Capecitabine versus 5-fluorouracil (5-FU)–based (neo)adjuvant chemoradiotherapy (CRT) for locally advanced rectal cancer (LARC): Long-term results of a randomized, phase III trial

Reporter: J. Nicholas Lukens, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 5, 2011

Presenter: Ralf Hofheinz, MD
Presenter's Institution: Day Treatment Centre at the Interdisciplinary Tumour Centre Mannheim, Universitätsmedizin Mannheim, Mannheim, Germany


  • 5-FU based chemoradiation prior to total mesorectal excision (TME) followed by additional 5-FU chemotherapy is the standard of care for locally advanced rectal cancer.
    • The German rectal trial demonstrated a benefit to preoperative chemoradiotherapy with infusional 5-FU and 50.4 Gy, compared with the same regimen given post-operatively. Benefits to preoperative therapy in the German trial included higher rates of sphincter-sparing procedures, decreased acute and long term toxicity, and improved local control rates.
    • The results of the German trial were published during the accrual of the current trial, which explains why patients were accrued in the adjuvant and then neo-adjuvant setting.
  • The oral fluoropyrimidine Capecitabine has been shown to be as effective as 5-FU/leucovorin in the adjuvant treatment of Stage III colon cancer, and non-inferior to infusional 5-FU in combination with Oxaliplatin as first line treatment for metastatic colon cancer.
  • Furthermore, Capecitabine has radiosensitizing properties that make it an attractive alternative to 5-FU in the treatment of rectal cancer.
  • Whether Capecitabine can replace infusional 5-FU for locally advanced rectal cancer has not been established in a randomized Phase III trial.
  • This study was conducted as a non-inferiority phase III trial investigating chemoradiation with Capecitabine in comparison with 5-FU in both the pre-operative and post-operative setting.


  • The study enrolled patients at least 18 years of age with locally advanced rectal cancer, defined as International Union Against Cancer (UICC) Stages II or III (T3/4 or N+) disease, 0-16 cm from the anal verge. Staging was by endoscopic ultrasound (EUS).
  • The study was designed as a two-arm, two-strata, randomized phase-III trial.
  • The 2 arms of the study were capecitabine (Arm A) and 5-FU (Arm B), and the 2 strata were adjuvant (S1) and neo-adjuvant (S2).
  • The dose of Capecitabine given concurrent with RT was 1,650 mg/m2, and the dose of adjuvant Capecitabine was 2,500 mg/m2. The dose of 5-FU concurrent with RT was 225 mg/m2 continuous infusion daily in the adjuvant setting, whereas in the neoadjuvant setting it was 1000 mg/m2 on days 1-5 and days 29-33. Bolus 5-FU dose was 500 mg/m2.
  • The 4 regimens were as follows:
    • Arm A, Stratum 1 (Capecitabine, adjuvant): TME -> 2 cycles of Capecitabine -> 50.4 Gy RT with concurrent Capecitabine -> 3 additional cycles of Capecitabine.
    • Arm A, Stratum 2 (Capecitabine, neo-adjuvant): 50.4 Gy RT with concurrent Capecitabine -> TME -> 5 cycles of adjuvant Capecitabine
    • Arm B, Stratum 1 (5-FU, adjuvant): TME -> 2 cycle 5-FU -> 50.4 Gy RT with concurrent 5-FU -> 2 additional cycles 5-FU
    • Arm B, Stratum 2 (5-FU, neo-adjuvant): 50.4 Gy RT with concurrent 5-FU -> TME -> 4 cycles 5-FU.
  • The primary endpoint was 5-year overall survival, and this was designed as a non-inferiority study. Secondary endpoints were disease-free survival and toxicity.


  • 230 patients were enrolled in the adjuvant stratum, and 161 in the neo-adjuvant stratum.
  • The Capecitabine and 5-FU arms were well balanced, with about 80% of patients having T3 or T4 tumors, and 60% having node positive disease in each arm.
  • The percentage of patients receiving all scheduled cycles of chemotherapy in the adjuvant setting was high in both arms, approximately 80%; in contrast, after neo-adjuvant chemoradiation and surgery, 35% of patients did not start post-operative adjuvant therapy, and less than half received all planned cycles.
  • Toxicity:
    • Leukopenia was more common in the 5-FU arm.
    • GI toxicity was greater in the Capecitabine arm, with proctitis developing in 31 Capecitabine patients and 10 5-FU patients, and diarrhea being more pronounced in Capecitabine patients during concurrent chemoRT.
    • Hand foot skin reaction was significantly more common in the Capecitabine arm, as was fatigue.
  • Focusing on the neo-adjuvant stratum, the rate of sphincter-sparing surgery was similar between the 2 arms.
  • More patients in the Capecitabine arm had a pathologic complete response (pCR) rate compared to 5-FU arm: 14% versus 5%, although this did not reach statistical significance (p = 0.16)
  • There was a trend towards improved T and N downstaging in patients receiving Capecitabine, with less ypN+ tumors (p=0.09).
  • The rates of local recurrence were low for both arms, 6% for Capecitabine and 7% for 5-FU.
  • There were fewer distant metastases in the Capecitabine arm, 19% versus 28% (p=0.04).
  • Disease free survival (secondary endpoint) at 3 years favored the Capecitabine arm: 75% versus 67% (p = 0.035)
  • For overall survival (primary endpoint) at 5 years, Capecitabine was non-inferior: 76% versus 67% (p = 0.001 for non-inferiority), and test for superiority was borderline significant (p = 0.053).
  • Hand foot skin (HFS) reaction has been shown to be a marker of good prognosis, and in this study, patients developing any HFS reaction to Capecitabine had better 3-year DFS and 5-year OS compared to those who did not.

Authors' conclusions

  • Both 5-FU and capecitabine are well tolerated, with GI toxicity and HFS reaction more common with capecitabine.
  • In the neo-adjuvant setting, capecitabine led to increased surgical downstaging and a higher pCR rate, although this did not reach significance.
  • Capecitabine was non-inferior to 5-FU with regard to 5-year OS, and showed a strong trend towards improved 5-year OS; 3 year DSF was significantly better with capecitabine.
  • The development of HFS reaction has prognostic significance.
  • Capecitabine may replace 5-FU in the peri-operative treatment of locally advanced rectal cancer.

Clinical/Scientific Implications

  • This study firmly establishes that Capecitabine is at least equivalent, if not slightly superior, to 5-FU in the neo-adjuvant and adjuvant treatment of locally advanced rectal cancer.
    • In particular, it appears to lead to improved downstaging (in the neo-adjuvant setting), fewer distant metastases, and improved DFS relative to 5-FU.
  • In practice, many clinical trials subsequent to this incorporate capecitabine rather than 5-FU in addition to other chemotherapy agents, so it is unclear whether this study will have a dramatic impact on clinical practice.
  • Capecitabine leads to increased GI toxicity relative to 5-FU, but not to the extent that it compromises the ability of patients to complete their planned cycles of chemotherapy, any more so than with 5-FU.
    • Nevertheless, in the neo-adjuvant setting, a significant number of patients enrolled in this study did not receive planned adjuvant chemotherapy (capecitabine or 5-FU).
  • This study confirms that the development of hand-foot skin reaction has prognostic significance, and raises the question of whether chemotherapy dose should be adjusted to produce this surrogate clinical marker.