Skip to main content
NCBI home page
BMJ Clinical Evidence logoLink to BMJ Clinical Evidence
. 2010 Apr 27;2010:0401.

Colorectal cancer treatment

Asad Qureshi 1,#, Anjali Verma 2,#, Paul Ross 3,#, David Landau 4,#
PMCID: PMC2907599  PMID: 21718569

Abstract

Introduction

Colorectal cancer is the third most common malignancy in the developed countries, and about a quarter of people present with intestinal obstruction or perforation. Risk factors for colorectal cancer are mainly dietary and genetic. Overall 5-year survival is about 50%, with half of people having surgery experiencing recurrence of the disease.

Methods and outcomes

We conducted a systematic review and aimed to answer the following clinical question: What are the effects of treatments for colorectal cancer? We searched: Medline, Embase, The Cochrane Library, and other important databases up to August 2008 (Clinical Evidence reviews are updated periodically, please check our website for the most up-to-date version of this review). We included harms alerts from relevant organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA).

Results

We found 57 systematic reviews, RCTs, or observational studies that met our inclusion criteria. We performed a GRADE evaluation of the quality of evidence for interventions.

Conclusions

In this systematic review we present information relating to the effectiveness and safety of the following interventions: adjuvant systemic chemotherapy, preoperative radiotherapy, and routine intensive follow-up.

Key Points

Colorectal cancer is the third most common malignancy in the developed world, and about a fifth of people present with intestinal obstruction or perforation.

  • Risk factors for colorectal cancer are mainly dietary and genetic.

  • Overall 5-year survival is about 50%, with half of people having surgery experiencing recurrence of the disease.

Adjuvant systemic chemotherapy reduces mortality compared with surgery alone in people who have Dukes' C colorectal cancer. We don't know whether adjuvant systemic chemotherapy improves mortality compared with surgery alone in people with Dukes' B colorectal cancer.

  • It has been suggested that people with high-risk Dukes' B may derive some benefit with adjuvant systemic chemotherapy compared with surgery alone. However, we found no direct evidence on this group.

  • We found some evidence that oral fluoropyrimidines (with or without leucovorin) may be as effective as intravenous fluorouracil (with or without leucovorin) regimens at reducing mortality.

  • The addition of oxaliplatin to fluorouracil plus leucovorin improves disease-free survival at 3 and 4 years compared with fluorouracil plus leucovorin alone in people with Dukes' B or C colon cancer.

  • Adding irinotecan to fluorouracil plus leucovorin does not reduce mortality any more than fluorouracil plus leucovorin alone in people with Dukes' C colorectal cancer and increases toxic effects.

Preoperative radiotherapy may modestly reduce local tumour recurrence and mortality compared with surgery alone in people with rectal cancer.

  • Preoperative radiotherapy may reduce local recurrence compared with postoperative radiotherapy. There may be no difference in overall survival between preoperative and postoperative radiotherapy.

Routine intensive follow-up may reduce the time to detection of recurrence and may increase survival compared with less-intensive follow-up in people with colorectal cancer.

About this condition

Definition

Colorectal cancer is a malignant neoplasm arising from the lining (mucosa) of the large intestine (colon and rectum). About two-thirds of colorectal cancers occur in the colon and the remainder in the rectum. Colorectal cancer may be categorised as Dukes' stage A, B, or C. Some studies in this review have reported staging of colorectal cancer as stage I, II, or III. Stage I corresponds to Dukes' stage A–B1; stage II corresponds to Dukes' stage B2–B3; stage III corresponds to Dukes' stage C. In this review we have included people with Dukes' A, B, and C (stage I, II, III) and excluded people with distant metastatic disease (Dukes' stage D or stage IV).

Incidence/ Prevalence

Colorectal cancer is the third most common cancer in the UK after breast and lung. There are about 36,000 new cases per year in the UK. It is the second highest cause of cancer death in the UK after lung cancer, with 16,000 deaths per year. Between 1979 and 1999 the incidence in men in the UK rose slowly before beginning to fall. Over the same period the incidence in women changed very little. In contrast to incidence trends, mortality has been falling since the early 1990s. Between 1997 and 2006 bowel cancer age-standardised mortality in the UK fell by 17%. The presentation of colorectal cancer can vary widely. Cancers of the proximal colon can present with weight loss and anaemia. Cancers of the distal colon and rectum are more likely to present with bleeding and altered bowel habit. In the UK, approximately one fifth of patients present with acute intestinal obstruction.

Aetiology/ Risk factors

Bowel cancer is generally more common in populations with a "Westernised" diet. Several studies have shown an increased risk associated with increased consumption of red meat and alcohol. Conversely, a diet high in fibre, fresh fruit, vegetables, and fish has been shown to reduce the risk of colorectal cancer. A minority of cases are directly associated with known genetic risk factors. However, people with a first-degree relative with bowel cancer are at twice the risk of developing it themselves.

Prognosis

As a result of earlier diagnosis and better treatment 5-year overall survival rose from about 20% in the early 1970s to approximately 50% in early 2000. Surgery is undertaken in over 80% of people, but about half experience recurrence of their cancer.

Aims of intervention

To reduce morbidity (e.g., bowel obstruction or perforation) and mortality associated with the tumour; to minimise adverse effects of treatment (e.g., avoiding permanent stoma by restoring intestinal continuity); to maximise quality of life.

Outcomes

Survival; disease-free survival; recurrence (local, metastasis); adverse effects of treatment.

Methods

Clinical Evidence search and appraisal August 2008. The following databases were used to identify studies for this systematic review: Medline 1966 to August 2008, Embase 1980 to August 2008, and The Cochrane Database of Systematic Reviews and Cochrane Central Register of Controlled Clinical Trials 2008, Issue 3 (1966 to date of issue). An additional search was carried out of the NHS Centre for Reviews and Dissemination (CRD) — for Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment (HTA). We also searched for retractions of studies included in the review. Abstracts of the studies retrieved from the initial search were assessed by an information specialist. Selected studies were then sent to the contributor for additional assessment, using pre-determined criteria to identify relevant studies. Study design criteria for inclusion in this review were: published systematic reviews of RCTs and RCTs in any language, at least single blinded, and containing more than 20 individuals of whom more than 80% were followed up. There was no minimum length of follow-up required to include studies. We excluded all studies described as "open", "open label", or not blinded unless blinding was impossible. We included systematic reviews of RCTs and RCTs where harms of an included intervention were studied applying the same study design criteria for inclusion as we did for benefits. In addition, we use a regular surveillance protocol to capture harms alerts from organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA), which are added to the reviews as required. To aid readability of the numerical data in our reviews, we round many percentages to the nearest whole number. Readers should be aware of this when relating percentages to summary statistics such as relative risks (RRs) and odds ratios (ORs). We have performed a GRADE evaluation of the quality of evidence for interventions included in this review (see table ). As part of this evaluation, we have selected outcomes on which to report, and have categorised the evidence relating to those outcomes as being of high, moderate, low, or very low, quality. In selecting outcomes, we have chosen clinical outcomes that are important to patients and doctors, rather than proxy or laboratory-based outcomes. The categorisation of the quality of the evidence (into high, moderate, low, or very low) is not necessarily a reflection of the overall methodological quality of any individual study providing evidence on that outcome. Rather, our categorisation relates directly to the evidence reported on our chosen outcome in our defined population of interest. This population and outcome of choice may represent a small subset of the total trial outcomes and populations. For further details of how we perform the GRADE evaluation and the scoring system we use, please see our website (www.clinicalevidence.com).

Table 1.

GRADE evaluation of interventions for colorectal cancer treatment

Important outcomes Mortality, recurrence (recurrence [local, metastasis], disease-free survival), adverse effects
Number of studies (participants) Outcome Comparison Type of evidence Quality Consistency Directness Effect size GRADE Comment
What are the effects of treatments for colorectal cancer?
At least 45 (at least 18,703) Mortality Adjuvant systemic chemotherapy v no adjuvant systemic chemotherapy 4 –1 0 –1 0 Low Quality point deducted for weak methods in some RCTs. Directness point deducted for inclusion of old regimens no longer in use affecting generalisability of results
At least 35 (at least 15,166) Recurrence (recurrence [local, metastasis], disease-free survival) Adjuvant systemic chemotherapy v no adjuvant systemic chemotherapy 4 –1 0 –1 0 Low Quality point deducted for weak methods in some RCTs. Directness point deducted for inclusion of old regimens no longer in use affecting generalisability of results
3 (at least 5840) Mortality Fluorouracil plus leucovorin plus levamisole v fluorouracil plus leucovorin alone 4 –1 0 0 0 Moderate Quality point deducted for no direct statistical analysis between groups in 1 RCT
2 (at least 4863) Recurrence (recurrence [local, metastasis], disease-free survival) Fluorouracil plus leucovorin plus levamisole v fluorouracil plus leucovorin alone 4 0 0 0 0 High
1 (4927) Mortality Low-dose leucovorin plus fluorouracil v high-dose leucovorin plus fluorouracil 4 0 0 0 0 High
1 (4927) Recurrence (recurrence [local, metastasis], disease-free survival) Low-dose leucovorin plus fluorouracil v high-dose leucovorin plus fluorouracil 4 0 0 0 0 High
8 (6962) Mortality Fluorouracil plus levamisole v fluorouracil plus leucovorin or v fluorouracil plus leucovorin plus levamisole 4 –1 –1 0 0 Low Quality point deducted for weak methods (incomplete reporting of results, no direct analysis between groups in 1 RCT, combing 2 groups in analysis in 1 RCT, use of co-intervention). Consistency point deducted for conflicting results
7 (6243) Recurrence (recurrence [local, metastasis], disease-free survival) Fluorouracil plus levamisole v fluorouracil plus leucovorin or v fluorouracil plus leucovorin plus levamisole 4 –1 –1 0 0 Low Quality point deducted for weak methods (incomplete reporting of results, combing 2 groups in analysis in 1 RCT, use of co-intervention). Consistency point deducted for conflicting results
1 (878) Mortality Fluorouracil plus leucovorin plus high-dose levamisole v fluorouracil plus leucovorin plus standard-dose levamisole 4 0 0 –1 0 Moderate Directness point deducted for alteration of regimen during course of trial
1 (878) Recurrence (recurrence [local, metastasis], disease-free survival) Fluorouracil plus leucovorin plus high-dose levamisole v fluorouracil plus leucovorin plus standard-dose levamisole 4 0 0 –1 0 Moderate Directness point deducted for alteration of regimen during course of trial
1 (1081) Mortality Fluorouracil plus leucovorin v fluorouracil plus lomustine plus vincristine (MOF) 4 0 0 0 0 High
1 (1081) Recurrence (recurrence [local, metastasis], disease-free survival) Fluorouracil plus leucovorin v fluorouracil plus lomustine plus vincristine (MOF) 4 0 0 0 0 High
4 (4275) Mortality Oral fluoropyrimidines v fluorouracil 4 –1 –1 0 0 Low Quality point deducted for incomplete reporting of results (in 2 RCTs). Consistency point deducted for conflicting results (subgroup analysis in rectal cancer alone and colon cancer alone)
5 (4441) Recurrence (recurrence [local, metastasis], disease-free survival) Oral fluoropyrimidines v fluorouracil 4 –1 –1 0 0 Low Quality point deducted for incomplete reporting of results (in 2 RCTs). Consistency point deducted for conflicting results (subgroup analysis in rectal cancer alone and colon cancer alone)
1 (172) Mortality Mitomycin C v oral fluoropyrimidines plus mitomycin C 4 –2 0 0 0 Low Quality points deducted for sparse data and unclear randomisation
2 (4738) Recurrence (recurrence [local, metastasis], disease-free survival) Fluorouracil plus leucovorin plus oxaliplatin v fluorouracil plus leucovorin alone 4 0 0 0 0 High
1 (1264) Mortality Fluorouracil plus leucovorin plus irinotecan v fluorouracil plus leucovorin alone 4 0 0 0 0 High
1 (1264) Recurrence (recurrence [local, metastasis], disease-free survival) Fluorouracil plus leucovorin plus irinotecan v fluorouracil plus leucovorin alone 4 0 0 0 0 High
1 (1945) Mortality Sequential methotrexate regimen v fluorouracil regimen 4 –1 0 0 0 Moderate Quality point deducted for change of intervention during course of trial (levamisole)
1 (1945) Recurrence (recurrence [local, metastasis], disease-free survival) Sequential methotrexate regimen v fluorouracil regimen 4 –1 0 0 0 Moderate Quality point deducted for change of intervention during course of trial (levamisole)
4 (2986) Mortality Infusion of fluorouracil v bolus fluorouracil 4 –1 0 –1 0 Low Quality point deducted for early termination of 2 RCTs. Directness point deducted diverse regimens among RCTs
4 (2986) Recurrence (recurrence [local, metastasis], disease-free survival) Infusion of fluorouracil v bolus fluorouracil 4 –1 0 –1 0 Low Quality point deducted for early termination of 2 RCTs. Directness point deducted diverse regimens among RCTs
20 (8163) Mortality Preoperative radiotherapy plus surgery v surgery alone 4 0 0 –2 0 Low Directness points deducted for clinical heterogeneity among RCTs (non-contemporary techniques) and significance dependant on analysis (whether published data or individual patient data)
At least 13 (at least 7464) Recurrence (recurrence [local, metastasis], disease-free survival) Preoperative radiotherapy plus surgery v surgery alone 4 0 –1 0 0 Moderate Consistency point deducted for statistical heterogeneity among RCTs in analysis
3 (1320) Mortality Preoperative radiotherapy v postoperative radiotherapy 4 –1 0 –2 0 Very low Quality point deducted for incomplete reporting of results. Directness points deducted for difference between groups in the chemotherapy regimens used, and in the proportion of people who received the allocated treatment in 1 RCT
3 (1320) Recurrence (recurrence [local, metastasis], disease-free survival) Preoperative radiotherapy v postoperative radiotherapy 4 –1 0 –2 0 Very low Quality point deducted for incomplete reporting of results. Directness points deducted for difference between groups in the chemotherapy regimens used, and in the proportion of people who received the allocated treatment in 1 RCT
9 (at least 2923) Mortality Intensive follow-up v less-intensive follow-up 4 0 0 –1 0 Moderate Directness point deducted for wide variation in follow-up regimens between RCTs
9 (at least 2923) Recurrence (recurrence [local, metastasis], disease-free survival) Intensive follow-up v less-intensive follow-up 4 0 –1 –1 0 Low Consistency point deducted for statistical heterogeneity. Directness point deducted for wide variation in follow-up regimens between RCTs
Open in a new tab

Type of evidence: 4 = RCT. Consistency: similarity of results across studies. Directness: generalisability of population or outcomes. Effect size: based on relative risk or odds ratio.

Glossary

Dukes' classification

Dukes' original classification of the pathological stages of carcinoma of the colon and rectum includes three stages: A, limited to mucosa and submucosa; B, penetration of the entire bowel wall and serosa or pericolic fat; C, stages A and B, and invasion into the regional draining lymph node system. More recently, stage D has been proposed to classify people with advanced and widespread regional involvement (metastasis).

High-quality evidence

Further research is very unlikely to change our confidence in the estimate of effect.

Low-quality evidence

Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.

Moderate-quality evidence

Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.

Very low-quality evidence

Any estimate of effect is very uncertain.

Disclaimer

The information contained in this publication is intended for medical professionals. Categories presented in Clinical Evidence indicate a judgement about the strength of the evidence available to our contributors prior to publication and the relevant importance of benefit and harms. We rely on our contributors to confirm the accuracy of the information presented and to adhere to describe accepted practices. Readers should be aware that professionals in the field may have different opinions. Because of this and regular advances in medical research we strongly recommend that readers' independently verify specified treatments and drugs including manufacturers' guidance. Also, the categories do not indicate whether a particular treatment is generally appropriate or whether it is suitable for a particular individual. Ultimately it is the readers' responsibility to make their own professional judgements, so to appropriately advise and treat their patients.To the fullest extent permitted by law, BMJ Publishing Group Limited and its editors are not responsible for any losses, injury or damage caused to any person or property (including under contract, by negligence, products liability or otherwise) whether they be direct or indirect, special, incidental or consequential, resulting from the application of the information in this publication.

Contributor Information

Asad Qureshi, Guy's & St. Thomas' NHS Foundation Trust, London, UK.

Anjali Verma, Maidstone & Tunbridge Wells NHS Trust, Maidstone, UK.

Paul Ross, Guy's & St. Thomas' NHS Foundation Trust, London, UK.

David Landau, Guy's & St. Thomas' NHS Foundation Trust, London, UK.

References

  • 1.Dukes CE. The classification of cancer of the rectum. J Pathol Bacteriol 1932;35:323–332. [Google Scholar]
  • 2.Office for National Statistics. Cancer statistics registrations: registrations of cancer diagnosed in 2005, England. ONS 2008; series MB1 no. 36. http://www.statistics.gov.uk/statbase/Product.asp?vlnk=8843 (last accessed 5 February 2010). [Google Scholar]
  • 3.Welsh Cancer Intelligence and Surveillance Unit. 2008. http://www.wales.nhs.uk/sites3/home.cfm?OrgID=242 (last accessed 5 February 2010). [Google Scholar]
  • 4.ISD Scotland. Colorectal cancer. 2009. http://www.isdscotland.org/isd/1425.html#Colorectal (last accessed 5 February 2010). [Google Scholar]
  • 5.Northern Ireland Cancer Registry. 2006. http://www.qub.ac.uk/research-centres/nicr/FileStore/PDF/Filetoupload,148094,en.pdf (last accessed 5 February 2010). [Google Scholar]
  • 6.Office for National Statistics. Cancer registration statistics. ONS 2007; series MB1 No. 38. http://www.statistics.gov.uk/statbase/Product.asp?vlnk=8843 (last accessed 5 February 2010). [Google Scholar]
  • 7.NHS Centre for Reviews and Dissemination. The management of colorectal cancer 1997. Eff Health Care 1997;3. [Google Scholar]
  • 8.Koehne CH, Midgley R, Seymour M, et al. Advanced colorectal cancer: which regimes should we recommend? Ann Oncol 1999;10:877–883. [DOI] [PubMed] [Google Scholar]
  • 9.Norat T, Bingham S, Ferrari P, et al. Meat, fish, and colorectal cancer risk: the European Prospective Investigation into cancer and nutrition. J Natl Cancer Inst 2005;97:906–916. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Larsson SC, Wolk A. Meat consumption and risk of colorectal cancer: a meta-analysis of prospective studies. Int J Cancer 2006;119:2657–2664. [DOI] [PubMed] [Google Scholar]
  • 11.Norat T, Lukanova A, Ferrari P, et al. Meat consumption and colorectal cancer risk: dose-response meta-analysis of epidemiological studies. Int J Cancer 2002;98:241–256. [DOI] [PubMed] [Google Scholar]
  • 12.Cho E, Smith-Warner SA, Ritz J, et al. Alcohol intake and colorectal cancer: a pooled analysis of 8 cohort studies. Ann Intern Med 2004;140:603–613. [DOI] [PubMed] [Google Scholar]
  • 13.Park Y, Hunter DJ, Spiegelman D, et al. Dietary fiber intake and risk of colorectal cancer: a pooled analysis of prospective cohort studies. JAMA 2005;294:2849–2857. [DOI] [PubMed] [Google Scholar]
  • 14.Koushik A, Hunter DJ, Spiegelman D, et al. Fruits, vegetables and colon cancer risk in a pooled analysis of 14 cohort studies. J Natl Cancer Inst 1996;99:1471–1483. [DOI] [PubMed] [Google Scholar]
  • 15.Coleman MP, Rachet B, Woods LM, et al. Trends in socioeconomic inequalities in cancer survival in England and Wales up to 2001. Br J Cancer 2004;90:1367–1373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Dube S, Heyen F, Jenicek M. Adjuvant chemotherapy in colorectal carcinoma. Results of a meta analysis. Dis Colon Rectum 1997;40:35–41. Search date 1993. [DOI] [PubMed] [Google Scholar]
  • 17.Sargent DJ, Goldberg RM, Jacobson SD, et al. A pooled analysis of adjuvant chemotherapy for resected colon cancer in elderly patients. N Engl J Med 2001;345:1091–1097. Search date not reported. [DOI] [PubMed] [Google Scholar]
  • 18.Benson AB 3rd, Schrag D, Somerfield MR, et al. American Society of Clinical Oncology recommendations on adjuvant chemotherapy for stage II colon cancer. J Clin Oncology 2004;22:3408–3419. [DOI] [PubMed] [Google Scholar]
  • 19.Figueredo A, Charette ML, Maroun J, et al. Adjuvant therapy for stage II colon cancer: a systematic review from the Cancer Care Ontario Program in evidence-based care's gastrointestinal cancer disease site group. J Clin Oncol 2004;22:3395–3407. [DOI] [PubMed] [Google Scholar]
  • 20.Figueredo A, Coombes ME, Mukherjee S. Adjuvant therapy for completely resected stage II colon cancer. In: The Cochrane Library, Issue 3, 2008. Chichester, UK: John Wiley & Sons, Ltd. Search date 2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Sakamoto J, Hamada C, Yoshida S, et al. An individual patient data meta-analysis of adjuvant therapy with uracil-tegafur (UFT) in patients with curatively resected rectal cancer. Br J Cancer 2007;96:1170–1177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Smith RE, Colangelo L, Wieand HS, et al. Randomized trial of adjuvant therapy in colon carcinoma: 10-year results of NSABP protocol C-01. J Natl Cancer Inst 2004;96:1128–1132. [DOI] [PubMed] [Google Scholar]
  • 23.Gray R, Barnwell J, McConkey C. QUASAR Collaborative Group. Adjuvant chemotherapy versus observation in patients with colorectal cancer: a randomised study. Lancet 2007;370:2020–2029. [DOI] [PubMed] [Google Scholar]
  • 24.Hafstrom L, Domellof L, Rudenstam CM, et al. Adjuvant chemotherapy with 5-fluorouracil, vincristine and CCNU for patients with Dukes' C colorectal cancer. The Swedish Gastrointestinal Tumour Adjuvant Therapy Group. Br J Surg 1990;77:1345–1348. [DOI] [PubMed] [Google Scholar]
  • 25.Ito K, Yamaguchi A, Miura K, et al. Oral adjuvant chemotherapy with carmofur (HCFU) for colorectal cancer: five-year follow-up. Tokai HCFU Study Group - third study on colorectal cancer. J Surg Oncol 1996;63:107–111. [DOI] [PubMed] [Google Scholar]
  • 26.O'Connell MJ, Mailliard JA, Kahn MJ, et al. Controlled trial of fluorouracil and low-dose leucovorin given for 6 months as postoperative adjuvant therapy for colon cancer. J Clin Oncol 1997;15:246–250. [DOI] [PubMed] [Google Scholar]
  • 27.Schippinger W, Samonigg H, Schaberl-Moser R, et al. A prospective randomised phase III trial of adjuvant chemotherapy with 5-fluorouracil and leucovorin in patients with stage II colon cancer. Br J Cancer 2007;97:1021–1027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.QUASAR Collaborative Group. Comparison of fluorouracil with additional levamisole, higher dose folinic acid or both as adjuvant chemotherapy for colorectal cancer: a randomised trial. Lancet 2000;355:1588–1596. [PubMed] [Google Scholar]
  • 29.De PlacidoS, Lopez M, Carlomagno C, et al. Modulation of 5-fluorouracil as adjuvant systemic chemotherapy in colorectal cancer: the IGCS-COL multicentre, randomised, phase III study. Br J Cancer 2005;93:896–904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Wolmark N, Rockette H, Mamounas E, et al. Clinical trial to assess the relative efficacy of fluorouracil and leucovorin, fluorouracil and levamisole, and fluorouracil, leucovorin, and levamisole in patients with Dukes' B and C carcinoma of the colon: results from National Surgical Adjuvant Breast and Bowel Project C-04. J Clin Oncol 1999;17:3553–3559. [DOI] [PubMed] [Google Scholar]
  • 31.Arkenau HT, Bermann A, Rettig K, et al. 5-Fluorouracil plus leucovorin is an effective adjuvant chemotherapy in curatively resected stage III colon cancer: long-term follow-up results of the adjCCA-01 trial. Ann Oncol 2003;14:395–399. [DOI] [PubMed] [Google Scholar]
  • 32.Tsavaris N, Genatas K, Kosmas C, et al. Leucovorin and 5-fluorouracil versus levamisole and 5-fluorouracil as adjuvant chemotherapy in rectal cancer. J BUON 2005;10:499–504. [PubMed] [Google Scholar]
  • 33.Bleeker WA, Mulder NH, Hermans J, et al. The addition of low-dose leucovorin to the combination of 5-fluorouracil- levamisole does not improve survival in the adjuvant treatment of Dukes' C colon cancer. IKN Colon Trial Group. Ann Oncol 2000;11:547–552. [DOI] [PubMed] [Google Scholar]
  • 34.Di Costanzo F, Sobrero A, Gasperoni S, et al. Adjuvant chemotherapy in the treatment of colon cancer: randomized multicenter trial of the Italian National Intergroup of Adjuvant Chemotherapy in Colon Cancer (INTACC). Ann Oncol 2003;14:1365–1372. [DOI] [PubMed] [Google Scholar]
  • 35.Link KH, Kornmann M, Staib L, et al. Increase of survival benefit in advanced resectable colon cancer by extent of adjuvant treatment: results of a randomized trial comparing modulation of 5-FU + levamisole with folinic acid or with interferon-alpha. Ann Surg 2005;242:178–187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Dencausse Y, Hartung G, Sturm J, et al. Adjuvant chemotherapy in stage III colon cancer with 5-fluorouracil and levamisole versus 5-fluorouracil and leucovorin. Onkologie 2002;25:426–430. [DOI] [PubMed] [Google Scholar]
  • 37.O'Connell MJ, Sargent DJ, Windschitl HE, et al. Randomized clinical trial of high-dose levamisole combined with 5-fluorouracil and leucovorin as surgical adjuvant therapy for high-risk colon cancer. Clin Colorec Cancer 2006;6:133–139. [DOI] [PubMed] [Google Scholar]
  • 38.Wolmark N, Rockette H, Fisher B, et al. The benefit of leucovorin-modulated fluorouracil as postoperative adjuvant therapy for primary colon cancer: results from National Surgical Adjuvant Breast and Bowel Project protocol C-03. J Clin Oncol 1993;11:1879–1887. [DOI] [PubMed] [Google Scholar]
  • 39.Twelves C, Wong A, Nowacki MP, et al. Capecitabine as adjuvant treatment for stage III colon cancer. N Engl J Med 2005;352:2696–2704. [DOI] [PubMed] [Google Scholar]
  • 40.Lembersky BC, Wieand HS, Petrelli NJ, et al. Oral uracil and tegafur plus leucovorin compared with intravenous fluorouracil and leucovorin in stage II and III carcinoma of the colon: results from National Surgical Adjuvant Breast and Bowel Project Protocol C-06. J Clin Oncol 2006;24:2059–2064. [DOI] [PubMed] [Google Scholar]
  • 41.Min JS, Kim NK, Park JK, et al. A prospective randomized trial comparing intravenous 5-fluorouracil and oral doxifluridine as postoperative adjuvant treatment for advanced rectal cancer. Ann Surg Oncol 2000;7:674–679. [DOI] [PubMed] [Google Scholar]
  • 42.Sugimachi K, Maehara Y, Ogawa M, et al. Postoperative chemotherapy for colorectal cancer by combining 5-fluorouracil infusion and 1-hexylcarbamoyl-5-fluorouracil administration after curative resection. Cancer 1996;77:36–43. [DOI] [PubMed] [Google Scholar]
  • 43.Kotake K, Ohashi Y, Kodaira S, et al. A multicenter randomized study comparing 5-fluorouracil continuous infusion (ci) plus 1-hexylcarbamoyl-5-fluorouracil and 5-FU ci alone in colorectal cancer. Oncol Rep 2005;14:129–134. [PubMed] [Google Scholar]
  • 44.Ito K, Yamaguchi A, Miura K, et al. Prospective adjuvant therapy with mitomycin C and carmofur (HCFU) for colorectal cancer, 10-year follow-up: Tokai HCFU Study Group, the first study for colorectal cancer. J Surg Oncol 1996;62:4–9. [DOI] [PubMed] [Google Scholar]
  • 45.Andre T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med 2004;350:2343–2351. [DOI] [PubMed] [Google Scholar]
  • 46.Kuebler JP, Wieand HS, O'Connell MJ, et al. Oxaliplatin combined with weekly bolus fluorouracil and leucovorin as surgical adjuvant chemotherapy for stage II and III colon cancer: results from NSABP C-07. J Clin Oncol 2007;25:2198–2204. [DOI] [PubMed] [Google Scholar]
  • 47.Saltz LB, Niedzwiecki D, Hollis D, et al. Irinotecan fluorouracil plus leucovorin is not superior to fluorouracil plus leucovorin alone as adjuvant treatment for stage III colon cancer: results of CALGB 89803. J Clin Oncol 2007;25:3456–3461. [DOI] [PubMed] [Google Scholar]
  • 48.Sobrero A, Frassineti G, Falcone A, et al. Adjuvant sequential methotrexate --> 5-fluorouracil vs 5-fluorouracil plus leucovorin in radically resected stage III and high-risk stage II colon cancer. Br J Cancer 2005;92:24–29 [erratum in: Br J Cancer 2005;93:733]. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.André T, Quinaux E, Louvet C, et al. Phase III study comparing a semimonthly with a monthly regimen of fluorouracil and leucovorin as adjuvant treatment for stage II and III colon cancer patients: final results of GERCOR C96.1. J Clin Oncol 2007;25:3732–3738. [DOI] [PubMed] [Google Scholar]
  • 50.Chau I, Norman AR, Cunningham D, et al. A randomised comparison between 6 months of bolus fluorouracil/leucovorin and 12 weeks of protracted venous infusion fluorouracil as adjuvant treatment in colorectal cancer. Ann Oncol 2005;16:549–557. [DOI] [PubMed] [Google Scholar]
  • 51.Poplin EA, Benedetti JK, Estes NC, et al. Phase III Southwest Oncology Group 9415/Intergroup 0153 randomized trial of fluorouracil, leucovorin, and levamisole versus fluorouracil continuous infusion and levamisole for adjuvant treatment of stage III and high-risk stage II colon cancer. J Clin Oncol 2005;23:1819–1825. [DOI] [PubMed] [Google Scholar]
  • 52.Arkenau HT, Rettig K, Porschen R, et al. Adjuvant chemotherapy in curative resected colon carcinoma: 5-fluorouracil/leucovorin versus high-dose 5-fluorouracil 24-h infusion/leucovorin versus high-dose 5-fluorouracil 24-h infusion. Int J Colorec Dis 2005;20:258–261. [DOI] [PubMed] [Google Scholar]
  • 53.Land SR, Kopec JA, Cecchini RS, et al. Neurotoxicity from oxaliplatin combined with weekly bolus fluorouracil and leucovorin as surgical adjuvant chemotherapy for stage II and III colon cancer: NSABP C-07. J Clin Oncol 2007;25:2205–2211. [DOI] [PubMed] [Google Scholar]
  • 54.NIH Consensus Conference. Adjuvant therapy for patients with colon and rectal cancer. JAMA 1990;264:1444–1450. [PubMed] [Google Scholar]
  • 55.Laurie JA, Moertel CG, Fleming TR, et al. Surgical adjuvant therapy of large-bowel carcinoma: an evaluation of levamisole and the combination of levamisole and fluorouracil. The North Central Cancer Treatment Group and the Mayo Clinic. J Clin Oncol 1989;7:1447–1456. [DOI] [PubMed] [Google Scholar]
  • 56.Moertel CG, Fleming TR, Macdonald JS, et al. Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 1990;322:352–358. [DOI] [PubMed] [Google Scholar]
  • 57.Wong RK, Tandan V, De Silva S, et al. Pre-operative radiotherapy and curative surgery for the management of localized rectal carcinoma. In: The Cochrane Library, Issue 3, 2008. Chichester, UK: John Wiley & Sons, Ltd. Search date 2006. [DOI] [PubMed] [Google Scholar]
  • 58.Colorectal Cancer Collaborative Group. Adjuvant radiotherapy for rectal cancer: a systematic overview of 8507 patients from 22 randomised trials. Lancet 2001;358:1291–1304. Search date not reported. [DOI] [PubMed] [Google Scholar]
  • 59.Peeters KC, Marijnen CA, Nagtegaal ID, et al. The TME trial after a median follow-up of 6 years: increased local control but no survival benefit in irradiated patients with resectable rectal carcinoma. Ann Surg 2007;246:693–701. [DOI] [PubMed] [Google Scholar]
  • 60.Jansson-Frykholm G, Glimelius B, Pahlman L. Preoperative or postoperative irradiation in adenocarcinoma of the rectum. Final treatment results of a randomised trial and an evaluation of late secondary effects. Dis Colon Rectum 1993;36:564–572. [DOI] [PubMed] [Google Scholar]
  • 61.Sauer R, Becker H, Hohenberger W, et al. German Rectal Cancer Study Group. Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med 2004;351:1731–1740. [DOI] [PubMed] [Google Scholar]
  • 62.Taher AN, El-Baradie MM, Nasr AM, et al. Locally advanced rectal carcinoma: preoperative radiotherapy versus postoperative chemoradiation, 10-year follow-up results of a randomized clinical study. J Egypt Nat Cancer Inst 2006;18:233–243. [PubMed] [Google Scholar]
  • 63.Marijnen CA, van de Velde CJ, Putter H, et al. Impact of short-term preoperative radiotherapy on health-related quality of life and sexual functioning in primary rectal cancer: report of a multicenter randomized trial. J Clin Oncol 2005;23:1847–1858. [DOI] [PubMed] [Google Scholar]
  • 64.Peeters KC, van de Velde CJ, Leer JW, et al. Late side effects of short-course preoperative radiotherapy combined with total mesorectal excision for rectal cancer: increased bowel dysfunction in irradiated patients - a Dutch colorectal cancer group study. J Clin Oncol 2005;23:6199–6206. [DOI] [PubMed] [Google Scholar]
  • 65.Birgisson H, Påhlman L, Gunnarsson U, et al. Late gastrointestinal disorders after rectal cancer surgery with and without preoperative radiation therapy. Br J Surg 2008;95:206–213. [DOI] [PubMed] [Google Scholar]
  • 66.Birgisson H, Pahlman L, Gunnarsson U, et al. Late adverse effects of radiation therapy for rectal cancer - a systematic overview. Acta Oncol 2007;46:504–516. [DOI] [PubMed] [Google Scholar]
  • 67.Ooi B, Tjandra J, Green M. Morbidities of adjuvant chemotherapy and radiotherapy for resectable rectal cancer: an overview. Dis Colon Rectum 1999;42:403–418. Search date 1998. [DOI] [PubMed] [Google Scholar]
  • 68.Jeffery M, Hickey BE, Hider PN. Follow-up strategies for patients treated for non-metastatic colorectal cancer. In: The Cochrane Library, Issue 3, 2008. Chichester, UK: John Wiley & Sons, Ltd. Search date 2006. 17253476 [Google Scholar]
  • 69.Tjandra JJ, Chan MK. Follow-up after curative resection of colorectal cancer: a meta-analysis. Dis Col Rectum 2007;50:1783–1799. [DOI] [PubMed] [Google Scholar]
  • 70.Stiggelbout AM, de Haes JCJM, Vree R, et al. Follow up of colorectal cancer patients: quality of life and attitudes towards follow up. Br J Cancer 1997;75:914–920. [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ Clin Evid. 2010 Apr 27;2010:0401.

Adjuvant systemic chemotherapy

Summary

MORTALITY Compared with no adjuvant systemic chemotherapy: Adjuvant systemic chemotherapy plus surgery may be more effective than surgery alone at reducing mortality in people with Dukes' C colorectal cancer. Adjuvant systemic chemotherapy plus surgery may be no more effective than surgery alone at reducing mortality in people with Dukes' B colorectal cancer ( low-quality evidence ). Fluorouracil plus leucovorin plus levamisole compared with fluorouracil plus leucovorin alone: Adding levamisole to fluorouracil plus leucovorin seems no more effective than fluorouracil plus leucovorin alone at reducing mortality at 3 years in people with Dukes' B and C colorectal cancer ( moderate-quality evidence ). Low-dose leucovorin plus fluorouracil compared with high-dose leucovorin plus fluorouracil: High-dose leucovorin plus fluorouracil seems no more effective than low-dose leucovorin plus fluorouracil at reducing mortality at 3 years in people with Dukes' B and C colorectal cancer ( high-quality evidence ). Fluorouracil plus levamisole compared with fluorouracil plus leucovorin or compared with fluorouracil plus leucovorin plus levamisole: We don't know whether fluorouracil plus leucovorin is more effective than fluorouracil plus levamisole or than fluorouracil plus leucovorin plus levamisole at reducing overall mortality in people with Dukes' B and C colorectal cancer (low-quality evidence). Fluorouracil plus leucovorin plus high-dose levamisole compared with fluorouracil plus leucovorin plus standard-dose levamisole: Adding high-dose levamisole to fluorouracil plus leucovorin seems no more effective than adding standard-dose levamisole to fluorouracil plus leucovorin in improving overall survival and may increase adverse effects (moderate-quality evidence). Fluorouracil plus leucovorin compared with fluorouracil plus lomustine plus vincristine (MOF): Fluorouracil plus leucovorin seems more effective than fluorouracil plus lomustine plus vincristine (MOF) at improving overall survival at 3 years in people with Dukes' B and C colon cancer (high-quality evidence). Oral fluoropyrimidines compared with fluorouracil: We don't know whether oral fluoropyrimidine regimens (with or without leucovorin) are more effective than fluorouracil (with or without leucovorin) at improving overall survival in people with Dukes' B and C colorectal cancer (low-quality evidence). Mitomycin C compared with oral fluoropyrimidines plus mitomycin C: Mitomycin C alone may be less effective than 1-hexylcarbamoyl-5-fluorouracil (HCFU) plus mitomycin C at reducing survival at 10 years in people with colorectal cancer (low-quality evidence). Fluorouracil plus leucovorin plus irinotecan compared with fluorouracil plus leucovorin alone: Adding irinotecan to fluorouracil plus leucovorin is no more effective than fluorouracil plus leucovorin alone at improving overall survival at 5 years in people with stage III colon cancer (high-quality evidence). Sequential methotrexate regimen compared with fluorouracil regimen: Adding methotrexate to fluorouracil plus leucovorin (with or without levamisole) seems no more effective than fluorouracil plus leucovorin (with or without levamisole) alone in improving overall survival at 5 years in people with Dukes' B2/B3 and C colon cancer (moderate-quality evidence). Infusion of fluorouracil compared with bolus fluorouracil: We don't know whether regimens including fluorouracil infusion are more effective than regimens including a bolus of fluorouracil at improving overall survival at 3 to 6 years in people with Dukes' B and C colorectal cancer (low-quality evidence). RECURRENCE (RECURRENCE [LOCAL, METASTASIS], DISEASE-FREE SURVIVAL) Compared with no adjuvant systemic chemotherapy: Adjuvant systemic chemotherapy plus surgery may be more effective than surgery alone at reducing recurrence and improving disease-free survival in some people with colorectal cancer. However, the significance of the results varied by the population studied (stage and site of disease), the chemotherapy regimen employed, and the analysis undertaken (low-quality evidence). Fluorouracil plus leucovorin plus levamisole compared with fluorouracil plus leucovorin alone: Adding levamisole to fluorouracil plus leucovorin seems no more effective than fluorouracil plus leucovorin alone at reducing recurrence at 3 years in people with Dukes' B and C colorectal cancer (high-quality evidence). Low-dose leucovorin plus fluorouracil compared with high-dose leucovorin plus fluorouracil: High-dose leucovorin plus fluorouracil seems no more effective than low-dose leucovorin plus fluorouracil at reducing recurrence at 3 years in people with Dukes' B and C colorectal cancer (high-quality evidence). Fluorouracil plus levamisole compared with fluorouracil plus leucovorin or compared with fluorouracil plus leucovorin plus levamisole: Fluorouracil plus leucovorin may be more effective than fluorouracil plus levamisole at improving recurrence/disease-free survival in people with Dukes' B and C colorectal cancer (low-quality evidence). Fluorouracil plus leucovorin plus high-dose levamisole compared with fluorouracil plus leucovorin plus standard-dose levamisole: Adding high-dose levamisole to fluorouracil plus leucovorin seems no more effective than adding standard-dose levamisole to fluorouracil plus leucovorin in improving disease-free survival and may increase adverse effects (moderate-quality evidence). Fluorouracil plus leucovorin compared with fluorouracil plus lomustine plus vincristine (MOF): Fluorouracil plus leucovorin seems more effective than fluorouracil plus lomustine plus vincristine (MOF) at improving disease-free survival at 3 years in people with Dukes' B and C colon cancer (high-quality evidence). Oral fluoropyrimidines compared with fluorouracil: We don't know whether oral fluoropyrimidine regimens (with or without leucovorin) are more effective than fluorouracil (with or without leucovorin) at decreasing recurrence or improving disease-free survival in people with Dukes' B and C colorectal cancer (low-quality evidence). Fluorouracil plus leucovorin plus oxaliplatin compared with fluorouracil plus leucovorin alone: Adding oxaliplatin to fluorouracil plus leucovorin is more effective than fluorouracil plus leucovorin alone in reducing recurrence at 3 years and improving disease-free survival at 3 to 4 years in people with stage II and III colon cancer (high-quality evidence). Fluorouracil plus leucovorin plus irinotecan compared with fluorouracil plus leucovorin alone: Adding irinotecan to fluorouracil plus leucovorin is no more effective than fluorouracil plus leucovorin alone at improving disease-free survival at 5 years in people with stage III colon cancer (high-quality evidence). Sequential methotrexate regimen compared with fluorouracil regimen: Adding methotrexate to fluorouracil plus leucovorin (with or without levamisole) seems no more effective than fluorouracil plus leucovorin (with or without levamisole) alone in improving disease-free survival at 5 years in people with Dukes' B2/B3 and C colon cancer (moderate-quality evidence). Infusion of fluorouracil compared with bolus fluorouracil: We don't know whether regimens including fluorouracil infusion are more effective than regimens including a bolus of fluorouracil at improving relapse free or disease-free survival at 3 to 6 years in people with Dukes' B and C colorectal cancer (low-quality evidence).

Benefits

In this option we have included RCTs of adjuvant systemic chemotherapy following presumed complete surgical resection of a colorectal tumour (Dukes' A–C, stage I–III), and have excluded RCTs that examined the effects of immunotherapy or biological agents, or of regional chemotherapy (such as portal vein infusion or intraperitoneal chemotherapy), or where there had been prior treatment with chemotherapy or radiation.

Adjuvant systemic chemotherapy versus no adjuvant systemic chemotherapy:

We found four systematic reviews in five reports that had different inclusion criteria and performed different analyses, one meta-analysis of individual patient data, two further reports of RCTs included in the reviews, which provided additional data, three additional RCTs, and one subsequent RCT.

The first review (search date 1993; see comment below) included RCTs of adjuvant chemotherapy in people with Dukes' C colon cancer, or chemotherapy and radiotherapy in people with Dukes' B or C rectal cancer, with a minimum 3 years' follow-up. It did not report the proportion of people with rectal cancer who also received radiotherapy. It found that adjuvant chemotherapy significantly reduced mortality compared with surgery alone at 5 years (29 RCTs [9 RCTs colon cancer, 17 RCTs colorectal cancer, 3 RCTs rectal cancer], 12,079 people; OR 0.91, 95% CI 0.83 to 0.99). However, included RCTs varied widely in methodological quality and chemotherapy regimen employed (publication dates of included RCTs ranged from 1959–1993). In a sensitivity analysis (restricting the analysis to studies of higher methodological quality), the review found a similar significant reduction in mortality (quality score greater than 30, scale not reported; OR 0.91, 95% CI 0.83 to 0.98).

The second review (search date not reported; 3351 people with colon cancer – 1446 [46%] people stage II and 1905 [57%] people stage III; see comment below) compared postoperative fluorouracil plus leucovorin (5 RCTs) or fluorouracil plus levamisole (2 RCTs) versus surgery alone. It found that fluorouracil-based adjuvant chemotherapy significantly increased overall survival compared with surgery alone at 5 years, and also significantly decreased recurrence at 5 years (7 RCTs, 3351 people; overall survival: 71% with fluorouracil-based treatment v 64% with surgery alone; HR for death from any cause 0.76, 95% CI 0.68 to 0.85; recurrence free: 69% with fluorouracil-based treatment v 58% with surgery alone; HR for recurrence 0.68, 95% CI 0.60 to 0.76).

The third review (search date 1987–2004; see comment below), which was reported in two papers, meta-analysed data on people with stage II colon cancer in RCTs that had at least one fluorouracil-based chemotherapy arm and in whom surgery had been undertaken with curative intent. It reported that data on people with stage II colon cancer usually comprised a subset of the total population of individual RCTs, which usually included a wider population. In people with stage II colon cancer, it found no significant difference between adjuvant therapy with fluorouracil-based chemotherapy and surgery alone in mortality (12 RCTs, 4187 people; RR 0.87, 95% CI 0.75 to 1.01; P = 0.07). However, these data included 1476 people who received portal vein infusion or immunotherapy, both of which are outside the inclusion criteria of this review. An analysis excluding people who had had immunotherapy (455 people) reported similar results (10 RCTs, 3732 people; RR 0.86, 95% CI 0.73 to 1.02; P = 0.08). The review reported that a further analysis also excluding people who had received portal vein infusion (1021 people) found similar results (further details not reported).

The fourth review (search date 1987–2007; see comment below) included RCTs in people with stage II colon cancer and also included RCTs in which a subset of people had rectal cancer. It reported that, overall, people in the RCTs had completely resected colon (7731 people) or rectal (7600 people) cancers, and four RCTs included only people with stage II colon cancer (3384 people). RCTs published before 1987 were excluded due to methodological concerns about trial design, lack of standardisation of surgery and staging, and inclusion of old chemotherapy regimens. Some RCTs included in the review used chemotherapy regimens no longer in current usage, and most RCTs included wider populations from which data on people with stage II disease was extracted. The review found no significant difference between adjuvant chemotherapy and surgery alone in overall survival in people with stage II disease (16 RCTs; 747/3682 [20%] with adjuvant systemic chemotherapy v 733/3415 [21%] with surgery alone; RR 0.96, 95% CI 0.87 to 1.05). It found that adjuvant therapy significantly improved disease-free survival compared with surgery alone (defined as recurrences: 18 RCTs; 773/4366 [18%] with adjuvant systemic chemotherapy v 912/4276 [21%] with surgery alone; RR 0.83, 95% CI 0.75 to 0.91). The relative proportion of people with stage II colon or rectal cancer was not reported for either analysis.

The patient-level meta-analysis included five RCTs comparing long-term adjuvant uracil-tegafur compared with surgery alone in people with presumed curative resection of rectal cancer. It included RCTs (2091 people in Japan, Dukes' A [13%], Dukes' B [33%], Dukes' C [54%]) identified after a "meticulous search", although details of the search strategy employed were not reported further. Treatment periods varied from 12 to 24 months, three RCTs also used additional chemotherapy (mitomycin C/5-fluorouracil), and the review reported that "most of the Japanese rectal cancer patients did not receive pre- or postoperative radiotherapy in any of the trials." It obtained individual data for each participant from the original trial investigators, and performed an intention-to-treat analysis. It found that, in people with rectal cancers, adjuvant uracil-tegafur-based regimens significantly increased survival (deaths: 290/1115 [26%] with adjuvant treatment v 311/976 [32%] with surgery alone; HR 0.82, 95% CI 0.70 to 0.97; P = 0.02) and significantly improved disease-free survival (HR 0.73, 95% CI 0.63 to 0.84; P less than 0.0001).

One RCT included in two reviews published in three reports reported additional 10-year follow-up results. The RCT compared adjuvant 5-fluorouracil-based chemotherapy in addition to MeCCNU and vincristine versus surgery alone in people with Dukes' B or C adenocarcinoma of the colon. The original 5-year disease-free survival and overall survival figures were significantly different between groups in favour of adjuvant chemotherapy (disease-free survival: RR 1.29, 95% CI 1.03 to 1.61; P = 0.03; overall survival: RR 1.29, 95% CI 1.01 to 1.66; P = 0.04). However, at 10 years' follow-up there was no significant difference between groups in disease-free survival or overall survival (disease-free survival: results presented graphically; HR 1.14, 95% CI 0.94 to 1.39; P = 0.17; overall survival: results presented graphically; HR 1.12, 95% CI 0.91 to 1.38; P = 0.27). The RCT noted that the specific chemotherapy regimen used in the RCT was no longer recommended as other regimens had been shown to be more effective.

One RCT in which a subgroup of people with stage II disease of the colon had been reported in a systematic review also reported additional data on a wider population. It included 3239 people (16 [0.5%] people stage I; 2963 [91%] people stage II; 260 [8%] stage III) with colorectal cancer (2291 [ 71%] people with colon cancer; 948 [29%] people with rectum cancer or both). It compared fluorouracil plus leucovorin versus surveillance after surgery. However, the chemotherapy regimen altered during the course of the trial. The majority of people with chemotherapy received fluorouracil plus low-dose leucovorin (63%), but some people received high-dose leucovorin, and some groups also received levamisole. Overall, it found that adjuvant chemotherapy significantly improved survival and significantly decreased the risk of recurrence compared with observation alone at a median follow-up of 5.5 years (death from any cause: 311/1622 [19%] with chemotherapy v 370/1617 [23%] with surgery alone; RR 0.82, 95% CI 0.70 to 0.95; P = 0.008; recurrences: 293/1622 [18%] with chemotherapy v 359/1617 [22%] with surgery alone; RR 0.78, 95% CI 0.67 to 0.91; P = 0.001).

The first additional RCT (334 people with Dukes' C colorectal cancer) found no significant difference between adjuvant 5-fluorouracil plus vincristine plus lomustine and surgery alone in overall survival or recurrence (survival: results presented graphically; P = 0.34; recurrence: results presented graphically; P = 0.16).

The second additional RCT (173 people with colorectal cancer) compared adjuvant carmofur (HCFU, an oral fluoropyrimidine) for 1 year versus surgery alone. Participants with all Dukes' stages were included in the trial (17 people Dukes' A/B1; 71 people Dukes' B2; 50 people Dukes' B3/C; 21 people Dukes' C; 4 people Dukes' D). The RCT reported no difference in overall survival between the groups at 5 years (further numerical details not reported; P value not reported). It found that adjuvant chemotherapy significantly improved disease-free survival compared with surgery alone (77% with adjuvant chemotherapy v 63% with surgery alone; P = 0.04).

The third additional RCT (317 people, stage II and III colon cancer) compared 5-fluorouracil plus leucovorin versus surgery alone. It found that, compared with surgery alone, adjuvant chemotherapy significantly increased the proportion of people relapse free at 5 years (results presented graphically; P = 0.004) and significantly increase overall survival at 5 years (results presented graphically; P = 0.02).

The subsequent RCT (500 people with stage II colon cancer) compared adjuvant 5-fluorouracil plus leucovorin versus surgery alone in people with potentially curative resection. The RCT found no significant difference between groups in overall survival at a median follow-up of 95 months (deaths: 55/252 [22%] with adjuvant chemotherapy v 58/248 [23%] with surgery alone; HR 0.88, 95% CI 0.61 to 1.27; P = 0.49). It found no significant difference between groups in relapse or disease-free survival at a median follow-up of 95 months (relapse [local recurrence or metastatic disease]: results presented graphically; HR 0.69, 95% CI 0.45 to 1.06; P = 0.09; disease-free survival [defined as recurrence, metastasis, second primary, or death]: 75 events with adjuvant chemotherapy v 76 events with surgery alone; HR 0.95, 95% CI 0.69 to 1.31; P = 0.77).

Fluorouracil plus leucovorin plus levamisole versus fluorouracil plus leucovorin alone:

We found one RCT (4863 people with Dukes' A [0.2%], B [27%], C [73%] colorectal cancer and no residual disease after surgery) comparing adjuvant chemotherapy with levamisole plus intravenous fluorouracil plus leucovorin versus intravenous fluorouracil plus leucovorin alone. It found no significant difference between groups in mortality or recurrence rate after 3 years (mortality: 31% with levamisole plus fluorouracil plus leucovorin v 29% with fluorouracil plus leucovorin alone; OR 1.10, 95% CI 1.00 to 1.22; recurrence: 903/2429 [37%] with levamisole plus fluorouracil plus leucovorin v 862/2434 [35%] with fluorouracil plus leucovorin alone; OR 1.07, 95% CI 0.97 to 1.17).

One factorial RCT (1327 people, colorectal cancer, stage II or stage III, numbers in each stage not reported) compared 5-fluorouracil alone, 5-fluorouracil plus levamisole, 5-fluorouracil plus leucovorin, and 5-fluorouracil plus levamisole plus leucovorin. People with rectal cancer also received radiotherapy. It found similar mortality in groups at 5 years but did not test the significance of differences between individual groups (mortality: 67/308 [21.7%] with 5-fluorouracil alone v 65/357 [18.2%] with 5-fluorouracil plus levamisole v 67/312 [21.5%] with 5-fluorouracil plus leucovorin v 65/350 [18.6%] with 5-fluorouracil plus levamisole plus leucovorin; overall between-group analysis or analysis between individual groups not reported). In a multivariable regression analysis, it found no significant effects on relapse or survival of adding levamisole (relapse: HR 0.99, 95% CI 0.81 to 1.21; death: HR 0.94, 95% CI 0.73 to 1.20) or of adding leucovorin (relapse: HR 0.89, 95% CI 0.73 to 1.09; death: HR 1.02, 95% CI 0.80 to 1.30).

We found one further three-armed RCT comparing fluorouracil plus leucovorin, fluorouracil plus levamisole, and fluorouracil plus leucovorin plus levamisole (see fluorouracil plus levamisole versus fluorouracil plus leucovorin or versus fluorouracil plus leucovorin plus levamisole).

Low-dose leucovorin plus fluorouracil versus high-dose leucovorin plus fluorouracil:

We found one RCT (4927 people with Dukes' A [0.2%], B [28%], or C [72%] colorectal cancer and no residual disease after surgery), which compared adjuvant high-dose leucovorin versus low-dose leucovorin in people given intravenous fluorouracil. It found no significant difference between groups in mortality or recurrence rate after 3 years (mortality: 30% with high-dose v 29% with low-dose; OR 1.04, 95% CI 0.94 to 1.15; recurrence: 888/2464 [36%] with high-dose v 888/2463 [36%] with low-dose; OR 1.00, 95% CI 0.91 to 1.09).

Fluorouracil plus levamisole versus fluorouracil plus leucovorin or versus fluorouracil plus leucovorin plus levamisole:

We found eight RCTs.

The first RCT (2151 people, Dukes' B [41%] and C [59%] colon cancer) compared adjuvant treatment with fluorouracil plus leucovorin, fluorouracil plus levamisole, and fluorouracil plus leucovorin plus levamisole. It found that fluorouracil plus leucovorin significantly increased disease-free survival compared with fluorouracil plus levamisole at 5 years (65% with fluorouracil plus leucovorin v 60% with fluorouracil plus levamisole; P = 0.04). It found no significant difference between groups in survival (74% with fluorouracil plus leucovorin v 70% with fluorouracil plus levamisole; P = 0.07). It found no significant difference between fluorouracil plus leucovorin plus levamisole and fluorouracil plus leucovorin alone, or between fluorouracil plus leucovorin plus levamisole and fluorouracil plus levamisole alone in disease-free survival or overall survival (results presented graphically; all P values above 0.05).

The second RCT (680 people, stage III colon cancer) compared adjuvant treatment with 5-fluorouracil plus leucovorin versus 5-fluorouracil plus levamisole. It found that 5-fluorouracil plus leucovorin significantly increased overall survival and recurrence-free survival compared with 5-fluorouracil plus levamisole after 82 months (median survival: 88.9 months with 5-fluorouracil plus leucovorin v 78.5 months with 5-fluorouracil plus levamisole; P = 0.0035; median time to relapse: 79.8 months with 5-fluorouracil plus leucovorin v 69.3 months with 5-fluorouracil plus levamisole; P = 0.012).

The third RCT (150 people, Dukes' B2 [47%] and C [53%]) compared adjuvant 5-fluorouracil plus leucovorin versus 5-fluorouracil plus levamisole in people with rectal cancer. Both groups also received adjuvant radiotherapy. People in the leucovorin arm received 6 months of adjuvant chemotherapy, whereas those in the levamisole arm received 12 months of chemotherapy. It found no significant differences between the groups in recurrence or mortality after a median follow-up of 7.4 years (results presented graphically: recurrence; P = 0.82; mortality; P = 0.65).

The fourth RCT (500 people with Dukes' C colon cancer), which compared 5-fluorouracil plus levamisole versus 5-fluorouracil plus levamisole plus leucovorin, found no significant difference between groups in overall survival or disease-free survival at 5 years (overall survival: results presented graphically, 55% with 5-fluorouracil plus levamisole v 59% with 5-fluorouracil plus levamisole plus leucovorin; P = 0.96; disease-free survival: results presented graphically, 49% with 5-fluorouracil plus levamisole v 46% with 5-fluorouracil plus levamisole plus leucovorin; P = 0.86).

The fifth RCT (1703 people, Dukes' B2–3 [48%] and C [52%] colon cancer), which compared adjuvant 5-fluorouracil plus levamisole versus 5-fluorouracil plus levamisole plus leucovorin, found no significant difference between groups in overall survival or disease-free survival at a median follow-up of 6.4 years (overall survival: 68% with 5-fluorouracil plus levamisole v 71% with 5-fluorouracil plus levamisole plus leucovorin; disease-free survival: 58% with 5-fluorouracil plus levamisole v 60% with 5-fluorouracil plus levamisole plus leucovorin; both results presented graphically, reported as not significant; P value not reported).

The sixth RCT (904 people, Dukes' B3 [8%] and C [92%] colon cancer) compared 5-fluorouracil plus levamisole, 5-fluorouracil plus levamisole plus leucovorin, and 5-fluorouracil plus levamisole plus interferon-alfa. We have not reported on results for the interferon arm as this is outside of the scope of this review. The RCT found that 5-fluorouracil plus levamisole plus leucovorin significantly improved recurrence-free survival and overall survival compared with 5-fluorouracil plus levamisole at a median follow-up of 4.6 years ([577 people in analysis] recurrence-free survival: 62% with 5-fluorouracil plus levamisole plus leucovorin v 52% with 5-fluorouracil plus levamisole alone; P = 0.007; overall survival: 72% with 5-fluorouracil plus levamisole plus leucovorin v 60% with 5-fluorouracil plus levamisole alone; P = 0.004).

The seventh RCT (155 people with stage III colon cancer) compared 5-fluorouracil plus levamisole for 12 months versus 5-fluorouracil plus leucovorin for 6 months or 5-fluorouracil plus leucovorin for 12 months. As, at interim analysis, there was no significant difference in recurrence or disease-free survival in the two leucovorin arms; results for the 6- and 12-months leucovorin groups were pooled together in the final analysis. The RCT found no significant difference between 5-fluorouracil plus levamisole and 5-fluorouracil plus leucovorin (for 6 or 12 months) in overall survival or disease-free survival (results presented graphically: overall survival: HR 0.99, 95% CI 0.54 to 1.82; P = 1.0; disease-free survival: HR 1.04, 95% CI 0.59 to 1.83; P = 0.9).

The eighth factorial RCT compared 5-fluorouracil alone, 5-fluorouracil plus levamisole, 5-fluorouracil plus leucovorin, and 5-fluorouracil plus levamisole plus leucovorin (see fluorouracil plus leucovorin plus levamisole versus fluorouracil plus leucovorin alone).

Fluorouracil plus leucovorin plus high-dose levamisole versus fluorouracil plus leucovorin plus standard-dose levamisole:

One RCT (878 people, with stage II [25%] or stage III [75%] colon cancer) examined the effects of giving levamisole at its maximum tolerated dose.It compared 5-fluorouracil plus leucovorin plus high-dose levamisole versus 5-fluorouracil plus leucovorin plus standard-dose levamisole. The initial high-dose levamisole regimen was reduced during the course of the RCT owing to neurological toxicity. It found no significant difference between groups in disease-free survival or overall survival (disease-free survival: results presented graphically; HR 1.0, 95% CI 0.8 to 1.3; P = 0.82; overall survival: results presented graphically; HR 1.0, 95% CI 0.8 to 1.3; P = 0.91).

Fluorouracil plus leucovorin versus fluorouracil plus lomustine plus vincristine (MOF):

One RCT (1081 people, Dukes' B [28%] or C [72%] colon cancer) compared adjuvant lomustine plus vincristine plus 5-fluorouracil (MOF) versus 5-fluorouracil plus leucovorin. The median follow-up was 47.6 months. It found that 5-fluorouracil plus leucovorin significantly increased disease-free survival compared with MOF at 3 years (results presented graphically, 73% with 5-fluorouracil plus leucovorin v 64% with MOF; P = 0.0004). It found that 5-fluorouracil plus leucovorin significantly increased overall survival compared with MOF at 3 years (results presented graphically, 84% with 5-fluorouracil plus leucovorin v 77% with MOF; P = 0.003).

Oral fluoropyrimidines versus fluorouracil:

We found five RCTs.

The first RCT (1987 people with stage III colon cancer) compared adjuvant capecitabine oral chemotherapy versus bolus intravenous fluorouracil plus leucovorin. Median follow-up was 3.8 years. It found that oral capecitabine significantly increased relapse-free survival compared with fluorouracil plus leucovorin (analysis excludes 39 people without relapse or who died from a cause unrelated to colon cancer: HR 0.86, 95% CI 0.74 to 0.99; P = 0.04). It found no significant difference between groups in disease-free survival or overall survival (disease-free survival: results presented graphically; HR 0.87, 95% CI 0.75 to 1.00; P = 0.05; overall survival: results presented graphically; HR 0.84, 95% CI 0.69 to 1.01; P = 0.07). The RCT was designed to prove equivalence of oral to intravenous chemotherapy but not to test any potential survival advantage.

The second RCT (1608 people with stage II [47%] and stage III [53%] colon cancer) compared oral uracil and tegafur (UFT) plus leucovorin versus intravenous fluorouracil plus leucovorin. People were followed up for a median of 62.3 months. It found no significant difference between groups in disease-free survival or overall survival (disease-free survival: results presented graphically; HR 1.00, 95% CI 0.85 to 1.19; overall survival: results presented graphically; HR 1.01, 95% CI 0.83 to 1.24).

The third RCT (166 people with rectal cancer, stage II [40%] and stage III [60%]) compared oral doxifluridine plus leucovorin versus intravenous 5-fluorouracil plus leucovorin. Both groups received postoperative adjuvant radiation. It found no significant difference between groups in rate of recurrence after a median follow-up of 15 months (local or systemic recurrence: 6/92 [7%] with oral doxifluridine plus leucovorin v 9/74 [12%] with intravenous 5-fluorouracil plus leucovorin; P = 0.94). It did not report on overall or disease-free survival.

The fourth RCT (251 people, 33 people Dukes' A/B1; 113 people Dukes' B2; 101 people Dukes' B3/C; 2 people Dukes' D colorectal cancer) compared 1 year of 1-hexylcarbamoyl-5-fluorouracil (HCFU), an oral agent, following induction intravenous 5-fluorouracil versus induction 5-fluorouracil alone. However, it did not perform an overall analysis. Rather, it reported results separately for colon (142 people) and rectal (109 people) cancer. In people with rectal cancer, it found that HCFU plus 5-fluorouracil significantly improved overall survival and recurrence rates compared with fluorouracil alone at 5 years (survival: results presented graphically, 83% with HCFU plus fluorouracil v 52% with fluorouracil alone; P less than 0.05; recurrences: 12/54 [22%] with HCFU plus fluorouracil v 21/52 [40%] with fluorouracil alone; P less than 0.05). In people with colon cancer, it found no significant difference between groups in overall survival or recurrence rates (survival: results presented graphically, 82.0% with HCFU plus fluorouracil v 81.6% with fluorouracil alone, reported as not significant; P value not reported; recurrences: 15/70 [22%] with HCFU plus fluorouracil v 16/71 [24%] with fluorouracil alone, reported as not significant; P value not reported).

The fifth RCT (429 people with stage II [45%] or stage III [55%] colorectal cancer) compared a 14-day course of 5-fluorouracil continuous infusion followed by oral HCFU for 1 year versus a 14-day course of 5-fluorouracil continuous infusion alone. It found no significant difference between groups in overall survival or disease-free survival at 5 years (overall survival: HR 0.96, 95% CI 0.58 to 1.57; P = 0.87; disease-free survival: HR 1.21, 95% CI 0.79 to 1.84; P = 0.38). In subgroup analysis, it found that that disease-free survival was significantly improved in the oral HCFU group in people with colon cancer (268 people; HR 1.87, 95% CI 1.03 to 3.38) but not in people with rectal cancer (161 people; HR 0.71, 95% CI 0.38 to 1.34). It found no significant difference between groups in overall survival or rectal cancer in people with colon cancer (overall survival: 268 people; HR 1.40, 95% CI 0.72 to 2.72; rectal cancer: 161 people; HR 0.58, 95% CI 0.27 to 1.27).

Mitomycin C versus oral fluoropyrimidines plus mitomycin C:

One RCT (172 people with colorectal cancer) compared a combination of mitomycin C (MMC) plus HCFU versus mitomycin C alone. It found that mitomycin C plus HCFU significantly increased survival compared with mitomycin C alone at 10 years (160 people; results presented graphically, 70% with MMC plus HCFU v 52% with MMC alone; P = 0.037). It did not report on other outcomes. It reported that randomisation was by the "envelope" method in order of operations performed (further details not reported).

Fluorouracil plus leucovorin plus oxaliplatin versus fluorouracil plus leucovorin alone:

We found two RCTs. The first RCT (2246 people with stage II [40%] or stage III [60%] colon cancer; the MOSAIC trial) compared adjuvant fluorouracil plus leucovorin plus oxaliplatin versus fluorouracil plus leucovorin alone following curative resection. The RCT found that fluorouracil plus leucovorin plus oxaliplatin significantly decreased recurrence and significantly increased disease-free survival after a median follow-up of 37 months (recurrence: HR 0.77, 95% CI 0.65 to 0.91; P = 0.002; disease-free survival: 78% with fluorouracil plus leucovorin plus oxaliplatin v 73% with fluorouracil plus leucovorin alone; P = 0.002). It found no significant difference between groups in overall survival after a median follow-up of 37 months (HR 0.90, 95% CI 0.71 to 1.13). However, it noted that these data were preliminary and immature and that it was too early to compare the groups statistically in terms of survival. Hence, no conclusion could be drawn on survival until longer follow-up was reported.

The second RCT (2492 people with stage II [29%] or stage III [71%] colon cancer) compared weekly bolus 5-fluorouracil plus leucovorin plus oxaliplatin versus 5-fluorouracil plus leucovorin alone. The RCT found that the addition of oxaliplatin significantly improved disease-free survival compared with 5-fluorouracil plus leucovorin alone at 4 years (results presented graphically; HR 0.80, 95% CI 0.69 to 0.93; P = 0.0034). The RCT reported that the protocol required analysis of survival at 5 years' follow-up, which would be reported in a subsequent publication.

Fluorouracil plus leucovorin plus irinotecan versus fluorouracil plus leucovorin alone:

One RCT (1264 people with completely resected stage III colon cancer) compared adjuvant treatment with standard weekly bolus 5-fluorouracil plus leucovorin with or without irinotecan (CPT-11). The RCT found no significant difference between 5-fluorouracil plus leucovorin plus irinotecan and 5-fluorouracil plus leucovorin alone in overall survival or disease-free survival at 5 years (overall survival: results presented graphically; P = 0.74; disease-free survival: results presented graphically; P = 0.85). It also found no significant difference between groups in recurrence-free survival (events include progression, new colon primary, or death with evidence of disease: results presented graphically; P = 0.84).The RCT found that the addition of irinotecan significantly increased toxic effects (see harms below).

Sequential methotrexate regimen versus fluorouracil regimen:

One RCT (1945 people, Dukes' B2 or B3 [55%], Dukes' C [45%] colon cancer) compared six 1-monthly cycles of 5-fluorouracil with leucovorin versus six 1-monthly cycles of sequential methotrexate and fluorouracil with leucovorin. Both groups also received levamisole. The RCT reported that only 60% of people eventually received levamisole as the trial occurred when the efficacy of levamisole was beginning to be questioned. The RCT found no significant difference between groups in overall survival or disease-free survival at 5 years (overall survival: results presented graphically, 77% with methotrexate v 77% with control; P = 0.90; disease-free survival: results presented graphically, 67% with methotrexate v 63% with control; P = 0.44).

Infusion of fluorouracil versus bolus fluorouracil:

We found four RCTs.

The first factorial RCT (905 people with stage II [43%] and stage III [57%] colon cancer) compared a semi-monthly regime consisting of leucovorin as a 2-hour infusion followed by a 5-fluorouracil bolus and 5-fluorouracil as a 22-hour continuous infusion versus a monthly regime consisting of 5 consecutive days of a 15-minute leucovorin infusion followed by 5-fluorouracil as a 15-minute infusion. People were also randomised to receive treatment for a period of 24 weeks versus 36 weeks. It found no significant difference between the regimens for overall survival or disease-free survival at 6 years (overall survival: results presented graphically; HR 1.02, 95% CI 0.77 to 1.34; P = 0.91; disease-free survival: results presented graphically; HR 1.01, 95% CI 0.81 to 1.27; P = 0.74). It found no significant difference between 24 weeks' and 32 weeks' treatment duration in overall survival or disease-free survival (overall survival: HR 1.11, 95% CI 0.84 to 1.45; P = 0.47; disease-free survival: HR 0.97, 95% CI 0.77 to 1.22; P = 0.63).

The second RCT (801 people with Dukes' B or C colorectal cancer) compared 6 months of bolus 5-fluorouracil plus leucovorin on a 28-day cycle versus 12 weeks of 5-fluorouracil given as a protracted venous infusion. Of 478/801 (60%) people with colon cancer, 211 (44%) people had Dukes' B cancer and 266 (56%) people had Dukes' C cancer. Of 323/801 (40%) people with rectal cancer, 114 (35%) people had Dukes' B cancer and 204 (63%) people had Dukes' C cancer. In total, 59/323 (18%) people with rectal cancer received either pre- or postoperative radiotherapy. The RCT found no significant difference between the bolus and protracted venous-infusion groups in overall survival or relapse-free survival at 5 years (overall survival: 71% with bolus regimen v 76% with protracted venous infusion; HR 0.79, 95% CI 0.61 to 1.03; P = 0.083; relapse-free survival: 67% with bolus regimen v 73% with protracted venous infusion; HR 0.8, 95% CI 0.62 to 1.04; P = 0.10).

The third RCT (1135 people, Dukes' B2 and C1 or C2 colon cancer) compared a continuous infusion of 5-fluorouracil for 56 days every 9 weeks for three cycles versus intravenous 5-fluorouracil plus leucovorin for 5 days every 28 days for a total of six cycles. Both groups also received levamisole every other week. People were followed up for a median of 6.52 years. The RCT found no significant difference between groups in overall survival or disease-free survival at 5 years (overall survival: results presented graphically; HR 1.16, 95% CI 0.93 to 1.44; P = 0.18; disease-free survival: results presented graphically; HR 1.05, 95% CI 0.86 to 1.3; P = 0.65). The RCT was terminated prematurely in 1999 after the first interim analysis demonstrated little likelihood of continuous infusion showing superiority within the stipulated hazard ratio.

The fourth RCT (145 people, stage III colon cancer) compared: 5-fluorouracil plus leucovorin for 5 days every 4 weeks for a total of six cycles; leucovorin as a 2-hour infusion plus a 24-hour infusion of 5-fluorouracil on days 1, 8, 15, 22, 29, 36, and 50 for a total of two cycles; and the same 5-fluorouracil regime for two cycles but without leucovorin. The RCT found no significant difference among groups in disease-free survival at 3 years (results presented graphically, reported as no significant difference among groups; P value not reported) or among groups in mean overall survival (results presented graphically, between group analysis P = 0.6). The RCT was terminated prematurely owing to slow accrual (the initial proposed size was 325 people) and lack of financial support.

Harms

Adjuvant systemic chemotherapy versus no adjuvant systemic chemotherapy:

The first review reported 27 deaths in the levamisole group (19 cancer-related) compared with seven with placebo (further details and statistical analysis of adverse effects not reported). The second review reported that, although it was not a randomised comparison, people treated with fluorouracil plus levamisole had significantly more leukopenia (P = 0.001) and nausea/vomiting (P = 0.05), whereas people treated with fluorouracil plus leucovorin had significantly more stomatitis (P = 0.001) and diarrhoea (P = 0.001). The third review gave no information on adverse effects. The fourth review found no significant difference between adjuvant chemotherapy and surgery alone in systemic chemotherapy deaths (15 RCTs, 4457 people; RR 0.99, 95% CI 0.89 to 1.11). The meta-analysis of patient-level data gave no information on adverse effects. One RCT included in the reviews provided additional data and reported that eight (0.5%) people died from non-colorectal cancer causes in the chemotherapy group compared with four (0.25%) in the surgery-alone group; however, only one death was thought to be possibly chemotherapy related. One additional RCT, which used 5-fluorouracil, lomustine (CCNU), and vincristine, reported that the toxicity experienced was "pronounced", and that 20% of people discontinued treatment because of adverse effects (mainly gastrointestinal and bone marrow), and one person developed acute lymphoblastic leukaemia (further details not reported; statistical analysis between groups not reported). Another additional RCT reported that adverse effects due to carmofur (HCFU) appeared in 18 (20%) people, and included pollakisuria (8 cases), a hot sensation (3 cases), dizziness (3 cases), numbness (2 cases), diarrhoea (2 cases), with one occurrence each of anorexia, micturition disorder, tinnitus, stomatitis, irritability, and nausea and vomiting. The remaining additional RCT reported no deaths associated with chemotherapy. The most common grade 3 or 4 toxicities with adjuvant chemotherapy were stomatitis (36%), diarrhoea (24%), leukoplakia (14%), and nausea (7%). The subsequent RCT reported that 4.5% of 245 people had grade 3 and 4 toxicities with 5-fluorouracil plus leucovorin, the majority being diarrhoea (6 people, 2.5%) and nausea (4 people, 1.6%; further details and statistical analysis between groups not reported).

Fluorouracil plus leucovorin plus levamisole versus fluorouracil plus leucovorin alone:

The first RCT reported that serious unexpected adverse effects were rare and almost all toxic effects seen were those to be expected with the regimens (further details of adverse effects not reported). It reported that treatment-related mortality was 0.1% (4 deaths). The second RCT stated that 13 toxic deaths (1%) were reported, with no significant difference among treatments (P value not reported). It did not statistically compare adverse effects between individual groups.

Low-dose leucovorin plus fluorouracil versus high-dose leucovorin plus fluorouracil:

The RCT reported a treatment-related mortality of 0.1% (see fluorouracil plus leucovorin plus levamisole versus fluorouracil plus leucovorin alone above).

Fluorouracil plus levamisole versus fluorouracil plus leucovorin or versus fluorouracil plus leucovorin plus levamisole:

The first RCT found that 18 people died while receiving chemotherapy (4 people with fluorouracil plus leucovorin v 3 people with fluorouracil plus levamisole v 11 people with fluorouracil plus leucovorin plus levamisole; statistical analysis between groups not reported). The percentage of people who experienced grade 3 or grade 4 toxicity on at least one course was 35% people with fluorouracil plus leucovorin versus 28% people with fluorouracil plus levamisole versus 36% people with fluorouracil plus leucovorin plus levamisole (statistical analysis between groups not reported).

The second and third RCTs gave no detailed information on adverse effects.

The fourth RCT noted that nausea (78%) was the most commonly reported gastrointestinal adverse effect. It found that fluorouracil plus levamisole plus leucovorin significantly increased mucositis, conjunctivitis, and alopecia compared with fluorouracil plus levamisole (mucositis: 11% with fluorouracil plus levamisole plus leucovorin v 1% with fluorouracil plus levamisole; P less than 0.001; conjunctivitis: 45% with fluorouracil plus levamisole plus leucovorin v 22% with fluorouracil plus levamisole; P less than 0.001; alopecia: 34% with fluorouracil plus levamisole plus leucovorin v 16% with fluorouracil plus levamisole; P = 0.008).

The fifth RCT found that fluorouracil plus levamisole plus leucovorin significantly increased gastrointestinal toxicity compared with fluorouracil plus levamisole (total gastrointestinal toxicity: 79% with fluorouracil plus levamisole plus leucovorin v 68% with fluorouracil plus levamisole; P value reported as less than 0.0000; grade 3 severity or more: 19% with fluorouracil plus levamisole plus leucovorin v 9% with fluorouracil plus levamisole; P value reported as less than 0.0000), although grade 3/4 leukopenia was significantly less (1% with fluorouracil plus levamisole plus leucovorin v 3% with fluorouracil plus levamisole; P = 0.02).

The sixth RCT reported one toxicity-related death occurred in one person receiving fluorouracil plus levamisole due to neutropenia followed by pneumonia, and grade 3/4 toxicities occurred in 18% (137/766) people which mainly consisted of neutropenia, nausea and vomiting, and diarrhoea.

The seventh RCT reported that no treatment-related deaths were seen, and generally toxicity was mild; the most common toxicities/adverse effects being nausea, loss of appetite, and leukopenia. It reported that more diarrhoea and stomatitis were seen in the leucovorin arms (grade 3–4 toxic events per cycle, diarrhoea: 2.0 with fluorouracil plus leucovorin v 0.2 with fluorouracil plus levamisole; stomatitis: 1.0 with fluorouracil plus leucovorin v 0.7 with fluorouracil plus levamisole; statistical analysis between groups not reported).

Fluorouracil plus leucovorin plus high-dose levamisole versus fluorouracil plus leucovorin plus standard-dose levamisole:

The RCT found that, compared with the standard levamisole regimen, the high-dose levamisole regimen significantly increased grade 3/4 neurocortical toxicities, lethargy, other neurological toxicities, nausea, and vomiting (grade 3/4 neurocortical toxicities: P less than 0.001; lethargy: P less than 0.001; other neurological toxicities: P less than 0.001; nausea: P = 0.04; vomiting: P = 0.01).

Fluorouracil plus leucovorin versus fluorouracil plus lomustine plus vincristine (MOF):

The RCT reported that a similar proportion of people in each group (6%) had toxicity of at least grade 4. The RCT found MOF increased haematological toxicity and alopecia (haematological toxicity; WBC count less than 2000/microlitre: 16% with MOF v 0.8% with fluorouracil plus leucovorin; alopecia [at least 50% thinning of hair]: 13% with MOF v 0.8% with fluorouracil plus leucovorin). It found that fluorouracil plus leucovorin increased diarrhoea (48% with MOF v 85% with fluorouracil plus leucovorin; statistical analysis between groups not reported).

Oral fluoropyrimidines versus fluorouracil:

The first RCT found that oral capecitabine significantly reduced the onset of predefined overall grade 3 or 4 toxic effects (diarrhoea, stomatitis, nausea, vomiting, hand–foot syndrome, alopecia, or neutropenia) compared with fluorouracil plus leucovorin (results presented graphically; P less than 0.001). With regard to individual grade 3/4 adverse effects, it found that capecitabine significantly reduced the risk of stomatitis and neutropenia (stomatitis: 2% with capecitabine v 14% with fluorouracil plus leucovorin; neutropenia: 2% with capecitabine v 26% with fluorouracil plus leucovorin), but increased the risk of hand–foot syndrome and hyperbilirubinaemia (hand–foot syndrome: 17% with capecitabine v less than 1% with fluorouracil plus leucovorin; hyperbilirubinaemia: 20% with capecitabine v 6% with fluorouracil plus leucovorin; all comparisons P less than 0.001).

The second RCT reported that 15 people died while receiving chemotherapy (9 [1.2%] with intravenous fluorouracil plus leucovorin v 6 [0.8%] with UFT plus leucovorin; statistical analysis between groups not reported), and that toxicities of the two regimens were similar. In total, about a third of people experienced a non-haematological grade 3/4 toxicity, with the most common being diarrhoea (any non-haematological grade 3/4 toxicity: 37.8% with intravenous fluorouracil plus leucovorin v 38.2% with UFT plus leucovorin; grade 3/4 diarrhoea: 28.5% with intravenous fluorouracil plus leucovorin v 29.4% with UFT plus leucovorin; statistical analysis between groups not reported).

The third RCT found that intravenous fluorouracil plus leucovorin significantly increased leukopenia and alopecia (leukopenia, grade I–IV: 30/74 [40%] with intravenous 5-fluorouracil plus leucovorin v 17/93 [18%] with oral doxifluridine plus leucovorin; alopecia, grade I–II: 21/74 [28%] with intravenous 5-fluorouracil plus leucovorin v 13/93 [14%] with oral doxifluridine plus leucovorin; P less than 0.05 for both comparisons), while oral doxifluridine plus leucovorin significantly increased diarrhoea (diarrhoea, grade I–IV: 15/74 [20%] with intravenous 5-fluorouracil plus leucovorin v 36/93 [39%] with oral doxifluridine plus leucovorin; P less than 0.05). Three people with diarrhoea with oral doxifluridine required admission to hospital. The fourth RCT reported no significant differences between groups in adverse effects. The fifth RCT reported that overall adverse effects were significantly higher in the HCFU-plus-fluorouracil group compared with fluorouracil alone (47/214 [22%] with HCFU plus fluorouracil v 28/215 [13%] with fluorouracil alone; P = 0.016). It reported that grade 3/4 toxicity occurred in six people in the HCFU group and three people in the fluorouracil alone group (no statistical analysis between groups reported), and that there were no treatment-related deaths.

Mitomycin C versus oral fluoropyrimidines plus mitomycin C:

The RCT reported that adverse effects due to mitomycin C were present in 27 (34%) people in the mitomycin C alone arm and in 27 (29%) people in the mitomycin C plus HCFU arm, and that adverse effects due to HCFU were present in 23 (25%) people in the mitomycin C plus HCFU arm (statistical analysis between groups not reported). It reported no severe adverse effects (statistical analysis between groups for individual adverse effects not reported).

Fluorouracil plus leucovorin plus oxaliplatin versus fluorouracil plus leucovorin alone:

The first RCT found that, compared with fluorouracil plus leucovorin alone, fluorouracil plus leucovorin plus oxaliplatin was associated with a significant increase of grade 3/4 adverse effects including paraesthesia, neutropenia, thrombocytopenia, nausea, diarrhoea, vomiting, allergic reactions, and neutropenia with fever or infection (paraesthesia: P = 0.001; neutropenia: P less than 0.001; thrombocytopenia: P = 0.001; nausea: P less than 0.001; diarrhoea: P less than 0.001; vomiting: P less than 0.001; allergic reactions: P less than 0.001; neutropenia with fever or infection: P less than 0.001). Overall, 92% of people in the oxaliplatin group had peripheral neuropathy during treatment, half of which was grade 1. In total, 137 (12%) of people had grade 3 peripheral neuropathy during treatment, 11 (1%) people had grade 3 peripheral neuropathy at 12 months, and five (0.5%) people had it at 18 months. Twelve people died within 1 month after the end of treatment (6 people in each group).

The second RCT reported that 13 people (1.0%) treated with fluorouracil plus leucovorin and 15 people (1.2%) with the oxaliplatin regimen died from any cause within 60 days of chemotherapy (P = 0.49). The oxaliplatin regimen significantly increased grade 3 or 4 diarrhoea (P = 0.003), nausea (P less than 0.001), vomiting (P less than 0.001), and dehydration (P less than 0.001). Infection associated with grade 3 or 4 neutropenia was seen in 2% of people with the oxaliplatin regimen compared with 1% with fluorouracil plus leucovorin alone (P less than 0.03). Grade 3 neurosensory toxicity was seen in 8.2% of people and grade 4 in 0.2% of people with the oxaliplatin regimen compared with 0.7% with grade 3 and 0% with grade 4 with fluorouracil plus leucovorin alone (P less than 0.001). A follow-up report of the RCT on neurotoxicity reported that oxaliplatin caused neurotoxicity, primarily in hands, during therapy (week 4, cycle 2: "quite a bit or more severe" hand/foot pain in cold: 26% with oxaliplatin regimen v 3% with fluorouracil plus leucovorin alone; P value not reported) and in feet during follow-up (18 months: "somewhat or more severe" feet numb/tingling: 22% with oxaliplatin regimen v 5% with fluorouracil plus leucovorin alone; P value not reported), and in more than 10% of people with oxaliplatin the neurotoxicity was long lasting (greater than 2 years).

Fluorouracil plus leucovorin plus irinotecan versus fluorouracil plus leucovorin alone:

The RCT found that a significantly greater proportion of people died within 6 months of study entry in the irinotecan group compared with the fluorouracil plus leucovorin-alone group (3% with irinotecan regimen v 1% with fluorouracil plus leucovorin alone; P = 0.008). The RCT found that fluorouracil plus leucovorin plus irinotecan significantly increased grade 3 or 4 adverse effects compared with fluorouracil plus leucovorin alone, including those involving "neutrophils/granulocytes, absolute neutrophil count", "leukocytes, total WBC", fatigue, vomiting, and maximum toxicity (not further defined) ("neutrophils/granulocytes, absolute neutrophil count": 43% with irinotecan regimen v 5% with fluorouracil plus leucovorin alone; P less than 0.0001; "leukocytes, total WBC": 19% with irinotecan regimen v 3% with fluorouracil plus leucovorin alone; P less than 0.0001; fatigue: 10% with irinotecan regimen v 7% with fluorouracil plus leucovorin alone; P = 0.02; vomiting: 10% with irinotecan regimen v 7% with fluorouracil plus leucovorin alone; P = 0.03; maximum toxicity: 70% with irinotecan regimen v 49% with fluorouracil plus leucovorin alone; P less than 0.0001).

Sequential methotrexate and fluorouracil:

The RCT reported that overall, both groups had similar levels of grade 3/4 adverse effects. It found that thrombocytopenia and mucositis were significantly more common in the methotrexate group (thrombocytopenia: 2.2% with methotrexate v 0.1% with control; P less than 0.0001; mucositis: 15% with methotrexate v 12% with control; P = 0.048), while diarrhoea was significantly higher in the control group (7% with methotrexate v 13% with control; P less than 0.0001). There were two toxic deaths in the methotrexate arm and none in the control arm (statistical analysis between groups not reported).

Infusion of fluorouracil versus bolus fluorouracil:

The first RCT gave no information on adverse effects.

The second RCT found that, compared with the bolus group, the protracted venous-infusion regimen significantly reduced the incidence of neutropenia, diarrhoea, stomatitis, nausea/vomiting, alopecia, and lethargy (all P less than 0.0001) as well as anaemia (P = 0.019) and thrombocytopenia (P = 0.025), but significantly increased the incidence of hand–foot syndrome (P less than 0.0001).

The third RCT did not report a statistical analysis between groups for adverse effects. It found that grade 4 toxicity occurred in 39% of people in the bolus arm compared with 5% in the continuous-infusion arm; febrile grade 3/4 neutropenia occurred in 14 people in the bolus arm compared with none in the infusional arm; and grade 3 to 4 gastrointestinal toxicity was three-fold more common in the bolus arm. It reported that grade 2 hand–foot syndrome occurred in 35% of people in the continuous-infusion arm compared with 3% of people in the bolus arm and grade 3 hand–foot syndrome in 7% of people in the continuous-infusion arm compared with less than 1% of people in the bolus arm. Almost twice as many people (106 with continuous-infusion v 64 with bolus) discontinued treatment early with continuous infusion, the reasons given being the logistics of pump therapy, clotting episodes, neck pain related to the catheter, and chronic hand–foot syndrome.

The fourth RCT reported that the main grade 3/4 adverse effects "were well balanced in all treatment arms (no significant difference between all arms)." It found that diarrhoea was significantly increased in the leucovorin-plus-fluorouracil group compared with the fluorouracil-alone group (P = 0.04).

Comment

The first four systematic reviews mentioned a National Institutes of Health (NIH) consensus conference recommendation published in 1990 that, based on the then current data, concluded that people with stage III colon cancer unable to enter a clinical trial should be offered adjuvant chemotherapy unless medical or psychosocial contraindications existed. This was based on one RCT published in 1989, which found a significant improvement in survival in subgroup analysis in people with stage III colorectal cancer, and on a subsequent RCT published in 1990, which found a significant increase in overall survival in people with stage III colon cancer. The same consensus statement also recommended combined postoperative chemotherapy and radiotherapy in people with stage II and III rectal cancer, based on the then available evidence. Current national guidance in the UK makes no distinction between colon and rectal cancer, and recommends that all people with node-positive disease be offered chemotherapy if deemed fit enough to tolerate its adverse effects.

The fourth review in people with stage II colon disease noted that, although there was no improvement in overall survival with adjuvant systemic chemotherapy, there was a significant improvement in disease-free survival (defined as recurrences). The review concluded, "It seems reasonable to discuss the benefits of adjuvant systemic chemotherapy with those stage II patients who have high-risk features, including obstruction, perforation, inadequate lymph node sampling, or T4 disease. The comorbidities and likelihood of tolerating adjuvant systemic chemotherapy should be considered as well. There exists a need to further define which high-risk features in stage II colon cancer patients should be used to select patients for adjuvant therapy."

Clinical guide

The role of adjuvant chemotherapy in stage III colorectal cancer has been clinically accepted and most recent clinical trials have focused on the role of chemotherapy in stage II colorectal cancer, the addition of newer anticancer agents, or comparing oral and intravenous agents. It is now accepted that fluorouracil plus levamisole is no more effective than fluorouracil alone and that high-dose leucovorin is no more effective than low-dose leucovorin at reducing mortality or preventing recurrence.

Substantive changes

Adjuvant systemic chemotherapy Previous option restructured to include only adjuvant systemic chemotherapy and to exclude RCTs that examined the effects of immunotherapy, biological agents, or regional chemotherapy. Two systematic reviews in three reports, one meta-analysis of individual patient data,two RCTs identified by the reviews and which provided additional data, three additional RCTs, and one subsequent RCT added to benefits and harms sections, which compared the effects of adjuvant systemic chemotherapy plus surgery versus surgery alone. Two already reported systematic reviews which compared the effects of adjuvant systemic chemotherapy plus surgery versus surgery alone re-evaluated and re-reported. Twenty-three RCTs comparing different adjuvant systemic chemotherapy regimens versus each other added, and one extended follow-up report on harms added. The previous conclusions confirmed that adjuvant systemic chemotherapy confers benefit compared with surgery alone in some people with colorectal cancer. Categorisation of adjuvant systemic chemotherapy unchanged (Beneficial).

BMJ Clin Evid. 2010 Apr 27;2010:0401.

Preoperative radiotherapy in people with rectal cancer

Summary

MORTALITY Preoperative radiotherapy plus surgery compared with surgery alone: Preoperative radiotherapy plus surgery may be marginally more effective than surgery alone at reducing overall mortality in people with rectal cancer ( low-quality evidence ). Preoperative radiotherapy compared with postoperative radiotherapy: We don't know whether preoperative radiotherapy is more effective than postoperative radiotherapy at improving overall survival in people with Dukes' B and C rectal cancer ( very low-quality evidence ). RECURRENCE (RECURRENCE [LOCAL, METASTASIS], DISEASE-FREE SURVIVAL) Preoperative radiotherapy plus surgery compared with surgery alone: Preoperative radiotherapy plus surgery seems more effective than surgery alone at reducing both local recurrence and any recurrence in people with rectal cancer ( moderate-quality evidence ). Preoperative radiotherapy compared with postoperative radiotherapy: Preoperative radiotherapy may be more effective than postoperative radiotherapy at reducing local recurrence at 5 years in people with Dukes' B and C rectal cancer, but we don't know whether it is more effective at increasing disease-free survival (very low-quality evidence).

Benefits

Preoperative radiotherapy plus surgery versus surgery alone:

We found one systematic review (search date 2006) comparing preoperative radiotherapy plus surgery versus surgery alone. The review included people of any age with localised resectable carcinoma of the rectum, and included RCTs of colorectal cancer where data for rectal cancer could be extracted. RCTs that used adjuvant chemotherapy were included provided chemotherapy was given to both arms. The review noted that, since the publication of earlier RCTs, radiotherapy techniques had evolved, and it examined the effects of different techniques through subgroup/sensitivity analysis (it reported that 14 included RCTs used non-contemporary radiotherapy techniques). The review noted that total mesorectal excision (TME) had been increasingly incorporated into clinical practice, and it performed a subgroup analysis on TME. It did not report the proportion of people with Dukes' stage A, B, or C in each analysis. It reported that, overall in included RCTs, the average proportion of people with Dukes' A disease was 17% in the control arm (range 0.7% to 37% in individual RCTs; no further details reported). It did not obtain individual patient data, but included data from one previously published analysis of individual patient data (including 8705 people) in a sensitivity analysis as this provided more-accurate and longer-term data than were available through published literature.

The review found that overall mortality was lower with preoperative radiotherapy compared with surgery alone; however, the result was of borderline significance (20 RCTs [published data only], 8163 people; HR 0.93, 95% CI 0.87 to 1.00; P = 0.04). In a further analysis using both individual patient data and published data, it found no significant difference in overall survival between groups (20 RCTs, 7710 people; HR 0.95, 95% CI 0.89 to 1.02; P = 0.1). Calculating the magnitude of any potential benefit based on the mortality curve from the largest included RCT with long-term 8-year follow-up (1168 people), the review reported that this was estimated to translate into a 2% survival improvement (from 75% to 77%) at 5 years and a 2% survival improvement (from 60% to 62%) at 8 years. In subgroup analysis, it found that preoperative radiotherapy significantly increased overall survival in RCTs including surgical techniques other than TME (non-TME: 19 RCTs, 6302 people; HR 0.92, 95% CI 0.86 to 0.99; P = 0.02), in those RCTs using higher biologically equivalent-dose radiotherapy (biological effective dose defined as 30 Gy or above: 9 RCTs, 5590 people; HR 0.91, 95% CI 0.84 to 0.98; P = 0.01) and in those RCTs with treatment fields focused to the posterior pelvis (4 RCTs, 3870 people; HR 0.85, 95% CI 0.76 to 0.95; P = 0.003).

The review found that preoperative radiotherapy significantly reduced local recurrence compared with surgery alone (13 RCTs, 7467 people; HR 0.71, 95% CI 0.64 to 0.78; P less than 0.00001). However, the result was significantly heterogeneous (P less than 0.0001 for heterogeneity). The absolute rate of local recurrence in the control groups was variable across the included RCTs (range 11–54%). The cause of the heterogeneity was not clear, and the review concluded that a single hazard ratio was inappropriate. It suggested that the heterogeneity may be due to difference in the baseline risk of recurrence. However, of the 13 individual RCTs included in the analysis, eight RCTs found a significant reduction in local recurrence with preoperative radiotherapy, one RCT was borderline significant in favour of radiotherapy, and four RCTs found no significant differences between groups or the differences were not estimable. Only one included RCT (1861 people) specifically required TME as the surgical technique, and it had found that preoperative radiotherapy significantly reduced local recurrence compared with surgery alone (OR 0.42, 95% CI 0.26 to 0.67). A subsequent report of this RCT found that the difference between groups was still significant at 5 years' follow-up (local recurrence: 6% with preoperative radiotherapy v 11% with surgery alone; P less than 0.001).

The review found that preoperative radiotherapy significantly reduced any recurrence compared with surgery alone (8 RCTs, 5177 people; HR 0.89, 95% CI 0.82 to 0.97; P = 0.006). The review reported that, using data from the largest RCT with available data (849 people), this would translate to a modest improvement of 2% in recurrence-free survival (75% to 77%) at 2 years, and 4% (60% to 64%) at 5 years.

Preoperative radiotherapy versus postoperative radiotherapy:

We found three RCTs.

The first RCT (471 people with Dukes' B or C rectal carcinoma, numbers with each stage not reported) compared preoperative radiotherapy (25.5 Gy in 1 week) versus postoperative radiotherapy (60.0 Gy over 7–8 weeks). Results were based on 413/471 (87%) people initially randomised. It did not report whether people also received adjuvant chemotherapy. It found that preoperative radiotherapy significantly reduced local tumour recurrence after 5 years (27/209 [13%] with preoperative radiotherapy v 45/204 [22%] with postoperative radiotherapy; P = 0.02). It found no significant difference in 5-year survival (results presented graphically, absolute numbers not reported; P = 0.5).

The second RCT (799 people, with Dukes' A/B1 [22%], B2/B3 [29%], C [32%], D [7%], rectal carcinoma [some people unknown histology]) compared preoperative radiotherapy (50.4 Gy in 28 fractions) versus postoperative radiotherapy (same schedule plus a 5.4 Gy in 3-fraction tumour-bed boost). Both groups also received chemotherapy with 5-fluorouracil during radiotherapy and 4 weeks after surgery (in the preoperative group) or 4 weeks after chemoradiotherapy (in the postoperative group). There were significant differences between groups in the proportion of people who received the full dose of radiotherapy or full dose of chemotherapy (full-dose radiotherapy: 92% with preoperative v 54% with postoperative; P less than 0.001; full-dose chemotherapy: 89% with preoperative v 50% with postoperative; P less than 0.001). The RCT found no significant difference between groups in survival or disease-free survival (survival: results presented graphically, 76% with preoperative chemoradiotherapy v 74% with postoperative chemoradiotherapy; P = 0.80; disease-free survival: results presented graphically, 68% with preoperative chemoradiotherapy v 65% with postoperative chemoradiotherapy; P = 0.32). The RCT found that preoperative treatment significantly reduced the cumulative incidence of local recurrences compared with postoperative treatment at 5 years (results presented graphically, 6% with preoperative chemoradiotherapy v 13% with postoperative chemoradiotherapy; P = 0.006).

The third RCT (50 people, Dukes' B [26%] or Dukes' C [74%]) compared preoperative radiotherapy versus postoperative chemoradiation in people with localised rectal cancer. People randomised to the first group received preoperative radiotherapy (46 Gy in 23 fractions over 4.5 weeks) followed by surgery with or without adjuvant chemotherapy (depending on staging). Those randomised to the second group had surgery followed by chemoradiotherapy (50 Gy in 25 fractions with 5-fluorouracil chemotherapy). The RCT found no significant difference between groups in overall survival or disease-free survival at 10 years (overall survival: results presented graphically, 63% with postoperative treatment v 60% with preoperative treatment; P = 0.698; disease-free survival: 65% with postoperative treatment v 66% with preoperative treatment; P = 0.816).

Harms

Preoperative radiotherapy plus surgery versus surgery alone:

The systematic review reported that acute radiotherapy adverse effects were poorly reported in included RCTs. The proportion of people with no toxicities ranged from 20% to 84%, and the most common adverse effect was diarrhoea (256/1254 [20%]). With regard to acute toxicities post surgery, the review found that preoperative radiotherapy was associated with a significantly lower rate of people with no toxicities compared with surgery alone (no toxicities: 6 RCTs; 962/1836 [52%] with preoperative radiotherapy v 1125/1879 [60%] with surgery alone; RR 0.88, 95% CI 0.82 to 0.94). It found significantly more pelvic/perineum wound infections with preoperative radiotherapy compared with surgery alone (6 RCTs, 3350 people; HR 1.36, 95% CI 1.00 to 1.83; P = 0.05). It found no significant difference between groups for wound infections (not otherwise specified), delayed wound healing, multiorgan failure, perforation, intestinal fistula, or intestinal necrosis (all P greater than 0.05). It meta-analysed data for postoperative mortality, but the result was significantly heterogeneous between RCTs (P = 0.003 for heterogeneity). Of the 16 included individual RCTs, 14 RCTs found no significant difference between groups for postoperative mortality. No meta-analysis of data for late toxicities was possible as included RCTs reported on different aspects of long-term outcome. The review found that preoperative radiotherapy significantly increased stool frequency compared with surgery alone (1 RCT, 171 people; RR 2.52, 95% CI 1.10 to 5.75) and problems with rectal continence (2 RCTs, 235 people; RR 2.12, 95% CI 1.49 to 3.03).

One RCT included in the systematic review subsequently reported on long-term harms for 990/1530 (65%) people included in the original trial. At 3 months, it found significantly worse activity level scores in the preoperative-radiotherapy group compared with the surgery-only group (P = 0.006). At 24 months, the incidence of faecal incontinence was significantly higher in the preoperative radiotherapy group (51% with preoperative radiotherapy v 37% with surgery only; P = 0.002). At 24 months after surgery, a decline in sexual activity was observed both for males and females, which was larger in preoperative-radiotherapy patients than in surgery-only patients. In males, 67% of the preoperative-radiotherapy group and 76% of the surgery-only group who were previously active were still sexually active (P = 0.06). For females, these figures were 72% and 90%, respectively (P = 0.01). At a median of 5 years' follow-up, 597/1530 (39%) people returned a further questionnaire. Among 362 people who did not have a stoma it found that, compared with surgery alone, preoperative radiotherapy significantly increased the rate and severity of faecal incontinence, the incidence of pad-wearing as a result of incontinence, anal blood loss, and mucus loss (faecal incontinence during day: 62% with preoperative radiotherapy v 38% with surgery alone; P less than 0.001; severity of faecal incontinence: results presented graphically; P less than 0.001; pad wearing: 56% with preoperative radiotherapy v 33% with surgery alone; P less than 0.001; anal blood loss: 11% with preoperative radiotherapy v 3% with surgery alone; P = 0.005; anal mucus loss: 27% with preoperative radiotherapy v 15% with surgery alone; P = 0.004).

Another RCT included in the systematic review subsequently reported on long-term adverse effects. It included 908/1147 (79%) of people initially randomised. It reported that the cumulative incidence of small bowel obstruction after 14 years of follow-up was estimated as 14% in people who had radiotherapy and 6% in people who did not have radiotherapy (P less than 0.001). It reported that people who had radiotherapy had an increased risk of late small bowel obstruction irrespective of the radiation technique used. Five people with late gastrointestinal disorders died, four in the irradiated group and one who received surgery alone (P = 0.18).

An additional systematic review (search date not reported) examined late effects related to radiotherapy. It included data from systematic reviews, RCTs, and clinical trials. It was narrative in character and did not meta-analyse data. Most included studies were RCTs comparing short-course preoperative radiotherapy versus surgery alone. The most common late adverse effects due to radiotherapy were bowel obstruction, faecal incontinence, evacuation problems, and sexual dysfunction (statistical analysis between groups not reported).

Preoperative radiotherapy versus postoperative radiotherapy:

The first RCT did not report a direct analysis for adverse effects between groups, but also included data from a previous study. The second RCT found no significant difference between groups in the overall rates of in-hospital mortality, postoperative complications, the rate of anastomotic leakage of any grade, delayed sacral wound healing, postoperative bleeding, or ileus (in-hospital mortality; P = 0.41; postoperative complications: P = 0.68; rate of anastomotic leakage of any grade: P = 0.77; delayed sacral wound healing: P = 0.10; postoperative bleeding: P = 0.5; ileus: P = 0.26). It found that, among people who had received radiotherapy as per the protocol (636 people), preoperative chemoradiotherapy significantly reduced the proportion of people with: grade 3/4 acute diarrhoea (12% with preoperative chemoradiotherapy v 18% with postoperative chemoradiotherapy; P = 0.04); any acute grade 3 or 4 toxic effect (27% with preoperative chemoradiotherapy v 40% with postoperative chemoradiotherapy; P = 0.001); long-term grade 3/4 strictures at the anastomotic site (4% with preoperative chemoradiotherapy v 12% with postoperative chemoradiotherapy; P = 0.003); and any long-term grade 3/4 toxic effect (14% with preoperative chemoradiotherapy v 24% with postoperative chemoradiotherapy). A further analysis, which included the 110 people in the postoperative group who did not receive radiotherapy, found no significant difference between groups in acute toxic effects (P = 0.78) or long-term toxic effects (P = 0.85). The third RCT reported that no people who received adjuvant chemotherapy had grade 3 or 4 haematological or gastrointestinal toxicities, and no people experienced any grade 3 or 4 late radiation sequelae. It reported that the proportion of people with acute radiation reaction was significantly higher with postoperative treatment (grade 3 radiation reaction: 9/26 [35%] with postoperative v 2/24 [8%] with preoperative; P = 0.039).

We found one systematic review (search date 1998, 9 RCTs) comparing preoperative versus postoperative radiotherapy, which reported on harms. It found that postoperative radiotherapy increased anastomotic complications (breakdown and stricture formation) compared with preoperative radiotherapy.

Comment

Clinical guide:

There is evidence of modest improvement in survival following preoperative radiotherapy compared with surgery alone for rectal cancer. There is evidence of a significant improvement in local recurrence but the risk of local recurrence for Dukes' A tumours is so low that preoperative radiotherapy is unlikely to provide much absolute benefit. There is evidence to favour preoperative versus postoperative radiotherapy. There are unresolved issues about the relative effectiveness of different radiotherapy schedules.

There is evidence for significant late toxicity with preoperative radiotherapy compared with surgery alone. There are also unresolved issues about the relative toxicity of different radiotherapy schedules.

Substantive changes

Preoperative radiotherapy in people with rectal cancer One systematic review added and one subsequent report of an RCT included in the review added, which compared the effects of preoperative radiotherapy plus surgery versus surgery alone. Two further RCTs added, which compared the effects of preoperative radiotherapy regimens versus postoperative radiotherapy regimens. Categorisation changed in light of additional evidence. Preoperative radiotherapy in people with rectal cancer categorised as Beneficial (changed from Trade off between benefits and harms).

BMJ Clin Evid. 2010 Apr 27;2010:0401.

Routine intensive follow-up

Summary

MORTALITY Intensive follow-up compared with less-intensive follow-up: Intensive follow-up seems more effective than less-intensive/minimal follow-up at reducing overall mortality at 5 years in people with Dukes' A, B, and C colorectal cancer. However, intensive follow-up regimens varied widely between RCTs ( moderate-quality evidence ). RECURRENCE (RECURRENCE [LOCAL, METASTASIS], DISEASE-FREE SURVIVAL) Intensive follow-up compared with less-intensive follow-up: We don't know whether intensive follow-up is more effective than less-intensive/minimal follow-up at reducing recurrence in people with Dukes' A, B, and C colorectal cancer. Intensive follow-up may be more effective than a less intensive/minimal follow-up at reducing the time to detection of recurrence (reducing mean time to recurrence), and at increasing the proportion of curative surgical procedures attempted for recurrence. However, intensive follow-up regimens varied widely between RCTs ( low-quality evidence ).

Benefits

Intensive follow-up versus less-intensive follow-up:

We found two systematic reviews.

The first systematic review (search date 2006, 8 RCTs, 2162 people) included RCTs in people with colorectal cancer at any stage (Dukes' stage A, B, and C) in whom surgery had been undertaken with curative intent and that compared different follow-up strategies. It reported considerable variation in follow-up strategies reported in the eight RCTs, and the frequency of follow-up visits and the investigations performed varied widely among included RCTs. The review found that more-intensive follow-up significantly improved overall survival compared with less-intensive/minimal follow-up at 5 years (mortality: 6 RCTs; 218/793 [27%] with intensive v 274/808 [34%] with minimal; OR 0.73, 95% CI 0.59 to 0.91). It found that intensive follow-up significantly reduced the mean time to recurrence compared with less-intensive/minimal follow-up (3 RCTs, 420 people; WMD –6.75 months, 95% CI –11.06 months to –2.44 months; significant heterogeneity between RCTs in analysis). It found no significant difference between groups in recurrence (7 RCTs, 1938 people; OR 0.91, 95% CI 0.75 to 1.10). The review reported that more surgical procedures were performed for recurrences in the experimental arms than in the control arms of RCTs (101/354 [28%] with intensive v 44/354 [12%] with less-intensive/minimal; further details not reported, statistical analysis between groups not reported). The review found that, compared with less-intensive/minimal follow-up, more-intensive follow-up was associated with a significantly higher rate of curative surgical procedures attempted for recurrences (6 RCTs; 95/818 [12%] with intensive v 40/795 [5%] with less-intensive/minimal; OR 2.41, 95% CI 1.63 to 3.54). The review noted that, although it found no statistical heterogeneity for most analyses, differences in study populations and study design (variety of components) may challenge the validity of combining the results in a meta-analysis. In addition, during recruitment of included RCTs (1983 to 2001), clinical care and surgical techniques had evolved considerably. Although it found an overall benefit with more-intensive follow-up, owing to wide variations between studies it was unable to infer what an optimal follow-up regimen might include.

The second systematic review (search date 2007, 8 RCTs, 2923 people) included seven RCTs included by the first review, excluded one RCT included by the first review on methods, and included one RCT excluded by the first review as trial recruitment was ongoing. The review found that more-intensive follow-up significantly reduced overall mortality compared with less-intensive/minimal follow-up (8 RCTs; 321/1474 [22%] with intensive v 373/1449 [26%] with less-intensive/minimal; OR 0.74, 95% CI 0.59 to 0.93; P = 0.01). It found no significant difference between groups in all-site recurrence (P = 0.68) and in local or distant metastases (P greater than 0.05). It found that intensive surveillance was significantly more likely to detect recurrence amenable to surgery compared with less-intensive/minimal follow-up (105/985 [11%] with intensive v 54/953 [6%] with less-intensive/minimal; P = 0.0002). It also found that curative reoperations were significantly more likely in people with intensive follow-up (7 RCTs; 86/354 [24%] with intensive v 35/353 [10%] with less-intensive/minimal; OR 2.81, 95% CI 1.65 to 4.79; P = 0.0001). The review concluded that the optimal follow-up strategy (frequency of follow-up and type of tests) was not clear and that further large-scale trials were needed to clarify some of these issues.

Harms

Intensive follow-up versus less-intensive follow-up:

The first review reported that one RCT reported on harms. It found two perforations and two gastrointestinal haemorrhages requiring transfusion from a total of 731 colonoscopies (no further details reported). The second review gave no information on adverse effects. It reported that the potential impact of the intensive surveillance programme on people's physical and psychological well-being, and the associated complications from surveillance procedures had not been well documented.

Comment

One RCT (212 people being followed up after treatment for colorectal cancer) included in the reviews found that follow-up increased feelings of reassurance but not quality of life. Most people in the trial said that they would still prefer follow-up even if it did not lead to earlier detection of recurrence. Current follow-up regimens are variable in frequency and intensity.

In a subgroup analysis, the second review found that RCTs with carcinoembryonic monitoring (CEA) and colonoscopy included had a significantly improved survival compared with RCTs not using these interventions or using them less (CEA v no CEA testing or more v less CEA testing; P = 0.0002; colonoscopy v no colonoscopy or more v less colonoscopy; P = 0.04). However, these data should be interpreted with caution as they do not examine the effects of these interventions alone, as CEA and colonoscopy were not the only differences between the different arms of the RCTs that included multiple diverse components. Hence, any difference in mortality could be due to other differences between arms other than CEA or colonoscopy.

Substantive changes

Routine intensive follow-up Two systematic reviews added, which came to similar conclusions as previously reported systematic reviews. Categorisation of routine intensive follow-up unchanged (Likely to be beneficial).

Articles from BMJ Clinical Evidence are provided here courtesy of BMJ Publishing Group

RESOURCES