Ritonavir and COVID-19: pragmatic guidance is important

We thank Joseph Heskin and colleagues¹ for highlighting the crucial issue of drug–drug interactions (DDIs) with ritonavir, the pharmacoenhancer or booster co-formulated with the novel SARS-CoV-2 protease inhibitor, PF-07321332 (Paxlovid, Pfizer [New York, NY, USA]).² Since Paxlovid will be primarily administered to non-hospitalised individuals and prescribed by clinicians who might not routinely manage complex interactions or have access to their full medication list, an awareness of the DDI potential and clear pathways to support safe decision making are essential, ideally led by pharmacists who have speciality knowledge in this area. If managed appropriately, DDI should, in most cases, not necessitate a change in antiviral management.

The onset of ritonavir's inhibitory effect on the CYP3A4 isoenzyme, and to a lesser degree CYP2D6, is rapid, but the inhibition is also lost rapidly after drug cessation, mostly within 2 days.³ This information is important to guide dose adjustment or pause of concomitant medication where advised. As Heskin and colleagues clearly highlight, ritonavir also induces several cytochrome P450 isoenzymes, but this induction effect is slow to develop and is unlikely to be of clinical importance when used in a short course. However, an important consideration in people established on strong CYP3A inducers, such as carbamazepine, phenytoin, and rifampicin, is that these inducers are likely to reduce nirmatrelvir exposure and, as induction persists for about 2 weeks after cessation, are a contraindication to its use.

The clinical impact of DDIs depends on a number of factors including: the therapeutic window of the co-administered drug; the degree to which co-administered drugs are metabolised via CYP3A4 (ie, higher DDIs magnitudes are anticipated for those extensively metabolised by CYP3A4, for instance simvastatin); and the clinical indication and relative benefit treatment for the individual. We, of course, advise prescribers to consult the relevant summaries of product characteristics, and appropriate prescribing tools. Heskin and colleagues¹ refer to the University of Liverpool HIV drug interaction checker, and, although this is an invaluable tool, we encourage clinicians to refer to their specific COVID-19 interaction checker, as the advice might differ for short-term ritonavir use. However, the real-life effect of known or predicted DDIs, and recommended practice, might differ from prescribing advice, and sources of advice might be inconsistent. Antiretrovirals are one example of potential DDIs. The current UK patient information leaflet warns that Paxlovid treatment can result in medicines used to treat HIV becoming less effective.⁴ However, any resulting DDIs are not considered clinically important, as reflected by the University of Liverpool interaction checker, and the UK prescribing advice does not warn of reduced antiretroviral effectiveness.⁵ Another example is anti-platelet agents. Ritonavir reduces exposure to the active metabolite of clopidogrel, resulting in reduced anti-platelet effectiveness. This DDI might be clinically important in the context of a recent vascular stent but the potential interaction is not included in the information for patients or prescribers.4, 5 Conversely, the University of Liverpool COVID-19 interaction checker advises not to co-administer Paxlovid and clopidogrel, but provides more nuanced advice based on clinical indication, advising that the period within 6 weeks after stenting is the highest risk and, beyond that, a transient loss in effectiveness might be acceptable versus a change in drug or less effective antiviral. However, we suggest that the association between COVID-19 and thrombotic events might warrant a more cautious approach and switch to an alternative antiplatelet therapy (such as prasugrel or ticagrelor) or COVID-19 treatment for the first 12 weeks after stent therapy, and potentially longer (up to 6 months in the presence of an acute coronary syndrome).

Careful review of concomitant medication, led where possible by a pharmacist, and clear guidance for prescribers, is essential to facilitate safe and pragmatic decision making. Rapid access to appropriate specialty advice will assist risk–benefit assessment in complex cases, but optimal COVID-19 treatment, with an alternative to Paxlovid if necessary, should not be delayed due to DDI concerns.

For more on the University of Liverpool drug interaction checker see https://www.hiv-druginteractions.org

For more on the COVID-19 interaction checker see https://www.covid19-druginteractions.org