A lower psoas muscle index predicts a poorer prognosis in metastatic hormone‐naïve prostate cancer

Abstract Introduction A recent investigation revealed that sarcopenia was associated with a poorer prognosis in some solid malignancies, including prostate cancer. In most reports, sarcopenia was defined as a low psoas volume on CT. This study investigated the association of sarcopenia, determined according to the psoas muscle volume and density on CT, with the prognosis in patients with metastatic hormone‐naïve prostate cancer (mHNPC). Methods A total of 66 patients initially diagnosed with mHNPC were enrolled in this study. Skeletal muscle was evaluated according to the psoas muscle index density (PMID) on computed tomography scans. The psoas muscle volume was calculated at the level of L3 and CT density was evaluated as the mean CT density at the psoas muscle area. We divided the patients into higher and lower PMID groups. Results The lower PMID group (on both sides) showed a poorer overall survival than the higher PMID group (Right: 32.5 vs 99.0 months in Rt PMID, P = .014; Left: 36.0 vs 100.0 months in Lt PMID, P = .029). The lower PMID group (on both sides) showed a shorter time to CRPC (Right: 9.0 vs 42.0 months in Rt PMID, P = .006; Left: 9.0 vs 31.0 months in Lt PMID, P = .005). A multivariate analysis showed that lower Rt PMID and Lt PMID were independent risk factors for poorer OS (HR:2.02, 95%CI: 1.04‐3.90, P = .037, HR:2.29, 95%CI: 1.18‐4.47, P = .015, respectively). For CRPC, both Rt and Lt lower PMID also showed independent risk factors for shorter time to CRPC (HR:2.39, 95%CI: 1.23‐4.62, P = .010, HR:2.43, 95%CI: 1.23‐4.78, P = .010, respectively). Conclusions Among mHNPC patients, both lower PMID groups showed a poorer overall survival and shorter time to CRPC than the higher PMID groups.


| INTRODUC TI ON
In 1989, Rosenberg suggested the concept of sarcopenia as decrease of skeletal muscle. 1 The term sarcopenia was coined as a combination of the words "sarx" and "penia." Recently, "sarcopenia" has been used to describe a decrease in skeletal muscle due to various reasons, in addition to the decrease associated with aging in elderly individuals. In 2010, the European Working Group on Sarcopenia in Older people (EWGSOP) recommended using both a low muscle mass and low muscle function (strength or performance) to define sarcopenia. 2 The Asian working group for sarcopenia (AWGS) also suggested a definition of sarcopenia in 2014. 3 Because the skeletal volume is not strictly correlated with muscle strength, the EWFSOP and AWGS included muscle strength in the definition of sarcopenia. 3 The presence of malignant disease was thought to be a reason for the development of sarcopenia and a recent study suggested that systemic chemotherapy and radiotherapy were associated with decreased muscle volume and influenced the prognosis of malignant diseases. 4,5 Previous sarcopenia-related studies have evaluated the area or volume of the psoas muscle and investigated the influence on the prognosis. 6 For pancreatic malignancy, the psoas area and density-predicted prognosis were shown to be independent risk factors for the prognosis. 7 However, the qualitative of psoas muscle has not been examined in prostate cancer. 7,8 In the present study, we examined the importance of the psoas muscle index density (PMID) as a qualitative prognostic factor in metastatic hormone-naïve prostate cancer (mHNPC).

| MATERIAL S AND ME THODS
A total of 66 patients with mHNPC and who were managed in Yokohama City University Medical Center (Yokohama, JAPAN) between 2008 and 2014 were enrolled in this study. All patients had de novo metastatic prostate cancer and received androgen deprivation therapy without up-front abiraterone or docetaxel. The exclusion criterion was enrollment in clinical trial for initial mHNPC treatment.
CT scans were used for calculating the prostate muscle area. The psoas muscle index (PMI) was calculated as the right/left area at the level of L3 divided by the square of body height. The PMID was calculated multiplied between PMI and CT density in the psoas muscle area ( Figure S1).
Receiver operator characteristic curves (ROCs) were used to determine the candidate cutoff points. For ROCs, we set the event as death at 4 years after the diagnosis and castration-resistant prostate cancer (CRPC) at 3 years after the diagnosis. The time to CRPC and the OS were analyzed using the Kaplan-Meier method, and the resultant curves were statistically tested by the log-rank method. A Cox proportional hazards model was used for univariable and multivariable analysis. Multivariable analysis was performed using the factors which showed significant differences by univariable analysis.
We also performed multivariable analysis using important clinical factors including Gleason Score, lactate dehydrogenase (LDH), and Rt PMID. P values of <.05 were considered to indicate statistical significance in all statistical analyses.

| RE SULTS
A total of 66 patients were enrolled in this study. This study was  (Figure 1). There were no marked differences between the higher and lower CT density groups ( Figure S2).
As shown in Figure  A multivariate analysis showed that both lower Rt and Lt PMID were an independent risk factor for a poorer OS (hazard  used CT, as CT scans are easy to analyze retrospectively. In particular, at the L3 level, the psoas muscle volume has been reported to be correlated with the whole-muscle volume, and the PMI is determined as the psoas muscle area divided by the square of body height. 9 However, the PMI does not reflect the muscle power. In pancreatic malignancy, the psoas area and density predicted prognosis were shown to be independent risk factors for the prognosis. 7 Thus, the present study used PMID to reflect the muscle power. We used the PMID as a multiplier between PMI and the CT density and a lower PMID was related to a poorer prognosis in patients with mHNPC.

| D ISCUSS I ON
Chakedis et al and Namn et al revealed the importance of the psoas muscle density on CT in patients with biliary tract cancers and pancreatic cancer. 7,8 Several limitations associated with the present study warrant mention. First, it was a nonrandomized retrospective study and enrolled a relatively small number of patients, so a further prospective study is needed in order to confirm these results.
Second, we were unable to clarify the mechanism underlying the relationship between low PMID and a poor prognosis. A previous report suggested the involvement of systemic inflammation and sarcopenia. 4,5 While this study was unable to confirm the relevant mechanism, sarcopenia might influence advanced prostate cancer development. Third, the HR in each side of PMID showed differences. Our previous report also revealed that the both psoas muscle volume were correlated. 10 Though the meanings of differences between right and left, both PMID would be a useful factors to predict prognosis. Further study is needed to confirm the HR by lower PMID.
Recent studies have identified sarcopenia (loss of skeletal muscle mass) as one of the important factors for poorer prognostic factors. 11 The PMI, which was a relatively simple method to represent skeletal muscle volume in whole body. This index is attracting attentions as one of the prognostic factors for malignancy, although it was popularized as an index to quantify sarcopenia at first. Based on the findings of the present study, sarcopenia would be associated with a poor prognosis in mHNPC patients. Psoas muscle volume and muscle CT density might be clues that can be used to predict the prognosis of patients with mHNPC. Abbreviations: Alb, albumin; ALP, Alkaline phosphatase; Cre, creatinine; Hb, hemoglobin; LDH, lactate dehydrogenase; PMID, psoas muscle index density; PSA, prostate-specific antigen.