Abstract GS03-02: Mammographic surveillance in early breast cancer patients aged 50 years or over: results of the Mammo-50 non-inferiority trial of annual versus less frequent mammography

Abstract

Introduction: Annual surveillance mammograms for an unspecified period, after treatment for early breast cancer, are widely practised in USA and Europe and represent a significant healthcare cost. Current UK guidelines recommend annual mammograms up to 5 years, then reverts to 3 year screening without specified risk stratification. Further evidence is needed to determine the optimum frequency and duration of mammographic surveillance. Methods: A multi-centre, randomised controlled, phase III trial of annual mammography versus 2-yearly for conservation surgery and 3-yearly mammograms for mastectomy patients up to 9 years. Women were eligible if aged 50 years or over at initial diagnosis of breast cancer (invasive or DCIS), and recurrence free 3 years post curative surgery. Primary outcome was breast cancer specific survival (BCSS). Secondary outcomes include recurrence free interval (RFI) and overall survival (OS). BCSS event was defined as deaths from breast cancer and RFI as any locoregional or distant invasive recurrence or new breast primary. 5000 women were needed to detect a 3% absolute non-inferiority (NI) margin for BCSS with 2.5% one-sided alpha and at least 85% power. Analyses were carried out on intention-to-treat basis. Results: 5235 women were randomised between April 2014 and September 2018. 4347 (83%) women were aged 55-75 years, 4203 (80%) had undergone conservation surgery, 4564 (87%) had invasive disease, 1162 (22%) had node positive disease, 4330 (83%) had ER positive tumours and 3812 (73%) were taking hormone therapy at the time of randomisation. Patient characteristics were balanced across arms. With a median of 5.4 years follow-up (interquartile range 4.6-5.9), 319 women have died; 104 of breast cancer (53 on annual arm; 51 on less frequent arm). BCSS at 5 years was 98.2% (95% CI 97.5-98.6%) on annual arm and 98.3% (95% CI 97.7-98.8%) on less frequent arm. Hazard ratio (HR) was 1.04 (95% CI 0.71 -1.54), demonstrating non-inferiority of less frequent mammograms at the 3% margin (NI p< 0.0001; critical value 2.71) and the 1% margin (NI p=0.02; critical value 1.56). 320 (6%) women had a new invasive breast cancer event (55 loco-regional recurrences, 85 new breast primaries, 139 distant recurrences and 41 with multiple invasive events); 164 on the annual arm compared to 156 on the less frequent arm. Five-year RFI was 94.2% (95% CI 93.2-95.1%) for the annual arm and 94.4% (95% CI 93.4-95.3%) for the less frequent arm; HR= 1.03; (95% CI 0.83-1.28) demonstrating non-inferiority at a 2% margin (NI p=0.006; critical value 1.36). OS at 5 years was 94.9% (95% CI 93.9-95.7%) on the annual arm and 94.3% (95% CI 93.3-95.2%) on the less frequent arm. Hazard ratio (HR) was 1.18 (95% CI 0.94 -1.47), demonstrating non-inferiority of less frequent mammograms at the 3% margin (NI p=0.003; critical value 1.61) and the 2.5% margin (NI p=0.02; critical value 1.51). A total of 14987 mammograms have been performed on the annual arm and 8047 on the less frequent arm. 1967 (75%) of 2618 women on the annual arm complied with their allocated schedule compared to 1775 (68%) of 2617 women on the less frequent arm. COVID-19 pandemic affected compliance; it is estimated that 345 (7%) women missed mammograms during the pandemic. A sensitivity analysis was performed on the 72% of women who fully complied with their scheduled mammograms as per protocol, and again NI was demonstrated for BCSS, RFI and OS. Conclusions: For patients aged 50 or older and 3 years post diagnosis, Mammo-50 demonstrated that less frequent mammograms were no worse than annual mammograms. These results provide evidence for less frequent mammographic surveillance for this patient population. Citation Format: Janet Dunn, Peter Donnelly, Nada Elbeltagi, Andrea Marshall, Alastair Thompson, Riccardo Audisio, Sarah Pinder, David Cameron, Amy Campbell, Sue Hartup, Lesley Turner, Annie Young, Helen Higgins, Eila Watson, Sophie Gasson, Peter Barrett-Lee, Claire Hulme, Bethany Shinkins, Peter Hall, Andy Evans. Mammographic surveillance in early breast cancer patients aged 50 years or over: results of the Mammo-50 non-inferiority trial of annual versus less frequent mammography [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr GS03-02.