June 2003

---Philip A. Brunell, MD

There is a wonderful article in the May 1, 2003 issue of The New England Journal of Medicine (348:1746), which compares the frequency of pneumococcal disease for the periods before and after the licensure of the pneumococcal conjugate vaccine (PCV7). The most interesting finding is the decline in pneumococcal disease in individuals who had not received the vaccine.

Surprise, but no surprise

That there was herd immunity is not surprising, as PCV7 has been shown to decreases carriage of vaccine serotypes. Thus, one might expect transmission of pneumococci from those vaccinated to contacts to be reduced. Of the most impressive decreases in pneumococcal disease was in the 20- to 40-year-old age group, ie, among those in the child-rearing age group. They had been shown previously to have high carriage rates presumably due to transmission of pneumococci from their young children. A similar decline of disease in the unimmunized was observed following the introduction of Haemophilus influenzae type b (Hib) conjugate vaccine.

Of particular interest, however, was the decline in the 65 and older age group who are at significantly increased risk of severe pneumococcal disease. Because of this, pneumococcal polysaccharide vaccine (23PS) is routinely recommended for them. Their rates are 2 ½ times greater than the 40- to 64-year-old group. In the 65 and older age group, there was an 18% decline in disease despite the fact that only seven of the 23 serotypes in the 23PS vaccine are contained in PCV7. Indeed the entire decline of disease in this age group could be accounted for by conjugate vaccine serotypes.

There was virtually no decline in the serotypes in the polysaccharide vaccine (23P), which were not contained, in the conjugate vaccine.

This raises some interesting questions. If one assumes there has not been a change in the utilization of the 23P, which has been recommended for routine use in this group, then the conjugate vaccine is adding protection that the polysaccharide vaccine does not adequately provide. The 23P vaccine has been reported to be between 50% and 75% effective in those who are vaccinated in this age group. Although one must be cautious in extrapolating to the pediatric age group, it is known that young children do not respond well to the 23P vaccine. In the past, 23P vaccine had been recommended for children at increased risk at age 2. Many, however, will not respond optimally to the more common serotypes until an older age.

At the current time one is given the option of using either of the two pneumococcal vaccines for children at moderate risk after the age of 2 years. Efficacy data for these vaccines are sparse in this age group. On the basis of the findings in this paper, one might be tempted to tilt toward the use of the conjugate vaccine despite its greater expense.

Catch-up

Because manufacture of PCV7 is back to normal, the use of the vaccine may be liberalized, and catch up for those who missed doses during the period of shortage can be addressed. Schedules for catch up are now available at www.cdc.gov/mmwr/preview/mmwrhtml/mm5219a6.htm. One of the other more interesting findings of the NEJM study was that all of these declines — including the striking declines of invasive pneumococcal disease in those under 12 months, between 12 and 23 and 24 to 36 months (69%, 68% and 44% respectively), occurred while there was considerable underuse of the vaccine.

The manufacturer is said to have sold 9 million and 15 million doses the first two years of licensure, yet, 32 million doses would have been required for a four-dose schedule for the estimated 4 million babies born annually. It was estimated that approximately 10% of the doses distributed were used for children more than 24 months of age.

As these results were achieved despite the fact that some of those vaccinated may have received fewer than the recommended number of doses, the investigators raise the interesting question as to whether we really need as many doses in the series as is now recommended. In the initial reports very few failures were in children vaccinated who failed to complete the series of doses. This may be true, but we do not know the number of these children at risk. Certainly curtailed series were given during the shortage. We may obtain more information on curtailed schedules retrospectively from this experience.

Fewer doses?

So, are all of the doses of this very expensive vaccine, which have been recommended, really needed or is it possible to give fewer doses? Changing the recommendations, which would not conform to the package insert, ie, FDA approval, would be very difficult. You are urged not to do this experiment in your practice but to continue to use the recommended schedule.

Although we refer to the PCV7 as a vaccine, it is in reality seven vaccines. This is one reason for its high cost. The separate vaccines have different antibody kinetics and although some stimulate a response to one or two doses others require three or four. Thus, reducing the number of doses might reduce the effectiveness of some of the contained vaccines. That is, we might expect to see failures to some of the strains with a reduced number of doses. One must balance this against reducing the cost of immunization. We are told by Stan Block, MD, who practices in rural Kentucky that many family practitioners do not stock the vaccine because of the cost. Thus, children must find another vaccine provider or do without.

One of the nagging questions raised by the routine use of PCV7 has been whether there will be replacement of the vaccine serotypes. In the current study, the rates of disease due to vaccine and vaccine-related serotypes in children under 2 had declined while the rate of serotypes not represented in the vaccine had increased 27%, which was not statistically significant. This certainly will bear watching.

The other major issue is the problem of pneumococcal resistance. Although the number of pneumococcal infections diminished during the period of observation, the proportion of non-susceptible types did not change during this period. However, the number of infections caused by resistant strains, not to mention their circulation, were probably reduced.

Are we there yet? The report through 2001 indicates that the disease rate in children younger than 5 years of age had dropped from 96.4 to 39.7 cases per 100,000. This represented a decline of 59%. We have not approached the success we have achieved with other vaccine preventable diseases or with Hib in particular. The interruption of supply for PCV7 did not help, but there are noticeable inequities among those sharing in this success story.

States with aggressive vaccine programs have done well. Blacks, although having a higher rate of disease have had a similar rate of decline following introduction of the vaccine. We may never achieve the success we have experienced with its predecessor conjugate vaccine with is directed against a single serotype but the finding of herd immunity is a helpful sign. Are we there yet? Not quite.