Modulation of light delivery in photodynamic therapy of brain tumours.

This study was performed to determine whether modulation of light delivery could improve tumour kill in photodynamic therapy (PDT) of brain tumours, as optimal dosimetry has not been fully established. One hundred and sixty-five adult Wistar rats were treated, of which 70 had an implanted C6 cerebral glioma. Haematoporphyrin derivative (HpD) was injected at doses between 0 and 20 mg/kg, 24 h prior to irradiation with 630 nm laser light. The total energy dose was varied from 0 to 1200 J/cm(2), with fluence rates of 625, 3125 or 9375 mW/cm(2). In some studies, the light delivered at 3125 mW/cm(2) was divided into 10 fractions of approximately 13 s, with refractory intervals of 60 s. The most striking finding was that HpD was much more potent than previously reported. All doses greater than 1.0 mg/kg resulted in normal brain damage with light doses above 50 J/cm(2). However, at 1.0 mg/kg, significant normal injury was not apparent until 1200 J/cm(2). Failure of drug-light dose reciprocity indicated that photobleaching occurred, protecting normal tissue. Selective tumour kill was observed to 2.2. mm depth (SE +/- 0.44 mm). Using lower power or fractionated light did not improve tumour kill and normal tissue injury occured with fluence rates of 9375 mW/cm(2). In conclusion, the doses of HpD currently used in clinical brain tumour trials may be too high to achieve selective tumour kill. Higher light fluence rates allowed shorter intraoperative irradiation times with no loss of efficacy. Photodynamic therapy continues to demonstrate potential as an effective treatment for local control of cerebral lesions.