Screening false positives/negatives

A major consideration in screening is the detection capability of the screen for both false negatives (lack of detection of an active drug) and propensity to find false positives (detection of a response to the compound not due to therapeutic activity of interest). Ostensibly, false positives might not be considered a serious problem in that secondary testing will detect these and they do not normally interfere with the drug discovery process. 

The use of animal models for depression has two main objectives. One is to provide a behavioural model that can be used to screen potential antidepressant treatments. For this, the behaviour does not have to be an animal analogue of depression all that is needed is for it to be consistently prevented by established antidepressant agents (i.e. no false negatives) but not by drugs which have no antidepressant effect in humans (i.e. no false positives). 

Assessment and definition of sensitivity are often described for quantitative analysis but are of equal importance for qualitative devices of the dip-stick type that are very popular for farm- or field-based screening assays. Because of the somewhat subjective nature of visually assessed assays, the assay s sensitivity must be validated using a number of observers to determine at what level a test is deemed positive. The number of false positives and false negatives must be carefully determined in order to balance consumer safety and potential economic loss to animal producers. 

An assay or screen is said to exhibit ROBUSTNESS when it has a high discriminatory power and produces a low number of FALSE NEGATIVE and FALSE POSITIVE results. 

Table 4.3. Scheme of test results for screening procedures tp true positive,)p false positive, tn true negative, fn false negative, n total number of tests...

The most commonly used screening method for HIV is an enzyme-linked immunosorbent assay, which detects antibodies against HIV-1 and is both highly sensitive and specific. False positives can occur in multiparous women in recent recipients of hepatitis B, HIV, influenza, or rabies vaccine following multiple blood transfusions and in those with liver disease or renal failure, or undergoing chronic hemodialysis. False negatives may occur if the patient is newly infected and the test is performed before antibody production is adequate. The minimum time to develop antibodies is 3 to 4 weeks from initial exposure. 

Competition between components of a mixture becomes more likely as the hit rate and pool size increase. For a 10% hit rate, competing hits will be found in 26% of mixtures of ten compounds, but only 2.7% in mixtures of three [11]. Unless mixtures are deconvo-luted, competition can lead to false negatives when the competing compounds have different affinities. For screens using protein detection, competition can also produce false positives due to the additive effect of multiple weak binders. Competition can be controlled by limiting the size of mixtures and pooling dissimilar compounds to reduce the likelihood of two compounds binding at the same site. 

Alternatively, samples are quickly screened by thermal methods, such as DSC or ITC. This alternative approach eliminates the necessity for stability set-downs hence cycle times and sample consumption are reduced. However, the data obtained are difficult to interpret and may be misleading false positives and negatives are routinely encountered [14]. 

The 3T3 NRU test may easily be performed under GLP conditions in contract research organizations (CROs) and a high-throughput screening (HTS) method was recently reported in the literature. The HTS method produced no false positives, although some false negatives were observed, suggesting that the standard 3T3 NRU test protocol remains necessary for the final selection of nonphototoxic compounds [42]. 

---