DOES THE INTERVAL BETWEEN FRACTIONS MATTER IN THE RANGE OF 4-8 H IN RADIOTHERAPY - A STUDY OF ACUTE AND LATE HUMAN SKIN REACTIONS

Accelerated radiotherapy has the potential to increase local control o f rapidly growing tumours. To determine the necessary time interval fo r complete repair of sublethal damage in normal tissue in a clinical s ituation, we have compared the acute and late skin reactions with 8 an d 24 h between fractions, using the same dose per fraction and total d ose. Forty-nine breast cancer patients participated in this study, and received bilateral parasternal irradiation to 50 Gy with 2 Gy per fra ction as part of their adjuvant postoperative radiotherapy. The time i nterval between daily fractions was always 8 h on the left field and 2 4 h on the right, and the total treatment time was 2.5 and 5 weeks, re spectively. The acute endpoint was erythema, measured by reflectance s pectrophotometry and an acute reaction score for erythema and desquama tion. The late endpoint was telangiectasia, scored on an arbitrary sca le. The results have also been compared with those in a previously tre ated group of patients with 4 and 24 h between fractions. The degree o f acute reactions was decreased with an 8-h interval compared with 24 h between fractions with the peak acute score as endpoint; no differen ce was seen with the peak reflectance measurements. The maximal expres sion occurs similar to 1 week earlier with the accelerated schedule, p ossibly as a consequence of the reduction of the treatment time. The p attern of the acute reaction for 8 h between fractions is similar to t hat for 4 h. The late reactions were more pronounced with the 8-h inte rval, compared with 24 h, equivalent to a dose increase of 10% at this effect level. An interval of 4 h gave the same late reactions as 8 h. Therefore, the results indicate a very slow component of repair of su blethal damage for late-reacting tissues, and that an interval of >8 h is necessary for complete repair. The exact interval, in the range 4- 8 h, did not produce a resolvable difference in the severity of the la te responses.