history of photodetection and photodynamic therapy, The

Photochemistry and Photobiology, Nov 2001 by Ackroyd, Roger, Kelty, Clive, Brown, Nicola, Reed, Malcolm

In a further study, in 1967, Lipson et al. (32) investigated the use of fluorescent bronchoscopy in 50 patients. Of 34 malignant tumors accessible to the bronchoscope, 32 exhibited fluorescence but none of the benign lesions did so. In the same year Gray et al. (33) studied cervical and vaginal lesions using HpD fluorescence techniques. Fluorescence was demonstrated in all but one of 34 malignant lesions, but disappointingly over half (13 of 23) of the benign lesions also fluoresced. Further histological review, however, revealed the presence of either carcinoma in situ or severe dysplasia in most of these lesions.

The following year the same group reported the use of HpD fluorescence in a large series of 226 patients, including 173 malignant tumors and 53 benign lesions (34). Fluorescence was detected in 84% of adenocarcinomas, 77% of squamous carcinomas and 62.5% of sarcomas. Increased fluorescence was observed in the squamous lesions when compared with either the adenocarcinomas or the sarcomas which only displayed low levels of fluorescence. In contrast, only 22% of the benign lesions, which included leg ulcers, postirradiation ulcers and burns, exhibited low levels of fluorescence. Though a significant difference existed between the fluorescence in benign and malignant lesions, it was, however, concluded that the low sensitivity and specificity of this technique limited the use in tumor detection. However, it was proposed that fluorescence detection may have a role in the assessment of the extent of tumor invasion, especially during endoscopic procedures.

In 1971, two otolaryngologists, Leonard from Philadelphia and Beck from Iowa, reported a study of tumor detection using HpD in 40 patients with suspected head and neck tumors (35). The typical red fluorescence of HpD was observed in 29 patients with biopsy proven malignancy, furthermore, in 5 patients the HpD fluorescence was used to aid detection of the lesions and the choice of biopsy site. Of the remaining cases three benign tumors and four inflammatory lesions exhibited no fluorescence, but two out of four biopsies of "normal tissue" were fluorescent. This finding was ascribed to the presence of lymphoid tissue in the biopsy specimen. It was, therefore, concluded that HpD fluorescence was of no value in predicting tumor margins or in detecting malignancies covered by an intact mucosa. The authors considered that the affinity of HpD for lymphoid structures would lead to considerable confusion in the diagnosis of head and neck malignancies.

Despite this problem, Lipson and coworkers (36) continued to produce encouraging results using HpD fluorescence in the detection of early cancers and premalignant changes in the cervix. However, several cases of squamous metaplasia, chronic cervicitis and one case with no obvious histological abnormality also exhibited fluorescence.

Some years later coworkers at the Mayo Clinic assessed HpD in the localization of early-stage lung cancer (37,38). Successful detection of carcinoma in situ was described using a system in which violet light from a filtered mercury lamp was alternated at high frequency with white light, allowing both tumor fluorescence and near normal visualization during endoscopic examinations.