2514 Anal. Chem. 1990, 62, 2514-2521 Near- Infrared Spectrometry of Microorganisms in Liquid Pharmaceuticals Leonard J. Galante and Michael A. Brinkley zyxwvuts Glaxo, Inc., Five Moore Drive, Research Triangle Park, North Carolina 27709 James K. Drennen and Robert A. Lodder* College zyxwvutsrq of Pharmacy, University of Kentucky Medical Center, Lexington, Kentucky 40536-0082 Biotechnology and phatmaceutkal research have created a number of new and potentlally life-saving drugs. Many of these hgs are formulated as Injectable proclucts. Some drug products zyxwvutsr do not w i v e autoclaving or other means of termkral sterllizatlon. An aseptlc flHlng process Is typically used to sterke such products, but it is less reliable than autoclaving, maklng detectlon of unsterHe unHs even more essential. In- vaslve mlcrobldoglcal methods and IurMdknetry are currently employed as lnspectlon techniques. These processes are time-consuming, destroy product, and may not detect low levels of contamlnatlon. Near-IR light scattering Is proposed as a new method of determlnlng low levels of contamlnatlon nonlnvasively and nondestructlvely. The method Is used successfully In the current study to detect contamlnatlon by a specles of yeast, mold, and bacteria In Intact plastic lnhrsion bags at levels as low as three colony-formlng units per mil- llllter for yeast. By use of the near-IR method, each lnject- able unit can be evaluated with Its Integrity malntalned, al- lowing the product to be dispensed or evaluated by another analytical method. INTRODUCTION Biotechnology has recently developed a number of new products that are now available for treating diseases. Many of the biotechnology products of pharmaceutical interest are fragile molecules (e.g., proteins) that cannot withstand ex- posure to the digestive tract or be formulated as a tablet or other oral dosage form. These products must be formulated as parenteral drugs (injectables) in a sterile vial or intravenous (IV) bag. Many parenteral drugs, including those derived from biotechnology and those used in conventional drug formula- tions, are also sensitive to the high temperatures required for heat sterilization (autoclaving). For these types of products, maintaining the stability of the drug (preventing decomposition) and ensuring the sterility of the product (preventing microbial growth) can be chal- lenging problems. Preservative systems and sterilization procedures must be well monitored (I) and tested by validated microbiological methods (2,3). Intravenous bags or vials can be filled aseptically as an alternative to heat sterilization. This process involves filling the product container with a solution of drug and excipients after it has passed through a series of filters (usually 0.22 pm) to remove any microorganisms. The number of parenteral products produced aseptically has increased over the last few years. This trend is alarming the Food and Drug Administration (FDA) because aseptic- filling procedures are not as rugged or safe as terminal ster- ilization. Autoclaving provides a sterility assurance level of * Author to whom correspondence should be addressed. lo4 or better (probability of an unsterile unit), while aseptic filling generally achieves an assurance level of only (4, zy 5). Accordingly, the FDA is trying to discourage the use of aseptic filling by requiring manufacturers of aseptically filled products to submit methods and data justifying why terminal sterilization cannot be used. These manufacturers must also describe the microbiologicalmonitoring and control procedures used to assure sterility (4, 6). The challenge to the analyst is to determine which products are contaminated and prevent their use. Perhaps the simplest method of assessing sterility involves the incubation of an injectable product for a period of time until the microorgan- isms in the container grow sufficiently that turbidity develops. The turbidity is then detected by visual examination. Un- fortunately, it can take a significant amount of time for turbidity to develop. Furthermore, the materials of some product containers hinder visual inspection. For example, some IV bags are composed of polymers that become trans- lucent when the bags are stored too closely together because water becomes trapped in the wall of the bag. Ordinarily, sterility assurance and microbial identifications are accomplished by selecting a number of units within a specific lot (batch sampling) and testing them by conventional microbiological methods. These methods are invasive and destroy the units of the product being examined. These methods also tend to be imprecise, laborious, and time-con- suming. An analytical method that enables the detection of low levels of microorganisms in parenteral products without la- borious random testing or the need for long incubation times would represent a significant advance in the analysis of parenteral products. Such a method could be used to detect contamination by bacteria, yeast, or molds in drug vials and IV bags. An analytical method based on near-infrared (near-IR) light scattering is proposed in this work as a method for detecting small quantities of microorganisms in sealed IV bags. This light-scattering method is noninvasive and non- destructive, preventing possible contamination of bags by the analytical method itself. Near-IR methods provide information on molecular structure, are well suited to the analysis of aqueous samples, and have been used in the detection of contaminated products zy (7,8). The energy of near-IR photons is low (approximately 1 eV), making destruction of even protein drugs by the probe beam unlikely. An analysis of the distribution quantiles of near-IR spectral data provides a powerful means of inter- preting light-scattering results. The principal advantage of the near-IR light scattering method is that every single unit of the product can be examined for sterility without invading the sample and destroying the product. Furthermore, the method appears to be able to differentiate between different classes of microorganisms as well as to isolate their location inside the container and determine their concentration in solution. 0003-2700/90/0362-2514$02.50/0 0 1990 American Chemical Society