http://www.biomedcentral.com/news/20030523/05

The replacement effect

Concern over a vaccine that causes unintended increase in related infections

By John Dudley Miller
 

Prevnar, a vaccine that is dramatically decreasing the number of American babies' infections from a common but serious bacterium, also increases the number of other infections the babies get, according to an Israeli researcher. No other vaccine in history has shown this unintended effect, said Ron Dagan, director of the Pediatric Infectious Disease Unit at Soroka University Medical Center in Beer-Sheva, Israel.

"[Its importance] is not minor," he said, "but for the moment, we cannot estimate totally the clinical effect because we still do not have enough vaccinated people."

Hypothetically, Prevnar's disease-enabling ability might someday become as powerful as its disease-preventing capacity, rendering it effectively useless, said Dagan, and necessitating a new vaccine design. Future vaccines utilizing the same approach of targeting specific strains in a larger microbial family might also create this effect, according to other researchers.

Prevnar, a 7-valent pneumococcal conjugate vaccine marketed by Wyeth, helps prevent routine fevers, headaches, and earaches in young children—as well as rarer, invasive infections with bacterial pneumonia or meningitis—caused by the seven most widespread varieties of 90 types of pneumococcus bacteria. Kids and adults routinely carry all 90 strains in their noses and throats, even when they are not sick.

Prevnar vaccinations have been recommended for all American babies since mid-2000. Between 1999 and 2001, "invasive" pneumococcal infections, the most serious category, dropped 68% in babies under age 2, according to a national study conducted by the US Centers for Disease Control and Prevention (CDC).

Prevnar also lowered pneumococcal infection rates in older people who had never received the vaccine. Over the same 2 years, invasive pneumococcal infections in the United States dropped 29% in adults aged 20 to 39 and 17% in people over 65.

Dagan attributes those decreases to the fact that vaccinated babies no longer carry the seven most prominent strains in their noses and throats, so they can't infect parents or grandparents.

Even so, Prevnar is already becoming somewhat less effective against infections of the middle ear, according to the CDC's Cynthia Whitney. "It's a big headache for us… the numbers just aren't as good as one would like," she said.

Between 1999 and 2001, another study in the Pittsburgh area (C. McEllistrem et al., unpublished data, 2003) found the percentage of pneumococcal middle ear infections caused by strains that Prevnar doesn't protect against more than doubled, from 16 to 37%.

This shift in prevalence from one kind of infection to another is not yet much of a problem with invasive infections, although Dagan said cases are increasing.

Dagan and others call the effect "replacement" because Prevnar effectively replaces infections by the seven targeted strains of pneumococci with similar infections by the other 83 strains. Prevnar vaccination also enables a slight increase in middle ear infections caused by two unrelated bacteria. These infections spread more easily because the vaccine wipes out their seven former competitors.

Dagan said the reason previous vaccines have not caused replacement is that healthy people don't harbor the bugs those vaccines protect against. With Prevnar, once it eliminates the seven selected strains, all the others proliferate inside babies' noses and throats, waiting for an opportunity to start an infection or to spread to someone else.

Steve Pelton, a pediatrician and pediatrics professor at the Boston University School of Medicine, points to an additional cause in the unexpected diversity of the 90 strains of pneumococci. Researchers had hoped that targeting the seven most common strains would provide more protection against the other 83, he explained. But if a replacement vaccine is eventually necessary, Pelton suggests a different approach: If you target something that's in all pneumococci, then you won't have this problem, he said.