Synthesis, photophysical properties, tumor uptake, and preliminary in vivo photosensitizing efficacy of a homologous series of 3-(1'-alkyloxy)ethyl-3-devinylpurpurin-18-N-alkylimides with variable lipophilicity.

Starting from methylpheophorbide-a, a homologous series of purpurinimides containing alkyl substituents at two different positions [as 3-(1(1)-O-alkyl) and 13(2)-N-alkyl] were synthesized. These compounds with variable lipophilicity (log P 5.32-16.44) exhibit long wavelength absorption near lambda(max)700 nm (epsilon: 45 000 in dichloromethane) with singlet oxygen ((1)O2) production in the range of 57-60%. The shifts in in vivo absorptions and tumor/skin uptake of these compounds were determined in C3H mice bearing RIF tumors by in vivo reflectance spectroscopy. The results obtained from a set of photosensitizers with similar lipophilicity (log P 10.68-10.88) indicate that besides the overall lipophilicity, the presence and position of the alkyl groups (O-alkyl vs N-alkyl) in a molecule play an important role in tumor uptake, tumor selectivity, and in vivo PDT efficacy. At present, all purpurinimide analogues are being evaluated at various doses, and experiments are underway to establish a quantitative structure-activity relationship on a limited set of compounds. The 1D and 2D NMR and mass spectrometry analyses confirmed the structures of the desired purpurinimides and the byproducts formed during various reaction conditions. The mechanisms of the formation of the unexpected 12-formyl- and 12-(hydroxymethyl)purpurinimides under certain reaction conditions are also discussed.

Photofrin photodynamic therapy can significantly deplete or preserve oxygenation in human basal cell carcinomas during treatment, depending on fluence rate.

At high fluence rates in animal models, photodynamic therapy (PDT) can photochemically deplete ambient tumor oxygen through the generation of singlet oxygen, causing acute hypoxia and limiting treatment effectiveness. We report that standard clinical treatment conditions (1 mg/kg Photofrin, light at 630 nm and 150 mW/cm2), which are highly effective for treating human basal cell carcinomas, significantly diminished tumor oxygen levels during initial light delivery in a majority of carcinomas. Oxygen depletion could be found during at least 40% of the total light dose, but tumors appeared well oxygenated toward the end of treatment. In contrast, initial light delivery at a lower fluence rate of 30 mW/cm2 increased tumor oxygenation in a majority of carcinomas. Laser treatment caused an intensity- and treatment time-dependent increase in tumor temperature. The data suggest that high fluence rate treatment, although effective, may be inefficient.

Parabolic quantitative structure-activity relationships and photodynamic therapy: application of a three-compartment model with clearance to the in vivo quantitative structure-activity relationships of a congeneric series of pyropheophorbide derivatives used as photosensitizers for photodynamic therapy.

An open three-compartment pharmacokinetic model was applied to the in vivo quantitative structure-activity relationship (QSAR) data of a homologous series of pyropheophorbide photosensitizers for photodynamic therapy (PDT). The physical model was a lipid compartment sandwiched between two identical aqueous compartments. The first compartment was assumed to clear irreversibly at a rate K0. The measured octanol-water partition coefficients, P(i) (where i is the number of carbons in the alkyl chain) and the clearance rate K0 determined the clearance kinetics of the drugs. Solving the coupled differential equations of the three-compartment model produced clearance kinetics for each of the sensitizers in each of the compartments. The third compartment was found to contain the target of PDT. This series of compounds is quite lipophilic. Therefore these drugs are found mainly in the second compartment. The drug level in the third compartment represents a small fraction of the tissue level and is thus not accessible to direct measurement by extraction. The second compartment of the model accurately predicted the clearance from the serum of mice of the hexyl ether of pyropheophorbide a, one member of this series of compounds. The diffusion and clearance rate constants were those found by fitting the pharmacokinetics of the third compartment to the QSAR data. This result validated the magnitude and mechanistic significance of the rate constants used to model the QSAR data. The PDT response to dose theory was applied to the kinetic behavior of the target compartment drug concentration. This produced a pharmacokinetic-based function connecting PDT response to dose as a function of time postinjection. This mechanistic dose-response function was fitted to published, single time point QSAR data for the pheophorbides. As a result, the PDT target threshold dose together with the predicted QSAR as a function of time postinjection was found.

Subcellular localization patterns and their relationship to photodynamic activity of pyropheophorbide-a derivatives.

To determine if subcellular localization is important to photodynamic therapy (PDT) efficacy, an in vitro fluorescence microscopy study was conducted with a congeneric series of pyropheophorbide-a derivatives in human pharyngeal squamous cell carcinoma (FaDu) cells and murine radiation-induced fibrosarcoma (RIF) mutant cells. In the FaDu cells the octyl, decyl and dodecyl ether derivatives localized to the lysosomes at extracellular concentrations less than needed to produce a 50% cell kill (LD50). At extracellular concentrations equal or greater than the LD50 the compounds localized mainly to mitochondria. The propyl, pentyl, hexyl and heptyl ether derivatives localized mainly to the mitochondria at all concentrations studied. This suggested that mitochondria are a sensitive PDT target for these derivatives. Similar experiments were performed with two Photofrin-PDT resistant RIF cell lines, one of which was found to be resistant to hexyl ether derivative (C6) mediated-PDT and the other sensitive to C6-PDT relative to the parent line. At extracellular concentrations of C6 below the LD50 of each cell line, the mutants exhibited lysosomal localization. At concentrations above these values the patterns shifted to a mainly mitochondrial pattern. In these cell lines mitochondrial localization also correlated with PDT sensitivity. Localization to mitochondria or lysosomes appeared to be affected by the aggregation state of the congeners, all of which are highly aggregated in aqueous medium. Monomers apparently were the active fraction of these compounds because equalizing the extracellular monomer concentrations produced equivalent intracellular concentrations, photoxicity and localization patterns. Compounds that were mainly aggregates localized to the lysosomes where they were rendered less active. Mitochondria appear to be a sensitive target for pyropheophorbide-a-mediated photodamage, and the degree of aggregation seems to be a determinant of the localization site.

pH-dependent chalcogenopyrylium dyes as potential sensitizers for photodynamic therapy: selective retention in tumors by exploiting pH differences between tumor and normal tissue.

Ideal photosensitizers have long-wavelength absorption and strong tumor selectivity with rapid clearance from normal tissues. The telluroselenopyrylium dye 1 that absorbs light at 795 nm (epsilon = 285,000 M-1 cm-1) has a novel property that enhances the tumor specificity and normal tissue clearance. After intralesional injection to both tumors and surrounding skin, it disappeared from the normal skin of BALB/c mice faster than it did from subcutaneously implanted Colon 26 tumors, which resulted in therapeutic selectivity. In vivo reflectance spectroscopy showed that the half-life in tumor was about 50 min while in skin it was around 12 min. This phenomenon appears to be related to the pH differences in normal skin versus tumor, because the rates of drug hydrolysis in solution were shown to be sensitive to changes in pH. Inhibition of tumor regrowth following intratumoral photosensitizer administration depended on both light dose and drug dose, as well as the time interval between dye injection and irradiation; selectivity depended on the time interval. Although treatment parameters were not optimized efficacy was superior to systemic Photofrin under our standard conditions. We discuss how new, more optimal, photosensitizers can be designed that use rates of hydrolysis to exploit the differences in pH between normal tissue and tumor.

Photofrin photodynamic therapy for treatment of AIDS-related cutaneous Kaposi's sarcoma.

OBJECTIVE: Kaposi's sarcoma, the most common malignancy in AIDS patients, often presents with painful cutaneous lesions that are difficult to treat effectively despite a wide variety of therapeutic approaches. We used photodynamic therapy in an attempt to provide effective palliative treatment for this disease. METHODS: Photodynamic therapy utilizes the activation by light of a photosensitizing drug that preferentially accumulates in tumor tissue such as Kaposi's sarcoma. We enrolled 25 patients who received 1.0 mg/kg of Photofrin 48 h before exposure to 100-400 J/cm2 of 630 nm light. RESULTS: Of the 348 lesions treated, 289 were evaluable: 32.5% had complete clinical response, 63.3% had partial clinical response and 4.2% were clinical failures. There was a strong correlation between response and light dose: 54% of lesions achieved a complete clinical response at optimum light dose (> 250 J/cm2). There was no correlation of response with CD4 cell count nor was there a change in CD4 cell count post-treatment. At 400 J/cm2 full field scabbing and necrosis occurred in 90% of the treated fields. Thus, the maximum tolerated dose was determined to be 300 J/cm2. At light doses of 250 J/cm2 and below the toxicities were limited to erythema and edema in the treatment field. Forty-three biopsies were taken 0.5 h to 4 months post-treatment. These showed little change in the B and T cell infiltrates identified. Kaposi's sarcoma cells disappeared post-treatment in certain lesions. CONCLUSION: Photofrin is effective palliative treatment for HIV-associated Kaposi's sarcoma.

Design and construction of a light-delivery system for photodynamic therapy.

We have developed a device to divide the output from a dye laser into as many as eight beams of equal power with negligible total power loss. In this system, 630-nm s-plane polarized laser light was split by a series of highly polarization-sensitive plate beamsplitters. Each of the beams was coupled to a 200, 400, or 600 microm diameter optical fiber. Brewster-window-type attenuators allowed the power of each beam to be individually set. It was possible to reconfigure the device to produce four, two, or one output(s). We discuss the design requirements of the beamsplitter device and describe its construction from mostly commercially available components. An apparatus for positioning and stabilizing each optical fiber relative to the skin surface of a patient is also described. The illumination from the fiberoptic supported by such an apparatus strikes a defined surface area and is independent of patient movement. Both the beamsplitter device and the optical fiber positioner are used routinely in photodynamic therapy (PDT) of malignant tumors in the clinic and in the laboratory.

An in vivo quantitative structure-activity relationship for a congeneric series of pyropheophorbide derivatives as photosensitizers for photodynamic therapy.

An in vivo quantitative structure-activity relationship (QSAR) study was carried out on a congeneric series of pyropheophorbide photosensitizers to identify structural features critical for their antitumor activity in photodynamic therapy (PDT). The structural elements evaluated in this study include the length and shape (alkyl, alkenyl, cyclic, and secondary analogs) of the ether side chain. C3H mice, harboring the radiation-induced fibrosarcoma tumor model, were used to study three biological response endpoints: tumor growth delay, tumor cell lethality, and vascular perfusion. All three endpoints revealed highly similar QSAR patterns that constituted a function of the alkyl ether chain length and drug lipophilicity, which is defined as the log of the octanol:water partition coefficient (log P). When the illumination of tumor, tumor cells, or cutaneous vasculature occurred 24 h after sensitizer administration, activities were minimal with analogs of log P < or = 5, increased dramatically between log P of 5-6, and peaked between log P of 5.6-6.6. Activities declined gradually with higher log P. The lack of activity of the least-lipophilic analogs was explained in large part by their poor biodistribution characteristics, which yielded negligible tumor and plasma drug levels at the time of treatment with light. The progressively lower potencies of the most lipophilic analogs cannot be explained through the overall tumor and plasma pharmacokinetics of photosensitizer because tumor and plasma concentrations progressively increased with lipophilicity. When compensated for differences in tumor photosensitizer concentration, the 1-hexyl derivative (optimal lipophilicity) was 5-fold more potent than the 1-dodecyl derivative (more lipophilic) and 3-fold more potent than the 1-pentyl analog (less lipophilic), indicating that, in addition to the overall tumor pharmacokinetics, pharmacodynamic factors may influence PDT activity. Drug lipophilicity was highly predictive for photodynamic activity. QSAR modeling revealed that direct antitumor effects and vascular PDT effects may be governed by common mechanisms, and that the mere association of high levels of photosensitizer in the tumor tissue is not sufficient for optimal PDT efficiency.

Correlation between site II-specific human serum albumin (HSA) binding affinity and murine in vivo photosensitizing efficacy of some Photofrin components.

Human serum albumin (HSA) is one of the key components in human blood that may influence drug distribution. As such, it is important to know the affinity of any drug for albumin. Previously, Photofrin, a mixture of monomeric, dimeric and oligomeric porphyrins, has been subjected to HSA binding studies. However, due to its complex nature, binding studies on Photofrin or other hematoporphyrin derivatives with HSA are inconclusive. In this report, the binding properties of some components (dimers and trimers) of Photofrin and the relationship between murine photosensitizing efficacy and those binding properties were investigated. The interaction of these porphyrins with HSA was investigated by direct ultrafiltration and fluorescent titration techniques with fluorescent probes such as dansyl-L-proline (DP), which is known to interact selectively with site II on HSA. Porphyrins also were tested for antitumor activity in a mouse model following intravenous administration and exposure to laser light. Together, the results suggest that the photosensitizers that were preferentially bound to site II of HSA were most effective at controlling murine tumor regrowth.

An assay for the quantitation of Photofrin in tissues and fluids.

A method for determining the concentration of Photofrin in tissues and biological fluids was developed. The procedure is based on the dissolution of biological material with Solvable, a commercially available tissue solubilizer, followed by porphyrin-specific fluorescence detection and measurement. It was found necessary to use a quadratic standard curve for the estimation of unknown Photofrin concentrations. While this method is limited to compounds that are stable in strong base, it has the advantages of being sensitive, rapid and low cost.

A preliminary pharmacokinetic study of intravenous Photofrin in patients.

Photofrin is a light-activated compound used for photodynamic therapy (PDT) of malignant tumors. Although PDT with this drug has been approved for clinical use in the United States, Canada, Japan, and the Netherlands there are few published reports on the biodistribution of Photofrin in humans. In this study we report measurable amounts of Photofrin in human serum up to approximately 1 year following injection of two different Photofrin doses. Concentration-time data were collected from 3, 12, 19, and 10 patients after 0.75, 0.875, 1, and 2 mg Photofrin/kg body weight. Patients who received 2 mg Photofrin/kg were scheduled to undergo intraoperative PDT for the treatment of mesothelioma or carcinoma of the lung. Patients receiving 0.75, 0.875, or 1 mg Photofrin/kg were treated for basal cell carcinoma; 1 mg Photofrin/kg is now a standard dose for PDT of cutaneous malignancies at this institute. For the 1 mg Photofrin/kg dose, a triexponential 3-compartment pharmacokinetic model was fitted to 30 data points pooled from the 19 patients, as if we had one "superpatient." The alpha, beta, and gamma halflives were approximately 16 h, 7.53 days, and 155.56 days, respectively. The mean (+/- SEM) serum concentrations 48 after injection (when most tumors are exposed to drug-activating light) of 0.875, 1, or 2 mg Photofrin/kg were 2.70 +/- 0.47, 4.00 +/- 0.66, and 3.47 +/- 0.97 micrograms Photofrin/ml, respectively. No porphyrin fluorescence could be detected in serum collected from patients 560 to 1335 days after Photofrin injection.

Alkyl ether analogs of chlorophyll-a derivatives: Part 1. Synthesis, photophysical properties and photodynamic efficacy.

The synthesis, preliminary in vivo biological activity, singlet oxygen and fluorescence yields of a series of alkyl ether derivatives of chlorophyll-alpha analogs are described. For short-chain carbon ethers (1-7 carbon units), it was observed that the biological activity increased by increasing the length of the carbon chain, being maximum in compounds with n-hexyl and n-heptyl chains. Related sensitizers prepared by reacting 2-(1-bromoethyl)-2-devinylpyropheophorbide-alpha with (sec)alcohols were found to be less effective. Under similar treatment conditions, photosensitizers containing cis- and trans- 3-hexenyl side chains were ineffective. Thus, both stereochemical and steric factors caused differences in sensitizing activity. In general, pyropheophorbide-alpha analogs were found to be more active than related chlorin e6 derivatives, in which the isocyclic ring (ring "E") was cleaved. Related photosensitizers in the 9-deoxy- series were found to be as effective as the corresponding pyropheophorbide-alpha analogs. The photosensitizers prepared from pyropheophorbide-alpha methyl ester and chlorin e6 trimethyl ester have long wavelength absorption at 660 nm (epsilon 45 000 to 50 000). Reduction of the carbonyl group in the pyropheophorbide-alpha to methylene (ring E) resulted in a blue shift to 648 nm (epsilon 38 000).

The validation of a new vascular damage assay for photodynamic therapy agents.

The therapeutic effect of photodynamic therapy (PDT: photodynamic sensitizer + light) is partly due to vascular damage. This report describes a new vascular photodamage assay for PDT agents and a validation of the assay. The method described here quantitates changes in tissue blood perfusion based on the relative amount of injected fluorescein dye in treated and untreated tissues. A specially designed fluorometer uses chopped monochromatic light from an argon laser as a source for exciting fluorescein fluorescence. The fluorescent light emitted from the tissue is collected by a six element fiberoptic array, filtered and delivered to a photodiode detector coupled to a phase-locked amplifier for conversion to a voltage signal for recording. This arrangement permits a rather simple, inexpensive construction and allows for the simultaneous use of the argon laser by other investigators. The routine assay for characterizing a specific photosensitizer at a standard dose consists of the sequential allocation of eight mice to a set of different light doses designed to span the dose-response range of fluorescein fluorescence exclusion (measured 8-10 min after fluorescein injection). The assay validation experiment used an anionic photosensitizer, 2-[1-hexyloxyethyl]-2-devinyl pyropheophorbide-a at a dose of 0.4 mumol/kg. The parameter estimates (n = 34 mice) from fitting the standard Hill dose-response model to the data were: median fluorescence exclusion light dose FE50 = 275 +/- 8.3 J/cm2 and Hill sigmoidicity parameter m = -3.66 +/- 0.28. Subsets of the full data set randomly selected to simulate a standard eight mice experiment yielded similar parameter estimates. The new assay provides reliable estimates of PDT vascular damage with a frugal sequential experimental design.

Photosensitization of murine tumor, vasculature and skin by 5-aminolevulinic acid-induced porphyrin.

The effects of topical and systemic administration of 5-aminolevulinic acid (ALA) were examined in several murine tumor systems with regard to porphyrin accumulation kinetics in tumor, skin and blood, vascular and tumor cell photosensitization and tumor response after light exposure. Marked, transient increases in porphyrin levels were observed in tumor and skin after systemic and topical ALA. Rapid, transient, dose-dependent porphyrin increases were also observed in blood; these were pronounced after systemic ALA injection and mild after topical application. They were highest within 1 h after ALA injection, thereafter declining rapidly. This matched the clearing kinetics of injected exogenous protoporphyrin IX (PpIX). Initially, vascular photosensitivity changed inversely to blood porphyrin levels, increasing gradually up to 5 h post-ALA, as porphyrin was clearing from the bloodstream. This pattern was again matched by injected, exogenous PpIX. After therapeutic tumor treatment vascular disruption of the tumor bed, while observed, was incomplete, especially at the tumor base. Minimal direct tumor cell kill was found at low photodynamic therapy (PDT) doses (250 mg/kg ALA, 135 J/cm2 light). Significant, but limited (< 1 log) direct photodynamic tumor cell kill was obtained when the PDT dose was raised to 500 mg/kg systemic ALA, followed 3 h later by 270 J/cm2, a dose that was however toxic to the animals. The further reduction of clonogenic tumor cells over 24 h following treatment was moderate and probably limited by the incomplete disruption of the vasculature. Tumor responses were highest when light treatment was carried out at the time of highest tumor porphyrin content rather than at the time of highest vascular photosensitivity. Tumor destruction did not reach the tumor base, regardless of treatment conditions.

Murine pharmacokinetics and antitumor efficacy of the photodynamic sensitizer 2-[1-hexyloxyethyl]-2-devinyl pyropheophorbide-a.

The combination of the new photodynamic sensitizer 2-[1-hexyloxyethyl]-2-devinyl pyropheophorbide-a (HPPH) and laser light of wavelength 665 nm showed antitumor activity against two s.c.-implanted murine tumors. HPPH also sensitized normal mouse foot tissue to light but photosensitivity decreased rapidly with time after HPPH administration. Mechanistic studies revealed that HPPH induced little direct tumor cell toxicity but was an effective mediator of vascular photodamage. Pharmacokinetic studies following intravenous injection of 1 mg [14C]HPPH per kilogram revealed a biexponential decay with time, with plasma alpha and beta half-lives of 0.69 and 21 h respectively. Fecal excretion was the primary route of elimination. The highest levels of [14C]HPPH were found in the liver, which also showed the greatest long-term retention. The sequence of decreasing uptake levels was the liver, adrenals, lung, spleen, kidney, urinary bladder, heart, eye, skin, pancreas, muscle, testes, fat and brain. This distribution correlated with the relative blood perfusion rates in the tissues.

Chlorin and porphyrin derivatives as potential photosensitizers in photodynamic therapy.

In order to find a photosensitizer with better optical properties and pharmacokinetics than Photofrin II, a series of new photosensitizers related to methyl pheophorbide-a and chlorin-e6 were synthesized. These compounds absorb at substantially longer wavelengths (lambda max 660 nm) than does Photofrin II (630 nm) and show promise for use in photodynamic therapy. Among the porphyrins, we observed that long carbon chain ether derivatives are better photosensitizers than their ester analogs. These sensitizers were tested for in vivo photosensitizing activity vis-a-vis Photofrin II, using the standard screening system of DBA/2 mice bearing transplanted SMT/F tumors. Most of these photosensitizers were found to have better tumoricidal photosensitizing activity than Photofrin II and demonstrated more rapid attenuation of normal tissue photosensitivity with time after administration vis-a-vis Photofrin II.

Potentiation of photodynamic therapy in mice with recombinant human tumor necrosis factor-alpha.

The tumoricidal response of subcutaneously growing SMT-F adenocarcinoma implanted into syngeneic DBA/2 mice to Photofrin II-sensitized photodynamic therapy (PDT) was statistically significantly enhanced by the addition of a single dose of intravenously administered recombinant human tumor necrosis factor-alpha (rHuTNF-alpha). The interaction appeared to be approximately additive, i.e. tumor response to PDT plus rHuTNF-alpha was about the same as that observed by doubling the PDT dose. Conversely, rHuTNF-alpha did not significantly potentiate the cutaneous phototoxicity in mouse feet due to PDT. These data suggest that combination therapy should be considered for improving tumor response while retaining treatment selectivity in human malignancies.

Protection of murine foot tissue and transplantable tumor against Photofrin-II-mediated photodynamic sensitization with WR-2721.

Four thiol-containing compounds, WR-2721, WR-149024, WR-168643 and WR-361, were compared as photoprotectors of murine feet. The protector doses were the maximal tolerated intraperitoneal doses, administered 24 h after injection of Photofrin II and 15 min before illumination with 630-nm laser light. While all four compounds were effective, only WR-2721 demonstrated a statistically significant attenuation of phototoxicity. WR-2721 was found to protect SMT-F tumors in the same mouse strain, using tumor growth delay and short-term control as endpoints. A comparison of the dose modification factors for foot and tumor responses indicated no therapeutic advantage in using WR-2721 during photodynamic treatment of these two tissues.

Distribution and elimination of Photofrin II in mice.

The distribution and elimination of [14C]PII, the radioisotopically-labeled equivalent of the mixture of porphyrins known as Photofrin II used in the photodynamic treatment of solid tumors, were determined in tumor-free and SMT-F tumor-bearing DBA/2 Ha-DD mice. Following i.p. injection, drug was absorbed from the peritoneum with a half-life of about 1 h; elimination from plasma was rapid, declining about 1.4 logs in concentration over 48 h following i.v. administration. However, some [14C]-activity was still detectable after 75 days. Normal tissues take up the drug within about 7.5 h after administration, with peak concentrations distributed as follows: liver, adrenal gland, urinary bladder greater than pancreas, kidney, spleen greater than stomach, bone, lung, heart greater than muscle much greater than brain. Only skeletal muscle, brain, and skin located contralaterally to subcutaneously implanted SMT-F tumors had peak [14C]-activities lower than tumor tissue; skin overlying SMT-F tumors showed concentrations not significantly different (P greater than 0.3) from tumor. After 75 days all tissues examined retained some fraction of [14C]-activity, ranging from 16% for kidney to 61% for spleen, of the initial peak tissue levels. The primary route of elimination of Photofrin II was through the bile-gut pathway, with greater than 59% of the administered [14C]-activity recovered in the feces, and only about 6% in the urine, over 192 h. HPLC analyses of fecal extracts showed that mostly monomeric and other low molecular weight porphyrin components of Photofrin II were eliminated. The higher molecular weight oligomeric fractions of Photofrin II were retained in liver and spleen up to 14 days after injection.

The time course of cutaneous porphyrin photosensitization in the murine ear.

This study was designed to investigate the time course of acute cutaneous photosensitivity following administration of Photofrin II using the murine ear swelling response (ESR) as an in vivo end-point. Ros:(ICR) mice were injected with 5 mg/kg Photofrin II and illuminated 7.5 h to 31 days later with 630-nm laser light; ESR was measured 24 h after illumination. There was a direct correlation between ESR and the concentration of [14C]Photofrin II in blood, while no relationship between ESR and the level of [14C]Photofrin II in the ear tissue of exsanguinated mice was evident. Photosensitivity in the mouse foot can be suppressed by preexposure to low doses of light via a photochemical destruction of tissue-bound sensitizer (Boyle and Potter, 1987, Photochem. Photobiol. 46, 997-1001). However, mouse ears pretreated with 84 J/cm2 of 630-nm light (28 J/cm2/day, given 2, 4 and 6 d after injection), a dose sufficient to reduce porphyrin fluorescence in ear tissue by about 75%, prior to the usual light dose (88.6 J/cm2, 630 nm, day 9 after injection) showed a mean ESR not significantly different (P less than 0.5) from that for ears which received only a single dose of 88.6 J/cm2 on day 9. It is concluded, for this animal model, that circulating porphyrin is the source of photoinduced ear-tissue edema and that photobleaching of tissue-bound sensitizer does not attenuate ear-tissue photosensitivity.