Tehran University Medical Journal (Jan 2016)

How accurate is MRI to predict response to neoadjuvant chemoradiation in locally advanced rectal cancer?

  • Afsaneh Alikhasi,
  • Monir Sadat Mirai Ashtiani,
  • Farshid Farhan,
  • Mehdi Aghili,
  • Mohammad Sadegh Fazeli,
  • Mohammad Babaei,
  • Afsaneh Maddah-Safai,
  • Peiman Haddad

Journal volume & issue
Vol. 73, no. 10
pp. 700 – 708

Abstract

Read online

Background: This study investigated compatibility between post chemoradiation magnetic resonance images and histologic findings after operation and chemoradiation in patients with locally advanced rectal cancer. Methods: In this prospective study, 63 patients referred to Cancer Institute of Emam Khomeini Hospital, Tehran, Iran, from October 2011 to October 2013 with locally advanced rectal cancer receiving neoadjuvant chemoradiation (50.4 Gy external beam radiation with concomitant capecitabine 825 mg/m2 PO twice a day with or without 60 mg/m2 oxaliplatin weekly). Patients had an MRI before chemoradiation and MRI assessment were used to identify Tumor (T) and lymph node (N) staging by an experienced radiologist. Patients were recommended to repeat MRI after surgery but it was not obligatory. Findings of post chemoradiation MRI and histopathologic reports were compared. Downstaging was defined as at least one stage decrease in T or N in histopathologic report comparing to their first MRI, on condition of no sign of disease progression. Results: 32 patients (50.79%) had T downstaging and 36 of them (57.14%) showed N downstaging: none had disease progression. In this study MRI had an accuracy of 55.5% for rectal tumor (T) restaging after chemoradiation comparing to pathology. MRI sensitivity for T restaging was 33.3% to 83.3%. There was a higher possibility to have errors in restaging of T1-2 stages. Specificity of MRI for T restaging was higher than its sensitivity, 66.6%. In this study lymph node involvement (N) was determined according to morphology and size. MRI has an accuracy of 42.8% for detecting lymph node involvement. Its sensitivity and specificity for N restaging were 50% and 66.6% respectively. All patients had MRI before chemoradiation, although 21 of them repeated MRI after chemoradiation since it was not mandatory. 19 of these 21 patients underwent surgery. Conclusion: Although MRI is a suitable imaging for staging locally advanced rectal cancer its use for restaging after chemoradiation is under question. According to this study, MRI accuracy rates for both T and N restaging were below the rates of previous studies.

Keywords