E-poster Presentation 2014 World Cancer Congress

Predictors of breast radiotherapy plan modifications: Quality assurance rounds in a large cancer centre (#920)

Timothy Lymberiou 1 , Susanne Galuszka 1 , Grace Lee 1 , Wei Xu 1 , Sue Su 1 , Anthony Fyles 1 , Tom Purdie 1 , Pam Catton 1 , Caroline Chung 1 , Rob Dinniwell 1 , Kathy Han 1 , Anne Koch 1 , Wilfred Levin 1 , Lee Manchul 1 , Joshua Sappiatzer 1 , Penny Mackenzie 1 , Fei-Fei Liu 1
  1. Princess Margaret Cancer Centre, Toronto, ON, Canada


Quality assurance (QA) in modern radiotherapy (RT) aims to detect inconsistencies that might adversely affect treatment outcome. Across communities, there are significant variations in the practice of QA reviews; here, we describe a single institutional experience on the breast QA process.


To report on the process and outcomes of our breast QA rounds, and identify factors that associate with plan modifications.


All curative breast RT plans were presented at weekly QA rounds prior to commencement of treatment. Comments regarding the plan were recorded in real-time, including type of modifications: no (A), minor (B), or major (C).


From January 1, 2010 to December 31, 2012, 2223 breast cases were reviewed; only 47 (2.1%) underwent a minor change (B); 52 (2.3%) required a major modification (C). The most common changes involved volume coverage, seroma contouring, addition of a boost, or use of bolus. On univariate analysis, plans using more than 2 fields (OR 2.57, p=0.0011), triple negative disease (OR 2.49, p=0.017), axillary node dissection (OR 1.76, p=0.045), and tumour size more than 2 cm (OR 2.01, p=0.025) were significantly associated with category C. After multivariate analysis, only the number of fields (OR 2.09, p=0.017), and triple negative disease (OR 2.34, p=0.027) remained significant. For cases who were node negative, margin negative, and only 2 fields with no boosts (n=561), modifications were required in only 0.89% of instances.


It is feasible to conduct weekly QA review for all radically treated breast cancer cases in a busy cancer centre. Techniques with more than 2 fields, and triple negative disease predicted for a higher likelihood of plan modifications. Conversely, less than 1% of node-negative cases with clear margins, treated with a 2-field technique and no boosts, required adjustment. These observations will contribute to streamlining future breast QA rounds.