Serum creatinine and cystatin C-based diagnostic indices for sarcopenia in advanced non-small cell lung cancer

Background Sarcopenia is an important prognostic factor of lung cancer. The serum creatinine/cystatin C ratio (CCR) and the sarcopenia index (SI, serum creatinine x cystatin C-based glomerular filtration rate) are novel screening tools for sarcopenia; however, the diagnostic accuracy of the CCR and SI for detecting sarcopenia remains unknown. We aimed to explore and validate the diagnostic values of the CCR and SI for determining sarcopenia in non-small cell lung cancer (NSCLC) and to explore their prognostic values for overall survival. Methods We conducted a prospective cohort study of adult patients with stage IIIB or IV NSCLC. Levels of serum creatinine and cystatin C were measured to calculate the CCR and SI. Sarcopenia was defined separately using CCR, SI, and the Asian Working Group for Sarcopenia (AWGS) 2019 criteria. Participants were randomly sampled into derivation and validation sets (6:4 ratio). The cutoff values for diagnosing sarcopenia were determined based on the derivation set. Diagnostic accuracy was analysed in the validation set through receiver operating characteristic (ROC) curves. Cox regression models and survival curves were applied to evaluate the impact of different sarcopenia definitions on survival. Results We included 579 participants (women, 35.4%; mean age, 58.4 +/- 8.9 years); AWGS-defined sarcopenia was found in 19.5% of men and 10.7% of women. Both CCR and SI positively correlated with computed tomography-derived and bioimpedance-derived muscle mass and handgrip strength. The optimal cutoff values for CCR and SI were 0.623 and 54.335 in men and 0.600 and 51.742 in women, with areas under the ROC curves of 0.837 [95% confidence interval (CI): 0.770-0.904] and 0.833 (95% CI: 0.765-0.901) in men (P = 0.25), and 0.808 (95% CI: 0.682-0.935) and 0.796 (95% CI: 0.668-0.924) in women (P = 0.11), respectively. The CCR achieved sensitivities and specificities of 73.0% and 93.7% in men and 85.7% and 65.7% in women, respectively; the SI achieved sensitivities and specificities of 75.7% and 86.5% in men and 92.9% and 62.9% in women, respectively. CCR-defined, SI-defined, and AWGS-defined sarcopenia were independently associated with a high mortality risk [hazard ratio (HR) = 1.75, 95% CI: 1.25-2.44; HR = 1.55, 95% CI: 1.11-2.17; and HR = 1.76, 95% CI: 1.22-2.53, respectively]. Conclusions CCR and SI have satisfactory and comparable diagnostic accuracy and prognostic values for sarcopenia in patients with advanced NSCLC. Both may serve as surrogate biomarkers for evaluating sarcopenia in these patients. However, further external validations are required.