Limitations of Testing Methodologies, Interpretation of Results and Algorithms for Confirming Positive/Negative Results Ken Henderson, Ph.D., Director Research Animal Diagnostic Services Charles River, Wilmington, MA [email protected]

Discussion Topics • General algorithms for screening and confirming positive findings • Variables impacting infectious agent detection during routine screening • Addressing positive findings, confirmation strategies, and associated limitations

The general algorithm for screening and confirmation is the same regardless of the infectious agent or the initial screening method

General “Detection → Confirmation” Algorithm Lab Animal Facility Routine screening: Samples or rodents If the positive result is an unexpected finding for an excluded agent, contact the testing lab to discuss: 1. Test results 2. Confirmation testing strategies Precautionary Steps Then Action

Diagnostic Laboratory Positive/equivocal finding

Confirmation testing If Positive

General Detection → Confirmation Algorithm Lab Animal Facility Routine screening: Samples or rodents

Diagnostic Laboratory

Screening versus Diagnosis • Screening: The examination, of a group of usually asymptomatic individuals, to determine if they meet a predetermined set of specifications • Diagnosis: The process of determining by examination the nature and circumstances of a clinical sign or abnormality Today – Few prevalent rodent infectious agents are first detected by observing clinical signs – Most laboratory methods focus on screening techniques used to detect a specific list of excluded agents

Transmission to Bedding Sentinels • Both observations and publications support that many commonly excluded agents may not be efficiently transferred to bedding sentinels – The agent may not remain viable outside the host – The quantity of agent transferred may be too low or below an infectious dose – The sentinel may not be susceptible (too old/less susceptible strain) – Agent may replicate slowly = late seroconversion or agent may be below limit of detection in sample – Prevalence on the racks may be low* – Bedding transfer amount/technique not optimal

Prevalent agents for which transmission/detection failure is often observed or reported • • • • • • • • • •

Murine norovirus Mouse parvovirus Helicobacter spp. P. pneumotropica M. pulmonis S. aureus Pneumocystis spp. Fur mites Pinworms Anecdotal evidence for several others

Successful testing is dependent on our knowledge of the excluded agent • When is the agent present – Does it persist or not? – What age groups or what time post-exposure are you likely to detect the agent?

• • • •

Where is it present on the host – what samples are best? What is the replication rate? When can seroconversion be expected? Are some rodent strains or age groups more susceptible than others? • Does the agent efficiently transmit to bedding sentinels?

What happens between mouse/sample submission and positive or equivocal findings?

Laboratory flow for Initial and secondary screening of rodents and samples for infectious agents Necropsy • Gross Pathology • Parasitology • Sample Collection for Screening Assays

Tissues collected if abnormalities found

Histopathology

Samples

Serum

Samples PCR

Bacteriology

Serology

= Initial screening = Follow-up testing

Methods used for primary screen and follow-up testing Screening Method Primary Serology MFIA/MFI ELISA LDH testing for LDV Parasitology Macroscopic and microscopic exam

Confirmation Secondary IFA PCR Seeing is believing -Second set of eyes to confirm

Fecal centrifugation concentration Bacteriology Colony and organism Biochemical ID morphology PCR- Select organisms

PCR

Screening Assays

Repeat screening and or reisolate NA Species specific assays Secondary targets

Tertiary HAI Western Blot

Sequencing

Sequencing

Today’s screening assays are more comprehensive and use multiple controls • Multiplexed bead-based serology – – – –

Multiple protein antigens/agent Positive (high and low) and negative controls Absent conjugate controls Anti-rodent immunoglobulin control

• PCR – – – – –

Multiple targets (signatures)/agent Quantified positive and negative controls Size markers for gel-based PCR to estimate product size Real-time Ct values estimate copy numbers OD reading/nucleic acid recovery controls/endogenous gene controls – PCR inhibition controls

Example: PCR inhibition control (384-well)

General Detection → Confirmation Algorithm Lab Animal Facility Routine screening: Samples or rodents

Diagnostic Laboratory Positive/equivocal finding

General Detection → Confirmation Algorithm Lab Animal Facility Routine screening: Samples or rodents If the positive result is an unexpected finding for an excluded agent, contact the testing lab to discuss: 1. Test results

Diagnostic Laboratory Positive/equivocal finding

Investigate unexpected positive/equivocal findings Immediately Contact the testing laboratory - Why? • No assay or screening procedure is 100% sensitive and specific – Laboratories take great precautions to limit false-positive and negative results, but they WILL occur.

• Although assays/processes are qualified or validated, there likely hundreds of steps from mouse shipment to receipt of the final report that can go wrong – Many mistakes can lead to false-positive results • Human errors will continue to occur regardless of all the controls used by the testing laboratory

By immediately contacting the laboratory you can… • Discuss the results or findings for primary and secondary tests – How positive/equivocal? - How confident are they that it is real? – Review photo documentation – How did the controls look? – Any obvious errors – Potential for assay-specific artifacts? • Determine if there are any additional tests the lab can perform on the positive sample – If not already done, can they repeat the test? • Ask that the sample be submitted to a second lab • Performing the above may reverse the call

General “Detection → Confirmation” Algorithm Lab Animal Facility Routine screening: Samples or rodents If the positive result is an unexpected finding for an excluded agent, contact the testing lab to discuss: 1. Test results 2. Confirmation testing strategies Precautionary Steps

Diagnostic Laboratory Positive/equivocal finding

Confirmation testing

If the initial positive still appears to be real, you should still obtain confirmation that the agent is present There remains many reasons why the initial positive findings could be false so confirmation should be performed before taking any expensive action

Taking precautions while you are waiting for confirmation results • Biosecurity precautions should be put in place to limit animal and staff movement • The reality is routine health monitoring is not an early detection system. If the positive is real, the agent has likely been in your facilities for months or longer.

Confirmation testing strategy • No different from the initial screening process, successful confirmation testing strategy depends on the knowing the agent – The knowledge base for infectious agents are constantly evolving as well as sampling and detection methods

• Your diagnostic laboratory is a good resource for helping you plan a confirmation testing strategy – You are likely not the first to receive a positive for the agent of concern – Laboratories have ongoing experience with what works, what “kind of” works, and what does not.

The first step – Test remaining or subsequent sentinel mice • When testing additional sentinels to confirm, consider all possible methods of detection when collecting samples • You can either submit for all testing methods at once or focus first on gold standard methods/or high confidence methods first and hold samples for secondary testing. – Examples: • For MHV and rodent parvoviruses, collect serum and mesenteric lymph nodes. • For bacteria, collect samples for culture and PCR

Subsequent sentinel testing to confirm positives has challenges • If you are using microisolation caging, the agent is not likely to easily spread to other cages so infection process may be limited to one simultaneous infection cycle OR, the agent by nature may not transfer well to bedding sentinels so initial “+” could have been a lucky finding • As a result, subsequent sentinels may not receive bedding containing a viable dose of the agent capable of causing infection

Example Scenario for MPV: Considerations for Confirmatory Submissions

Log10 Copies per mg Sample

8 7 6 5 4 3 2 1 0

BALB/c C57BL/6

7

MLN mRNA

MPV-1e Infection and Fecal Shedding in BALB/c and C57BL/6 Mice

MLN DNA

6 5 4 3 2 1 7

Feces DNA

6 5 4 3 2 1 123 5 7 10

14

21

28

42

Days Post Inoculation

56

Timing is not in your favor: Hypothetical infection time line for MPV = Sentinel Placement

Fecal Shedding

Index Cage

-8

= Infectious Shedding Period

Bedding Sentinels

-4

0

4

8

Weeks

12

16

20

24

Timing is not in your favor: Hypothetical infection time line for MPV = Sentinel Placement

Fecal Shedding

Index Cage

-8

= Infectious Shedding Period = Susceptible Sentinel

Bedding Sentinels

-4

0

4

8

Weeks

12

16

20

24

Other challenges for confirmation of MPV in subsequent sentinels • Infectious dose MPV-1e ~ 1000 particles for BALB/c mice (~1.5 log10 higher for C57B/6) • Virus is diluted in bedding sentinel cage by negative cage contributions What if you use open-top cages? • MPV is most likely throughout the room so confirmation testing should not be challenging – If you are breeding in the room, consider testing 8-12 week old mice by serology and PCR – Environmental swabs of room exhaust ducts

What if the current or subsequent sentinels are negative? Hold your breath? • If subsequent sentinels are negative, maybe it was only a cage or two and has burned out. • Increase surveillance – Increase serology only for MPV and bleed sentinel every 4 weeks or use overlapping sentinels mice every 4 weeks. Hold lymph nodes/spleen for confirmation – Test culled research mice to understand prevalence and impact for researchers – May be wrong strategy if breeding

OR

Confirmation by testing study rodents - How far do you go? • General concerns for MPV – Not all mice within a positive cage may be infected – Mice on the racks may have been moved since infection weeks or months before initial positive was detected

Approaches for MPV testing of study rodents • Serum collection – Very successful… if all animals are bled. Might fail if only a representative mouse from each cage is bled. – Animal stress and Technician stress – very labor intensive

• PCR on feces – Used successfully by some, but detection may be hit or miss because of waning fecal titers. – If done should be performed immediately after initial sentinel mouse is determined to be positive. – Expensive! Pooling among cages decreases cost, but also decreases sensitivity

• Another route: If looking to confirm try PCR swab of rack exhaust filter or horizontal plenum

If confirmation testing does not provide support for initial finding… • Increase surveillance – Increase serology only for MPV and bleed sentinel every 4 weeks or use overlapping sentinels mice every 4 weeks. Hold lymph nodes/spleen for confirmation – Test culled research mice to understand impact to researchers and prevalence

• Consider continuing precautionary steps – Mouse lines moved to other buildings should be tested – Designated staff or last room maintained in the day – Etc.

General “Detection → Confirmation” Algorithm Lab Animal Facility Routine screening: Samples or rodents If the positive result is an unexpected finding for an excluded agent, contact the testing lab to discuss: 1. Test results 2. Confirmation testing strategies Precautionary Steps Then Action

Diagnostic Laboratory Positive/equivocal finding

Confirmation testing If Positive

Questions?