Environmental and Personnel Monitoring Programs—A Risk-Based Case Study of Cutibacterium acnes

The Microorganism Cutibacterium acnes

The Gram-positive aerotolerant anaerobic bacillus C. acnes is a pleomorphic nonmotile bacterium first described by Orla-Jensen in 1909 (8). A single bacterium is 0.4–0.5 μm wide and 0.8–0.9 μm long (9), and they are frequently microscopically observed in chains of two or more cells. As a member of the Actinobacteria, Proprionibacteriales, it was initially classified into the genus Corynebacterium in 1923 before transfer to the genus Propionibacterium in 1933 (9) and then inclusion into the new genus Cutibacterium in 2016 (10).

Although this lipophilic microorganism is usually regarded as a strict anaerobe, it can tolerate oxygen and grows at reduced rates (11). Cutibacterium acnes may survive as long as 8 months under anaerobic conditions and prefers low redox potential environments (12).

The application of early phenotypic methods, including serological agglutination, cell wall sugar analysis, bacteriophage, and fermentation profiling revealed three C. acnes clades termed type I, II, and III (13⇓⇓–16). Cutibacterium acnes type I clade can be subdivided into closely related phylotypes: IA1, IA2, IB, and IC; phylotype IA1 has been associated with acne, whereas types II and III are generally isolated in association with healthy skin (7).

Cutibacterium acnes possesses the potential to generate virulence factors that may assist in the adhesion and colonization of habitats on the human skin. These virulence factors include hyaluronate lyase (17), sialidases, and endo-glycoceramidases putatively involved in host tissue degradation (18).

Cutibacterium acnes hydrolyzes triglycerides in sebum and releases free fatty acids, including lauric and linoleic acid, as well as short-chain fatty acids (e.g., propionic acid) that contribute to an acidic skin pH and antimicrobial activity (19, 20). Cutibacterium acnes is also known to produce bacteriocin-like molecules that may be responsible for its successful colonization in the follicle and on the skin surface (21, 22).

The whole genome sequencing of C. acnes (strain DSM 16,379) describes 2,560,265 base pairs coding for a repertoire of genes that support impressive adaptive capabilities; including the capacity to grow under microaerobic as well as anaerobic conditions (18). The genes encode all of the components for the conservation of energy by oxidative phosphorylation, substrate-level phosphorylation via the Embden–Meyerhof pathway and pentose phosphate pathway, and several amino acid–degrading pathways. This array of metabolic pathways and catabolic options equip C. acnes with a significant degree of flexibility in surviving and/or proliferating in environments with changing oxygen concentrations. Surprisingly, several gene sequences demonstrate similarity to immune reactive proteins shared by Mycobacterium tuberculosis and Mycobacterium leprae (18); the authors not only question why, the general significance, but also what specific consequence, if any, this might have in our field.

Cutibacterium acnes and the Skin

Metagenomic analysis (23) of the human body has permitted the accurate identification and mapping of the diverse microbial communities and their interactions hitherto impossible with culture-based methodologies. Thus, we are now very much aware of our own inherent microflora, or more accurately our “microbiome” (24). The HMP (25) has illustrated the abundance of microorganisms that reside on and in our bodies and in particular the surface and subsurface layers of our skin. This in turn has illuminated two major concerns for the pharmaceutical microbiologist. Firstly, the human skin microbiome is extraordinarily large and diverse in microorganisms; their interactions representing complex abundant sources of microorganisms capable of contaminating controlled environments, processes, and products. One simple numeric illustration is that up to 10¹¹/m² bacteria inhabit the human skin (26). Secondly, most microorganisms that inhabit the skin are either in a quiescent state or unrecoverable by traditional culture media and conditions rendering this microbiological risk for the most part unseen by the standard means of monitoring environments, personnel, processes, or products. The complexity and dynamics of the human microbiome and the newly understood risks to pharmaceutical manufacturing and control are exemplified by C. acnes.

In general, we can divide the skin into three broad regional habitats relevant to our skin microbiome: (1) moist regions including the axilla, perineum, and toe webs; (2) oily or sebaceous areas including the head and neck; and (3) dry areas such as the forearms and legs. Whereas certain abundant species of microorganisms are recovered in one or two of the skin regions, Cutibacterium has been recovered from all areas of healthy individuals (27, 28). Although ubiquitous to all regions of the skin, C. acnes also appears to be the predominant species in oily, sebaceous skin areas (29). The skin's sebaceous glands manufacture a hydrophobic mixture of lipids (30) termed sebum and purposed to lubricate and protect the skin and hair. As such, the chemical constituents of the lipid rich sebum are regulated and altered depending upon several factors, such as environmental conditions (e.g., temperature) and age. Although sebum generally exerts antimicrobial activity, C. acnes hydrolyzes the triglycerides present in sebum, releasing free fatty acids that promote bacterial adherence, and then colonizes the sebaceous units (6). The hair follicle and sebaceous glands of the skin represent hospitable environments for C. acnes, which communicate using autocrine signal molecules to modulate biofilm generation along the lumen of the hair follicle and the hair follicle itself (31). A simplified illustration of a hair follicle and the location of C. acnes is provided in Figure 1. It is a valid assertion that cells of C. acnes are commonly physically “coated” with sebum and that they benefit from its protective properties. A propensity to generate biofilm appears linked to certain phylotypes of C. acnes (32) and therefore has recently been associated with the causation of acne vulgaris (7, 33). In the follicular environment, C. acnes exhibits niche competition with other microorganisms, such as Staphylococcus species, and is known to produce a thiopeptide antibiotic, cutimycin (34). In contrast to the classical assertion that C. acnes is the infectious agent causing acne vulgaris, contemporary science provides evidence that C. acnes is not only ubiquitous to all skin locations but is also equally abundant in both unaffected and acne-affected follicles (7). It could be feasible that the choice of the skin sampling techniques might bias the conclusion regarding the abundance of C. acnes on the skin surface versus follicular niches; however, Hall et al. (35) compared different techniques an concluded that there was negligible difference. Cutibacterium acnes is a commensal microorganism that, although associated with acne vulgaris, is a well-recognized opportunistic pathogen associated with implant surgery (36⇓⇓–39) yet conversely mutualistically assists in the health and homeostasis of the skin (40).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Generalized diagram illustrating the distribution of Cutibacterium acnes on the skin and within hair follicles.

Culturing Cutibacterium acnes

The majority of studies regarding C. acnes are in the clinical context of infection evaluation, investigation, and clinical infection control. Few studies exist regarding the isolation, recovery, and culturing of C. acnes in pharmaceutical manufacturing environments. Cutibacterium acnes is known to grow on a variety of media and under various incubation conditions; culture media it grows on include blood agar, chocolate agar, Brucella agar, and brain heart infusion broth, whereas C. acnes fails to grow on MacConkey agar with lactose (41). The ideal culture temperature of C. acnes is 30°C–37°C (42). The variety of media that permit the recovery and growth of C. acnes may reflect the metabolic diversity of C. acnes; however, extended incubation conditions of 10–13 days might be required for culture media that are suboptimal for C. acnes (41, 43). In clinical infections studies, the limitation of the incubation time to 7 days has resulted in the failure to associate and identify the presence of C. acnes (44). Other authors have also demonstrated a similar incubation period of 6–10 days can be adequate to recover C. acnes (45). A review of the literature illustrates the fastidious nature of C. acnes in preferring a complex growth medium rich in soluble growth factors derived from blood or animal tissue digests. Blood agar (46) or brain heart infusion (47, 48) agars are commonly the media of choice and provide the necessary growth factors. The absence of these growth factors would prevent recovery of isolates of C. acnes or at minimum result in retarded growth rates. Hyde (49) reported that the inclusion of Tween-80 in the media also assists the growth of C. acnes. Furthermore, C. acnes demonstrates a clear preference for anaerobic growth conditions and low redox potential environments. Recovery and growth under aerobic conditions is generally reported as slow (11, 50); however, supplementation of the environment with 5% (v/v) CO₂ is beneficial. Rieber et al. (43) demonstrated an increased speed of recovery of C. acnes with supplemented nutrient-rich media. The authors pointed out the importance of maintaining a continuous anaerobic environment during culturing, otherwise growth is impaired. Table I summarizes the source literature describing those culture conditions successfully used for culturing of C. acnes on agar plates. Details quantifying the precise numbers of colony forming units (CFUs) or recovery efficiency from the various media are not included. These details would be valuable in gauging the different abilities of the media to recover C. acnes; however, the diverse, varied, clinical contexts of the methods and sampling make this impossible. In general, we can conclude that successful consistent recovery of C. acnes on agar requires a complex medium and anaerobic incubation at 30°C–37°C for at least 7 days. Other authors reported the recovery and culturing of C. acnes in liquid broth such as thioglycollate broth (43, 57).

Adequacy of Industry Standard Monitoring and Sterility Testing For Cutibacterium acnes

It is important to recognize the inherent constraints of all environmental and personnel monitoring programs in that they are imperfect “snap shots” of microorganisms in the environment and on personnel. One of the greatest constraints of environmental and personnel monitoring programs is that they will only detect a small fraction of the total microorganisms present (58). Monitoring programs are neither quantitatively precise nor encompassing of all microorganisms. The cadence of sampling, sample location, technique, and choice of growth media and incubation conditions for the recovery of microorganisms are all factors of compromise chosen with a specific end purpose in mind. All environmental monitoring and personnel monitoring programs are purposed to assist in the assessment of the adequacy of the controls protecting the environment, process, and product from extraneous sources and vectors of microorganisms. An adequate environmental and personnel monitoring program should therefore be an early diagnostic of potential end product risk, incorporating the optimum choice of factors to fulfill that purpose. Effective environmental and personnel monitoring programs must therefore be more effective in recovering and identifying potential risk than the limited end product testing program is as a means of detecting contamination. No single program is capable of comprehensively demonstrating the complete array of microorganisms and the associated risks to environment, process, or product; however, all programs must be adequate in describing those risks. As such, the diagnostic effectiveness of environmental and personnel monitoring programs is primarily dependent upon the choice of growth media and the incubation conditions to adequately capture those risks by the recovery of appropriately representative microorganisms. In this respect, the adequacy of environmental monitoring and personnel monitoring programs relies upon the recovery of microorganisms that are indicative of this risk and the assumption that the recovery and the patterns of recovery for these microorganisms are a sufficient marker. All effort dedicated to optimized sampling frequency and location are all confounded if the choice of growth medium is inadequate for recovering microorganisms. As we understand far more clearly the actual types and populations of microorganisms in the clean room environment (59, 60) and upon personnel, the selection of the growth media to recover microorganisms that are adequately representative of overall microbiological risk must be subject to reassessment. To this end, almost all environmental and personnel monitoring programs use the industry standard media of trypticase soy agars for the recovery of bacteria and Sabouraud dextrose agar for the recovery of fungi. A cursory review of the literature providing data supporting the choice of the monitoring media and the incubation conditions is summarized in Table II. None of these studies evaluated by comparison the recovery of human commensals such as C. acnes by inclusion of alternative growth media and incubation conditions. A comparison of the industry-standard agar-based media and incubation conditions per Table II with the data in the prior section (see Table I) clearly shows that C. acnes would seldom be consistently recovered by the monitoring programs. Although inconsistent and irregular recovery of C. acnes has been reported on aerobically incubated TSA supplemented with Tween-80 (4–7 days incubation at 35°C–37°C), the authors postulate that a skin-associated growth factor may have aided recovery on the growth medium (67).

The absence of C. acnes from clean room microflora profiling assessments (68, 69) is likely not because of the physical absence of this organism but rather because of the inability of the growth medium to recover this microorganism. Also, some environmental monitoring technologies, such as active air samplers, may fail to deposit the microorganisms or may impart forces that damage the microorganisms and render them unculturable. Cutibacterium acnes has a moderate level of oxygen tolerance, superoxide dismutase and catalase activity, and a capacity to endure in aerobic environments for at least many hours (70⇓–72). It would seem that industry standard environmental and personnel monitoring programs have adopted TSA as the preferred growth medium and incubation conditions tailored to aerobic mesophilic microorganisms. In so doing, the programs could be erroneously designed in their failure to recognize that a more abundant microbiological risk to the process and the product may be present as microorganisms that would not fit in this category and could therefore fail to be recovered. Cutibacterium acnes would likely not be consistently recovered by this industry standard, and yet contemporary scientific literature suggests that this microorganism is one of the most, if not the most, prevalent skin-borne microorganism. It appears somewhat ironic that monitoring regimens specifically chosen with the rationale of recovering the microorganisms most likely present from human origin could be failing to do so by the exclusion of the most abundant because of the selection of the growth medium and the incubation conditions.

In stark contrast to the standard growth media and incubation conditions adopted by industry for environmental and personnel monitoring programs, the compendial sterility test incorporates a test primarily intended for the culture of anaerobic bacteria. The sterility test incorporates fluid thioglycollate media, known to support the growth of C. acnes (57), and an incubation duration of 14 days. L-cysteine and thioglycollate within this medium (as indicated by colorless resazurin) achieve a redox potential (Eh) below −110mV. Resazurin remains pink when the Eh is above −51 mV. The appropriately applied sterility test is therefore likely to recover the presence of viable cells of C. acnes within a test sample. The industry standard application of environmental and personnel monitoring (TSA culture medium incubated aerobically) together with end product sterility testing (including thioglycollate medium to recover anaerobes) may be fundamentally flawed and of ill design. As they stand, the identification of microbiological risk through environmental and personnel monitoring clearly fails the intended purpose of prospective early signaling of risk and patient safety and relies more upon the statistically and technically flawed sterility test (73).

FDA Form 483s concerning Cutibacterium acnes

Although FDA Form 483 observations frequently fail to include comprehensive details and an accurate context for the quality or compliance issue identified, they do assist in assessing purportedly significant microbiological risks. A review of recent FDA Form 483 observations reveals that C. acnes features in several quality and compliance issues spanning at least two decades and across a diverse number of pharmaceutical and biotechnology firms. The risk that C. acnes represents to aseptically manufactured product is neither limited by the type of sterile product manufactured nor by the manufacturing process. This organism possesses a number of unique features that confer a remarkable potential for product and patient risks. A summary of many (but not all) of those FDA Form 483 observations that specifically identify C. acnes by name as a sterility assurance and patient risk are provided in Table III.

Reviewing the details contained in each Form 483 in Table III, we may make several conclusions:

1. Cutibacterium acnes has been and continues to be recoverable in controlled clean room environments. Clean room environments that are in many ways managed and controlled according to current Good Manufacturing Practices. Controls include appropriate gowning and garbing and clean room sanitization practices. Recovery of this microorganism has been achieved by firms adopting growth media and incubation conditions similar to those discussed previously.

2. Cutibacterium acnes is not routinely and consistently recoverable on the typical industry standard culture medium of TSA and the associated aerobic incubation durations and temperatures (see previous). Nor have firms appropriately qualified their environmental and personnel monitoring programs to demonstrate the recovery of this aerotolerant anaerobe.

3. Sterility test fails associated with C. acnes are not rare nor extraordinary events but are commonly associated with the personnel and the intervention activities performed during manufacturing or sterility testing.

4. Firms continue to fail to associate C. acnes with significant sterility assurance risks and fail to accommodate appropriate monitoring genuinely capable of measuring or assessing this risk. It is quite possible that there exists a pervasive misconception that anaerobes cannot endure our aerobic environment, requiring anaerobic conditions to pose a risk, and do not represent viable contamination risks to sterile product.

In addition to the Form 483s summarized in Table III, the authors reviewed and identified several other related nonconformance reports that did not reference C. acnes by name but did reference the inadequate consideration of anaerobic conditions within the personnel and the environmental monitoring programs. It is highly likely that many firms beyond those listed here have and are, knowingly or unknowingly, experiencing patient risk from this microorganism.

Patient Risk from Cutibacterium acnes

During aseptic manufacturing processes, microorganisms represent a hazard for the sterile end product in which contamination with a single microorganism constitutes a loss of sterility, nonconforming product, and a potential for adverse patient impact. At a fundamental level, microorganisms originating from a specific source gain access into the manufacturing processor or the finished drug product by a transfer process.

Humans represent the major source of microbiological hazards to all aseptically manufactured products. Each individual human possesses approximately 10¹³ bacteria (26) with an average skin surface area of 1.5–2.0 m² (26 m² if the follicular surface area is included (74)) and hosts up to 10⁷ CFU/cm² (75) as determined using culture-based techniques. As previously mentioned, Cutibacterium and C. acnes may be more abundant than other microorganisms, represent a higher proportion of the population of different species, and are exceptionally stable members of the skin microbiome (76). Fully garbed personnel discharge 1–2 CFU/m³ air in clean rooms of 90 m³ volume and volumetric air change rate of 20/h (77, 78), shedding 0.22–0.38 CFU/s (79, 80). Therefore, the quantity of C. acnes entering the environment is likely significant. It should be noted, however, that these data for CFU shed from gowned personnel were derived using growth media (TSA) purposed to recover aerobic microorganisms (79). It is quite feasible that the actual microbiological risk and the actual microbiological shedding from personnel has been underestimated. Many specific steps of transfer are recognized; however, mechanistically, the transfer of microorganisms from a source to a vulnerable process or product may occur from either airborne transfer or touch contamination, the transfer from a contaminated surface to another surface (81). Recent data demonstrate that the surface-to-surface transfer efficiency is of a magnitude of approximately 15%–35% within the constraints of experimental parameters (82). Furthermore, residual skin bacteria left on objects can be matched to the individual who touched the object (83). Airborne transfer has been effectively modeled (84). The risk assessment of aseptic processes was well described in the seminal work of Whyte and Eaton (85, 86) and related frameworks (87). The application of these methodologies clearly illustrate the potential magnitude of microbial risk posed by C. acnes, and that C. acnes could remain a risk “unseen” by the industry standard environmental and personnel monitoring programs. An additional and important consideration of risk from this organism is that the “coating” of sebum over cells of C. acnes provides an additional protective barrier that is perhaps not so prevalent on other skin-borne microorganisms. This hydrophobic envelope affords physical and diffusional (88) protection to C. acnes that may extend to disinfectants and chemical sterilants.

In terms of direct patient impact, C. acnes is a significant causal agent of certain clinically relevant infections (89). As previously stated, C. acnes is the cause of patient morbidity and mortality as an opportunistic infectious agent, mostly in association with implant surgeries (36⇓⇓–39). Specific isolates of C. acnes are linked to disease (90, 91); furthermore, phylotypes IC and II have been specifically associated with opportunistic infections of deep tissue (92). The authors have found no corroborated reports of infection from parenteral products contaminated with this microorganism; however, there are many literature references concerning blood products, the infusion of blood products, the associated recovery of C. acnes, and potential patient risk. Care is warranted when evaluating many of these reports as the study methodologies adopted do not always consider the specific culture requirements needed for recovery of C. acnes. For example, Cunningham and Cash (93) evaluated the microbial contamination of 1000 platelet concentrates stored at 20°C and did not report the recovery of C. acnes. Closer examination of the methods illustrates that the duration of the incubation of the tests (96 h in total) was unlikely to recover C. acnes. Jacobs et al. (94) reported on the relationship between bacterial contamination and patient impact by surveying platelet infusions; C. acnes was not featured as a recovered contaminant. Unsurprisingly the recovery methodology, specifically the 48 h duration of the culture incubation, was insufficient to recover C. acnes. Current practice in the culturing of platelet concentrates reflects these culture requirements; generally, a maximum duration of 5–7 days using the rapid microbiological test technology BacT/Alert (95) confirms the occasional presence of C. acnes (96). A recent study (97) stated that 95% of contaminated platelet units contained C. acnes; this study employed anaerobic culture for 7 days incubation using the BacT/Alert rapid test system. There is unequivocal data that evidences that C. acnes is a recognized bacterial contaminant of platelet concentrates and is implicated in transfusion-transmitted bacterial infections (98, 99). Nevertheless, Brecher and Hay (99) are clear in stating “Although a few cases of Propionibacterium-contaminated units have been infused into patients, no long-term sequelae have been reported. Debate continues as to the value of an anaerobic culture in this context.”

A recent comprehensive NIH study evaluated several modern rapid sterility test technologies versus the compendial referee method using 118 isolates (77 bacteria, 8 yeasts, and 33 molds) representing 93 organisms from a NIH cGMP facility, 3 organisms from contaminated product, and 22 reference strains; C. acnes was not included (100). The absence of C. acnes is noteworthy and may indicate that this microorganism is not a concern in the NIH cGMP facilities and contaminated sterile products or simply that the monitoring and test conditions are not conducive to recovery. All these facts together regarding the clinical recovery and significance of C. acnes may indeed suggest the potential presence of this microorganism in aseptic manufacturing environments, which could result in unsterile product. However, the severity of the patient impact from infusion of the microorganism into the blood stream is questionable.

Conflict of Interest Declaration

The authors declare that they have no competing interests.