From the data reviewed (see Table 1 in the report), the committee concludes that the evidence favors rejection of a causal relationship between hepatitis B vaccine administered to adults and incident multiple sclerosis. The committee also concludes that the evidence favors rejection of a causal relationship between hepatitis B vaccine administered to adults and multiple sclerosis relapse. There are no epidemiological data regarding the relationship of hepatitis B vaccination in infants and young children and the risk for MS. The committee cannot extend the causality conclusion based on epidemiological studies in adults to include a possible risk to infants and young children.

Based on the data reviewed (see Table 2 in the report), the committee concludes that the evidence is inadequate to accept or reject a causal relationship between hepatitis B vaccine and the first episode of a central nervous system demyelinating disorder. A first episode of CDD might or might not be indicative of MS, which requires more than one episode for diagnosis. The evidence derives from unpublished data, for the most part, and the one published study had methodological weaknesses.

Only one uncontrolled ecological study was available regarding ADEM (Sadovnick and Scheifele, 2000). The committee concludes that the evidence is inadequate to accept or reject a causal relationship between hepatitis B vaccine and ADEM.

Only one unpublished study was available regarding optic neuritis (DeStefano, 2002) (see also Table 3). Thus, the committee concludes that the evidence is inadequate to accept or reject a causal relationship between hepatitis B vaccine and optic neuritis.

Only case reports were available regarding transverse myelitis, GBS, and brachial neuritis. Thus, the committee concludes that the evidence is inadequate to accept or reject a causal relationship between hepatitis B vaccine and transverse myelitis, GBS, and brachial neuritis.

Three mechanisms—molecular mimicry, bystander activation, and superantigen stimulation—can be posited as the theoretical ways in which immunization of any kind (either through infection or vaccination) could cause the development of demyelinating diseases of the central and peripheral nervous systems. Molecular mimicry refers to the process of a microbial antigenic determinant cross-reacting with a self-protein. If the self-protein is a myelin-related protein, the subsequent immunological response could lead to autoimmune demyelination. Bystander activation refers to a process of a microbial infection (or other stimulus) leading to the release of a large quantity of normally sequestered host proteins and the subsequent destruction of host tissue, which could include cen-

Front Matter (R1-R14)

Executive Summary (1-16)

Immunization Safety Review: Hepatitis B Vaccine and Demyelinating Neurological Disorders (17-90)

Appendix A: Committee Conclusions and Recommendations from Previous Reports (91-99)

Appendix B: Public Meeting Agenda (100-102)

Appendix C: Chronology of Important Events Regarding Vaccine Safety (103-106)