In 1993, ‘sarcopenia’ was coined by Evans and Campbell (1) from a combination of the Greek ‘sarx’ for flesh and ‘penia’ for loss (2). The term sarcopenia, which was used for age-related loss of skeletal muscle mass and function, has also been adapted to refer to the severe loss of muscle mass and function associated with adverse outcomes in oncology (3,4). In the clinical definition of cancer cachexia, sarcopenia is part of the diagnostic criteria along with weight loss and low Body mass index (BMI) (5). In clinical practice, ovarian cancers are often diagnosed at advanced stages, complicated by large tumors, massive ascites, and comorbidities, such as deep vein thrombosis (DVT) and pulmonary embolism (PE). As a result, many patients with ovarian cancers experience cachexia and sarcopenia because of appetite loss and systemic inflammation (6,7).

Several reports, in which digital axial computed tomography (CT) was used, have suggested that sarcopenia is associated with poor prognoses in several solid cancers, including ovarian cancers (8,9). In those papers, sarcopenia was assessed by the total area of the skeletal muscle mass at the third lumbar vertebra, which was demarcated using specialized imaging software that analyzed the number of Hounsfield units (HU) in the CT scans (8,9). On the other hand, Rutten et al (10) reported that the psoas muscle area at the third lumbar vertebra was not representative of the total skeletal muscle area for the assessment of sarcopenia in ovarian cancer.

The purpose of this study is to evaluate the association between the prognoses of epithelial ovarian cancer patients and sarcopenia assessed based on the psoas major muscle area at the fifth lumbar vertebra, which was easily detected relative to the position of the ilium and sacrum (11).

In the present study, the PMI evaluated at the fifth lumbar vertebra as an index for sarcopenia was identified as an independent unfavorable prognostic factor in Japanese patients who received upfront treatment for ovarian cancers. To our knowledge, this is the first report using multivariate analyses to demonstrate the association between the loss of psoas major muscle mass and risk of poor outcome in ovarian cancer patients who received TC therapy. This finding provides evidence that we might be able to evaluate only one muscle to determine the prognosis of epithelial ovarian cancer in primary therapy using the TC regimen. PMI was associated with overall survival in ovarian cancer patients, and the significant impact of PMI on prognosis was confirmed by two independent multivariate analyses.

Conrad et al (13) reported that pre-operative PMI at the fourth lumbar vertebra was not associated with poor prognosis and that PMI, in combination with hypoalbuminemia, was associated with unfavorable prognosis in advanced ovarian cancer. In their paper, the mean PMI was 2.80 cm²/m². On the other hand, in this study, the mean PMI was 5.48 cm²/m². We speculate that the difference in muscle volume distribution due to the range of stages had a strong influence on survival.

Rutten et al (10) insisted that the change in psoas muscle area was not representative of the change in total muscle area and should not be used as a substitute for the total skeletal muscle in predicting survival in patients with ovarian cancer. In their study, the median psoas muscle area at the third lumbar vertebra was 13.3 cm². On the other hand, in this study, the median psoas muscle area at the fifth lumbar vertebra was 12.7 cm². We speculated that our patients had a smaller muscle mass because the fifth lumbar vertebra has a larger muscle area than the third lumbar vertebra (Fig. 3). In our study, the proportion of overweight and obese patients (18.1%) was small compared to the proportion (35.7%) in Rutten's report. As shown in our previous study, there is a positive and moderate correlation between BMI and PMI, which might have influenced the different results between their study and ours (11).

The psoas area is often measured at a standard lumbar vertebral landmark (L3 or L4), but sometimes, unreliable soft tissue landmarks, such as the umbilicus, have been used (14). Previous papers mainly reported on sarcopenia assessed by the total area of skeletal muscle mass at the third lumbar vertebra (3,6,8-10). It is easy to measure the psoas major muscle areas when the fifth lumbar vertebra, where the ilium bones are located in the lower half dorsal side in the horizontal section of CT imaging, is used as a landmark. When evaluating sarcopenia using the psoas major muscle only, the belly of the muscle is not located at the third lumbar vertebra but actually at the fifth lumbar vertebra (Fig. 3). In this study, we used a PMI calculated using the elliptical region of interest that could be obtained with a simple method, as previously described (11). We believe that easier detection and measurement of PMI are of great importance for clinical practice.

In a recent meta-analysis not including Asian countries, the skeletal mass index was significantly associated with survival (15). Conversely, in a cohort study from the United States of America, sarcopenia was not associated with poor survival outcomes in epithelial ovarian cancer patients receiving upfront platinum and taxane-based chemotherapy (16). In a report from Asia, Kim et al (17) found that the skeletal mass index was not associated with worse survival outcomes, but patients with a high fat-to-muscle ratio showed significantly worse OS in advanced-stage, high-grade serous ovarian carcinoma. Conversely, Huang et al (18) reported that the skeletal muscle index was independently associated with poor OS in patients with stage III epithelial ovarian cancer treated with primary debulking surgery and adjuvant platinum-based chemotherapy. Although sarcopenia has been recognized as an important prognostic factor, sarcopenia remains controversial in terms of what its optimal index is and what the optimal cut-off values are. Moreover, values will differ according to the number of lumbar vertebra and the patient's race.

In pancreatic cancer, sarcopenic obesity has been recognized as a poor prognostic factor in meta-analysis (19). In ovarian cancer, a Korean group reported that a high fat-to-muscle ratio was related to significantly worse OS (17). Furthermore, in the present study, being overweight (BMI ≥25 kg/m²) was a poor prognostic factor, independent of sarcopenia, for OS in multivariate analysis #2. These results are congruent with those reported by Kim et al (17). Sarcopenic obesity is associated with elevated markers of inflammation such as interleikin-6 (IL-6), tumor necrosis factor (TNF) and C-reactive protein (CRP) in postmenopausal women (20). Inflammation that causes cachexia and sarcopenia causes sarcopenic obesity, and obesity itself has a negative effect on inflammation and tumor-promoting effects (21). We believe that sarcopenic obesity potentially has a significant impact on survival, even in ovarian cancer.

In conclusion, sarcopenia assessed using PMI values measured at the fifth lumbar vertebra had a significant and independent impact on OS in patients with epithelial ovarian cancers, though further analyses including a large number of patients are needed. Moreover, the PMI value should be evaluated along with impacts of other drugs, such as molecularly targeted drugs and immune checkpoint inhibitors. The measurement of the psoas major muscle areas at the fifth lumbar vertebra could be an important indicator for the prognosis of epithelial ovarian cancers.