159622-173

Assessing solid tumor response with and without RECIST.

Category: 
Cancers of the Colon, Rectum, and Anus
Session Type and Session Title: 
Poster Session C: Cancers of the Colon, Rectum, and Anus
Abstract Number: 

504

Poster Board Number: 
Poster Session C Board #A17
Citation: 
J Clin Oncol 34, 2016 (suppl 4S; abstr 504)
Author(s): 
Aileen Deng, Benjamin E Leiby, Russell J. Schilder, William Kevin Kelly, Sandeep Deshmukh, Ashwin Reddy Sama; Thomas Jefferson University Hospital, Philadelphia, PA; Division of Biostatistics, Department of Pharmacology and Experimental Therapeutics, Jefferson Medical College of Thomas Jefferson University, Philadelphia, PA; The Sidney Kimmel Cancer Center at Thomas Jefferson University, Philadelphia, PA

Abstract Disclosures

Abstract: 

Background: Response Evaluation Criteria in Solid Tumors (RECIST) has become widely accepted as gold standard for response evaluation in clinical trials. It remains underutilized in routine clinical practice. We compared tumor response assessment made with and without RECIST. Methods: This study included patients with solid tumors who underwent imaging from January 2013 to December 2014 at a single academic center. Tumor response was assessed by a radiologist using RECIST and by an oncologist (Onc) and resident (Res) without using RECIST (standard report). Tumor response was classified as progressive disease (PD), stable disease (SD), partial response (PR) and complete response (CR). Agreement in assessment between RECIST and standard report was determined by percent agreement and Kappa statistic. Results: 292 imaging studies were included. Concordance between RECIST and Onc-interpreted standard report is presented in Table 1. Overall agreement between RECIST and Onc-interpreted standard report was 56% (95% CI: 46-65%) and Kappa was 0.31 (95% CI: 0.19-0.44). Similar results were seen between RECIST and Res-interpreted standard report (Table 1). Overall agreement between RECIST and Res-interpreted report was 54% (95% CI: 44%-63%) and Kappa was 0.26 (95% CI: 0.13-0.40). Conclusions: Our study found variability in tumor response assessment between clinicians and radiologists. RECIST-classified PD was often interpreted as SD and vice versa, a distinction that affect treatment decisions. Our study highlights the need to standardize tumor response assessment.

Agreement between RECIST and standard report (# reports; % agreement).

RECIST
Report
Oncologist Assessment
Resident Assessment
Total
CRPDPRSDCRPDPRSD
CR120081001920
60%0%0%40%5%0%0%95%
PD15402305631978
1.3%69.2%0%29.5%0%71.8%3.9%24.4%
PR2484003153654
3.7%7.4%14.8%74.1%0%5.6%27.8%66.7%
SD142710890371885140
10%19.3%7.1%63.6%0%26.4%12.9%60.7%
Total29851816019636159292