Photodynamic therapy (PDT) for cancer patients has developed into an important new clinical treatment modality in the past 25 years. PDT involves administration of a tumor-localizing photosensitizer or photosensitizer prodrug (5-aminolevulinic acid [ALA], a precursor in the heme biosynthetic pathway) and the subsequent activation of the photosensitizer by light. Although several photosensitizers other than ALA-derived protoporphyrin IX (PpIX) have been used in clinical PDT, ALA-based PDT has been the most active area of clinical PDT research during the past 5 years. Studies have shown that a higher accumulation of ALA-derived PpIX in rapidly proliferating cells may provide a biologic rationale for clinical use of ALA-based PDT and diagnosis. However, no review updating the clinical data has appeared so far.
A review of recently published data on clinical ALA-based PDT and diagnosis was conducted.
Several individual studies in which patients with primary nonmelanoma cutaneous tumors received topical ALA-based PDT have reported promising results, including outstanding cosmetic results. However, the modality with present protocols does not, in general, appear to be superior to conventional therapies with respect to initial complete response rates and long term recurrence rates, particularly in the treatment of nodular skin tumors. Topical ALA-PDT does have the following advantages over conventional treatments: it is noninvasive; it produces excellent cosmetic results; it is well tolerated by patients; it can be used to treat multiple superficial lesions in short treatment sessions; it can be applied to patients who refuse surgery or have pacemakers and bleeding tendency; it can be used to treat lesions in specific locations, such as the oral mucosa or the genital area; it can be used as a palliative treatment; and it can be applied repeatedly without cumulative toxicity. Topical ALA-PDT also has potential as a treatment for nonneoplastic skin diseases. Systemic administration of ALA does not seem to be severely toxic, but the advantage of using this approach for PDT of superficial lesions of internal hollow organs is still uncertain. The ALA-derived porphyrin fluorescence technique would be useful in the diagnosis of superficial lesions of internal hollow organs.
In the first step of the heme biosynthetic pathway, 5-aminolevulinic acid (ALA) is formed from glycine and succinyl coenzyme A (CoA). The last step is the incorporation of iron into protoporphyrin IX (PpIX), which takes place in the mitochondria under the action of the enzyme ferrochelatase. With the addition of exogenous ALA, PpIX may accumulate because of the limited capacity of ferrochelatase. Porphobilinogen deaminase is another enzyme of the heme synthesis pathway (catalyzing the formation of uroporphyrinogen from porphobilinogen). Its activity is higher in some tumors,1-3 whereas that of ferrochelatase is lower,2-8 so that PpIX accumulates with some degree of selectivity in such tumors. Because PpIX is an efficient photosensitizer, ALA has been introduced as a drug for clinical photodynamic therapy (PDT) of cancer.9, 10 PDT involves, in general, systemic administration of a tumor-localizing photosensitizer or photosensitizer prodrug and the subsequent activation of the photosensitizer by light. In 1990, Kennedy et al.9 first applied topically ALA-based PDT in a clinical setting. Today ALA-PDT is successfully used for the treatment of a variety of neoplastic and nonneoplastic diseases.
ALA can be applied topically or systemically for PDT of skin and other tumors (such as skin basal cell carcinoma and gastrointestinal adenocarcinoma).9, 11-17 It can also be used for diagnostic evaluations of tumors of the skin, bladder, gastrointestinal tract, and lung.16-19 ALA is hydrophilic and does not easily penetrate through intact skin,9, 20, 21 or through cell membranes.22 When ALA is applied topically to cutaneous tumors, the tumor selectivity is also caused by an increased permeability of the skin tumor. Nodular skin tumors with a relatively intact keratinized surface layer are refractive to topical ALA-PDT because ALA does not penetrate to their base. We have therefore made progress in the field of ALA-based PDT by studying a number of lipophilic ALA ester derivatives. These are more lipophilic and may penetrate more easily through the keratinized layer and deeper into tumors than ALA itself.23, 24 The esterase activity in cells and tissues leads to cleavage of ALA from the ALA ester derivatives.
ALA-based PDT has been the most active area of PDT research during the past 5 years.25-37 The number of published articles reporting clinical research on ALA-PDT has increased exponentially since the year 1992. As no comprehensive review of clinical ALA-PDT has appeared since 1992, we now review the recent data on clinical ALA-based PDT and diagnosis and discuss the future challenges of this promising treatment modality. Brief sections on regulation of heme synthesis and on light dosimetry for ALA-PDT are also included.
REGULATION OF HEME SYNTHESIS
The initial step in the heme synthesis pathway is the formation of ALA. In mammals and photosynthetic bacteria, ALA is formed from glycine and succinyl-CoA by the enzyme ALA synthase (ALAS). In vertebrates, there are two ALAS isoenzymes, a housekeeping ALAS and an erythroid specific isoenzyme. The enzyme is located on the matrix side of the inner mitochondrial membrane,38 loosely associated with the membrane.39 The enzyme has the main regulatory function of the pathway.
The next enzyme in the pathway, ALA dehydratase, is located in the cytosol and induces the condensation of two molecules of ALA to yield porphobilinogen (PBG) with the elimination of two water molecules. The combined action of PBG deaminase (PBGD) and uroporphyrinogen III (co)syntase40 condenses in a head-to-tail manner four molecules of PBG and cyclizes the tetrapyrrole chain to form uroporphyrinogen III. Both enzymes are located in the cytosol, and the action of PBGD is the rate-limiting step. A series of decarboxylations and oxidations have to take place before iron can be inserted into the tetrapyrrole ring. The first part of this process is performed in the cytosol by uroporphyrinogen decarboxylase. This enzyme removes four acetic acid carboxyl groups from uroporphyrinogen to form the tetracarboxylic coproporphyrinogen. Coproporphyrinogen III, to be used for heme synthesis, is now exposed to coproporphyrinogen oxidase, which is situated in the intermembrane space of the mitochondria.41, 42 The enzyme decarboxylates and oxidizes the propionic side chains in ring A and B to vinyl groups, and protoporphyrinogen IX is formed. The final step in the synthesis of PpIX is the oxidation of the tetrapyrrole ring by removal of six hydrogens from protoporphyrinogen IX, catalyzed by protoporphyrinogen oxidase. The enzyme is embedded in the inner mitochondrial membrane with its active site on the matrix side of the membrane.43 It is an oxygen-dependent enzyme with high substrate specificity.44 The tetrapyrrole structure is now ready for the incorporation of iron, which is catalyzed by ferrochelatase (EC 188.8.131.52). Ferrochelatase is located in the inner mitochondrial membrane.
Regulation of the Heme Synthesis Pathway
All the enzymes in the heme pathway act irreversibly. The pathway is partly regulated by substrate availability and feedback inhibition of ALAS. The concentrations of substrates and intermediates are usually far below the Michaelis constants of all the enzymes involved.45 Of all the enzymes in the pathway, ALAS has the lowest activity, followed by PBGD, whereas the other enzymes have much higher activities. In human erythroid cells, ferrochelatase activity is also low, being only about three-fold higher than that of ALAS.45
A main regulatory step in the heme pathway is linked to ALAS activity (Fig. 1 (12K)). Heme can inhibit the enzyme directly46 as well as the transcription, translation, and transport of the protein into mitochondria. The direct inhibition of the enzyme may, however, be of minor importance, because the inhibition occurs only at around 10-5 M, whereas the formation of ALAS is controlled at 10-7 M. It has been suggested that a free heme pool at a concentration of about 10-7 M is involved in this regulation.47 The housekeeping ALAS, expressed in all tissues, and the erythroid specific isoenzyme are regulated somewhat differently.48-59 More details of the regulation of heme synthesis and degradation have recently been reviewed by us.37
DISTRIBUTION AND TOXICITY OF PPIX INDUCED BY TOPICAL APPLICATION OF ALA
Although little information exists about the tissue distribution of ALA after topical application, the fluorescence of ALA-derived PpIX in normal and diseased human skin has been found to increase with time after topical ALA application, with a plateau of approximately 4-14 hours, depending on ALA concentrations (2-40%) in formulas, amounts of preparations (30-50 mg/cm2), and application time (up to 24 hours) used.60-62 In general, topical application of ALA alone for less than 4 hours produces PpIX only at the site of ALA application, whereas the administration for a longer time (up to 14 hours) or combined with skin penetration enhancers (such as dimethylsulfoxide [DMSO]) leads to a generalized photosensitization of the skin (Peng et al., unpublished data). Six hours after topical ALA application (5-40%), a minor increase of porphyrins in erythrocytes and plasma of patients was observed; normal levels returned before 24 hours had passed. Blood count, transport proteins, and enzymes were not significantly influenced.63 Generally, ALA-derived PpIX fluorescence can not be detected in the skin 24 hours after completion of topical ALA application. ALA itself does not seem to be toxic to tissues when concentrations <50% in water/oil emulsion by weight are topically applied for at least 48 hours. Moreover, no evidence shows toxicity of ALA-derived PpIX on tissues before light exposure. However, during and a few hours after light irradiation, most patients experience a pruritus, prickling or burning sensation in light-irradiated areas, a sensation similar to that observed in porphyria patients shortly after sun exposure. Some patients cannot even tolerate this pain. Such irritant reaction can significantly be reduced by use of 2% lignocaine gel9 or "Emla" cream (containing 2.5% lignocaine and 2.5% prilocaine),12 by local intracutaneous anesthesia of 1% lignocaine62 or 2% mepivacaine,64 or by spray of a preparation containing 10% lignocaine (Warloe et al., unpublished data). Thus, anesthetic drugs should routinely be included in the cream preparations. Occasionally, some treated lesions can develop bacterial superinfection.
PHARMACOKINETICS AND TOXICITY OF ALA AND ALA-DERIVED PPIX AFTER SYSTEMIC ALA ADMINISTRATION
Several hospitals have started to use systemic administration of ALA for fluorescence diagnosis and PDT of skin, gastrointestinal, and lung cancers.16, 17, 19, 65, 66 However, it is still not clear whether ALA itself is toxic after systemic administration. This issue has been debated for a long time. Although some patients suffer from mild, transient nausea or/and transient abnormalities of liver function, it appears that systemic administration of exogenous ALA at a dose lower than 60 mg/kg (oral) or 30 mg/kg (intravenous) does not result in any neurotoxic symptoms. Moreover, several earlier studies have shown no porphyric symptoms in cancer patients or healthy volunteers with transient or sustained high plasma ALA levels after single or repeated systemic administration of exogenous ALA.67-71 The plasma concentration of ALA (26.8 μmol/L) peaked at 60 minutes after a single oral administration of 3.3 mg/kg ALA in a normal human subject, with a half-life (T1/2) of 50 minutes.71 Regula et al.16 measured the plasma ALA concentrations during fractionated oral administration (at hourly intervals) of 30 or 60 mg/kg ALA in 13 patients with gastrointestinal tumors. They found that the mean plasma ALA concentration in 11 patients 6 hours after a fractionated dose of 30 mg/kg was 63 μmol/L (standard deviation, 33 μmol/L). The ALA levels in 2 other patients 6 hours after 60 mg/kg also as a fractionated dose were 116 and 205 μmol/L, respectively. In contrast, Gorchein and Webber72 found that the maximum plasma ALA levels in 2 patients with acute intermittent porphyria were only 9 and 12 μmol/L, but with severe neurologic deficit, including respiratory paralysis, quadriplegia, and extensive autonomic abnormalities. Obviously, administration of ALA to cancer patients for PDT treatment has led to much higher plasma ALA levels than those in porphyric patients. Why was ALA-PDT treatment associated only with mild nausea and occasional vomiting, without any forms of neurovisceral symptoms often seen in porphyric patients? The reasons for this are not known. Nevertheless, it has been reported that exogenous ALA may penetrate across the blood-brain barrier and the central nervous system itself may synthesize porphyrins from exogenous ALA.73-75 Therefore, much care should be taken in clinical trials of systemic ALA administration, particularly for the patients with porphyria or severe diseases of the liver and kidneys, because acute attacks of hepatic porphyrias with neurovisceral symptoms are always associated with high urinary excretion of ALA,76 and in this case ALA is generally considered to be the most likely neurotoxic compound.77, 78 Unfortunately, only a marginal amount of knowledge is now available concerning the pharmacokinetics and toxicity of exogenously administered ALA and the relationship between the pharmacokinetics of ALA and that of ALA-derived PpIX in plasma and tissues.
The findings that the plasma of porphyric patients contains certain ALA levels might suggest that the rates of ALA excretion from different types of cells prior to the formation of ALA-derived PpIX might be one of the reasons for the variability of PpIX production in various types of cells and tissues in vivo.
ALA-Derived PpIX in Blood
Little information is available about the pharmacokinetics of ALA-derived PpIX in humans. Lofgren et al.79 reported that the highest levels of ALA-derived PpIX in the plasma of rabbits occurred 1 hour after an intravenous (i.v.) dose of 50 mg/kg or 100 mg/kg ALA, and 2 hours after a dose of 200 mg/kg. The PpIX concentrations declined to the control level by 24 hours, with a half-life of approximately 60 minutes. A similar pharmacokinetic pattern was also observed by Henderson et al.80 in the serum of mice receiving intraperitoneal (i.p.) injection of 250-1000 mg/kg ALA. The value of serum PpIX over a 5-hour period after an i.p. dose of 1000 mg/kg ALA was similar to that just after an i.v. dose of 7 mg/kg exogenous PpIX.80 Recently, Webber et al.81 reported a pharmacokinetic study of ALA-derived PpIX in 4 cancer patients after oral administration of 60 mg/kg ALA. They found that the half-life of exogenous ALA-derived PpIX was approximately 8 hours after a brief distribution phase. Similar results were also obtained by Egger et al.82 in dogs receiving i.v. injection of 100 mg/kg ALA. Clearly, more work is needed concerning the pharmacokinetics of ALA-derived PpIX.
ALA-Derived PpIX in the Skin
Numerous pharmacokinetic studies of ALA-derived PpIX in the skin of various species have been performed. These investigations may reflect the accumulation of circulating PpIX. In most studies, the techniques used were based on a noninvasive spectrophotofluorometric method to measure in vivo PpIX fluorescence in skin surface in situ after systemic (i.p./i.v./oral) administration of various doses of ALA.83 In the skin of mice, dogs and humans, ALA-derived PpIX peaks at approximately 3-8 hours and is almost completely eliminated within 24 hours after systemic ALA administration.60, 83 Similar results have also been obtained by fluorescence microscopy84 and chemical extraction techniques.80, 84, 85 In fact, such a phenomenon has been noted for a long time.67, 86-88
An important issue is whether ALA-derived PpIX present in the skin originates from bone marrow and liver via the blood circulation, or is locally synthesized in the skin itself, or both. A considerable amount of evidence has shown that the synthesis of ALA-derived PpIX can take place in situ in the skin. For example, local (topical, intradermal, and intracutaneous) application of exogenous ALA to normal and diseased skin of various species led to a porphyrin fluorescence and subsequently light-induced photosensitization localized only to the site of previous ALA application.10
Considering that Photofrin (Quadra Logic Technologies, Vancouver, Canada) contains approximately 5-10% PpIX and Photofrin and hematoporphyrin (a more polar dye) are well known to be retained in the skin for several weeks,89, 90 why can ALA-derived PpIX fluorescence not be detected 24 hours after systemic administration of exogenous ALA? The most likely explanation is that a single high dose of exogenous ALA leads to a temporary production of a high amount of PpIX in the skin. After that, most ALA-derived PpIX is quickly either metabolized into nonfluorescing heme/bilirubin in the skin or released from the skin and transported to the liver via the blood. The PpIX and heme/bilirubin in the liver are further metabolized in the intestines and excreted into the feces.37 Such a explanation needs to be experimentally confirmed. Another possibility is that the persisting fluorescence in the skin of patients given Photofrin is due to other porphyrins than PpIX.
PDT OF HUMAN PRIMARY NONMELANOMA SKIN TUMORS AFTER TOPICAL ALA APPLICATION
Nonmelanoma skin cancer is the most common form of cancer in fair-skinned populations. The majority of nonmelanoma skin cancer is basal cell carcinoma (BCC) and squamous cell carcinoma (SCC).91, 92 In the United States alone, more than 500,000 BCCs and 100,000 SCCs are diagnosed annually. In Australia, the annual incidence of treated nonmelanoma skin cancer is estimated to be 823 in 100,000, and the rates for BCC and SCC are estimated to be 657 and 166 in 100,000, respectively.93 In the United Kingdom, approximately 190,000 new skin tumors are diagnosed every year.94 The mortality from nonmelanoma skin cancers is low compared with that from other malignancies, but both mortality and incidence are rising and affecting younger people.
BCC and SCC arise from the epidermis or its appendages. About 45-60% of BCCs are noduloulcerative and 15-35% are superficial.95, 96 Currently, both surgical and nonsurgical treatments are used for nonmelanoma skin cancer, including excisional surgery, Mohs' surgery, cryosurgery, electrodesiccation and curettage, topical chemotherapy, and radiotherapy.96-98 Systemically administered HpD/Photofrin has also been tried in the PDT treatment of nonmelanoma skin cancer.99-108 Moreover, topically TPPS-based PDT has shown promising results in the treatment of BCCs.109, 110 In 1990 Kennedy et al.9 reported the first treatment of 80 BCCs using topical ALA-PDT with success; this was followed by Wolf and Kerl's report in 1991 that xerodermal pigmentosum was removed with topical ALA-PDT.111 ALA-PDT is now widely applied in the treatment of cutaneous tumors, although these clinical trials are still at Phases I-II.
Standard Procedures for Topical ALA-PDT
So far no proprietary agent has been marketed for topical PDT of human primary nonmelanoma skin tumors, although a number of different formulations have been used in various trials.9, 11-15, 109, 110, 112 In the standard procedure, an oil-in-water emulsion of ALA is applied to a skin lesion. This emulsion layer is covered by a semipermeable dressing, and the skin lesion is exposed to light, causing a singlet oxygen-induced photodamage to the lesion. The concentration of ALA in the emulsion is usually 20%, but it can be varied from 2% to 40%, depending on the application time. For example, a cream containing 2-5% ALA applied for a time longer than 8-12 hours produces an amount of PpIX similar to that produced by a cream with 20% ALA applied for 3 hours (Peng et al., unpublished data). The optimal dose is still not known, and the concentration of 20% ALA is likely to be an overdose in some clinical treatments. All of the oil/water emulsions (cream, lotion, and ointment) applied so far are commercially available (Glaxal [Roberts Pharm. Corp., Ontario, Canada],9 Unguentum [Merck, Germany],12, 14, 62 Doritin [Chemofux, Vienna, Austria],11, 64 Essex [Schering Corp., Kenilworth, NJ],13 and Decoderm [Merck]61). It seems that all the emulsions work well, and none appears to be superior. The time for the topical ALA application is usually 3-8 hours, to allow penetration of ALA into the lesion and synthesis of PpIX. The light source used in most cases is a laser with a wavelength of approximately 630 nm, but incoherent light sources such as tungsten lamps, xenon lamps, and halogen lamps with suitable red filters are also often used. In general, the total light dose is 60-250 J/cm2 with an intensity of 50-150 mW/cm2 when a laser is used, whereas the dose is 30-540 J/cm2 with dose rates ranging from 50 to 300 mW/cm2 when a lamp is used. The temperature of the skin lesions can rise to 39.5-42.5°C during topical ALA-PDT when an intensity of 100 mW/cm2 is used.113 The response to the treatment is usually evaluated clinically within 1-2 months after treatment. The criteria of therapeutic effectiveness adopted for most clinical studies are as follows: tumor complete response (CR) is defined as the absence of clinically evidence of tumor at the site of treatment; partial response (PR) is defined as a reduction of 50% or more in tumor size; no response (NR) is defined as a reduction of less than 50% in tumor size.
Topically ALA-Based PDT for the Treatment of Human Skin BCC and SCC
During the past 5 years, more than 10 articles have reported the use of topically ALA-based PDT for the treatment of BCCs. Table 1 summarizes the clinical results. In a total of 826 superficial BCC lesions treated, the weighted average rates of CR, PR, and NR were 87%, 5%, and 8%, respectively, whereas among 208 nodular BCC lesions, the corresponding rates were 53%, 35%, and 12%, respectively. For superficial BCCs, most trials obtained good results, with CR rates ranging from 79% to 100%. The two exceptions were trials by Cairnduff et al.12 and Lui et al.,116 both of which reported a CR rate of only 50%. The reasons for these exceptions are not known. For nodular lesions, the majority of reports demonstrated a CR rate lower than 50% after a single treatment, but higher ALA concentrations and longer application times tended to increase ALA-derived PpIX in the lesions65 and, consequently, improve the outcome of the treatment.117
Table 1. Summary of Published Clinical Studies Using Topical ALA-PDT in BCC
Total no. of lesions (sBCC and *nBCC)
ALA concentration in oil/water emulsion (w/w) and time applied (hrs)
Table 2 shows the results of topical ALA-PDT for the treatment of SCCs. The weighted average rates of CR, PR, and NR were 81%, 14%, and 5%, respectively, for a total of 41 superficial SCC lesions treated; these rates were similar to those for superficial BCCs. However, nodular SCCs did not respond well to topical ALA-PDT with the current protocol, although only few nodular SCC lesions have been treated so far.
Table 2. Summary of Published Clinical Studies Using Topical ALA-PDT in SCC
Total no. of lesions (sSCC and *nSCC)
ALA concentration in oil/water emulsion (w/w) and time applied (hrs)
It should be pointed out that the superficial lesions of BCC and SCC, when evaluated clinically, are often found to be deeply penetrating lesions examined by histopathology.116 Because there is no clear line of demarcation between a "thin" and a "thick" BCC/SCC lesion, errors resulting from clinical evaluation can strongly affect the results of ALA-PDT.
The current protocols of topical ALA-PDT are far from ideal for the treatment of nodular BCCs and SCCs. They have gained low CR rates and high recurrence rates (Tables 1 and 2), although several efforts have focused on a prolonged application of ALA, adding some other useful chemical additives in cream base and repeated PDT procedure (see below) to improve the therapeutic effectiveness. The reasons why the success was only partial are not fully known. Limited ALA penetration into deep layers of the nodular lesion is at least one of the causes. The capacity of ALA-derived PpIX production in various histopathologic types of the tumors may also have been related. Therefore, analysis of histologic localization of ALA-derived PpIX is useful for optimization of topical ALA-PDT. Selective localization of ALA-derived PpIX fluorescence has been shown in the superficial BCC lesions rather than in the adjacent normal epidermis after topical application of ALA for 3 hours,21 but the deep layers of nodular BCCs demonstrated little fluorescence.60, 65, 118 The penetration of ALA into the deep BCC lesions could be increased by prolonging the time of topical application of ALA to 12-48 hours.65, 119 Moreover, both the penetration of ALA and production of ALA-derived PpIX could be enhanced by using topical ALA plus DMSO,65 a skin penetration enhancer,120 and desferrioxamine (DFO), an inducer of porphyrin synthesis.64 However, significant variability and heterogeneity of the ALA-derived PpIX fluorescence have been observed between and within the BCC lesions,65, 118 probably due to a short duration of ALA application, varying ALA penetrating abilities, varying ALA/PpIX excretion rates in different lesions, or variations in the histopathologic type of BCC.119, 121 For example, the morphea type of BCC has little or only spotty inhomogenous PpIX fluorescence.119 It is noteworthy that oral122 or intravenous65 administration of ALA allows PpIX production throughout the superficial,65, 122 nodular,65, 122 and even morphea types122 of BCC, thereby providing a significant advantage over topical ALA application.
Topically ALA-Based PDT for the Treatment of Bowen's Disease and Actinic Keratosis
Nearly all reports (Table 3) demonstrate that topical ALA-PDT for the treatment of Bowen's disease (intraepidermal SCC) has obtained promising CR rates, ranging from 89% to 100%. An exception is the study of Fijan et al.,64 which demonstrated a CR rate of only 50%. Furthermore, actinic keratosis may be most sensitive to the treatment modality, with a 92% weighted average CR rate of 116 lesions treated (Table 4).
Table 3. Summary of Published Clinical Studies Using Topical ALA-PDT in Bowen's Disease
Total no. of lesions
ALA concentration in oil/water emulsion (w/w) and time applied (hrs)
Improvement of the Therapeutic Effectiveness of Topical ALA-PDT by Repeated Treatments
Topical ALA-PDT can be repeated for the lesions that fail to respond well to previous treatment(s). Repeated treatments were generally much more effective than a single treatment, particularly for the nodular BCC lesions (Table 5). For example, Svanberg et al.13 found that only 16 of 25 nodular BCCs (64%) had a CR after a single treatment, whereas 100% CR was achieved with one additional treatment. Studies of Warloe et al.14 and Fijan et al.64 also showed that repeated treatments increased CR rates of nodular BCC from 34% to 68% and from 32% to 59%, respectively.
Table 5. Comparison of CR Rates of Primary Nonmelanoma Skin Tumors after Single and Repeated Topical ALA-PDT
Improvement of the Therapeutic Effectiveness of Topical ALA-PDT by the Use of DMSO/EDTA/DFO or Curettage
The relatively poor results of topical ALA-PDT in the treatment of nodular BCCs and SCCs may be due to a limited tissue penetration of ALA and an inadequate production of ALA-derived PpIX. Warloe et al.14 have treated a large number of BCC lesions with ALA cream containing DMSO and ethylenediamine tetraacetic acid (EDTA). Although the CR rate was not improved in the case of superficial BCCs, it was significantly increased in the nodular lesions, especially in the lesions less than 2 mm thick (Table 6). Good results were also obtained by Orenstein et al.61 in the treatment of nodular BCCs with DMSO/EDTA. Thus, the therapeutic effectiveness of topical ALA-PDT for nodular lesions may be improved by using skin penetration enhancers in combination with porphyrin production inducers. However, the actual role EDTA plays in the clinical treatment is still not clear. In addition, DFO enhanced the fluorescence intensity of PpIX in the skin lesions after topical application of ALA for 20 hours,64 and a better therapeutic effect would be expected in such cases. Recently, Warloe et al. have tried a curettage procedure to reduce tumor volume and remove the surface structure of 152 nodular tumors before ALA-PDT. Such a procedure achieved 85% CR with a follow-up of 3-6 months (Warloe et al., unpublished data).
Table 6. Comparison of Topical ALA-PDT of BCCs with or without DMSO/EDTAa
Comparison of CR Rates between Initial Clinical Evaluation and "Long Term" Follow-Up or Histopathologic Evaluation after Topical ALA-PDT
Although topical ALA-PDT has achieved promising results in the treatment of superficial skin lesions, the clinical response rates have usually been evaluated within 1-2 months after treatment. In most studies the follow-up is too short to draw any sensible conclusions. Table 7 provides the information available so far in the literature as to the difference in CR rates between initial clinical evaluation and "long term" follow-up or histopathologic evaluation after treatment of various cutaneous diseases. All initial clinical CR rates decreased after "long term" follow-up or histopathologic evaluation except in the study of Calzavara-Pinton, which still demonstrated 100% CR rates of Bowen's disease and keratoacanthoma after a long term follow-up of 24-36 months.15 In most studies the initial CR rates did not significantly decrease after long term follow-up, but in four trials of BCC (two superficial and two nodular),12, 15, 116 the CR rates were remarkably decreased from initial 67-88% to 33-50% after a median follow-up of 17-36 months or 3 months of histopathologic evaluation. Thus, a full picture of the therapeutic effectiveness of topical ALA-PDT for cutaneous lesions requires data on a long term follow-up or histopathologic confirmation.
Table 7. Summary of Short Term versus "Long Term" CR Rates in Topical ALA-PDT of Primary Nonmelanoma Skin Tumors
b Used to treat a total of 393 sBCC lesions, 141 receiving 20% ALA alone, 125 receiving ALA plus 2-20% DMSO/2-4% EDTA, and 127 receiving 50-90% DMSO applied 15 min prior to application of ALA alone or ALA plus DMSO/EDTA. Used to treat a total of 326 nBCC lesions, 80 receiving ALA alone, 110 receiving ALA/DMSO/EDTA, and 136 receiving DMSO as pretreatment.
Effect of Light Dose on the Results of Topical ALA-PDT
Little information is available regarding the effect of light dose on the response rates of skin lesions to topical ALA-PDT. So far the light dose applied is within a wide range of 60-250 J/cm2 for laser sources and 30-540 J/cm2 for nonlaser sources. In many clinical trials the light exposure has been overdosed. Is the CR rate of treatment proportional to the light dose applied? Warloe et al.14 failed to find a clear proportional relation between the CR rate and light dose used. However, it appears that doses ranging from 50 to 90 J/cm2 at an intensity of 150 mW/cm2 are required to achieve good results of the treatment in both superficial and nodular BCCs.14 For the very superficial lesions, such as Bowen's disease, even lower doses may be used. For nodular tumors, interstitial insertion of fiberoptic cylinders into deep lesions may be useful, but this still remains to be determined. In addition, fractionated irradiation could result in a faster regression of the lesions, but the effects of split light dose and light intensity on the CR rate need to be studied.
TOPICALLY ALA-BASED PDT OF OTHER TUMORS
Warloe et al.14 treated patients with nevoid basal cell carcinoma syndrome (Gorlin's syndrome) who had a total of 11 superficial and 26 nodular BCC lesions, and the CR rates of the superficial and nodular lesions were only 61% and 12%, respectively. Karrer et al.123 found good results in a patient with Gorlin's syndrome who had multiple BCCs and failed to respond to conventional methods including surgical excision, cryotherapy, and ionizing radiotherapy.
Eighteen patients with vulval or vaginal carcinomas in situ were treated with topical ALA-PDT at the Norwegian Radium Hospital (Kristensen et al., unpublished data). All the tumors showed a strong fluorescence after topical ALA application for 4 hours, but only approximately 50% of the treated lesions had CR. The reason for this is not understood, but we found that the ALA-derived porphyrin fluorescence in the treated lesion biopsies was completely photobleached by the light exposure (100-150 J/cm2) (Peng et al., unpublished data).
Recently, CR of cutaneous T-cell lymphoma has been reported after topical ALA-PDT.117, 124 Wolf et al.125 emphasized that repeated topical ALA-PDT is important in treating the cutaneous T-cell lymphoma, because the results obtained by both Ammann and Hunziker126 and Svanberg et al.13 were disappointing after a single topical ALA-PDT. We observed a strong fluorescence of ALA-derived porphyrins in the tumor cells of a patient with cutaneous T-cell lymphoma (Peng et al., unpublished data). Similarly, by means of laser-induced fluorescence measurements, Svanberg et al.13 found the ratio of fluorescence intensity between T-cell lymphomas and surrounding normal tissue to be 5:1. This may have been due to a lack of ferrochelatase in the mitochondria of the aberrant T-lymphocytes that led to an accumulation of endogenous porphyrins.127
So far, all three reported clinical trials of topical ALA-PDT for metastatic nodular breast carcinoma have achieved poor results,9, 12, 115 probably due to the fact that the periphery of the metastatic tumors lay beneath the normal skin, where it is difficult for ALA to penetrate. Wolf et al.11 have also reported ALA-PDT to be a therapeutic failure in the treatment of metastases from malignant melanoma.
TOPICALLY ALA-BASED PDT OF HUMAN NONNEOPLASTIC SKIN DISEASES
Although topical ALA-PDT has most often been employed to date in the treatment of skin tumors, its potential use is far beyond dermatologic oncology. Boehncke et al.128 treated 3 patients with chronic plaque-stage psoriasis every other day with PDT, using a topical application of 10% ALA for 5 hours before light exposure at a dose of 25 J/cm2 (70 mW/cm2), and achieved promising results. Nelson et al.129 treated 14 patients with psoriasis with 10-20% ALA and UVA light exposure weekly for a total of 4 times. About half of the treated lesions improved by more than 50% after 4 weekly treatments. We studied 20 psoriatic biopsies taken from 6 patients after topical 20% ALA application and found that psoriatic lesions can produce a strong but unevenly distributed PpIX fluorescence (Peng et al., unpublished data). Similar results were obtained by others.130
Both Kennedy et al.9 and Ammann et al.131 observed a poor response of refractory verrucae vulgaris to topical ALA-PDT with application of a 20% ALA cream for 3-6 hours followed by light exposure from a slide projector.
Frank et al.132 treated 7 genital condyloma acuminatum (CA) lesions, applying 20% ALA topically for 14 hours before light exposure of a argon dye laser with a dose of 100 J/cm2 at an intensity of 75 or 150 mW/cm2. They obtained CR in 4 of 7 lesions after 3 months. Fehr et al.133 studied in detail the distribution of ALA-derived PpIX in the vulvar CA lesions of 24 patients at various time intervals after the topical application of 2.5% or 20% ALA. They found that both in vivo fluorescence imaging in situ and fluorescence microscopy of biopsies showed selective fluorescence of ALA-derived PpIX in the labia minora and vestibule of condylomas within short time intervals, particularly in the lesions located in the areas of non-hair-bearing skin, indicating that ALA-PDT could have a potential for ablation of genital CA. Furthermore, the ratio of epithelial condyloma fluorescence to adjacent skin 1.5 hours after ALA application was higher with 2.5% ALA than with 20% ALA.133 Similarly, we studied 3 cases of vulval CA and found that all the lesions demonstrated a strong ALA-derived PpIX fluorescence after topical 20% ALA application for 4 hours (Peng et al., unpublished data).
It is noteworthy that topical ALA application to skin induces an accumulation of PpIX not only in the epidermis but also in its adnexa (including hair follicles and sebaceous glands) in mice,84, 134 dogs, and humans.60 Consequently, topical ALA-PDT could provide potential uses for treatment of disorders originating from the skin appendages. A preliminary study has shown that topical ALA-PDT could be useful in treating hirsutism by permanently damaging hair follicles.135 Grossman et al.135 reported that 3 months after PDT with topical 20% ALA and 200 J/cm2, only 50% of the treated sites had hair regrowth, and the adjacent dermis was not damaged. Furthermore, acne, a disorder of sebaceous glands, could be another potential indication for this modality. We have also observed some fluorescence of ALA-derived PpIX in eczematous lesions (Peng et al., unpublished data).
SYSTEMICALLY ADMINISTERED HEMATOPORPHYRIN/HEMATOPORPHYRIN DERIVATIVES-BASED PDT FOR HUMAN PRIMARY NONMELANOMA SKIN TUMORS
In 1978 Dougherty et al.136, 137 reported a pioneering clinical study in which systemically administered hematoporphyrin derivative (HpD)-PDT was used to treat 5 BCC lesions with a 100% CR rate at a follow-up of 12 months. Since then, a number of similar clinical trials have been performed in the treatment of primary nonmelanoma skin tumors.99-106 Table 8 presents a summary of the majority of published data. In addition, the reports of McCaughan,153 Bandieramonte et al.,154 Gregory and Goldman,112 Waldow et al.,143 and Petrelli et al.155 have shown promising results for BCCs, SCCs, and Bowen's disease, although they are not included in Table 8. As can be seen from Table 8, the majority of the studies employed HpD/Photofrin and laser systems with a time interval of 24-120 hours between systemic drug administration and light exposure. Some used also hematoporphyrin (Hp) or Photosan-3 (Seelab, Wesselburenerkoog, Germany), a similar agent to HpD. In 15 trials of PDT, involving a total of 553 BCC lesions, the average CR, PR, and NR rates were 86%, 13%, and 5%, respectively. However, there was wide variation among CR rates in the different studies. For example, 7 trials in which a total of 120 lesions were treated achieved a 100% initial CR rate,115, 136, 138, 143, 146, 148, 150 whereas 3 studies involving 31 lesions only obtained a CR rate of approximately 50%.140, 142, 147 Moreover, recurrence rates after treatment varied from 0%146 to 100%,142 with most follow-ups longer than 10 months. Obtaining results similar to those with ALA-PDT, Wilson et al.149 found that a second treatment of BCC for PR and recurrent lesions from the first PDT increased the CR rate from 88% to 97% among 151 treated lesions.
Table 8. Summary of Published Data on Hp/HpD/Photofrin/Photosan-3-based PDT for the Treatment of Primary Nonmelanoma Skin Tumors
In 5 trials involving a total of 60 SCCs, the averages of CR, PR, and NR were 72%, 20%, and 8%, respectively. Two of the 5 studies had a 100% initial CR rate and no recurrences at a follow-up of 14-48 months,146, 148 whereas the other 3 trials achieved only 44-81% initial CR rates with 40-50% recurrence rates at a follow-up of 6-12 months.141, 142, 147 In contrast, all 4 studies of PDT, involving a total of 560 lesions of Bowen's disease, had a consistent 100% CR rate with recurrence of only 1 lesion at a follow-up of 6-24 months.145, 147, 151, 152
COMPARISON OF ALA-PDT WITH HPD/PHOTOFRIN-PDT AND WITH CONVENTIONAL TREATMENT MODALITIES
Different studies have shown a wide variation in the responses of nonmelanoma primary skin tumors to ALA-PDT and HpD/Photofrin-PDT. This could be due to a lack of controlled clinical PDT trials (including treatment protocols and patient selection criteria). In general, PDT outcome depends on the type and amount of sensitizing agent absorbed by the tumor, light wavelength, depth of light penetration into the tumor, and light energy delivered. In most studies, the light source was a laser emitting at approximately 630-635 nm, but some investigations were performed with other light sources, making direct comparison difficult. Moreover, the treated tumors ranged in size from a few mm to more than 20 cm and had pigmentation of varying degrees.
Advantages and disadvantages of topical ALA-PDT and systemically administered HpD/Photofrin-based PDT of primary nonmelanoma skin tumors are summarized in Table 9. Although both modalities are suitable for treatment of superficial cutaneous tumors, Photofrin is the most widely used photosensitizer in clinical PDT trials and is the only agent that has been approved for several clinical indications in Japan, Canada, the Netherlands, the United States, and France. Moreover, with the current protocol, Photofrin-PDT appeared more efficient than topical ALA-PDT in destroying cutaneous lesions. The main disadvantage of using Photofrin-PDT is the risk of prolonged skin photosensitivity.
Table 9. Comparison of Topical ALA-PDT and Systemic HpD/Photofrin-PDT for the Treatment of Skin Cancer
1) Convenient; available on an outpatient basis
2) Low-cost (ALA is cheaper than Photofrin, and ordinary lamps with suitable filters can be used)
3) No toxicity or interaction with other medications
4) High selectivity leaving the surrounding normal skin intact and functional
5) Several separate lesions can be treated simultaneously
6) The same lesion(s) can be repeatedly treated
7) Cosmetic results are superior to conventional modalities
8) No risk of skin photosensitivity after 24 hrs
9) Local anesthesia is often required during light exposure
10) Efficient for superficial lesions
1) Relatively inconvenient; patients often stay in the hospital for a few days
2) Expensive (laser is used in most cases)
3) No systemic toxicity or interaction with other medications
4) Selectivity leaving the surrounding normal skin intact and functional
5) Several separate lesions can be treated simultaneously
6) The same lesion(s) can be repeatedly treated
7) Cosmetic results are equal or superior to conventional modalities in most cases
8) Risk of skin photosensitivity for at least 4-6 wks
9) Local anesthesia is sometimes required during light irradiation
10) More efficient than ALA-PDT in treatment of nodular lesions
Several studies have shown that the location of BCCs is an important factor affecting PDT results.149, 150, 156 For example, BCCs located on the nose or eyelid had higher PR and higher recurrence rates after Photofrin-PDT than those located at other sites.149, 150 Such a "site effect" has also been seen in topical ALA-PDT of solar keratoses (SK). Wolf et al.157 achieved 93.6% CR in 204 SK lesions on the face, scalp or neck, whereas only 48.9% CR was achieved in lesions on the forearm or the dorsum of the hand. Similar findings were also obtained by Szeimies et al.158 Apparently, the amount and tissue distribution of ALA-derived PpIX fluorescence can vary from one part to another of the skin10 as well as of skin lesions.157 Bandieramonte et al.154 treated 42 BCC lesions with HpD-PDT and obtained a CR rate of approximately 50%. They found that small, persistent areas of BCC appeared to be related to high pigmentation of the lesions or to the "border effects" (an insufficient dose of light at the border) when irradiation was performed with multiple adjacent fields. Similarly, a study by Calzavara-Pinton showed no effect of topical ALA-PDT on pigmented BCCs.15
Cutaneous SCCs are not as sensitive as BCCs to PDT with ALA or HpD/Photofrin.138 The neoplastic cells of SCCs may not produce ALA-derived PpIX or selectively uptake HpD/Photofrin as much as the neoplastic cells of BCCs (Peng et al., unpublished data). In addition, some superficial SCCs evaluated clinically were actually those that infiltrated into deeper layers of the skin, where the lesions might not receive enough ALA and/or light irradiation.116
It should be pointed out that PDT is, in general, still considered a palliative modality rather than a first treatment for most cancer patients. Therefore, most patients receive multiple therapies prior to PDT, such as ionizing radiation, surgical excision, cryotherapy, topical 5-fluorouracil, electrodesiccation, or curettage. In other words, the majority of patients fail or recur on multiple other therapies prior to PDT. This situation may reduce PDT efficiency, particularly in cases of topical ALA-PDT. Moreover, a comparison of the treatment results of skin tumors achieved with different therapies should be limited to lesions of similar size, location, and histopathologic type. In addition, general medical conditions of patients should be considered.
Topical ALA-PDT has several potential advantages over conventional therapies. It is noninvasive, has a short photosensitization period, produces excellent cosmetic results, and is well tolerated by patients. Moreover, it can be used to treat multiple superficial lesions in short treatment sessions, patients who refuse surgery or have pacemakers and bleeding tendency, and lesions in specific locations such as the oral mucosa or the genital area. It can be used as a palliative treatment, and it can be applied repeatedly without cumulative toxicity. However, for a new modality to become clinically acceptable as a routine treatment, it must possess a therapeutic advantage over existing conventional treatments. For example, as shown in Table 10,159 several conventional modalities are available for the treatment of BCCs with low recurrence rates during short term and long term follow-ups, although the results vary substantially, probably due to variation in the location, size, and histopathologic subtype of BCCs and to the physician's experience. With current protocols, PDT in which ALA or HpD/Photofrin is used does not seem to be superior to conventional treatments for skin tumors, but some individual clinical PDT trials have achieved comparable or favorable results with outstanding cosmetics, particularly in cases of large and multiple lesions. In addition, successful results with PDT of skin tumors have recently been obtained using second-generation photosensitizers, such as benzoporphyrin derivative monoacid ring A (BPD-MA),160, 161 tin-ethyl etiopurpurin (SnET2),162 and mono-l-aspartyl chlorin e6 (NPe6).163, 164 These second-generation dyes have a larger absorption peak (approximately 660-690 nm) than PpIX/HpD/Photofrin and much less risk of prolonged cutaneous photosensitivity.
Table 10. Summary of Short Term versus Long Term BCC Recurrence Rates with Conventional Treatment Modalitiesa
Short term (<5 yrs)
Long term (5 yrs)
BCC: basal cell carcinoma.
Adapted by permission of the publisher from Rowe et al.159 Copyright 1989 by Elsevier Science Inc.
Curettage and electrodesiccation
SYSTEMICALLY ADMINISTERED ALA-BASED PDT FOR HUMAN TUMORS OF THE AERODIGESTIVE TRACT
In 1993 Grant et al.165 reported that ALA-derived PpIX peaked at 4-6 hours in oral cavity SCCs of all 4 patients examined after oral administration of 30-60 mg/kg ALA, and returned to background within 24 hours. Similar kinetics of ALA-derived PpIX were also obtained in sigmoid colorectal adenocarcinoma of 3 patients subsequent to oral administration of ALA at doses of 30 or 60 mg/kg.166 In general, there is no gastrointestinal (GI) tumor selectivity (relative to surrounding normal mucosa) of ALA-derived PpIX with a dose lower than 40 mg/kg, although wide variation was seen from one patient to another (Peng et al., unpublished data). However, the selectivity of ALA-derived PpIX appeared to be improved by using a higher dose (60 mg/kg),166 and the PpIX ratio of tumor to normal mucosa was found to be about 5:1 in colon carcinomas.16, 167 In addition, ALA-derived PpIX levels were found to be higher in tumors of the esophagus, duodenum, and lowest part of the large bowel than in colorectal tumors, but doubling the ALA dose increased significantly the amount of PpIX in the colorectal tumors.16, 167 With ALA-PDT, Fan et al. obtained CR only in 2 of 7 oral SCCs after oral administration of 60 mg/kg ALA (divided into 3 equal fractions over 2 hours), followed by light exposure (628 nm) up to 200 J/cm2 (up to 200 mW/cm2), but all 13 premalignant lesions treated obtained full-thickness epithelial necrosis and elimination of dysplastic epithelium.168 Similarly, 8 of 10 patients with GI tumors given a red laser light exposure (628 nm) at a dose of 50-100 J/cm2 (50 mW/cm2) after ALA administration demonstrated a only superficial necrosis of the tumors (0.5-1.5 mm in depth).16 Mlkvy et al.169 compared the effect of ALA-PDT (oral, 60 mg/kg, 6 hours before light irradiation) with that of Photofrin-PDT (i.v., 2 mg/kg, 48 hours before light exposure) in the treatment of duodenal and colorectal polyps in 6 patients with familial adenomatous polyposis. They found that the PDT-induced tumor necrosis was only superficial (up to 1.8 mm in depth) in the case of ALA but much deeper in the case of Photofrin. Warloe et al. treated 9 patients with rectal tubulovillous adenomas with ALA-based or Photofrin-based PDT after the main bulk of the primary tumors had been endoscopically resected.170 Nine patients were treated during a total of 14 PDT sessions, 5 receiving Photofrin and 9 receiving ALA, respectively. The tumors in all 5 Photofrin-PDT sessions showed complete regression. However, they all recurred 4-20 months after treatment. Four of 9 ALA-PDT recipients achieved CR, and no recurrence was seen after 3-10 months. In addition, two of the cases with PR after the first treatment were given a second ALA-PDT, and both of them showed CR. Thus, systemically administered ALA-based PDT is simple and safe and may be a promising technique for the treatment of small and superficial mucosal precancerous and cancerous lesions of the aerodigestive tract, such as dysplasia in Barrett's esophagus and small tumors.171 Optimization of the technique parameters is required for the larger lesions.
DETECTION OF EARLY BLADDER CARCINOMA BY ALA-DERIVED PPIX FLUORESCENCE
Precancerous and cancerous urothelial lesions, such as tiny dysplasia, carcinoma in situ, or flat papillary tumors, can be easily missed during conventional cystoscopy under white light. Recently, Kriegmair et al.172-174 used intravesical instillation of a pH-neutral 3% ALA solution for 2-3 hours followed by fluorescence cystoscopy with violet light from a krypton ion laser (406.7 nm) for excitation of ALA-derived PpIX. A sharply marked red fluorescence induced from ALA in the urothelial lesions could be easily observed with the naked eye during the fluorescence cystoscopy. The mean ratio of fluorescence intensity between urothelial carcinoma and normal urothelium was 17:1. The fluorescence microscopy revealed that the PpIX fluorescence was limited mainly to the urothelial layer. Little was detected in the submucosal or muscle layers of the bladder wall, indicating that there may be no direct phototoxic damage to vessels and muscle cells of the bladder wall. In a group of 104 patients with bladder carcinoma, the sensitivity of the ALA-derived PpIX fluorescence cystoscopy in detection of neoplastic urothelium was 96.9%, significantly higher than that of conventional white light cystoscopy (72.7%).18 A similar finding was also obtained by Jichlinski et al.175 Thus, ALA-derived PpIX fluorescence cystoscopy may be useful for detecting the precise sites of bladder urothelial lesions, especially in cases of suspicious or positive urine cytologic findings. Moreover, a decrease in recurrence rates may be expected for transurethral resection of bladder carcinoma performed under violet light after intravesical ALA instillation. Little information exists as to the use of PDT with intravesical instillation of ALA for the treatment of superficial urothelial tumors.176 However, intravesical, oral, or i.v. administration of ALA to rats or pigs led not only to an accumulation of ALA-derived PpIX in the urothelium and bladder tumors, but also to a destruction of the lesions after light exposure.177-181 Photofrin and some other fluorescent agents have also been tried to detect early stages of bladder carcinoma.182, 183 However, the procedures are subject to considerable disadvantages. For example, Photofrin is usually given systemically to a patient, with a risk of skin photosensitivity. Moreover, the fluorescence quantum yield and/or the absolute amount of Photofrin are so low that highly sensitive devices are required to detect the Photofrin fluorescence in the urothelial lesions. Finally, the ratio of the fluorescence intensity of Photofrin between chemically induced rat bladder tumors and normal rat bladder urothelium was found to be only 2-5:1 after i.v. injection,184 whereas the ratio of the ALA-derived PpIX fluorescence between the same animal tumor model and normal bladder mucosa was shown to be 20:1 after topical application of ALA to the urinary bladders,185 showing a highly selective accumulation of ALA-derived PpIX in the malignant urothelium.
DETECTION AND PDT OF EARLY-STAGE LUNG CARCINOMA WITH ALA-DERIVED PPIX
Inhalation of ALA has potential for detecting bronchial malignancies. Baumgartner et al.186 have reported that 4 patients with positive sputum cytology but negative white light bronchoscopy received a 10% NaCl-ALA solution by means of a conventional nebulizer. Three hours after inhalation, patients were examined by fluorescence bronchoscopy. ALA-derived PpIX fluorescence spectra could be clearly recorded, and the PpIX fluorescence in several dysplastic areas was imaged. Moreover, Huber et al.19 studied 7 patients with lung malignancies; 250 or 500 mg ALA dissolved in 5 ml saline were inhaled with a PARI-boy jet-nebulizer, and the endobronchial ALA deposition was estimated to be 25 mg and 50 mg, respectively. Fluorescence bronchoscopy was also performed 3 hours after inhalation, and biopsies were taken. A strong selective PpIX fluorescence was found only in the areas of tumor, dysplasia, or severe inflammation, although a weak PpIX fluorescence was observed in the normal or inflamed areas. With this technique, patients coughed during inhalation, but the peak expiratory flow did not change. Awadh and Lam66 examined the efficacy of ALA-derived PpIX as a photosensitizer for photodetection and PDT of early-stage lung carcinoma in 5 patients with 8 sites of carcinomas in situ after oral administration of 25-60 mg/kg ALA. They claimed that ALA-derived PpIX fluorescence was poor for selective photodetection of tumors, with diffuse false positive PpIX fluorescence in areas of inflammation or metaplasia. However, after light exposure with 630 nm from a KTP-dye laser at a light dose of 200 J/cm2 (using microlens or a cylindrical diffuser), complete response was achieved in 7 of 8 sites treated, with a follow-up of 1-12 months. Moreover, no skin photosensitivity was observed. This indicates that systemically administered ALA-PDT has promise as a treatment for early stages of lung carcinoma.
DETECTION OF MALIGNANT GLIOMA BY ALA-DERIVED PPIX FLUORESCENCE
Complete tumor removal by surgery is crucial for long term survival of patients with malignant glioma. However, uncritical resection may be deleterious to neurologic function. Thus, techniques are required to provide clear intraoperative tumor identification for optimal tumor resection. Stummer et al.187 tried intraoperative photodetection of ALA-derived PpIX fluorescence in 3 patients with multiform glioblastoma. The patients were given 10 mg/kg orally 3 hours before anaesthesia. The PpIX fluorescence was excited by violet blue xenon light and visualized with longpass filter goggles. Thirty-five biopsies were taken from fluorescing, neighboring, and nonfluorescing tissues for histopathologic evaluation. Normal brain tissues showed no PpIX fluorescence, whereas tumor tissues that infiltrated adjacent normal brain tissues demonstrated a clear red fluorescence. Such a method provided 85% sensitivity and 100% specificity for the detection of malignant glioma tissues. This suggests that ALA-PDT of brain tumors could be possible with a high therapeutic selectivity.
LIGHT DOSIMETRY FOR ALA-PDT
Choice of Light Source
A number of different light sources are being used in clinical and experimental PDT--lasers as well as nonlaser light sources (including light-emitting diode arrays, fluorescent tubes, and incandescent lamps), continuous as well as pulsed sources. A laser offers significant advantages whenever fiberoptics are needed to reach the tumor. However, ALA-PDT is mostly used in the treatment of tumors at the surfaces of organs (the skin, bladder, and aerodigestive tract). In such cases, lamps may be as well suited as lasers. Nonlaser light sources emit significant fluences of infrared radiation together with light useful for PDT.188 Infrared radiation should be filtered out to avoid hyperthermia, although some investigators find that mild hyperthermia (40-42 °C) acts additively or synergistically with PDT.189, 190 To avoid hyperthermia, a fluence rate lower than 150 mW/cm2 should be used. Figure 2 (9K) shows the emission spectrum of a halogen lamp with filters constructed by one of our colleagues, H. B. Steen (of the Biophysics department at our institution) for ALA-PDT, together with the absorption spectra of PpIX and its photoproducts. It has been reported that pulsed light may have a deeper penetration into tissue than continuous wave (CW) light.191 If this is true, the effect must be due to saturation of the normal absorbers in tissue, mainly melanin and hemoglobin. Extremely short light pulses would be needed to reach saturation because the lifetimes of excited states of the tissue chromophores are very short.191 Thus, some investigators find little difference in the efficiency of pulsed light and CW light with respect to PDT efficiency.192, 193
Oxygen Depletion during ALA-PDT
Oxygen is needed in PDT reactions. We have found that the PDT efficiency is halved when the concentration of O2 is reduced to 1% (∼14 μM) from 20% in well-oxygenated tissue, which is similar to what was found for ionizing radiation.194 During PDT the concentration of O2 in tissue is reduced in two ways: through damage of the vascular system and through O2 consumption in the oxidative reactions taking place. Thus, the blood perfusion is a main determinant for the limiting light fluences above which O2 depletion occurs. It has been observed that in ALA-PDT a given exposure at a high fluence rate leads to less skin damage than the same exposure given at a low fluence rate (Kennedy J C, unpublished data). Usually, at clinical doses of the commonly used sensitizers for i.v. injection, O2 depletion is of concern at fluence rates above 50 mW/cm2. In the case of ALA-PDT, however, the PpIX concentration is so low that significantly higher fluence rates can be used without any risk of reducing the efficiency by O2 depletion. It should be noted that what one may call the "PDT-dose rate" is proportional to the product of the sensitizer concentration in the tissue, the extinction coefficient of the sensitizer at the treatment wavelength, and the fluence rate of the light. Thus, in the case of ALA-PDT with light within the Soret band of PpIX, one would expect to find O2 depletion at a fluence rate of only 5% of that giving O2 depletion when applying light at 630 nm. In any case, surface irradiation would lead to O2 depletion only in the upper part of the tissue (which may be good, because it would help save normal skin) because the space irradiance decreases very rapidly with increasing distances from the surface. Typically, the space irradiance is halved per 1 mm (at 400 nm) to 3 mm (at 630 nm) down into the tissue.
Choice of Optimal Wavelength for ALA-PDT
In most cases of ALA-PDT, light at 630 nm is applied. However, down to about 2 mm from the surface in human skin and muscle tissues as well as in BCC lesions, light in the Soret band (410 nm) would give the largest cell inactivation, whereas at depths exceeding 2 mm, 635 nm light may be optimal (Fig. 3 (7K)).195 Similar findings were obtained by others.196 Basically, the choice of the optimal wavelength for PDT should be made on the basis of the appropriate action spectrum. One convenient method would be to measure the action spectrum of photobleaching of the dye, since that process is caused by generation of singlet oxygen, which is also the cytotoxic photoproduct.197 As photosensitizing photoproducts with an absorption peak around 670 nm are formed during ALA-PDT, it may be advantageous to use a broad-band light source with an emission spectrum that also covers part of the action spectrum of the photoproducts (Fig. 2 (9K)).
Dosimetry, in relative units, can also be conveniently determined by measuring the rate of photobleaching of the sensitizer.197
PHOTODEGRADATION OF PPIX
PpIX is rapidly degraded during ALA-PDT.198-200 In view of this, the real PDT dose is where C(Fε t) is the concentration of sensitizer in tissue, decreasing as a function of the product of fluence rate (F), extinction coefficient (ε), and time (t). Photodegradation of PpIX may occur at a low rate even in the absence of O2.201, 202 During ALA-PDT it is mainly the PpIX present in the upper 0.1 mm of tissue that is degraded to any significant extent.200 If the concentration of PpIX in the skin and other tissues overlying a tumor is so low that most of it is photodegraded before irrepairable PDT-induced damage is caused, photobleaching can be taken advantage of. In such cases there is no upper limit to the light exposures one can apply and it is possible to eradicate tumors, which usually contain significantly higher concentrations of sensitizer than the normal surrounding tissues, without any unacceptable damage to normal tissue.199
Photodegradation Products of PpIX
When PpIX is exposed to light, several chlorin-type photoproducts are formed.203-207 When proteins are present, some of the protein-derived photoproducts react with PpIX to produce secondary PpIX-derived photoproducts. One of the major products is photoprotoporphyrin, which is itself a good photosensitizer but also a photolabile molecule.201, 208-210 Being a chlorine, it has a relatively strong absorption at about 670 nm,195, 201, 209 which some of the PDT lamps have been constructed to cover.195 However, in most cases the amount of photoprotoporphyrin formed is so small that it does not play a major role for ALA-PDT.211
MAJOR CURRENT CHALLENGES
Much knowledge has already been obtained about the metabolism and biodistribution of ALA and porphyrin precursors in the heme biosynthetic pathway. Recent studies have shown a high uptake of ALA by more rapidly proliferating cells. Together with possibly low activity of ferrochelatase, this favors porphyrin accumulation by tumor cells, thus providing a biologic rationale for the clinical use of ALA-based diagnosis and PDT. Clinical applications of topical ALA-PDT have already achieved promising results, indicating that this modality is an effective and practical method for the treatment of superficial benign and malignant diseases of the skin and internal hollow organs.212 Future research should be intensified to determine what mechanisms are responsible for recurrence of some skin tumors treated with ALA-PDT, for wide individual variation (among cells, tissues, and patients) of the concentration and localization of ALA-derived PpIX, and for poor gastrointestinal tumor selectivity. Moreover, the parameters affecting ALA-based diagnosis and PDT must be optimized, and the efficacy of the modality with a long term follow-up has yet to be compared with those of conventional therapies in controlled clinical trials.
Topical ALA-PDT for Skin Disorders
Clearly, patients with certain categories of skin lesions do benefit from ALA-PDT. These are (1) patients with lesions that if resected would result in significant cosmetic or functional impairment (i.e., lesions in midface, including the nose and the perioral and periocular areas; ear lesions; etc.), (2) patients with multicentric lesions, (3) patients who have failed previous therapy, and (4) patients who are medically unfit to undergo surgery and/or general anesthesia (e.g., recent myocardial infarction). Currently, however, the techniques of topical ALA-PDT for skin disorders have not been optimally established. In particular, inhomogeneous distribution or lack of selective accumulation of ALA-derived PpIX in nodular and infiltrating skin tumors (e.g., BCC and SCC) suggests that topical ALA application with the current delivery procedure may not be a reliable regimen for ALA-PDT treatment of such diseases.
The efficacy of ALA-PDT may be improved by the following approaches:
1Certain agents may improve ALA penetration into deep lesions without reducing selectivity. The choice of solvents may be an important factor in determining the properties of the enhancer. Several physical methods may also enhance ALA penetration, such as tape-stripping, partial curettage of lesions, ultrasound, microwave irradiation, and iontophoresis.
2Potent inhibitors of ferrochelatase and protoporphyrinogen oxidases and modulators of heme and chlorophyll biosynthetic pathway could be used to manipulate cellular biochemistry (including intracellular iron metabolism) for enhancing the production of ALA-derived porphyrins. Preclinical studies have already shown that several compounds, such as DMSO, EDTA, DFO, 2-allyl-2-isopropylacetamide, 1,10-Phenanthroline, and 1,2-diethyl-3-hydroxypyridin-4-one, have potential for increasing porphyrin accumulation from ALA. In addition, some cell-stimulating compounds, such as lectin, may help cells and tumors to accumulate ALA-derived PpIX selectively.
3Determining the optimal time for topical ALA application would allow sufficient ALA penetration and ALA-derived porphyrin production in whole lesions, providing a clinical practical convenience for both physicians and patients.
4Direct intralesional injection of ALA could be administered in some cases.
5Systemic administration (oral/i.v.) of ALA could lead to a more homogenous tissue accumulation of ALA-derived PpIX.
6Simpler, cheaper, and more efficient light delivery systems should be constructed with respect to optimal wavelengths of photoactivating light for both PpIX and its chlorin-type of photoproducts.
7The photobleaching property of ALA-derived PpIX can be used to increase the selectivity of the ALA-PDT effect.
8Intermittent applications of activating light could increase the ALA-PDT efficiency.
9Repeated ALA-PDT treatments may be advantageous because this modality has no side effects or cumulative toxicity.
10A better understanding of light distribution in tissue and improved dosimetry procedures will lead to improvements of ALA-PDT.213
PDT with ALA Ester Derivatives
Topical ALA-PDT is, to some extent, restricted by the rate of uptake of the hydrophilic ALA by neoplastic cells and/or its poor diffusion through the skin lesions, particularly in the case of thick lesions. ALA ester derivatives would show better biologic availability in the cutaneous lesions due to their lipophilic character. Such ALA derivatives are deesterified by esterase in cells and tissues. We therefore studied several ALA ester derivatives (ALA esterified with C1-C3 and C6-C8 chained aliphatic alcohols), and found that in both WiDr and NHIK human carcinoma cell lines in vitro, esterification of ALA with the long chain (C6-C8) alcohols produced ALA-derived PpIX more efficiently than did nonesterified ALA (Fig. 4 (5K)). Short chain ALA esters (C1-C3) were less efficient than ALA in inducing ALA-derived PpIX. Similar results have also been recently reported.214 The PpIX induced from nonesterified or esterified ALA was found to be equally efficient in sensitizing the tumor cells to photoinactivation.22 In animal studies, by means of a fiberoptic point monitoring system and fluorescence microscopy, the ALA esters with the short chain alcohols were found to be deesterified and converted into porphyrins in the normal skin of mice. The porphyrin fluorescence produced from the ALA esters was similar or stronger than that induced by ALA (in contrast with results in vitro) (Peng et al., unpublished data). Moreover, preliminary studies have shown that topical ALA methylester-PDT resulted in a stronger growth inhibition of WiDr tumor xenografts in vivo than topical ALA-PDT (Peng et al., unpublished data). In clinical trials with nodular BCC lesions, the porphyrin fluorescence derived from ALA esters with the short or long chain alcohols was found to be stronger and a more homogenously distributed and to have a better selectivity than that induced by ALA (Peng et al., unpublished data). It is noteworthy that topical ALA-ester-PDT leads to significantly less pain during or after light exposure than ALA-PDT. The reasons for this are not fully understood. Because it results in less pain and higher selectivity with better therapeutic effectiveness expected, ALA methylester is used as a preferable drug for topical PDT at the Norwegian Radium Hospital. Overall, the ester derivatives of ALA may have advantages over nonesterified ALA in topical ALA-PDT of superficial lesions. Furthermore, the high selectivity of PpIX induced by ALA esters may have potentials for clinical diagnostic purposes. However, more controlled clinical work is needed; in particular, the pharmacokinetics and toxicity of ALA esters should be carefully studied.
PDT Using a Combination of ALA and Photofrin
The major side effect associated with Photofrin-based PDT is the prolonged risk of skin photosensitivity. This restricts clinical PDT application to a considerable extent. The main advantage of using ALA-PDT is that ALA-derived PpIX is cleared from the body within 24-48 hours after systemic ALA administration. This would reduce or avoid the risk of prolonged skin phototoxicity. Our previous studies have shown that efficient eradication of tumor by PDT requires destruction of both cellular components and vascular stroma of tumor.215, 216 Because PpIX synthesized endogenously from ALA localizes within tumor cells and Photofrin distributes mainly in vascular stroma of tumors (Fig. 5 (60K)), PDT with a mixture of ALA (at a therapeutic dose) and Photofrin (at a low dose that would reduce or avoid the risk of skin photosensitivity) may destroy both neoplastic cells and vascular stroma of tumor tissue. Thus, we combined ALA with Photofrin in the treatment of human WiDr tumor xenografts. The dose of ALA applied was 250 mg/kg, and that of Photofrin was 1 mg/kg, a dose 5-20 times lower than therapeutic doses for such tumor xenografts and one that does not induce any skin photosensitivity. PDT with such a combination inhibited the growth of the tumors more efficiently than PDT with ALA (250 mg/kg) or Photofrin (5 mg/kg) alone (Peng et al., unpublished data). These preliminary studies strongly suggest that the combination of ALA and Photofrin not only synergistically enhances PDT efficiency of tumor but also avoids the risk of photofrin-induced skin phototoxicity.
ALA-PDT for Superficial Lesions of Internal Hollow Organs
It is generally accepted that PDT has a curative effect on small superficial lesions, probably mainly due to a sufficient light penetration of such small tissue volumes. Preferential accumulation of ALA-derived PpIX in the mucosal lesions of the aerodigestive tract, the genitourinary tract, the bronchial tree, and, to a much lesser extent, the submucosal and muscle layers (in contrast with HpD and Photofrin, which distribute mainly in the submucosal vascular stroma) allows selective destruction by PDT of the small epithelial precancerous and cancerous lesions of the hollow organs with low risk of damage to deeper layers. The tumor selectivity of ALA-derived PpIX may be improved by fractionated ALA administration8, 217, 218 or by the use of liposome-encapsulated ALA.219 In addition, ALA-PDT may be used as an intraoperative adjuvant modality to destroy residual tumor cells after surgical debulking of the tumor, because ALA-derived PpIX localizes in individual malignant cells rather than in tissue stroma.
ALA-Derived PpIX Fluorescence for Detection of Tumor, Estimation of Optimal Time for Light Activation, and Prediction of PDT Response
A preferential distribution of ALA-derived porphyrins in tumor cells provides the possibility of in situ detection of the porphyrin fluorescence in the superficial transformed epithelial cells of the skin and internal hollow organs by means of fiberoptic point monitoring systems or of fluorescence imaging systems.220, 221 Such a procedure can be performed noninvasively in vivo with immediate results and can also be used for pharmacokinetic studies.83, 220-223 Topical and local internal administration of ALA may be used for such diagnostic purposes because intravesical, intrauterine, and inhalant administrations do not produce generalized cutaneous photosensitivity. Currently, promising data are available concerning diagnosis of early stages of bladder carcinoma. ALA ester derivatives may be more suitable for such fluorescence detection because they appear to induce PpIX with better tumor selectivity. Furthermore, such fluorescence detection in situ may be used to estimate the optimal time for light activation and the rate of PpIX photobleaching after therapeutic light irradiation, and may also be used to predict the outcome of PDT. The correlation of the fluorescence values obtained by such fluorescence measurements with actual ALA-derived PpIX levels is required, as the PDT effect depends, to some extent, on the tissue concentration of photosensitizer.177, 224 Furthermore, such macroscopic fluorescence measurements may not necessarily agree with the microscopically time-dependent intracellular/extracellular localization patterns of ALA-derived PpIX.
Other Potential Uses for ALA-PDT
Attractive possibilities exist for the use of ALA to detect and treat malignant cells in blood, as highly preferential accumulation of ALA-derived PpIX has been shown to occur in the circulating transformed cells7, 127, 225 (Peng et al., unpublished data). Thus, flow cytometry of blood or marrow cells of a cancer patient incubated with ALA in vitro may permit detection of very low concentrations of certain types of malignant cells. Subsequently, one may hope that the malignant cells can be selectively killed by light exposure before autotransplantation of the blood and marrow. In addition, the modality may have potentials for photoinactivation of virus in blood products and parasitized erythrocytes. ALA in combination with light may also be used as both a diagnostic and a therapeutic means of cardiovascular application.36
The authors thank Drs. J. M. Gaullier and G. B. Kristensen for fruitful cooperation and V. Iani, H. Heyerdahl, E. Hellesylt, and W. Danielsen for excellent technical assistance.