The study aim was to profile cytochrome P450 CYP1 family (including CYP1A1/1A2 and CYP1B1) protein expression during the malignant progression of primary melanoma and subsequent metastatic disease. Tissue microarrays of primary (n = 75) and metastatic melanoma (n=104) were constructed with ethical approval and the patient demographics were as follows: (1) Primary melanoma; age 22 to 93 (median 59); sex M/F 36/44; Breslow thickness 0.4 to 15 mm (median 2.5); ulceration 25/80, and (2) metastatic melanoma; age 26 to 92 (median 60); sex M/F 54/49; ulceration 30/103; number of nodes 1 to 15 (median 2); extra-capsular spread 20/95. CYP1A1/1A2 and CYP1B1 were detected immunohistochemically using fully validated and selective poly- and monoclonal antibodies. In order to facilitate the spectral resolution of CYP1 staining from melanin vector SG (grey) was used with nuclear fast red counterstain. CYP1 staining intensity was scored visually (negative 0, weak 1, moderate 2, and strong 3) in consultation with a pathologist (P.I.R). Spectral imaging microscopy (SIM) was also used to accurately quantify CYP1 staining intensity at every pixel of a captured image of each melanoma core. Reference spectra of the individual chromophores were used to spectrally un-mix CYP1 (grey), from the nuclear stain (red), and melanin (brown) before the mean normalised absorbance intensity of CYP1 staining in the entire core was determined. Melanoma specimens were graded according to the 2002 AJCC classification system: primary stage I n = 27 (1A 8, 1B 19), and stage II n = 48 (2A 22, 2B 16, 2C 10), lymph node metastasis stage III n = 98 (3B 53, 3C 45), visceral metastasis stage IV n = 6. Normal skin (n = 27), benign naevi (n = 14), and dysplastic naevi (n = 21) were also included in the study. CYP1B1 was not present in normal skin but was over-expressed in both primary (71%) and metastatic (65%) melanoma scored visually and at a higher frequency when quantified by SIM (91% and 83%, respectively). There was no significant difference in CYP1B1 between primary sub-types but primary melanoma (stage I & II) was significantly (p = 0.004) greater than metastasis (stage III & IV). CYP1B1 expression did not correlate with ulceration or Breslow thickness but did correlate with N stage lymph node metastasis. CYP1B1 protein expression in dysplastic naevi indicated up-regulation at an early stage of melanoma progression. In marked contrast CYP1A1/1A2 is not expressed in normal skin nor primary/metastatic melanoma. CYP1A1/1A2 but not CYP1B1 is expressed in normal liver but only the latter was over-expressed in liver metastasis. In conclusion, CYP1B1 protein expression is maintained with advancing AJCC stage from primary through to visceral metastasis. Future work will seek to correlate protein expression with functionality with a view to exploiting CYP1B1 in the enzyme/prodrug therapy of malignant melanoma.

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA