Quality of life after breast-conserving therapy and adjuvant radiotherapy for non-low-risk ductal carcinoma in situ (BIG 3-07/TROG 07.01): 2-year results of a randomised, controlled, phase 3 trial

Abstract

Background: BIG 3-07/TROG 07.01 is an international, multicentre, randomised, controlled, phase 3 trial evaluating tumour bed boost and hypofractionation in patients with non-low-risk ductal carcinoma in situ following breast-conserving surgery and whole breast radiotherapy. Here, we report the effects of diagnosis and treatment on health-related quality of life (HRQOL) at 2 years.

Methods: The BIG 3-07/TROG 07.01 trial is ongoing at 118 hospitals in 11 countries. Women aged 18 years or older with completely excised non-low-risk ductal carcinoma in situ were randomly assigned, by use of a minimisation algorithm, to tumour bed boost or no tumour bed boost, following conventional whole breast radiotherapy or hypofractionated whole breast radiotherapy using one of three randomisation categories. Category A was a 4-arm randomisation of tumour bed boost versus no boost following conventional whole breast radiotherapy (50 Gy in 25 fractions over 5 weeks) versus hypofractionated whole breast radiotherapy (42·5 Gy in 16 fractions over 3·5 weeks). Category B was a 2-arm randomisation between tumour bed boost versus no boost following conventional whole breast radiotherapy, and category C was a 2-arm randomisation between tumour bed boost versus no boost following hypofractionated whole breast radiotherapy. Stratification factors were age at diagnosis, planned endocrine therapy, and treating centre. The primary endpoint, time to local recurrence, will be reported when participants have completed 5 years of follow-up. The HRQOL statistical analysis plan prespecified eight aspects of HRQOL, assessed by four questionnaires at baseline, end of treatment, and at 6, 12, and 24 months after radiotherapy: fatigue and physical functioning (EORTC QLQ-C30); cosmetic status, breast-specific symptoms, arm and shoulder functional status (Breast Cancer Treatment Outcome Scale); body image and sexuality (Body Image Scale); and perceived risk of invasive breast cancer (Cancer Worry Scale and a study-specific question). For each of these measures, tumour bed boost was compared with no boost, and conventional whole breast radiotherapy compared with hypofractionated whole breast radiotherapy, by use of generalised estimating equation models. Analyses were by intention to treat, with Hochberg adjustment for multiple testing. This trial is registered with ClinicalTrials.gov, NCT00470236.

Findings: Between June 1, 2007, and Aug 14, 2013, 1208 women were enrolled and randomly assigned to receive no tumour bed boost (n=605) or tumour bed boost (n=603). 396 of 1208 women were assigned to category A: conventional whole breast radiotherapy with tumour bed boost (n=100) or no boost (n=98), or to hypofractionated whole breast radiotherapy with tumour bed boost (n=98) or no boost (n=100). 447 were assigned to category B: conventional whole breast radiotherapy with tumour bed boost (n=223) or no boost (n=224). 365 were assigned to category C: hypofractionated whole breast radiotherapy with tumour bed boost (n=182) or no boost (n=183). All patients were followed up at 2 years for the HRQOL analysis. 1098 (91%) of 1208 patients received their allocated treatment, and most completed their scheduled HRQOL assessments (1147 [95%] of 1208 at baseline; 988 [87%] of 1141 at 2 years). Cosmetic status was worse with tumour bed boost than with no boost across all timepoints (difference 0·10 [95% CI 0·05-0·15], global p=0·00014, Hochberg-adjusted p=0·0016); at the end of treatment, the estimated difference between tumour bed boost and no boost was 0·13 (95% CI 0·06-0·20; p=0·00021), persisting at 24 months (0·13 [0·06-0·20]; p=0·00021). Arm and shoulder function was also adversely affected by tumour bed boost across all timepoints (0·08 [95% CI 0·03-0·13], global p=0·0033, Hochberg adjusted p=0·045); the difference between tumour bed boost and no boost at the end of treatment was 0·08 (0·01 to 0·15, p=0·021), and did not persist at 24 months (0·04 [-0·03 to 0·11], p=0·29). None of the other six prespecified aspects of HRQOL differed significantly after adjustment for multiple testing. Conventional whole breast radiotherapy was associated with worse body image than hypofractionated whole breast radiotherapy at the end of treatment (difference -1·10 [95% CI -1·79 to -0·42], p=0·0016). No significant differences were reported in the other PROs between conventional whole breast radiotherapy compared with hypofractionated whole breast radiotherapy.

Interpretation: Tumour bed boost was associated with persistent adverse effects on cosmetic status and arm and shoulder functional status, which might inform shared decision making while local recurrence analysis is pending.

Funding: National Health and Medical Research Council, Susan G Komen for the Cure, Breast Cancer Now, OncoSuisse, Dutch Cancer Society.