Abstract
Of more than 150 000 published studies evaluating new biomarkers, fewer than 100 biomarkers have been implemented for patient care. One reason for this is lack of rigorous testing by the medical community to validate claims for biomarker clinical relevance, and potential reluctance to publish negative results when confirmation is not obtained. Here we sought to determine the utility and reproducibility of immunohistochemical detection of hepatocyte growth factor (HGF) in melanoma tissue, an approach of potential assistance in defining patients with innate resistance to BRAF inhibitor therapy. To this end, a published and a revised method that retained sensitivity but with greater specificity for HGF detection, were evaluated in cells known to endogenously express HGF, and in models where HGF is upregulated via cytokine induction and via overexpression by gene transfection. Consequent patient evaluation in collaboration with the Melanoma Institute Australia of a cohort of 41 melanoma specimens with extensive clinical annotation failed to validate HGF immunohistochemistry as a predictor of response to BRAF inhibitors. Targeted therapies for advanced melanoma and other cancers show great promise, and rigorous validation studies are thus indicated for approaches that seek to personalize such therapies to maximize therapeutic efficacy.