Performance Characteristics Of Automated LAL Tests 

By Tim Sandle, Ph.D, M.A., BSc (Hons), CBiol, MSBiol., MIScT

Introduction

The use of automated Limulus amebocyte lysate (LAL) methods to perform Bacterial Endotoxin Testing (BET) is now commonplace in most pharmaceutical laboratories. By automated I am referring to the use of software driven heated tube and plate readers for turbidimetric or chromogenic LAL testing for laboratories testing against Ph. Eur. 2.6.14 or USP <85>.

These ‘semi' - automated methods have a number of advantages over the gel-clot method in both testing terms, such as easier demonstration of inhibition / enhancement testing, and in aiding cGMP through the various reports that can be obtained. The Gel-clot test is still widely used and, by virtue of its relative simplicity, remains a robust method. However, laboratories facing ever more regulation are more often employing quantitative methods.

Testing using the automated methods requires validation, which any reputable laboratory demonstrates, in simple terms, by testing three different lots / batches at a sample within the maximum valid dilution and demonstrate freedom from interfering substances. The automated instruments themselves will have been calibrated by the supplier for factors like uniformity of temperature.

Not all laboratories, however, have examined these instruments for their performance characteristics in the way that a biochemist would look at the equipment used for conducting more classical assays. Yes – for some microbiologists it's news that the LAL test is an assay. Due to the fractions of endotoxin (in terms of weight of lipopolysaccharide) that the test detects it cannot fall within the tight scope that the biochemist would apply to an assay, for aspects like accuracy or precision or other parameter pulled out of the ICH guidelines.

This doesn't mean that microbiologists can ignore the performance of their instruments. This article sets out some of the performance characteristics which could be considered for automated LAL instruments. I am not suggesting that these all need to be looked at or that already busy microbiology lab have to grapple with yet another validation task. However, the microbiologist responsible for such systems should at least understand their strengths and limitations. Types of automated LAL instruments Automated LAL instruments fall into two broad categories depending upon the nature of the test: turbidimetric (where the rate of turbidity (or the ‘gelation') of the lysate is proportional to the amount of endotoxin present) and chromogenic (where a synthetic ‘yellow' chromogenic substrate is used in the clotting cascade and the intensity of this yellow is related to the amount of endotoxin). Turbidimetric tests are read using photometric instruments (such as spectrophotometers) and chromogenic tests are more often read using ELISA plate readers. This article doesn't attempted to weigh up the pros or cons of competing instrumentation (or step into the cut throat world of competing LAL reagent suppliers) except to acknowledge that different types of software affect the test and different manufacturer's lysates differ considerably, not least in the sensitivity to LAL Reactive Materials, such as, glucans.

Applying performance characteristics

I am probably taking it as read that some performance characteristics should be applied to automated LAL tests. Understanding the abilities of the instrumentation is important in examining and interpreting test results and in defending those test results against misunderstanding from Production and QA departments. Some characteristics will be included in annual or six-monthly calibrations of the instrument, such as, temperature distribution, and others are shown through demonstration of the suitability of the instrument over time, such as the results from the negative control samples (the ‘blanks' to assayist).

What assay characteristics can be applied to automated LAL testing so that the performance of the instruments and test can be judged? In this article the following will be examined:

Limit of Detection
Limit of Quantitation
Reproducibility
Repeatability
Precision
Accuracy
Robustness

I would suggest that these factors need to be performed, examined and filed as part of the instrument's equipment validation.

Tim Sandle, Ph.D, M.A., BSc (Hons), CBiol, MSBiol., MIScT – Dr. Sandle is the Head of Microbiology at the UK Bio Products Laboratory. Dr. Sandle is a chartered biologist and holds a first class honors degree in Applied Biology; a Masters degree in education; and obtained his doctorate from Keele University. His role involves overseeing a range of microbiological tests, batch review, microbiological investigation and policy development. In addition, he is an honorary consultant with the School of Pharmacy and Pharmaceutical Sciences, University of Manchester and is a tutor for the university's pharmaceutical microbiology M.Sc course. Dr. Sandle serves on several national and international committees relating to pharmaceutical microbiology and cleanroom contamination control (including the ISO cleanroom standards). He is currently chairman of the PharMIG LAL action group and serves on the NBS cleaning and disinfection committee. He has written over eighty book chapters, peer reviewed papers and technical articles relating to microbiology. He is currently the editor of the Pharmaceutical Microbiology Interest Group Journal and runs an on-line microbiology forum (www.pharmig.blogspot.com). Dr. Sandle is an experienced auditor and frequently acts as a consultant to the pharmaceutical and healthcare sectors.

Posted by Tim Sandle