Residual Tumor Burden in Rectal Cancer: A New Method of Assessing Pathologic Response to Neoadjuvant Therapy
Atin Agarwal, George Chang, Yuan-Chung Hu, Melissa Taggart, Asif Rashid, Scott Kopetz, Dipen Maru. Baylor College of Medicine, Houston, TX; University of Texas MD Anderson Cancer Center, Houston, TX
Background: We assessed implications of residual tumor burden categories and tumor regression grade (TRG) as predictors of recurrence-free survival (RFS) in a large patient population with long term (median:70 months) follow up.
Design: Retrospective study of 251 patients with locally advanced rectal adenocarcinoma treated with neoadjuvant therapy followed by resection was performed. The extent of residual carcinoma was assessed irrespective of nodal status by estimating the proportion of residual cancer cells in relation to the total tumor area. Four subsets were analyzed: no residual cancer cells (complete response), ≤5% residual cancer cells(near complete response), >5% and <50% residual cancer cells(major response) and ≥50% residual cancer cell (minor response). TRG was categorized as: no viable tumor cells (0), small groups of cancer cells (1), residual cancer outgrown by fibrosis (2) and minimal or no tumor kill(3). The pathologist was blinded from all clinicopathologic parameters.
Results: 157 men and 94 women were studied. ypT0N0 was in 19%, ypT1-2N0 in 27%, ypT3-4N0 in 21%, and N+ in 33%. Lymphovascular invasion in 24% and perineural invasion in 10%. 9 had positive margin. Pathology response was complete in 21%, near complete in 20%, major in 37% and minor in 22%. TRG 0 in 21%, 1 in 33%, 2 in 38% and 3 in 8%. Log-rank test indicated difference in RFS by categories of response and TRG was significant (p<.001). The 3 & 5 years RFS, by categories of residual tumor burden: complete 98% & 95%; near complete 88% & 88%; major 74% & 69% and minor 71% & 62% (p<.001). The 3 and 5 years RFS by categories of TRG: 0, 98% & 96%; 1, 83% & 81%; 2, 67% & 59% (p<.001) and 3, 95% & 95%. RFS between patients with TRG 2 & major response was better than TRG 2 & minor response (p=.02). In univariate analysis, factors associated with shorter RFS were major or minor response, high grade (HR:1.98 CI:1.07-3.65), lymphovascular invasion (HR:2.2 CI:1.24-3.72), perineural invasion, higher pT, N+ disease, and positive margin (HR:3.01 CI:1.19-7.63). Major (HR:3.11 95% CI: 0.99-9.71) & minor response (HR: 4.32 95% CI: 1.31-14.19) and perineural invasion (HR:2.44 95% CI: 1.13-5.26) remained significant in the multivariate analysis.
Conclusions: Residual tumor burden is a predictor of RFS in patients with resected rectal cancer after neoadjuvant therapy, and if validated may be a surrogate biomarker suitable for propsepctive clinical trial. This parameter is complementary in further stratifying prognosis of patients with TRG 2.
Tuesday, March 5, 2013 1:00 PM
Poster Session IV # 140, Tuesday Afternoon