A dietary supplement, commonly taken by people with HIV may prevent many antiretrovirals working properly, according to a recently published US study. Researchers at the US National Institutes for Health observed that garlic supplements sharply reduced the blood levels of the protease inhibitor saquinavir. Garlic induces, or speeds up, the activity of the CYP 450 enzyme responsible for metabolizing protease inhibitors and the NNRTIS nevirapine and efavirenz.
The report is of particular concern as many people taking protease inhibitors also take garlic supplements in the hope of controlling the cholesterol increases seen on PI treatment.
Nine HIV-negative volunteers were given 1200 mg of saquinavir three times a day with food for three days, with blood samples taken at regular intervals to determine baseline saquinavir levels. They were then given commonly available garlic capsules twice daily for three weeks. When blood samples were analysed, it was found that plasma levels of saquinavir had fallen from baseline levels by an average of 51 per cent. Both saquinavir and garlic were then discontinued for a ten day "wash-out" period. Finally, participants received a further three days of treatment with saquinavir, with blood samples again taken at regular intervals, which showed that levels of saquinavir had not returned to anticipated baseline levels, being on average 35 per cent lower than expected.
Patient adherence was checked with a self-completed questionnaire. One trial participant missed a single dose of saquinavir, but the research team did not believe this to be significant.
"In the presence of garlic supplements, blood concentrations of saquinavir decreased by about 50 per cent among our study participants," said report co-author Dr Judith Falloon. "We saw a definite, prolonged interaction. The clear implication is that doctors and patients should be cautious about using garlic supplements during HIV therapy," she added.
The same NIH research team had previously established that the herbal remedy St John's wort, effective in treating mild or moderate depression, prevented the protease inhibitor indinavir working properly.
Saquinavir, rather than more commonly prescribed protease inhibitors was selected for the trial because it is known that it does not impact substantially on CYP450. Exactly how garlic capsules disrupt the uptake of saquinavir was not determined by the team.
It is also unclear if taking saquinavir boosted by ritonavir will prevent the alterations in saquinavir blood levels caused by the garlic capsules. A study presented at the Interscience Conference on Antimicrobial Agents and Chemotherapy last year in Toronto showed that garlic did not significantly reduce levels of ritonavir, a protease inhibitor that has a very strong inhibitory effect on CYP 450. However, that study only looked at the effect of garlic on one dose of ritonavir.
Garlic supplements, which have a popular reputation as a natural cholesterol lowering agent, are often taken by people prescribed protease inhibitors whose cholesterol levels have increased as a side effect of therapy, prompting Dr Falloon to warn, "it's clear from this study that any patient using saquinavir as the sole protease inhibitor should avoid garlic supplements."
The report authors conclude with the caution that dietary supplements should not be considered benign therapies, as "some of these properties may have potent pharmacological actions and may alter the blood levels of concomitant medications...health care professionals can expect the identification of additional herbal interactions."
This warning is evidenced by a recent review in the British Journal of Clinical Pharmacology. The anti-coagulant warfarin, and St John’s wort, were found to be the most frequently reported drug and herbal therapy respectively in published cases of drug-herb interactions.
Piscitelli SC et al. The effect of garlic supplements on the pharmacokinetics of saquinavir. Clinical Infectious Diseases 34 (Electronic Edition), 2002.
Fugh-Berman A et al. Herb-drug interactions: Review and assessment of report reliability. British Journal of Clinical Pharmacology 2001;52:587-595.