Levels of the drugs AZT (zidovudine, Retrovir) and 3TC (lamivudine, Epivir) are generally higher in seminal fluid than they are in blood, a new study of these drugs’ pharmacokinetics (PKs) has found.
However the researchers also found – unexpectedly, they say – that intracellular levels (inside the cell) of the active metabolite of these drugs, the triphosphate, was lower than expected in white cells (lymphocytes) within semen than it was in comparable blood cells. In the case of AZT, intracellular levels of the drug in seminal T-lymphocytes reached only one-third of the level seen in similar cells in the blood.
The findings, published in the June edition of the Journal of Acquired Immune Deficiency Syndromes, suggest that even though free virus in semen may be successfully suppressed by nucleoside analogues, individual variations between patients in drug metabolism may leave some with lymphocytes in semen capable of transmitting HIV infection.
The study, conducted by the Universities of North Carolina and Puerto Rico and funded by GlaxoSmithKline, was the first to attempt to measure longitudinal levels of nucleoside reverse transcriptase inhibitor (NRTI) drugs in semen over a 12-hour period, either within or outside the cell.
This is important because the way drugs are metabolised within the body means that blood and semen levels may not peak and trough at the same time. One-off measurements taken at a specific time after dosing may therefore provide a very inaccurate picture of the true ratio of drug concentration in blood plasma to seminal plasma (BP/SP ratio).
A series of longitudinal measurements enables the calculation of the Area Under the Curve (AUC), or total drug exposure over a specific period, which gives a much more reliable estimate of this crucial ratio. The authors calculated that if one-off measurements were used, the lowest likely value of the BP/SP ratio (the lower 95% confidence boundary) could be anything from 60% to 285% of the observed value. But when the AUC0-12 (the Area Under the Curve from dosing to 12 hours afterwards) in blood and semen were compared, the true lowest likely BP/SP value would be between 47% and 87% of the observed value – much less variable.
The reason a longitudinal study has not been done before is partly technical, in that collecting, processing and analysing cells to determine drug levels inside them is a complex and expensive procedure compared with simply measuring plasma levels, and becomes more so if a number of time points are sampled. But it is mainly logistical.
There is no more sophisticated method of extracting samples of semen from men than via masturbation – a procedure presenting obvious difficulties if five seminal samples are required over a 12-hour period. Even if patients are capable of this, the samples extracted and drug concentrations within them would not have typical values.
To get over this problem the study took semen samples twice on the first day – immediately before dosing and 12 hours afterwards – and then got the men to provide samples on three separate out-patient visits at, respectively, three hours, six hours and nine hours after dosing.
This was a small study of 14 men studied between May 2000 and June 2003. They were all in their 30s (age range 34-40) with half being white and half African-American. They were not patients in need of HAART and were slow progressors; their average CD4 count was 512 cells/mm3 and their average viral load at study entry was 400 copies/ml. All were free of other sexually transmitted infections (STIs).
Because of the degree of viral suppression achieved, five patients were given AZT plus 3TC without a third drug. Another five took a protease inhibitor (unboosted: three nelfinavir (Viracept) and two indinavir (Crixivan)) and four took an NNRTI (three efavirenz (Sustiva), one nevirapine (Viramune)) alongside AZT/3TC. As indicated above, all but one of the men had viral loads below 50 copies/ml by the end of two weeks (average follow-up, four weeks).
The SP/BP ratio for plasma (extracellular) AZT was 2.28 – in other words the AUC0-12 was over twice as high in semen as in blood plasma (3790 nanograms per millilitre per hour (ng/mL)in semen compared with 1479ng/mL in blood). In the case of 3TC, the SP/BP ratio was 6.67 or nearly seven times as much exposure to 3TC in semen as in blood (31084 ng/mL in semen versus 4924 ng/mL in blood).
In terms of intracellular levels of the phosphorylated drugs (AZT-triphosphate (TP) and 3TC-TP) the SP/BP ratio for 3TC-TP was 1.0, in other words as much drug in semen cells as in blood cells. But in the case of AZT-TP the ratio was 0.36, in other words only slightly over a third as much AZT in seminal as in blood lymphocytes.
These ratios may vary according to factors like the degree of HIV and STI-related immune activation. AZT is more easily metabolised by activated T-lymphocytes than by quiescent ones, whereas in 3TC the reverse applies; there might therefore have been higher levels of AZT-TP observed in men with current or recent inflammatory STIs or higher HIV viral loads. The researchers mention that a previous non-longitudinal study of tenofovir (Viread), a drug which penetrates cells more easily as it is already phosphorylated, had found a SP/BP ratio of 1.0.
Intracellular half-lives of AZT-TP and 3TC-TP are 10.2 and 25 hours respectively, and the study found quite constant levels of intracellular drug, with AZT-TP varying about twofold over 12 hours and 3TC-TP hardly at all. Levels of extracellular drug in plasma were much more steady in the case of semen, with at most an eightfold variation in AZT levels over the 12 hours, than in blood, where there was a 900-fold difference in AZT levels. In the case of 3TC there was an eightfold variation in blood but at most a 50% increase in semen.
The study subject who had the lowest intracellular levels of AZT and 3TC in his semen was also the only one failing to achieve an undetectable viral load according to study protocol. Although this study is far too small (14 subjects) to draw any conclusions from this finding, especially as the person concerned was not tested for drug resistance, it may suggest that inter-patient variability in intracellular levels of the nucleoside drugs in semen could possibly be a factor in the transmission of drug-resistant HIV.
While bringing good news in general about seminal drug concentrations, the study is too small to correlate these drug levels with viral load and CD4 response. The patient with the lowest intracellular drug levels in semen was also the one that failed to achieve undetectability according to the study protocol. However this was defined very strictly; the man had been taking AZT/3TC dual therapy for 3.5 years before study entry and had been undetectable. He had a single viral blip to 880 copies/ml in his semen, but not in his blood, at week two, and then became undetectable again and stayed that way for 32 weeks of follow-up.
The main use of this study will be to act as calibration for larger studies: it means, as the researchers say, that in future “a single SP/BP concentration ratio may serve as a reasonable proxy for overall intracellular drug exposure in the male genital tract.” They called for further studies to be done in men with ongoing HIV replication and active STIs.
Dumond JB et al. Differential extracellular and intracellular concentrations of zidovudine and lamivudine in semen and plasma of HIV-1-infected men. J Acqu Immune Defic Syndr 48(2):156-162, 2008.