LER: Microbiology’s Hottest Urban Myth

Monday, November 30, 2015

Endotoxin Consultant

Endotoxins in gram negative bacteria (GNB) are remarkably complex biomolecules that share a common architecture. Endotoxin is only found in the outer layer of GNB. It exists as vesicles containing lipopolysaccharide (LPS) molecules embedded in surface proteins, lipoproteins and phospholipids. Lipopolysaccharide (LPS) is the biologically active component in endotoxin, but in its pure form, does not exist in nature. Should endotoxin gain access to body tissues, it induces dose-dependent biological effects from mild to life-threatening conditions. Concern about low lipopolysaccharide recovery (LLR), initially described inaccurately as low endotoxin recovery (LER), led to uncertainty about the bacterial endotoxin test (BET), new FDA expectations with regard to endotoxin testing of biologics, and a needless escalation of rabbit pyrogen testing (RPT). This report describes elements of the LER/LLR controversy and suggests a path forward to resolve this concern.

Overview

Over the past three years, investigations into the so-called LER phenomenon have revealed the nature of this controversial mystery. During endotoxin hold-time studies, LLR was reported in monoclonal antibody (MAb) formulae that contained a divalent-cation chelating buffer and a polysorbate, termed an LLR condition. Recent reports described how LPS was poorly recovered and lost pyrogenicity after incubation in LLR conditions. In contrast, laboratory-grown, naturally occurring endotoxin (NOE) was fully recovered by BET methods and was pyrogenic in rabbits1-5. The following information was gleaned from various investigators:

  1. LPS and endotoxin do not behave the same in all milieu;
  2. LLR occurs when a chelating buffer and polysorbate are excipients in a MAb formula;
  3. LLR is only observed upon attempts to recover LPS in undiluted drug product or substance;
  4. LLR occurs with and without the presence of the drug substance being tested;
  5. NOEs are recoverable and don’t need de-masking agents in so-called LER conditions;
  6. FDA expects the industry to devise a way to detect LPS in so-called LER conditions;
  7. FDA reviewers assert that LAL no longer protects the public from endotoxin;
  8. A well-designed study may distinguish LER, LLR and other types of BET inhibition;
  9. Data on the FDA website indicates that LER does not present a health hazard;
  10. LER, as originally described, apparently does not exist.

Endotoxin and Lipopolysaccharide are Not the Same

Scientists endeavored to study endotoxin more efficiently by isolating and purifying LPS from GNB cell wall. Six decades ago Westphal and Lüderitz furthered methods for extracting lipopolysaccharide (LPS) from various Gram-negative bacteria (GNB) and described the chemical nature of LPS6. Endotoxin is composed of a hydrophilic polysaccharide covalently linked to a highly conserved, hydrophobic lipid region. Harsh methods such as hot phenol/water extraction of GNB yielded potent, protein-free and soluble LPS. The availability of large quantities of pure LPS enabled critical studies of the nature and toxicity of LPS, including LAL (Limulus amebocyte lysate) discovery by Levin7. Early endotoxin investigators also reported that LPS activity varied due to the presence of various salts and detergents5.

Emerging LAL technology prompted the need for an endotoxin standard. As interest in LAL as a replacement for the RPT grew dramatically during the 1970s, LPS from E. coli 055.B5 was widely used for positive controls8. The FDA and USP contracted Rudbach circa 1976 to prepare reference material that was free of proteins and had average endotoxin activity9. A 30g batch of reference LPS that met these criteria was prepared from E. coli 0113:H10:K(-). All batches of EC (E. coli) reference standard contained this bulk material and fillers. The RSE (Endotoxin Reference Standard) is primarily used by LAL producers to determine the potency of LAL reagents and to standardize CSE (Control Standard Endotoxin) reagents used as positive controls in a USP <85> BET (Bacterial Endotoxins Test). The compendia for BET is silent on how a CSE might be calibrated for activity with RSE.

It is naive to assume that LPS and endotoxin are identical in behavior: one is natural, the other doesn’t exist in nature. The experience of the LAL industry is that LPS is a reliable source of positive-control material in a compendial BET, but is more unstable than native endotoxin. There are many anecdotal stories of unpurified endotoxin being used for recovery studies in the form of mixed endotoxin material, such as tap water, or laboratory-derived preparations. Bowers and Tran10 prepared endotoxin (NOE) from several GNB to assess the stability of endotoxin across various matrices, production processes and containers. Since CSE was not representative of endotoxin that might actually be present in a contamination event, preparation of high-concentration NOEs enabled Pfizer to add endotoxin to starting biopharmaceutical processes. A kinetic BET study confirmed the endotoxin activity at each time of use.

As reviewed by McCullough11, a positive aspect of this controversy is a greater appreciation among investigators and regulators of the fundamental differences in properties between endotoxin and LPS (RSE and CSE). Use of the terms endotoxin and LPS interchangeably must be done with great caution. The secret to understanding LLR is that the LPS in endotoxin is shielded in cell-wall vesicles and generally protected from chelators and detergents, whereas LPS forms micelles and other aggregate types that are unprotected from disaggregation factors. A review of potential LLR mechanisms support the notion that chelators and detergents act together to drive LPS toward a disaggregated state that is no longer LAL or RPT active5.

The LLR and LER Phenomena

A discussion of low endotoxin recovery at the PDA Annual Meeting in 2013 prompted concern about limitations of the BET12. The report stated that endotoxin (actually LPS) could not be recovered after being introduced into undiluted biologics containing chelating buffers and tween. Other reports more accurately described this phenomena as LLR, low LPS recovery1,3. Bolden and colleagues2 spiked approximately 18-EU/mL concentrations of laboratory-derived E. coli endotoxin and E. coli LPS into three different batches of a monoclonal antibody (MAb) that were formulated in a citrate/polysorbate-80 buffer. Samples were stored at 2-8° C for 10 days and checked periodically for recovery of activity. The LPS was not recovered after a few hours, whereas the endotoxin was fully recoverable over a 10-day period. Platco3 compared the recoveries of a laboratory-derived endotoxin and an LPS after inoculation (spiking) into two different undiluted MAb preparations that were formulated in a citrate/polysorbate buffer system. The LPS was not recoverable after 6 hours. The NOE, derived from Enterobacter cloacae, was fully recoverable at all intervals tested during a 14-day incubation period. Divalent cation replacement alone did not resolve recovery of LPS.

Dubczak1,4 extended the NOE and LPS comparison to include BET analyses and RPTs (rabbit pyrogen tests). The LPS was RSE (Endotoxin Reference Standard), the international reference standard for the BET. The NOE was Enterobacter cloacae. The NOE and RSE were inoculated into a normal saline control and an LLR formula containing a citrate/ polysorbate buffer. The rabbits received the same endotoxin-unit (EU) dose of the NOE and RSE. Both NOE and RSE in normal saline were fully recovered by kinetic chromogenic assays at the start of the incubation and at 24 hours; both were pyrogenic in rabbits at the same time periods. Results were different when the LLR solution was used. The NOE was fully active in BET and RPT at zero time and at 24 hours. However, RSE was BET and RPT active at zero time, but was nonreactive with the BET and non-pyrogenic at 24-hours.

Curiously, Genentech/Roche (GR) is the only manufacturer of MAbs containing LLR excipients that has reported a lack of success using NOEs for hold-time studies. In an informal survey, at least a score of similar MAb compounds produced by other parenteral manufacturers did not observe significant loss of LAL activity when using NOE. The FDA would benefit from an impartial study by experts in the industry regarding the use of NOE for these studies.

Distinguishing LLR from Other Causes of BET Inhibition

Interference with the BET is caused by conditions that either alter LAL efficiency or change the behavior of LPS, such as LLR13. The inhibition produced by a combination of divalent-cation-chelation and polysorbate detergent (an LLR condition) must be distinguished from inhibition caused by pH, chelating agents, cationic adsorption or other inhibitory conditions. A screening test may be applied to clarify the true nature of an inhibition event and demonstrate mitigation of LLR with an NOE. As described in Figure 1, a protocol to assess the recovery of LPS and NOE, in parallel, will distinguish LLR and LER in undiluted matrices. Drug products that contain LLR ingredients inhibit the recovery of LPS within a few hours, but generally allow recovery of an NOE for long periods after incubation. Drug products that do not meet these criteria require targeted study to determine the role of drug substance or excipients in poor recovery of LPS or endotoxin, such as a positively-charged (cationic adsorber) active ingredient. There may be isolated instances where the complex interaction of LLR agents and drug product produce a condition that is not amenable to recovery of NOE or LPS in undiluted drug.

 Figure 1. Decision tree for LLR assessment when conducting parallel recovery studies of lipopolysaccharide and native endotoxin in undiluted MAb matrices. Recovery of only endotoxin is consistent with LLR.

The experimental design for conducting recovery studies for LPS and NOE embodies critical components, such as choice and activity of NOE and LPS analytes, timing, nature of the standard curve, storage conditions and controls. Hughes14 reported that the varied protocols for NOE that were reviewed had produced variable results. Pyrogen tests with LPS were also highly variable. In the absence of a guidance document, it would be prudent for laboratory stakeholders to design a uniform recovery protocol and asses the recovery of both analytes with consistent methods.

LLR Presents No Health Hazard

No pyrogenic outbreaks have occurred in the LAL era due to inability to detect endotoxin with the BET. (Table 1) Further, quality by design and bioburden control eliminate pyrogens. The presence of drug product did not mask the detection of endotoxin in gentamicin associated with the pyrogenic outbreak of 1998-9915. Although gentamicin was administered to treat GNB infections, pyrogenic reactions were readily observed in 155 patients.

Table 1. Prominent Pyrogenic Outbreaks15

Bolden16 reported that the endotoxin in an MAb formulated in impure tap water was recovered over a 60-day period. The distrust of LAL over failure to detect LPS in LLR conditions pales in comparison with the scores of RPT and LAL comparisons conducted during the past four decades. Masking is an inaccurate synonym for LLR because it implies that the LPS will be expressed later; this notion is speculative in that there is no data to support delayed expression. The pyrogen test is highly variable, non-specific and at least one-to-two orders of magnitude less sensitive than LAL reagent for endotoxin detection. FDA’s database for adverse effects for biologics is free of pyrogenic reactions.

Changes at the FDA

The report that endotoxin could not be detected in bulk MAb product understandably placed FDA reviewers in an awkward regulatory position. To further complicate matters, the FDA has been overwhelmed with compelling misinformation that: 1) LAL reagent was an inadequate test for endotoxin, 2) there was an imminent health risk from masked pyrogens, and 3) NOEs were contrived and unsuitable for hold-time studies. In response, the FDA: 1) rejected NOE, the logical analyte for these studies; 2) required rabbit pyrogen testing for end-product release for fear that endotoxin was masked from LAL reagent; and, 3) expected the industry to discover a way to recover LPS controls or use new detection methods. Failure to resolve the LLR controversy may suppress development of MAb biosimilars.

The industry expects the FDA to take a fresh look at this controversy and consider relevant scientific data. The first corrective step is to terminate unwarranted pyrogen testing. The second is to allow firms to use suitable NOEs to mitigate LLR conditions and bring an end to the so-called LER controversy. The Agency could benefit from an industry-led study of parallel LPS and NOE recovery that would be conducted under the aegis of a harmonized protocol.

Summary

No one disputes that low recovery of LPS occurs in LLR conditions, such as chelating buffers and detergents. The so-called LER phenomena arose because recovery of LPS activity failed to mimic recovery of endotoxin activity in LLR formulations,. A uniform screening test can reveal LLR conditions and legitimize the use of NOE preparations for endotoxin challenge studies. The LER myth will fade away when FDA accepts NOE for endotoxin hold-time studies for an MAb and recognizes that such studies otherwise have no value added.

References

  1. Bolden J, Platco C, Dubczak J, Cooper J, McCullough KM. 41(5) Stimuli to the Revision Process: The Use of Endotoxin as an Analyte in Biopharmaceutical Product Hold-Time Studies. Pharmacopeial Forum, 2015.
  2. Bolden JS, Clarebout ME, Miner MK, et al. Evidence against a bacterial endotoxin masking effect in biologic drug products by Limulus amebocyte lysate detection. J Parenter Sci Technol. 2014;68(5):472-477.
  3. Platco C. Low lipopolysaccharide recovery versus low endotoxin recovery in common biological product matrices. American Pharmaceutical Review, 2014;17(Endotoxin Detection Suppl Part II):4-7.
  4. Dubczak J. A comparative in vitro and in vivo low endotoxin recovery (LER) assessment. PDA Annual Global Conference on Pharmaceutical Microbiology. Bethesda, MD, 2014.
  5. Tsuchiya M. Possible mechanism of low endotoxin recovery. American Pharmaceutical Review, 2014;17(7)1-5.
  6. Westphal O, Luderitz O. Uber die extraktion von bakterien mit phenol/wasser. A. Naturforsch. 1952 7b: 148-155. Levin J, Bang FB. The role of endotoxin in the extracellular coagulation of Limulus blood. Bull. Johns Hopkins Hosp. 1964, 115:265-274.
  7. Levin J, Bang FB. The role of endotoxin in the extracellular coagulation of Limulus blood. Bull. Johns Hopkins Hosp. 1964, 115:265-274.
  8. Cooper JF, Levin J, Wagner HN. Quantitative comparison of in vitro (Limulus) and in vivo (rabbit) methods for the detection of endotoxin. J. Lab. Clin. Med. 1971; 78:138-148.
  9. Rudbach JA, Akiya J, Elin RJ. Preparation and properties of a national reference standard endotoxin. J. Clin. Microbiol. 1976; 3:21-25.
  10. Bowers K, Lynn T. Creation of an in-house naturally occurring endotoxin preparation for use in endotoxin spiking studies and LAL sample hold time analysis. American Pharmaceutical Review. 2011;14(6):92-97.
  11. McCullough K. LER frequently asked questions. American Pharmaceutical Review. 2015, (Endotoxin Detection Supplement Part III):4-7.
  12. Chen J, Vinther A. Low endotoxin recovery (LER) in common biologics products. Parenteral Drug Association Annual Meeting, Orlando, 2013.
  13. Cooper JF. Resolving LAL test interferences. J. Parent. Sci. Tech. 1990, 44(1): 13-15.
  14. Hughes PF, Thomas C, Suvarna K. et al. Low endotoxin recovery: an FDA perspective. BioPharma Asia. 2015;4(2). www.biopharma-asia.com.
  15. Cooper, J.F.” Investigation of endotoxin out-of-specification and unexpected results,” in Microbial Risks and Investigations, eds. McCullough, K.Z. and Moldenhauer, J., 563-578. Bethesda, MD: PDA/DHI, 2015.
  16. Bolden, J. “Low LPS and endotoxin recovery: keeping an eye on the big picture. “ Presented at the PMF Bacterial Endotoxins Summit, Iselin, NJ, September 2015.

Author Biography

Beginning in the late 1960s, Dr. James F. Cooper pioneered development of the LAL test as a sensitive detector of endotoxin (pyrogen) in parenteral products. His publications span the history of LAL technology. In 1987, he founded Endosafe Inc., a producer of LAL reagent. He consults on endotoxin issues and is a retired Professor of Pharmaceutical Sciences at the Medical University of South Carolina College of Pharmacy, Charleston. He holds a Doctor of Pharmacy and a Bachelor of Sciences in Pharmacy from the University of Tennessee and a Master of Science from The Johns Hopkins University. He has held many offices in the American Pharmacy Association, Society of Nuclear Medicine, and Parenteral Drug Association, including the PDA Board of Directors (1983-85). He was presented the PDA’s James P. Agalloco Award for Excellence in Education in 2013. He served on the USP Council of Experts for Sterile Compounding (2005-2010) for Chapter <797>.

  • <<
  • >>

Join the Discussion