Factors affecting acute skin toxicity in patients having breast irradiation after conservative surgery: a prospective study of treatment practice at the Royal Marsden Hospital.

The results are presented of a prospective study of acute skin toxicity in 197 patients with early stage breast cancer, who were treated by conservative surgery and postoperative radiotherapy. We have examined the factors determining the severity of the acute skin reaction with particular reference to the degree of dry or moist desquamation at the completion of treatment. One hundred and ten patients had treatment with radiotherapy alone. The remaining 87 received synchronous chemotherapy with breast irradiation, using either the 3M or the 2M regimen, consisting of mitoxantrone and methotrexate, with (3M) or without Mitomycin-C (2M). Patients were analysed according to both the severity and the site of the skin reaction, age, dose, dose variation across the central outline, treatment technique, beam energy, field separation and breast size. A univariate analysis of these results, which has been presented as an odds ratio of the likelihood of developing a moderate or severe reaction in comparison with those scored as mild, has shown that several factors are associated with an increase in the acute skin reaction. These include the use of the semi-supine technique (odds ratio (OR) = 7.3 (95% CI 3.7-14.6)), beam energy (60Co: 6-10 MV photons OR = 5.9 (95% CI 2.6-13.4)), field separation (> or = 20 cm: < 20 cm OR = 4.1 (95% CI 2.2-7.8)), dose variation across the central outline (> or = 10%: < 10% OR = 9.7 (95% CI 2.6-36.4)), inclusion of the axilla (OR = 4.6 (95% CI 2.4-8.9)), and bust size (bra cup size C and D: A and B OR = 4.6 (95% CI 2.7-11.9)). Using multivariate logistic regression, the technique of radiation delivery and bust size were shown to be independently significant variables affecting acute skin reaction. In view of the high correlation between variables (e.g. radiotherapy technique and beam energy) it is still not possible to specify with definite certainty which is the primary variable causing the skin reaction. However 20/57 (35%) of patients treated by the semisupine technique sustained a severe skin reaction, with > 10% dry or moist desquamation in the treatment field. This compares with only 6/140 (4%) patients treated by the supine method. A possible mechanism by which treatment using the semisupine technique may enhance acute toxicity is discussed. We conclude that there are both treatment and patient related factors that will increase the acute skin reaction after breast irradiation.