Negative reagent controls in diagnostic immunohistochemistry: do we need them? An evidence-based recommendation for laboratories throughout the world.

During the development and application of immunohistochemical techniques to diagnostic pathology, controls have played an important role in ensuring that the appropriate result has been obtained. The positive control is a tissue or collection of cells that contains the target protein (antigen) and confirms that the primary antibody and detection system have worked properly. The negative control, used primarily for primary antibody validation, is a specimen that does not contain the target protein and, as a result, should show no immunoreactivity. The negative reagent control (NRC) is an additional tissue section (frozen or paraffin) or cytology smear prepared from the patient specimen that is treated identically to that of the test slide(s) except for the omission of the primary antibody. Its primary purpose is to identify nonspecific reactivity due to factors or components other than the primary antibody. In the early days of diagnostic immunohistochemistry, the NRC was important because of the use of crude detection systems that often showed cross-reactivity with endogenous immunoglobulins (Fig. 1A) or with other proteins present in the test specimen that could be confused with specific immunoreactivity. In addition, with the introduction of avidin-biotin–based detection, nonspecific reactivity with endogenous biotin was frequently observed, especially following heat-induced epitope retrieval complicating interpretation (Fig. 1B). The introduction of sensitive and specific polymer-based detection systems has eliminated most of these issues.1 A critical examination of the use of NRCs in diagnostic immunohistochemistry today reveals several issues in play. First and foremost, they consume precious tissue or cells that could be used for additional testing. This is especially true today, when laboratories are seeing more fine-needle aspirate biopsy specimens and diagnostic needle cores that require ancillary immunohistochemical and/or molecular testing for targeted therapies for patients with cancer. Second, depending on the type of laboratory running the immunohistochemical tests (eg, general surgical pathology vs. a specialty laboratory such as urology, GI, or GYN), NRCs can represent as much as 50% of the slides in a given run (personal observation). This overwhelming volume can be a major burden on the staff and the instruments processing these slides. Finally, laboratories incur a major expense in terms of labor, supplies, and reagents for running these additional slides, for which there is no reimbursement. As a result, the routine use of NRCs in diagnostic immunohistochemistry has been challenged.2 In 2008, I made a decision to stop running NRCs routinely except for antibodies requiring avidin-biotin detection. My decision was based on 3 factors: (1) the