) have appeared in recent years as effective treatments for CRPC. Although ARATs have expanded the treatment options for CRPC, the therapeutic effect of ARATs is poor in some cases. However, it is unclear what factors are involved in the therapeutic effects of ARATs.
Various biomarkers that are involved in the therapeutic effects of ARATs have been reported in the literature (9). Expression of androgen receptor (AR) splice variant-7 in circulating tumour cells, pre-treatment tumour nuclear AR over-expression and CYP17 expression in the metastatic lesion, and over-expression of the Ets-related gene (ERG) which is an E26 transformation–specific transcription factor family member, have been reported as biomarkers that are involved in the therapeutic effects of ARATs (10, 11). In clinical practice, it has been observed that the effect of the second agent is lower than the first agent when ARATs are administered consecutively due to cross-resistance between ARATs (12). Therefore, various guidelines recommend sequential therapy of ARATs should be avoided (13, 14). However, most studies about these biomarkers have reported oncological or pharmacological factors. Owing to the extension of the treatment period for CRPC due to the advent of new therapeutic agents, opportunities to treat elderly patients with CRPC who may be treated with ARATs are increasing. Consequently, there is a need for research on CRPC biomarkers from the viewpoint of not only oncological or pharmacological, but also patient factors.
Sarcopenia is a state of reduced physical performance resulting from a natural decline in whole-body muscle mass and strength due to aging (15). It is classified as either primary sarcopenia due to aging or secondary sarcopenia caused by inactivity, disease, or nutritional deficiencies. Diseases that cause sarcopenia include organ failure, inflammatory diseases, cancer, and endocrine diseases. Cancer patients develop cachexia as cancer progresses, which frequently causes sarcopenia. In addition, many cancer treatments, particularly chemotherapeutic agents, can cause both indirect (anorexia, nausea, and fatigue) and direct damage to muscle tissue by pathways that upregulate proteasome activity, activate mitogen-activated protein kinase and extracellular regulated kinase signalling, and induce mitochondrial dysfunction irrespective of anorexia or nutrition (16).
In recent years, sarcopenia has been reported as a prognostic factor for various cancers, such as pancreatic cancer, colorectal cancer, lung cancer and breast cancer (17). Even for prostate cancer (PC), sarcopenia has been reported to be a poor prognostic factor (18, 19). However, most of these reports analysed patients with CRPC, focusing on taxanes or both taxanes and ARATs without distinguishing them. Thus, there are no studies focused on the relationship between sarcopenia and the therapeutic effects of ARATs.
Therefore, this study aimed to reveal the relationship between the therapeutic effect of ARATs, which are the gold standard for CRPC treatment, and sarcopenia.
Patients and Methods
Patients. We reviewed the electronic medical records of 138 consecutive patients treated with ENZ or ABI as 1st-line therapy for CRPC at our two hospitals between January 2015 and January 2022. After excluding 11 patients whose height and body weight records were unavailable, 127 patients were included in the analysis. Age, Gleason score at the time of prostate biopsy, initial PSA, clinical T stage at PC diagnosis, time to CRPC, PSA nadir during initial ADT, previous treatment history, presence or absence of metastasis, 1st-line ARATs (ENZ or ABI), presence or absence of post-chemotherapies, body mass index (BMI), skeletal muscle index (SMI), and survival outcomes were retrospectively evaluated. The study protocol was approved by the Institutional Review Boards of both hospitals (IRB No. 21122) and the study was performed in accordance with the Declarations of Helsinki.
Treatment. The treating physician and patients determined the treatment selection. In addition to ADT, all patients received one of the following treatments: ENZ administered orally at a dose of 160 mg/day or ABI administered orally at a dose of 1,000 mg (4×250 mg tablets daily) and prednisone administered orally at a dose of 5 mg twice a day. Patients continued to receive treatment until disease progression or unacceptable toxicity occurred. Diagnostic imaging studies, such as computed tomography (CT), magnetic resonance imaging, and bone scintigraphy, were performed at the time of PC or CRPC diagnosis and before changes in treatment. The interval between subsequent imaging studies and all therapeutic decisions was left to the discretion of the attending physician. The definition of CRPC was based on that defined by the Prostate Cancer Working Group 2 (PCWG2).
Endpoints. The primary endpoint was progression-free survival (PFS), defined as the time from treatment initiation to the first of the following events: prostate specific antigen (PSA) progression; radiographic progression defined by the modified RECIST version 1.0 as progression locally, to lymph nodes, or distant metastases; or death from PC. PSA progression was defined as a PSA level of at least 2.0 ng/ml higher than and 25% elevation from the nadir PSA level, which was confirmed by a second PSA test at least 4 weeks later, based on the PCWG2. The secondary endpoints were overall survival (OS) and PSA response rates. The OS was defined as the time from randomisation to death from any cause. The PSA response rate was defined as the proportion of subjects with a decline of 50% or more in PSA levels from baseline within the first 12 weeks.
Image analysis. Muscle mass was measured by analysing the cross-sectional muscle area (CSA, cm2) at the level of the third lumbar vertebra from the CT images taken at CRPC diagnosis, using a macro developed by the National Institutes of Health software Image J (Bethesda, MD, USA). The CSA of each image was related to the body area to establish the skeletal muscle index (SMI, cm2/m2). The SMI was calculated by dividing the skeletal muscle area by the square of the height (m). Sarcopenia was defined according to sarcopenia guidelines as in (20): SMI less than 43 with a BMI of less than 25 or a SMI of less than 53 with a BMI of 25 or more (Figure 1A). Patients outside the definition of sarcopenia were included in the non-sarcopenia group (Figure 1B). BMI was calculated as weight (kg)/height squared (m2). To ensure uniformity, CT images were analysed by one researcher in a blinded fashion, and the average value of triplicate measurements was used.
Computed tomography images used to measure body composition. A) Computed tomography image in the axial plane at the L3 level of a sarcopenic patient. B) Computed tomography image in the axial plane at the L3 level of a non-sarcopenic patient. The muscle structures forming the cross-sectional areas surrounded by yellow lines were used to calculate the skeletal muscle index.
Statistical analysis. The patient characteristics were compared between the sarcopenic and non-sarcopenic groups. Continuous variables were compared using the Mann-Whitney U-test, and categorical variables were compared using the chi-squared test or Fisher’s exact test. PSA response rates were compared using Fisher’s exact test. The Kaplan-Meier method was used to draw the survival curve, and the log-rank test was used to compare the PFS and OS between the two groups. Univariate and multivariate Cox proportional hazards survival analyses were used to identify the factors associated with PFS and OS. Continuous variables were dichotomized using round values close to the median as cutoffs. The concept of events per variable set the number of independent variables, and elements with a p-value of less than 0.10 in the univariate analysis were included in the multivariate analysis. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was set at a p-value of less than 0.05.
Results
Patient characteristics. Baseline patient characteristics comparing the sarcopenic and non-sarcopenic groups are shown in Table I. Age, Gleason score at the time of prostate biopsy, initial PSA, clinical T stage at PC diagnosis, time to CRPC, PSA nadir during initial ADT, previous treatment history, presence or absence of metastasis, 1st-line ARAT (ENZ or ABI), presence or absence of post-chemotherapies, and BMI were not significantly different between the two groups. The median age was 78 years (range=65-95 years). Seventy-two patients had a Gleason score of nine and the median initial PSA value was 115.9 ng/ml (2.13-23,356). The T stage at the diagnosis of PC was more progressive than that of T3 in 87 patients. The median time to CRPC was 33 months (range=1-201 months), and the median PSA nadir value during the initial ADT was 0.282 ng/ml (0.005-615). Sixty-two patients received two or more anti-androgen therapies and 25 received estramustine, which is an estrogen and cytostatic antineoplastic agent. With regards to metastasis: 84 patients had bone metastases, 57 had lymph node metastasis, and 18 had visceral metastasis. As 1st-line ARATs for CRPC, 83 patients received ENZ and 44 patients received ABI. The baseline PSA value before 1st-line ARATs was 7.97 ng/ml (0.027-697.7). Furthermore, 34 patients received DOC therapy and 18 received CBZ therapy after 1st-line ARATs. The median BMI was 21.17 (13.09-29.76) and the median SMI was 38.39 (23.77-59.75).
Patient characteristics.
PSA response. The waterfall plot of the maximum percentage change in PSA level from baseline after 1st-line ARATs treatment in the sarcopenia group is shown in Figure 2A. Eighty-three of the 99 patients (83.8%) exhibited a PSA response and 90 of the 99 patients (90.9%) exhibited a PSA decrease. The waterfall plot of the maximum percentage change in PSA level from baseline after 1st-line ARATs treatment in the non-sarcopenia group is shown in Figure 2B. Nineteen of the 28 patients (67.9%) exhibited a PSA response and 21 of the 28 patients (75.0%) exhibited a PSA decrease. The sarcopenic group had a better PSA response rate than the non-sarcopenia group, but the difference was not statistically significant (p=0.103).
Waterfall plots showing the best responses regarding changes in prostate-specific antigen (PSA) levels from baseline after androgen receptor-axis-targeted therapies (ARATs) in the sarcopenia group and the non-sarcopenia group. A) Waterfall plots showing the best responses regarding changes in PSA levels from baseline in the sarcopenia group. B) Waterfall plots showing the best responses regarding changes in PSA levels from baseline in the non-sarcopenia group.
PFS and OS. The median PFS in the sarcopenia group was 26.21 months [95% confidence interval (CI)= 21.25-70.43 months], and the median PFS in the non-sarcopenia group was 16.96 months (95%CI=10.00-36.25 months); hence, the PFS in the sarcopenia group was significantly longer than that in the non-sarcopenic group (p=0.013) (Figure 3). The median OS in the sarcopenia group was 63.61 months (95%CI=35.57-NR months), and the non-sarcopenia group was 62.93 months (95%CI=47.86-80.00 months); hence, there was no significant difference in OS between the two groups (p=0.776) (Figure 4).
Kaplan-Meier curves for progression-free survival in the sarcopenia and non-sarcopenia groups.
Kaplan-Meier curves for overall survival in the sarcopenia and non-sarcopenia groups.
Prognostic factors for PFS and OS. For PFS, the univariate analysis using the Cox proportional hazards model showed that three variables, including the previous use of oestrogen preparations [p=0.011, hazard ratio (HR)= 2.097, 95%CI=1.188-3.701], previous use of ENZ (p=0.025, HR=0.545, 95%CI=0.321-0.925), and SMI (p=0.0270, HR=0.546, 95%CI=0.320-0.933), showed statistical significance (Table II). In the multivariate analysis, the previous use of estramustine (p=0.012, HR=2.084, 95%CI=1.176-3.694), 1st line ARATs (p=0.014, HR=0.512, 95%CI=0.301-0.879), and SMI (p=0.026, HR=0.543, 95%CI=0.315-0.928) were independent predictive factors for PFS (Table II).
Univariate and multivariate analysis for progression-free survival.
For OS, the univariate analysis using the Cox proportional hazards model showed that four variables, including age (p=0.014, HR=2.096, 95%CI=1.161-3.785), T stage at diagnosis (p=0.04644, HR=0.511, 95%CI=0.2634-0.989), bone metastasis (p=0.036, HR=2.199, 95%CI=1.051-4.599) and baseline PSA before 1st line treatment (p<0.001, HR=3.092, 95%CI=1.690-5.560), had statistical significance (Table III). In the multivariate analysis, age (p<0.001, HR=2.987, 95%CI=1.495-5.966), bone metastasis (p<0.001, HR=4.015, 95%CI=1.700-9.481), and baseline PSA before 1st line ARATs (p<0.001, HR=3.092, 95%CI=1.690-5.560) were independent predictive factors for OS (Table III).
Univariate and multivariate analysis for overall survival.
Discussion
Muscle weakness or loss of muscle mass is a significant complication of ADTs. It has been reported that PC cases treated with ADTs for 36 months or longer lose about 2.4% of lean body mass (21). ARATs, which inhibit androgen signalling like ADTs, have the same effect as ADTs in reducing skeletal muscle mass. Fischer et al. reported that ENZ and ABI caused a loss of skeletal muscle mass of approximately 5.2% and ABI caused a loss of skeletal muscle mass of approximately 3.0% (22). However, no studies have revealed the relationship between sarcopenia and the therapeutic effects of ARATs. This study is the first to reveal a relationship between sarcopenia and the therapeutic effects of ARATs.
Approximately 20-30% of patients with non-metastatic CRPC (nmCRPC) were included in both groups. Therefore, the PFS and OS in this study were longer than those of the COU-AA-302 (5) and the PREVAIL trials (7). The PFS of the sarcopenia group was significantly longer than that of the non-sarcopenia group. Furthermore, the sarcopenia group had a slight advantage regarding PSA response, although there was no significant difference between the two groups. This suggests that ARATs may effectively treat patients with PC patients and sarcopenia. Although there is currently no clear explanation for the effectiveness of ARATs in PC with sarcopenia, insulin-like growth factor-1 (IGF-1) may provide a clue.
IGF-1 enhances protein synthesis and promotes muscle hypertrophy (23). Although IGF-1 is a hormone produced mainly in the liver, it has also been shown to be expressed in skeletal muscles. A decrease in IGF-1 with age causes loss of muscle mass; therefore, an imbalance of IGF-1 is one of the causes of sarcopenia (24). Patients with high IGF-1 levels at the start of initial hormonal therapy may not become sarcopenic because they do not have significant muscle atrophy. After CRPC develops, a ligand-independent phosphorylation mechanism of cell proliferation occurs. Cell proliferation mechanisms by ligand-independent phosphorylation are known to induce transcriptional activation of ARs by IGF-1, keratinocyte growth factor, and epidermal growth factor, even in the absence of androgens (25). Therefore, in the non-sarcopenic group with high IGF-1 levels, IGF-1 may induce the transcriptional activity of ARs and reduce the therapeutic effect of ARATs. In practice, IGF-1 has been reported to be associated with PC development and progression (26). However, both IGF-1 and pro-inflammatory factors produced by cancer cells, such as tissue necrosis factor-α and interleukin-6, play essential roles in the pathological mechanisms of sarcopenia (27), and further studies are needed to verify this hypothesis.
In addition to sarcopenia, prior use of estramustine was an independent factor associated with PFS in multivariate analysis. A retrospective study analysing the effect of prior use of ethinylestradiol, which is estrogen preparation same as estramustine, on ENZ treatment for CRPC revealed that cross-resistance between them was observed in the setting of pre-DOC (28); hence, prior use of ethinylestradiol was an independent factor for PFS but not for OS, which is similar to our results. In women, oestrogen depletion is associated with muscle weakness and bone loss (15). Furthermore, oestrogens are constituents of steroid regimens administered to castrated male cattle to stimulate muscle growth and improve carcass quality, indicating that oestrogens may have anabolic effects on male muscle mass (29). Therefore, it is possible that oestrogen administration changed the SMI of patients with CRPC.
The 1st-line ARATs was also an independent factor associated with PFS in the multivariate analysis. Several studies have compared the therapeutic effects of ENZ and ABI as 1st-line therapy for CRPC (30, 31). Although most of them showed no significant difference in OS between ENZ and ABI, ENZ was significantly better in terms of improving PFS (30, 31). ENZ not only acts as an antagonist of the AR, but also does not promote the translocation of ARs to the cell nucleus and prevents the binding of ARs to deoxyribonucleic acid and to coactivator proteins (32). Therefore, ENZ inhibits androgen signalling more strongly than ABI, which acts upstream of the androgen signalling pathway. Moreover, ENZ suppresses PSA mRNA expression and directly inhibits PSA production, lowering PSA levels beyond its actual antitumour effect (33). Hence, the 1st-line ARATs was considered an independent factor related to PFS.
Age, bone metastasis, and PSA level at the start of 1st-line ARATs were independent factors associated with OS. Bone metastasis was an independent factor associated with OS, partly because this study included a mixed cohort of patients with metastatic and non-metastatic CRPC. However, visceral metastasis may not have been an independent risk factor because of the small number of cases. Regarding PSA levels at the start of 1st-line therapy, we previously reported that OS was significantly prolonged when treatment was started at a PSA level of 10 ng/ml or less in patients with CRPC treated with ARATs (34), and similar results were obtained in this study. In addition, this study was a cohort including nmCRPC patients, and it is possible that PSA at the time of CRPC diagnosis correlates with prognosis in nmCRPC patients as well. Miyake et al. also reported that PSA at the time of nmCRPC diagnosis is a prognostic factor for metastatic-free survival (35).
This study has several limitations. First, it is a retrospective analysis of a small number of facilities. Second, we used a standard definition of sarcopenia. Patients with CRPC develop muscle weakness after long-term ADT; therefore, using a standard definition of sarcopenia leads to its diagnosis in many patients with CRPC. Sarcopenia was diagnosed in more than 80% of registered cases in this study. Thus, we believe it is inappropriate to apply the general criteria for sarcopenia in patients with PC, particularly those with CRPC who have undergone long-term hormone therapy. Hence, it is necessary to analyse in detail how muscle mass changes with long-term ADT in patients with PC and to establish criteria for sarcopenia unique to PC. Finally, data on physical activity and nutrition was not collected. Resistance exercise training has been reported to mitigate or reverse some adverse effects of ADT on body muscle composition (21).
In conclusion, this study revealed that ARATs could more effectively treat patients with CRPC and sarcopenia than patients with CRPC without sarcopenia. Sarcopenia may positively influence the therapeutic effects of ARATs. Furthermore, sarcopenia may not be a poor prognostic factor for patients with CRPC treated with ARATs. Further research is required to elucidate the underlying mechanisms.
Acknowledgements
The Authors would like to thank Editage (www.editage.com) for English language editing.
Footnotes
Authors’ Contributions
Conception: Tasuku Hiroshige. Study design: Tasuku Hiroshige. Data acquisition and analysis: Tasuku Hiroshige, Naoyuki Ogasawara, Hisaji Kumagae, Kosuke Ueda, Katsuaki Chikui, Kei-ichiro Uemura, Kiyoaki Nishihara, Makoto Nakiri, Mitsunori Matsuo, Shigetaka Suekane. Article writing: Tasuku Hiroshige. Critical review: Tsukasa Igawa.
Conflicts of Interest
The Authors declare that they have no conflicts of interest in relation to this study.
- Received February 5, 2023.
- Revision received March 7, 2023.
- Accepted March 8, 2023.
- Copyright © 2023, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).