Is Bacillus subtilis Objectionable in a Non-Sterile Oral Liquid Solution?

Introduction

The U.S. Federal Good Manufacturing Practices (GMP) Regulations, i.e., 21 CFR 211.113 Control of Microbiological Contamination require pharmaceutical manufacturers to exclude objectionable microorganisms from their non-sterile pharmaceutical drug products. What microorganisms that may be considered objectionable can be determined from the clinical literature, microbial infection outbreak investigations especially associated with drugs and medical devices, and to a lesser degree U.S. drug product recalls. In the absence of regulatory guidance, this issue was the subject of the 2014 PDA Technical Report No. 67 Exclusion of Objectionable Microorganisms from Non-sterile Pharmaceutical and OTC Drug Products, Medical Devices and Cosmetics. The PDA Technical Report concluded that what are objectionable microorganisms will depend on the pharmaceutical dosage form, the formulation, physicochemical and microbiological attributes of the individual drug product and its intended use. A more recent book chapter by the author updated this material (Cundell, 2020).

Based on a survey of U.S. product recalls of non-sterile products published by Sutton and Jimenez (2012), there are around 15-20 recalls annually. Perhaps surprisingly to some, the major reason for the recalls is the presence of objectionable microorganisms in these products (72%), not exceeding the recommended microbial limit.

During a 7-year period from 2004-2011, 144 non-sterile drug products were subject to voluntary recall. The recalls by product types were over-the-counter drug products (42%), cosmetics and soaps (31%), medical devices (14%), dietary supplements and probiotics (8%), and pharmaceuticals (5%). Because of the strict adherence to cGMPs and greater financial and technical resources, it may not be unexpected that prescription pharmaceuticals had the least number of recalls amongst these product types. Of the 144 recalls, the frequency of implication of different microorganisms was Burkholderia cepacia (34 recalls, 24% of all recalls), unspecified fungi (19 recalls, 13%), Bacillus cereus (9 recalls, 6%), Pseudomonas aeruginosa (6 recalls, 4%), Elizabethkingia meningoseptica (5 recalls, 4%), Pseudomonas putida (3 recalls, 2%), Pseudomonas spp. (2 recalls, 2%), and Salmonella spp. (1 recall, 1%).

The role of Bacillus cereus in minor, self-limiting intestinal disturbances may be a reason for the recalls for this bacterium. According to the FDA Bad Bug Book, the presence of large numbers of B. cereus (greater than 106 organisms/g) in a food is indicative of active growth and proliferation of the organism is consistent with a potential human health hazard due to production of bacterial enterotoxins. These numbers are far in excess of the Total Aerobic Microbial Count of not more than 100 cfu/g recommended for an oral liquid.

The author’s interest was provoked by the December 18, 2019 voluntary nationwide recall of the anti-epilepsy drug Levetiracetam Oral Solution, 100 mg/mL due to microbial contamination by the manufacturer Lannett Company, Inc., Philadelphia, PA. The recall notice published on the FDA website (www.fda.gov/recalls) stated that the product was contaminated because the Gram-positive, spore-forming bacterium Bacillus subtilis was found in one of the pharmaceutical ingredients used to manufacture the product.

This article will explore whether this common environmental bacterium, if found with counts less than 100 cfu/mL in this oral solution, should be considered objectionable. The author believes that the position taken by the manufacturer is questionable and would be unlikely to be taken after a comprehensive microbial contamination risk assessment.

The risk statement associated with the product recall stated: “Bacillus subtilis is ubiquitous in the environment and although the pathogenic potential has been described as low, serious systemic infections have been reported. The likelihood of the health hazard depends on the degree of microbial contamination, the dose and duration of treatment, and the patient’s underlying conditions. It is possible that a severe infection may occur in immunecompromised patients. Lannett has not received any reports of adverse events related to this recall to date”.

The author believes that the product formulation, route of administration and the dosage, i.e., an oral liquid administered twice daily using a medicine dropper or cup is not associated with systemic infections and the target patient populations, i.e., epileptics, whom are not likely to be immunecompromised, mitigates the risk.

Product Formulation

According to the package insert, Levetiracetam Oral Solution, 100 mg/mL, is a clear, colorless, grape-flavored liquid, and supplied in 473 mL (16 fl . oz.) HDPE bottles. It contains 100 mg of levetiracetam per mL. Inactive ingredients: acesulfame, potassium, citric acid, flavor, glycerin, methyl paraben, propyl paraben, purified water, sodium citrate, and sorbitol solution. The function and microbiological attributes of the ingredients are summarized in Table 1.

Key physicochemical and microbiological parameters that will affect the ability of Bacillus subtilis spores to germinate and grow in the product at the expense of the ingredients, include the pH and water activity of the oral liquid and the efficacy of the preservative system against B. subtilis. Although the bacterium can metabolize sorbitol, the low pH (4-5), and reduced water activity (0.7-0.8) will prevent bacterial growth, and the methyl paraben/propyl paraben preservative system will be active against any vegetative B. subtilis formed after spore germination (Handbook of Pharmaceutical Excipients).

Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special offers
from American Pharmaceutical Review – all delivered right to your inbox! Sign up now!

Based on the 2001 NSF/ANSI Standard for Non-potentially Hazardous Foods, potentially hazardous foods do not include those with a pH of 4.6 or less and a water activity (a_w) of 0.85 or less. Note that the minimum pH and a_w requirements for growth of the related B. cereus are 4.9 and 0.93 respectively. Even more pertinent, the same document states that the maximum a_w for the so-called no time/temperature control for safety of food containing spores but not vegetative cells is 0.92. A recent publication from the laboratory of internationally-renown expert on bacterial spores Peter Setlow (Rao et al, 2017) demonstrated that humectants that lowered the a_w ≤ 0.90, prevented the germination of B. subtilis, B. megaterium, and B. cereus spores by inhibiting the activation of the spore cortex peptidoglycan hydrolysis and/or dipicolinic acid release. Water activity will exert microbiological control by two mechanisms, namely preventing the spore germination and inhibiting growth of germinated spores. In addition, B. subtilis will not grow in an oral solution with a pH less than 4.9 and if the pH is higher and growth occurs the vegetative cells will be killed by the paraben preservative system. Based on the hurdle technology concept, this product will not support the growth of microorganisms and vegetative microorganisms will not survive in the products (Leistner, 1994).

What would we expect to happen to the ingested B. subtilis spores? Bacillus species are a minor component of the gut microbiota in slightly higher numbers than may be expected than suggested from the ingestion of plant materials. Spores will survive the acidic gastric juices of the stomach and bile salts of the upper intestine, may germinate within the intestine, and pass out in fecal material as vegetative cells or spores. B. subtilis, being a strict aerobe, would not be expected to grow well in the anoxic conditions of the lower intestine (Hong et al, 2005).

Although the FDA Warning Letter does not disclose the identity of the pharmaceutical ingredient that contained B. subtilis, it was assumed by the author that it was Sorbitol Solution, 70%, USP. The current USP monograph does not have a microbiological requirement, but the monograph for the companion Sorbitol Powder does. The latter has a specification of the total aerobic count using the Plate Method is NMT 1000 cfu/g, and the total combined molds and yeasts count is NMT 100 cfu/g. The author has addressed the microbiological attributes of pharmaceutical excipients in an earlier publication (Cundell, 2005).

According to the Handbook of Pharmaceutical Excipients, Sorbitol is most likely to be manufactured by high-pressure hydrogenation with a nickel catalyst of high fructose corn syrup. The water activity of 70% Sorbitol is around 0.8 so it will not support spore germination and bacterial growth, but the bacterial endospores that enter the product post-synthesis may persist in this material and be transferred into the oral solution. Given the broad industrial use of 70% Sorbitol it would be unreasonable to prohibit the use of this ingredient by the pharmaceutical industry because it contained spores from a ubiquitous environmental bacterium like B. subtilis.

What Members of the Genus Bacillus are Objectionable in a Non-sterile Drug Product?

Obviously the more invasive major bacterial pathogen B. anthracis would be objectionable in any non-sterile drug product, but it is highly unlikely that a drug product manufactured under GMP conditions would contain this bacterium. Although it is the enterotoxins and not the actual organisms that produce the intestinal disturbances, a case may be made for designing B. cereus as an objectionable microorganism depending on the dosage form, formulation, and intended use of a non-sterile drug product. Table 2 provides some of the characteristics of six common Bacillus species.

Bacillus subtilis in Fermented Food and Dietary Supplements

The safety of ingesting B. subtilis spores becomes more obvious from a review of fermented foods and probiotics. Poly-λ-glutamic acid synthesis-positive strains of B. subtilis, are used as starter cultures for the production of Natto, a popular fermented soybean food in the Japanese diet (Kubo et al, 2011). Natto is reported to contain spore counts in excess of 108 cfu/g. Probiotics containing billions of B. subtilis spores are marketed in Eastern Europe, Asia and South America (Hong et al, 2005). Comprehensive evaluations have demonstrated the safety of these probiotic products in terms of toxicity and virulence (Sanders et al, 2003; Hong et al, 2008; Jezewska-Frackowiak et al, 2018). Although the FDA has not approved as a policy probiotic strains as Generally Recognized As Safe (GRAS), the Center For Food Safety and Applied Nutrition has not disagreed with the safety evidence found in probiotic manufacturers submissions.

Conclusions

The author has observed that some pharmaceutical manufacturers take the expediency, on receipt of adverse FDA Form 483 observations, of recalling batches of product for containing objectionable microorganisms to demonstrate a due diligence of caution, as part of a strategy to avoid a follow-up Warning Letter. This is often done without making a comprehensive risk assessment as to whether microorganisms found in a non-sterile drug product are truly objectionable. Without the full details, this article presents evidence that B. subtilis, if they survive the manufacturing process, and are found in numbers less than 100 cfu/mL in the product would not be objectionable. The reduced water activity, low pH, and robust preservative system will prevent the growth of spore-forming bacteria in the oral solution. B. subtilis is not considered a frank pathogen and as a strict aerobe will not proliferate in the human intestine. These factors mitigate the risk of bacterial infection in all patient populations.

Furthermore, recalling a non-sterile drug product because B. subtilis spores were found in a pharmaceutical ingredient is too conservative an action, especially as many of these materials are industrial commodities. A more nuanced approach is needed.

References

1. Cundell. A. M. 2005 Managing the microbiological quality of pharmaceutical excipients. PDA J. Pharm. Sci & Technol. 59 (6): 381-395
2. Cundell, T. 2020 Chapter 11 Exclusion of objectionable microorganisms from non-sterile pharmaceutical drug products. In Pharmaceutical Microbiological Quality Assurance and Control – Practical Guide for Non-sterile Manufacturing D. Roesti and M. Goverde (editors) John Wiley & Sons, Hoboken, New Jersey pp371-400
3. Hong, H. A., L.H. Duc and S. M. Cutting 2005 The use of bacterial formers as probiotics FEMS Microbiol. Rev. 29: 813-835
4. Hong, H.A., J.M. Huang et al, 2008. The safety of Bacillus subtilis and Bacillus indicans as food probiotics J. Appl. Microbiol. 105: 510-520
5. Jezewska-Frackowiak, J., K. Seropczynska et al, 2018. The promises and risks of probiotic Bacillus species ABP Biochica Polonica 65(4): 509-519
6. Kubo, Y., A. P. Rooney et al 2011 Phylogenetic analysis of Bacillus subtilis strains applicable to Natto (Fermented Soybean) Production Appl. Environ. Microbiol. 77(18):
7. Leistner, L 1994 Further developments in the utilization of hurdle technology for food preservation J. Food Eng, 22: 421-432
8. Rao, L., F.E. Feeherry et al, 2017 Eff ects of lowering water activity by various humectants on germination of spores of Bacillus species with different germinants Food Microbiology 72:112-117
9. Sanders, M.E., L. Mortelli and T.A. Tompkins 2003 Spore formers as probiotics: Bacillus, Sporolactobacillus and Brevibacillus. Compr. Rev. Food Sci. Food safety 2:101-110
10. Sutton, S. and L. Jimenez 2012 A review of reported recalls involving microbiological control 2004-2011. Amer. Pharm. Rev. 15 (1): 42-52