Quantitative HLA-G Expression in Metastasising and Non-Metastasising Primary Thin Cutaneous Melanomas

the original pathology report), Breslow thickness, physical site (head/neck, torso, lower limb, upper limb) and age. Individual patient data sets were coded to maintain confidentiality. Specimens were co-processed for standardisation. Specimens were deparaffinised using xylene, rehydrated through a graded series of ethanol and rinsed in phosphate-buffered saline. Sections were submitted to antigen retrieval at high temperature and pressure in 10 m M sodium citrate buffer (pH 6.0). Endogenous peroxidase was neutralised by incubation in 2% hydrogen peroxide. Nonspecific antibody binding was prevented through Dako protein block solution (Dako Canada Inc., Mississauga, Canada). Sections were incubated with primary monoclonal antibody against native and denatured HLA-G heavy chain (4H84, IgG1; gift of M. McMaster, University of California, San Francisco, Calif., USA). Omission of the primary antibody served as a negative control. Human placenta served as positive tissue control. Primary antibodies were revealed using the DakoCytomation streptavidin-biotin universal detection system according to the manufacturer’s instructions (Dako Canada Inc.). Sections were counterstained with Mayer’s haematoxylin and mounted. Digital images were acquired using the Nikon Coolpix 990 digital camera on a Nikon Eclipse 660 microscope (Nikon Canada, Mississauga, Canada). Quantification of staining intensity was carried out using the Imaging Processing and Analysis module of the Simple PCI Imaging System (Compix Inc., Cranberry Township, Pa., USA). Briefly, for each section all positively and negatively staining areas were delineated and the absolute light transmission of the positively staining areas measured at ! 20 magnification. All positive areas representing melanocytes were included, with care being taken to avoid inclusion of lymphocytes and other inflammatory cells. The transmission of each defined positive area was divided by the total area of positive staining to obtain an average transmission for each section. Average transmission is inversely related to staining intensity; 4–6 sections were processed for each lesion and the values averaged. All measurements were taken by 2 independent, blinded observers.