LBA518 Background: The START Trials (ST-A and ST-B) test the hypothesis that breast cancer is as sensitive to fraction (Fr) size as late reacting normal tissues, with an a/β value of about 4Gy. Methods: The phase III randomised START Trials tested hypofractionated post-operative RT in women with completely excised invasive breast cancer (T1–3, N0–1, M0). Centres opted for either ST-A or ST-B. ST-A tested 50Gy in 25Fr (5 wks) vs 41.6Gy vs 39Gy, both in 13Fr (5 wks). ST-B tested 50Gy in 25Fr (5 wks) vs 40Gy in 15Fr (3 wks). Stratification was by centre, surgery and boost. The primary endpoint was local-regional (LR) relapse. Late normal tissue effects (NTE) were assessed by breast photographs, clinical examination and quality of life (QL) questionnaires. Survival analysis methods were used to estimate rates of relapse and NTEs, and hazard ratios (HR) (with 95%CI). Smoothed estimates of absolute differences in relapse rates were obtained from the rates in the 50Gy control arms and the HR. Results: 2236 (ST-A) and 2215 (ST-B) patients were recruited from 35 UK centres during 1999–2002. Median follow-up is 5.1 years (ST-A) and 6.0 years (ST-B). There were 93 LR relapses in ST-A (4.1% at 5 years, 3.2- 5.0%), with absolute differences in LR relapse rates at 5 years compared with 50Gy of 0.2% (−1.3%−2.6%) for 41.6Gy and 0.9% (−0.8%−3.7%) for 39Gy. The a/β estimate for tumour control was 5.0Gy (−2.7–12.7). In ST-B, there were 65 LR relapses (2.8% at 5 years, 2.1–3.5%), with an absolute difference in LR relapse rates at 5 years of −0.6% (−1.7%−0.9%) for 40Gy vs 50Gy. In ST-A the rate of mild/marked change in photographic breast appearance was lower in 39Gy vs 50Gy (HR 0.69, 0.52–0.91), and similarly for 40Gy vs 50Gy in ST-B (HR 0.83, 0.66–1.04). The a/β estimate for change in breast appearance was 3.1Gy (1.6–4.6). Rates of induration, telangiectasia and breast oedema were lower in 39Gy (ST-A) and 40Gy (ST-B) compared with the 50Gy arms. QL results were consistent with the clinical findings. Conclusions: The fractionation sensitivity of breast cancer is comparable to that of late reacting normal tissues, confirming the results of a recent pilot trial. These results support the use of hypofractionated RT schedules for early breast cancer. No significant financial relationships to disclose.
The UK Standardisation of Breast Radiotherapy (START) Trial B of radiotherapy hypofractionation for treatment of early breast cancer: a randomised trial