1. ¹Experimental Therapy (H6), The Netherlands Cancer Institute/Antoni van Leeuwenhoek Ziekenhuis, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands
2. ²Biometrics and Statistics, The Netherlands Cancer Institute/Antoni van Leeuwenhoek Ziekenhuis, Amsterdam, The Netherlands
3. ³Medical Oncology, The Netherlands Cancer Institute/Antoni van Leeuwenhoek Ziekenhuis, Amsterdam, The Netherlands

Correspondence: Dr FA Stewart, E-mail: f.stewart@nki.nl

Received 4 July 2002; Revised 20 September 2002; Accepted 1 October 2002.

Abstract

Clinical photodynamic therapy (PDT) schedules are based on the assumption that optimum drug–light intervals are times at which there is a maximum differential between photosensitiser retention in the tumour and surrounding normal tissue. However, vascular-mediated effects contribute to tumour destruction by PDT; therefore, plasma sensitiser levels and endothelial cell drug exposure could also be important determinants of PDT response. The purpose of this study was to investigate the influence of tumour, tissue and plasma concentrations of the photosensitiser Foscan^® (meta-tetrahydroxyphenylchlorin, mTHPC) on PDT response. Groups of BalbC nude mice, bearing human mesothelioma xenografts (H-MESO1) were injected (i.v.) with a single dose of ¹⁴C-labelled mTHPC, or with two doses, separated by 72 h. Drug levels in plasma, tumour and normal tissues were measured at 5 min to 120 h after drug administration. The PDT tumour and skin responses were evaluated by illuminating separate groups mice at intervals of 5 min to 120 h after injection of Foscan (nonlabelled). Drug levels in both tumour and skin increased during the first 24 h after a single injection, and remained almost constant for at least 120 h. The second injection produced a further, rapid increase in mTHPC levels in tumours and skin, with steady state being maintained from 20 min to 120 h. By contrast, PDT response of both tumours and skin were maximal for illumination at 1–3 h after drug, with very little response when illumination was given 48–120 h after drug. There was no significant correlation between tumour or skin drug level and PDT response. There was, however, a significant correlation between plasma drug levels and tumour or skin response, excluding an initial distribution time of 20 min. These studies demonstrate a pronounced disassociation between tumour drug levels and optimum drug–light intervals for PDT response with Foscan. We suggest that the PDT effect, in both tumours and normal tissues, is largely mediated via vascular damage and that the selectivity of PDT is not based on differential tumour drug uptake.