CMV DNA IU/mL to Copies/mL Converter
CMV DNA IU/mL to Copies/mL Converter
The WHO cytomegalovirus standard assigns a potency in international units and no copy number at all. Convert with the factor published for your assay, and see how far apart three platforms put the same specimen.
CMV DNA IU/mL → copies/mL
IU/mL + named assay → copies/mLCMV DNA 1,800 IU/mL, measured on an Abbott RealTime CMV assay (copies/mL = IU/mL ÷ 1.56)
The conversion, and why it is not a constant
- IU/mL
- international units per mL, calibrated to the 1st WHO International Standard for human cytomegalovirus (NIBSC 09/162), which is assigned 5 million IU on reconstitution in 1 mL
- copies/mL
- a legacy unit: how many genome copies a particular manufacturer’s assay counted per international unit. The WHO standard’s instructions for use assign no copy number whatsoever
- copies per IU
- 0.641 for Abbott RealTime CMV, 1.099 for the Roche COBAS AmpliPrep/COBAS TaqMan CMV test, 0.610 for the Qiagen artus CMV QS-RGQ. A 1.8-fold spread, with 1 inside it
- log₁₀ offset
- log₁₀ of the factor. Abbott’s insert states it directly: log copies/mL = log IU/mL minus 0.19, which is log₁₀(1/1.56)
Worked example
CMV DNA 1,800 IU/mL, measured on an Abbott RealTime CMV assay (copies/mL = IU/mL ÷ 1.56)
1,800 ÷ 1.56 = 1,154 copies/mL
log₁₀ 1,800 = 3.26 log₁₀ IU/mL
log₁₀ 1,154 = 3.06 log₁₀ copies/mL
The difference is −0.19 log₁₀, exactly the figure the Abbott insert prints
On the Roche COBAS AmpliPrep/COBAS TaqMan assay the same 1,800 IU/mL is 1,800 × 1.099 = 1,978 copies/mL — 1.7 times the Abbott figure, and the conversion has moved the number in the opposite direction
A reader who assumed 1 copy per IU would have reported 1,800 and been wrong by 36% one way or 10% the other, depending on a machine they never saw
Published copies-per-IU factors, by platform
| Assay | Manufacturer | Copies per IU | IU per copy | What the source prints |
|---|---|---|---|---|
| RealTime CMV | Abbott Molecular | 0.641 | 1.56 | copies/mL = IU/mL divided by 1.56; log copies/mL = log IU/mL minus 0.19 |
| COBAS AmpliPrep/COBAS TaqMan CMV | Roche | 1.099 | 0.91 | 1 copy/mL = 0.91 IU/mL |
| artus CMV QS-RGQ | Qiagen | 0.6098 | 1.64 | 1 copy/mL = 1.64 IU/mL |
What international units fixed, and what they did not
| Measurement | Spread across seven systems |
|---|---|
| The WHO standard material itself | IQR 0.07–0.43 log₁₀ IU/mL; differences 0.22–0.67 log₁₀ |
| Clinical specimens, assays’ own in-kit calibration | IQR 0.44–1.05 log₁₀ IU/mL; median spread 1.02 log₁₀ |
| Clinical specimens, recalibrated against the WHO standard | median spread 0.86 log₁₀ IU/mL |
| Clinical specimens, recalibrated against a commercial secondary standard | median spread 0.71 log₁₀ IU/mL |
An international unit is a vial, not a count
Why there is no single number of copies per international unit is worth stating plainly, because it is the same reason for every virus on this site. An international unit is defined by a physical object: the 1st WHO International Standard for human cytomegalovirus, NIBSC 09/162, a lyophilised preparation assigned 5 million IU on reconstitution in a millilitre. Everything calibrated against that vial is on a common scale. What the vial does not come with is a copy number — its instructions for use say nothing relating international units to genome copies, because that depends on how a given assay extracts, amplifies and counts.
So the factors are properties of machines. Abbott’s RealTime CMV insert states that copies/mL is IU/mL divided by 1.56, which is 0.641 copies per IU. A comparison of two platforms calibrated against the same WHO standard found 1 copy/mL equivalent to 0.91 IU/mL on the Roche COBAS AmpliPrep/COBAS TaqMan assay — 1.099 copies per IU — and to 1.64 IU/mL on the Qiagen artus QS-RGQ, which is 0.610. Those three span 1.8-fold and straddle 1, so a reader cannot guess even the direction of the correction without knowing the platform.
The deeper point is that standardising the unit did not standardise the measurement. One set of specimens run across seven commercial systems showed the WHO material quantified fairly consistently, an interquartile range of 0.07 to 0.43 log₁₀ IU/mL, while real clinical specimens spread by 0.44 to 1.05 log₁₀ — and recalibrating every system against the WHO standard moved the median spread only from 1.02 to 0.86 log₁₀. Patients’ plasma is not commutable with a vial of lyophilised virus, and no shared calibrator makes it so.
Which is why this page prints no threshold. Pre-emptive therapy cut-offs for CMV are set locally, against the assay and specimen type a centre uses, precisely because a number from one laboratory does not transfer. A result is comparable only with results from the same assay; a change across platforms is not a change in the patient. What does transfer is the log change on one assay, which the viral load log change calculator works through. This supports a clinician’s judgement rather than replacing it: it is arithmetic on the figures entered, and knows nothing about the patient.
Frequently asked questions
What is the conversion factor from CMV IU/mL to copies/mL?
There is no universal factor. Abbott’s RealTime CMV insert gives 0.641 copies per IU (copies = IU ÷ 1.56); a published comparison gives 1.099 for the Roche COBAS AmpliPrep/COBAS TaqMan assay and 0.610 for the Qiagen artus QS-RGQ. The factor belongs to the platform.
Why does the WHO standard not define a copy number?
Because it is a material, not a count. NIBSC 09/162 is a vialled preparation assigned 5 million international units per millilitre on reconstitution; how many genome copies an assay reads out of it depends on that assay’s extraction, target and amplification, so the instructions for use assign no copy equivalence at all.
Are CMV viral loads from two laboratories comparable?
Not reliably, even in IU/mL. Across seven commercial systems the same clinical specimens varied by 0.44 to 1.05 log₁₀ IU/mL, and recalibrating against the WHO standard left most still spread by more than 0.5 log₁₀. Monitor a patient on one assay.
Should a CMV result be reported in IU/mL or copies/mL?
IU/mL. It is what current assays issue and what the WHO standard defines. Convert to copies/mL only to make sense of a historical result, and only when you know which assay produced it.
Does the unit affect the log change between two CMV results?
No, as long as both came from the same assay. The conversion multiplies the value, which is a constant addition in log space, and that constant cancels on subtraction. The absolute figure is unit-dependent; the change is not.
Related calculators
References
- Abbott Molecular. Abbott RealTime CMV package insert, 51-608290. (Standardised to NIBSC 09/162; LLOQ 50 IU/mL; “results can be converted to copies/mL using a conversion factor of 1.56”.)
- Çolak D, Sağlık İ, Can Sarınoğlu R, et al. Comparison of two commercial quantitative cytomegalovirus (CMV) polymerase chain reaction tests calibrated by World Health Organization international CMV standard. Mikrobiyol Bul. 2020;54(2):257–65. doi:10.5578/mb.68628
- NIBSC. 1st WHO International Standard for Human Cytomegalovirus for Nucleic Acid Amplification Techniques, code 09/162: instructions for use. Potters Bar: National Institute for Biological Standards and Control. (5 × 10⁶ IU on reconstitution in 1 mL; no copy equivalence is assigned.)
- SeraCare / LGC Clinical Diagnostics. AccuSpan CMV DNA Linearity Panel datasheet 2410-0174. (“1 copy/mL = 0.91 IU/mL, 1 IU/mL = 1.1 copies/mL” for the Roche COBAS AmpliPrep/COBAS TaqMan CMV test.)
- What challenges remain in harmonizing cytomegalovirus viral load quantification across laboratories? Preprint, bioRxiv 2024. doi:10.1101/2024.12.09.627549. (Seven laboratories, seven commercial systems; variation per specimen mostly above 0.5 log₁₀ IU/mL even after recalibration against the WHO standard.)
Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/