April 15, 2016 (Vol. 36, No. 8)
John Dubczak General Manager Charles River Laboratories
Novel Approach to Achieving Consistent Results with the Limulus Amebocyte Lysate Test
The standard tool for screening endotoxins—highly toxic, heat-stable lipopolysaccharides that constitute the outer cell membrane of gram-negative bacteria—is the Limulus amebocyte lysate (LAL) test. It is so-named for a compound found in horseshoe crabs that reacts with exquisite sensitivity to the presence of gram-negative bacteria. In 1977, the Food and Drug Administration (FDA) specified the conditions for the use of the LAL test as an end-product test for endotoxins for human biological products and medical devices.
The use of the LAL test for human and animal injectable drug products was initially described in an FDA draft guideline. Drug companies were asked to submit data that compared their experience in the parallel testing of rabbits to that of LAL testing. This information as well as lessons learned from this experience was critical in the preparation of the final “Guideline on Validation of the Limulus Amebocyte Lysate Test as an End-Product Test for Human and Animal Parenteral Drugs, Biological Products, and Medical Devices,” which was issued by the agency in 1987.
The 1987 document described the procedures that the FDA considered necessary to demonstrate the absence of interferences (positive and negative controls) and procedures associated with routine testing of drug products. The Guideline did not, however, indicate that an endotoxin standard series had to be run with each set of tests if consistency of standard curves was demonstrated in the test laboratory. In essence, what the agency was referring to was an archived standard curve.
Too Much Variation
Unfortunately, that consistency has been extremely difficult to demonstrate using conventional methods. Conventional LAL reagents have always been freeze-dried in order to impart stability. LAL reagents are reconstituted and typically used during the course of a day. Once reconstituted, the activity of the LAL reagent will gradually change over time. That change in activity is especially seen in quantitative assays and, as a result, precludes the use of an archived standard curve.
One also finds variation with the standard used in each and every LAL test. The United States Pharmacopeia (USP) standard and the calibrated secondary standards provided by licensed LAL manufacturers are purified lipopolysaccharides (the biological active moiety of endotoxin). It is well known that the endotoxin standards used in LAL assays are profoundly influenced by the materials and techniques used by trained LAL analysts. The type of pipettes used, the type of dilution tubes used, the length of time used, the length of time used to vortex endotoxin standards, and the stability of the liquid stock endotoxin solution are all variables that laboratory personnel manage every day.
This variability highlights the need for a method that does not require endotoxin curves with every assay and that makes the reconstituted stability of the LAL reagent a nonissue. One such method that we developed is the Endosafe® Portable Test System™ (PTS), which utilizes a kinetic chromogenic technique and an archived standard curve. All of the reagents required for a compendial LAL assay have been applied to a plastic cartridge the size of a pack of chewing gum.
A handheld spectrophotometer warms the cartridge, draws the sample to three different mixing stations, and then measures a color within reaction wells (Figure 1). Because the reagents are pre-loaded and because the reconstitutions of each reagent is controlled to the millisecond, the LAL activity of a specific lot of cartridges can be measured and then used to develop an archived standard curve.
As a result, errors that operators seldom spot are eliminated from routine LAL testing.
Creating the Archived Standard Curve
To create an archived standard curve, representative samples of a manufactured lot of cartridges are submitted to a production laboratory for activity testing. There, the USP Reference Standard Endotoxin (RSE) is used to create three concentrations that bracket a 2-log range for the sensitivity of the cartridge lot. For example, one of the most common sensitivity ranges utilized in LAL laboratories encompasses 5.0, 0.5, and 0.05 EU/mL.
The reaction times for each of the three RSE standards are determined and then used to calculate the potency of the positive product control (PPC) that has been applied to two of the four cartridge channels. Separate negative control tests are conducted to ensure the absence of background contamination.
Once completed, a representative number of cartridges from the manufactured lot are presented to a separate quality control (QC) laboratory. The second QC lab will prepare a high, medium, and low concentration of RSE for a precision/accuracy test that challenges the previously developed archived standard curve and PPC spike.
Stability of the Archived Standard Curve
To demonstrate the utility of the archived standards, we conducted a stability study that examined the data prospectively and retrospectively. In this analysis, an Excel spreadsheet was developed. The spreadsheet contained all of the calculations necessary for the construction of a linear regression model. Regression models were then used to predict one variable from one or more other variables.
In this case, the initial (zero time) LAL reaction times were entered onto the regression model and the calculations fixed. The zero time LAL/PTS reaction times thus served as an archived standard curve. Reaction time data obtained from freshly prepared standards were also entered for each stability test interval from a specific batch of cartridges. All of the reaction time data were entered onto the spreadsheets and interpolated against the regression model.
Since the PPC is added to cartridges, the reactivity of the PPC is independent of the preparation of the endotoxin standards. And since each stability testing interval involved the preparation of a fresh standard curve (5.0, 0.5, and 0.05 EU/mL endotoxin standards), we were able to interpolate the reaction times of the PPC against an Excel regression model using the fresh standards for that interval and against the reaction times obtained for the archived standard curve.
So what did we learn? The interpolated (back-predicted) value (Figure 2) from each stability interval demonstrates the utility of the archived standard curve. The interpolated endotoxin concentrations, obtained from the reaction times of an endotoxin standard that was prepared at each stability interval, is more variable in the LAL test.
As can be seen in the graphs, the standards were weak at the six-month testing interval, and as a result they overestimated the PPC value. These data indicate that an archived standard curve will provide more consistent results on a day-to-day, week-to-week, or month-to-month basis.