Comparison of Metabolic and Morphological Response Criteria for Early Prediction of Response and Survival in NSCLC Patients Treated With Anti-PD-1/PD-L1

Introduction/Aim: Immunotherapy with immune checkpoint inhibitors (ICIs) has positively changed the history of several malignant tumors. In parallel, new challenges have emerged in the evaluation of treatment response as a result of their peculiar anticancer effect. In the current study, we aimed to compare different response criteria, both morphological and metabolic, for assessing response and outcome in patients with advanced non-small cell lung cancer (NSCLC) treated with ICI. Materials and Methods: Overall, 52 patients with advanced NSCLC candidate to ICI were prospectively evaluated. Inclusion criteria comprised whole-body contrast-enhanced CT and (18)F-FDG PET/CT at baseline and at the first response evaluation 3 or 4 cycles after ICI. Response assessment on CT was performed according to RECIST 1.1 and imRECIST criteria, whereas metabolic response on PET was computed by EORTC, PERCIST, imPERCIST, and PERCIMT criteria. The concordance between the different tumor response criteria and the performance of each criterion to predict progression-free survival (PFS) and overall survival (OS) were calculated. Results: Inclusion criteria were fulfilled in 35 out of 52 patients. We observed a low agreement between imRECIST and imPERCIST (κ = 0.143) with discordant response in 20 patients, particularly regarding stable disease and progressive disease groups. Fair agreement between imRECIST and EORTC (κ = 0.340), and PERCIST (κ = 0.342), and moderate for PERCIMT (κ = 0.413) were detected. All criteria were significantly associated with PFS, while only PERCIMT and imPERCIST were associated with OS. Of note, in patients classified as immune stable disease (iSD), imPERCIST, and PERCIMT well-differentiated those with longer PFS (p < 0.001, p = 0.009) and OS (p = 0.001, p = 0.002). In the multivariate analysis, performance status [hazard ratio (HR) = 0.278, p = 0.015], imRECIST (HR = 3.799, p = 0.026), and imPERCIST (HR = 4.064, p = 0.014) were predictive factors for PFS, while only performance status (HR = 0.327, p = 0.035) and imPERCIST (HR = 3.247, p = 0.007) were predictive for OS. Conclusions: At the first evaluation during treatment with ICI, imPERCIST criteria correctly evaluated treatment response and appeared able to predict survival. Moreover, in patients with iSD on CT, imPERCIST were able to discriminate those with longer survival. This advantage might allow for earlier therapy modification based on metabolic response.