Negative Biopsy Specimens in the Setting of High A Priori Probability of Disease

Abstract

More and more, newly published management guidelines are touted as evidence driven with extensive data-backed assessments of risk of disease. These risks are determined by outcomes data generated in clinical studies and formal clinical trials. As any pathologist knows, the data derived from any study are only as good as the study inputs and methodology. For instance, if the goal is to identify the risk of disease, one needs to have optimal methods and tools in place to find evidence of the disease if it is present. In addition, and looking more carefully into the process, one needs to appreciate how reproducible the targeted diagnosis actually is and how the sampling of tissue can affect detection sensitivity. It is only when these parameters are appropriately considered that reality-based assessments of risk can be input into management guidelines. But it is not just guideline development that is important in this discussion. Individual patient results are equally affected by the process noted above. If you apply an extant management guideline to a patient, that application also relies on a sound pathology diagnosis. Hence, the use of an optimized interpretive technique must also apply equally in the setting of the primary diagnosis in order for the guidelines to actually work. Therefore, both crafting and clinical use of any guideline must use optimal—and similar—pathology procedures. Risk-based management guidelines for cervical cytology have been recently published that were crafted using extensive evidence-based literature reviews and expert opinion.1 These guidelines were based on the outcomes data grounded by extensive follow-up studies aimed at identifying the risk of high-grade cervical lesions over time for the various cytologic abnormalities. Based on risk levels observed in these studies, appropriate management could be defined. …