ACT I
UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil and mitomycin in epidermoid anal cancer (ACT I)
Phase 3N = 585
- Population
- Patients with epidermoid (squamous, basaloid or cloacogenic) carcinoma of the anal canal or anal margin, staged clinically with the 1985 UICC TNM classification. Anal margin cancers and T1 cancers, usually excluded from other randomised trials, made up 23% and 13% of the trial population.
- Design
- Multicentre randomised trial of the UK Co-ordinating Committee on Cancer Research. Of 856 patients considered for entry, 585 were randomised (1:1; December 1987 to March 1994, 56 centres): 290 to 45 Gy radiotherapy in 20 or 25 fractions over 4 to 5 weeks, and 295 to the same radiotherapy plus infused fluorouracil and mitomycin. Clinical response was assessed 6 weeks after initial treatment; good responders were recommended for boost radiotherapy and poor responders for salvage surgery. Main endpoint local failure from 6 weeks after initial treatment; secondary endpoints overall and cause-specific survival. A 13-year follow-up report (median 13.1 years) re-analysed 577 eligible patients.
- Primary endpoint
- Local failure
- Arms
- Radiotherapy alone (control, n = 290): 45 Gy in 20 or 25 fractions over 4 to 5 weeks; boost radiotherapy recommended for good responders and salvage surgery for poor responders at the 6-week assessment
- Radiotherapy with fluorouracil and mitomycin (experimental, n = 295): The same radiotherapy plus fluorouracil (1000 mg/m2 for 4 days or 750 mg/m2 for 5 days) by continuous infusion in the first and final weeks and mitomycin 12 mg/m2 on day 1 of the first course
- People
- UKCCCR Anal Cancer Trial Working Party, UKCCCR Anal Cancer Trial Working Party (UK Co-ordinating Committee on Cancer Research)
Δ Local failure−23 pointsRFSHR 0.70 (95% CI 0.58–0.84)Δ median RFS+3.3 yearsΔ 60-month RFS+12.9 pointsLocoregional failureHR 0.46 (95% CI 0.35–0.60)Δ 60-month Locoregional failure−24.8 points
Show the numbers
| Endpoint | Analysis | Radiotherapy with fluorouracil and mitomycin | Radiotherapy alone | Effect |
|---|---|---|---|---|
| Local failurePatients analysed for local failure: 279 radiotherapy and 283 combined modality therapy, of 585 randomised; the abstract states that analysis was by intention to treatLocal failure from 6 weeks after initial treatment. The 2010 follow-up report describes the original definition as residual or recurrent locoregional disease (biopsy or clinical determination), colostomy for treatment-related morbidity, or a pretreatment colostomy still open 6 months after treatment. | primarymedian follow-up 42 monthsUKCCCR 1996 | 36% | 59% | P<0.0001Local failure in 164 of 279 radiotherapy patients and 101 of 283 combined modality patients; median follow-up 42 months (interquartile range 28-62). The abstract reports a 46% reduction in the risk of local failure, relative risk 0.54 (95% CI 0.42-0.69). Risk of death from anal cancer 0.71 (0.53-0.95), p = 0.02. No overall survival advantage: 0.86 (0.67-1.11), p = 0.25. Early morbidity was significantly more frequent with combined modality therapy (p = 0.03); late morbidity occurred at similar rates. The abstract labels only 0.54 'relative risk' and does not name the measure of the 0.71 and 0.86 values; the 2010 long-term report calls the 1996 values 0.54, 0.71 and 0.86 hazard ratios. They stay out of the hazard-ratio field because the 1996 abstract's own label is 'relative risk'. |
| RFS577 eligible randomised patients (285 radiotherapy alone, 292 combined modality therapy)Time to any first relapse or death (13-year follow-up report). | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | median 4.6 years; 46.6% at 5 y; 29.7% at 12 y | median 1.3 years; 33.7% at 5 y; 17.7% at 12 y | HR 0.70 (95% CI 0.58–0.84)P<0.001Hazard ratio for combined modality therapy versus radiotherapy alone. Landmarks are the 5-year and 12-year relapse-free survival rates from the report's efficacy table; the report gives the absolute risk difference as +12.9% at 5 years. Data on relapse and death were collected until October 2007. |
| Locoregional failure560 of 577 eligible patients, after excluding 17 (8 radiotherapy alone, 9 combined modality) who did not survive 6 weeks beyond the end of initial radiotherapyTime to first locoregional relapse (disease in the anal canal, margin or pelvic region), analysed from 6 weeks after the end of initial treatment (13-year follow-up report). | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | 32.3% at 5 y; 33.8% at 12 y | 57.1% at 5 y; 59.1% at 12 y | HR 0.46 (95% CI 0.35–0.60)P<0.001Hazard ratio for combined modality therapy versus radiotherapy alone. Landmarks are the 5-year and 12-year locoregional relapse rates. The abstract reports that, 12 years after treatment, for every 100 patients given chemoradiation there are an expected 25.3 fewer with locoregional relapse (95% CI 17.5-32.0 fewer) than with 100 given radiotherapy alone. Only 11 patients had locoregional relapse as a first event after 5 years. |
| OS577 eligible randomised patients (285 radiotherapy alone, 292 combined modality therapy) | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | median 7.6 years (5.9–9.9); 58.1% at 5 y; 41.5% at 10 y | median 5.4 years (3.6–6.8); 53% at 5 y; 35.8% at 10 y | HR 0.86 (95% CI 0.70–1.04)P=0.12Hazard ratio for combined modality therapy versus radiotherapy alone; the interval includes 1. Landmarks are the 5-year and 10-year overall survival rates. Deaths from causes other than anal cancer were more frequent with chemoradiation in the first years: the abstract reports a 9.1% increase in non-anal cancer deaths in the first 5 years (95% CI +3.6 to +14.6) that disappeared by 10 years; deaths from second malignancies were 34 vs 18 (P = 0.03), chemoradiation vs radiotherapy alone. |
| Death from anal cancer577 eligible randomised patients (285 radiotherapy alone, 292 combined modality therapy) | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | 30.5% at 5 y; 36.2% at 12 y | 41.8% at 5 y; 48.7% at 12 y | HR 0.67 (95% CI 0.51–0.88)P=0.004Hazard ratio for combined modality therapy versus radiotherapy alone. Landmarks are the 5-year and 12-year anal cancer death rates. The abstract reports 12.5 fewer anal cancer deaths per 100 patients treated with chemoradiation at 12 years (95% CI 4.3-19.7 fewer). |
| Colostomy-free survival577 eligible randomised patients (285 radiotherapy alone, 292 combined modality therapy)Time to colostomy or death; all colostomies, whether for disease or for morbidity, were included. Colostomy data were collected up to 2000. | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | median 4.7 years; 46.9% at 5 y; 29.6% at 12 y | median 1.8 years; 36.8% at 5 y; 20.1% at 12 y | HR 0.76 (95% CI 0.63–0.91)P=0.004Hazard ratio for combined modality therapy versus radiotherapy alone. Landmarks are the 5-year and 12-year colostomy-free survival rates. The report states that no colostomies were recorded after November 1995 and that colostomy information was not routinely collected after 2000, which it lists as a limitation. |
| Endpoint | Analysis | Radiotherapy with fluorouracil and mitomycin | Radiotherapy alone | Effect |
|---|---|---|---|---|
| Local failurePatients analysed for local failure: 279 radiotherapy and 283 combined modality therapy, of 585 randomised; the abstract states that analysis was by intention to treatLocal failure from 6 weeks after initial treatment. The 2010 follow-up report describes the original definition as residual or recurrent locoregional disease (biopsy or clinical determination), colostomy for treatment-related morbidity, or a pretreatment colostomy still open 6 months after treatment. | primarymedian follow-up 42 monthsUKCCCR 1996 | 36% | 59% | P<0.0001Local failure in 164 of 279 radiotherapy patients and 101 of 283 combined modality patients; median follow-up 42 months (interquartile range 28-62). The abstract reports a 46% reduction in the risk of local failure, relative risk 0.54 (95% CI 0.42-0.69). Risk of death from anal cancer 0.71 (0.53-0.95), p = 0.02. No overall survival advantage: 0.86 (0.67-1.11), p = 0.25. Early morbidity was significantly more frequent with combined modality therapy (p = 0.03); late morbidity occurred at similar rates. The abstract labels only 0.54 'relative risk' and does not name the measure of the 0.71 and 0.86 values; the 2010 long-term report calls the 1996 values 0.54, 0.71 and 0.86 hazard ratios. They stay out of the hazard-ratio field because the 1996 abstract's own label is 'relative risk'. |
| RFS577 eligible randomised patients (285 radiotherapy alone, 292 combined modality therapy)Time to any first relapse or death (13-year follow-up report). | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | median 4.6 years; 46.6% at 5 y; 29.7% at 12 y | median 1.3 years; 33.7% at 5 y; 17.7% at 12 y | HR 0.70 (95% CI 0.58–0.84)P<0.001Hazard ratio for combined modality therapy versus radiotherapy alone. Landmarks are the 5-year and 12-year relapse-free survival rates from the report's efficacy table; the report gives the absolute risk difference as +12.9% at 5 years. Data on relapse and death were collected until October 2007. |
| Locoregional failure560 of 577 eligible patients, after excluding 17 (8 radiotherapy alone, 9 combined modality) who did not survive 6 weeks beyond the end of initial radiotherapyTime to first locoregional relapse (disease in the anal canal, margin or pelvic region), analysed from 6 weeks after the end of initial treatment (13-year follow-up report). | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | 32.3% at 5 y; 33.8% at 12 y | 57.1% at 5 y; 59.1% at 12 y | HR 0.46 (95% CI 0.35–0.60)P<0.001Hazard ratio for combined modality therapy versus radiotherapy alone. Landmarks are the 5-year and 12-year locoregional relapse rates. The abstract reports that, 12 years after treatment, for every 100 patients given chemoradiation there are an expected 25.3 fewer with locoregional relapse (95% CI 17.5-32.0 fewer) than with 100 given radiotherapy alone. Only 11 patients had locoregional relapse as a first event after 5 years. |
| OS577 eligible randomised patients (285 radiotherapy alone, 292 combined modality therapy) | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | median 7.6 years (5.9–9.9); 58.1% at 5 y; 41.5% at 10 y | median 5.4 years (3.6–6.8); 53% at 5 y; 35.8% at 10 y | HR 0.86 (95% CI 0.70–1.04)P=0.12Hazard ratio for combined modality therapy versus radiotherapy alone; the interval includes 1. Landmarks are the 5-year and 10-year overall survival rates. Deaths from causes other than anal cancer were more frequent with chemoradiation in the first years: the abstract reports a 9.1% increase in non-anal cancer deaths in the first 5 years (95% CI +3.6 to +14.6) that disappeared by 10 years; deaths from second malignancies were 34 vs 18 (P = 0.03), chemoradiation vs radiotherapy alone. |
| Death from anal cancer577 eligible randomised patients (285 radiotherapy alone, 292 combined modality therapy) | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | 30.5% at 5 y; 36.2% at 12 y | 41.8% at 5 y; 48.7% at 12 y | HR 0.67 (95% CI 0.51–0.88)P=0.004Hazard ratio for combined modality therapy versus radiotherapy alone. Landmarks are the 5-year and 12-year anal cancer death rates. The abstract reports 12.5 fewer anal cancer deaths per 100 patients treated with chemoradiation at 12 years (95% CI 4.3-19.7 fewer). |
| Colostomy-free survival577 eligible randomised patients (285 radiotherapy alone, 292 combined modality therapy)Time to colostomy or death; all colostomies, whether for disease or for morbidity, were included. Colostomy data were collected up to 2000. | long term-updatemedian follow-up 13.1 yearsNorthover 2010 | median 4.7 years; 46.9% at 5 y; 29.6% at 12 y | median 1.8 years; 36.8% at 5 y; 20.1% at 12 y | HR 0.76 (95% CI 0.63–0.91)P=0.004Hazard ratio for combined modality therapy versus radiotherapy alone. Landmarks are the 5-year and 12-year colostomy-free survival rates. The report states that no colostomies were recorded after November 1995 and that colostomy information was not routinely collected after 2000, which it lists as a limitation. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| RFS | 0.7 | 0.58–0.84 |
| Locoregional failure | 0.46 | 0.35–0.6 |
| OS | 0.86 | 0.7–1.04 |
| Death from anal cancer | 0.67 | 0.51–0.88 |
| Colostomy-free survival | 0.76 | 0.63–0.91 |
- Radiotherapy alone (control)
- Radiotherapy with fluorouracil and mitomycin (experimental)
Data table (modelled values)
| Months | Radiotherapy alone | Radiotherapy with fluorouracil and mitomycin |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 39 | 52.0% | 62.2% |
| 78 | 26.2% | 39.2% |
| 117 | 20.4% | 33.1% |
| 156 | 17.7% | 29.7% |
- Radiotherapy alone (control)
- Radiotherapy with fluorouracil and mitomycin (experimental)
Data table (modelled values)
| Months | Radiotherapy alone | Radiotherapy with fluorouracil and mitomycin |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 32 | 73.0% | 76.0% |
| 65 | 50.5% | 55.8% |
| 97 | 41.0% | 46.7% |
| 130 | 35.8% | 41.5% |
Appears in
Sources
UKCCCR Anal Cancer Trial Working Party. Epidermoid anal cancer: results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and mitomycin. Lancet. 1996;348(9034):1049-54.
Northover J, Glynne-Jones R, Sebag-Montefiore D, James R, Meadows H, Wan S, et al. Chemoradiation for the treatment of epidermoid anal cancer: 13-year follow-up of the first randomised UKCCCR Anal Cancer Trial (ACT I). Br J Cancer. 2010;102(7):1123-8.