Improving patient survival with the colorectal cancer multi-disciplinary team


  • MacDonald M provided substantial statistical analysis, and co-authored; Hooton G carried out initial patient data collection; McKay G provided statistical analysis; Grose D carried out oncological data collection; Mohammed N was the lead oncological clinician for the patient group after MDT, and co-authored; Porteous C was the lead surgical clinician for all patients, was responsible for the paper’s inception, and co-authored.

  • This paper is a continuation of a study presented to The European Society of Coloproctology in Malta, September 2007, published as the abstract: MacDermid E, Mackay G, MacDonald M, Hooton G, Porteous C. The Colorectal Cancer Multi-Disciplinary Team Meeting Fails To Improve Patient Survival. Colorectal Dis 2007 Volume 9, Supplement 3, October 2007.

Ewan MacDermid, 9 Terry Close, Swinger Hill, ACT 2606, Australia.


Objective  There is little information on the impact of the colorectal multi-disciplinary team (MDT) in the United Kingdom. Our single operator presented his patients before and after the inception of an MDT meeting in June 2002. The aim of this study was to assess the effect of this on his patients’ survival, and trends in the use of adjuvant chemotherapy.

Method  Data were collected on all patients (n = 310) undergoing colectomy for colorectal cancer by one surgeon. Excluding patients with Dukes A stage, the pre-MDT cohort from January 1997 to May 2002 was 176 and the post-MDT cohort from June 2002 to December 2005 was 134. Three-year survival rates were calculated using Kaplan–Meier life table analysis. Prognostic factors were analysed using Cox-proportional hazard regression, and chemotherapy data analysed using the chi-squared test. Independent prognostic indicators of chemotherapy prescription were examined using binary logistic testing.

Results  MDT status was shown to be an independent predictor of survival on hazard regression analysis (P = 0.044). A significantly greater number of patients were prescribed adjuvant chemotherapy in the post-MDT cohort (P = 0.0002). MDT status was shown to be a significant prognostic indicator of chemotherapy prescription (P < 0.0001). Three-year survival for Dukes C patients was 58% in the pre-MDT group, and 66% in the post-MDT group (P = 0.023).

Conclusion  There was a significant increase in patients undergoing adjuvant postoperative chemotherapy after the inception of the MDT. This was associated with a significant survival benefit in patients with Dukes C disease. The data suggest that the MDT process has resulted in an increase in the prescription of adjuvant chemotherapy, with 3-year survival being greater after its inception.


The multi-disciplinary team (MDT) has evolved over the last 7 years to become the accepted forum for decision-making for patients with cancer in the United Kingdom. There is evidence that the MDT process improves patient selection and overall survival in oesophageal, gastric, hepato-biliary, breast and ovarian cancer [1–4]. The implementation of the colorectal cancer MDT was prompted by recommendations by the Calman–Hine report in 1995 [5,6], and a survey performed in 2002 showed that 82% of lead clinicians in English cancer networks had access to an MDT meeting for their colorectal cancer patients [7]. There is little published evidence examining whether the colorectal cancer MDT meeting improves survival.

Our district general hospital’s weekly colorectal cancer MDT meeting includes the department’s colorectal surgeons, a consultant radiologist with an interest in magnetic resonance imaging (MRI), a consultant pathologist, a specialist colorectal consultant clinical oncologist, colorectal nurse specialists and the MDT audit clerk.

One consultant colorectal surgeon (CP) has been operating at the hospital since 1996 and has routinely presented to an MDT meeting since June 2002 all his patients with Dukes B and C stage and metastatic disease after primary surgery. This date also coincided with the appointment of a specialist colorectal clinical oncologist. Before the introduction of the MDT meeting, patients were discussed on an ad hoc basis with an oncologist. We initially performed an analysis on the post-MDT cohort of patients from June 2002 to March 2007 and found a nonsignificant trend towards improved survival compared with patients treated before the MDT was initiated [8]. In this study, we reviewed the results on patients treated from June 2002 to December 2005, giving a longer period for the assessment of survival.


Three hundred and ten patients with colorectal cancer resection excluding those with Dukes A stage performed by or supervised by our single surgeon (CP) from January 1997 to December 2005 were included. Before the introduction of the MDT meeting, the data were collected retrospectively from theatre log-books and case notes. After the MDT was formed, this information was recorded prospectively by the West of Scotland Managed Clinical Network For Colorectal Cancer. Information recorded included hospital number, age at time of original surgery, gender, site of cancer, tumour node metastases (TNM) and Dukes stage from histological reports, the presence of metastatic disease detected on imaging or at surgery, whether the patient was referred to an oncologist after surgery and the type and quantity of adjuvant chemotherapy or radiotherapy.

Dates of death were obtained from the Registry of Births, Deaths and Marriages in Edinburgh and from the hospital’s patient record system. Only patients who had resection of the primary tumour, whether curative or not, were included in the study. All patients had open resection through a mid-line laparotomy, and the surgical technique did not change over the study period. Data on the presentation of adjuvant chemoradiotherapy were extracted from retrospective case notes (before 2002) and from a prospective database maintained by the regional oncology centre (after 2002). Information included the specific chemo- and radiotherapy regime and whether the patient completed it or not.

Statistical analysis

Patients were grouped into those operated before and after the introduction of the MDT (June 2002). Survival was calculated over the period from the date of surgery to October 2007. Kaplan–Meier life-table and the Log-rank test were used to express survival. The chi-squared test was used to assess the differences in categorical variables. Cox-proportional hazard analysis was used to determine the independent effect of the MDT on survival after adjusting for the following variables: gender, age, Dukes’ stage, site of cancer and year of surgery. The site of cancer was divided into right (caecum, ascending colon, hepatic flexure) and left (transverse colon, splenic flexure, descending colon). Sigmoid cancers were considered separately, and recto-sigmoid cancers were included as rectal cancer. Binary logistic regression analysis was performed to identify the factors that independently predicted the prescription of adjuvant chemotherapy. Med-Calc™ software was used in all the analyses. A P-value of less than 0.05 was taken as the level of significance.


Baseline characteristics

Data were analysed on 176 patients before the commencement of the MDT, and on 134 after. Their baseline characteristics are shown in Table 1. The mean age of the pre-MDT group was 69.9 years (median: 70.7), and of the post-MDT group 69.5 years (median: 70.6). The respective gender ratios (male:female) were 1:0.83 and 1:1.16.

Table 1.   Baseline characteristics of patients before and after commencement of MDT meeting.
Female %45.553.7
Age mean (median) years69.94 (70.74)69.5 (70.6)
Peri-operative death (%) 6 (4.5%) 8 (6%)
Dukes B (%)67 (38%)51 (38.1%)
Dukes C (%)69 (39.2%)58 (43.3%)
Metastatic disease (%)40 (23.3%)25 (18.6%)
Right sided cancer (%)47 (26.1%)38 (28.4%)
Left sided cancer (%)17 (9.4%)16 (11.9%)
Sigmoid cancer (%)28 (15.9%)40 (29.9%)
Rectal cancer (%)86 (48.9%)40 (29.9%)
Receiving palliative chemo (%)22 (12.5%)13 (9.7%)
Receiving adjuvant chemo (%)23 (13%)42 (31.3%)


Among Dukes B patients, 3-year survival was 76% before the MDT and 70% after (P = 0.486) (Fig. 1). In patients with Dukes C stage, 3-year survival was calculated at 58% before the MDT and 66% after (P = 0.023) (Fig. 2). Median survival for patients with metastatic disease operated before the MDT was 8 months, compared with 11.9 months after (P = 0.234) (Fig. 3).

Figure 1.

 Kaplan Meier analysis of patients with Dukes B disease.

Figure 2.

 Kaplan Meier analysis of patients with Dukes C disease.

Figure 3.

 Kaplan Meier analysis of patients with metastatic disease.

The Cox-proportional hazards model showed age (P = 0.003), Dukes stage (P < 0.0001) and MDT status (P = 0.044) to be independent predictors of mortality (Table 2).

Table 2.   Cox-propotional hazard model examining the influence of independent variables on overall survival.
VariableHazard ratio95% confidence intervalsP-value
Dukes stage2.582.10–3.18 < 0.0001
Site of cancer0.950.85–1.060.362


Significantly more patients were prescribed primary adjuvant chemotherapy after inception of the MDT (31.3%vs 13%, P = 0.0002) (Dukes B, P = 0.0019; Dukes C, P = 0.0039). There was no significant increase in the number of patients with metastatic disease being prescribed palliative chemotherapy before and after (P = 0.431). Data were unobtainable on 10 patients (Table 3). Binary logistic regression analysis demonstrated that Dukes’ stage (P < 0.001) and MDT status (P < 0.001) were independent predictors of chemotherapy prescription (Table 4). The average ages of patients receiving adjuvant chemotherapy before and after the inception of the MDT were 65.7 years (median: 62.6) and 65.2 years (median: 67.3).

Table 3.   Proportions of patients receiving chemotherapy in the pre- and post-MDT cohorts.
StagePrescribed adjuvant CCompleted adjuvant CPrescribed palliative CRecords unavailable
  1. C, chemotherapy.

 B (n = 67)1 (1.5%) 01 (1.5%)2 (3%)
 C (n = 69)22 (31.9%)15 (21.7%)8 (11.6%)0
 Metastatic (n = 40)   13 (32.5%)3 (7.5)
 B (n = 51)9 (17.6%)5 (9.8%) 02 (3.9%)
 C (n = 58)34 (58.6%)25 (43.1%)1 (1.7%)3 (5.1%)
 Metastatic (n = 25)  11 (44%)0
Table 4.   Binary logistic model demonstrating association of independent variables with prescription of primary adjuvant chemotherapy.
VariableCoefficientOdds ratioP-value95% CI

Preoperative radiotherapy

Before MDT 21 (24.4%) patients of 86 with rectal cancer received preoperative radiotherapy, all having the short course (5 × 5 Gy) regime. After MDT, all rectal cancer patients were staged with thin slice MRI and were treated with long course chemo-radiotherapy if the mesorectal margin was threatened or involved. In the post-MDT cohort, 13 (32.5%) patients of 40 were given preoperative radiotherapy (P = 0.462).


Meta-analysis has shown a survival benefit with adjuvant postoperative chemotherapy, particularly in patients with a Dukes C lesion [9,10], and this has been well recognized for at least 17 years [11,12]. Improved surgical technique and postoperative care have also contributed to an improvement in colorectal cancer disease-free and overall survival rates [13]. Although our study was small and subject to the hazards of retrospective analysis [14], we managed to demonstrate a survival benefit since the commencement of the MDT meeting and the significant increase in patients receiving adjuvant chemotherapy, particularly in Dukes C disease. Our hypothesis is that the MDT meeting has facilitated access to oncological care and has thereby increased the numbers being offered adjuvant chemotherapy. This is reassuring, given the considerable amount of time and resources which are allocated to maintaining the MDT process. The benefits of adjuvant chemotherapy had been known for some time [12], which would have been expected to be reflected in higher rates of chemotherapy in our pre-MDT cohort, but our data suggest that it was the inception of the MDT meeting which resulted in an increase. The study also demonstrated a trend towards oncology sub-specialization after the inauguration of the MDT meeting and the increasing use of MRI for preoperative staging. There is no suggestion that older patients are being put forward for adjuvant therapy, and this is borne out by the almost identical mean age in the two cohorts undergoing chemotherapy (65.7 vs 65.2). Although in recent years oxaliplatin containing compounds have been shown to offer a survival advantage over 5-FU [15], this was not a confounding factor in our study as almost none of our patients in either cohort received it as primary adjuvant chemotherapy. One striking aspect of the two cohorts was a drop in the number of rectal cancers being resected by our operator, although this has may be due to a change in the classification of cancers. It may also reflect changes in the workload and staffing at our department. It has been impossible to assess whether the associated increase in adjuvant chemotherapy has affected patient quality of life with this retrospective study. Other factors we were unable to examine were the proportion of patients requiring emergency surgery, and those who were referred at a later date for hepatic or pulmonary metastectomy. These factors have been shown to be independent prognostic indicators of overall survival [16–18]. Due to a lack of data in the pre-MDT cohort, we were also unable to examine the effect of the MDT on patients who were deemed unsuitable for any resection. It may be that the MDT has its greatest effect in allocating such patients to palliative treatment or surgery.

Along with the benefit in overall survival, there has been an improvement in the patient management process with the advent of the MDT meeting. This may be hard to quantify statistically, but it has been recognized that improved communication between different specialities makes the patient management process more efficient and helps prevent unnecessary investigation and interventions [19]. One extremely valuable aspect of the MDT meeting is that it facilitates multi-disciplinary review of imaging such as CT and MRI, although this is highly dependent on local resources and facilities [20]. It also provides a valuable training opportunity for junior doctors and other specialists.


We thank Fiona Marjoriebanks, Inverclyde Royal Infirmary Audit Office, the staff of the Registry of Births, Deaths and Marriages (Scotland), and the medical records staff of the Royal Alexandra Hospital and Beatson West of Scotland Cancer Centre.