Vaccine
Safety > Research
Safety of
Acellular Pertussis Vaccine:
Follow-Up Studies
As published in Brown F, Greco D, Mastrantonio P, Salmaso S, Wassilak
S: Pertussis Vaccine Trials. Dev Biol Stand. Basel, Karger, 1997, vol 89, p. 373-375
Author: |
Robert T. Chen, M.D., M.A.
Chief, Vaccine Safety and Development Activity
Epidemiology and Surveillance Division
National Immunization Program (MS-E61)
Centers for Disease Control and Prevention
Atlanta GA 30333 |
Although the clinical trial
data presented at this meeting clearly show that the acellular pertussis
vaccines as DTaP are safer than the whole-cell pertussis vaccine as DTP,
several important questions about the safety of DTaPs remain to be answered.
In the Swedish Stockholm trials I and II, the apparent high incidence of
hypotonic, hyporesponsive episodes (HHE) observed warrants continued study
to understand better the characteristics of these events and the recovery
status of these infants. In addition, the change in pertussis vaccine
exposure in Sweden since the 1980's and the availability of various disease
registries makes Sweden an attractive site to conduct ecological or other
studies examining relationship between pertussis vaccination and certain
adverse events with insidious/delayed onset (e.g., asthma, diabetes
mellitus).
While the efficacy of
various DTaPs have been compared critically by several speakers, similar
critical comparison of the "safety profiles" between the DTaPs
have been limited. Almost all the discussion on safety at this meeting has
focused on comparison of DTaP vs. whole-cell DTP and DT. The NIH-sponsored
phase II trials of acellular pertussis vaccines did not show important
differences in the frequencies of common local and systemic reactions.
Nevertheless, a meta-analysis or summary paper which focuses on comparison
of rates of more serious events (seizures, fever>40oC., HHE's)
would be helpful.
Clinical trials, despite
their admittedly methodological purity and elegance, are limited in sample
size, duration, and population heterogeneity. Rare or delayed adverse
reactions, or reactions in sub-populations can only be detected after
product licensing and general use. One of the major concerns with the safety
of whole-cell pertussis vaccination relates to rare serious events such as
acute encephalopathy, estimated to occur at 0.0-10.5 per million doses. The
size of the phase I/II trials (in the hundreds to low thousands) and phase
III trials (in the low tens of thousands) for acellular pertussis vaccines
in infants are clearly too small to address such an issue. While the greater
safety of these vaccines in prelicensing clinical trials are promising, the
safety of each vaccine regarding association with the most serious adverse
events can only be examined by postlicensure monitoring.
Traditionally,
post-marketing surveillance systems have relied on essentially passive
monitoring such as the Vaccine Adverse Event Reporting System (VAERS) in the
United States (1). Such surveillance systems, however, have great difficulty
in distinguishing causal from coincidental events and surveillance artefacts.
Review of VAERS data from 1991-1993 provides a first perspective on the
safety of DTaP vaccines in widespread use . During this period, five million
net doses of DTaP have been distributed in the U.S. for the fourth and fifth
doses and the overall reporting ratio for serious adverse events in children
age 1-4 years appeared to have declined to one third that of whole-cell DPT
vaccine (2). Similar surveillance after use of DTaP in infants, as well as
any new combination vaccines, will be needed. Comparison of the "safety
profile" in VAERS reports by each DTaP vaccine type may also provide
some greater insights into their relative safety. The development and use of
standard protocols for follow-up of serious adverse events may improve the
scientific use of such reports.
Passive surveillance
systems will always be severely limited, however, by their biases and
limited information content. Active surveillance or definitive
epidemiological studies are required to provide validation of signals of
concern generated by passive surveillance (3). Ad hoc studies (e.g.,
National Childhood Encephalopathy Study) cannot be done on a very timely
basis and are also limited in the small number of hypothesized adverse
events that can be studied. Pharmacoepidemiologists studying treatments have
been relying on pre-organized Large-Linked DataBases (LLDB) to provide more
flexible and timely studies on the safety of drugs (3). In the U.S., CDC has
worked with several health maintenance organizations since 1991 to create
the Vaccine Safety Datalink (VSD) project. The VSD currently conducts
comprehensive surveillance on approximately 500,000 children under seven
years of age (2% of the U.S. age-specific population) annually. The
computerized vaccination records (lot-specific) on these children are linked
with medical outcomes (hospital, outpatient, emergency room, laboratory),
and covariates (e.g., from census data tapes) for epidemiological studies
(4). About 100,000 doses of DTaP have been given to this population to date.
As more DTaP is used, the VSD should provide a more rigorous means of
monitoring vaccine safety. Other countries (e.g., U.K., Sweden, Denmark)
have also begun or are planning to embark on LLDB studies.
Finally, several lessons
are available to us on the importance of monitoring vaccine safety. The
first is that public confidence, once lost, is difficult to regain (3). This
painful lesson was learned with public concerns regarding safety of DTP in
many countries during the 1970's and 1980's. Furthermore, the principles of primum
non nocere(first do no harm) are, if anything, more important in public
health than in clinical medicine, since vaccines, unlike most drugs, are
given to healthy persons, usually infants. The World Health Organization's
Expanded Programme on Immunizations (EPI) has recently called for all
national immunization programs to conduct surveillance of vaccine adverse
events following vaccination (5).
The area of risk
communications will become increasingly important as immunization programs
successfully reduce vaccine-preventable diseases to the point where the
frequency of vaccine-associated adverse events exceeds that of the disease.
Risk communications will be challenging when most health care providers as
well as parents have never seen a case of vaccine-preventable disease. In
this new era, it may be more productive for us to speak of "vaccine
safety" research, a positive concept vs. "adverse event"
research. We always speak of "safety and efficacy" of a vaccine,
placing safety before efficacy; we should never forget that.
References
-
Chen RT, Rastogi SC,
Mullen JR, Hayes S, Cochi SL, Donlon JA, Wassilak SG. The Vaccine
Adverse Event Reporting System (VAERS). Vaccine 1994;12:542-50.
-
Rosenthal S, Chen R,
Hadler SC. The safety of acellular pertussis vaccine versus whole cell
pertussis vaccine: a post-marketing assessment. Arch Dis Child (in
press).
-
Chen RT. Special
methodological issues in pharmacoepidemiology studies of vaccine safety.
In: Strom BL, ed. Pharmacoepidemiology. Sussex: John Wiley & Sons,
1994.
-
Chen RT, Glasser JW,
Rhodes P et al. The Vaccine Safety Datalink Project: A new tool
for improving vaccine safety monitoring in the United States.
Pediatrics 1997 (in press).
-
Expanded Programme on
Immunization. Surveillance of adverse events following immunization:
field guide for managers of immunization programs. Geneva:WHO/EPI/TRAM/93.2;
1993.
|