2563 Background: The purpose of our study was to compare how effectively RECIST 1.1 captures lymph node disease and treatment response in advanced prostate cancer compared to RECIST 1.0 based on "per lymph node" (LN) analysis Methods: 30 study group pts (med. 69 yrs, PSA 55.55 (0.06 - 2625) ng/ml, Gleason 8) are part of an ongoing prospective clinical imaging trial evaluating the diagnostic utility of FDG PET/CT in advanced prostate cancer. Patients underwent a baseline scan followed by 3 scans done at an 4 monthly intervals. 30 validation cohort pts (med. 70 yrs, PSA 253 (9.61 - 2290) ng/ml, Gleason 8) were identified from 3 IRB approved therapeutic trials for castrate-resistant prostate cancer. 2 consecutive scans done after the start of new therapy in both group of patients were evaluated for the extent of disease and for response by RECIST 1.0 & 1.1 criteria. Number of abnormal LN lesions, defined as target lesions based on morphological appearance using RECIST 1.0 (CTLD ≥10 mm) and RECIST 1.1 (CTSD ≥15mm), were calculated for both groups of pts. The response of concordant target LNs was assessed as CR, PR, SD and PD based on RECIST 1.0 and 1.1 cut-off points.
RESULTS
Of 75 LNs in the study group considered as target lesions by RECIST 1.0, only 25 (33.4%) were considered abnormal by RECIST 1.1. Similarly, of 83 individual LN lesions in the validation group by RECIST 1.0, only 41 (49.4%) were considered target lesions by RECIST 1.1 measurement criteria. Of the 20 lesions that were considered PD in both groups combined by RECIST 1.0, only 10 (50%) were captured by 1.1 criteria. Only 3 (15.8%) LN lesions were considered CR by 1.0 compared to 19 by 1.1 criteria.
CONCLUSIONS
RECIST 1.1 criteria for categorizing lymph nodes as normal or pathologic based on more stringent measurement rules may lead to delayed estimates of disease progression or fail to detect it entirely. RECIST 1.1 may also categorize more patients as complete responders than RECIST 1.0. We will further evaluate these cohorts for effect on individual outcome parameters. These preliminary findings need to be assessed in prospective trials and have major implications for comparing trials done with RECIST 1.0 or 1.1.