HIV-1 RNA Copies/mL to IU/mL Converter
HIV-1 RNA Copies/mL to IU/mL Converter
HIV-1 RNA is reported in copies/mL and the international standard is in IU/mL. Convert between them with the factor published for the assay that produced the result, and see why the two numbers differ by about 0.23 log.
HIV-1 RNA copies/mL → IU/mL
copies/mL + named assay → IU/mLHIV-1 RNA 120,000 copies/mL, measured on a Roche cobas HIV-1 assay (1 copy = 1.70 IU)
The conversion, and what it is not
- copies/mL
- what the assay reports and what every HIV treatment threshold is written in. Unlike hepatitis B and C, routine HIV reporting never moved to international units
- IU/mL
- the unit calibrated to the WHO International Standards for HIV-1 RNA, NIBSC 97/656 and 12/224. It appears on some European reports, in trial protocols and in external quality assessment
- IU per copy
- published per platform: 1.70 for the Roche cobas assays, 1.74 for Abbott m2000 RealTime, 1.63 for Abbott Alinity m. There is no universal value, and the standard itself assigns no copy number
- log₁₀ offset
- log₁₀ of the factor, 0.23 for the default option — constant at every viral load, which is why a log change is unit-independent and an absolute value is not
Worked example
HIV-1 RNA 120,000 copies/mL, measured on a Roche cobas HIV-1 assay (1 copy = 1.70 IU)
120,000 × 1.70 = 204,000 IU/mL
log₁₀ 120,000 = 5.08 log₁₀ copies/mL
log₁₀ 204,000 = 5.31 log₁₀ IU/mL
The difference, 0.23 log₁₀, is log₁₀ 1.70 — the same at a viral load of 50 and at a viral load of 5,000,000
On an Abbott Alinity m assay the same 120,000 copies/mL would read 195,600 IU/mL rather than 204,000, a 4% difference with no change in the patient
Run the other way: a result of 1,000 IU/mL on the Roche factor is 1,000 ÷ 1.70 = 588 copies/mL, which is above the 200 copies/mL failure threshold where the raw IU figure would have looked five times worse than it is
Published IU-per-copy factors, by platform
| Assay | Manufacturer | Copies per IU | IU per copy | What the source prints |
|---|---|---|---|---|
| cobas HIV-1 on 5800/6800/8800, and COBAS AmpliPrep/COBAS TaqMan HIV-1 v2.0 | Roche | 0.5882 | 1.7 | 1 copy = 1.70 IU, 1 IU = 0.60 copies |
| m2000 RealTime HIV-1 | Abbott | 0.5747 | 1.74 | 1 copy = 1.74 IU, 1 IU = 0.57 copies |
| Alinity m HIV-1 | Abbott | 0.6135 | 1.63 | 1 copy = 1.63 IU, 1 IU = 0.61 copies |
The thresholds, and the unit each one is written in
| Term | Definition | Unit |
|---|---|---|
| Virologic suppression | A confirmed HIV RNA level below the lower limit of detection of the assay in use | copies/mL |
| Low-level viraemia | A confirmed level above the lower limit of detection but under 200 | copies/mL |
| Virologic failure | Failure to achieve or to maintain suppression to under 200 | copies/mL |
| Virologic blip | After suppression, an isolated detectable level followed by a return to suppression | copies/mL |
Why HIV kept copies/mL when hepatitis did not
Hepatitis B and hepatitis C both moved their routine reporting to international units, for a good reason: a “copy” was never a standardised quantity, each manufacturer defined its own relationship between the genome copies an assay counted and the number it printed, and results were not comparable between laboratories. HIV did not follow. The WHO International Standards for HIV-1 RNA exist — NIBSC 97/656 and its replacement 12/224 — and modern assays are calibrated against them, but they go on reporting copies per millilitre, because that is the unit in which every treatment threshold, every trial endpoint and thirty years of literature are written.
So the conversion here is unusual: it runs from the unit everyone uses towards the unit the standard is defined in. The factors are small and they agree with each other — 1.63 to 1.74 IU per copy across two manufacturers and four platforms, a far tighter cluster than hepatitis C manages. But none of them is 1. An international unit and a copy are different quantities, and a result that changes unit without changing patient moves by about 0.23 on the log scale.
That is not small here. Repeat testing of one specimen varies by roughly 0.3 to 0.5 log₁₀ from imprecision alone, and a change under about 0.5 log₁₀ is conventionally read as noise — see the viral load log change calculator. A unit change therefore contributes about half of a “real” change by itself, so a copies/mL result compared with an IU/mL result from another laboratory can look like a rise that never happened.
The practical advice is the dull one. Stay in copies/mL; it is what the assay issues and what the guidelines are written against. Convert only to read a trial protocol or a European report, and only when you know which platform produced the number. A result is comparable only with results from the same assay; a change across platforms is not a change in the patient. The viral load log converter handles the absolute-to-log step for any virus, and the HCV RNA converter and HBV DNA converter make the same point for hepatitis, where the factors are much further apart. 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
Is HIV viral load reported in copies/mL or IU/mL?
Copies/mL, essentially everywhere. The WHO International Standards for HIV-1 RNA exist and current assays are calibrated to them, but routine reporting never moved to international units as hepatitis B and C did, because every treatment threshold and trial endpoint is in copies/mL.
What is the conversion factor between HIV-1 RNA copies/mL and IU/mL?
It belongs to the assay, not to the virus. The manufacturers’ panel datasheets give 1.70 IU per copy for the Roche cobas assays, 1.74 for Abbott m2000 RealTime and 1.63 for Abbott Alinity m. If your platform is not among them, it has no published factor and this page refuses to supply one.
Does the unit change affect a log change between two results?
No, provided both results came from the same assay. The conversion is a multiplication, which in log space is a constant addition, and a constant cancels when one log value is subtracted from another. A 1-log fall is a tenfold fall in either unit.
How big is the unit change on the log scale?
About 0.23 log₁₀ at 1.70 IU per copy. That matters because repeat testing of the same specimen varies by roughly 0.3 to 0.5 log₁₀ from imprecision alone, so switching unit contributes about half of what would otherwise be read as a real change.
Can I compare an IU/mL result against the 200 copies/mL failure threshold?
Not directly. The threshold is a copies/mL figure, so an IU/mL number is about 1.7 times too high against it — 170 IU/mL looks close to the limit when the copies figure behind it is 100. Convert first, with the factor for the assay that produced the result.
Related calculators
References
- Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the Use of Antiretroviral Agents in Adults and Adolescents with HIV: virologic failure. Department of Health and Human Services, updated 12 September 2024.
- SeraCare / LGC Clinical Diagnostics. AccuSpan HIV-1 RNA Linearity Panel datasheets 2410-0221. (Copies-to-IU factors stated per platform for Roche cobas and COBAS AmpliPrep/COBAS TaqMan HIV-1 v2.0, Abbott m2000 RealTime HIV-1 and Abbott Alinity m HIV-1.)
- NIBSC. WHO International Standards for HIV-1 RNA for nucleic acid amplification techniques, codes 97/656 (1st) and 12/224 (2nd). Potters Bar: National Institute for Biological Standards and Control.
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/