CDC Report on Missouri H5N1 Serology Testing

Topline Summary of Findings

In August 2024, healthcare providers in Missouri had a patient who had gastrointestinal symptoms with a history of chronic respiratory illness. The person was hospitalized and tested for multiple respiratory pathogens, and was positive for influenza A. They recovered from their illness and were discharged. Follow-up surveillance testing was conducted at the Missouri Department of Health and Senior Services Public Health Laboratory using polymerase chain reaction (PCR) that is part of routine national influenza surveillance. Testing found that the patient tested presumptive positive for avian influenza A(H5) ("H5 bird flu"). There was no history of exposure to infected animals or humans. The specimen was forwarded to CDC as part of routine protocol. CDC confirmed the specimen as H5N1 bird flu and this was reported as a case on September 6, 2024.

Missouri state and local officials performed a lengthy retrospective investigation of everyone who came into close contact with the patient and identified seven people (6 health care workers and one family member) who had symptoms that warranted additional investigation given their potential exposure to this patient. CDC, in support of this investigation, conducted a series of tests on blood specimens from 6 of these people – and the original patient – to help identify signs of previous infection or exposure to H5N1 well after full resolution of their symptoms. This was not possible for the others, whose suggestive symptoms had completely resolved at the time of interview. None of the results of this extensive testing and investigation support that human-to-human spread occurred. The health care workers who were tested showed no signs of previous H5N1 infection. They were all "sero-negative." This finding rules out person-to-person spread between the MO case patient and any of health care workers tested. The Missouri case and a household contact both had some evidence – though inconsistent – which suggested exposure to – or a previous infection with – H5N1 using these serologic tests.

To date, human-to-human spread of H5 bird flu has not been identified in the United States. CDC believes the immediate risk to the general public from H5N1 bird flu remains low, but people with exposure to infected animals are at higher risk of infection.

Following these tests, CDC continues to assess that the risk that someone who has not had contact with an infected animal will become infected with H5N1 remains low.

Background on serological testing methodology:

Serological testing for H5N1 was performed on single time point serum specimens taken from retrospectively identified, previously symptomatic, exposed people after enough time had passed to allow their immune system to create antibodies. These tests assess whether the antibodies produced by patients or close contacts of the patient are able to neutralize or bind to the H5N1 virus. If they are able to neutralize or bind to the virus, that suggests a previous infection or exposure to the virus.

The samples were collected by state and local officials from:

• One confirmed H5 case (reported on Sept 6, 2024) who tested positive for H5 by PCR through routine influenza surveillance;
• One household contact of that case who was not tested at the time because their symptoms had resolved;
• Five health care workers with possible work-related exposures to the case who had not been tested since their symptoms had resolved.

Note that there was also one additional previously identified health care worker with exposure to the confirmed H5 case. This person was tested for influenza at the time and was found to be negative, so no serological testing was performed on this person. Since such a low amount of virus was detected in the clinical specimen from the confirmed case in Missouri, a complete genomic sequence from the virus could not be recovered. CDC was able to obtain some sequence data from what little virus was present in the clinical specimen. This sequence data – publicly available as "A/Missouri/121/2024" – showed some changes in the hemagglutinin (HA) that might impact how the patient's immune system responded to the infection (antigenicity). These antigenic changes meant that the antibodies that developed by a person's immune response after exposure to this virus might not bind to the H5N1 viruses isolated from other H5N1 cases used in CDC's H5 serology test.

To ensure accurate serology results, CDC experts developed two reverse genetics (RG) viruses with the same antigenic changes that were documented in the A/Missouri/121/2024 HA sequence. The creation of those viruses took 3 weeks. CDC then used those RG viruses to test the blood specimens from the seven people from MO to look for an immune response to that specific virus, which would suggest exposure/infection.

The human population has complex pre-existing immunity to influenza viruses. In public health investigation of possible influenza infections, blood specimens taken at two different times are recommended for serological testing to assess for infection, one specimen collected within seven days of illness onset, and a second specimen collected after the patient has recovered at ≥ 21 days (ideally 21 to 28 days) after symptom onset. The rise in antibodies measured between the two specimens provides definitive evidence of an immune response to infection. In the Missouri investigation, only one specimen was available, which was collected weeks after the initial case was identified. This added another layer of complexity for serologic analysis in this situation given the retrospective nature of this investigation.

Laboratory Methods:

There is no commercially available H5N1 serology test since such testing does not currently have a clinical role in patient care. CDC does flu serology testing mainly to inform public health investigations and policy. Unlike clinically available serologies for other infections, non-clinical serologic testing for seasonal influenza viruses takes about a week. Serologic testing for H5 takes even longer because of the need for biosafety level 3 enhanced (BSL-3E) handling. Testing in this particular situation required additional steps to ensure valid results. Testing was conducted against four different target viruses: the two RG viruses developed for this investigation (described above), the H5N1 virus isolated from the first human case during this outbreak (posted as "A/Texas/37/2024"), and a seasonal A(H1N1)pdm09 virus. Both RG viruses developed by CDC were used in testing in order to provide additional assurance regarding the validity of the findings.

The following testing was conducted in multiple replicates:

• Microneutralization (MN) assay: The main serology test is called a microneutralization (MN) test. This is a highly sensitive and specific test for identifying influenza virus-specific antibodies in blood sera. The MN assay was performed to measure for neutralizing antibodies against the RG H5 viruses.
• Hemagglutinin inhibition (HI) assay: The HI assay tests for antibodies to the hemagglutinin of the specific influenza virus using red blood cells (RBC). Multiple types of RBCs were assessed in the HI assays to detect antibodies in the sera.
• The Multiplex Antibody Detection Assay (MIADA): a test which looks for antibody binding to 28 different influenza antigen targets (28-plex), it also measures the total immunoglobulin (Ig), IgG, IgM and IgA antibody responses.
• Seasonal Influenza Serum Absorption: an additional step that removes antibodies against seasonal influenza viruses that might be cross-reactive prior to testing against A(H5N1) viruses was also performed. This step helps to rule out the possibility of cross-reactivity from prior exposure to seasonal influenza viruses (through infection or vaccination) which might generate a false positive H5N1 serology result.

Serologic Findings and Interpretation:

CDC and historic WHO criteria for H5N1 seropositivity suggestive of infection requires two positive tests: a MN assay and either an HI or another immunologic assay, like MIADA (See End Note).

• The 5 health care workers who had exposure to the MO case patient were all seronegative using the MN, HI, and MIADA assays. This means they did not have any serologic evidence of past infection with H5N1 bird flu.
• The Missouri case and their household contact had some evidence of a humoral immune response to H5N1 bird flu virus suggestive of possible infection, but it was not detected across assays.
  • The MN assay detected neutralizing antibodies against the RG viruses in the index patient and their household contact. (Neutralizing antibodies are antibodies that defend a cell from a pathogen by disrupting its ability to infect a cell.)
  • The HI assay was negative in both the case and their household .
  • In the MIADA assay, neither the MO case or their household contact were positive.

Conclusion of Missouri Investigation:

The 5 health care workers were definitively seronegative against H5 in all assays, which means they were not infected with H5. This finding rules out person-to-person spread between the index MO case patient and any of five health care workers. Symptoms reported by these contacts were not caused by their exposure to this patient.

Testing results of the sera from the MO case patient and their household contact were similar: both showed evidence of an antibody immune response to H5 in only one assay (that detects H5 neutralizing antibodies), but not on the other serologic assays used to detect infection. The weak immune signal suggests that it is possible that both of these people may have been exposed to H5 bird flu despite the fact that they did not meet accepted thresholds for seropositivity. These similar immunologic results coupled with the epidemiologic data that these two individuals had identical symptom onset dates support a single common exposure to bird flu rather than person-to-person spread within the household. Intensive epidemiologic investigation has not identified an exposure to an animal or animal product exposure to explain these possible infections, and these serologic data cannot further elucidate the exposure leading to these possible infections.

End Note:

Historical World Health Organization serological case definition using single serum

Single convalescent serum, collected at ≥ 21 days after symptom onset (or last exposure to infected animals/persons), with a microneutralization antibody titer ≥ 1:40 to an A(H5) virus,

AND either

• 1) a positive result using a different serological assay (e.g., hemagglutination inhibition (HI) antibody titer≥40) or
• 2) an H5-specific positive result from another immunological assay such as ELISA or multiplex binding antibody assay.