Building a better pertussis vaccine

June 2005

Jim Cherry and I have had somewhat differing views on pertussis since we authored a piece on this subject with David Karzon at the request of the AAP (Pediatrics 1988;81:939-984). I have the greatest respect for Jim Cherry’s knowledge of pertussis. He is clearly one of the global leaders in the field. But I differ with him on a number of issues.

Admittedly, there could be a blip in cases of pertussis in teenagers as a result of decreased usage of some poor vaccine during the period when they were infants.

One of the two whole-cell vaccines in use at that time prior to the licensure of the acellular vaccines almost 15 years ago was less than 40% effective compared with placebo. At the same time, there was reluctance on the part of some parents to have their children receive these whole-cell vaccines and an over interpretation of contraindications by many physicians because of the reactions these vaccines produced. Thus, the poor potency, in addition to poor acceptance of whole-cell vaccines used until in the early 1990s, could have resulted in a large number of poorly protected teenagers at this time. Whether continuing to immunize those who are becoming teenagers at the present time who have received the newer, more effective acellular vaccines will benefit from this and subsequent groups is unclear. Many European countries have not even used a fourth dose of pertussis vaccine.

I must take issue with Jim’s statement: “The serologic diagnosis of B. pertussis infection is now accepted worldwide and is extremely useful in diagnosing pertussis in adolescents and adults.” My informants at the CDC indicated to me that they do not accept PCR or serologic diagnosis except those done at the State Laboratory Institute in Jamaica Plain, Mass. Nor does WHO accept serologic diagnosis.

The state of Massachusetts has the best serologic laboratory for diagnosis of pertussis. Its test was reported to detect 65% of culture-proven cases. Within two years after the introduction of serologic testing in 1987, the number of cases in Massachusetts increased more than threefold, the vast majority of the increase based on the newly introduced serologic test. Coincident with this increase was a rise in the number of cultures submitted to the State Laboratory Institute. As the number increased, the proportion that was positive decreased, reflecting the intensified effort to find cases. The authors stated, “both increased diagnostic awareness of pertussis and increased investigation of institutional outbreaks appear to have contributed to an increased number of bacteriologically confirmed cases over time” (J Infect Dis. 1994;169,1297).

At a recent Advisory Committee on Immunization Practices (ACIP) meeting, data were presented which indicated that the rate of pertussis in Massachusetts in recent years was six to 90 times the teenage rate for the rest of the United States. What makes the data particularly troubling is that Massachusetts manufactured its own pertussis vaccine prior to 1996. As noted in my prior commentary, Massachusetts and a handful of other states contribute the bulk of reported cases in the 10- to 20-year age group in the United States. The error in estimating the actual increase in cases may be amplified even further by the acceptance of epidemiologically linked cases. Thus, cases diagnosed on the basis of contact with erroneously “laboratory confirmed” cases will amplify the original error. The CDC will accept the diagnosis submitted from the states regardless of the test used for “laboratory confirmation.” New York has reported that many PCR results used for laboratory confirmation of cases reported to them were incorrect when tested in the State Laboratory Institute (J Clin Microbiol. 2002;40:2801-2805).

What may be more relevant than the proportion of teenagers and adults with chronic cough that have high antibody levels of pertussis toxin (PT) antibody is the rising rate in PT in the general population. These estimates are quite variable. In the most recent eight-city study, which I consider the best done, the estimate was 1.3%. About 80% of these are thought to be asymptomatic infections. How much of this apparent disease is due to loss of pertussis immunity, the use of poor vaccines in the past, the failure to vaccinate or the use of unapproved tests is uncertain. How much of this is correctable, if real, by immunization of teenagers also is uncertain. There are no accepted serologic correlates of protection to predict what will occur clinically. In light of this, clinical trials were required to evaluate the original acellular pertussis vaccines. Clinical trials to evaluate the effectiveness of pertussis vaccine in adults in the United States have been frustrating due to the scarcity of confirmed cases of pertussis in this population.

A ‘Hawthornesque’ effect

The Hawthorne effect, which I refer to in this context, does not involve the participant of a study, but rather those who seek out cases and report them. There has been much discussion of the effect of simply doing a study on the results, since it was first described by Elton Mayo in a General Electric plant in Hawthorne, Ill., in the late 1920s. Performance of workers increased simply as a result of being studied. Sociologists attributed this to the fact that the workers felt they were being observed or wanted to gain the esteem of cohorts. I am certain that the avidity with which cases of pertussis in adolescents are sought out has been enhanced by the reports that everyone else is finding them. I better not miss one. I must admit that no one was formally studying the seekers. I will yield on this one, and call it a “Hawthornesque” effect.

My colleague Ted Eickhoff, chief medical editor of Infectious Disease News indicates that he will vigorously promote the use of pertussis vaccine in adults and adolescents if recommended by the ACIP. The FDA has approved it. The English epidemiologist, Paul Fine, first suggested that the pertussis vaccine decreased disease more than transmission. If it is to have any effect on decreasing transmission of infection, we must all support it vigorously. There have not been any significant side effects in adolescent vaccinees in three other countries who have implemented this program. I would have preferred to wait until they found out whether it also produced the results postulated. The interpretation of the results will be complicated by the fact that pertussis toxoid, which is used in the serologic assay, is also in the vaccine. Thus, if one accepts serologic confirmation of infection, serology probably will be useless in evaluating the effectiveness of the vaccine.

If vaccination of teenagers produces the effect postulated, it certainly will be a plus. I would have much preferred that the tens and ultimately hundreds of millions of dollars spent on this program be spent on developing a better pertussis vaccine, serologic correlates of protection and more reliable methods of diagnosis. One of my concerns is that the years consumed in determining the effect of this program will delay this effort. This is by far the poorest bacterial vaccine in routine use and one of the oldest. Although in 2000 there were 35 cases of tetanus and one case of diphtheria, last year there were more than 12,000 reported cases of pertussis, yet DTaP are contained in the same vaccine. The Japanese developed acellular vaccines a generation ago. Is it not time that we turn our attention to making a better pertussis vaccine?