[Quantitative Agreement Between Kits in Cytomegalovirus Viral Load Measurement: Where Are We?]

Monitoring of viral load is of critical importance in the clinical management of patients at risk of cytomegalovirus (CMV)-related complications following transplantation. Quantitative real-time polymerase chain reaction (qRT-PCR) is one of the most commonly used methods for CMV DNA detection. For this purpose, commercial test kits calibrated according to the International Standard of Quantitation (ISQ) defined by the World Health Organization (WHO) are used. However, measurement variability between different test systems still constitutes a significant problem. The aim of this study was to compare the test results of two different commercial kits, CMV Cobas Ampliprep/Cobas Taqman (CMV-CAP/CTM) (Roche Diagnostics, Mannheim, Germany) and NeuMoDx CMV Quant Assay (Qiagen, Ann Arbor, USA) which have been calibrated with the WHO CMV IQS. The results of 478 plasma samples run simultaneously with the CMV-CAP/CTM and NeuMoDx CMV PCR tests were analyzed. Fully automated steps including extraction, real-time amplification and result analysis were performed according to the manufacturer’s recommendations. CMV DNA was detected in 216 (45.18%) samples and not detected in 82 (17.15%) samples in both tests. A total of 180 (37.65%) samples had discordant results. Statistically, a moderate level of qualitative agreement was found between the qualitative results of both tests (kappa= 0.28, p< 0.001). When the quantitative results obtained in the dynamic measurement range of both tests (n= 104) were examined, the median viral load values measured by CMV-CAP/CTM and NeuMoDx CMV tests were calculated as 1598 IU/mL (range= 137.41-115570) and 2600 IU/mL (range= 55-220000), respectively. According to the correlation analysis, a very strong correlation was found in the comparison of the results of both tests (r= 0.83, p< 0.001). According to the Bland-Altman analysis; the average difference between the CMV-CAP/CTM test and NeuMoDx CMV test values was found to be 0.296 log10 (standard deviation: 0.412) IU/mL (the lowest difference was 0.005 and the highest difference was -1.559 log10 IU/mL), and the CMV-CAP/CTM test gave lower measurements than the NeuMoDx CMV test. In 104 samples with results in the dynamic measurement range of both tests, for log10 IU/mL results, the measurement difference was within ±0.5 log10 IU/mL in 71 (quantitative agreement: 68.3%) samples and the measurement difference was greater than ±0.5 log10 in 33 (31.7%) samples (median= 0.7 log10 IU/ml; range: 0.51-1.56). A measurement difference of more than ±1 log10 was detected in two samples (1.9%). As a result, in the measurements made with CMV-CAP/CTM and NeuMoDx CMV PCR tests, a moderate level of agreement was found for qualitative results in plasma samples, a strong correlation for quantitative values and a biologically significant viral load difference in one-third of the samples was detected. Our study shows that the measurement differences observed between CMV PCR test platforms do not completely disappear even if calibration is provided with the WHO ISQ. In post-transplantation follow-up, considering the measurement differences between test systems and preferring the same sample type and the same test platform in CMV viral load monitoring is important in terms of ensuring consistency in patient management.