Photochemistry and Photobiology, 1999, 70(4): 656-662
5-AminolevuIinic Acid Photosensitization of Dysplastic Barrett’s Esophagus: A Pharmacokinetic Study Roger Ackroyd‘, Nicola Brown’, David Vernon2, David Roberts2,Timothy Stephenson3,Stuart Marcus4, Christopher Stoddardl and Malcolm Reed*’ ’Department of Surgical and Anaesthetic Sciences, University of Sheffield, Sheffield, UK; *Centre for Photobiology and Photodynamic Therapy, University of Leeds, Leeds, UK; 3Departmentof Histopathology, Royal Hallamshire Hospital, Sheffield, UK and 4DUSA Pharmaceuticals, Inc., Valhalla, NY, USA Received 31 March 1999; accepted 6 July 1999
ABSTRACT Photodynamic therapy (PDT) using 5-aminolevulinicacid (ALA)-induced protoporphyrin M (PpM) may have a role in the treatment of dysplastic Barrett’s esophagus. Before ALA-induced PDT can be used clinically, optimum treatment parameters must be established. In this study of 35 patients, the issues of drug dosage, time interval between drug and light delivery and side effects of oral ALA administration are addressed. Spectrofluorometric analysis of tissue samples demonstrates that oral ALA administration induces porphyrin accumulation in esophageal tissues, with maximum levels at 4-6 h. Highperformance liquid chromatography confirms the identity of this porphyrin as PpM, and fluorescence microscopy analysis demonstrates that it preferentially accumulates in the esophageal mucosa, rather than in the underlying stroma. Side effects of ALA administration included malaise, headache, photosensitivity, alopecia, transient derangement of liver function, nausea and vomiting. Fewer side effects and less hepatic toxicity was seen with 30 mgkg than 50 mgkg ALA. In conclusion, oral ALA administration induces preferential PpIX accumulation in the esophageal mucosa, with peak PpM fluorescence noted at 4 h and minimal systemic toxicity at a dose of 30 mgkg.
INTRODUCTION Photodynamic therapy (PDT)? is a form of treatment, in which tissue damage is achieved by the action of light on a photosensitizing agent. It may be of use in the treatment of *To whom correspondence should be addressed at: Department of Surgical and Anaesthetic Sciences, University of Sheffield, Glossop Road, Sheffield S10 2JF, UK. Fax: +44-114-271-3791; e-mail:
[email protected] tAbbreviarions: ALA, 5-aminolevulinic acid; AST, aspartate transaminase; CCD, charge-coupled device: FBC, full blood count; yGT, gamma-glutamyl transferase; H&E, hematoxylin and eosin; HpD. hematoporphyrin derivative; LFT, liver function tests; PDT, photodynamic therapy: PpIX, protoporphyrin IX;U+E, urea and electrolytes. 0 1999 American Sociely for Photobiology 0031-8655/99
$5.00+0.00
dysplastic Barrett’s esophagus. Until recently the clinical use of PDT has been restricted by the side effect of long-lasting cutaneous photosensitization. However, recent studies describe the use of 5-aminolevulinic acid (ALA), a naturally occumng intermediate in the heme biosynthetic pathway, as a novel photosensitizer (1-5). 5-Aminolevulinic acid is a prodrug that has no photosensitizing properties but is metabolized through various intermediate porphyrins to protoporphyrin IX (PpIX), a fluorescent compound (6). It is the presence of PpIX in tissue that is exploited in ALA-induced PDT (2). 5-Aminolevulinic acid-induced PpIX has several advantages over previous photosensitizers, such as hematoporphyrin derivative (HpD). In particular, it is rapidly excreted, such that the resulting photosensitization lasts only 24 h (2,3,7,8). Also, PpIX accumulates preferentially in the gastrointestinal mucosa, rather than the underlying stroma (1,9), that may reduce damage to deeper layers, with consequent reduction in the risk of perforation or stenosis that have been seen with other photosensitizers (10,ll). There is evidence that PpIX may accumulate preferentially in tumor rather than normal tissue, which has implications for targeting treatment (1). Finally, ALA can be given orally, providing a convenient route of administration (3,7-9,12,13). However, the ideal drug dose and time interval to light delivery are unknown. Also, the safety of ALA is undocumented, with little data available on the systemic effects of oral ALA. The effect of a continuous enteral infusion of ALA has been studied in one human volunteer (1 2). with no effect on full blood count (FBC), urea and electrolytes (U+E) or liver function tests (LlT). In a study of four patients with oral cancer, given 30 m g k g ALA, the FBC and U+E remained unchanged. The only effects were a raised serum aspartate transaminase (AST) in three patients and bilirubin in two patients (8). All had returned to normal by 3 days. In a more recent study of 18 patients with gastrointestinal tumors given oral ALA, no effects were seen on FBC or U+E at 0, 24 or 48 h, but in two thirds of those given 60 m g k g ALA and one fifth of those given 30 m g k g ALA, a transient rise in serum AST was seen (3). In all cases, this had returned to normal by 72 h. A slight increase in serum
656
Photochemistry and Photobiology, 1999, 70(4) 657 bilirubin (three patients), gamma-glutamyl transferase (yGT) (two patients) and alkaline phosphatase (one patient) was observed, but none exceeded more than twice normal. No further increase was seen with repeated exposure to ALA. With regard to symptoms, a study of ALA ingestion in four human volunteers demonstrated cutaneous photosensitization lasting 24 h (7,14). More recent studies have shown variable side effects, including transient photosensitivity, nausea and vomiting or no symptoms (3,8,12). To date there have been no detailed reports of the effects of ALA administration in any larger series of patients. The aim of this study was to establish optimum treatment parameters for ALA-induced PDT in the treatment of dysplastic Barrett’s esophagus, in particular to: ( 1 ) identify side effects of oral ALA administration and determine the safe drug dose, (2) determine the accuracy of endoscopic diagnosis of Barrett’s esophagus, (3) assess whether PpIX accumulates in Barrett’s mucosa, (4) determine the time interval between ALA administration and peak PpIX levels, (5) confirm that PpIX is the predominant porphyrin following ALA administration and (6) identify the site of PpIX localization within the esophageal wall. A detailed pharmacokinetic study was performed to address these issues.
PATIENTS AND METHODS Thirty-five patients with low-grade dysplasia in Barrett’s esophagus were given either ALA (in orange juice) or placebo (orange juice alone), followed at 2 , 4 or 6 h by an endoscopy with multiple esophageal biopsies. Using sealed envelopes, patients were randomized with regard to drug dose and time interval to endoscopy into seven groups: 30 m g k g ALA with endoscopy at 2 h (n = 5). 4 h (n = 5 ) or 6 h (n = 5 ) . or 50 m g k g ALA with endoscopy at 2 h (n = 5). 4 h (n = 5) or 6 h (n = 5 ) , or placebo (endoscopy at 2 [n = I], 4 [n = 21 or 6 [n = 21 h). At the specified time, each patient underwent an endoscopy, and the extent of the Barrett’s epithelium was recorded. Twelve standard biopsies were taken, 8 from the Barrett’s mucosa and 4 from normal squamous epithelium. These were divided into four groups of three, each containing two samples of Barrett’s mucosa and one of normal squamous epithelium that were processed as follows: (1) Histological analysis: placed in formalin and sent unlabeled for analysis. The pathologist was asked to identify the epithelium and diagnosed Barrett’s only when specialized columnar epithelium (with intestinal metaplasia) was seen ( 1 5 ) ; (2) spectrofluorometric analysis: dried on filter paper, weighed, wrapped in foil, placed in cryovials and stored in liquid nitrogen for future analysis; (3) HPLC: dried on filter paper, weighed, wrapped in foil and stored in cryovials in liquid nitrogen for future analysis; and (4) fluorescence microscopy: dipped in freezing iso-pentane, wrapped in foil and stored in cryovials in liquid nitrogen for future analysis by fluorescence microscopy. Baseline venous blood samples were taken before ALA (or placebo) administration and analyzed for FBC, U + E and LIT. A further sample was centrifuged at 5000 rpm for 10 min and the plasma removed. This was then stored at -20°C for future HPLC analysis. A repeat venous blood sample was obtained at the time of endoscopy for similar analysis. Throughout the day patients remained in semidarkness, to reduce the risk of cutaneous photosensitization, and were allowed home after dark with instructions lo avoid bright light for 24 h. All were visited at home at 2 and 7 days, when further venous blood samples were obtained. Tissue porph-yrin analysis bv spectrojuorometry. The porphyrin content of the biopsies was measured by spectrofluorometry. following a physicochemical extraction procedure (16). Each biopsy was placed into a test tube (the two Barrett’s biopsies were added together for analysis), to which was added 4 mL of 5050 methanol/ 0.9 M perchloric acid. The tissue was homogenized for 5 min at
24000 rpm, and the resulting homogenate was centrifuged at 3000 rpm for 10 min. The supernatant was decanted off and the fluorescence measured at the appropriate wavelength for PpIX (excitation 406 nm; emission 604 nm). The PpIX concentration was then quantified against a standard fluorescence curve, from which the tissue PpIX concentration was calculated. The effect of protein quenching was eliminated by serial dilution of tissue extracts and repeat measurement of fluorescence in all cases, to ensure that the fluorescence decrease was proportional to the concentration decrease. HPLC analysis of plasma and esophageal tissue. Plasma samples and esophageal tissue from a selection of seven patients who received varying doses of ALA and one who received placebo were analyzed for porphyrins by HPLC. For plasma porphyrin extraction, 1 mL of solvent (ethyl acetate :glacial acetic acid 4: 1 vol/vol) was added to 200 p L of plasma. This was homogenized and centrifuged at 3000 rpm for 5 min. The lower phase and solid pellet were discarded. The upper organic phase was dried under a stream of nitrogen and dissolved in 25 pL methanol containing 10% 1 M hydrochloric acid. To effect tissue porphyrin extraction, 200 p L of 1.O M sodium hydroxide was added and samples were incubated overnight at 20°C. The extraction procedure was as for the plasma samples. The HPLC analysis was performed using a Waters Nova Pak C l 8 analytical column, with organic solvents (1) 10% acetonitrile/90% ammonium acetate (1 M, pH 5.16) and (2) 10% acetonitrile/90% methanol. Porphyrins were detected using a Waters 474 scanning fluorescence detector, with excitation and emission wavelengths of 400 nm and 620 nm, respectively, and were identified against the elution times of porphyrin standards. The area under each peak was calculated by computer integration, to allow quantitative analysis. Fluorescence microscop-y analysis of esophageal biopsies. Fluorescence images of tissue sections were obtained using a highly sensitive charge-coupled device (CCD) camera system (Astrocam 3200 LN/S, Astrocam, Cambridge, UK), linked to a computerized image processor (1,17). Frozen samples were cut into 8 pm sections, mounted on glass coverslips and allowed to dry at room temperature in the dark for 30 min. Sections were examined under phase-contrast illumination to locate structures of interest. To prevent PpIX photobleaching, the microscope illuminator was fitted with a 665 nm long-pass filter (Oriel, Leatherhead, UK), to remove the PpIX-activating wavelengths. Fluorescence microscopy was performed using an epifluorescence microscope (Leitz Wetzlar GmbH, Germany) fitted with the liquid nitrogen-cooled CCD camera, with illumination provided by a xenon arc source (18). producing 300 mW of light at 630 ? 15 nm via a 5 mm-diameter liquid light guide (Ultrafine 380 series). This was directed onto a filter block (Oriel XF46; excitation 632 ? 3 nm, emission 665-690 nm) in the epifluorescence head of the microscope. Images were taken using X I 0 and X40 oil-immersion fluorescence phase objectives, with exposure times of 30 and 20 s, respectively. Several fluorescence images were captured at each magnification, along with corresponding phase-contrast images. Image acquisition and processing was performed using Imager 2 software (Astrocam, Cambridge, UK). Following fluorescence microscopy, sections were fixed in formal-saline and stained with hematoxylin and eosin (H&E). Calibration images of a uniformly fluorescent glass plate were collected using both X 10 and X40 objectives. to verify the stability of the fluorescence imaging system and correct for nonuniformity in the illumination system. High-frequency noise was reduced using a single-pass median filter. Each image was corrected using a smoothed calibration image of the appropriate magnification and a shade-correction algorithm. The microscopic field corresponding to each fluorescence image was relocated on the appropriate H&E-stained section to confirm histology and facilitate identification of tissue structures. Fluorescence mcasurements in normal squamous and Barrett’s mucosa and associated stroma were made by multiple box superimposition to determine the relative fluorescence intensities in each. In an attempt to correct for tissue autofluorescence, the mean fluorescence in different components of control tissues (not treated with ALA) were subtracted from those of corresponding components in ALA-treated tissues (assuming that the tissue autofluorescence properties are unaffected by ALA). This simultaneously corrected for any fluorescence by the optical components of the imaging system.
658 Roger Ackroyd et a/. Separate quantitative analyses were performed on the epithelium
8
and underlying stroma of both Barrett’s and normal squamous epithelia from which fluorescence intensity ratios were calculated for (1) Barrett’s epithelium : Barrett’s stroma, (2) normal epithelium : normal stroma and (3) Barrett’s epithelium : normal epithelium, thus
.. .
allowing an assessment of the spatial localization of PpIX within esophageal tissue following ALA administration. Statistical analysis was by Mann-Whitney (I-test as appropriate (SPSS for Windows, version 6.0, SPSS Inc.).
.a
RESULTS 0
Systemic effects of oral ALA administration No significant changes were seen in any patient at any time in FBC,U+E, total protein, albumin or globulins. A clinically insignificant increase in serum bilirubin was seen at 48 h in five patients, but this returned to normal by 7 days. A minimal rise in serum alkaline phosphatase was seen in one third of patients at 48 h, including a two-fold increase in one patient given 50 m g k g ALA, but all had returned to normal by 1 week. The serum y-GT level was raised up to five-fold at 48 h in one third of patients. All those showing a marked increase had received 50 m g k g ALA, although overall this rise was not significant. All values had again returned to normal by 7 days. The most significant changes occurred in serum transaminase levels, with alanine transaminase and AST levels elevated in over half the patients at 48 h. However, these increases were dose dependent, with significantly higher levels in patients receiving 50 m g k g ALA than those given 30 m g k g ALA, and no effect was seen with placebo. No serious or lasting side effects were observed. Mild symptoms included malaise, headache, mild skin sensitivity, nausea and vomiting, predominantly in patients given 50 mg/ kg, but all had settled by 48 h. Another finding was esophageal discomfort, probably due to the endoscopic light source activating PpIX in the esophageal mucosa. The only unacceptable side effect was mild scalp alopecia in one patient given 50 m g k g ALA, although complete regrowth occurred by 1 month.
Histological analysis Histological examination of biopsies confirmed the presence of Barrett’s or normal squamous epithelium and correlated exactly with the endoscopic appearance in all cases, confirming our ability to differentiate normal squamous and Barrett’s epithelium at endoscopy. Thus we were satisfied that the biopsies taken for other analyses had been correctly identified as either normal squamous or Barrett’s epithelium.
Tissue porphyrin analysis by spectrofluorometry Fluorescence values for normal squamous epithelium were (mean 2 SEM) 0.3 2 0.1, 2.6 rt: 0.3, 3.2 2 0.4, 3.1 2 0.4, 2.2 2 0.2, 5.4 2 0.7 and 4.8 2 0.5 for placebo, 30 m g k g ALA (2, 4 and 6 h) and 50 mgkg (2, 4 and 6 h), respectively. The respective values for Barrett’s epithelium were 0.7 2 0.1, 2.5 ? 0.5, 3.2 2 0.4, 1.8 2 0.1, 2.3 2 0.1, 3.2 2 0.4 and 4.8 2 0.7. In all patients given ALA, the tissue PpIX levels were significantly greater than placebo (P < 0.001) and were dose-related with significantly greater levels in patients given 50 mgkg than 30 m g k g (P = 0.01 1) (Fig. 1a). The relationship between tissue PpIX concentration (ngl mg) and the time interval following ALA administration is
10
20
30
40
50
ALAdw(wWd b
0
2
4
6
TrmeaftaALAd ” . ‘bn(h0lW)
Figure 1. Tissue PpIX concentration (ng/mg)in esophageal biopsies versus (a) dosage of ALA (at 4 h time point only) and (b) time after ALA administration (for 30 mgkg dose only).
shown in Fig. lb. In patients given 30 mgkg, peak levels were found at 4 h for both normal squamous and Barrett’s mucosa, while in the higher dose group, maximum levels occurred at 4 h in normal squamous mucosa and 6 h in Barrett’s epithelium. Overall, no significant difference was observed between the PpIX levels in Barrett’s and normal squamous epithelium (P = 0.206). However, the PpIX levels were greater in normal squamous than Barrett’s epithelium at 6 h in the 30 mgl kg group (P = 0.009) and at 4 h in the 50 m g k g group ( P = 0.021).
HPLC analysis of plasma and esophageal tissue following ALA administration The HPLC analysis of plasma from the patient given placebo revealed small deflections corresponding to endogenous porphyrins, but the only measurable porphyrins were coproporphyrin I (23.9%) and PpIX (76.1%). Analysis of plasma from the seven patients given ALA produced the same deflections in all cases, but PpIX eluted at 53 min with a magnitude 100-fold greater than any background eluate fluorescence. Following ALA administration, uroporphyrin (I and III) and coproporphyrin I levels remained low ( blue > green > red > black.) The corresponding H&Estained sections are shown in c and d.
After 6 h, the proportion of PpIX was further increased to around 90% in all cases. The ALA dose had no effect on the proportions of porphyrins present. The HPLC analysis of esophageal tissue from the patient given placebo revealed one small deflection, corresponding to PpIX. In all patients given ALA, the same single peak was evident, but the magnitude of the deflection was again
Table 1. Quantitative fluorescence values obtained by fluorescence microscopic analysis of tissue samples taken at endoscopy following oral ALA administration* Randomization group
BE
30 m a g ; 2 h Mean 4500 SD 1500 30 mg/kg; 4 h Mean 5200 SD 1100 30 mg/kg; 6 h Mean 2000 SD 50 m a g ; 2 h Mean 4500 SD 1100 50 mg/kg; 4h Mean 6400 SD 4800 50 mg/kg; 6 h Mean 11 700 SD 1600
BS B E B S NE
NS NE/NS B E N E
700 500
6.4 5.6
5300 5000
1.1
0.9
2200 1100
2.7 1.2
8900 3900
2.3
0.7
400
4.8
9900
0.3
5000
1300 600
3.6 1.9
3400 1300 1100
4100 2500
2.1 1.6
6500 6300
4200 800
2.9 0.8
9700 2700 5200
3.1
1.3 0.7 1.4 1.7
1.3
1.1 0.5
*BE, Barrett’s epithelium; BS, Barrett’s stroma; NE,normal epithelium; NS, normal stroma; SD, standard deviation.
greatly increased, up to 200-fold. In all cases PpIX was the only porphyrin identified.
Fluorescence microscopy analysis of esophageal biopsies Samples from the five patients given placebo were used as controls to correct for autofluorescence. Of the remaining 30 patients, analysis could not be performed on all biopsies due to difficulties in obtaining thin sections from such tiny samples, especially those taken from the normal squamous epithelium that tended to be smaller and more superficial, with little underlying stroma. Overall, analysis was possible on 17 samples of Barrett’s epithelium, all of which contained intact underlying stromal tissue, and 12 samples of normal squamous epithelium, of which only 4 contained enough underlying stroma to permit analysis. However, analysis was performed on both Barrett’s and normal epithelia from at least one subject in each of the randomization groups. Examples of the fluorescence micrographs obtained together with the corresponding H&E-stained sections are shown in Fig. 2. The quantitative fluorescence values are given in Table 1. In all groups where several readings were obtained, wide variation in the degree of fluorescence was observed, producing a large standard deviation for each group. Statistical analysis of the results was difficult due to the small numbers in some groups, but the major findings were: (1) For both drug doses and at all times, fluorescence was greater in the Barrett’s epithelium than in the underlying stroma (P < 0.05). (2) For both doses and at all times, fluorescence was greater in normal epithelium than in the underlying stroma, but this difference was less apparent than in the Barrett’s tissue and did not reach statistical signifi-
660 Roger Ackroyd et a/.
cance. (3) There was no significant difference in the degree of fluorescence obtained in the Barrett’s epithelium between the two drug doses (30 vs 50 mgkg). (4) The degree of fluorescence observed in the Barrett’s stroma was greater in patients receiving 50 mgkg than 30 mgkg (P C 0.05), producing a significantly greater Barrett’s epithelium/Barrett’s stroma ratio in patients receiving 30 mgkg than 50 mgkg (P< 0.05). (5) With 30 mgkg ALA, more fluorescence was observed in normal squamous than Barrett’s mucosa (P < 0.05), this ratio increasing with time after ALA administration. (6) At 50 mgkg ALA, the reverse was seen, with greater fluorescence detected in Barrett’s than normal squamous mucosa, although this difference did not reach statistical significance. (7) With 30 mgkg ALA, maximum fluorescence levels were seen at 4 h in both Barrett’s epithelium and normal squamous mucosa, although these levels were not statistically significantly greater than those at either other time point. (8) After 50 mgkg, peak fluorescence was observed at 4 h in normal squamous mucosa and 6 h in Barrett’s epithelium, although again these levels were not statistically significantly greater than those at either other time point.
Photodynamic therapy for gastrointestinal tumors has been reported, primarily using HpD as the photosensitizer (1921). Although tumor necrosis has been observed, there have been occasional reports of treatments complicated by stenoses, perforations, fistulae and hemorrhage, as the depth of tissue necrosis produced may be difficult to control (19,20,22). However, following ALA administration, the preferential mucosal PpIX uptake potentially allows for the selective destruction of epithelial lesions, such as Barrett’s esophagus (1). Although a promising technique, previous reports are limited, and optimum treatment parameters are unknown. This paper aims to address this by means of a pharmacokinetic study. Our findings are similar to previous reports, with oral ALA having no effect on hemopoietic or renal function. Similarly, the total protein, albumin and globulin levels remained unchanged. However, significant dose-related effects were seen on other LFT, particularly the transaminases. This is not unexpected, as ALA is metabolized in the liver and thus induces upregulation of liver enzymes. However, the effects are reversible and transient, and although the biochemical effects of higher doses are unknown, it appears that the oral administration of 30 or 50 mgkg ALA is safe. Data on the side effects of ALA are limited. In one study, mild photosensitivity was seen in two patients after 2 4 4 8 h and mild nausea and vomiting in five patients within 12 h of oral ALA administration (3). In another, local discomfort was noted following ALA-induced PDT but only after light administration (8). In a third study, a human volunteer given high dose oral ALA displayed no side effects (12). Our findings are consistent with these studies, with many patients experiencing no side effects. Where side effects were seen, they were more common and severe in patients given 50 mgkg ALA. It may therefore be preferable to limit the ALA dose to 30 mgkg, although it is not known whether this is sufficient to induce a photodynamic effect.
The greater level of tissue PpIX in patients receiving ALA than those given placebo confirms that oral ALA results in tissue PpIX accumulation, while in the latter group the only PpIX is endogenous. The direct correlation between ALA dose and tissue PpIX concentration confirms that the tissue PpIX level is dose related. However, although no overall difference was seen between the PpIX levels in the two tissue types, the finding of increased PpIX content in squamous epithelium at 6 h in the 30 m g k g group and 4 h in the 50 mgkg group is difficult to explain. The effect of systemic ALA has been considered in several previous studies, mostly in laboratory animals (1,9,16,23,24)but also in human studies with a small number of patients. In a study of patients with colonic or oral cavity tumors, given 30-60 m g k g ALA orally (24), peak levels were seen at 6 and 5 h, respectively. In another study of patients with oral cavity tumors, peak fluorescence was found 4-6 h after oral ALA ingestion (8). These findings are comparable with ours, where maximum PpIX levels were seen at 4-6 h. The main difference was that in other studies there was selective uptake of ALA, with increased PpIX fluorescence in tumor, especially at higher doses (24). In our study, there was no difference in fluorescence between normal and Barrett’s, and although the latter is not tumor, all cases in this series had low-grade dysplasia. In our study, 50 m g k g ALA produced significantly greater PpIX fluorescence than 30 mgkg, but both doses produced greater fluorescence than placebo. The PpIX concentration required for PDT is unknown, as is whether 30 mg/ kg ALA is adequate, although previous work indicates that it may be sufficient, with superficial mucosal necrosis observed following light exposure in patients with esophageal and duodenal tumors (3). In our study, at this dose, peak tissue fluorescence was observed after 4 h, and at this time there was no significant difference between 30 and 50 mg/ kg. It was therefore concluded that the oral administration of 30 mgkg ALA with light exposure at 4 h may provide optimum treatment parameters for esophageal PDT. Although the endoscopic appearance of Barrett’s epithelium is distinctive (25), as confirmed by the complete correlation between the endoscopic and histological findings in our study, histological confirmation is still necessary (26). In particular, while it may be superfluous in differentiating columnar from squamous mucosa, it may assist in differentiating between true Barrett’s mucosa and columnar epithelium in a hiatal hernia, and it is imperative in the detection of dysplasia. The HPLC analysis of plasma revealed PpIX as the predominant porphyrin, although others were detected, the composition depending on the ALA dose and the time since administration. However, HPLC analysis of esophageal tissue revealed a different pattern, with PpIX the only porphyrin identified, although the amount was lower, reflecting the fact that PpIX is the major endogenous porphyrin. The reason for these differences is easily explained. Uro- and coproporphyrin and the corresponding porphyrinogens are rapidly excreted from cells before being metabolized to PpIX, while the lipophilic PpIX remains in the cell, particularly in the cell membranes. This also explains the observation that in plasma early after ALA administration, all the porphyrins
Photochemistry and Photobiology, 1999, 70(4) 661 are present (due to loss from cells), while later the predominant porphyrin is PpIX (due to slower loss from cells). There have been no previous reports of HPLC analysis of human plasma following ALA administration and very few studies of tissue analysis. In a study of porphyrin analysis of rat colonic and gastric tissues after intravenous ALA, PpIX was found to constitute at least 95% of the total porphyrin content, with small traces of coproporphyrin (
