Assays reportable value

The accuracy of any quantitative assay depends on the use of standards that have been thoroughly characterized by accepted and independent methods. Without careful preparation of standards, the reported values for samples will be systematically higher or lower than the true value. Chiron has devoted considerable effort to the development of gold standard preparations of RNA from HIV-1 and HCV and DNA from HBV for use in the bDNA assays. These standards have been made available to the U.S. Food and Drug Administration and the World Health Organization. 

Another proposed in vitro assay for muscle irritancy for injectable formulations is the red blood cell hemolysis assay (Brown et al., 1989). Water-soluble formulations in a 1 2 ratio with freshly collected human blood are gently mixed for 5 min. The percentage of red blood cell survival is then determined by measuring differential absorbance at 540 nm this value is then compared to values for known irritants and nonirritants. Against a very small group of compounds (four), this assay reportedly accurately predicts muscle irritation. 

A Km value of 2.8 mM for phosphoramidate serving as a phosphoryl donor with untreated rat liver microsomes has been observed at pH 6.5 by Parvin and Smith (44), who also reported values of approximately 0.45 for the ratio of phosphoramidate-glucose phosphotransferase/PPi-glucose phosphotransferase with both rat liver and kidney microsomes. The latter assays were carried out at pH 6.5 with 0.4 M glucose and 20 mM phosphate substrates. 

Biomimetic artifical membrane-paracellular pathways-Renkin function The purpose of this study was to construct and examine the prediction model for total passive permeation through the intestinal membrane. The paracellular pathway prediction model based on Renkin function (PP-RF) was combined with a bio-mimetic artificial membrane permeation assay (BAMPA), which is an in vitro method to predict transcellular pathway permeation, to construct the prediction model (BAMPA-PP-RF model). The parameters of the BAMPA-PP-RF model, for example, apparent pore radius and potential drop of the paracellular pathway, were calculated from BAMPA permeability, the dissociation constant, the molecular radius, and the fraction of a dose absorbed in humans consisting of 80 structurally diverse compounds. The apparent pore radius and the apparent potential drop obtained in this study were 5.61-5.65 A and 75-86 mV, respectively, and these were in accordance with the previously reported values. The mean square root error of the BAMPA-PP-RF model was 13-14%. The BAMPA-PP-RF model was shown to be able to predict the total passive permeability more adequately than BAMPA alone. 

A validation protocol states the purpose of the validation method, the intended substance being tested, the definition of the reportable value, the validation approach, the specific directions for conducting the validation assays, the statistical approach for analyzing the resulting data, and the nonambig-uous acceptance criteria. The protocol should allow no room for ambiguity in the execution of the protocol or in the acceptance criteria. The content of a protocol is described next. There are many ways to format a protocol herein are suggestions for the content of a protocol. 

Other procedural details such as test lot preparation to aliquot all the samples needed for the validation experiments should also be described briefly in this section of the protocol.] Referring to SOP 123, the reportable value is defined as the average of the three replicates. For this validation, accuracy, precision, and linearity of the replicate values will be assessed consequently, the precision of the results reported is at the replicate level not the reportable value level. [Note this practice is done for two reasons first, to conserve resources, and second, when it is not possible to repeat an additional assay with exactly the same sample preparation. The repeatability component of the precision is defined as within-assay variance. Using the replicate values yields a within-assay variance.]... 

ICH Q2A defines repeatability as the variability of the assay results under the same operating conditions over a short interval of time. Repeatability is also termed intra-assay precision. A reportable value of an assay is often the average of a specified number of replicate values, where the replicates are processed using the same sample preparation. Thus, it is not possible to obtain a true repeat of the assay s reportable value. Repeatability can be modeled as the within-assay variability of the replicates however, it should be noted that this precision is at the level of the replicate and not at the level of the reportable value. If the reportable value is an average of K replicates the variance of the reportable values will be less than that of the replicates by a factor of l/K (the standard deviation will be less by a factor of (1/ k ). 

A method is linear or operating in the linear region if the measurement of the product parameter is proportional to the quantity being measured. Although there seems to be some debate regarding this issue, linearity does not refer to the linear range of a calibration curve used within the method or assay for calculation of the reportable value, but rather to the proportionality of the method s reportable value to the quantity being measured. Here too, it is important that the validation... 

Analysis of accuracy and precision can be accomplished using the same statistical model with a slightly different response variable. For accuracy, the response variable (here this is the replicate value and differs from the reportable value of the assay as run outside of the validation) is observed mass... 

When there is more than one observation (whether it be replicate values or more than one reportable value over a number of assays) at each level of the analyte, then a lack-of-fit analysis can be conducted. This analysis tests whether the average response at each level of the analyte is a significantly better model for average assay response than the linear model. A significant lack-of-fit can exist even with a high correlation coefficient (or high coefficient of determination) and the maximum deviation of response from the predicted value of the line should be assessed for practical significance. 

As the clinical usefulness of the microbiological assay for folate became more widely recognized so the number of variations and modifications to the method increased. As a consequence there was a wide range in the reported values found in normal subjects and a considerable variation in the level below which a patient was considered folate deficient. Some of these are shown in Table 1. Some workers have even suggested that folate levels in serum be graded as low, intermediate, or normal. In general it appears that serum folate levels below 3.0 xg/liter reflect a deficient state. 

De Waal et al. (204) reported the characterization of six isoforms of SSS isolated from C. roseus cell cultures. All isoforms were shown to be glycosylated. The pH optima found are broader (6-7.5) than previously reported. For all of the isoforms, very similar kinetics were found however, they differ considerably from the previously reported data (200). First of all, no tryptamine inhibition (up to 5 mM) could be observed. Measuring the reaction progress curve rather than using a stopped assay, the determined for all isoforms is around 8.2-9.4 fiM (average 8.9 /u.Af) the Vmax is 153-312 nkat/mg (average 234 nkat/mg). No substrate inhibition could be found, although strictosidine inhibits the enzyme (K-, 248-442 /iAf). Particularly, the values found are much lower than the previously reported values of 0.9-6.6 mM (200). 

The study by Reece and Turner (1937) included assays of more than 300 hypophyses. The reported values (summarized in Table XVI) show a tenfold higher concentration of prolactin in the gland of (jalves than in the pituitary of fetuses, this observation being in contrast to the findings... 

FIGURE 3.1 Understanding risk, precision, and accuracy coexistence on inferences made from reportable quality measurements. Illustrated is the distribution (bel-shaped curve) of measurements generated by an analytical method with a true accuracy of 100% (0% bias) and true precision of approximately 0.5% standard deviation. This is an example for a sample having a true average of 97% (1% below the specification of 98%), and only measurement error accounts for any deviation of reported values from 97%. The x axis represents an analyte assay value and the y axis represents the frequency of a corresponding test result. 

---