Clinical utility of cycle threshold values in the context of COVID-19

Background The ability to predict likely prognosis and infectiousness for patients with COVID-19 would aid patient management decisions. Diagnosis is usually via real-time PCR and it is unclear whether the semi-quantitative capability of this method, determining viral load through cycle threshold (Ct) values, can be leveraged. Objectives We aim to review available knowledge on correlations between SARS-COV-2 Ct values and patient- or healthcare-related outcomes to determine whether Ct values provide useful clinical information. Sources A PubMed search was conducted on 1st June 2020 based on a search strategy of (Ct value OR viral load) AND SARS-CoV-2. Data was extracted from studies reporting on the presence or absence of an association between Ct values, or viral loads determined via Ct value, and clinical outcomes. Data from 18 studies were relevant for inclusion. One study reported on the correlation between Ct values and mortality and one study reported on the correlation between Ct values and progression to severe disease; both reported a signicant association (p < 0.001 and p = 0.008, respectively). Fourteen studies reported on the correlation between Ct value or viral loads determined via Ct value and disease severity and an association was observed in 8 (57%) studies. Studies reporting on the correlation of viral load with biochemical and haematological markers showed an association with at least one marker, including increased lactate dehydrogenase (n = 4), decreased lymphocytes (n = 3) and increased high-sensitivity troponin I (n = 2). Two studies reporting on the correlation with infectivity showed that lower Ct values were associated with higher viral culture positivity. Data suggest Ct values may and that Ct values may be useful in predicting the and prognosis of with COVID-19; however, further studies are warranted to conrm clinical value. one used used both The real-time PCR targets varied between included N, E genes and the 5’ untranslated genome region. Six studies analysed SARS-CoV-2 Ct values at multiple time points for each 10,11,18-21 and seven studies determined Ct values at hospital admission or diagnosis. 12-17,22 Fourteen studies reported on the direct correlation of outcomes with Ct values. 10-14,16,17,20,22-27 Three studies reported on the correlation of outcomes with viral load, determined using standard curves of Ct values versus RNA copy number, 15,19,21 and one study correlated outcomes with the inverse of Ct values, taken as a proxy for viral load. inorganic


Introduction
Patients infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) display disparate disease severity, ranging from an absence of symptoms to requiring intensive care and fatal outcomes. 1 Therefore, the ability to predict the likely prognosis and infectiousness of patients at diagnosis would greatly aid treatment and patient management decisions.
The standard molecular method for coronavirus disease 2019 (COVID-19) diagnosis is via real-time reverse transcription polymerase chain reaction (RT-PCR). 2 Real-time RT-PCR cycle threshold (Ct) values represent the number of ampli cation cycles required for the target gene to exceed a threshold level. Ct values are therefore inversely related to viral load and can provide an indirect method of quantifying the copy number of viral RNA in the sample; however, the use of Ct values as a proxy of viral load is in uenced by the assay itself (correlation would stand in the linear dynamic range of the speci c RT-PCR assay used) and factors within the sample matrix that can affect ampli cation e ciency. 3 It has previously been suggested that the viral load of SARS-CoV-2 may be an important factor in determining both disease severity and likelihood of transmission. [4][5][6] Although there are many differences between the current SARS-CoV-2 pandemic compared to the SARS-CoV epidemic of 2002, evidence from SARS-CoV indicated that higher viral load was associated with increased need for intensive care and overall worse prognosis. [7][8][9] In a clinical setting, the results of SARS-CoV-2 RT-PCR diagnostic tests are usually reported qualitatively as a binary positive or negative result using a speci ed cut-off, either based on Ct or integrated by an automatic algorithm interpreting different parameters of the potential ampli cation; Ct values themselves are not normally reported. It is currently unclear as to whether SARS-CoV-2 Ct values could be leveraged to guide patient management decisions.
In this review, we assessed the available global literature to determine whether there is evidence that SARS-CoV-2 Ct values correlate with clinical outcomes and therefore, whether they could provide valuable information to clinicians for more tailored decision-making.

Methods
This review was undertaken according to the principles outlined in the Cochrane handbook. A comprehensive search of PubMed was conducted on 1st June 2020 based on the following search strategy: (Ct value OR viral load) AND SARS-CoV-2. Titles and abstracts were screened for relevance by two independent reviewers, and a third reviewer resolved con icts. All studies that were conducted in humans diagnosed with COVID-19 and reported on the presence or absence of an association between real-time RT-PCR Ct values, or viral load speci cally determined via real-time PCR Ct value, and clinical or healthcare-associated outcomes were eligible for inclusion. Studies that reported only on the time course of SARS-CoV-2 viral load or that only compared viral load in different sample types or using different methodologies were not included. Pre-review articles, animal studies and reviews were excluded, but additional publications were identi ed by manual citation searching of appropriate reviews. The full-texts of relevant studies were assessed for inclusion by two independent reviewers and key data from all included studies were captured using a data extraction form. All extracted data were veri ed by an independent reviewer.

Disease progression
One study reported that SARS-CoV-2 Ct values at hospital admission negatively correlated with the probability of progression to severe disease in 62 patients who presented with mild-moderate disease (Table 1). 17 Lower Ct values were observed in specimens from patients who became severely ill during hospitalisation than those who did not (24 vs. 29; p = 0.008).

Disease severity
Eleven studies (with numbers of PCR-positive patients ranging from 10 to 308) reported on the correlation between Ct value and disease severity, [10][11][12][13][14]16,17,20,23,25,26 ower Ct values from respiratory samples were associated with more severe disease in 7 (64%) of these ( Table 2). [11][12][13][14]16,17,23 Three studies (with numbers of PCR-positive patients ranging from 23 to 114) reported on the correlation between viral load determined via Ct values and disease severity 15,19,21 and one of these (which included 96 patients) reported that higher viral loads were signi cantly associated with more severe disease ( Table 2). 19 Of the 15 studies reporting on the correlation between Ct value or viral load determined via Ct value and disease severity, 11 were performed in hospitalised patients [10][11][12][13][14][15][16][17]19,21,23 and three included non-hospitalised patients. 20,25,26 Of the eleven studies performed in hospitalised patients only, eight (73%) reported an association between Ct value and disease severity, [11][12][13][14]16,17,19,23 of which six showed statistical signi cance. [12][13][14]17,19,23 None of three studies that included non-hospitalised patients reported that patients with severe disease had higher viral loads compared with those with mild disease. 20,25,26 Biochemical and haematological markers All ve studies (with numbers of PCR-positive patients ranging from 12 to 308) reporting on the correlation of Ct value with biochemical and haematological markers showed a correlation with at least one marker (Table   2). 11,13,14,18,22 Lower Ct values were signi cantly associated with: higher lactate dehydrogenase levels (n = 4); 11,13,18,22 lower lymphocyte counts and/or percentages (n = 3); 11,13,14 lower T-cell counts (n = 3); 11,13,18 lower serum albumin levels (n = 2); 11,14 increased levels of creatinine kinase myocardial band (n = 2); 11,18 and increased levels of high-sensitivity troponin 1 (n = 2). 11,13 Two studies showed that lower Ct values were associated with higher neutrophil counts and/or percentages, 11,14 whereas one study showed a negative correlation. 18 One study in 12 patients showed that C-reactive protein levels negatively correlated with Ct value (r = −0.584; p = 0.03), 14 whereas another in 25 patients showed no signi cant association (p = 0.07). 22 Associations were also reported between Ct values and angiotensin II, 14 IL-2R, 13 basophil and eosinophil counts, and levels of myoglobin, N-terminal pro-brain natriuretic peptide, inorganic phosphorus and calcium. 11 Infectivity Two studies reported on the correlation between Ct value and infectivity and showed that lower Ct values were associated with higher probability of a positive viral culture (Table 2). 24,27 In one study of 155 patients, multivariate logistic regression analyses using time from symptom onset to test, age and gender as independent variables showed a signi cant effect of Ct value on the culture positivity of samples (OR 0.64 [95% con dence interval 0.49-0.84], p < 0.001) suggesting that for every one unit increase in Ct, the odds of positive culture decreased by 32%. 27 The results demonstrated that infectivity (de ned as growth in cell culture) was signi cantly reduced when RT-PCR Ct values were greater than 24 (p < 0.001).

Discussion And Conclusion
The majority of the 18 studies identi ed in this review reported an association between SARS-CoV-2 Ct values or viral load determined via Ct values and clinical outcomes. Higher Ct values generally correlate with lower viral loads, although Ct value and log viral load may not be directly proportional due to the linear dynamic range of the assay and potential presence of inhibitory factors within clinical samples. 28 Fifteen (79%) of the studies included in this review investigated the direct association of Ct values with clinical outcomes, rather than viral load itself, but it was assumed by authors that Ct values are an appropriate surrogate for viral load.
Clinical knowledge of COVID-19 is constantly evolving, with studies being published at a high rate; however, there is currently only limited data relating to the correlation of viral loads with patient prognoses, such as mortality or disease progression. Only one study reported on the association between mortality and SARS-Cov-2 Ct value and showed that lower Ct values correlated with increased risk of death, 11 which is consistent with data for previous epidemic-causing coronaviruses. 9,29 Given the wide range in disease course for COVID-19, the ability to predict which patients are at particularly high risk of deterioration and negative outcomes would be of particular value in the clinical setting; it would therefore be useful to continue to assess the value of SARS-CoV-2 Ct as further data become available.
Eleven studies reported on the correlation of Ct values with symptom severity at presentation and seven of these indicated that lower Ct values were associated with more severe disease. This is consistent with some previous studies of Ct values in other respiratory infections, [29][30][31] although other studies do not show correlation. 32 Whilst correlation between Ct value and disease severity was observed for 73% of studies in hospitalised patients, correlation between Ct or viral load determined via Ct value and disease severity was observed in none of the studies that included patients with COVID-19 who were not hospitalised. Studies in hospitalised patients are unlikely to include asymptomatic patients or those with very mild symptoms, but are likely to be more controlled, making correlations with Ct more probable. The role that symptoms play in viral shedding remains to be determined; in a large study of 5,830 patients with COVID-19, which was pre-review at the time that this review was conducted, viral load determined via Ct values in nasal swabs of asymptomatic and symptomatic patients was not statistically different (median 4.7 log 10 copies/ml vs. 5.0 log 10 copies/ml; p = 0.51). 33 Ct values were found to correlate with a number of clinical markers. Lower Ct values were associated with elevated LDH levels in all four studies in which it was assessed, which is consistent with reports that elevated LDH can act as in indicator of poor prognosis in patients with COVID-19. 34,35 Increased LDH re ects tissue destruction and in interstitial pulmonary brosis is seen as an important prognostic marker for lung injury. 36 Lower Ct values were associated with lower lymphocyte levels in all three studies in which it was assessed, which is consistent with reports that lymphopaenia could act as a predictor of higher disease severity in patients with COVID-19. 35,37 Similarly, correlation of Ct values was also seen with high-sensitivity troponin I, which has been suggested as a marker of COVID-19 disease progression and mortality. 35 Both studies that investigated the correlation between SARS-CoV-2 Ct values and infectivity showed that samples with higher Ct values had lower culture positivity. It has been shown that following resolution of COVID-19 symptoms, people can have prolonged positive SARS-CoV-2 real-time PCR results for several weeks 38 and at late time points, Ct values are often very high representing low copies of viral RNA. 20 Therefore, as suggested previously, 4 considering Ct values in conjunction with the clinical context of patients may help in patient management decisions such as the need for isolation, use of PPE and testing resources.
To the best of our knowledge, this is the rst report to systematically assess the globally available literature data relating to the predictive value of SARS-CoV-2 Ct values; however, it is associated with a number of limitations. The majority of the studies included in this review contained a relatively small number of patients; only four studies included more than 100 patients with COVID-19. 11,15,18,24 The viral load of SARS-CoV-2 is known to vary during the course of infection. 10,11,13,19,20,[39][40][41][42] The time from onset of symptoms to sampling varied between studies, and in most of the included studies, varied between patients. Time from onset of symptoms was included as an independent variable in only one of the analyses presented 27 and therefore this may be a confounding factor in many of the studies reported. The type of sample used varied between studies. Sample type is known to affect the Ct values and detected viral load, 20,40 and therefore this may have affected results. The variability within and between the included studies is not consistent with previously reported considerations regarding variability around factors involved (sample type, work ow, assay) in robust viral load measurement using RT-PCR. 3 Reproducible experimental layouts to assess viral load from patient samples are key to establish any correlation to patient outcome.
Reporting of qualitative SARS-CoV-2 test results as positive or negative is su cient for diagnosis, but the totality of currently available data indicate that the reporting of Ct values may offer bene t to clinicians in making clinical and patient-management decisions for patients with COVID-19, as well as guide infection control, public health and occupational health decisions. However, additional data and prospective studies are required to support this. Table 1 Summary of reported data relating to SARS-CoV-2 realtime PCR Ct value correlation with mortality, disease progression and severity   Figure 1 PRISMA ow diagram