Comprehensive geriatric assessment and predicted 3-year survival in treatment planning for frail patients with early breast cancer

Background Endocrine therapy alone has been a popular treatment for oestrogen receptor-positive breast cancer in elderly patients, although it may be inadequate in those surviving more than 2-3 years. The aim of this study was to estimate 3-year survival in frail patients with early breast cancer, to inform treatment decisions. Methods A risk score was created to estimate 3-year survival in individual patients using data from patients who had Comprehensive Geriatric Assessment (CGA) in a specialist clinic before decisions about their breast cancer treatment were made. The data were analysed using logistic regression. Results Ninety-seven (29.6 per cent) of the 328 patients had died by 3 years. Four components of the assessment proved strongly associated with survival: Mini Mental State Examination, Barthel Index of Activities of Daily Living, instrumental Activities of Daily Living and American Association of Anesthesiologists fitness grade. The derived CGA risk score gave an adequate level of discrimination and calibration, with an area under the receiver operating characteristic (ROC) curve of 0.75 (95 per cent c.i. 0.67 to 0.82) (Hosmer-Lemeshow statistic X2 = 7.9, P = 0.448). Conclusion Detailed assessment can allow prediction of survival probability in frail elderly patients. Good scores indicate good survival prospects and a likely benefit from surgery; poor scores are associated with reduced survival, although with wide variation. CGA is recommended before making decisions on best treatment. © 2015 BJS Society Ltd Published by John Wiley & Sons Ltd.