How many infective viral particles are necessary for successful mass measles immunization by aerosol?

Abstract

Objectives

This study characterized the performance of the “Classical Mexican Device”, which immunized 4 million children against measles, demonstrating the efficacy of aerosol vaccination.

Methods

Using plaque-forming units to quantify virus, the particle size distribution (Next Generation Pharmaceutical Impactor) and rate of output coupled with age specific patterns of breathing allowed the calculation of expected pulmonary deposition.

Results

The estimated immunization dose for infants was 30 pfu's, for small children 50, and for older children and adolescents it was 130 and 225, respectively.

Conclusions

These performance characteristics can be used to develop newer battery operable devices that are licensable.

Introduction

World Health Organization (WHO) statistics indicate that more than three quarters of a million children die each year from measles. The cause of death is predominantly measles pneumonia making it one of the most common respiratory diseases leading to death in children. With proper immunization, measles is a preventable disease. Barriers against mass immunization include the lack of trained personnel to administer the subcutaneous injection, lack of health care resources, problems dealing with potentially infective waste, and many other logistical issues throughout much of the developing world. Measles vaccination by the aerosol route has considerable appeal. It has been shown to be capable of immunizing infants despite the presence of maternal measles antibodies [1]. Using a ‘home built’ system in Mexico, approximately 4 million children were exposed to measles vaccine aerosols, and observations suggested high efficacy of this route in preventing disease [2]. The system was highly efficient producing 93–100% seroconversion, which compared favourably with the expected seroresponse to subcutaneous administration in children over 9 months of age. The system used an IPI jet nebulizer driven by a 3/4 hp Evans industrial air compressor to deliver an aerosol by face mask of the reconstituted Edmonston Zagreb strain of attenuated measles vaccine virus. This strain of virus and route of administration has been shown to be superior to subcutaneous injection in re-vaccination studies of school children [3], [4]. The children were exposed to the output of the device for a 30 s period via a paper conical mask held over the mouth and nose. The disadvantage of the system is both its bulk and the power consumption of the compressor, which would create difficulties in areas of the world with no access to electricity. The very noisy compressor may frighten small children and is not approved for human use, which would make international licensing complicated. There is a need for an aerosol delivery system that could be licensed internationally and have less power demands than the “Mexican” system, while retaining the same level of performance in terms of efficacy in vaccination.

The particle size of inhaled droplets is the most important determinant of the probability of deposition in the lungs. For aerosolized soluble pharmaceuticals, the volume of the droplet and hence the amount of drug carried is proportional to the third power of the radius. Small particles, such as those <1 μm, may be numerous but carry little drug. Larger particles, such as those >5 μm, while carrying much more drug may be too large to penetrate below the vocal cords in adults [5], [6], [7]. The fraction of the drug carried in particles sufficiently small as to enter the lungs when inhaled is known as the respirable fraction (RF). The data [7] supporting an RF of 5 μm for adults or fully-grown children is derived from deposition studies that looked at this factor specifically. There is supportive data from recalculating a deposition study of Wildhaber et al. [8] for a choice of 4 μm for children of 4 years. There is little direct in vivo evidence in small children and infants, but there are numerous suggestions that the choice of a 5 μm cut off appropriate in adults [7] would be too large [8], [9] , especially for infants inhaling through a face mask [10]. This is supported by data from an ingenious in vitro model [11] of the upper airway of a 9-month-old infant that would suggest that the RF would be ≤ 3 μm and inversely dependent on inspiratory flow. Most of this work has been done with pharmacological agents, not viruses that will replicate, making “dose” calculations more complicated. Furthermore, field studies would suggest a wide safety margin for the measles vaccine [1], [3] since the route of delivery likely resulted in large variations in the amount of virus deposited below the vocal cords. Many children undoubtedly received a much larger dose than that necessary for immunization, but undue side effects were not reported. Currently, the dose required for aerosol immunization remains unknown. Furthermore, while there is agreement that the wild type measles virus spreads through the respiratory system, it is not definitively known whether lung deposition or upper airway deposition or both would be necessary for successful immunization via the aerosol route. The key to being able to develop a system analogous to the proven “Mexican” device, but more practical will be a better understanding of the likely viral load delivered to the lungs of infants and children of various sizes as well as that likely to deposit in the respiratory tract as a whole.

For pharmacological agents in solution, there is good evidence that the drug is uniformly distributed in the droplets. Automated particles size measuring devices that measure only droplet size give identical results to those that quantify the actual drug in a size range such as cascade impaction, providing care is taken to prevent heat transfer to the aerosol [12]. However, for suspensions such as budesonide, this may not be the case [13], [14] since particles may be too large to be distributed evenly throughout the aerosol resulting in many small droplets that are “empty.” In such a situation, particle sizing the “active” aerosol requires an inertial impaction technique such as a cascade impactor [14] with quantification of drug within the particle size distribution. The purpose of this research was to determine the output and particle size distribution of measles vaccine aerosolized by the “Classical Mexican Device” (CMD). From this data and the known respiratory pattern of children, it will be possible to estimate the necessary “infective” dose carried in various particle size ranges, providing a guide for the design of a more practical system for large immunization campaigns in the developing world.