Abstract
The laboratory toxicologist is frequently faced with the challenge of selecting appropriate vehicles or developing utilitarian formulations for use in in vivo nonclinical safety assessment studies. Although there are many vehicles available that may meet physical and chemical requirements for chemical or pharmaceutical formulation, there are wide differences in species and route of administration specific to tolerances to these vehicles. In current practice, these differences are largely approached on a basis of individual experience as there is only scattered literature on individual vehicles and no comprehensive treatment or information source. This approach leads to excessive animal use and unplanned delays in testing and development. To address this need, a consulting firm and three contract research organizations conducted a rigorous data mining operation of control (vehicle) data from studies dating from 1991 to present. The results identified 65 single component vehicles used in 368 studies across multiple species (dog, primate, rat, mouse, rabbit, guinea pig, minipig, chick embryo, and cat) by multiple routes. Reported here are the results of this effort, including maximum tolerated use levels by species, route, and duration of study, with accompanying dose limiting toxicity. Also included are basic chemical information and a review of available literature on each vehicle, as well as guidance on volume limits and pH by route and some basic guidance on nonclinical formulation development.
Formulation, even at the rudimentary level employed for initial studies used to evaluate the safety of new drug candidates, is perhaps the weakest link in both pharmacology and toxicology for drugs and industrial and agricultural chemicals.
In the preclinical safety assessment of potential new drugs, it is required that the material of interest must be suitably formulated in a manner that allows adequate administration of the test substance, with little or no effects in test animals that is attributable to the vehicles used in producing such a formulation. The formulation must be suitable for the intended route of administration, maintain the stability of the active ingredient, and preferably maximize the systemic bioavailability of the drug. Occasionally, the vehicle(s) or formulation is specified by the sponsor of a study, but more frequently is the informed choice of the laboratory conducting specified safety studies.
Because the process of vehicle selection has been mostly one of custom or personal choice, there are many vehicles which have seen use in preclinical formulation. However, results as to their suitability, utility, and limitations on their use are not generally reflected in the literature or taught in any formal manner. Certainly such information is not attainable in any readily available place. This paper is intended to at least begin to fill this gap.
General Preclinical Formulation Principles
Dosing formulations for preclinical studies should be selected with consideration of a number of desirable characteristics as summarized in Table 1 (Gad 1994; Gad and Chengelis 1997).
Animal use and care guidance provides limits to volumes that may most commonly be administered by each of the common routes. These limits are presented in Table 2 (Gad 1994; Gad and Chengelis 1997).
To provide information to all those who in the future must select suitable vehicles for formulation for nonclinical dosing of animals, a data mining operation was undertaken.
METHODS
Four separate organizations undertook a review of their files to identify and collect data on control vehicles they had used in studies over the previous 15 years (1991 to present). Each in vivo study conducted during this period had its vehicle control group evaluated. If the vehicle was other than water, the highest no-observable-adverse-effect level (NOAEL) for the vehicle formulation used was determined and this was added to the database. Extracted in this manner by the participating organizations (Calvert, CIT, Gad Consulting Services [GCS], and MPI) were the maximum nontoxic volume identified in these reviewed studies by species, route, and duration. The nature of any dose limiting toxicity for a vehicle or vehicle combination was also identified. The number of single-vehicle formulations disclosed, as reflected in the body of this paper, was large, with more than 60 different formulations being reported here. There were also a number of combinations that are not included in this paper.
GCS then assembled the provided data into the tables presented here, with materials listed in alphabetic order (but with these tables also indexed in Table 69). Table 69 also provides basic physical chemical data and general toxicity references on these vehicles.
The actual acceptable vehicle usage data is presented in Tables 3 to 68, with a single vehicle being presented per table and arranged in alphabetical order according to common name.
Subsequent to the publication of this article, GCS will post this data to a website (www.gadconsulting.com), with a mechanism to submit additional vehicle data. GCS is committed to updating and maintaining such a free access website database for at least the next 5 years, through 2011.
Formulation for drugs, and in a less sophisticated manner, all test materials for evaluation in intact animal systems (whether done for efficacy/pharmacology testing or for toxicology) is a field of expertise of its own. Although there are a few books on formulation of human drugs (Racz 1989, for example), no such volume known to the authors exist for preclinical formulations.
Although the development of pharmaceutical formulations for marketed clinical products is done in a rigorous manner, what is employed for nonclinical testing is much more pragmatic. The reader is referred to Racz (1989), Yalkowsky (1999), and Weiner and Kotkoskie (1999) for more detail than is offered here as to the principles of vehicle and formulation component selection.
DISCUSSION
Although some of the vehicle options and choices presented here may seem unusual, it should be noted that what is reported here represents what is used from the discovery phase of drug development through early nonclinical safety assessment, and in many cases, all the way through to use in marketed products (the reader is invited to inspect the FDA’s [Food and Drug Administration] inactive ingredient list, at www.fda.gov/cder/drug/iig/inact.pdf, to verify this point). Lactose as a formulation component for inhalation may seem contralogical, yet marketed products employ it for dispersion of drug powder and metered dose inhaler (MDI) dosage formulation.
For all formulations, the ability to accurately administer an aliquot of what has been prepared, with each aliquot being of uniform content, is a primary requirement. With fewer exceptions (i.e., capsule fills), this means achieving a solution or (second choice) stable suspension.
Although the first choice for any systemic route is always a modification of an aqueous based vehicle, the physiochemical characteristics of the test material dictate available options (Yalkowsky 1999; Racz 1989). One starts with either a polar or nonpolar solvent to begin with, depending on which achieved adequate dissolution of the test material, and works from there.
Although it is common in much of the pharmaceutical industry to stick to those components that have already seen use in other drugs from as early a point in development as possible, such is nowhere near a universal case.
In hopes of making the database presented here as accessible as possible and to broaden its content, GCS has set up an online version with free access (go to www.gadconsulting.com), which provides an online mechanism to submit new data, and will maintain and update the electronic site for at least 5 years after publication of this article (until 2011).