The TME Trial after a Median Follow-up of 11 Years

Reviewer: Lara Bonner Millar, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: November 1, 2010

Authors: C. A. Marijnen1, W. van Gijn1, I. D. Nagtegaal2, E. Klein Kranenbarg1, H. Putter1, T. Wiggers3, H. J. T. Rutten4, L. Pahlman5,B. Glimelius5, C. J. H. van de Velde1
1 Leiden University Medical Center, Leiden, Netherlands
2 University Medical Center St. Radboud, Nijmegen, Netherlands
3 University Medical Center Groningen, Groningen, Netherlands
4 Catharina Hospital, Eindhoven, Netherlands
5 Uppsala University Hospital, Uppsala, Sweden
Author: Lara Bonner Millar


  • Local recurrence is a major problem in rectal cancer treatment, and may contribute to both decreased overall survival and quality of life.
  • The Swedish rectal trial attempted to improve on surgery-only results by randomizing patient to hypofractionated radiation therapy with a pre-operative 5 Gray (Gy) x 5 fraction radiation scheme vs surgery alone.
  • The older surgical technique for rectal cancer, employed in the Swedish trial, involved a blunt rectal dissection. This approach frequently left gross disease or positive margins.
  • The TME (total mesorectal excision) procedure was popularized by Drs. Ryald and Heald in the 1980s and 1990s and involves sharp dissection along the fascial plane with removal of perirectal tissue en mass.
  • TME appears to improve patient outcomes when compared to incomplete TME or lesser surgery. The Dutch study built on the Swedish study by evaluating the role of RT in the context of TME, an oncologically superior surgery. Long term results of the Dutch study are presented here.

Materials and Methods

  • 1861 patients with resectable rectal cancer were randomized to 1) pre-op 5 x 5 Gy radiotherapy followed by TME or 2)TME alone. No chemotherapy was allowed.
  • There was no age restriction.
  • Surgery, radiotherapy, and pathologic examination were standardized.
    • RT delivery: No CT planning; 3 or 4 field technique with conventional planning
    • Total treatment time did not exceed 10 days—patients went to surgery within 3-5 days of completing RT
    • If surgery-only patients had + margins (</=1mm), then mandatory post-op RT was given
  • The primary endpoint was local control. The analysis was done in intent-to-treat fashion.
  • The authors report findings after 10 years of follow-up.


  • Local recurrence was 5% in the irradiated group and 11% in the surgery alone group (p <0.001).
    • The most common site of local failure was pre-sacral
    • For every 100 pts treated, pre-op RT prevented 6 local recurrences; therefore for the group as a whole, the NNT to prevent one LR was 17.
    • The NNT decreased with increase in stage; for stage I pts, the NNT is 38.
  • Overall recurrence was significantly lower in the irradiated group (28.8% vs. 33.6%, p = 0.042), but there was no difference in overall survival demonstrated between the two groups.
  • In subgroup analysis, for patients with a negative circumferential resection margin (CRM), both freedom from local recurrence and cancer specific survival were higher in the irradiated group. There was no impact on overall survival.
    • LR 3% ( pre-op RT) vs 8.7% (no RT).
    • 10 year over all survival 56.4% (pre-op RT) vs 57% (no RT).
  • Other subgroup analyses found radiotherapy to reduce local recurrence in lymph node positive (Stage III) patients and also for those with tumors more than 5 cm from the anal verge.
    • For lymph node positive (Stage III) patients with a negative CRM, preoperative radiotherapy improved 10 year survival from 41% to 51%, p = 0.02.
    • For low rectal cancers (0-5 cm from the verge), NNT= 50; for mid-rectal cancer (5-10 cm from verge), NNT =10, and for high rectal cancers (10-15 cm from verge) NNT = 20 to avoid one local recurrence. RT thus appeared to most effective in treatment of mid-rectal cancers.
  • Higher incontinence rates were observed at 5 years in the group receiving RT (62% for RT vs 38% for TME alone), as well as higher bleeding and mucous discharge rates

Author's Conclusions

  • Preoperative short-term RT for resectable rectal cancer reduces the local recurrence rate by more than 50%.
  • In subgroup analyses, radiotherapy seems most effective in patients with a negative CRM, patients with TNM stage III and patients with a tumor height greater than 5 cm.
  • For the group as a whole, there was no overall survival benefit, with addition of RT; however, an overall survival benefit was observed for stage III (lymph node positive) patients on sub-group analysis.

Clinical Implications

  • This is a landmark trial whose results have been reproduced (MRC 07 Trial: 3 yr results of pre-op RT 5x5 then TME vs TME followed by 45 Gy if + margin had similar local control)
  • A next step in the research process may be to attempt to reduce overtreatment: 31% of patients enrolled in this trial were stage I
    • In the US, these Stage I pts would not get radiation
    • Use of CT scans, endoscopic ultrasound , and/or MRI as part of pre-operative evaluation may enable us to better select patients that would benefit from addition of RT.
  • There was a very short interval between RT and surgery, so patients did not have time to be downstaged by the RT. The Lyon R 90-01 trial randomized clinical T2-3 pts to RT followed by surgery two weeks later vs. surgery after 6-8 weeks and found a higher path CR rate and sphincter preservation rate with waiting longer before surgery.
  • Hypofractionation raises concern about late effects- there was a significantly higher incontinence rate at 5 years (62% for RT vs 38% for TME alone), as well as higher bleeding and mucous discharge rates
    • Similarly, the Swedish rectal trial also found increased late toxicity in the RT group. Bowel obstruction requiring surgery occurred in 5.5% of patients treated with surgery alone vs 13.9% for the pre-op 5 Gy x 5 group
  • No chemotherapy was given; can we infer that distant recurrence and overall survival would have been affected had chemoRT been delivered rather than RT alone? This question was not addressed by the current study.
  • We do not have evidence that short course RT should replace long course neoadjuvant RT with concurrent chemo for pts with locally advanced disease
    • Other short course trials are currently enrolling (TROG, Berlin, Stockholm), and may provide answers to outstanding questions.