Objective: Preoperative radiation significantly decreases the number of lymph nodes (LNs) retrieved from rectal cancer patients, but little is known of its prognostic value under these circumstances.Methods: Surveillance, Epidemiology, and End Restults Program (SEER)-registered ypⅢ stage rectal cancer patients, and patients from Fudan University Shanghai Cancer Center (FUSCC) were combined and analyzed.Optimal cutoff value for number of LNs retrieved was first determined by X-tile program.Next, Kaplan-Meier curves and Log-rank tests were adopted to generate and compare the survival rates for ypNO patients using separated cutoff value from 2 to 20.Results: We identified the optimal cutoff for retrieved LNs as 10 (x2=12.182, P=0.014), which was validated as an independent prognosis factors in a Cox regression model.Further analysis showed that the LN count was only a prognosis factor with cutoffs from 8 to 16(except for 13).After a cutoff of 16, the 5-year survival rate decreased gradually.Among the FUSCC patients, there were fewer LNs retrieved from patients with good tumor regression compared to patients with poor regression, and tumor response grade was an independent prognosis factor.Conclusions: Our results confirmed that the number of LNs in ypⅢ stage rectal patients was a prognosis factor only with cutoffs from 8 to 16(except for 13).Using the total mesorectal excision technique with an adequate pathologic examination, a large number of LNs retrieved(≥ 17) might indicate worse tumor response grade and poorer survival.