Background: The management of subclinical disease is an indication for radiotherapy (RT) in many cancers. To determine if there is therapeutic gain from adding hyperthermia (HT) to RT in this setting, it is important to evaluate late normal tissue effects. This study represents an interim report from a prospective clinical trial comparing late effects in heated vs. unheated portions of the chest wall, in which each patient functions as her own control.
All patients were treated electively: none had visible disease at the initiation of thermoradiotherapy, since macroscopic disease was eliminated by pre-treatment chemotherapy and/or surgery.

Methods: Fifty five patients with high-risk breast carcinoma have been treated on study. Ultrasound HT was given simultaneously with four of the chest wall RT treatments for the first 46 cases. After sufficient follow up to establish late tolerance to four fractions of HT, the last eight cases received eight HT treatments.
The 12x12 cm ultrasound field covered the primary tumor site plus an elective region, which was randomized to lie either medial or lateral to the primary tumor site. The strip of chest wall on the opposite side of the primary tumor bed was the unheated control region and received the same dose of RT. RT to the chest wall and regional lymphatics was 50-60 Gy in 1.8-2 Gy fractions. Heated and control sectors received the same RT dose.

Results: With a mean follow up of 26 months, there have been 3 locoregional failures: one in the regional lymph nodes (radiated but not heated) and two diffusely in the chest wall (equal involvement of both heated and control sectors). A total of 29 patients had grade > 2 chest wall morbidity (usually asymptomatic pigment changes). In 6/23 cases one sector was noted to have morbidity while the other did not: in four cases the heated sector displayed grade 2 morbidity and in two the control sector displayed grade 2 morbidity.

Conclusions: The post treatment sequellae of 4-8 elective HT treatments appear to be clinically acceptable, with no significant morbidity difference between heated and control sectors. Locoregional control to date is promising.