Sarcopenia is defined by the European Working Group on Sarcopenia in Older People 2 (EWGSOP2) as progressive and generalized skeletal muscle degeneration with reduction in muscle strength, quality, and/or performance [1], which is further classified according to etiology into primary (attributed to the aging process) or secondary (caused by cytokine-mediated inflammation in chronic diseases such as cancer, endocrine disease, or organ failure) [2]. Sarcopenia in cancer patients is associated with lower survival and worse tolerance to chemotherapeutic agents [2,3,4,5,6]. The standard radiological method to estimate sarcopenia is to measure skeletal muscle mass at the level of the third lumbar vertebral body (L3) on axial computed tomography (CT) images using either manual or deep learning automated segmentation [7,8,9,10,11]. Software packages frequently provide measurements of visceral and subcutaneous fat as well.

In the article recently published in European Radiology by Surov et al [12], the authors investigated the predictive value of skeletal muscle mass (SMM) for dose-limited toxicity (DLT) and survival in patients suffering from primary CNS lymphoma. They found a higher objective response rate (ORR) in patients with higher muscle density. In other words, sarcopenia was associated with poor ORR. The authors used freely available software to measure skeletal muscle area (SMA), skeletal muscle index (SMI), visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), and visceral to subcutaneous adipose ratio (VSR) [12]. Measurements were done utilizing pre-defined HU thresholds (− 190 to − 30 to define fat and − 29 to 150 to define muscles) on post-contrast (portal phase) axial CT images at mid L3. Although this method does not require precise training or long post-processing times, it requires post-contrast CT scans of the abdomen, which might not be routine for patients with CNS lymphoma given that PET has a higher diagnostic yield than CT [13]. Another limitation is that routine measurements of sarcopenia are performed on non-contrast images to avoid contrast-related changes in muscle density (HU), which may in turn affect the definition of tissues that was performed using pre-defined HU (as mentioned earlier). The authors of this article claim that the area of adipose tissue and muscle would not be affected [12]. Finally, the researchers choose to define complete/partial response or progressive/stable disease based on the size of enhancing lesions. Complete response (CR) was defined as complete disappearance of contrast enhancement on MRI. Partial response (PR) was defined as a 50% decrease in enhancing tumor diameters. Progressive disease (PD) was defined as a 25% increase in the enhancing lesions or the appearance of any new CNS or non-CNS site of disease. Any other situation was characterized as stable disease (SD). Although this method is widely applied, new consensus recommendations are to use volumetric measurements combined with functional data from DWI, perfusion MRI, and PET for systemic staging evaluation in clinical practice and clinical trials [14].

Those limitations aside, the main strength of this study is that it shows a strong association between low skeletal muscle mass (LSMM) and objective response rate, which may have clinical implications around patient selection for treatments and trials and for inclusion oncology nutritionists in the PCNS lymphoma treatment team.