The vaccines protect in a novel manner. Animal studies strongly indicate that the OspA antibodies elicited by the vaccines prevent human infection by neutralizing the spirochetes in the gut of the vector tick between the time of attachment and the subsequent transmission of B. burgdorferi to the host [6-8]. This unique mechanism of action suggests that development of effective neutralizing antibody levels is the critical determinant of vaccine efficacy. An additional protective effect may occur in the host's bloodstream, although B. burgdorferi undergoes substantial antigenic change, resulting in repression of OspA expression when in contact with warm blood.

The safety profile in vaccine trials that included more than 10 000 persons 15 years of age and older is highly reassuring. Participants had mild local discomfort at the injection site but did not experience serious systemic adverse reactions. Nevertheless, the FDA Advisory Committee expressed ambivalence and raised four main issues about the vaccines that require future attention.

1. Duration of protection, need for boosters, and alternative dosage schedules. Randomized clinical trials have established the efficacy of the Lyme disease vaccines at 20 months from the first dose, but longer follow-up studies are not yet available [4, 5]. Antibody levels seem to decrease rapidly, and frequent (perhaps annual) boosters will probably be necessary. Now that it is clear that a third dose is needed to achieve adequate protection levels, dosage schedules other than the current 0-, 1-, and 12-month schedules must be explored with the goal of achieving complete immunization within a single Lyme disease season.

2. Use of the vaccines in children. Children are at the highest risk for Lyme disease, but no efficacy or safety studies have been completed that will allow FDA approval of the use of Lyme disease vaccines in children. Pediatric caregivers will probably encounter substantial pressure to use the vaccines "off-label."

3. Immunopathogenicity resulting in rare or late adverse reactions. Persons with a history of immune-mediated neurologic disease or arthritis, immuno-compromised persons, and pregnant women were excluded from the phase III vaccine trials. Therefore, the full safety profile is unknown, and the vaccines are not recommended for these groups. Persons with HLA-DR4 haplotype have an increased risk for chronic Lyme arthritis, a condition associated with OspA reactivity in synovial fluid. The theoretical possibility that vaccination with recombinant OspA vaccines may lead to an immunopathogenic response based on molecular mimicry led the FDA Advisory Committee to express caution about vaccinating persons who have arthritis until further studies are completed. In current studies, persons who report a history of Lyme disease do not seem to have an increased risk for serious adverse reactions, but long-term adverse events have not been evaluated [9].

4. Effect on serodiagnosis and masking of other tick-borne diseases. Because antibodies to OspA are the basis of current enzyme-linked immunosorbent assays that are used to test for exposure to B. burgdorferi, Lyme disease vaccines will cause false-positive results. This will require the use of other diagnostic tests, such as the more expensive Western blot assay, for serologic evaluation of vaccinated patients for active Lyme disease. Western blot studies in current trials indicate that the vaccines fully protect against infection with B. burgdorferi and do not result in subclinical or modified infections that may cause late complications of Lyme disease [4, 5]. This is reassuring but requires further study, particularly in patients in whom adequate protection is not maintained through booster immunizations. Because patients with recognized Lyme disease should be carefully evaluated for concomitant babesiosis and ehrlichiosis, there is concern that patients exposed to deer ticks will not be evaluated promptly for these more serious diseases in the absence of a sentinel manifestation of Lyme disease.

Humans represent a dead end for B. burgdorferi, and although vaccination will protect individual persons, it will not interrupt the epidemiologic transmission of the spirochete [10]. When a public health priority is assigned to Lyme disease vaccines, it is important to note that Lyme disease is not communicable from human to human and is associated with negligible mortality. In addition, most patients with Lyme disease (approximately 85%) develop the characteristic erythema migrans rash, which allows early recognition of disease and implementation of effective treatment with inexpensive oral antibiotics. Nevertheless, the public perception of risk for Lyme disease has increased to a level of extreme concern in areas highly endemic for this disease, and surveys indicate that demand for the vaccines will be high.

The distribution of Lyme disease is highly regional. More than 90% of cases occur in 10 states (Wisconsin, Minnesota, and states on the mid-Atlantic and New England seaboards). Even in these states, the incidence of Lyme disease varies by county and by smaller geographic divisions. Although the major determinant of risk is geographic exposure, other recognized risk factors in endemic areas include outdoor work, recreational or leisure activities, and failure to take personal protective measures to avoid tick bites [10].

The uneven distribution of Lyme disease and the high degree of variation in the risk to any particular person within an endemic area confound efforts to establish uniform recommendations for vaccine use. The vaccines will not be used universally, and individual patients and health care providers (including public programs) will exercise considerable choice in their use. In regions where Lyme disease is highly endemic (incidence, >1% per year), the vaccines will probably be recommended for persons 15 to 65 years of age, excluding shut-ins and urban dwellers, who have reduced risk for tick exposure [10]. In regions of intermediate risk (incidence, 0.1% to 1% per year), recommendations will target persons whose activities increase their risk for tick exposure. The vaccines will not be recommended for persons in the regions with the lowest incidence.

These stipulations leave uncertainty about whether to use the vaccine in many cases, especially in persons who travel to highly endemic areas (many of which are prime tourist destinations). In these situations, physicians will make decisions individually on the basis of duration of exposure, type of activity, and level of personal concern, just as physicians who advise persons about international travel must weigh individual factors before deciding which preventive measures to take. Perhaps a new field of "domestic emporiatics" (advice for persons traveling in North America) will be spawned.

The excellent safety profiles of the Lyme disease vaccines and great public demand for them will make it difficult to deny vaccination to highly concerned persons. This will create problems for insurers and public vaccine programs that would like to be selective in use and reimbursement. Although the cost of the vaccines has not been determined, it is expected to be $50 to $100 per year; cost studies indicate a high price, especially compared with other measures (such as early recognition and treatment of Lyme disease, promotion of personal protective measures, and environmental control activities to reduce exposure to deer ticks). The need to vigorously pursue these less costly measures will not be obviated by the introduction of the vaccines.

Lyme disease occurs in all age groups [10]. It is anticipated that current studies will establish the safety and immunogenicity of the Lyme disease vaccines in children and will result in the addition of these vaccines to the standard pediatric immunization schedule in areas where Lyme disease is highly endemic. Additional studies are needed in elderly persons. More user-friendly schedules (for example, 0-, 1-, and 2-month schedules) will probably be evaluated and adopted.

The current Lyme disease vaccines should be considered first-generation vaccines that are useful but limited additions to our arsenal. A goal of future vaccines will be to provide longer-term protection and broader protection against variant strains of B. burgdorferi. The recent sequencing of the entire DNA genome of B. burgdorferi (>1.4 million base pairs) will undoubtedly facilitate future vaccine development [11].