Medical Oncology Department
St-Luc University Hospital
Université Catholique de Louvain
Colorectal cancer (CRC) has a high incidence (145,290 new cases in 2005 in the USA) and, compared with prostate and breast cancers, a higher mortality rate (37% after five years). That mortality rate, however, decreased by 1.7% per year between 1985 and 1997.(1,2)
The tumour–node–metastasis (TNM) system is usually chosen to stage the extension of the disease.(3) About 70% of patients present with locoregional stage without distant metastasis, with an estimated 20% having a TNM stage I (superficial cancer with very low risk of relapse), approximately 25% a stage II (deeper cancer in the intestinal wall without lymph node metastasis) and 25% a stage III (with lymph node metastasis and with about 30–60% risk of relapse and death).
The standard treatment of colon cancer is surgery of the primary tumour with resection of the regional lymph nodes, frequently through laparoscopy,(4) whereas in the case of rectal cancer, total mesorectal resection is the treatment of choice. Preoperative radiotherapy is recommended in locally advanced rectal cancer, only to prevent local relapse.(5)
Due to the positive effects of adjuvant treatment on the risk of relapse and death, most stage III and many stage II patients with poor prognosis also receive chemotherapy.(6,7) The criteria for giving adjuvant chemotherapy in stage II CRC are still under discussion, but some of them, such as depth of wall invasion (TNM T4), intestinal perforation, anatomopathological diagnosis of lymphatic or venous permeation, presence of macroscopic or microscopic residual cells left by the surgeon and a small number of resected lymph nodes (<8), are well established. Others, such as perinervous permeation, poor tumour differentiation and intermediate (8–12) number of resected lymph nodes, are still a matter of debate. Possible biological markers, such as low number of microsatellite instabilities in the tumour and loss of heterozygosity, also remain to be confirmed.(8)
For CRC metastases, which are most frequently localised in the liver, lungs, distant lymph nodes or peritoneum, chemotherapy is the preferred treatment. Advances made in anticancer drugs in the last decade have doubled the objective response rate (ORR) to above 40% and the median survival time to more than 20 months in some series.(9) As a result of the high ORR of chemotherapy and advances in surgery, resection of liver metastasis has recently attracted increasing attention, leading to a five-year survival rate of around 40% in a population of selected patients.(10)
Drug treatment of CRC
Table 1 shows a survey of current regimens used in CRC treatment. The cornerstone of CRC treatment remains 5-fluorouracil (5-FU), usually in association with folinic acid. It may be administered as bolus, such as in the Mayo Clinic regimen,(11) or by continuous infusion (CI) with a pump, as in the AIO (Arbeitsgemeinschaft Internisticher Onkologen) schedule(12) or in other schedules.(13) The continuous form has higher ORR (22%) than the bolus form (14%), at the cost, however, of more hand-and-foot syndrome (34% vs 13%). The latter regimen is preferred in the USA, although it is accompanied by more grade 3 and 4 neutropenia, while in Europe the continuous form is more frequently used.(14)
In future, 5-FU may be replaced by its oral prodrug form capecitabine (Xeloda), which is as effective and less toxic both in the adjuvant and the metastatic settings.(15) Capecitabine mimics continuous infusion of 5-FU and retains its toxic effect on the skin, particularly the hand-and-foot syndrome, which is dose-limiting. Its recommended dosage in fit people is 1,250mg/m(2)/12h during two weeks, to be repeated every three weeks. Another oral form of 5-FU, UFT (tegafur + uracil), with benefits and toxicities similar to those of 5-FU, has also been described.(16) Exclusive use of a 5-FU derivative for metastatic CRC treatment results in a median survival time of 11.5 months.(17)
Oxaliplatin (Eloxatin) is effective, in combination with CI of 5-FU, in the first- and second-line treatment of metastatic CRC. Different schedules have been described, such as FOLFOX 4.(18) This association in the adjuvant setting (MOSAIC study) prolongs the disease-free survival (DFS) of some stage II and stage III CRC patients and will therefore soon be registered for this indication.(19) Its principal toxicity is cumulative and results in a slowly reversible sensitive neuropathy of the extremities, particularly upon exposure to cold. Oxaliplatin and irinotecan (a drug that is also used in CRC treatment) also produce general secondary effects similar to those of many other chemotherapeutics compounds (blood cell cytopenia, nausea and vomiting, fatigue, etc).
Irinotecan (Campto(®)) is active on its own,(20) as well as in association with bolus or infusional 5-FU,(21) both in first- and second-line treatment of metastatic CRC, but its benefit in the adjuvant setting has not yet been demonstrated. Its most dangerous toxic effects, especially when given alone at doses >300mg/m(2), are diarrhoea and dehydration, requiring immediate and sustained treatment with loperamide. The ORRs induced by either oxaliplatin or irinotecan (with 5-FU) in metastatic CRC are close to 50%, and the median survival is close to 15 months. Administration of both drugs together in the same combination increases toxicity but not the ORR. In contrast, depending on comorbidities of the patient, they can be used consecutively in any order.(22) Patients who received both drugs during the course of their disease had a median survival time of about 18 months.(23) Oxaliplatin and irinotecan may be associated with either 5-FU or capecitabine with comparable results, but the capecitabine associations are not yet registered.
Epidermal growth factor (EGF) stimulation and angiogenesis are two mechanisms promoting the progression of tumour cells. Recently, monoclonal antibodies have been developed against the EGF receptor and the vascular endothelial growth factor (VEGF), the latter being one of the factors that activate angiogenesis. Two drugs, cetuximab and bevacizumab, have been registered in several countries and will be reimbursed in different situations.
Cetuximab (Erbitux(®)) is an IgG1 antibody that prolongs the DFS of patients whose EGFR-positive cancerous cells are refractory to irinotecan. It is administered intravenously, once a week, alone or in association with irinotecan (Bond regimen).(24) The two drugs have a synergistic potential despite the resistance of the CRC cells to irinotecan.(25) Interestingly, the benefit is significantly related to the most significant toxic effect, an acneiform rash, which has also been observed with other EGF pathway inhibitors.(26) On the other hand, no relation is detected between the intensity of the immunostaining and the sensitivity of the cells to the drug and, recently, patients with EGFR-negative cancerous cells were also demonstrated to be sensitive to cetuximab.(27) Several ongoing studies are aiming at verifying whether cetuximab could be associated with other chemotherapeutic drugs, such as oxaliplatin, in CRC and earlier stages of the disease. ABX-EGF, another humanised monoclonal antibody, is given every two weeks and does not induce hypersensitivity.(28)
Bevacizumab (Avastin(®)) is a humanised monoclonal antibody against VEGF – a promoter of neovascularisation in adults. Thus, its use should prevent vascularisation and, hence, growth of macroscopic tumours larger than a few millimetres. Bevacizumab has been tested in different clinical situations, in particular in the first-line treatment of metastatic CRC patients. In this setting, when associated with IFL (irinotecan, bolus 5-FU and leucovorin), it increased median survival time from 15.6 to 20.3 months compared with IFL alone.(29) An equivalent benefit was observed in combination with the FOLFOX regimen.(30) Bevacizumab is associated with proteinuria and hypertension and rare lethal arterial thrombotic or haemorrhagic events. Gastroenteric perforations and wound healing delays have also been described.
The best way to associate these four new drugs is still unclear. For medical and economical reasons, these drugs should be used as early as possible in the course of the disease, and should, therefore, be tested in the adjuvant setting. In the case of metastatic disease, the many possible combinations must be compared before declaring one schedule superior to the other for the general population or even for a particular patient. Tailoring the treatment to both the tumour biology and the patient’s genes is certainly one avenue for further progress.(31,32)
- Colorectal cancer. Facts and figures, special edition 2005. American Cancer Society; 2005.
- CA Cancer J Clin 2005;55:10-30.
- Greene FL, et al, editors. AJCC cancer staging handbook. 6th ed. New York, Springer, 2002.
- N Engl J Med 2004;350:2050-9.
- N Engl J Med 2004;351:1731-40.
- N Engl J Med 2001;345:1091-7.
- J Clin Oncol 2004;22:1797-806.
- Arch Pathol Lab Med 2000;124:979-94.
- N Engl J Med 2005;352:476-87.
- Annu Rev Med 2005;56:139-56.
- Adv Exp Med Biol 1988;244:173-84.
- Ann Oncol 1994;5:233-7.
- J Clin Oncol 1997;15:808-15.
- J Clin Oncol 1998;16:301-8.
- J Clin Oncol 2001;19:4097-106.
- J Clin Oncol 2002;20:3605-16.
- J Clin Oncol 2004;22:3766-75.
- J Clin Oncol 2000;18:2938-47.
- N Eng J Med 2004;350:2343-51.
- Lancet 1998;352:1413-8.
- Lancet 2000;355:1041-7.
- J Clin Oncol 2004;22:229-37.
- J Clin Oncol 2004;22:1209-14.
- N Engl J Med 2004;351:337-45.
- Hematol Oncol Clin North Am 2002;16:1041-63.
- Proc Am Soc Clin Oncol 2003;22:204a (Abstract 817).
- J Clin Oncol 2005;23:1-8.
- J Clin Oncol 2004;22 Suppl:247s (Abstract).
- N Engl J Med 2004;350:2335-42. Proc Gastrointestinal Cancers Symposium 2005;168 (Abstract).
- Proc Gastrointestinal Cancers Symposium 2005;168 (Abstract).
- Br J Cancer 2004;90:8-11.
- Pharmaco‑genomics 2004;5:835-43.
Colorectal cancer. Facts and figures, Special edition 2005. American Cancer Society.
Colorectal Cancer Network
National Cancer Institute