Skeletal muscle index impacts the treatment outcome of elderly patients with diffuse large B cell lymphoma.

Sarcopenia is a crucial factor in the physical fitness of elderly individuals. This study investigated the prognostic values of multiple parameters of sarcopenia in association with established prognostic factors in elderly Japanese patients with diffuse large B cell lymphoma (DLBCL). As candidate indicators for sarcopenia, the skeletal muscle index (SMI) (cm 2 /m 2 ), the psoas muscle index, the erector spinae muscle index, the visceral fat index, the subcutaneous fat index, and the visceral to subcutaneous fat area ratio at the third lumbar level were assessed by computed tomography at their initial diagnosis in 102 patients with DLBCL over 75 years old those were diagnosed and treated in our institute from 2007 to 2020. The primary endpoint was overall survival (OS), and the secondary endpoint was progression-free survival (PFS). The median age of patients analyzed was 80 years at diagnosis. The sex-specific cut-offs for the indices adopted two approaches: (i) the historical cut-off values established in the previous study for healthy Japanese individuals (Hamaguchi Y. J Cachexia Sarcopenia Muscle. 2018), and (ii) each sex-specific lowest quartile in our cohort. As the results, SMI evaluated by the historical cut-off and sex-specific lowest quartile was identified as the most influential independent prognostic factor for both OS and PFS among various parameters for sarcopenia. Furthermore, we developed an elderly sarcopenia prognostic index (ESPI). ESPI, which combines SMI evaluated by the historical cut-off and LDH > ULN, demonstrated statistically significant prognostic impacts on OS and PFS. Moreover, compared to the R-IPI, ESPI showed the ability to identify intermediate-risk groups and indicated a trend toward improved predictive accuracy. Our study revealed that SMI is the most appropriate assessment method for evaluating sarcopenia and the critical prognostic factor in OS and PFS of elderly patients with DLBCL.