ESTRO 2019 – Total neoadjuvant therapy for patients with locally-advanced rectal cancer: which is the better sequence of treatment?

  • Elena Riboldi — Agenzia Zoe
  • Univadis
L'accesso ai contenuti di questo sito è riservato agli operatori del settore sanitario italiano L'accesso ai contenuti di questo sito è riservato agli operatori del settore sanitario italiano


  • More patients with locally-advanced rectal cancer treated with total neoadjuvant therapy (TNT) achieve a pathological complete response (pCR) if preoperative chemoradiotherapy (CRT) is administered prior to chemotherapy (CT), and not conversely.

Why this matters

  • Optimal scheduling of preoperative CRT and CT has yet to be determined.
  • This is the first trial to report safety and efficacy of TNT sequences.
  • Results from this trial led to the selection of the TNT sequence for further phase 3 comparison with standard preoperative CRT.


Study design

  • The multicentre, phase 2, “pick the winner” design, CAO/ARO/AIO-12 trial enrolled 311 patients with stage II-III rectal cancer.
  • Patients were randomly assigned to induction CT prior to CRT (group A) or to consolidation CT following CRT (group B).
  • CRT consisted of 50.4 Gy (28 fractions) plus infusional 5-fluorouracil (5-FU), and oxaliplatin.
  • CT consisted of oxaliplatin (Ox), leucovorin, and infusional 5-FU (3 cycles).
  • In both groups, surgery was scheduled on week 18.
  • The primary endpoint was pCR.


Key results

  • Compliance to CRT was higher in group B: full dose RT, 5-FU, Ox was received, respectively, by 91%, 78%, and 76% in group A and 97%, 87%, and 93% in group B.
  • Compliance to CT was higher in group A: 92% vs 85% of patients in group A and B, respectively, completed the 3 cycles of Ox.
  • CRT-related grade 3-4 toxicity was lower in group B (37% vs 27%); CT-related grade 3-4 toxicity was 22% in both groups.
  • In the intention-to-treat population, a pCR was achieved in 17% (95% CI 12%-24%) in group A and in 25% (95% CI 18%-32%) in group B.
  • Only group B (P=0.0002) but not group A (P=0.210) fulfilled the predefined statistical hypothesis that TNT can increase the pCR rate from 15% (expected after standard CRT) to 25%.