CT based radiomic approach on first line pembrolizumab in lung cancer

Abstract

Clinical evaluation poorly predicts outcomes in lung cancer treated with immunotherapy. The aim of the study is to assess whether CT-derived texture parameters can predict overall survival (OS) and progression-free survival (PFS) in patients with advanced non-small-cell lung cancer (NSCLC) treated with first line Pembrolizumab. Twenty-one patients with NSLC were prospectively enrolled; they underwent contrast enhanced CT (CECT) at baseline and during Pembrolizumab treatment. Response to therapy was assessed both with clinical and iRECIST criteria. Two radiologists drew a volume of interest of the tumor at baseline CECT, extracting several texture parameters. ROC curves, a univariate Kaplan-Meyer analysis and Cox proportional analysis were performed to evaluate the prognostic value of texture analysis. Twelve (57%) patients showed partial response to therapy while nine (43%) had confirmed progressive disease. Among texture parameters, mean value of positive pixels (MPP) at fine and medium filters showed an AUC of 72% and 74% respectively (P < 0.001). Kaplan-Meyer analysis showed that MPP < 56.2 were significantly associated with lower OS and PFS (P < 0.0035). Cox proportional analysis showed a significant correlation between MPP4 and OS (P = 0.0038; HR = 0.89[CI 95%:0.83,0.96]). In conclusion, MPP could be used as predictive imaging biomarkers of OS and PFS in patients with NSLC with first line immune treatment.

Figure 1

Final population enrollment flow-chart.

Figure 2

Figure shows the graphical representation of receiving operator curve (ROC) of pretreatment texture parameters with the best area under the curve (AUC), in particular is possible to appreciate standard deviation and MPP parameters as follows: MPP with Spatial Filter Scale (SSF) 2, 3 and 4 showed an AUC of 0.726, 0.743 and 0.726 respectively (all P < 0.001) and a sensitivity of 72.2, 74.4 and 72.9 and specificity of 70.3, 70.8 and 65,6 respectively. Otherwise, both Standard Deviation at SFF 2 and 3 found an AUC of 0.70 (all P < 0.001) with sensitivity of 69.2 for SSF 2 and 3 and specificity of 71.3 and 69.2 respectively.

Figure 3

Figure shows results of Kaplan–Meier analysis for MPP Texture parameters at Spatial Filter Scale (SSF) 2 for Overall Survival (OS) and Progression Free Survival (PFS). When dichotomized at the optimal threshold identified in Kaplan–Meier analysis, MPP under 56.22 at fine scale (SSF = 2) was significantly associated with lower OS and PFS after administration of Pembrolizumab (red line), while MPP above 56.22 is significantly associated with higher OS and PFS (green line), P = 0.0035 both.

Figure 4

Picture shows two explanatory cases of pretreatment contrast enhanced CT at portal-venous phase and the associated Texture analysis extracted with a dedicated software (TexRAD version 3.2.0, Feedback Medical Ltd,Cambridge, UK; https://fbkmed.com/texrad-landing-2/). (A) shows a patient with good prognosis with an Overall Survival (OS) of 24 months and partial response to treatment while (E) represents a patient with poor prognosis, an OS of 3 months and a progression free survival (PFS) of two months concluded with death; Texture analysis at different spatial scale filters (SSF) is shown for fine SFF in (B,F), medium in (C,G) and coarse in (D,I). In line with the cut-off of 56.22 found on the Kaplan–Meier analysis, above which patients show good OS and PFS: patient reported in (A), with excellent OS had an MPP value at SSF 2 of 74.76 while patient depicted in (E) with poor OS had an MPP value at SFF 2 of 38,97.