Photodynamic therapy: a clinical reality in the treatment of cancer

Transcription

1 Photodynamic therapy Review Photodynamic therapy: a clinical reality in the treatment of cancer Colin Hopper Photodynamic therapy (PDT) is a minimally invasive treatment with great promise in malignant disease. It can be applied before, or after, chemotherapy, ionising radiation, or surgery, without compromising these treatments or being compromised itself. Unlike radiotherapy and surgery, it can be repeated many times at the same site. Response rates and the durability of response with PDT are as good as, or better than, those with standard locoregional treatments. Furthermore, there is less morbidity and better functional and cosmetic outcome. PDT is valuable for premalignant conditions such as mucosal dysplasia and carcinoma-in-situ. The excellent cosmetic outcome makes it valuable for skin lesions and for lesions of the head, neck, and oral cavity, where another advantage is that it has negligible effects on underlying functional structures. With endoscopic delivery of light to hollow structures, PDT has been successful in the treatment of early gastrointestinal cancers, such as oesophageal cancer, and lung cancer. The superficial effects of PDT can be exploited in the treatment of large areas such as the pleura and peritoneum, where curative radiation doses cannot be tolerated by underlying normal tissue. PDT is an ideal adjuvant therapy when surgical resection of solid tumours might leave behind residual microscopic disease. Interstitial light delivery, where light is fed directly into solid tumours, allows PDT to be used for large, buried tumours that would otherwise require extensive surgical resection. Lancet Oncol 2000; 1: Photodynamic therapy (PDT) is a minimally invasive technique and, unlike ionising radiation, can be applied repeatedly at the same site. It has particular appeal in oncology because the use of chemotherapy, ionising radiation, or surgery does not preclude the use of PDT, and all of these approaches can be used in a patient who has received PDT. PDT uses a photosensitising drug that is activated by exposure to light of a specific wavelength. Illumination of the target tumour site by light at the activating wavelength results in destruction of cells by a non-free radical oxidative process. PDT is a cold photochemical reaction, and the photosensitising agents used are of inherently low systemic toxicity. The low morbidity and negligible functional disturbance that result from this combination offer many advantages in the treatment of certain malignant and premalignant disorders. The interaction of drugs and light has been recognised for several thousand years, but it has only recently become a viable clinical option in the treatment of cancer through 212 Figure 1. The principles of PDT. developments in photosensitising agents, light activation devices, and light delivery systems. Evidence has accumulated that the response rates and the durability of the response are as good as, or better than, those with standard locoregional treatments. This review summarises the current status of the technique and considers its future in management of cancer. Basics of PDT The three fundamental elements of PDT are oxygen, a photosensitiser, and visible light (Figure 1). The photosensitiser is activated by the light and interacts with molecular oxygen to produce an excited state reactive singlet oxygen. This moiety is highly cytotoxic, with a short lifetime (<0.04 s) and a short radius of action (<0.02 m). The direct cytotoxic activity and microvascular damage contribute to the destruction of tumour cells, which is manifested as swelling and formation of necrotic tissue. This tissue eventually sloughs away (or is resorbed), and there is normal healing and re-epithelialisation of the treated site. Because PDT is a cold photochemical process, there is no tissue heating, and connective tissues such as collagen and elastin are largely unaffected. There is therefore much less risk to the integrity of underlying structures than with thermal laser techniques and surgery. Most photosensitisers are administered systemically, although some can be applied topically in the treatment of skin cancer. After a period to allow the photosensitiser to collect in the target tissue, low-power light of the appropriate wavelength is directed onto the tumour. The depth of effect that can be achieved depends on the photosensitiser used. CH is at the Maxillofacial Unit, University College London Hospitals NHS Trust, the Eastman Dental Institute, and the National Medical Laser Centre, London, UK Correspondence: Mr Colin Hopper, National Medical Laser Centre, Department of Surgery, Royal Free and University College London Medical School, Riding House Street, London W1W 7EJ, UK. Tel: Fax: c.hopper@ucl.ac.uk. THE LANCET Oncology Vol 1 December 2000 Courtesy of A Kübler.

2 Review Photodynamic therapy Human tissue transmits light most effectively in the red part of the visible spectrum, and the newer photosensitisers with a strong absorption band in this region (630 nm and above) can be activated to a depth of 1 cm. Most photosensitisers have more than one absorption band, a property that can be exploited to control the depth of effect for different clinical applications. Systemic administration of the photosensitiser leads to a period of unwanted residual photosensitivity that must be managed until the drug is eliminated. Treatment times vary substantially, and the time is related to the absorption of light by the photosensitiser and the efficiency of transfer of light energy to oxygen. Typical treatment times for first-generation sensitisers are about 1000 s; newer photosensitisers can induce effective cell killing with the application of around 200 s of low-power laser light. Light-emitting diodes and xenon lamp sources are commonly used for dermatological applications, but lasers are the most convenient and controllable light source. Laser light is coherent and monochromatic and can be directed along fibreoptic cables, allowing light to be introduced into hollow organs and deep-seated tumours (Figure 2). Diode lasers, which are much cheaper and more portable than metalvapour or tuned-dye lasers, have become the preferred light source. The simplicity of the process has much to offer the clinician. Superficial tumours can generally be treated with local anaesthesia and sedation. The lack of major systemic toxicity (other than residual photosensitivity) limits the need for other medication after treatment. Superficial treatments do not require sterile theatre conditions and can be given in an outpatient clinic. Photosensitisers Photosensitisers are probably incorporated directly into cellular membranes, but they do not seem to accumulate within cell nuclei. This conclusion is supported by observations that PDT is not mutagenic in vitro. There is evidence that photosensitisers are taken up and retained preferentially by neoplastic tissue, but as yet this effect is not sufficiently pronounced to allow a selective clinical response. The exception is in brain tumours, where the ratio of photosensitiser in tumour to normal brain tissue reaches about 12:1. Because activated photosensitisers can also fluoresce while returning to the ground state, selective accumulation can be exploited in the fluorescence detection of tumours. The only three photosensitisers to have received approval by regulatory authorities to date are Photofrin (porfimer sodium), 5-aminolaevulanic acid (ALA) and Verteporfin (BPD, benzoporphyrin derivative). Verteporfin is approved in the USA and some other countries for age-related macular degeneration. A potent second-generation photosensitiser, Foscan (temoporfin; meta-tetrahydroxyphenyl chlorin), has been submitted for approval in the USA and Europe. First-generation photosensitisers Porfimer sodium was the first photosensitiser to receive approval, and it is now licensed for use in the oesophagus, lung, stomach, cervix, and bladder. However, it is only moderately active in tissue because the wavelength of light needed for activation (630 nm) penetrates tissue only slightly, Figure 2. CT scan showing placement of three fibres for interstitial light delivery during PDT. and the absorption band at this wavelength is weak, therefore the depth of effect is limited to 0.5 cm. The efficiency of transfer of energy from light to cytotoxic products (quantum yield) is moderate (about 0.5), and skin photosensitivity persists for many weeks. However, porphyrins have been used to great effect in the treatment of early cancers of the stomach, lung, and head and neck. ALA is a naturally occurring precursor in the heme biosynthetic pathway, and to date it has received approval only for a non-malignant disorder, actinic keratosis. It is converted to the endogenous photosensitiser protoporphyrin IX, which can be activated by red, green, and even blue light. It can be delivered in topical, oral, or intravenous formulations and can be triggered by many light sources. Its uses are constrained by its depth of effect (<0.2 cm) to cutaneous lesions such as basal-cell carcinoma, especially in disorders such as basal-cell naevus syndrome. Second-generation photosensitisers Second-generation synthetic photosensitisers have shorter periods of photosensitivity, longer activation wavelengths, and therefore increased depth of effect, higher yields of singlet oxygen, and better tumour selectivity. The groups that have been most actively investigated are the chlorins, texaphyrins, purpurins, and phthalocyanines. The synthetic chlorin temoporfin is a very potent sensitiser (with a quantum yield of singlet oxygen of about 0.87) activated at 652 nm, with a residual photosensitivity of only 2 weeks. Phase IIb studies in head and neck cancer have been completed, and this photosensitiser has also been studied in bronchial and oesophageal tumours and in some gynaecological applications. Studies have begun in human mesothelioma, liver and bone metastases, and nasopharyngeal, prostate, and pancreatic cancer. Several other drugs are under investigation, such as tin etiopurpurin and phthalocyanines. A developmental THE LANCET Oncology Vol 1 December

3 Photodynamic therapy Review Table 1. Selected clinical trials of PDT in premalignant conditions Tumour site Photosensitiser N Main results Ref Barrett s mucosa Porfimer sodium 100 Conversion of 75 80% to normal squamous epithelium in all patients; 1 dysplasia eliminated in 78; 10 of 13 tumours eliminated; stricture in 34% of patients Barrett s oesophagus Systemic ALA 5 High-grade dysplasia eradicated in all patients; no recurrence in follow-up of months Field cancerisation of Porfimer sodium patients showed CR in 6 8 weeks; 1 had residual leucoplakia; 2 3 oral cavitydeveloped new areas within 12 months but not in treated sites Oral dysplasia and Oral ALA 12 All patients showed improvement in dysplasia; treated areas healed without 4 squamous-cell carcinoma scarring; ALA effect too superficial to be effective in squamous-cell carcinoma Field change and confirmed Temoporfin 20 6 of 6 T3 lesions, 3 of 6 T4 lesions, and 9 of 14 T1/T2 cleared after one 5 oral cancer treatment Refractory transition-cell Porfimer sodium 36 58% CR at 3 months; 10 of 21 with CR had recurrences at mean follow-up 6 carcinoma of bladder of 12 months (range 9 48); 7 patients had bladder contracture Refractory carcinoma-in-situ HpD (74%) CR at 3 months; recurrence (superficial or low-grade papillary 7 of the bladder tumours) at 2 years in 14 of 18 CR followed up; 10 of 34 alive at 64 months mean follow-up; transient reduction in bladder capacity Superficial bladder cancer Intravesical ALA 10 CR in 4 patients; partial response in 2; no change in 3; progressive disease in 1; 8 mean follow-up 15 months; no bladder shrinkage CR = complete response; HpD = haematoporphyrin derivative. bacteriochlorin product SQN400 (m-thpbc, metatetrahydroxyphenyl bacteriochlorin), which absorbs light at 740 nm, has entered phase I studies for the treatment of colorectal liver metastases. Lutetium texaphyrin shows strong absorption at 732 nm, and it has been investigated in patients with unresectable or metastatic cancers of the skin and subcutaneous tissue. Skin photosensitivity lasts for less than 72 h, and a phase II study is now under way. PDT in the treatment of malignant disease Trials in many cancers have shown that PDT can achieve control rates similar to those achieved with the standard techniques of surgery and radiotherapy. The real advantages of PDT are the lower morbidity rates, improved functional and cosmetic outcome, and simplicity of the technique. PDT has therefore been investigated most extensively in disorders for which standard treatment has low efficacy or unacceptable side-effects. Examples include premalignant conditions and tumours that are inaccessible by surgery or would be associated with unacceptably high morbidity. There have been a great many abstracts and preliminary reports on PDT in cancer treatment, but until recently few studies had been done on large numbers of patients or had acquired data prospectively. Now various phase IIb studies have shown that there are several areas of oncology for which PDT will be an effective addition to current treatments. Surface illumination of malignant lesions Treatment of premalignant conditions Premalignant dysplastic lesions and non-invasive cancers are common in the mucosa of the aerodigestive and urinary tracts for example, mucosal dysplasia of the oral cavity, Barrett s oesophagus associated with high-grade dysplasia, and carcinoma-in-situ of the lung and bladder. Treatment of a premalignant lesion before it becomes invasive is clearly desirable and preferable to treatment of a solid tumour. Eradication of the lesion at this stage will also minimise the risk of metastatic disease. However, current treatments for premalignant conditions are far from satisfactory. Surgery for 214 Barrett s oesophagus (oesophagectomy) requires general anaesthesia and carries significant morbidity, including functional impairment and disfigurement. Radiotherapy for carcinoma-in-situ is effective, but it is rarely justified because of the acute and long-term adverse effects and it can typically be administered only once. PDT is emerging as an attractive option for mucosal dysplasia and carcinoma-in-situ, because a large area of mucosa, including areas of apparently normal mucosa, can be treated superficially (Table 1). Porfimer sodium and ALA have both been investigated in Barrett s oesophagus, 1,2 for which only a limited depth of effect is required. Overholt and Panjehpour 1 reported significant response rates in 100 patients treated with PDT. The main unwanted effect of circumferential treatment with PDT is stricture, which was noted to varying degrees in 34% of patients. It was treated with dilation, and the risk of this complication can be minimised by use of centring balloons during treatment. The hospital stay was substantially shorter than that required for oesophagectomy. In the oral cavity, most premalignant lesions are excised or vapourised. PDT is valuable in treating mucosa; superficial epithelial necrosis can be achieved, leaving little scarring and no cumulative toxicity. Porfimer sodium, ALA, and temoporfin have all been studied for dysplasia of the oral cavity. 3 5 Treatment of the bladder with PDT is an attractive alternative to cystectomy in patients with extensive superficial bladder tumour. Several early studies with porfimer sodium and haematoporphyrin derivative showed promising results, which have been confirmed in more recent trials. 6,7 The major unwanted effect (besides residual photosensitivity) with porfimer sodium was bladder contracture. ALA has also been investigated in superficial bladder cancer, 8 producing similar remission rates but with fewer cases of bladder irritability or contracture. These findings show the importance of matching the depth of effect with the extent of disease. Treatment of cutaneous malignant disease Skin is readily accessible to treatment by light, and much experience has been gained with PDT for malignant disorders THE LANCET Oncology Vol 1 December 2000

4 Review Photodynamic therapy Table 2. Selected clinical trials of PDT in cutaneous malignant disease Tumour site Photosensitiser N Main results Ref BCC of skin Porfimer sodium 37 patients, CR of 88%, and 12% PR; 10% recurrence at 1 year; low morbidity, lesions most common side-effects moderate pain and oedema BCC of skin Topical ALA 80 lesions 90% CR; 7.5% PR; good cosmetic results, high acceptability to patients; 10 only 3 months follow up BCC of skin; Topical ALA 16 lesions, BCC: 50% CR at 17 months; well tolerated, EMLA (topical anaesthetic) 12 Bowen s disease 36 lesions used to minimise discomfort Bowen s disease: CR of 89% at median follow up; 9% recurrence after 18 months BCC (34 superficial, 22 Topical ALA 49 patients CR in 88% of superficial BCC, 32% of nodular BCC, 81% of solar 11 nodular); solar keratoses (43) (single treatment) keratoses, 30% of Bowen s disease; repeat therapy required in 4 Bowen s disease (10) superficial BCC and 6 nodular BCC; follow up to 20 months Primary non-melanomatous Temoporfin 18 patients, 93% CR; 7% PR; mean follow up 15 months; excellent cosmetic 13 tumours of skin 97 lesions outcome, high patients satisfaction Chest wall recurrence in DPE 20 patients 20% CR; 45% PR; 35% no change; duration of response short (average 14 breast cancer 2.5 months); complications included pain, blistering, ulceration, and necrosis BCC = basal-cell carcinoma; CR = complete response; PR = partial response; DPE = dihaematoporphyrin ether. of the skin (Table 2). Most experience has been in the treatment of cutaneous basal-cell carcinoma with ALA and porfimer sodium, but there are also preliminary results in Bowen s disease and in chest-wall recurrences of breast cancer. In aqueous solution, ALA passes readily through abnormal but not normal keratin, thus inducing a degree of selective photosensitisation in abnormal tissue. Large studies have reported complete response rates with PDT of around 80%, 9 11 but a study with longer follow-up noted only 50% complete response at median follow-up of 17 months. 12 Treatment with ALA and porfimer sodium is limited to tumours of depth 0.2 cm and 0.5 cm, respectively. Thicker tumours can be treated with more powerful photosensitisers: Kübler and colleagues have reported 93% complete response with low doses of temoporfin to treat primary nonmelanomatous tumours of the head and neck. 13 PDT of skin lesions clearly has great potential, because cosmetic outcome is very important in these exposed tumours. All three of the major photosensitisers have shown good results and can be selected to achieve the desired depth of effect. PDT of chest-wall recurrences with porfimer sodium was less successful, but the investigators suggested that use of a photosensitiser activated at longer wavelengths may penetrate tissue better with improved outcomes. 14 Cancers of the head, neck, and oral cavity Skin and mucosal lesions in the head and neck region are difficult to treat satisfactorily with standard approaches. Surgery and radiotherapy, used alone or in combination, produce good success rates with excellent durability of response. However, the complex nature of this anatomical region means that control is commonly achieved at the expense of substantial functional disturbance and disfigurement or other long-lasting complications of radiotherapy, such as xerostomia and osteoradionecrosis. Furthermore, despite advances in surgical and radiotherapeutic techniques, control rates have not improved significantly during the past 50 years. PDT is particularly well suited to the treatment of lesions of the head, neck, and oral cavity because it has little effect on underlying functional structures and has an excellent cosmetic outcome. The healing that follows the necrotic stage is a normal process of re-epithelialisation, which produces negligible scarring that diminishes in most cases over the next 6 12 months. For this reason, many studies have looked at PDT in the treatment of oral cancers (Table 3). Table 3. Selected clinical trials of PDT in malignant disorders of the head, neck, and oral cavity Tumour site Photosensitiser N Main results Ref Oral cavity DPE 41 lesions Carcinoma-in-situ: CR 87%; PR 13%; 25% recurrence over months Early-stage tumours: CR 56%; PR 24% Advanced disease: results equivalent to standard therapy Oral cavity, intranasal, Porfimer sodium (33), 40 patients CR 100%; 3 patients with recurrent T1 SCC had recurrences within 15 nasopharyngeal temoporfin (7) 8 months (mean follow up 46 months); 1 patient with two T1 tumours developed metastases 2 months after PDT Oral cavity SCC Temoporfin 108 patients Carcinoma-in-situ T1 and T2: 87% CR; sustained at 1 year by 82%, 17 and at 2 years by 75%; mean survival at least 2 years Oral cavity SCC (second Temoporfin 96 patients CR 58% and overall response (including PR) 67%; mean duration of 18 primary and recurrent disease) response 11 months Oral cavity SCC (incurable by Temoporfin 64 patients, 53% overall clinical benefit; 33% objective palliative response 20 surgery or radiotherapy) 92 treatments to treatment; tumour cleared in 17% of patients SCC of lip Temoporfin 14 patients 96% CR at 12 weeks; recurrence in 2 patients at 4 and 18 months 19 CR = complete response; PR = partial response; DPE = dihaematoporphyrin ether; SCC = squamous-cell carcinoma. THE LANCET Oncology Vol 1 December

5 Photodynamic therapy Review Table 4. Selected clinical trials: PDT in the treatment of gastrointestinal and other malignant disease Tumour site Photosensitiser N Main results Ref Inoperable lung cancer Porfimer sodium 100 patients Mean endoluminal obstruction fell from 86% to 18%; improvement in 22 FVC and FEV1; overall 2-year survival 19%; mean follow up 72 months; no treatment-related mortality Early-stage lung cancer Porfimer sodium 54 patients, 50 (85%) of assessable carcinomas CR; median duration of CR 14 months; tumours 5 recurrences within 18 months SCC of lung Porfimer sodium 21 patients, CR in 15 patients; duration of response longer than 1 year in 11 patients tumours (mean follow-up 68 months); 5 patients had recurrence after 12 months Early SCC of oesophagus Temoporfin (14), 28 patients, CR of 84% in early cancers; 2 stenoses and 2 fistulas; mean follow up 24 HpD (9), porfimer 31 tumours 2 years sodium (8) Early bronchial and Temoporfin 27 patients, 83% of early cancers showed no recurrence after mean disease-free 29 oesophageal cancers 40 tumours follow up of 15.3 months; complications reduced with green light Advanced oesophageal cancer Porfimer sodium vs 218 patients Improvement in dysphagia equivalent, but tumour response 32% with 25 Nd:YAG laser PDT and 20% after laser ablation; 9 vs 2 CR Oesophageal carcinoma Porfimer sodium 77 patients Minimal complications and no treatment-related mortality; length of 28 palliation equal to, or better than other treatments Inoperable oesophageal cancer Porfimer sodium 65 patients Dysphagia relieved in all patients; 7 still alive after 2 30 months; mean 26 survival 8 months; survival influenced by performance status Early gastric cancer Temoporfin 22 patients CR in 80% of patients with intestinal cancer and 50% with diffuse 27 Lauren s carcinoma; mean follow up 12 and 20 months, respectively CR = complete response; FVC = forced vital capacity; FEV1 = forced expiratory volume in 1 s; SCC = squamous-cell carcinoma; HpD = haematoporphyrin derivative. Large retrospective studies have shown a clear benefit of porfimer sodium and the other porphyrin preparations over surgery and radiotherapy in situations such as field change disease. 15,16 Applications of ALA in oral cancer are limited by the small depth of effect. Figure 3. A 68-year-old patient with an infiltrative squamous-cell carcinoma on the right oral commissure, (a) before and (b) 3 months after PDT treatment with temoporfin. Courtesy Graham Putnam For the first time in cancers of this region, a series of prospective phase IIb studies are producing results with significant numbers of patients. Interim data from three major studies on early oral cancer, recurrent oral cancer, and a palliative study on advanced cancer with temoporfin as the photosensitiser have shown control rates in the treatment of early and recurrent disease similar to those with standard approaches, with the benefit of less morbidity, excellent cosmetic and functional outcome, and the ability to treat synchronous or metachronous disease several times without cumulative toxicity. In early disease, 17 complete response rates of 87% were obtained with one treatment of PDT, and this response was sustained at 1 year by 82% of patients (75% at 2 years). Figure 3 shows a patient with an infiltrative squamous-cell carcinoma, before and after PDT treatment with temoporfin. A small number of patients whose tumours were not completely cleared by PDT were successfully treated with surgery or radiotherapy, confirming that these procedures can be carried out at sites previously treated with PDT. In recurrent and second primary disease, 18 complete response rates of 58% were obtained with PDT alone, and there was an overall response (including partial responses) of 67%. The proportion with sustained response at 1 year was 78%. In this study a few patients received multiple treatments of PDT. In advanced disease, 20 when all other therapeutic options have been exhausted, PDT with temoporfin provided significant palliative benefit in 53% of patients. Surprisingly, given the advanced nature of the disease in this palliative trial, complete tumour control was achieved in 17% of patients. Taken together, these studies show that PDT achieves the best response rates in early disease as is the case with most treatments. However, they also show that PDT unlike radiotherapy and some surgical procedures can be repeated many times at the same site without 216 THE LANCET Oncology Vol 1 December 2000

6 Review Photodynamic therapy compromising salvage therapies. Wenig and colleagues confirmed that PDT can be used to treat very large tumours through multiple applications of drug and light to progressively debulk the tumour. 20 Treatment of other malignant disease Light can be delivered endoscopically to any hollow structure, such as the oesophagus or bronchial tree, thus avoiding major excision. PDT has been used successfully in the treatment of early gastrointestinal cancers and lung cancer (Table 4). Furuse and colleagues 21 reported complete response rates of 85% in patients with early-stage lung cancer, and Moghissi and co-workers 22 found that PDT with porfimer sodium can be effective in the palliation of inoperable lung cancer, relieving obstruction and improving forced vital capacity in a group of patients for whom other options had been exhausted. Cortese and colleagues 23 estimated that 22% of patients treated with PDT for early-stage squamous-cell lung cancer could be spared surgical resection as a result. None of these studies reported serious adverse events. A substantial amount of data has been collected on the treatment of oesophageal cancer with PDT. The standard treatment is total oesophagectomy, which is associated with a high degree of morbidity and mortality, or thermal laser ablation. Most cases of oesophageal cancer present at an advanced stage with dysphagia, and palliative therapy is commonly carried out with expandable metal stents. However, long-term complications remain. Early squamous-cell carcinoma of the oesophagus can be effectively eradicated with temoporfin or porfimer sodium PDT, 24 with few complications. Other investigators have compared PDT with thermal ablation. 25 The two treatments were equivalent in relieving dysphagia, but tumour response was sustained for longer and there were more complete tumour responses with PDT. Temporary photosensitivity is seen with PDT, but there were fewer acute adverse events, such as perforations, that can result from thermal laser therapy. Moghissi and colleagues 26 have used PDT successfully in the palliation of patients with oesophageal cancer who could not be treated with surgery, because of the advanced nature Figure 4. Placement of light delivery fibres in interstitial application of PDT (see overleaf). of the disease or the general condition of the patient. All the patients had dysphagia and a substantial proportion could not swallow liquids. Dysphagia was relieved in all the treated patients, although survival time was only a few months owing to the advanced nature of the disease. Similarly, McCaughan and co-workers 28 confirmed that PDT of advanced disease caused negligible complications and the extent of palliation was longer than that reported for other treatments. PDT is a minimally invasive technique for the treatment of oesophageal cancer but there is a risk that stenosis may develop. Grosjean et al 29 report that these and other complications can be avoided by use of the different absorption peaks of photosensitisers. They used green light at nm with temoporfin to control the depth of effect and to produce superficial areas of tissue necrosis without scarring. Intraoperative and adjunctive treatment of surface cancers The superficial effects of PDT can be exploited in the treatment of large areas, such as the pleura and peritoneum. Cancers that arise in, or metastasise to, serosa are spread over large surface areas. Although surgical resection is possible to remove gross lesions, removal of microscopic disease is not feasible. Local recurrence or persistence is common in surgical resection of peritoneal carcinomatosis, malignant Table 5. Selected clinical trials of PDT used intraoperatively and as an adjunctive treatment for surface cancers Site Photosensitiser N Main results Ref Pleural neoplasms Porfimer sodium 54 patients 4 treatment-related complications and 1 death 3 days after surgery 30 Pleural cancers Temoporfin 5 patients No deaths at 30 days; 4 patients alive with no signs of recurrence at months Malignant pleural Porfimer sodium 31 patients Estimated median overall survival 12 months; survival of stage III and 32 mesothelioma stage IV patients 8 months, and stages I and II 21 months Diffuse malignant Temoporfin 8 patients Local tumour control achieved, although distant spread not prevented 33 mesothelioma of chest THE LANCET Oncology Vol 1 December

7 Photodynamic therapy Review Table 6: Early results of interstitial application of PDT to solid tumours Site Photosensitiser N Main results Ref Subcutaneous and Porfimer sodium 9 patients CR rate for interstitial treatments 52%, rising to 81% with higher light 35 cutaneous tumours 50 tumours and drug doses; tumour volumes up to 60 cm 3 successfully treated Head and neck tumours Temoporfin 26 patients 38 treatments. All tumours responded to PDT and normal tissue 36 recovered well after treatment CR = Complete response mesothelioma, and pleural carcinomas. Radiotherapy is limited by the inability to deliver a curative radiation dose that can be tolerated by underlying normal tissue. PDT has great potential in the treatment of surface serosal cancers and, in combination with surgical debulking, in the eradication of residual malignant disease. 30 Light can be delivered to large surfaces in a short time, by deliberate use of reflected and scattered light in hollow cavities. The limited depth of effect is an advantage, in that underlying organs are spared significant damage. In malignant pleural mesothelioma, complete surgical resection of the primary tumour has a significant rate of local failure. The addition of PDT to surgery may improve local control by eradicating the remaining occult cancer cells Similarly, in a large number of solid tumours, surgical resection leaves behind residual microscopic disease that may lead to local recurrence or metastatic disease. PDT seems an ideal adjuvant therapy especially when the risk of local failure is high, such as in the abdominal cavity after resection of gastrointestinal tumours, retroperitoneal or small-bowel sarcomas, radical prostatectomy, and malignant gliomas. A review of data on treatment of more than 310 patients concluded that there is a clear trend of improved median survival after surgical resection of malignant brain tissue and one PDT treatment. 34 Several preliminary reports have appeared, which are summarised in Table 5. Interstitial treatment PDT is not limited to the treatment of surface cancers, whether cutaneous or in hollow organs. Light can be delivered directly into solid tumours by feeding laser fibres through needles placed under image guidance in the tumour (Table 6). There is a clear parallel with the early days of interstitial radiotherapy, and the principles of fibre placement are similar to those of the Paris system. This type of treatment requires all light sources to be placed in an equidistant, parallel arrangement to ensure even light distribution, through overlapping spheres of light (Figure 4). By a pull-back technique, layers of tumour can be treated, resulting in necrosis throughout even very large tumours. The necrotic tissue is resorbed rather than sloughing off. Interstitial light delivery allows PDT to be used in the treatment of large, buried tumours and is particularly suitable for those in which surgery would involve extensive resection (Figure 5). Up to 60 cm 3 of subcutaneous tumour tissue has been successfully treated in this way with porfimer sodium, 35 and consistent tumour necrosis was achieved in 38 treatments with temoporfin. 36 The most important limitation of the technique is the proximity of the tumour to major blood vessels, but otherwise this minimally invasive treatment is applicable to any site in the body. Figure 5: Pretreatment coronal MRI scans of an adenoid cystic carcinoma (a) filling the maxillary antrum and (b) the same area after interstitial PDT. 218 Conclusion Recent developments in photosensitisers and light delivery systems have substantially reduced treatment times and residual photosensitivity, while increasing the achievable depth of necrosis. Compared with standard approaches, PDT can achieve equivalent or greater efficacy in the treatment of many cancers, particularly in the head and neck THE LANCET Oncology Vol 1 December 2000

8 Review Photodynamic therapy and basal-cell carcinoma, with greatly reduced morbidity and disfigurement. The technique is simple, can commonly be carried out in outpatient clinics, and is highly acceptable to patients. PDT is not confined to the treatment of small superficial tumours. It can be repeated to debulk large tumours progressively, and it can also be applied through interstitial light delivery to large solid tumours. PDT is currently approved in the palliation of locally advanced cancers and a few early-stage diseases, but we should now consider its potential in situations that capitalise on the strengths of the technique. These include first-line treatment in early malignant and premalignant disease, adjuvant therapy for surgery, and interstitial treatment of deep-seated tumours. References 1 Overholt BF, Panjehpour M. Photodynamic therapy for Barrett s esophagus: follow-up in 100 patients. Gastrointest Endosc 1999; 49: Barr H, Shepherd NA, Dix A, et al. Eradication of high-grade dysplasia in columnar-lined (Barrett s) oesophagus by photodynamic therapy with endogenously generated protoporphyrin IX. Lancet 1996; 348: Grant WE, Hopper C, Speight PM, et al. Photodynamic therapy of malignant and premalignant lesions in patients with field cancerization of the oral cavity. J Laryngol Otol 1993; 107: Fan KF, Hopper C, Speight PM, et al. Photodynamic therapy using 5-aminolevulinic acid for premalignant and malignant lesions of the oral cavity. Cancer 1996; 78: Fan KF, Hopper C, Speight PM, et al. Photodynamic therapy using mthpc for malignant disease in the oral cavity. Int J Cancer 1997; 73: Nseyo UO, Shumaker B, Klein EA, et al. Photodynamic therapy using porfimer sodium as an alternative to cystectomy in patients with refractory transitional cell carcinoma of the bladder. J Urol 1998; 160: Uchibayashi T, Koshida K, Kunimi K, et al. Whole bladder wall photodynamic therapy for refractory carcinoma in situ of the bladder. Br J Cancer 1995; 71: Kriegmair M, Baumgartner R, Lumper W, et al. Early clinical experience with 5-aminolevulinic acid for the photodynamic therapy of superficial bladder cancer. Br J Urol 1996; 77: Wilson BD, Mang TS, Stoll H, et al. Photodynamic therapy for the treatment of basal cell carcinoma. Arch Dermatol 1992; 128: Kennedy JC, Pottier RH, Pross DC. Photodynamic therapy with endogenous protoporphyrin IX: basic principles and present clinical experience. J Photochem Photobiol B 1990; 6: Fijan S, Honigsmann H, Ortel B. Photodynamic therapy of epithelial skin tumours using delta-aminolaevulinic acid and desferrioxamine. Br J Dermatol 1995; 133: Cairnduff F, Stringer MR, Hudson EJ, et al. Superficial photodynamic therapy with topical 5-aminolaevulinic acid for superficial primary and secondary skin cancer. Br J Cancer 1994; 69: Kübler AC, Haase T, Staff C, et al. Photodynamic therapy of primary non-melanomatous skin tumours of the head and neck. Lasers Surg Med 1999; 25: Sperduto PW, DeLaney TF, Thomas G, et al. Photodynamic therapy for chest wall recurrence in breast cancer. Int J Radiat Oncol Biol Phys 1991; 21: Biel MA. Photodynamic therapy using Photofrin and Foscan and the treatment of malignancies of the head and neck. Proc SPIE-Int Soc Opt Eng 1998; 3247: Gluckman JL. Hematoporphyrin photodynamic therapy: is there truly a future in head and neck oncology? Reflections on a 5-year experience. Laryngoscope 1991; 101: Hopper C. Photodynamic therapy with Foscan (temoporfin) in primary squamous cell carcinoma of the head and neck. Proceedings Fifth International Congress on Head and Neck Oncology; San Francisco, USA. July Tan IB. Foscan-mediated photodynamic therapy (Foscan-PDT) in recurrent and second primary oral cavity cancers. Laryngo-Rhino- Otology 2000; 79 (suppl 1): S318. Search strategy and selection criteria Published data up to 1998 were already available to me for this review. More recent publications were identified through our own database of photodynamic therapy publications and PubMed. All preclinical and scientific papers and papers on non-malignant conditions (eg psoriasis, age-related macular degeneration) were excluded. Papers submitted on the same topic by the same group of investigators to several journals were reviewed to identify one in each case that was representative of the results. Case reports and (in most cases) preliminary results were excluded, and larger studies with long follow-up of patients responses were selected instead. For each tumour site, I selected, where possible studies on the largest number of patients that used a range of photosensitisers. Reviews and other reports are included in the electronic references. 19 Kübler A. Photodynamic therapy for the treatment of lip cancer. Proceedings of the 15th European Association for Cranio-Maxillofacial Surgery; 2000 Sept 5 9; Edinburgh, UK. Harcourt Brace; Wenig BL, D Cruz A, Iqbal A, Bryce RP. Foscan-mediated photodynamic therapy (Foscan PDT) in the palliative treatment of advanced head and neck cancer. Proc Am Soc Clin Oncol 2000; 19 (abstr 165). 21 Furuse K, Fukuoka M, Kato H, et al. A prospective phase II study on photodynamic therapy with photofrin II for centrally located earlystage lung cancer: the Japan Lung Cancer Photodynamic Therapy Study Group. J Clin Oncol 1993; 11: Moghissi K, Dixon K, Stringer MR, et al. The place of bronchoscopic photodynamic therapy (PDT) in advanced unresectable lung cancer: experience of 100 cases. Eur J Cardiothoracic Surg 1999; 15: Cortese DA, Edell ES, Kinsey JH. Photodynamic therapy for early stage squamous cell carcinoma of the lung. Mayo Clin Proc 1997; 72: Savary JF, Grosjean P, Monnier P, et al. Photodynamic therapy of early squamous cell carcinomas of the esophagus: a review of 31 cases. Endoscopy 1998; 30: Lightdale CJ, Heier SK, Marcon NE, et al. Photodynamic therapy with porfimer sodium versus thermal ablation therapy with Nd:YAG laser for palliation of esophageal cancer: a multicenter randomized trial. Gastrointest Endosc 1995; 42: Moghissi K, Dixon K, Thorpe JAC, et al. The role of photodynamic therapy (PDT) in inoperable oesophageal cancer. Eur J Cardiothorac Surg 2000; 17: McCaughan JS Jr, Ellison EC, Guy JT, et al. Photodynamic therapy for esophageal malignancy: a prospective twelve-year study. Ann Thorac Surg 1996; 62: Ell C, Gossner L, May A, et al. Photodynamic ablation of early cancers of the stomach by means of mthpc and laser irradiation: preliminary clinical experience. Gut 1998; 43(3): Grosjean P, Savary JF, Mizeret J, et al. Photodynamic therapy for cancer of the upper aerodigestive tract using tetra (m-hydroxyphenyl)chlorin. J Clin Laser Med Surg 1996; 14: Pass HI, DeLaney TF, Tochner Z, et al. Intrapleural photodynamic therapy: results of a phase I trial. Ann Surg Oncol 1994; 1: Baas P, Murrer L, Zoetmulder FAN, et al. Photodynamic therapy as adjuvant therapy in surgically treated pleural malignancies. Br J Cancer 1997; 76: Takita H, Dougherty TJ. Intracavity photodynamic therapy for malignant pleural mesothelioma. Semin Surg Oncol 1995; 11: Ris HB, Altermatt HJ, Nachbur B, et al. Intraoperative photodynamic therapy with m-tetrahydroxyphenylchlorin for chest malignancies. Lasers Surg Med 1996; 18: Kostron H, Obwegeser A, Jakober R. Photodynamic therapy in neurosurgery: a review. J Photochem Photobiol B 1996; 36: Lowdell CP, Ash DV, Driver I, et al. Interstitial photodynamic therapy: clinical experience with diffusing fibres in the treatment of cutaneous and subcutaneous tumours. Br J Cancer 1993; 67: Hopper C, Suhr M, Jones L, et al. Interstitial photodynamic therapy in head and neck cancer. Br J Cancer 2000; 83 (suppl 1): abstr 79. A list of references for further reading appears on The Lancet Oncology s website: THE LANCET Oncology Vol 1 December