Mitomycin/5-FU with concurrent radiation therapy remains the standard of care for anal cancer, however, the use of cisplatin/5-FU is also a reasonable approach in some patients for instance those with HIV or in whom the toxicity of mitomycin would be considered unacceptable.
A large prospective randomised trial, RTOG 98-11,r compared mitomycin/5-FU in combination with radiation therapy, directly to cisplatin/5-FU in combination with radiation therapy. 682 patients with anal canal carcinoma were randomised to either mitomycin 10 mg/m2 day 1 and day 29 combined with infusional 5-FU 1000 mg/m2/day day 1 to 4 or cisplatin 75 mg/m2 day 1, day 29, day 57 and day 85 combined with infusional 5-FU 1000 mg/m2/day day 1 to 4, day 29 to 32, day 57 to 60 and day 85 to 88. Both regimens were given radiation therapy at a minimum dose of 45 Gy in 25 fractions. Patients with established AIDS were excluded.r
The primary endpoint was 5-year disease-free survival (DFS). Secondary end points were overall survival (OS) and time to relapse.
ACT II,r a large randomised multicentre trial investigated the role of replacing mitomycin with cisplatin in improving responses with acceptable toxicity and whether maintenance treatment after chemoradiation improves survival. Between June 2001, and December 2008, 940 patients with histologically proven anal squamous, basaloid or cloacogenic carcinoma were equally randomised in a 2x2 factorial design to receive 5-FU 1000 mg/m2/day days 1 to 4 and days 29 to 32 and either mitomycin 12 mg/m2 day 1 or cisplatin 60 mg/m2 day 1 and day 29. Radiation therapy of 50.4 Gy was delivered in 28 daily fractions over 5.5 weeks with a two-phase technique. 4 weeks after chemoradiation, patients were also randomised to receive maintenance therapy (2 cycles of cisplatin and 5-FU weeks 11 and 14) or no maintenance.
The primary endpoints for comparing mitomycin with cisplatin were complete response (26 weeks from start of chemoradiation) and grade 3 or 4 toxicities (4 weeks after chemoradiation); and for comparing maintenance with no maintenance was progression-free survival (PFS). Secondary endpoints were colostomy-free survival (CFS), recurrence rate, cause-specific survival and OS.
Efficacy
RTOG 98-11rr
At the time of initial analysis, the median follow-up for all patients was 2.51 years. Although the 5 year DFS and OS rates were similar between both arms, the cumulative colostomy rate was significantly higher in the cisplatin/5-FU arm (10% versus 19%; p=0.02).r
Long-term analysis performed on 649 eligible patients out of the 682 accrued patients demonstrated significantly enhanced 5 year DFS and OS rates in the mitomycin/5-FU group compared to the cisplatin/5-FU group. There was a trend towards statistical significance for 5 year locoregional failure (LRF), CFS, colostomy failure (CF) and distant metastasis (DM) rates for patients treated with mitomycin/5-FU versus cisplatin/5-FU.r
Gundersonr |
Mitomycin/5-FU |
Cisplatin/5-FU |
p-value |
5 year DFS (%, 95% CI) |
67.8 (62.3-72.6)
|
57.8 (52.1-63.0)
|
0.006 |
5 year OS (%, 95% CI) |
78.3 (73.2-82.5)
|
70.7 (65.2-75.4)
|
0.026 |
5 year LRF (%, 95% CI) |
20 (15.6-24.4)
|
26.4 (21.5-31.3)
|
0.087 |
5 year CFS (%, 95% CI) |
71.9 (66.5-76.5)
|
65 (59.4-70.0)
|
0.05 |
5 year CF (%, 95% CI) |
11.9 (8.3-15.4)
|
17.3 (13.1-21.5)
|
0.074 |
5 year DM (%, 95% CI) |
13.1 (9.3-16.8)
|
18.1 (13.8-22.4)
|
0.12 |
Disease-free survival (A) and overall survival (B)r
© J Clin Oncol 2012
ACT IIr
The median follow-up was 5.1 years. 77% mitomycin group (365/472) and 72% cisplatin group (338/468) completed chemoradiation without delays or dose reductions. 46% mitomycin group (105/226) and 41% cisplatin group (91/222) completed both courses of maintenance treatment without delays or dose reductions. There were no significant differences in the complete responses (absolute difference -0.9%, 95% CI -4.9 to 3.1; p=0.64) between the mitomycin and the cisplatin arms. PFS did not differ significantly between patients who took mitomycin and cisplatin during chemoradiation (HR 0.95, 95% CI 0.75–1.19; p=0.63). The addition of maintenance treatment did not improve the 3 year PFS. The 3 year PFS was better in patients with T1-T2 and node negative patients in comparison with T3-T4 and node positive patients. There was no significant difference in the 3 year OS between mitomycin and cisplatin arms and the maintenance and no maintenance arms.
Primary tumour response at 26 weeksr |
Mitomycin group
(n=432) |
Cisplatin group
(n=431) |
Complete response |
391 (90.5%) |
386 (89.6%) |
Partial response |
14 (3.2%) |
24 (5.6%) |
Stable disease |
5 (1.2%) |
6 (1.4%) |
Progressive disease |
22 (5.1%) |
15 (3.5%) |
Efficacyr |
Mitomycin
maintenance |
Cisplatin
maintenance |
Mitomycin,
no maintenance |
Cisplatin,
no maintenance |
3 year PFS (%, 95% CI) |
73 (66-78) |
74 (68-80) |
73 (67-78) |
72 (66-78) |
3 year CFS (%, 95% CI) |
73 (67-79) |
75 (68-80) |
75 (68-80) |
72 (66-77) |
3 year OS (%) |
82 |
83 |
86 |
84 |
Progression free survival (A) maintenance versus non maintenance (B) all four groupsr
© Lancet Oncol 2013
Anal cancer mortality (A) cisplatin versus mitomycin (B) all four groupsr
© Lancet Oncol 2013
© Lancet Oncol 2013
Toxicity
RTOG 98-11r
The rate of grade 3 or 4 overall late treatment-related toxicities was not statistically different between the mitomycin- and cisplatin-based groups (13.1% versus 10.7%; p=0.35). Haematologic grade 3 or 4 toxicities showed a significantly greater incidence in the mitomycin-based group (61.8%) compared to the cisplatin-based group (42.0%) (p < 0.001).r
Treatment-related adverse eventsr
© J Clin Oncol 2012
ACT IIr
Patients in the mitomycin arm had more acute grade 3/4 haematological toxicities (26% versus 16%; p< 0.001) but no higher rates of febrile neutropenia (3.1% versus 3.2%; p=0.93). Non-haematologic grade 3/4 toxicities were similar between arms (61% versus 65%; p=0.22).
Treatment-related adverse eventsr
© Lancet Oncol 2013