P-315 - Predictors of pathologic complete response after neoadjuvant treatment in rectal adenocarcinoma

Date 04 July 2015
Event WorldGI 2015
Session Posters
Topics Anticancer Agents
Pathology/Molecular Biology
Colon and Rectal Cancer
Basic Scientific Principles
Biological Therapy
Presenter M. Ayadi
Citation Annals of Oncology (2015) 26 (suppl_4): 1-100. 10.1093/annonc/mdv233
Authors M. Ayadi1, A. Mezlini1, F. Letaief-ksontini1, S. Ayadi2, T. Messai1, M. Kacem3
  • 1Institut Salah Azaiz, Tunis/TN
  • 2La Rabta Hospital, Tunis/TN
  • 3Rabta, Tunis/TN



Rectal adenocarcinoma is common worldwide and in Tunisia. The management of locally advanced stages is based on neoadjuvant radiation therapy (RT) followed by surgery with total mesorectal excision (TME). Nowadays, the combination of preoperative chemotherapy tends to become a standard, by potentiating the effects of RT. Our aim was to assess the response rate to neoadjuvant therapy and identify the predictors of histological regression after neoadjuvant RT or radiochemotherapy (RCT).


Between January 2000 and December 2011, 64 patients with resectable cancer of the lower and the middle rectum (T3 /T4 and/or N+) with no evidence of distant disease were assigned to receive neoadjuvant RT or RCT. Histological response was defined on resection specimen according to Dworak score and patients were classified into non responders (NR): grade 0; having a pathologic partial response (pPR): grade 1,2,3 or a pathologic complete response (pCR): grade 4.


Our study included 36 men and 28 women. The average age was 57 years. At diagnosis, the tumor was classified as T2 in 14 (pts), T3 in 38 (pts) and T4 in 12 (pts). Forty-seven (pts) were considered (cN +). Twenty-four (pts) had neoadjuvant RCT and 40 RT alone. An abdominoperineal resection (AAP) was performed in 29 (pts) (45%). Anterior resection with TME and colorectal or coloanal anastomosis was performed in 34 (pts) (53%). One patient had local resection. Histologically, 12 (pts) (19%) had a pCR, 24 (pts) (73%) had a pPR and 28 (44%) were NR. Four (pts) among those with pCR had an (AAP). In univariate analysis, predictors of PCR were: non-fixed tumors in rectal examination at diagnosis (p = 0,003), absence of lymph node involvement on initial imaging (cN0) (p = 0.021), normal initial CEA level (p = 0.01), hemoglobin level ≥ 12 g / dl (p = 0.009) and concomitant chemotherapy with RT (p = 0.021). After neoadjuvant therapy, tumor regression in rectal examination (p = 0.002) and tumor downstaging in imaging (p = 0.001) were correlated with histological regression. Multivariate analysis shows that the main predictors of p CR were combination of CT with neoadjuvant RT, (c N0) stage and tumor regression on imaging after neoadjuvant therapy.


Identifying predictors of pathological complete response to neoadjuvant therapy could select good-responder patients to whom we could propose a non-aggressive treatment and even surveillance.