MD consultで調べると

Mandell: Mandell, Douglas, and Bennett's Principles and Practice of Infectious Diseases, 7th ed.
Measurement of the Immune Response

Response to vaccines is often gauged by measuring the appearance and concentration of specific antibodies in the serum.[30] For some viral vaccines, such as those for measles, rubella, and hepatitis B, the presence of circulating antibodies correlates with clinical protection. Although this has served as a dependable indicator of immunity, seroconversion measures only the humoral parameter of the immune response. Secondary vaccine failure occurs when an individual who had previously had an adequate immune response loses measurable antibodies over time. This waning immunity can be attributed to a loss of long-lived memory B or T cells, in the absence of repeated exposure to the pathogen. Evaluating persistence of antibody has been used to determine duration of vaccine-induced immunity. However, the absence of measurable antibody may not mean that the individual is unprotected. Although a fall in titer takes place for some vaccines over time (e.g., measles, rubella, hepatitis B), on revaccination or challenge, a rapid secondary response is observed in IgG antibodies with little or no detectable IgM response, suggesting persistent protection. With some vaccines and toxoids, the mere presence of antibodies is not sufficient to ensure clinical protection, but rather a minimal circulating level of antibody is required (e.g., 0.01 IU/mL of tetanus antitoxin). Functional antibody is important in assessing immunity to bacterial polysaccharide vaccines. Opsonophagocytic activity (OPA) is considered the assay of choice for monitoring vaccine response[31] because the vaccines also induce nonfunctional antibodies that are detected in standard enzyme immunosorbent assay (EIA), although the EIA can be used as a proxy. Some immune responses may not in themselves confer immunity but may be sufficiently associated with protection that they remain useful proxy measures of protective immunity (e.g., vibriocidal serum antibodies in cholera). The measurement of cell-mediated immunity, which would be helpful in assessing the degree of ongoing protection in many circumstances, is usually limited to research laboratories and to only a few vaccines.

Long: Principles and Practice of Pediatric Infectious Diseases, 3rd ed.
CHAPTER 7 – Active Immunization

Measurement of Response

Ideally, reliable laboratory tests should be available to measure the presence and strength of each of the major effectors of protection against the disease for which the vaccination is administered. In practice, a wide variety of tests for presence of antibody are available, and include;
radioimmunoassay (RIA),
enzyme immunoassay (EIA),
complement fixation,
polymerase chain reaction,
and immunofluorescent techniques,

but these tests often do not measure the presence of functional (neutralizing or opsonizing) antibody. Tests for cell-mediated immunity are available in reference laboratories and research facilities; different tests are required to determine cytotoxic and helper T-lymphocyte (memory) functions.

For certain diseases, such as hepatitis B, poliovirus, and measles, reliable tests exist and antibody levels that correlate with protection are known; however, inexpensive tests are only widely available for hepatitis B. For other diseases, such as rubella, commercial tests are available, often using EIA methods, but their specificity is often less well defined, and their sensitivity is lower than sensitivity of neutralization assays. For some diseases, such as pertussis, no serologic correlate of protection has been defined. Development of better laboratory methods to measure protection and to permit rapid diagnosis of acute disease continues to be a priority of vaccine-preventable disease control programs.