RANDOMIZED TRIAL OF ONE VERSUS 2 ADJUVANT HYPERTHERMIA TREATMENTS PERWEEK IN PATIENTS WITH SUPERFICIAL TUMORS

One test for thermotolerance development in a clinical situation is to evaluate the effects of altering the hyperthermia fractionation inter val on tumour response to thermoradiotherapy. Between 1983 and 1990 44 evaluable advanced superficial tumours of miscellaneous origin in 41 patients were randomized to receive either once-weekly or twice-weekly external microwave hyperthermia treatments combined with radiation th erapy. The mean age of patients was 62 years, and 85 % had failed prev ious therapy. All lesions were less than 8 x 8 x 4 cm (L x W x D) and were heated by external 915 MHz microwaves. The mean radiation dose wa s 44 +/- 3 Gy (mean +/- SE) in the once-weekly group and 46 +/- 3 Gy i n the twice-weekly group (p = 0.64). The mean volume of the lesions he ated once weekly was 17 +/- 6 versus 23 +/- 5 cm3 for those heated twi ce weekly (p = 0.45). Hyperthermia was administered once weekly for 4. 6 +/- 0.2 sessions (range 3-7) or twice weekly for 8.1 +/- 0.3 session s (range 4-10). Thermometry was performed using 3.4 +/- 0.2 catheters and 5.1 +/- 0.6 thermal sensors per tumour in the once-weekly group, a nd 2.7 +/- 0.2 catheters and 5.8 +/- 0.3 thermal sensors per tumour in the twice-weekly group. Of the 44 evaluable randomized lesions a comp lete response (CR) at 2 months post-treatment was observed in 59% (13/ 22) heated once weekly and 55% (12/22) in those heated twice weekly. T he prognostic factors predictive of tumour complete response were foun d by logistic regression analysis to be radiation dose and tumour volu me, while the prognostic factors predictive of duration of response (C ox proportional hazards analysis) were median minimum tumour temperatu re (T(min)), minimum tumour temperature during the first heat treatmen t (T(min)) and tumour volume. The duration of local control in lesions with T(min) less-than-or-equal-to 39.5-degrees-C was 11.7 +/- 1.9 mon ths while for lesions with T(min) > 39.5-degrees-C it was 23.0 +/- 4.2 months (p = 0.01). The ED50 was calculated by logistic regression to be 40 Gy (95% CI = 22-54 Gy) for once- and twice-weekly heated lesions . There was not a significant difference in tumour response or duratio n of response between populations randomized to receive once- versus t wice-weekly hyperthermia treatments. There was also no difference in s kin reaction rates between once- and twice-weekly hyperthermia treatme nts, nor could a correlation be found between any thermal parameter an d skin reactions. It is concluded that similar complete response rates , similar duration of response and similar skin reaction rates can be obtained with once- or twice-weekly hyperthermia regimens.