Abstract 740P
Background
N+I improved Overall survival and overall response rate in intermediate/poor-risk pts with a clear cell aRCC. However, 20 % of pts didn’t respond to immunotherapy. Identification of early progression is crucial to initiate rapidly a new therapy. We aimed to compare two radiological volumetric methods versus RECIST 1.1 gold standard unidimensional measurement in clear cell aRCC treated by N+I.
Methods
Forteen pts with clear cell aRCC treated with N+I were included in a retrospective monocentric non-interventional study. Follow-up CT scan explorations were reviewed by two blinded- radiologists. Unidimensional RECIST and volumetric measurements were compared at each time-point. The main criteria was the inter-observer agreement for each method. The secondary criteria was the tumoral response assessment based on three different items: RECIST 1.1, spherical volumetric method (SVM), ellipsoidal volumetric method (EVM). Last criteria was median time to progression in each tumoral response assessment.
Results
Forteen aRCC pts were identified (median age 63.6). Intraclass coefficient correlation in volumetric method (0.986 [95% CI: 0.980, 0.990]) was higher than in RECIST (0.903 [95% CI: 0.861, 0.928]). Relative measurement differences with Bland and Altman plot was lower in volumetric method with shorter limits of agreement (0.8%; upper LOA95%: 36.5; lower LOA95%: −35), versus those in RECIST (-5.1 % (upper LOA95%: 46; lower LOA95%: −57). Volumetric method (especially ellipsoidal one) assesses progression disease earlier than RECIST for 57% (N= 8) pts. Median time to progression was 17 months, 18 months and 22 months, respectively with SVM, EVM and RECIST. No difference for partial response assessment was identified.
Conclusions
Volumetric assessment for tumoral response in metastatic RCC may predict earlier the progression disease with a higher inter-observer agreement.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.