Loading
START-A is the dose-finding half of the British experiment that shortened breast radiotherapy. By giving thirteen larger treatments over the same five weeks at two different total doses, it measured how sensitive breast cancer is to the size of each dose, and the answer is what made three weeks and later one week possible.
START-A and START-B were run together in 35 United Kingdom radiotherapy centres from 1999 to 2002. START-A kept the overall time at five weeks and varied only the fraction size, comparing 41.6 Gy or 39 Gy in thirteen fractions against the international standard of 50 Gy in twenty-five. Holding the time constant is what makes it a radiobiology experiment rather than a convenience trial: it isolates the effect of fraction size.
At a median 9.3 years, 139 local-regional relapses had occurred among the 2,236 women. Ten-year local-regional relapse was 6.3 percent with 41.6 Gy (95 percent confidence interval 4.7 to 8.5) against 7.4 percent with 50 Gy (5.5 to 10.0), a hazard ratio of 0.91 (0.59 to 1.38, p=0.65), and 8.8 percent with 39 Gy (6.7 to 11.4), a hazard ratio of 1.18 (0.79 to 1.76, p=0.41). Moderate or marked breast induration, telangiectasia and breast oedema were significantly less common after 39 Gy than after 50 Gy; normal-tissue effects after 41.6 Gy and 50 Gy did not differ.
The number that mattered is the one the two doses bracket. Tumour control was slightly better at 41.6 Gy and slightly worse at 39 Gy, while normal tissue improved steadily as the dose fell, which means breast cancer responds to fraction size much as normal breast tissue does. That sensitivity, a low alpha to beta ratio, is the physics behind START-B's 40 Gy in fifteen fractions and FAST-Forward's 26 Gy in five.
Numbers are from the trial as recorded here; see the source links in the table below. This is orientation, not medical advice: ask your team how closely the trial population matches you.
2,236 randomised.
95 percent confidence interval 4.7 to 8.5 · 95 percent confidence interval 6.7 to 11.4 · 95 percent confidence interval 5.5 to 10.0
Source| Endpoint | Arm | n | Value | HR (95% CI) | p | Source |
|---|---|---|---|---|---|---|
| Local-regional relapse at 10 yearsprimary | 41.6 Gy in 13 fractions | - | 6.3% | - | - | link |
| 39 Gy in 13 fractions | - | 8.8% | ||||
| 50 Gy in 25 fractions | - | 7.4% |
Shares START-B (UK Standardisation of Breast Radiotherapy), FAST-Forward, Hypofractionation (fewer, larger radiotherapy doses), HER2-positive breast cancer.
Shares FAST-Forward, Hypofractionation (fewer, larger radiotherapy doses), Hypofractionated radiotherapy, HER2-positive breast cancer.
Shares Hypofractionation (fewer, larger radiotherapy doses), Hypofractionated radiotherapy, HER2-positive breast cancer, Breast cancer (all types).
Shares Hypofractionated radiotherapy, The Institute of Cancer Research, HER2-positive breast cancer, Breast cancer (all types).
Shares Hypofractionated radiotherapy, HER2-positive breast cancer, Breast cancer (all types), IMRT / IGRT (modern external beam).
Shares Hypofractionation (fewer, larger radiotherapy doses), HER2-positive breast cancer, Breast cancer (all types), IMRT / IGRT (modern external beam).
Shares Hypofractionated radiotherapy, HER2-positive breast cancer, Breast cancer (all types), IMRT / IGRT (modern external beam).
Shares Alpha/beta ratio, Hypofractionation (fewer, larger radiotherapy doses), Hypofractionated radiotherapy, The Institute of Cancer Research.