Annals of Surgery

Skip Navigation LinksHome > April 2011 - Volume 253 - Issue 4 > Preoperative High-resolution Magnetic Resonance Imaging Can...
Annals of Surgery:
doi: 10.1097/SLA.0b013e31820b8d52
Original Articles

Preoperative High-resolution Magnetic Resonance Imaging Can Identify Good Prognosis Stage I, II, and III Rectal Cancer Best Managed by Surgery Alone: A Prospective, Multicenter, European Study

Taylor, Fiona G.M MBBS, FMRCS*; Quirke, Philip PhD, BM, FRCPath; Heald, Richard J MB, Bch, FRCS; Moran, Brendan MB, Bchir, FRCSI; Blomqvist, Lennart MD, PhD§; Swift, Ian MS, FRCS, FICS*; Sebag-Montefiore, David J FRCP, FRCR; Tekkis, Paris BMBS, MD, FRCS**; Brown, Gina MBBS, MD, FRCR for the MERCURY study group††

Collapse Box


Objective: To assess local recurrence, disease-free survival, and overall survival in magnetic resonance imaging (MRI)–predicted good prognosis tumors treated by surgery alone.

Background: The MERCURY study reported that high-resolution MRI can accurately stage rectal cancer. The routine policy in most centers involved in the MERCURY study was primary surgery alone in MRI-predicted stage II or less and in MRI “good prognosis” stage III with selective avoidance of neoadjuvant therapy.

Patients and Methods: Data were collected prospectively on all patients included in the MERCURY study who were staged as MRI-defined “good” prognosis tumors. “Good” prognosis included MRI-predicted safe circumferential resection margins, with MRI-predicted T2/T3a/T3b (less than 5 mm spread from muscularis propria), regardless of MRI N stage. None received preoperative or postoperative radiotherapy. Overall survival, disease-free survival, and local recurrence were calculated.

Results: Of 374 patients followed up in the MERCURY study, 122 (33%) were defined as “good prognosis” stage III or less on MRI. Overall and disease-free survival for all patients with MRI “good prognosis” stage I, II and III disease at 5 years was 68% and 85%, respectively. The local recurrence rate for this series of patients predicted to have a good prognosis tumor on MRI was 3%.

Conclusions: The preoperative identification of good prognosis tumors using MRI will allow stratification of patients and better targeting of preoperative therapy. This study confirms the ability of MRI to select patients who are likely to have a good outcome with primary surgery alone.

© 2011 Lippincott Williams & Wilkins, Inc.


Article Tools


Article Level Metrics