Summary The objective of this Phase II project is to improve outcomes and survival for patients with melanoma by reducing the number of false negative sentinel lymph node biopsies wherein metastatic disease is missed through conventional tissue processing and histological characterization. It is estimated that 4,200 patients are misdiagnosed in this manner in the US per year which results in more conservative treatment regimens for these patients than is required to effectively treat their disease. The outcome of patients not receiving the proper treatment regimen is that they experience a three-year reduction in ten-year survival which on a quality adjusted life year basis translates into a $1.2 billion economic loss. To address this problem, we have developed a three-dimensional tissue imaging and digital analysis approach which allows for the complete characterization of sentinel lymph node biopsy tissues and the identification of isolated tumor cells and micro- metastases that are commonly missed with traditional histopathology. We have demonstrated through our preliminary studies that using this approach we can identify these cancer cells in tissues that were previously characterized as node negative and that we can differentiate true negative from false negative samples. The focus of this project is to take this approach and transform it into a CLIA diagnostic assay that can be offered as a laboratory developed test to patients with negative sentinel lymph node biopsies at the conclusion of this Phase II work. Initially, the test will be offered following traditional tissue evaluation but eventually could be used as a primary diagnostic approach for all melanoma sentinel lymph node biopsy tissues. The development of this assay will be completed through conducting a retrospective clinical study in partnership with Cedars-Sinai with archived negative sentinel lymph node biopsy tissue blocks where the approach will be used to characterize these samples in their entirety. Through this study, we will demonstrate the accuracy, specificity and sensitivity of the test and will be able to quantify the extent to which it reduces the incidence of false negatives in the characterization of sentinel lymph node biopsies. Through the execution of this clinical study, we will build a statistical model that will threshold samples based upon their underlying three- dimensional features (e.g. total number of cancer cells, cancer cell aggregate volume, density of cancer cells) and classify them as ‘no metastases present’ (i.e. true negative) or ‘metastases present’ (i.e. true positive). Furthermore, the software we develop will for positive samples describe where a section should be taken from the sample for confirmation by a Visikol pathologist using traditional histopathology such that the report from the assay fits into the traditional classification paradigm. Lastly, we will transform our digital pathology software analysis approach into a 21 CFR part 11 c...