B27
Pragmatism in breast cancer clinical trial design encouraged wider participation of centres yet yielded a homogenous patient profile
Jayant Vaidya1, on bahalf of Targit Trialists's group2
1University College London, UK, 2TARGIT, London, UK
In 1999, we designed a randomised controlled trial to test whether TARGeted Intraoperative radioTherapy (TARGIT) was equivalent to postoperative external beam radiotherapy (EBRT).
To cater for a wide level of equipoise, our design was pragmatic with minimal restrictions for age (>=45) and tumour size (preferably smaller than 3.5cm), and no restrictions for grade and nodal status. At the outset, each centre specified these options in a treatment policy document.
This analysis of the treatment policies includes 1674 patients from 24 centres ransomised until April 2009. The minimum age at entry was specified to be 40, 45/48, and 50 by 2, 8 and 14 centres that randomised 243(14.5%), 514(30%) and 917(55%) patients. However, 1566 (93.5%) of patients randomised were >=50; 45-49y = 83 (5%) and 40-44y = 21 (1.25%). 10 (525 patients) centres did not restrict tumour size while 8 (800 patients), 1(187 patients), 5 (152 patients) centres restricted the size to <=2cm, <=2.5cm, <=3cm. However, 84% patients had tumour size <=2cm and <4% were >3.5cm. Grade 3 was excluded by only 4 centres (278 (16.6%) patients), but only 13% of all randomised patients had grade 3 cancers.
TARGIT could also be delivered either as a first or second procedure, and 37% more patients were randomised because of this. Furthermore, if patients randomised and given intraoperative radioatherapy were found to be high risk of elsewhere- recurrence (e.g. lobular cancers or EIC or other prespecified features) EBRT could be added within the protocol which essentially tested the two strategies and not techniques. Only 10% patients randomised to IORT received additional EBRT.
Allowing clinicians to be liberal in their intended inclusion criteria increased appeal and encouraged wider participation, yet led to a relatively homogeneous patient sample, demonstrating an unexpected conservatism in this pragmatic trial.
