Introduction
Mycoplasmas (the trivial name for Mollicutes) comprise the class of broadly distributed, wall less free living bacteria with one of the smallest known genomes supporting bacterial self replication [1]. While many mycoplasmas are commensals colonizing a wide range of plant, insect, reptilian, avian, mammalian, and human hosts, a number of mycoplasma species are pathogenic and cause disease in their natural hosts. Mycoplasmas, particularly species of the genera Mycoplasma and Acholeplasma, are also frequent contaminants of vaccine substrates, i.e., continuous cell lines, and less frequently, animal-derived tissues and primary cell cultures [2]. This presents a serious concern regarding the risk of mycoplasma contamination for research laboratories and commercial facilities developing and manufacturing cell-derived biological and pharmaceutical products.
Even though mycoplasmas are generally considered to be host specific, there are reports that mycoplasmas may cross species barriers [3]. Sporadic mycoplasma infections of immunocompetent and immunocompromised persons that originated from domestic and wild animals have been reported [4, 5]. These findings raise a concern about potential susceptibility of some individuals, particularly children and people with congenital or acquired immunodeficiencies, to non-human infectious mycoplasma agents. Thus, the risk of accidental mycoplasma contamination of biological products, including egg-based live and inactivated vaccines, may be considered a public health issue. In addition to the concerns that a contaminating adventitious agent could be transmitted to a vaccine recipient, mycoplasmas interact with the host cells in culture and can alter many metabolic and biochemical features of the cells, such as cell function, morphology, chromosomal aberrations, decreased or increased viral yields, and cytokine production [2]. That is, mycoplasma contamination can affect virtually every parameter, function, and activity of cultured cells [1, 2].
To minimize the risk to individuals receiving biologics (biologicals) including vaccines, monitoring for adventitious agents such as mycoplasma is performed during the manufacture of biologics produced in cell culture substrates (See Regulatory Aspects, below).
Regulatory Aspects of Mycoplasma Testing for Biologics
In the 1950’s and 1960’s, many cell cultures and established cell lines were found to be contaminated with mycoplasmas, which raised concerns for the safety of vaccines produced in cells [2] and led to the requirement for mycoplasma testing of cell substrates. In 1962 the United States Public Health Service (USPHS) established a mycoplasma test requirement for viral vaccines produced in cell cultures (Title 21 of the Code of Federal Regulations (21 CFR found that did not grow in conventional mycoplasma broth and agar media, additional recommendations for a tissue culture (indicator cell) assay were included, i.e., “Points to Consider (PTC) in the Characterization of Cell Lines Used to Produce Biologicals” (initially published by FDA in 1987 and updated in1993). These recommendations applied not only to viral vaccines, but also to other licensed biologic products produced in cell substrates.
The recommendations were most recently incorporated into a draft Guidance from FDA/CBER: “Draft Guidance for Industry: Characterization and Qualification of Cell Substrates and Other Biological Starting Materials Used in the Production of Viral Vaccines for the Prevention and Treatment of Infectious Diseases (2006)” [6]. The 2006 cell substrate guidance also stated that PCRbased assays may be used to detect mycoplasma, provided that such an assay can be shown to be comparable to the agar and broth media procedure and the indicator cell culture procedure. 
The recommendations for live and inactivated vaccines manufactured using cell substrates include testing of the cell banks (master cell bank and working cell bank) of the cell substrates used for virus propagation, vaccine virus seed used for infecting the cell substrate culture, crude virus harvest, and control cells. The testing procedure recommended in the 2006 cell substrate guidance includes two test methods, cultivation in agar/broth media and indicator cell culture procedures. The agar/broth media procedure is aimed at the detection of “cultivable” mycoplasmal agents that grow in defined broth media under aerobic or anaerobic conditions. The indicator cell culture method, which includes the use of suitable mammalian cell culture for mycoplasma enrichment followed by staining of cells with DNA-binding fluorochrome (e.g. bisbenzimide or Hoechst stain [2]), is primarily used to detect fastidious “non-cultivable” mycoplasma species unable to replicate in the broth and agar media utilized for routine mycoplasma growth. Although the combination of these two methods enables highly efficient mycoplasma detection in cell substrates and cell-derived products, the overall testing procedures require skilled interpretation of results and are timeconsuming (a minimum 28 days).
The long time period required for the broth incubation and subculture to agar is an important limitation that does not permit the use of these culture methods for timely testing of the products with shelf-lives shorter than the turnaround testing time, routine in-process testing, and a rapid “go/no go” decision at the harvest step during product manufacture.
There are other important shortcomings related to the mycoplasma testing using these methods:
• Because both methods detect only viable mycoplasma, inadvertent inactivation of mycoplasmas in the test article during freezing, storage, or transport to certified testing laboratories can occur. This might result in false-negative results for the mycoplasma test.
• Because mycoplasmas demonstrate fastidious growth significantly dependent on nutritional factors of media and environmental conditions, the use of unstandardized media and the absence of accepted mycoplasma standards for calibration of mycoplasma methods might also affect the results of mycoplasma testing. Thus, development of mycoplasma reference standards with well-determined titers represents an important quality control issue to minimize the risk of false-negative results during mycoplasma testing. Recently, such mycoplasma standards were established for European Pharmacopoeia by European Directorate for the Quality of Medicine (EDQM) [7].
The application of alternative nucleic acid-based methods, particularly in combination with efficient sample preparation procedures, could provide defined advantages over conventional microbiological methods in terms of analytical sensitivity, simplicity, and turnaround time required for testing. However, it remains unclear whether PCR-based methods can provide a limit of detection comparable or superior to those of the culture methods. Furthermore, nucleic acid amplification technique (NAT) methods do not allow for accurate discrimination between viable and nonviable mycoplasma contaminants, which might lead to falsepositive results. 610.30)). When “non-cultivable” strains of mycoplasma were found that did not grow in conventional mycoplasma broth and agar media, additional recommendations for a tissue culture (indicator cell) assay were included, i.e., “Points to Consider (PTC) in the Characterization of Cell Lines Used to Produce Biologicals” (initially published by FDA in 1987 and updated in1993). These recommendations applied not only to viral vaccines, but also to other licensed biologic products produced in cell substrates. The recommendations were most recently incorporated into a draft Guidance from FDA/CBER: “Draft Guidance for Industry: Characterization and Qualification of Cell Substrates and Other Biological Starting Materials Used in the Production of Viral Vaccines for the Prevention and Treatment of Infectious Diseases (2006)” [6]. The 2006 cell substrate guidance also stated that PCRbased assays may be used to detect mycoplasma, provided that such an assay can be shown to be comparable to the agar and broth media procedure and the indicator cell culture procedure.
Use of Fowl Eggs as Substrates for the Production of Live and Inactivated Virus Vaccines
Growth of influenza viruses in embryonated hens’ eggs has been the predominant production method since the first inactivated influenza vaccines were produced [8]. In some cases specific (specified) pathogen-free (SPF) fowl eggs are used, for example, for production of a live attenuated influenza vaccine [9]. The use of embryonated hens’ eggs and primary chick embryo cell cultures for manufacture of biologics (e.g., egg-based vaccines) also poses a potential risk of mycoplasma contamination due to broad distribution and vertical and horizontal transmission of mycoplasma infection in chicken flocks. It is important to emphasis that currently there have been no clinical reports of mycoplasma-related adverse events associated with the administration of egg-derived viral vaccines. The absence of those mycoplasma-related adverse events may be attributed to the use of SPF chicken eggs, efficient mycoplasma testing methods during manufacture of certain egg-based vaccines, and inactivation procedures that are expected to eliminate any viable mycoplasma contaminants. Of note, the European Pharmacopeia prescribes the procedures to ensure the SPF status of a flock used for vaccine production [10]. The SPF status of the flock is maintained by means of systematic cultural and serological testing of the flocks for the absence of specified avian viral and bacterial infectious agents that may be horizontally and vertically transmissible [10].
Mycoplasma monitoring of the SPF flock and SPF eggs includes detection of the two most common avian mycoplasmal pathogens, M. synoviae and M. gallisepticum. However, it does not assure that the SPF chickens and their eggs are also free from other mycoplasmas that can be recovered from avian species (e.g., M. lipofaciens, M. gallinarum, M. gallinaceum, M. glycophilum, M. meleagridis, M. iowae and A. laidlawii); these mycoplasma species are also able to infect embryonated eggs and replicate in ovo. Because testing for these mycoplasma species is discretional, some risk of accidental contamination of SPF eggs with these species exists.
While the testing requirements and recommendations for mycoplasma are well described for US-licensed viral vaccines produced in cell substrates [6] and for other US licensed biological products produced in cell substrates (1993 PTC on Cell Substrates), similar recommendations have not been provided for viral vaccines produced in eggs that are licensed in the US, possibly because the first inactivated influenza vaccines were licensed in the US in the 1940’s [8], before there was an awareness of the risk of mycoplasmas in biologic products.
The term “cell substrate” generally refers to primary cells or tissues, diploid and continuous cell lines having a cell banking system [6]. The recommendations for eggs or for primary cell cultures derived from eggs or other animal tissues are not as clear, but the principles for monitoring for the absence of adventitious agents still apply. Vaccines are expected to be free of extraneous infectious microorganisms and potential oncogenic agents [6]. The regulations also require that cell lines used for manufacturing biological products shall be tested for the presence of detectable microbial agents (21 CFR 610.18(c)(1)(iv)). While mycoplasma testing of vaccines produced in avian eggs is not specifically prescribed in the CFR and guidances, FDA may require testing deemed necessary to assure safety of the product. For example, the purity regulation (21 CFR 610.13) states that “products shall be free of extraneous material except that which is unavoidable in the manufacturing process described in the approved biologics license application.” The recommendations for monitoring animal sources including embryonated avian eggs for a vaccine or related product are provided in the FDA Guidance for Industry: Content and Format of Chemistry, Manufacturing and Controls Information and Establishment Description Information for a Vaccine or Related Product [11], and the information should include the results of adventitious agent screening. Moreover, the European Pharmacopoeia 6.8 recommendations for manufacture of Influenza Vaccine (Split Virion, Inactivated) states that the virus is grown in the allantoic cavity of embryonated hens’ eggs from healthy flocks. The virus seed lot should be produced in embryonated eggs from SPF chicken flocks and the virus seed should be tested for mycoplasmas using compendial methods [12]. Thus, although mycoplasma testing of eggs is not specifically required by the CFR or recommended by FDA guidance, it is prudent that manufacturers assure that viable mycoplasma contaminants are not present in human vaccines, and appropriate measures for this assurance will need to be described in the product Biologics License Application.
Control of Mycoplasma and Other Potential Adventitious Agents in the Production of Inactivated Vaccines Produced in Embryonated Hens’ Eggs
Although conventional mycoplasma testing procedures could be used in principle for efficient mycoplasma detection during the production of egg-derived viral biologics, manufacturers historically have used a risk assessment approach to ensure the safety of viral vaccines. The approach includes validation of the inactivation of potential adventitious agents by the inactivation and purification steps used for the viral vaccine production. This safety concern arose from the recognition that avian leukosis virus (ALV) and other extraneous agents might be present in the flocks of chickens providing the eggs [8, 13]. FDA may request that the manufacturer of an inactivated vaccine produced in eggs validate the inactivation and/or removal of mycoplasmas and other microorganisms in the Biologics License Application prior to approval. A recent review article provides an overview of the chemical and physical procedures related to the mechanisms of viral inactivation [14]. The validation of the efficiency of mycoplasma inactivation during manufacture is currently included in the testing requested by FDA for approval of egg-derived viral vaccines, including inactivated influenza vaccine.
Evaluation of the efficiency of inactivation of mycoplasma contamination during inactivated viral vaccine production includes several considerations:
• Selection of appropriate testing methods that can adequately determine the viability status of mycoplasmas in analyzed samples.
• Use of concentrations of chemical agents corresponding to that of the manufacturing process (e.g., concentration, time of inactivation and temperature conditions).
• Selection of appropriate mycoplasma reference species/ strains, potentially including strains that exhibit some resistance to the tested chemicals.
• Assessment of the efficiency of mycoplasma inactivation using the tested chemicals.
The data obtained during evaluation of efficiency of mycoplasma inactivation at each vaccine manufacturing step that has the potential to inactivate (e.g., elevated temperatures, exposure to chemicals that are mycoplasmacidal) and/or remove mycoplasmas (e.g., ultracentrifugation, ultra-filtration, chromatographic purification) should be used to assess the risk of residual contamination in the final vaccine product.
Chemicals Used for Preparation of Inactivated Virus Vaccines
Several different chemical agents are used during the manufacture of inactivated viral vaccines, including inactivated influenza vaccines. As an example, influenza vaccines are inactivated with beta-propiolactone (BPL), formalin [8] or binary ethylenimine (BEI) via chemical reaction of these reagents with viral capsid proteins and nucleic acids [14]. Disruption of the viral envelope, which produces subvirion or split virus preparations and releases separate structural proteins, is achieved with detergents and surfactants such as cetyltrimethyl ammonium bromide (CTAB), Triton X-100, Triton N101, Tri N-butyl phosphate, and sodium deoxycholate (DOC) [8].
Concentration of the chemicals, inactivation conditions, and time of the chemical treatment vary greatly among manufacturers and different vaccines, and this topic is beyond of the aim of this article and we would like to refer readers to several excellent articles regarding the chemical inactivation for viral vaccines [14, 15].
Experimental Data on Mycoplasma Inactivation by Chemicals Used for Inactivated Virus Vaccine Manufacture
The validation test for mycoplasma inactivation/removal typically relies on the use of several mycoplasma reference species, including two mandatory avian mycoplasma species M. synoviae and M. gallisepticum, to determine the titer reduction of mycoplasmas spiked into matrices taken at the beginning of selected vaccine manufacturing steps, which are able to result in mycoplasma inactivation. CBER/FDA has conducted a study that was aimed at evaluating of the efficiency of mycoplasma inactivation under conditions that could be used for the production of inactivated virus vaccines, including inactivated influenza virus (split virion) vaccine (Wilson David, et al., Evaluation of Mycoplasmas Inactivation during Production of Biologics: Egg-Based Viral Vaccines as a Model. Applied and Environmental Microbiology 2010; in press). The contribution of several chemical reagents (BPL, formalin, CTAB, triton X-100, and DOC) to the mycoplasma inactivation process was tested in separate experiments using reagent concentrations commonly used or representing a “worst case scenario” for influenza vaccine production. Twenty-two different mycoplasma species, including avian, animal and human mycoplasmas, were experimentally tested to assess the mycoplasmacidal activity of the chemicals listed above.
The results obtained for BPL and formaldehyde demonstrated that all 22 tested mycoplasmas could be completely inactivated within 3-24 hours at room temperature (RT) in allantoic fluid at 0.2% (≥ 66.60 mM) formaldehyde and 0.1% (≥ 13.87 mM) BPL. The results of the study using formaldehyde and BPL were in concordance with earlier published data obtained for those reagents using only three mycoplasma species (M. gallisepticum, M. canis, and A. laidlawii) spiked into VERO and DK cell culture suspensions [16]. Similar to the findings reported in the Koski study, incomplete inactivation of many of the tested species was observed at formaldehyde concentration below 0.02%. A significant improvement in mycoplasma inactivation could be observed only when the formaldehyde concentration was increased to 0.2% [16]. Similar to the results obtained by other authors [16], our data also demonstrated the enhanced resistance of A. laidlawii (compared to all other tested mycoplasma species) during the first five hours of incubation with 0.2% formaldehyde. However, A. laidlawii was unable to survive after 24 hour incubation with this concentration of formaldehyde. A. laidlawii was also found to be the most resistant mycoplasma species to BPL. It was able to survive in the presence of 0.1% BPL for at least 2 hours, which correlated well with the stability of this species published previously [16].
In the majority of protocols used to produce inactivated virus vaccines [17, 18], the concentration of BPL and formaldehyde generally exceeds the critical value 0.1% -0.2% found to be sufficient for complete mycoplasma inactivation at the inactivation conditions (time and temperature) used by manufacturers of inactivated virus vaccines. Thus, the use of the inactivation conditions described above for vaccine production dramatically reduces the risk of survival of any accidental mycoplasma contamination of embryonated eggs used for vaccine virus growth. However, the situation can change if a low incubation temperature is utilized for virus inactivation using formaldehyde or BPL. We demonstrated that reduction of the temperature from RT to +4 °C resulted in a considerable increase in survival of all tested mycoplasmas at all formaldehyde and BPL concentrations including 0.2% and 0.1%, respectively.
Strong temperature dependence of mycoplasmacidal activity was also observed for one of the surfactants, i.e., cetyltrimethyl ammonium bromide. It is intriguing that the mycoplasmacidal activity of two other surfactants, triton X-100 and DOC, was found to be temperature independent in the range from +37°C to +4°C. This effect is likely to result from the difference in mechanisms used by these surfactants to disrupt the integrity of mycoplasma cellular membranes.
Among all tested chemicals, cetyltrimethyl ammonium bromide at a concentration higher than 0.08% was found to be most efficient mycoplasma inactivating agent. Complete inactivation of mycoplasma by CTAB was observed during the first 30 min regardless of the temperature used for the inactivation procedure. Our data showed that concentrations of CTAB from 0.075% to 0.15% would inactivate all of the mycoplasma test species during the first 30 min of exposure.
Other Important Aspects of the Evaluation of Mycoplasma Inactivation
As stated above, although mycoplasma testing of eggs is not specifically required by the CFR or recommended by FDA guidance, it is prudent that manufacturers assure the absence of viable mycoplasma contaminants in vaccine products, and appropriate measures for this assurance will need to be described in the product Biologics License Application. FDA guidances do not provide information about mycoplasma reference species (strains) that specifically should be used for the validation study. Generally, M. gallisepticum, M. synoviae, M. orale, M. pneumoniae, and A. laidlawii are used as reference mycoplasma species to ensure the inactivation of potential mycoplasma contamination of eggs. Of these species, M. synoviae and M. gallisepticum represent well known avian pathogens, M. pneumoniae is human pathogen, A. laidlawii is a widely distributed species in the environment, and M. orale is a member of the human commensal microflora and may represent a contaminant derived from personnel. Thus, the selection of these mycoplasma species for validation is primarily based on their medical and veterinary significance, as well as representing possible cultural and biochemical features of mycoplasmas that are important for the detection and identification of possible mycoplasma contaminants. The sensitivity of these species to the chemicals used during manufacture of inactivated virus vaccines was never previously assessed and compared with other mycoplasma species that potentially may cause accidental contamination of eggs or vaccine virus seeds. In our study, we evaluated survival rates of 22 different mycoplasmas in the presence of all five tested chemicals. The list of mycoplasmas tested included species isolated from human, animal and avian hosts. The analysis of inactivation profiles obtained for different species showed that mycoplasma species might differ substantially in their sensitivity to formaldehyde, BPL and CTAB, particularly when the chemicals were used at the lowest concentrations tested.
The detailed evaluation of the stability of five mycoplasmas, M. synoviae, M. gallisepticum, M. orale, M. pneumoniae, and A. laidlawii, currently suggested for validation of mycoplasma inactivation during virus vaccine manufacturing, showed that these species are well suited for validation purposes. Analysis of inactivation profiles obtained for tested mycoplasmas under different concentrations of chemicals showed that M. synoviae, M. gallisepticum, M. orale, M. pneumoniae, and A. laidlawii felt into a group of species demonstrating lesser sensitivity to chemicals. Thus, M. synoviae and A. laidlawii inactivation usually required longer incubation times compared to other mycoplasmas including M. gallisepticum, M. orale, and M. pneumonia, when lower concentrations of formaldehyde, beta-propiolactone and CTAB were used. It is noteworthy that this difference was observed only with formaldehyde, BPL, and CTAB at concentrations below 0.2%, 0.1% and 0.04%, respectively.
Thus, the use of the species M. synoviae, M. gallisepticum, M. orale, M. pneumoniae, and A. laidlawii represents a sufficient set of strains to perform adequate validation of mycoplasma inactivation by BPL, formaldehyde, CTAB, Triton X-100, and DOC commonly used during inactivated virus vaccine production, including egg-based influenza vaccine.
Acknowledgements
The authors thank Drs. Selwyn Wilson David, Rajesh Gupta, and Konstantin Chumakov for critical comments and help during the manuscript preparation and editing.
References
1. Razin S, Yogev D, Naot Y: Molecular biology and pathogenicity of mycoplasmas. Microbiol Mol Biol Rev 1998, 62(4):1094-1156.
2. Rottem S, Barile MF: Beware of mycoplasmas. Trends Biotechnol 1993, 11(4):143-151.
3. Pitcher DG, Nicholas RA: Mycoplasma host specificity: fact or fiction? Vet J 2005, 170(3):300-306.
4. Lierz M, Jansen A, Hafez HM: Avian Mycoplasma lipofaciens transmission to veterinarian. Emerg Infect Dis 2008, 14(7):1161-1163.
5. Madoff S, Pixley BQ, DelGiudice RA, Moellering RC, Jr.: Isolation of Mycoplasma bovis from a patient with systemic illness. J Clin Microbiol 1979, 9(6):709-711.
6. USA Food and Drug Administration: Draft Guidance for Industry: Characterization and Qualification of Cell Substrates and Other Biological Starting Materials Used in the Production of Viral Vaccines for the Prevention and Treatment of Infectious Diseases. Bethesda; 2006. http://www.fda.gov/BiologicsBloodVaccines/ GuidanceComplianceRegulatoryInformation/Guidances/ Vaccines/ucm074801.htm
7. Milne C, Daas A: Establishment of European pharmacopoeia Mycoplasma reference strains. Pharmeuropa Bio 2006, 2006(1):57-72.
8. Bridges C, Katz JM, Levandowski RA, Cox NJ: Inactivated influenza vaccines. In: Vaccines. Edited by Plotkin S, Orentstein W, Offit P, 5th edn: Elsevier; 2008: 266-269.
9. Belshe RB, Walker R, Stoddard JJ, Kemble G, Maassab HF, Mendelman PM: Influenza Vaccine (Live). In: Vaccines. Edited by Plotkin S, Orentstein W, Offit P: Elsevier; 2008: 296.
10. European Pharmacopoeia 6.6.: 5.2.2. Chicken flocks free from specified pathogens for the production and quality control of vaccines. 2010: 5137-5139.
11. USA Food and Drug Administration: Guidance for Industry: Content and Format of Chemistry, Manufacturing and Controls Information and Establishment Description Information for a Vaccine or Related Product. Bethesda; 1999. http://www.fda.gov/BiologicsBloodVaccines/ GuidanceComplianceRegulatoryInformation/Guidances/ Vaccines/ucm076612.htm
12. European Pharmacopoeia 6.0.: Influenza Vaccine (Split Virion, Inactivated). 2008: 801-803.
13. Levandowski RA: Regulatory perspective in the United States on cell cultures for production of inactivated influenza virus vaccines. Dev Biol Stand 1999, 98:171-175.
14. Stauffer F, El-Bacha T, Da Poian AT: Advances in the development of inactivated virus vaccines. Recent Pat Antiinfect Drug Discov 2006, 1(3):291-296.
15. Budowsky EI: Problems and prospects for preparation of killed antiviral vaccines. Adv Virus Res 1991, 39:255-290.
16. Koski TA, Christianson GG, Cole FL: Inactivation of mycoplasmas by use of phenol, formalin and betapropiolactone. J Biol Stand 1976, 4(2):151-154.
17. Budowsky EI, Friedman EA, Zheleznova NV: Principles of selective inactivation of viral genome. VII. Some peculiarities in determination of viral suspension infectivity during inactivation by chemical agents. Vaccine 1991, 9(7):473-476.
18. Budowsky EI, Friedman EA, Zheleznova NV, Noskov FS: Principles of selective inactivation of viral genome. VI. Inactivation of the infectivity of the influenza virus by the action of beta-propiolactone. Vaccine 1991, 9(6):398-402.
Dr. Dmitriy Volokhov holds a Visiting Associate position in the Laboratory of Methods Development at the Center for Biologics Evaluation and Research (CBER), US FDA. Dr. Volokhov’s areas of expertise include Virology, Microbiology, Infectious Diseases, and Molecular Biology. He is the author of more than 20 publications in microbiology in scientific peer-reviewed journals.
Dr. Vladimir Chizhikov holds a Principal Investigator position in the Laboratory of Methods Development at the Center for Biologics Evaluation and Research (CBER), FDA. Dr. Chizhikov’s areas of expertise include Virology, Microbiology, and Molecular Biology. He is the author of more than 80 publications in scientific peer-reviewed journals.
Dr. Donna K.F. Chandler is currently an independent consultant for vaccine development. Previously, she was Deputy Director, Divison of Vaccines and Releated Products Applications, Office of Vaccines Research and Review, Center for Biologics Evaluation and Research, Food and Drug Administration (1997-2006). She was a reviewer at FDA from 1989 to 1997, and researched pathogenicity of mycoplasmas and food-borne organisms at FDA from 1978-1989.
Disclaimer: “The findings and conclusions in this article have not been formally disseminated by the Food and Drug Administration and should not be construed to represent any Agency determination or policy.”