The Prognostic Value Of Different Efficacy Evaluation Methods In Advanced Non-small-cell Lung Cancer Treated With Immune Checkpoint Inhibitors Combined With Anti-angiogenesis Therapy

Objective: The study was aiming to apply three-dimensional visualization technology to tumor response evaluation and establish a tumor volume-based response evaluation criteria in solid tumors(VB-RECIST),which was compared with three previous tumor diameter assessment-based response evaluation criteria(RECIST 1.1,ir RECIST,i RECIST)to explore the predictive value of tumor volume and CT value changes before and after treatment in the immunotherapy combined with anti-angiogenesis treatment model for survival.Methods: Patients with advanced or metastatic NSCLC who treated with immune checkpoint inhibitors combined with anti-angiogenesis therapy at Oncology Center of China Medical University from January 2018 to December 2021 were enrolled in this retrospective analysis.The final follow-up date was December 31,2022.CT or MR images of evaluable lesions were collected from enrolled patients at following timepoints: 2 weeks before treatment,baseline,and every 6-8 weeks after treatment(for at least 6 months),and the DICOM data of their images were uploaded to Touch Viewer3.0 software to calculate the 3D volume of evaluable target lesions and the mean CT value within the 3D volume of the target lesions using its 3D visualization technology.The calculated sums of diameters or volumes at each examination were compared to all previous examinations that were available.In univariate loess-regression,we explored the relation between the relative volumetric and diameter size changes.We use RECIST1.1,i RECIST,ir RECIST and VB-RECIST to evaluate efficacy.Kaplan-Meier was used to calculated PFS and OS.Log-rank test is used to compare the OS of patients with different response evaluated by the four methods.Results: A total of 61 patients with 106 target lesions were measured.The resulting dataset had 258 changes in diametric size and matching changes in volumetric size.Regression analysis yielded volumetric thresholds of-44.3% for a diameter change of-30%,and +43.3% for a diameter change of +20%,which was used as the threshold for assessing partial response and progression of target lesions by the VB-RECIST criteria.With a 36.1%(22/61)incidence of progression under the RECIST 1.1 criteria.i RECIST,ir RECIST and VB-RECIST had a concordance rate of 72.7%,72.7%,and 68.1%,respectively.Among the patients assessed consistently,in terms of time to progression of the target lesion,VB-RECIST detected progression one treatment cycle earlier(6-8weeks)in 4 patients than RECIST 1.1 criteria.Compared with RECIST 1.1,i RECIST,and ir RECIST,the VB-RECIST criteria performed best in differentiating patients with different response,and the OS were significantly different in patients with different response when this criteria was used.The efficacy of mean CT value change(ΔCT%)for predicting PFS and OS was assessed by ROC curve analysis.The AUC was 0.633 when PFS was used as the status variable,and 0.507 when OS was used as the status variable,suggesting that ΔCT% was better for PFS prediction and less effective for OS prediction.The best cutoff value for ΔCT% prediction of PFS was further calculated to be-19.63%.ΔCT% ≥-19.63% was significantly associated with shorter PFS(HR=4.82;P=0.001).Conclusion: Using the VB-RECIST to measure tumor three-dimensional volume to assess tumor efficacy can better reflect the actual tumor burden compared with the radial measurement in RECIST 1.1,which can identify tumor progression early and more accurately and thus guide treatment;and the rate of change of CT value within the outlining of tumor volume before and after treatment has certain value for predicting anti-tumor efficacy.