|Home | About | Journals | Submit | Contact Us | Français|
Genotypic tests are commonly used in clinical practice to infer human immunodeficiency virus type 1 (HIV-1) susceptibility to licensed antiretroviral drugs (3). Viral RNA extracted from patient plasma by standardized methods is used as the source material for reverse transcription (RT)-PCR followed by DNA sequencing. Viremia levels of >1,000 HIV-1 RNA copies per milliliter are normally recommended to perform the test. However, most clinicians are eager to have a genotyping result at lower viral loads (2). In fact, early detection of drug-resistant species has the potential to appropriately support treatment changes halting further development of resistance and cross-resistance (1, 5).
The success of genotyping is critically dependent on availability of an intact plasma-derived RNA. This can in turn be influenced by specimen handling, storage, and shipping. Due to the relative complexity of genotyping techniques, many clinics send plasma samples to a reference laboratory rather than setting up their own facilities. As a consequence, a reference laboratory may serve both a local and multiple distant HIV care units. The HIV Monitoring Laboratory (HML) at the University Hospital of Siena, Siena, Italy, has offered genotypic antiretroviral resistance testing as a public health service since 1996, generating >10,000 HIV-1 pol sequences mostly obtained from samples shipped from remote clinics. To examine the possible drawbacks of remote sampling versus local sampling, we recently analyzed the HML database and selected all the HIV-1 genotypic tests performed with the same and most updated procedure (4) on plasma samples sent by the local clinic of the same hospital and by remote clinics. Samples collected locally were obtained at the same building hosting the HML and sent as citrated blood at room temperature within 2 h after drawing. Upon arrival, plasma was collected immediately and stored at −70°C until examination 1 to 2 weeks later. In contrast, samples collected remotely were sent as frozen plasma and had been drawn 1 to 4 weeks earlier and kept frozen at −70°C at the local clinic. The query results were filtered for availability of the viral load measurement on the same samples provided by the clinic to the national Antiretroviral Resistance Cohort Analysis database (www.hivarca.net). Based on these inclusion criteria, genotyping data from 1,506 local samples and 2,558 remote samples from 39 different clinics were computed (Table (Table1).1). The rate of amplification failure was significantly higher with remotely collected samples at all viral load strata except when the viral load was >10,000 HIV-1 RNA copies/ml. At those high copy numbers, there was an extremely high success rate with both types of samples (>98%), hampering detection of a possible small difference. The apparent trend of requesting HIV-1 genotyping at very low viral loads (<200 HIV-1 RNA copies/ml) relatively more often at the local clinic probably resulted from increased confidence in the feasibility of the test at such viremia levels with respect to the remote clinics. Attempts to find any association between this finding and any specific clinic or time period were not fruitful. However, inquiries at some of the remote clinics revealed improper specimen handling and/or storage conditions as a result of a lack of standard operating procedures. Potentially relevant issues included time delays in obtaining plasma from whole blood and/or in freezing of plasma, as well as storage at −20°C until shipping. Due to the large number of different clinics and the difficulty in obtaining detailed information on sample processing, it was not possible to systematically investigate the reason(s) for the comparatively lower success rate with remote sampling. Nevertheless, our data emphasize the need for compliance with rigorous procedures for sample handling, storage, and shipping when low-viremia genotyping is to be performed at a remote reference laboratory. Based on our partial survey, inappropriate storage temperatures and temperature shifts appear to be the most probable causes for a loss of assay sensitivity. Specific programs targeting the personnel involved in such preanalytical stages at the clinic and implementing adequate standard operating procedures are highly advisable to exploit the full potential of high-sensitivity antiretroviral drug resistance genotyping.
This study was supported by the European Community's Seventh Framework Programme (FP7/2007-2013) under the project Collaborative HIV and Anti-HIV Drug Resistance Network (CHAIN), grant agreement number 223131.
Published ahead of print on 21 April 2010.