Differences in Viral Load Testing Methodology May Play an Important Role in Optimizing the Initial Management of HIV-Infected Pregnant Women

One of the major success stories of HIV management is the significant drop in perinatal transmission rates. ¹ Much effort has been made in achieving a rate of <200 cases/year in the United States. As part of routine management, providers rely heavily on blood tests to manage HIV-infected pregnant patients. Determination of maternal viral load (VL) and viral genotype are paramount in determining the optimal maternal and newborn antiretroviral regimen. Current methods of detecting HIV VL are polymerase chain reaction (PCR)-based testing. There are, however, subtle differences in HIV VL detection methods that health care providers may not be familiar with. A recent case in our clinic highlights the complexities of testing for HIV during pregnancy.

A 35-year-old HIV-positive pregnant woman originally from South Africa presented to our clinic at 13 weeks’ gestation. In addition to living in South Africa, she also spent time in her adolescence and early adulthood in Zambia before coming to the United States 10 years ago. She was unsure how she acquired HIV but suspects she was infected either through a blood transfusion, given her history of sickle cell disease, or heterosexual transmission. Both her current and previous partner tested negative, and she was antiretroviral treatment naive. She reported that prior to coming to the United States, HIV testing was indeterminate and that she was “HIV negative.” Her CD4 count was >1100 cells/mm³, by her report. Unfortunately, previous medical records were not available to verify her history. Testing was repeated at our center, with the HIV enzyme-linked immunosorbent assay (ELISA) as well as all Western blot bands returning positive, VL <75 copies/mL, and CD4 count 690 cells/mm³. Viral load was performed using Versant HIV kPCR (Siemens Diagnostics, Malvern, PA, USA). We found it surprising that her VL was undetectable without therapy. After discussion with the National Clinician’s Consultation Center Perinatal HIV Hotline [phone]), additional VL testing was sent to a laboratory that used real-time HIV-1 PCR (Abbott Molecular Diagnostics, Des Plaines, IL, USA) which has better sensitivity for groups N and O. The result from this assay returned with a VL of 108 copies/mL. Testing for HIV-1 DNA PCR was positive and HIV-2 DNA/RNA PCR was negative. An empiric regimen of abacavir (ABC)/lamivudine (3TC) and lopinavir/ritonavir (LPV/r) was started in order to avoid hemosuppression that can be seen with zidovudine (ZDV). Repeat VL was undetectable thereafter using real time PCR. She delivered a healthy full-term infant whose HIV DNA PCRs remained negative.

Numerous HIV-infected individuals with diverse viral genetic backgrounds are living in the United States. ² Choosing the correct testing method to determine antiretroviral therapy is essential in providing optimal care. Are current standards for HIV testing adequate to best manage these individuals? For the majority of infected pregnant patients, the answer is likely yes. Presently, 7 commercially available assays (only 4 with US Food and Drug Administration [FDA] approval) provide results with high sensitivity, specificity, and reproducibility. All are equipped to detect group M along with most of its 9 subtypes. Only the real-time PCR and Versant HIV kPCR can detect group O. Group N, however, can only be detected using the real-time PCR assay. ³ Groups N and O are more commonly found in West-Central Africa, particularly Cameroon. Our patient came from a region that has higher rates of group O and may have been infected by someone from that region. In this situation, real-time PCR would be the correct method to determine the VL. Even though we believe the correct assay was used, our patient’s VL remained low, indicating that she was either an elite controller or the original PCR testing was inadequate. We suspect the former. As expected, commercial genotype testing failed due to a low VL. A blood sample sent to a research laboratory for further genotyping (using RT-PCR) determined that our patient had a pansensitive group M, subtype C virus (94.9% similarity to protease gene and 94.3% similarity to the reverse transcriptase gene) using the Stanford University HIV Drug Resistance Database. ⁴

Health care providers should be mindful that there are subtle differences in VL testing methods. It is imperative that the type of assay chosen be based on the geographic location from where the infection was likely acquired. Although this does not assure that the correct assay will always be selected, it should improve the detection of the suspected HIV group and subtype. We advocate for sensible selection of PCR testing among the choices available and eagerly anticipate the development of more sensitive genotyping. This is particularly important in the management of HIV-positive pregnant women in whom viremic control is crucial to prevent perinatal HIV transmission.