Virology

Direct detection methods.

SARS-CoV-2 can be cultured from clinical specimens using routine methods and forms visible plaques in VeroE6 and Vero CCL81 cells within 2 to 3 days of inoculation. While useful in studying the basic biology of the virus, viral culture requires biosafety level-3 (BSL-3) safety precautions and is not recommended for routine diagnosis. Further, the sensitivity of culture methods may be inferior to NAATs as indicated by an inability to recover virus from respiratory specimens with less than 10 ⁴ RNA copies/mL.

RADTs are an inexpensive and direct method to detect the presence of SARS-CoV-2 in clinical specimens. These tests are simple-to-use lateral flow assays that can be read either visually, based on the appearance of colored indicator lines, or automatically, using a simple device to detect a fluorescent signal. At the time of writing, five such assays have received emergency use authorization (EUA) by the FDA. A major benefit of these tests is the ability to deploy them outside of a high-complexity laboratory, at or near the POC to rapidly identify persons with SARS-CoV-2 infection. Importantly, these tests have generally been approved for use in symptomatic patients within 1 week of symptom onset when viral load (VL) is the highest to maximize sensitivity. An independent clinical evaluation of two RADTs (BD Veritor, Becton, Dickinson and Co., Sparks, MD and Sofia 2, Quidel Corp., San Diego, CA) demonstrated approximately 80 to 85% sensitivity compared to a PCR-based test for SARS-CoV-2 in nasal specimens collected from patients less than 7 days from symptom onset. Importantly, the performance of these tests has not been evaluated when used as a screening test in asymptomatic populations where VL (and thus sensitivity) may be lower. A commercially available RADT without EUA designation found the test to be 3 log ₁₀ less sensitive than viral culture and 5 log ₁₀ less sensitive than PCR in analytic limit of detection (LoD) analysis. Because of potential shortcomings in sensitivity, the CDC and WHO recommend that negative results in symptomatic patients be considered presumptive and confirmed by a PCR-based test to rule out infection. ^, Conversely, if used to screen asymptomatic individuals (an off-label application), positive results may require PCR confirmation in low-incidence populations.

NAATs provide the most sensitive method for direct detection of SARS-CoV-2 in clinical specimens and are considered the “gold standard” method for detection of SARS-CoV-2 and diagnosis of Coronavirus Infectious Disease 2019 (COVID-19). Approximately 6 months following the first reported transmission in the United States, greater than 200 different NAATs were made commercially available through the FDA EUA regulatory pathway. This includes high-complexity batched testing platforms, moderate complexity sample-to-answer platforms, and “rapid” POC tests. The absolute sensitivity of these NAATs is difficult to ascertain due to lack of true “gold standard” and is likely influenced by multiple factors including specimen type, collection method, and test design (e.g., genomic target, detection chemistry, etc.). The majority of available tests are designed to detect two or three distinct genomic targets specific to SARS-CoV-2 and/or are conserved among the subgenus Sarbecovirus which includes 2003 SARS-CoV and other nonhuman SARS-CoVs. The goal of utilizing multiple assay targets is to increase sensitivity and provide redundancy to insulate against the impact of potential viral mutation events on test performance. However, in instances for which solely pan- Sarbecovirus targets are detected, results are typically considered preliminarily positive given the possibility of cross-reactivity with other viruses in the family. Multiple split-sample studies have compared the performance of available laboratory-based RT-PCR tests and report 96 to 100% positive agreement (i.e., sensitivity) among these tests based on consensus result. A rapid isothermal amplification test has been designed for use at the POC and delivers a result in as little as 5 minutes; however, reduced sensitivity of 75 to 90% has been reported compared to “traditional” laboratory-based RT-PCR assays. ^{, ,} This difference in sensitivity is likely due to a difference in the LoD between laboratory-based RT-PCR tests (39 to 779 copies/mL) and the POC test (3000 to 20,000 copies/mL). ^{, ,}

The sensitivity of NAATs can also be influenced by the type of specimen analyzed and when in the course of infection the specimen is collected. Among upper respiratory specimens, nasopharyngeal, throat, saliva, and anterior nares have all been evaluated. A limited number of studies have directly compared the sensitivity of viral detection in each specimen type. Results appear to support similar statistical performance among most specimen types with slightly higher sensitivity observed in NP than throat. ^{, ,} Viral RNA may be detected as early as 1 week prior to symptom onset, but typically peaks at or near the time of recognized symptoms. Following symptom onset, the VL begins to decrease in all respiratory specimens but remains detectable in lower respiratory specimens such as sputum or BAL significantly longer than upper respiratory specimens such as nasal or nasopharyngeal swabs. ^, In addition, positive BAL specimens have been reported in symptomatic and immunocompromised patients with negative NP or throat results. Therefore consideration may be given to testing a lower respiratory specimen in patients with compatible symptoms and a negative NP or nares result. Importantly, persistence in the lower respiratory tract is frequently at less than 10 ⁴ copies/mL, which as mentioned above correlates with a failure to recover virus in culture. The significance of long-term, low-level persistence as it related to risk of relapse or transmission is currently unclear.

Indirect detection methods.

The use of serologic methods to indirectly detect individuals who are currently infected or have been exposed to SARS-CoV-2 has been widely explored. Multiple technologies including highly automated laboratory-based IAs and POC “rapid” lateral flow tests are commercially available and have obtained EUA status from the US FDA ( https://www.fda.gov/medical-devices/emergency-situations-medical-devices/eua-authorized-serology-test-performance , accessed June 29, 2020). These assays most commonly utilize purified viral nucleocapsid or spike proteins to detect the presence of one or more antibody classes (e.g., IgG, IgM, IgA, total antibody). The sensitivity of these assays ranges from approximately 75% to greater than 99%, with reported specificities of approximately 90% to greater than 99%. The sensitivity of serologic tests is dependent on the time from exposure to the virus, known as the “diagnostic window.” For SARS-CoV-2, the development of specific IgM and IgG class antibodies appears to occur nearly simultaneously, with 95% of individuals generating a detectable antibody response by 14 days postexposure, and essentially 100% by 21 days postexposure. ^{, ,} This response reaches a plateau within 1 week of detection and then both IgG and IgM titers begin to decline. The decline in IgG titer is less abrupt than IgM; however, at this point it is not clear how long antibody remains detectable after initial exposure. Cross-reactivity with the four “human coronaviruses” appears to be negligible for most commercially available assays, though this may vary based on assay design.

The utility and interpretation of serologic results remains in question. As of time of writing, the seroprevalence in much of the world is likely 1 to 5%. This low prevalence renders even the most specific serologic assays a PPV as low as 80%. Further, the extent and duration of protective immunity correlated with a positive serologic result has not been established. Therefore the current utility of serologic assays lies primarily in the epidemiologic assessment of exposure to SARS-CoV-2 in various populations, which may be used to guide RT-PCR based screening strategies, and in retrospective diagnosis of late-stage infections (>2 weeks after symptom onset) when direct detection tests including RT-PCR may have converted to negative.

Serologic tests for CMV.

Serologic tests hold no utility for the diagnosis or monitoring of CMV disease. The titer of anti-CMV IgG is not correlated with active disease, and anti-CMV IgM is commonly associated with episodes of asymptomatic viral reactivation. The chief utility of CMV serology is to establish the serostatus of two specific populations: women, as part of a prenatal or perinatal screen, and potential donors and recipients prior to SOT or HSCT. The presence of CMV-specific IgG and IgM is determined using enzyme immunoassays (EIAs), chemiluminescent immunoassays (CIA), or indirect immunofluorescence assays (IFAs). For transplant patients determination of serostatus aids in stratifying risk and severity of potential CMV disease, and also guides dose and duration of antiviral therapy following transplant. In one study, donor positive/recipient negative (D+/R−) transplant carried a 19 to 31% risk of subsequent CMV disease compared with only 2 to 3% risk in D-/R− transplants. Extension of antiviral prophylaxis in D+/R− organ recipients may reduce the incidence of late onset CMV disease.

Beyond establishment of serostatus, differentiation between primary and past infection holds prognostic implications for the risk of congenital CMV disease. The presence of anti-CMV IgM should not be independently interpreted as evidence of primary infection due to long-term persistence (6 to 9 months) following primary infection, presence during episodes of viral reactivation, and cross-reactivity with other Herpesviridae ; however, absence of IgM effectively rules out recent infection. ^, Anti-CMV IgG avidity testing can be used to distinguish primary from past infection in women with dual positive IgG and IgM results. The presence of high avidity IgG is an indication that primary infection occurred 18 to 20 weeks prior to testing, that is, past infection and is associated with a 1 to 2% risk of congenital CMV infection compared with a risk of 12 to 75% in women with low avidity IgG. ^, Despite these prognostic implications, routine serologic screening of pregnant women for CMV is not currently recommended due to the remaining potential for congenital infection, regardless of maternal immune status, and lack of therapeutic intervention to prevent congenital transmission. IgG avidity testing has also been used as a prognostic tool in post-transplant patients. In this group, failure to develop high avidity antibodies following primary infection may be associated with increased risk for severe CMV infection or organ rejection. ^, Several ELISA-based CMV IgG avidity tests are commercially available; however, none have received FDA clearance to date. Laboratories that offer testing must conduct internal assay validation studies and may establish different thresholds to define high and low avidity results which contributes to variability in analytic and prognostic performance of these assays. Therefore results must be interpreted in the context of other laboratory and clinical findings.

Qualitative NAATs for CMV.

Qualitative NAATs are used to aid in diagnosis of localized infections such as pneumonia, retinitis, meningitis, gastrointestinal, or other “end organ disease.” Potential specimens include tissue biopsy, bronchoalveolar lavage (BAL), cerebrospinal fluid (CSF), and vitreous fluid. Direct detection of CMV in these specimens is important because 50 and 70% of patients with end organ disease have undetectable CMV in whole blood (WB) or plasma specimens. The high sensitivity and negative predictive value (NPV) of qualitative NAATs can be used to essentially rule out active CMV infection in cases of presumed CMV pneumonitis, retinitis, or gastrointestinal disease. Conversely, a positive result should not be used to define “proven infection” due to an inability to differentiate asymptomatic shedding from end organ disease. While higher VLs have been associated with increased specificity for disease, specimen heterogeneity and a lack of assay standardization has precluded the development of reliable quantitative thresholds to define active infection. Therefore all positive results obtained from fluids and tissues should be confirmed by a more specific method such as histologic examination and must be correlated with clinical symptoms to support a probable diagnosis of CMV disease.

There are currently no FDA-approved qualitative CMV NAATs, but several tests are commercially available as analyte specific reagents (ASRs) or have received CE marking for use in the European Union ( Table 89.1 ). An evaluation of five of these NAATs, as well as an LDT, was conducted using 200 prospectively collected clinical samples representing respiratory, urine, CSF, biopsy tissue, and other clinical specimens. The analytic LoD ranged from 10 ² to 10 ³ copies/mL for all six tests. The clinical sensitivity of 5 of 6 assays was 100%, with one of the assays demonstrating only 89% (41/46) sensitivity for detection of CMV in the clinical specimens. These data support the utility of qualitative NAATs as a high-sensitivity screen for the presence of CMV in various specimen types and an effective method to rule out CMV as the etiologic agent in these cases.

Quantitative NAATs for CMV.

An important factor in predicting and monitoring CMV disease is the change in blood or plasma VL. Rapid changes in VL are prognostic for stratifying risk of disease and severity of symptoms. ^, Results of VL tests also impact decisions to initiate or discontinue therapy and aid in early recognition of antiviral resistance. Therefore the use of quantitative NAAT is recommended by several national and international guidelines as an important component in patient management. ^,

The type of specimen tested impacts test sensitivity and the quantitative CMV VL value obtained, which in turn affects the interpretation of results. Assays using WB quantify latent, cell-associated virus in addition to infective extracellular virions. Therefore WB-based assays are more sensitive for the qualitative detection of CMV, and quantitative VL values are approximately 0.8 to 1.2 log higher than matched plasma specimens, albeit with a poor correlation coefficient ( r value, 0.19 to 0.79). Importantly, the increased sensitivity of WB assays has not demonstrated a prognostic advantage for determination of CMV disease compared to plasma-based tests. ^, The rate of virologic recurrence was similar in patients with undetectable VL in matched WB and plasma (23.6%) as it was in patients with a quantifiable VL in WB and CMV negative plasma (23.1%), owing to the detection of latent inactive virus in WB specimens.

Increasing analytical sensitivity of quantitative NAATs has lowered the threshold of detection to as few as 6 to 150 copies/mL in WB or plasma specimens. ^, At these low levels of detection, the presence of cell-associated virus or free viral DNA may contribute to a detectable VL, even following effective antiviral treatment. This is supported by the finding of detectable VL in both WB (∼70%) and plasma (∼48 to 59%) specimens in asymptomatic patients. ^, These data suggest that even when using plasma, a low VL does not indicate a risk of CMV disease. Further, the inability to achieve an undetectable VL can result in unnecessary antiviral therapy and increase the risk of resistance if a “treat to negative” paradigm is used. ^,

Absolute quantitative VL thresholds ranging from 1000 to 10,000 CMV copies/mL have been evaluated as clinical decision points to define active infection and drive initiation or discontinuation of antiviral therapy. ^{, ,} Unfortunately, the establishment of a single reliable threshold to predict disease has been hindered by a lack of standardization among commercially available and laboratory developed tests, including the use of different calibration materials, which causes interassay variability of 0.5 to 2.0 log ₁₀ copies/mL.

A significant step toward standardization of CMV VL testing was made with the introduction of an international standard by the World Health Organization (WHO) in 2010. This allows laboratories to calibrate VL results to a single international unit (IU) standard regardless of which test is being used, thereby enabling a more accurate comparison of viremia values across institutions. Standardized IVD assays can achieve precision within as little as 0.1 log ₁₀ copies/mL, which may be less variation than what is observed biologically during chronic infection. Specifically, a multicenter study of an FDA-cleared plasma-based NAAT calibrated to the IU standard demonstrated a narrow 95% confidence interval (CI) of 0.14 to 0.17 log ₁₀ copies/mL for specimens tested across five different laboratories. However, a comparison of ten different CMV NAATs, all calibrated to IU, demonstrated a variance of up to 2.8 log ₁₀ IU/mL among clinical specimens analyzed by each assay, with over 40% of specimens giving values greater than 0.5 log ₁₀ IU/mL from the mean. Further, the use of IU will not allow for interchangeable comparison of values obtained from different specimen types such WB and plasma. These data demonstrate that even when using a standardized calibrator and reporting results as IU/mL, assay-specific factors including the use of different specimen types, CMV gene targets, cycling conditions, and nucleic acid extraction methods can cause significant inter-assay variability. Taken together, these data underscore the difficulty in obtaining commutable VL results across different laboratories and the barriers to development of a “global” CMV VL threshold for use as a clinical decision point. Until these variables can be addressed, serial VL monitoring for preemptive therapy or identification of therapy failure should be conducted by the same laboratory to avoid potentially misleading results.

An alternative approach to absolute VL thresholds is the observation of change in VL values over time. Several studies have demonstrated that a significant or rapid change in VL is a better prognostic indicator than a single value, especially when the VL is relatively low (10 ² to 10 ³ copies/mL) where higher variability in assay results are expected. ^{, ,} CMV replication kinetics or “doubling time” (T _d ) has been used to stratify the risk of disease and necessity of antiviral therapy following transplant. ^, No definitive T _d threshold exists; however, one study found that use of a T _d less than 2 days as criteria for initiation of therapy was associated with significantly fewer days of viremia and antiviral therapy compared with the use of a VL threshold of 1000 copies/mL. While effective, this approach requires frequent (2- to 3-day interval) collection of specimens which may be inconvenient or impractical for outpatients. Importantly, once antiviral therapy has been initiated serial CMV VL tests should be conducted no more frequently than once per 5 to 7 days since the half-life of CMV DNA in blood is 3 to 8 days. Further, an increase in VL can be expected within 72 h of initiation of antiviral therapy in some patients and is not indicative of therapy failure.

Current assays that have attained FDA and CE regulatory status for IVD use include the Qiagen artus CMV RGQ MDx, the Roche COBAS AmpliPrep/COBAS TaqMan CMV test, and the Roche COBAS CMV test (run on cobas 6800 or 8800 systems). These assays all contain internal quantitative controls calibrated to the IU standard and are indicated for use only with plasma specimens. Each system includes automated nucleic acid extraction followed by RT-PCR amplification and detection of target DNA. The assays have an LoD of 34 to 91 IU/mL and a linear range of ∼10 ² to 10 ⁷ IU/mL. Additional quantitative CMV assays that have attained CE Mark or are commercially available as ASRs for the development of LDTs are available for use on various real-time PCR platforms and demonstrate similar limits of detection and quantification to FDA-cleared tests. However, the use of different extraction methods, thermocyclers, and calibration materials can contribute to variability of VL results. Therefore regardless of assay, it is important to establish a baseline VL for patients entering a health care system or for those with VL results received from a different laboratory.

Antiviral resistance testing in CMV.

Antiviral treatment or prophylaxis plays an integral role in the management of patients at risk for CMV infection and disease. The gold standard method to assess antiviral susceptibility involves observation of viral plaques or CPE in cultured cell lines in the presence of increasing concentration of an antiviral agent. This phenotypic method has the advantage of determining a specific inhibitory concentration for an antiviral and is independent of specific resistance mechanisms. The major drawbacks to this approach are the technical expertise required to carry out viral culture, the requirement for isolated virus, and the incubation time necessary to observe CPE, which can be as long as 4 weeks. In contrast, genotypic methods rely on the detection of specific mutations in the viral genome that are associated with antiviral resistance. These methods directly analyze virus present in clinical specimens (i.e., do not require isolation of the virus) and can be completed in as little as 1 to 2 days; however, the accuracy and reliability of results is dependent on knowledge of specific mutations leading to resistance. To date, the majority of mutations resulting in resistance to commonly used antiviral agents ganciclovir, cidofovir, and foscarnet map to specific regions of two CMV genes: UL97, encoding a tyrosine kinase required for activation of ganciclovir, and UL54, the viral DNA polymerase. Molecular assays using PCR to amplify selected regions of the UL97 and UL54 genes, followed by Sanger sequence analysis, have been developed by clinical and reference laboratories and are widely used to identify resistant CMV in clinical specimens. While mutations in UL97 are most common and result in resistance to ganciclovir only, CMV may also develop mutations in UL54 that result in resistance to one or more antiviral agents. Therefore analysis of both UL97 and UL54 should be conducted when patients fail therapy. ^,

Serologic tests for VZV.

Serologic diagnosis of primary varicella is not routinely performed because it requires comparison of acute and convalescent sera, which delays diagnosis by 10 to 14 days. The establishment of varicella serostatus is however beneficial for screening of health care workers and pretransplant assessment of individuals with no record of vaccination or natural varicella infection. Common methods to assess serostatus include latex agglutination (LA), ELISA, fluorescent antibody to membrane antigen (FAMA), and CIA ( Table 89.2 ).

The FAMA test is based on detection of varicella envelope-specific antibodies in cultured virus. FAMA has been considered the gold standard for the establishment of serostatus because of its high sensitivity and correlation with protective immunity; however, the assay is technically demanding and time consuming. These factors have prevented widespread use of this method in clinical laboratories. LA tests are inexpensive, require no additional equipment, are simple to perform, and can be completed within 10 to 15 minutes. The sensitivity of LA tests (89 to 98%) is similar to FAMA and equivalent or superior to that of traditional ELISA but specificity is low when compared to FAMA. ^{, ,} This may be due to the subjective interpretation of agglutination reactions. Newer ELISA and CIA tests based on purified envelope glycoprotein (gpELISA) have demonstrated better sensitivity (87 to 100%) than traditional ELISA tests (72 to 98%). ^, Additional benefits of ELISA and CIA tests include an objective result and the ability to automate testing, which enables high-throughput screening of sera.

Establishment of serostatus in immunized persons can be difficult because the antibody response to the vOka vaccine is 10-fold lower than the response to natural infection. Additionally, vaccine induced antibodies begin to decline and may become undetectable in 5 to 30% of individuals within 5 to 15 years post vaccination, resulting in 61 to 85% sensitivity of FAMA and LA in immunized populations. ^{, ,} New gpELISA tests demonstrate better performance in detecting seroconversion following immunization (87 to 99% sensitive). However, long-term follow-up studies have not been conducted to determine if sensitivity remains high despite declining antibody concentration.

Qualitative NAATs for VZV.

NAATs are the preferred method for laboratory diagnosis of acute primary varicella and secondary HZ disease because of the increased speed and sensitivity compared with culture or direct detection (e.g., direct fluorescent antibody [DFA]) methods. In patients with rash consistent with HZ, culture was found to be 20 to 53% sensitive and DFA was found to be 82% sensitive when compared to NAAT. ^, VZV NAATs may also be multiplexed to include HSV-1 and HSV-2. This can be valuable in the assessment of cutaneous and mucocutaneous lesions given the similarity in appearance. ^, Specifically, one study using a multiplexed VZV and HSV NAAT reported 11% of all positive VZV detections to be from male or female genital sites. In these cases, a VZV-specific test may not have been ordered based on clinical presentation and location of the lesion, supporting the value of combined target NAATs.

Qualitative NAATs also provide a sensitive method to detect VZV in sterile specimens such as CSF, BAL, or vitreous fluid where the presence of any amount of virus is likely causal. Pulmonary varicella or HZ can be severe in adults and if untreated carries up to 30% mortality. Supportive therapy and early treatment with acyclovir can greatly reduce mortality, however the clinical diagnosis of VZV pneumonitis is difficult because physical and radiographic findings are non-specific. ^{, ,} Direct analysis of BAL specimens using NAATs has enabled rapid and definitive detection of VZV in cases of pneumonitis, resulting in early appropriate therapy to improve patient outcome. ^, Similarly, clinical symptoms of necrotizing retinitis are nonspecific among the various herpesviruses commonly associated with this condition (VZV, CMV, HSV). ^, VZV NAATs have been successfully used to analyze vitreous specimens and provide a rapid and definitive diagnosis. The use of aqueous humor rather than vitreous is less invasive, requires as little as 10 to 20 μL fluid, and also appears to be acceptable for laboratory diagnosis of viral retinitis.

Several FDA or CE-Mark qualitative VZV NAATs are commercially available in addition to analyte-specific reagents and published primer sequences ( Table 89.3 ). FDA and CE-Mark assays are often cleared for specific specimen types, typically cutaneous lesions, and require additional verification studies to enable reporting of results on alternative specimen types.

Importantly, up to 5% of individuals receiving the vOka vaccine may develop a characteristic rash. ^, Discrimination between vaccine-induced lesions and acute infection with wild-type VZV in these patients may impact prognosis and infection prevention strategies. Importantly, none of the currently marketed assays differentiate wild-type VZV from vOka. Differentiation has been reliably achieved using additional primers that target a specific polymorphism in ORF 62 and other vaccine-specific SNPs. Currently, this testing is available through the US CDC and other specialized reference laboratories.

It is difficult to assess the absolute sensitivity of NAATs for detection of VZV in various specimens because “gold standard” reference methods including culture are significantly less sensitive than NAAT. Analytic studies typically report a lower LoD of 10 to 200 copies/mL; however, a lack of standardization of methods among LDTs makes the clinical comparison of assays difficult. For example, the initial description of a VZV LDT reported 94% sensitivity when compared to a composite method of culture, DFA, and serology. A subsequent study reported the LDT to be only ∼60% sensitive compared to a commercially available VZV NAAT. This highlights the importance of standardized test methods and the impact of the chosen gold standard comparator when reviewing the performance or comparative performance of a molecular assay(s).

NAATs for HHV-6.

Both qualitative and quantitative NAATs have been employed for the detection of HHV-6 in clinical specimens using various targets and methodologies. There are currently no FDA-cleared NAATs for detection of HHV-6 in serum, but CE-Marked assays are commercially available and several reference laboratories offer quantitative or qualitative testing ( Table 89.4 ).

A multiplexed panel for qualitative detection of HHV-6 in CSF has recently been cleared by the FDA for IVD use (see meningitis section of this chapter). The specific syndrome or site of infection dictates the specimen most appropriate for analysis by NAAT. These include CSF, BAL fluid, and blood/serum (when disseminated disease suspected).

Quantitative NAATs are the preferred method for laboratory diagnosis of HHV-6 reactivation disease in post-transplant or other at-risk populations. Whole blood, isolated PBMCs, or serum can be used to monitor HHV-6 VL. HHV-6 DNA has been detected in 30 to 90% of peripheral blood or PBMC samples obtained from asymptomatic individuals, frequently at levels of 10 ³ copies/mL or higher. ^, This complicates interpretation of a single VL result in a patient with compatible symptoms. A rapid increase of 3 to 4 log ₁₀ in HHV-6 VL, reaching 10 ⁵ to 10 ⁶ copies/10 ⁶ PBMC, has been observed between 0 and 14 days following onset of symptoms. ^, The delayed rise in VL precludes the use of post-transplant serial monitoring to predict HHV-6 reactivation disease; however, a rise in HHV-6 VL can be useful in confirming the involvement of HHV-6 when other viral etiologies (CMV, VZV) may also be in the differential. Additionally, VL is a useful prognostic marker since high VL is correlated with delayed tissue engraftment, and VL decreases following antiviral therapy. In contrast to peripheral blood or PBMCs, HHV-6 DNA is only rarely detected in sera of asymptomatic individuals and likely represents viral DNA released from lysed PBMCs rather than active viral replication. ^, This suggests the detection of HHV-6 DNA in serum may be more specific for active disease compared to peripheral blood, though sensitivity may be reduced.

The lack of assay standardization and an international calibrator complicate interpretation of quantitative VL results and impede the establishment of a universal threshold for clinical significance. A greater than 15-fold variability between HHV-6 assays at the upper end of quantitation and greater than 200-fold variability in VL results for reference specimens containing ∼3 log ₁₀ copies/mL has been reported. Recent development of the first WHO International Standard for HHV-6 has the potential to reduce interlaboratory and interassay variation in reported VL values; however, differences in assay target, extraction method, and PCR instrumentation still contribute to variation in VL values obtained from different tests. Therefore it is recommended to use the same test and laboratory when monitoring HHV-6 VL with importance being placed on the change in VL rather than the absolute value.