Early detection of premalignant conditions in the colon by fluorescence endoscopy using local sensitization with hexaminolevulinate

• Introduction
• Patients and methods
• Results
• Discussion
• Conclusion
• References

Background and study aims: We aimed to determine the feasibility of obtaining selective fluorescence of precancerous/cancerous lesions in the colon with a new fluorescence video endoscope system in combination with the selective photosensitizer precursor hexaminolevulinate (HAL), and to carry out a dose-finding study with evaluation of the optimal dose and application time.

Patients and methods: 12 patients with colorectal lesions underwent sensitization with locally applied HAL enemas in two concentrations (0.8 mmol and 1.6 mmol). The examination was conducted either 30 or 60 minutes after rectal administration of the sensitizer, using a special light source capable of delivering either white or blue excitation light. Red fluorescence induced by illumination with blue light was detected via a prototype fluorescence video colonoscope. Biopsies were taken from suspicious areas found with white or blue light. Corresponding endoscopic, fluorescence, and microscopic findings were compared.

Results: Using histological findings as the gold standard, 52/53 of the premalignant/malignant lesions showed red fluorescence under the photodynamic diagnosis (PDD) examination; 38/53 were detected with white-light endoscopy. The PDD mode showed 28 % more polyps than did white-light endoscopic imaging. The greatest fluorescence intensity in precancerous lesions was found with retention for 60 minutes of 500 ml of 1.6 mmol HAL.

Conclusions: Administration of HAL enema induces selective lesion fluorescence and increases the lesion detection rate in patients with colorectal adenoma and early carcinoma.

Introduction

Carcinoma of the lower gastrointestinal tract is among the leading malignancies in the western world, and early detection and removal of adenomas plays a crucial role in surveillance strategies. Colonoscopy requires extensive training and about 10 % of all adenomas larger than 10 mm in size are missed at standard white-light colonoscopy [1]. Consequently, several attempts have been made to enhance the diagnostic yield of colonoscopy, by means of chromoendoscopy using either methylene blue [2] or indigo carmine [3], but also by using fluorescence techniques in conjunction with light of selected wavelengths [4] [5].

Photosensitizers may be used in the diagnosis or treatment of malignant conditions: in gastrointestinal endoscopy, photodynamic diagnosis (PDD) using local or systemic administration of 5-aminolevulinic acid (5-ALA) has been described for the detection of premalignant/malignant conditions of the upper [6] [7] [8] and the lower gastrointestinal tract [9] [10]. Derivatives of 5-ALA, such as hexaminolevulinate (HAL) are more selective and effective [11], and have an even lower rate of phototoxic side effects [12] [13]. However, investigations in this field have been hampered due to the lack of a flexible fluorescence video endoscopy system.

In the present study, we investigated the performance, in patients with suspected colorectal dysplastic lesions, of a new fluorescence video colonoscope with white and blue light functionality, used in combination with the selective HAL photosensitizer (Hexvix; PhotoCure ASA, Oslo, Norway) administered locally as an enema with different concentrations and application times. We aimed to determine the feasibility of obtaining selective fluorescence of cancerous and precancerous colonic lesions with this new procedure. A second aim was to carry out a dose-finding study for HAL, to evaluate the optimal dose and application time for local sensitization for PDD.

Patients and methods

A total of 12 patients (seven men, five women) with known or highly suspected colonic adenoma or carcinoma participated in this clinical pilot study. The study protocol was approved by the Institutional Review Board of the University of Erlangen-Nürnberg and the national authority for drugs and medical devices (BfArm), and complied with good clinical practice.

Patients older than 18 years with suspicion of colorectal adenoma or carcinoma were asked to participate in the study, and written informed consent was obtained from all of them. Exclusion criteria for this study were: liver cirrhosis of Child stage B or C; acute or chronic hepatitis; elevated liver test values of unclear etiology, with elevation of transaminases to more than three times normal levels; known porphyria; pregnancy; or expected lack of compliance.

Patient history, a complete blood count, and biochemical levels were obtained before and 24 hours after administration of HAL. Vital signs including blood pressure and heart rate were monitored at baseline and every 4 - 6 hours during the first 24 hours. Potential skin photosensitivity was tested by comparing skin erythema upon blue-light illumination with a UVA lamp (UVA A/B Phototherapy System UV120-2; TheraLight Europe, Fischen am Ammersee, Germany) at baseline and 2 hours after the examination. Adverse events, including gastrointestinal symptoms reported by patients, were recorded.

For the HAL enema, 200 mg HAL HCl powder (PhotoCure ASA, Oslo, Norway) was dissolved in 1000 ml sterile phosphate buffered saline (0.8 mmol) or in 500 ml (1.6 mmol). After bowel cleansing with a polyethelene glycol-based solution, the HAL enema was administered immediately with a plastic tube (Trimline Long, rectal catheter; E-Z-EM Inc., Westbury, New York, USA) that had an inflatable blocking balloon at the top to prevent leakage of the enema. The enema was instilled over 30 minutes by turning the patient to the left side, their back, and their right side, and was retained for 30 (n = 6) or 60 minutes (n = 6) prior to endoscopy.

The endoscopic examination of the colon was done with a prototype flexible fluorescence video endoscope system. The equipment consisted of a PDD video colonoscope (13 902 PIKS; Karl Storz, Tuttlingen, Germany) with a long-pass observation filter of 450 nm integrated in the tip of the endoscope and a short-arc xenon lamp (300 watt, D-light C; Karl Storz) provided with a special switchable bandpass filter (wavelengths 380 - 430 nm). Switching between blue excitation light and white light was done by means of a head button on the proximal end of the video endoscope. Fluorescence images in the red spectral range at wavelengths above 600 nm were detected via the video colonoscope.

To investigate the diagnostic feasibility of PDD and potential improvement in diagnostic sensitivity for adenomas, the primary colon inspection was done using white light followed by blue excitation light, in sections from the cecum to the rectum. All suspected lesions identified using blue and/or white light were biopsied or resected to obtain tissue material for histological examination. Also mucosa that was non-suspicious under blue and white light in the colon transversum and rectum was routinely biopsied. All examinations were recorded. Corresponding endoscopic, fluorescence, and microscopic findings were compared, with histology being used as the gold standard.

The histopathological examinations were carried out by the pathology department of the Munich-Pasing Hospital.

For different HAL concentrations (0.8 mmol versus 1.6 mmol) and enema retention times (30 minutes versus 60 minutes), the fluorescence intensity of each lesion was rated on a scale of 1 to 5, where 1 represented no fluorescence and 5 brilliant fluorescence. The average fluorescence rating was calculated for each patient, and the mean fluorescence intensity ratings in the different dose-finding groups were compared with one another to find the optimal concentration and enema retention time.

Results

A total of 12 patients were examined in this study, starting with three patients with 1.6 mmol HAL concentration and 30 minutes’ retention time, followed by three patients with 0.8 mmol HAL and 30 minutes of retention of the enema, then three patients with 0.8 mmol HAL and 60 minutes, and finally three with 1.6 mmol HAL and 60 minutes. The patient data with the mean fluorescence intensity rating for each one are shown in [Table 1].

In the 12 colonoscopies that were performed, a total of 64 polyps were detected (52 dysplastic adenomas, 1 colon cancer and 11 hyperplastic or inflammatory polyps) with sizes of 1 - 50 mm (mean 8 mm, median 5 mm). Of these, 26 polyps were located in the right colon (that is, between the cecum and the right flexure).

Areas showing red fluorescence in the blue-light mode were documented as fluorescence-positive, and non-fluorescing areas as fluorescence-negative. All areas macroscopically manifesting features suspicious for tumor, such as a conspicuously irregular mucosal surface or polypoid mucosal elevations, were classified as white-light-positive. All biopsies revealing dysplastic or carcinoma cells at the histological work-up were classified as histology-positive.

In the specimen-based analysis of the 64 polyps evaluated it was found that, among the 53 histology-positive biopsies, there were 52 that were true-positive, i. e. with red fluorescence in the blue-light mode, and there were 38 true positives in the white-light examination. There was one false-negative polyp with the PDD procedure and 15 false-negative results with white light.

[Fig. 1] shows the fluorescence endoscopy appearance of an adenomatous polyp in white-light mode and the corresponding appearance in blue-light mode.

At PDD there were 10 false-positive results, of which six were in the group with HAL concentration of 1.6 mmol HAL and four in the group with a concentration of 0.8 mmol HAL. In white-light mode two false-positive results were seen, one of which was a true negative at the corresponding PDD examination. False-positive findings in blue-light mode related to inflammatory tissue and in white-light mode they were related to hyperplastic, inflamed mucosa.

The adenomatous polyps that were false negatives in white-light mode (n = 15) were obtained from nine patients, and could not be seen with the standard illumination, either because of adverse location behind a fold, in the flexures, or under the valve, or because of a very flat contour (see for example [Fig. 2]). Eight of these polyps were found in the right colon. These non-visible adenomas ranged in diameter from 1 to 10 mm, with a mean of 3.7 mm. [Fig. 3] shows the appearance of a nonadenomatous polypoid lesion in the colon with white light and with blue light.

Overall, with HAL-induced fluorescence endoscopy we saw 28.3 % more polyps than with white-light endoscopy The preliminary statistical data are shown in [Table 2], with the restrictions due to the small number of patients.

We saw no phototoxicity of the skin, gastrointestinal symptoms, or cardiovascular problems related to the administration of 200 mg HAL as a 0.8 mmol or 1.6 mmol enema. A slight isolated elevation of the liver enzyme aspartate aminotransferase (AST) of maximally one-third of baseline level occurred in two patients who had a 1.6 mmol HAL enema. In one further patient in the 1.6 mmol HAL group there was an isolated elevation of bilirubin from 1.4 mg/dl to 2.2 mg/dl (normal value < 1.0 mg/dl). No further systemic or local adverse reactions were reported.

Overall, administration of 500 ml of 1.6 mmol HAL for 60 minutes revealed the best fluorescence intensity rating for precancerous lesions, with a mean fluorescence intensity rating of 3.3, followed by 0.8 mmol HAL and 60 minutes’ retention time with a mean fluorescence intensity of 2.9 (see [Table 1]).

Discussion

Until now, only one clinical study has been performed to investigate the pharmacokinetics of HAL-induced tumor fluorescence for optimization of PDD in the gastrointestinal tract. A phase I study by Endlicher et al. included 10 patients with known rectal adenoma or cancer who received a 250-ml rectal clyster of HAL 3.2 mmol [12]. The fluorescence was variable in these patients. The best selectivity for adenomas was obtained with 60 minutes’ retention time before endoscopy. Carcinomas (n = 5) exhibited red fluorescence only at tumor margins, but fluorescence microscopy revealed selective fluorescence in the epithelium.

In our study, we demonstrate for the first time that fluorescence video endoscopy with HAL applied as an enema is suitable for diagnosis of neoplastic lesions not only in the rectum, but throughout the colon. Compared with white-light endoscopy, we missed only one small polyp. This PDD-negative, 3 mm sessile adenoma was located in the cecum. A flat polyp of 10 mm at the same location, but in the opposite wall of the cecal mucosa showed fluorescence, and we presume that the small false-negative polyp in the cecum did not come into contact with the HAL solution (1.6 mmol and 500 ml fluid). No adenomas were missed in any other cases, either in the left or - more importantly - in the right colon.

However, the improvement in diagnostic sensitivity for dysplastic lesions of the colon using PDD is associated with a slight decrease in specificity. The fluorescence mode was associated with 10 false-positive findings caused by an increased accumulation of HAL in inflamed or regenerative mucosa.

In addition, a prototype of a new fluorescence video endoscope, that was adjusted to the emission wavelength of HAL, was evaluated for the first time.

In practice, particularly, flat adenomas and dysplastic lesions in hidden places such as in the flexures, behind folds, and under the valve can be detected with this procedure. Additionally the PDD mode might allow a visual judgment on whether adenomatous lesions have been removed completely with the snare or biopsy forceps.

Conclusion

Administration of HAL enema induces selective fluorescence and increases the lesion detection rate in patients with colorectal adenoma and early carcinoma. Further investigations with larger groups of patients are needed to corroborate the present data.

Competing interests: None

References

• 1 Pickhardt P J, Nugent P A, Mysliwiec P A. et al . Location of adenomas missed by optical colonoscopy.  Ann Intern Med. 2004;  141 352-359
  CrossrefPubMedSearch in Google Scholar
• 2 Kiesslich R, Moenk S, Reinhardt K. et al . Methylene blue-aided chromoendoscopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative colitis.  Gastroenterology. 2003;  124 880-888
  CrossrefPubMedSearch in Google Scholar
• 3 Hurlstone D P, Cross S S, Slater R. et al . Detecting diminutive colorectal lesions at colonoscopy: a randomised controlled trial of pan-colonic versus targeted chromoscopy.  Gut. 2004;  53 376-380
  CrossrefPubMedSearch in Google Scholar
• 4 Messmann H, Endlicher E, Freunek G. et al . Fluorescence endoscopy for the detection of low and high grade dysplasia in ulcerative colitis using systemic or local 5-aminolaevulinic acid sensitisation.  Gut. 2003;  52 1003-1007
  CrossrefPubMedSearch in Google Scholar
• 5 Mayinger B, Guenther K, Horner P. et al . Diagnosis of rectal cancer and adenoma with light-induced autofluorescence spectroscopy.  J Photochem Photobiol B. 2003;  70 13-20
  CrossrefPubMedSearch in Google Scholar
• 6 Mayinger B, Reh H, Hochberger J, Hahn E G. Endoscopic photodynamic diagnosis: oral aminolevulinic acid is a marker of GI cancer and dysplastic lesions.  Gastrointest Endosc. 1999;  50 242-246
  CrossrefPubMedSearch in Google Scholar
• 7 Mayinger B, Neidhardt S, Reh H. et al . Fluorescence induced with 5-aminolevulinic acid for the endoscopic detection and follow-up of esophageal lesions.  Gastrointest Endosc. 2001;  54 572-578
  CrossrefPubMedSearch in Google Scholar
• 8 Endlicher E, Knuechel R, Hauser T. et al . Endoscopic fluorescence detection of low and high grade dysplasia in Barrett’s oesophagus using systemic or local 5-aminolaevulinic acid sensitisation.  Gut. 2001;  48 314-319
  CrossrefPubMedSearch in Google Scholar
• 9 Messmann H, Endlicher E, Freunek G. et al . Fluorescence endoscopy for the detection of low and high grade dysplasia in ulcerative colitis using systemic or local 5-aminolaevulinic acid sensitisation.  Gut. 2003;  52 1003-1007
  CrossrefPubMedSearch in Google Scholar
• 10 Ochsenkuhn T, Tillack C, Stepp H. et al . Low frequency of colorectal dysplasia in patients with long-standing inflammatory bowel disease colitis; detection by fluorescence endoscopy.  Endoscopy. 2006;  38 477-482
  Thieme ConnectPubMedSearch in Google Scholar
• 11 Endlicher E, Rümmele P, Hausmann F. et al . Detection of dysplastic lesions by fluorescence in a model of chronic colitis in rats after local application of 5-aminolevulinic acid and its esterified derivatives.  Photochem Photobiol. 2004;  79 189-192
  CrossrefPubMedSearch in Google Scholar
• 12 Endlicher E, Gelbmann C M, Knüchel R. et al . Hexaminolevulinate-induced fluorescence endoscopy in patients with rectal adenoma and cancer: a pilot study.  Gastrointest Endosc. 2004;  60 449-454
  CrossrefPubMedSearch in Google Scholar
• 13 Klem B, Lappin G, Nicholson S. et al . Determination of the bioavailability of [¹⁴C]-hexaminolevulinate using accelerator mass spectrometry after intravesical administration in human volunteers.  J Clin Pharmacol. 2006;  46 456-460
  PubMedSearch in Google Scholar

1 Both authors contributed equally.

B. Mayinger, MD, PhD

Department of Medicine II,
Hospital Munich-Pasing

Steinerweg 5
81241 Munich
Germany

Fax: +49-89-88922411

Email: brigitte.mayinger@krankenhaus-pasing.de

References

• 1 Pickhardt P J, Nugent P A, Mysliwiec P A. et al . Location of adenomas missed by optical colonoscopy.  Ann Intern Med. 2004;  141 352-359
  CrossrefPubMedSearch in Google Scholar
• 2 Kiesslich R, Moenk S, Reinhardt K. et al . Methylene blue-aided chromoendoscopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative colitis.  Gastroenterology. 2003;  124 880-888
  CrossrefPubMedSearch in Google Scholar
• 3 Hurlstone D P, Cross S S, Slater R. et al . Detecting diminutive colorectal lesions at colonoscopy: a randomised controlled trial of pan-colonic versus targeted chromoscopy.  Gut. 2004;  53 376-380
  CrossrefPubMedSearch in Google Scholar
• 4 Messmann H, Endlicher E, Freunek G. et al . Fluorescence endoscopy for the detection of low and high grade dysplasia in ulcerative colitis using systemic or local 5-aminolaevulinic acid sensitisation.  Gut. 2003;  52 1003-1007
  CrossrefPubMedSearch in Google Scholar
• 5 Mayinger B, Guenther K, Horner P. et al . Diagnosis of rectal cancer and adenoma with light-induced autofluorescence spectroscopy.  J Photochem Photobiol B. 2003;  70 13-20
  CrossrefPubMedSearch in Google Scholar
• 6 Mayinger B, Reh H, Hochberger J, Hahn E G. Endoscopic photodynamic diagnosis: oral aminolevulinic acid is a marker of GI cancer and dysplastic lesions.  Gastrointest Endosc. 1999;  50 242-246
  CrossrefPubMedSearch in Google Scholar
• 7 Mayinger B, Neidhardt S, Reh H. et al . Fluorescence induced with 5-aminolevulinic acid for the endoscopic detection and follow-up of esophageal lesions.  Gastrointest Endosc. 2001;  54 572-578
  CrossrefPubMedSearch in Google Scholar
• 8 Endlicher E, Knuechel R, Hauser T. et al . Endoscopic fluorescence detection of low and high grade dysplasia in Barrett’s oesophagus using systemic or local 5-aminolaevulinic acid sensitisation.  Gut. 2001;  48 314-319
  CrossrefPubMedSearch in Google Scholar
• 9 Messmann H, Endlicher E, Freunek G. et al . Fluorescence endoscopy for the detection of low and high grade dysplasia in ulcerative colitis using systemic or local 5-aminolaevulinic acid sensitisation.  Gut. 2003;  52 1003-1007
  CrossrefPubMedSearch in Google Scholar
• 10 Ochsenkuhn T, Tillack C, Stepp H. et al . Low frequency of colorectal dysplasia in patients with long-standing inflammatory bowel disease colitis; detection by fluorescence endoscopy.  Endoscopy. 2006;  38 477-482
  Thieme ConnectPubMedSearch in Google Scholar
• 11 Endlicher E, Rümmele P, Hausmann F. et al . Detection of dysplastic lesions by fluorescence in a model of chronic colitis in rats after local application of 5-aminolevulinic acid and its esterified derivatives.  Photochem Photobiol. 2004;  79 189-192
  CrossrefPubMedSearch in Google Scholar
• 12 Endlicher E, Gelbmann C M, Knüchel R. et al . Hexaminolevulinate-induced fluorescence endoscopy in patients with rectal adenoma and cancer: a pilot study.  Gastrointest Endosc. 2004;  60 449-454
  CrossrefPubMedSearch in Google Scholar
• 13 Klem B, Lappin G, Nicholson S. et al . Determination of the bioavailability of [¹⁴C]-hexaminolevulinate using accelerator mass spectrometry after intravesical administration in human volunteers.  J Clin Pharmacol. 2006;  46 456-460
  PubMedSearch in Google Scholar

1 Both authors contributed equally.

B. Mayinger, MD, PhD

Department of Medicine II,
Hospital Munich-Pasing

Steinerweg 5
81241 Munich
Germany

Fax: +49-89-88922411

Email: brigitte.mayinger@krankenhaus-pasing.de