The influence of systemic therapy on local recurrence following breast conservation There is clear evidence that chemotherapy alone does not reduce local recurrence risk after BCS (Fisher et al., 1989, 1995). The data on the inXuence of systemic therapy in addition to adjuvant RT on local recurrence is scarce. Retrospective series are confounded by competing risks, with those selected for systemic therapy more likely to develop distant disease before local recurrence and thus do not have any local recurrence recorded. In addition, there is little data on those patients with positive margins. The JCRT have found treatment with adjuvant chemotherapy to be associated with reduced risk of local failure following BCS/RT and RT (Rose et al., 1989). However this paper does not include data on the status of the surgical margin. Randomized data from NSABP B-13 indicates that in patients with clear margins adjuvant chemotherapy does appear to reduce risk of ipsilateral breast recurrence from 13.4% to 2.6% at 8 years (Fisher et al., 1996). It should be noted, however, that a tumour bed boost was not used in this trial and the total tumour bed dose was only 50 Gy. For patients with involved margins, a multivariate analysis of the Stanford series found the use of concurrent adjuvant chemotherapy to be independently associated with better local control in those with margins that were not_2 mm clear. In the group with margins positive, indeterminate or _2 mm clear local control was 98% compared with 86% for those who did not receive chemotherapy (Smitt et al., 1995). However, only 81 patients received chemotherapy and only 17 patients in the whole series of 289 women had more than focally positive margins. Other series including patients with involved margins have not found use of adjuvant chemotherapy to be associated with reduced local recurrence risk (Ryoo et al., 1989; Kurtz et al., 1990a; Anscher et al., 1993; Borger et al., 1994; Spivack et al., 1994; Touboul et al., 1999). In answer to our third question there is, therefore, insuYcient data available to conclude that adjuvant chemotherapy can normalize the high risk of local recurrence following BCS/RT and RT associated with a more than focally positive margin. Limited data suggests that use of adjuvant chemotherapy, at least if given either concurrently, following the radiotherapy or in a sandwich fashion, may further reduce risk of local recurrence in those with margins that are not more than focally positive (Rose et al., 1989; Smitt et al., 1995; Fisher et al., 1996).