Methyl-ALA–Induced Fluorescence in Photodynamic Diagnosis of Basal Cell Carcinoma Prior to Mohs Micrographic Surgery

Topical application of methylaminolevulinic acid (methyl-ALA) cream induces selective accumulation of the photosensitizer protoporphyrin IX (PpIX) in basal cell carcinoma (BCC). Irradiation of the porphyrin-enriched BCC with a Wood light leads to emission of a brick red fluorescence.¹ This procedure may be termed photodynamic diagnosis (PDD). In the present pilot study, we evaluate the use of PDD for the detection of BCC and delimitation of tumor size, with retrospective comparison of fluorescent images and histopathologic results from Mohs micrographic surgery (MMS) excision margins.

Twenty patients were treated using MMS (Table). Tumors were first inspected under ordinary white light illumination in each patient, and all suspect areas were documented by photography. Methyl-ALA cream (Metvix; Photocure-ASA, Oslo, Norway) was applied as a layer approximately 1-mm thick, and a margin of 5 to 10 mm was included. Treated areas were covered by an occlusive dressing (Tegaderm; 3M, Leicestershire, England) for 3 hours. Afterwards, patients were examined under a Wood light. All suspect red-fluorescing areas were photographed (Clearstone UV-DA; DigiMed Systems, Madrid, Spain).

All BCCs were excised by MMS within 1 week after fluorescence imaging without taking into consideration the fluorescence results. In the operating theater, the tumors were removed following the clinically defined margins. The first-step margins were evaluated according to the Breuninger² guidelines (slow Mohs) using paraffin-embedded specimens and hematoxylin-eosin staining. The final wound measurement after MMS, including the wound expansion, was compared with the size and PDD images of the lesion present at the initial evaluation.

The PDD fluorescence corresponded exactly to histologic tumoral margins in 14 of the 20 patients (70%). In 7 of these 14, the PDD correlated well with both the clinical extension and the histopathologic borders of the tumors (Figure 1). In the other 7, the fluorescence area was bigger than clinically expected (Figure 2). Histopathologic study of fluorescent borders showed tumoral cells. The PDD area clearly exceeded that of the white light inspection for tumor delimitation. In 3 patients, the PDD was falsely positive (2 recurrences on recent scars) (Figure 1). In 3 other cases, no fluorescence was found (false negative): 1 sclerodermiform BCC and 2 BCCs with trichilemmal differentiation without apparent epidermal connection.

Fluorescence imaging is a superficial technique that gives information about the superficial extent of tumors. It provides no information about the depth of a BCC, which can be determined only by histopathologic evaluation. The selective accumulation of PpIX in tumor tissue after 5-ALA applications depends on the permeation of the ALA through the stratum corneum³ and its diffusion through the dermis and epidermis. Because of its enhanced lipophilicity and better penetrative properties, methyl-ALA has a higher tissue specificity with a preferential enrichment of porphyrins in lesional skin than 5-ALA.⁴ The fluorescence ratios (porphyrin resistance) of BCC to normal skin are different with methyl-ALA (9:1) and 5-ALA (2:1).⁵

Photodynamic diagnosis has proven to be a highly sensitive method to detect neoplastic tissues, whereas PpIX, although sensitive, has a high false-positive rate because it accumulates in benign conditions like scars and inflammatory tissues. Generally, normal skin surrounding lesions presents a weak fluorescence, varying with each patient and location. The contrast between this basal fluorescence and the pathologic fluorescence needs to be intense and sharp enough to obtain a good demarcation in PDD. Occasionally, in patients with severe actinic damage, actinic keratoses close to BCCs will also fluoresce under PDD, which could increase the excision area unnecessarily. Future methods for fluorescence quantification should reduce the rate of false-positive lesions. The varying depths of the tumors or their low cellularity might provide an explanation for the absence of fluorescence in false-negative cases.

Our results suggest that fluorescence of BCC after methyl-ALA incubation, excited in the UV region of the spectrum, has characteristic features that differ from those seen in normal tissue, and this difference may be exploited for noninvasive diagnosis and the detection of lateral tumor margins. This technique could reduce the number of excisions in MMS, thus speeding up this procedure.

Correspondence: Dr Redondo, Department of Dermatology, University Clinic of Navarra, 31008 Pamplona, Spain (predondo@unav.es).

Financial Disclosure: None reported.