Polly Clayden, HIV i-Base
The new WHO paediatric guidelines now recommend immediate antiretroviral therapy for children with confirmed HIV infection aged <24 months. However identifying HIV-infected infants and linking them to treatment programmes, particularly fast, is easier said than done.
DNA-PCR testing is generally used for early infant diagnosis (EID), but it is expensive and requires sophisticated, centralised laboratories and trained technicians. Although DNA-PCR has been used in resource-limited settings, its long turnaround time contributes to infant loss-to-follow-up and loss of benefit of immediate initiation of treatment. Currently, no point-of-care (POC) HIV tests are available for infants.
In a special session at IAS 2010, Susan Fiscus presented an excellent overview of the current tools available and the prospects of POC technology for early EID. [1,2]
She began with a list of desirable qualities for a POC diagnostic test:
- Rapid (< one hour)
- Sensitive (> 95%)
- Specific (> 98%)
- Inexpensive (<$5 per test)
- Simple (equipment: battery operated, few moving parts, small footprint/technique: minimum training required)
- Robust no cold chain required
- Commercially available
- CE marked/FDA cleared
But she quoted Bill Rodriguezs remark on the subject: Cheap, fast or accurate. Pick two.
The current tests used for EID are HIV DNA and total nucleic acid assays. The gold standard is the Roche AMPLICOR HIV DNA assay, version 1.5. This test is used in many countries and can use whole blood pellets or dried blood spots (DBS).
The Roche Qualitative Total Nucleic Acid Assay has also been introduced. This test works on whole blood and DBS. In one study it was shown to be 100% sensitive and 99.7% specific.
Abbott is also developing a DNA assay.
These tests need large, expensive equipment and are probably only suitable for sophisticated, centralised laboratories.
For resource limited settings, HIV DNA assays need to be POC. Dr Fiscus described three tests in development.
Researchers at the Centre for Innovation in Global Health Technologies (CIGHT) at Northwestern University are working on a POC DNA-PCR. At CROI in February, they reported a lower limit of detection of 5 copies/reaction and good sensitivity and specificity.  This assay uses a small, portable, battery-operated analyser, which can assemble the reaction and perform fluorescence detection and thermal cycling. The analyser card integrates DNA extraction, PCR reagent storage without refrigeration and PCR amplification.
Data from Micronics Real Time PCR was also presented at CROI 2010.  This assay uses a credit card sized device and microfluidic principles for both nucleic acids extraction and amplification. The investigators reported good sensitivity and specificity in this study.
The Biohelix Isolamp is another simple HIV DNA test under development. This assay couples helicase-dependent isothermal amplification (HDA) with amplicon detection using a disposable cassette. Early data were presented at the 2010 HIV Diagnostics Conference. 
She explained that the CIGHT test is not yet ready for field-testing and is on hold while the group focuses on a POC p24 test. Both the Micronics and BioHelix tests appear to be in the proof of concept stage and are not ready for field testing yet either.
It is possible to use qualitative HIV RNA assays as an alternative to HIV DNA. The qualitative Gen-Probe Aptima is the only HIV RNA test approved by the FDA for diagnosis. Although the FDA approval is for plasma or serum, this system works well with DBS. It is very sensitive and specific and is being used by the State of New York for EID.
She mentioned that it is unclear whether HIV RNA assays will be as sensitive when infants are being prophylaxed or if mothers are receiving antiretrovirals and breastfeeding the child.
Several other HIV viral load assays are currently commercially available, but are not POC, require large expensive equipment, and are suitable for centralised laboratories.
Dr Fiscus described three POC RNA assays that are in development. The SAMBA (simple amplification based assay) is currently being developed by the University of Cambridge and Diagnostics for the Real World. Data were recently published in JID.  This test uses isothermal amplification and visual detection by dipstick. It has a limit of detection of 75 copies/mL using 250 mL of plasma, and 400 copies/mL using 100 mL whole blood. No cold chain is required and it can be battery operated. It is simple to operate and little training is needed. There will be a clinical trail for regulatory approval in 2011.
Dr Fiscus showed recent unpublished data from her own research group. The IQuum LIAT quantative POC HIV assay is a real time PCR, which can be battery operated, is easy to use and requires little training. It gave 92% correlation with Abbott m2000 with 75 plasma samples. It has not yet been tested with whole blood. The assay takes 60 minutes to perform but does need a cold chain.
Inverness Medical Innovations CLONDIAG, uses a microarray, real time detection method. It can use fingerstick, whole blood or plasma. The sample is applied directly onto the test cartridge, which is processed by a compact, battery driven instrument. Preliminary data provided by the manufacturer are promising.
Finally p24 antigen tests, which have limited use in adult diagnostics, can be used for EID. The ultrasensitive, heat dissociated p24 antigen assay has been shown to work well with both plasma and DBS.
As far as POC is concerned, Dr Fiscus showed results from the CIGHT p24 antigen rapid test that were recently published ahead of print in JAIDS.  This assay is performed, by adding 25mL of plasma to 75mL buffer. This mixture is heated in a water bath at 90 degrees for four minutes. A test strip is then inserted and gives a read out after 20 minutes. Trials in Cape Town showed 95% sensitivity and 99% specificity.
She also described an improved CIGHT POC p24 antigen rapid test under development. This assay uses whole blood and heat shock to increase sensitivity. It
consists of a plasma separator, reaction tube, reaction buffer and rapid test strip. It is battery operated and each test should cost $1-2.
In keeping with Bill Rodriguezs remark she presented the most likely future tests in a table. See Table 1.
Table 1: Cheap, fast or accurate. Pick two Susan Fiscus
|Cheap: < $5 USD||Fast:
< 60 min
|Accurate: Sensitivity >95% Specificity > 98%||Whole blood||Robust (battery operated and no cold chain)|
|IQuum LIAT||?||Yes||Yes||In development||Needs cold chain|
She concluded that promising POC assays for EID today include: IQuums LIAT, SAMBA, CIGHT p24 and possibly CLONDIGs viral load assay.
This was an incredibly useful overview and it looks like we can be optimistic about having a POC test for EID in the next couple of years.
Other presentations at the conference dealt with the challenges of access to EID. In the same session, Shaffiq Essajee noted that EID access has improved and in some countries more than 50% of exposed infants are tested.  But infant testing is usually linked to PMTCT and if coverage is low, so is infant testing coverage. Globally only 15% of exposed infants get a test. Even when there is access to EID, infected infants do not necessarily get ART.
Laura Guay showed that there are similar coverage cascades with EID to that of PMTCT.  Losses occur at each step of the cascade. She showed data from a programme of the Elizabeth Glaser Pediatric AIDS Foundation in which there were 4226 infants with known exposure. Of these 4099 (97%) had EID drawn, 895 (70%) had results returned from the lab with 449 (15%) positive results. Then only, 230, (51%) received results, 200 (87%) enrolled in care and 178 (89%) infants initiated treatment. Overall, she explained, there were 633 infected children, of which 71% were identified and 28% treated. And this she said is a good programme.
A related poster showed Ministry of Health data from 84 sites in Cambodia, Namibia, Senegal and Uganda with >21,000 infants tested.  Although the study was called Increasing uptake of HIV early infant diagnosis services in four countries and showed steady increases in sample volume, in 2008, it was still low in three of the countries reviewed: Cambodia 14%, Senegal 9% and Uganda 21%. Namibia however achieved 86-100% EID coverage. Less than half these infants ever tested via EID were tested in their first two months of life. Coverage of optimal service (early testing) is consequently even lower. And of those infants tested HIV positive via EID, attrition is significant 72%, 67% and 67% were not alive and on ART in Uganda, Cambodia and Senegal respectively.
The investigators concluded: Significant strides to establish and increase access to and uptake of EID testing have been made across all countries reviewed. When decentralising, it is important for programming to focus on early identification and access to the full package of exposed infant services including EID.
So as well as a POC test for EID, which will take care of some of the obstacles to successful diagnosis and treatment, programmes need to ensure good systems for patient management and links between services to ensure that they can take full advantages of the promise of these new technologies.
- Scaling Up Early Infant Diagnosis of HIV as the Bridge between Prevention, Care and Treatment: Successes, Challenges and Potential Solutions. 18th IAS. July, 2010. Vienna. Special session SUSS03. http://pag.aids2010.org/session.aspxs=150
- Fiscus S. EID – current tools and prospects of point of care technology. 18th IAS. July, 2010. Vienna. Special session SUSS0302.
- Jangam S et al. A point-of-care DNA PCR test for infants. 17th CROI. February 2010. San Francisco. Poster abstract 891.
- Granade T et al. rapid extraction and amplification of HIV-1 DNA from whole blood using a disposable microfluidics device. 17th CROI. February 2010. San Francisco. Poster abstract 945.
- Jordan JA et al. Evaluation of the IsoAmp HIV detection kit using whole blood samples. HIV Diagnostics Conference, March 24-26, Orlando, Florida. Poster abstract 34.
- Lee H et al. Simple Amplification-Based Assay: A nucleic assay based point-of-care platform for HIV-1 testing. Jour Infect Dis 2010:201 (Supp 1) S65-72.
- Parpia ZA et al. p24 antigen rapid test for diagnosis of acute pediatric HIV infection. J Acquir Immune Defic Syndr. Published ahead of print August 2010.
- Guay L. Early infant diagnosis of HIV: successes, challenges and potential solutions. 18th IAS. July, 2010. Vienna. Special session SUSS0301.
- Tripathi S et al. Increasing uptake of HIV early infant diagnosis (EID) services in four countries (Cambodia, Namibia, Senegal and Uganda. 18th IAS. July, 2010. Vienna. Poster abstract TUPDB205. http://pag.aids2010.org/Abstracts.aspx?SID=185&AID=14996