Managing potential drug-drug interactions between gastric acid-reducing agents and antiretroviral therapy: experience from a large HIV-positive cohort

Abstract

Drug-drug interactions between antiretroviral therapy and other drugs are well described. Gastric acid-reducing agents are one such class. However, few data exist regarding the frequency of and indications for prescription, nor risk assessment in the setting of an HIV cohort receiving antiretroviral therapy. To assess prevalence of prescription of gastric acid-reducing agents and drug-drug interaction within a UK HIV cohort, we reviewed patient records for the whole cohort, assessing demographic data, frequency and reason for prescription of gastric acid-reducing therapy. Furthermore, we noted potential drug-drug interaction and whether risk had been documented and mitigated. Of 701 patients on antiretroviral therapy, 67 (9.6%) were prescribed gastric acid-reducing therapy. Of these, the majority (59/67 [88.1%]) were prescribed proton pump inhibitors. We identified four potential drug-drug interactions, which were appropriately managed by temporally separating the administration of gastric acid-reducing agent and antiretroviral therapy, and all four of these patients remained virally suppressed. Gastric acid-reducing therapy, in particular proton pump inhibitor therapy, appears common in patients prescribed antiretroviral therapy. Whilst there remains a paucity of published data, our findings are comparable to those in other European cohorts. Pharmacovigilance of drug-drug interactions in HIV-positive patients is vital. Education of patients and staff, and accurate data-gathering tools, will enhance patient safety.

Keywords

HIV AIDS treatment antiretroviral therapy acid reducing therapy proton pump inhibitors protease inhibitors drug-drug interactions

Introduction and background

Clinically-significant drug-drug interactions (DDIs) involving antiretroviral (ARV) medications are common, occurring in up to 40% of HIV patients on treatment.¹ The consequences of DDI include fluctuations in ARV plasma levels leading to toxicity, diminished efficacy and increased risk of ARV resistance. In the current era of an ageing HIV cohort² polypharmacy and therefore DDIs are likely to become more prevalent.³

One example of a class of drug with potential for DDI is gastric acid–reducing agents. The bioavailability of ARVs may be affected by gastric pH (e.g. atazanavir, rilpivirine) or else reduced through binding to divalent cations (raltegravir, elvitegravir, dolutegravir). Gastric acid–reducing agent use is common; proton pump inhibitors (PPIs) are some of the most prescribed drugs in the United Kingdom (UK),⁴ with 33.2 million primary care prescriptions in England for lansoprazole and omeprazole combined in 2009.⁵ Both ranitidine and omeprazole are available over the counter in the UK.⁶ The easy availability and frequent prescription of gastric acid-reducing agents is therefore a concern in the context of HIV treatment; of all groups of medications that interact with ARVs, gastrointestinal medications are among the most frequently co-prescribed.⁷ Despite this, there are very few published data regarding the rate of prescription of gastric acid-reducing agents in HIV-positive cohorts, and none from the UK.

Aims and methods

The HIV service at the Royal Liverpool University Hospital, UK, provides care to approximately 1100 patients, with numbers increasing at a rate of 100 patients per year. As part of an on-going assessment of service provision, safety and quality we assessed the frequency of prescription of gastric acid-reducing agents in patients prescribed combination antiretroviral therapy (cART). We examined locally-held electronic clinic letters for all patients taking cART within the HIV service in December 2013 and recorded current cART regimen and concomitantly prescribed or over-the-counter gastric acid-reducing medications. In patients who received gastric acid-reducing agents, we evaluated the potential for clinically significant DDI and whether this had been adequately managed by the prescriber, using the Liverpool HIV drug interactions resource.⁸ We also recorded the indication documented for the use of gastric acid-reducing therapy, if this was available.

Table 1 describes the DDIs that are considered to be potentially significant. There is a potential for interaction between raltegravir and PPIs; plasma concentrations of the former have been shown to be 3–4 times increased in healthy volunteers taking concomitant omeprazole.⁹ No clinical toxicity has been reported from this interaction, however, and so we have not included this as a potentially significant DDI in our analysis.

Table 1.

Significant DDI between gastric acid–reducing therapy and ARVs (adapted from Liverpool HIV interaction resource)⁸ – indicates no interaction, + indicates potential for interaction and ++ indicates serious interaction.

	PPI	H2 receptor antagonist	Antacid
Atazanavir	++	+	+
Darunavir	–	–	–
Lopinavir	–	–	–
Saquinavir	–	–	–
Efavirenz	–	–	–
Etravirine
Nevirapine	–	–	–
Rilpivirine	++	+	+
Raltegravir	+	–	+
Maraviroc	–	–	–
Abacavir	–	–	–
Emtricitabine	–	–	–
Lamivudine	–	–	–
Tenofovir	–	–	–

ARV: antiretroviral; DDI: drug-drug interaction; PPI: proton pump inhibitor.

Results

Rates of gastric acid-reducing medication use

Of our cohort, 701 patients were on ART. Of these, 67 (9.6%) were taking gastric acid-reducing therapy. The vast majority of these patients – 59/67 (88.1%) – were taking PPIs. In all, 8/67 (12.0%) were taking H2-receptor antagonists and 2/67 (3.0%) were taking simple antacids containing alginate and calcium carbonate and/or magnesium. Table 2 illustrates the proportion of gastric acid-reducing therapy prescribed by cART regimen.

Table 2.

Gastric acid-reducing therapy and proton pump inhibitor (PPI) use across different combination antiretroviral therapy (cART) regimens.

cART backbone	Total patients	Number of patients taking PPI	Number of patients taking H2 antagonist	Number of patients taking antacid	Total number of patients taking any acid-reducing therapy
Darunavir	184	23 (12.5%)	1 (0.5%)	0 (0%)	24 (13%)
Atazanavir	34	0 (0%)	1 (2.9%)	0 (0%)	1 (2.9%)
Lopinavir	33	2 (6.1%)	0 (0%)	0 (0%)	2 (6.1%)
Saquinavir	1	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Total – PI	252	25 (9.9%)	2 (0.8%)	0 (0%)	27 (10.7%)
Efavirenz	273	21 (7.7%)	1 (0.4%)	1 (0.4%)	23 (8.4%)
Nevirapine	60	6 (10.0%)	1 (1.7%)	0 (0%)	7 (11.7%)
Rilpivirine	60	0 (0.0%)	2 (3.3%)	0 (0%)	2 (3.3%)
Etravirine	16	0 (0.0%)	0 (0.0%)	0 (0%)	0 (0.0%)
Total – NNRTI	409	27 (6.6%)	4 (9.8%)	1 (0.2%)	32 (7.8%)
Raltegravir	31	5 (16.1%)	0 (0.0%)	1 (3.2%)	6 (19.4%)
Others (inc Maraviroc)	9	2 (22.2%)	0 (0.0%)	0 (0%)	2 (22.2%)
Total – other cART	40	7 (17.5%)	0 (0%)	1 (2.5%)	8 (20%)
Total – patients on cART	701	59 (8.4%)	6 (0.9%)	2 (0.3%)	67 (9.6%)

PI: protease inhibitor; NNRTI: non-nucleoside reverse transcriptase inhibitor; cART: combination antiretroviral therapy; PPI: proton pump inhibitor.

Potential drug-drug interactions

We identified only four potential DDIs between gastric acid-reducing agents and ARVs in the cohort, with no serious DDIs identified. In all cases a potential interaction had been identified by the clinician and doses timed to reduce the risk of the interaction occurring. One patient taking atazanavir and two patients taking rilpivirine were taking concomitant H2-receptor antagonists (ranitidine); in all cases the dose of ranitidine was appropriately separated in time from the ARV dose. One patient taking raltegravir was taking concomitant antacid; these doses were also appropriately separated in time. All four of the patients with potential DDI were virologically suppressed. Five patients were prescribed concomitant raltegravir and PPIs but, as described above, we have not included this as a potentially significant DDI in our analysis.

Indication for gastric acid-reducing agent

The indication for gastric acid-reducing agent was recorded in 42/67 (62.7%) of patients taking these medications. The overwhelming majority of recorded indications were for treatment of symptomatic gastritis, oesophagitis or peptic ulcer disease (40/42 [95.2%]); the remainder were for primary prophylaxis of gastrointestinal bleeding (2/42 [4.8%]).

Discussion

Rates of gastric acid-reducing therapy use

Gastric acid-reducing therapy is common in our large cohort of patients. There are no peer-reviewed published data on the rates of gastric acid-reducing therapy use in HIV-positive UK cohorts for comparison; a German study in 2007 using self-administered questionnaires found that, of 424 HIV-positive patients completing the survey, 92 (22%) were taking gastric acid-reducing therapy.¹⁰ Data on the type of acid-reducing agent is presented only for those patients taking HIV protease inhibitor (PI)-based cART; 201 of the 424 patients were taking PI-based cART, and of these, 74 (37%) had taken gastric acid-reducing agents in the last 6 months. The majority of these (78/74 [65%]) were PPIs. Differing study methodologies make direct comparison with our data problematic, but it seems reasonable to conclude that gastric acid-suppressing therapy and PPI use is common amongst European HIV-positive people on ARVs.

Potentially significant DDIs were rare in our cohort, and where they exist they have been mitigated against by temporal separation of doses. This likely represents good clinician awareness of potential DDIs, and highlights the importance of educating health care staff in DDI related to ARVs, as well as a thorough medication history of both prescribed and over-the-counter medications.

Drug-drug interactions: HIV PIs

There is in general a paucity of published data regarding the interactions between ARVs and gastric acid-reducing therapy. One exception to this is the clear pharmacokinetic data that atazanavir absorption is reduced by gastric acid-reducing agents in HIV patients and healthy volunteers.¹¹ Its absorption declines with increasing pH, and so its bioavailability is adversely affected by all agents that reduce gastric acid – though the effect is most pronounced with PPIs.¹² Current advice from the European Medicines Agency is to avoid the combination. For H2-receptor antagonists the manufacturer suggests that these can be safely co-administered if administration of boosted atazanavir is at least 10 hours after and 2 hours before H2-receptor antagonists.¹³ This strategy is supported by pharmacokinetic data (with famotidine) in HIV-positive and -negative people.^13,14 Adherence to this guidance is reflected in our data, with the one patient who was taking boosted atazanavir and ranitidine separating their administration in time, and remaining virologically suppressed.

Levels of darunavir and lopinavir are thought to be unaffected by concomitant administration of gastric acid-reducing agents. There is pharmacokinetic evidence that lopinavir absorption is unaffected by gastric acid-reducing agents¹¹ and one published peer-reviewed pharmacokinetic study in 18 healthy volunteers that showed unchanged pharmacokinetic parameters when darunavir was administered with ranitidine or omeprazole.¹⁵

Drug-drug interactions: non-nucleoside reverse transcriptase inhibitors

Rilpivirine’s absorption is related to gastric pH and any gastric acid-reducing agent will reduce the bioavailability of rilpivirine.¹⁶ There are data from pharmacokinetic studies in healthy volunteers that omeprazole coadministration reduces rilpivirine levels to a subtherapeutic dose¹⁷ and the manufacturer advises against coadministration of PPIs with rilpivirine.¹⁶ H2 receptor antagonists are considered safe so long as they are administered 12 hours before or 4 hours after rilpivirine,¹⁶ though data are limited and there are only pharmacokinetic data in healthy HIV-negative volunteers to support this strategy with famotidine.¹⁷ There are no published peer-reviewed studies examining the coadministration of antacid preparations and rilpivirine, but the manufacturers suggest that antacids can be safely administered 2 hours before or 4 hours after rilpivirine.¹⁶ There are no published data, pharmacokinetic or otherwise, on outcomes in HIV-positive people taking rilpivirine and gastric acid-reducing therapy. The two patients in our cohort who were taking rilpivirine and concomitant ranitidine were appropriately separating the administration in time, and were virologically suppressed.

Data on coadministration of the other non-nucleoside reverse transcriptase inhibitor (NNRTIs) with gastric acid-reducing medication are sparse but they are thought to be safe. Administration of efavirenz¹⁸ and nevirapine¹⁹ with simple antacids in healthy volunteers does not alter their plasma levels; there are no data on their administration with PPIs or H2-receptor antagonists. Etravirine levels are increased by coadministration with omeprazole and ranitidine in healthy volunteers,²⁰ but this is well tolerated.

Drug-drug interactions: Raltegravir

Absorption of the integrase inhibitor raltegravir is reduced by binding to divalent cations, as found in magnesium- and aluminium-containing antacids.²¹ This DDI is not thought to be due to changes in gastric pH. The manufacturers suggest that magnesium- or aluminium-containing antacids should not be taken within 2 hours of raltegravir. Raltegravir is better absorbed at a higher pH and co-administration of ranitidine and omeprazole has been shown to increase levels of raltegravir in healthy volunteers.⁹ This was well tolerated in the healthy volunteers and a subgroup analysis of raltegravir phase III trials did not find any association between PPI or H2-receptor antagonist use and adverse events.⁹

One patient in our cohort was taking raltegravir and antacid, appropriately separated in time, and was virally suppressed. Five patients were taking raltegravir and a PPI, with no adverse effects reported.

Drug-drug interactions: NNRTIs and maraviroc

There are no published peer-reviewed data on the outcomes of coadministration of antacids, H2-receptor antagonists or PPIs and tenofovir, abacavir, lamivudine, emtricitabine or maraviroc; their coadministration is considered safe on theoretical grounds.

Conclusion

Our study has provided an insight into gastric acid-reducing therapy use in a large HIV-positive cohort, an area where data are sparse. Gastric acid-reducing therapy use – largely PPIs – is not uncommon within our cohort, though significant DDIs remain rare, and clinician awareness of minimising the risk of potential interactions is good. This is likely to be at least in part owing to the practice of referring to the Liverpool University HIV drug interactions website⁸ at our centre. The indication for gastric acid-reducing therapy was poorly recorded, but where it was recorded it was almost exclusively for symptomatic relief rather than primary prophylaxis of GI bleed.

Our study has clear limitations; it is retrospective and is likely to under-report gastric acid-reducing medication use, if clinicians did not record all medications at each encounter, and if patients did not remember or do not report all medications, including over-the-counter medications. The patient records used to gather data do not record whether concomitant medications are prescribed or purchased over the counter, so we are unable to provide a breakdown of where and how patients in our cohort are obtaining gastric acid-reducing medications. Furthermore, the number of potential DDIs are small and so it is impossible to draw any robust conclusions about the best strategies for achieving good virological and clinical outcomes in the face of potential DDIs – although adherence to the manufacturer’s guidance for temporal separation of medications was achieved in all cases in our cohort where we identified potential DDIs.

The data available to guide clinicians in the management of DDIs are largely pharmacokinetic studies in small populations of healthy volunteers. Virological and clinical outcome data are often lacking. Furthermore, the magnitude of DDI might be different in HIV-positive as compared to HIV-negative people, as alterations in gastric pH have been reported for HIV-infected patients.²² Further audit of large local or national cohorts could provide increased patient numbers and observational outcome data to complement these PK studies and guide clinical practice.

In conclusion, pharmacovigilance of those caring for patients on cART remains vital, and DDIs remain a serious barrier to effective treatment for those on ARVs. However, it is difficult to draw conclusions about the significance of drug interactions with commonly prescribed medications in the absence of good quality data describing the prevalence of their use, or clinically relevant outcome data. Local audit, as presented here, and potential future national audit are tools that can identify the scale of the problem, and provide pragmatic clinically relevant outcome data to complement pharmacokinetic studies and guide clinical practice.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Marzolini

Elzi

Gibbons

, et al. Prevalence of comedications and effect of potential drug-drug interactions in the Swiss HIV Cohort Study. Antivir Ther 2010; 15: 413–423.

Nguyen

Holodniy

. HIV infection in the elderly. Clin Interv Aging 2008; 3: 453–472.

Holtzman

Armon

Tedaldi

, et al. Polypharmacy and risk of antiretroviral drug interactions among the aging HIV-infected population. J Gen Intern Med 2013; 28: 1302–1310.

Prescriptions Dispensed in the Community, Statistics for England – 2003-13. The Health and Social Care Information Centre, UK, 2014.

Prescriptions Dispensed in the Community: Statistics for England – 1999-2009. The Health and Social Care Information Centre, UK, 2010.

Medicines and Healthcare Products Regulatory Agency. List B: consolidated list of substances in authorised medicines for general sale, http://www.mhra.gov.uk/Howweregulate/Medicines/Licensingofmedicines/Legalstatusandreclassification/Listsofsubstances/#l2 (2010, accessed Feb 2014).

Seden

Bradley

Miller

, et al. The clinical utility of HIV outpatient pharmacist prescreening to reduce medication error and assess adherence. Int J STD AIDS 2013; 24: 237–241.

Liverpool University HIV drug interaction resource, www.hiv-druginteractions.org (accessed December 2013).

Iwamoto

Wenning

Nguyen

B-Y

, et al. Effects of omeprazole on plasma levels of raltegravir. Clin Infect Dis 2009; 48: 489–492.

10.

van Lunzen

Liess

Arastéh

, et al. Concomitant use of gastric acid-reducing agents is frequent among HIV-1-infected patients receiving protease inhibitor-based highly active antiretroviral therapy. HIV Med 2007; 8: 220–225.

11.

Béïque

Giguère

la Porte

, et al. Interactions between protease inhibitors and acid-reducing agents: a systematic review. HIV Med 2007; 8: 335–345.

12.

Reyataz Summary of Product Characteristics, Bristol-Myers Squibb Pharmaceuticals Ltd, April 2012.

13.

Agarwala

Thal

Nettles

, et al. Further information on the administration of h2-receptor antagonists with atazanavir. J Acquir Immune Defic Syndr 2006; 42: 516–516.

14.

Wang

Boffito

Zhang

, et al. Effects of the H2-receptor antagonist famotidine on the pharmacokinetics of atazanavir-ritonavir with or without tenofovir in HIV-infected patients. Clin Epidemiol Res 2011; 25: 509–515.

15.

Sekar

Lefebvre

De Paepe

, et al. Pharmacokinetic interaction between darunavir boosted with ritonavir and omeprazole or ranitidine in human immunodeficiency virus-negative healthy volunteers. Antimicrob Agents Chemother 2007; 51: 958–961.

16.

Edurant Summary of Product Characteristics, Janssen-Cilag Ltd, November 2011.

17.

Crauwels

van Heeswijk

Stevens

. Clinical perspective on drug-drug interactions with the non-nucleoside reverse transcriptase inhibitor rilpivirine. AIDS Rev 2013; 15: 87–101.

18.

Sustiva Summary of Product Characteristics, Bristol-Myers Squibb Pharmaceuticals Ltd, July 2011.

19.

Viramune Summary of Product Characteristics, Boehringer Ingelheim International GmbH, January 2012.

20.

Schöller-Gyüre

Kakuda

De Smedt

, et al. A pharmacokinetic study of etravirine (TMC125) co-administered with ranitidine and omeprazole in HIV-negative volunteers. Br J Clin Pharmacol 2008; 66: 508–516.

21.

Isentress Summary of Product Characteristics, Merck Sharp & Dohme Ltd, August 2013.

22.

Welage

Carver

Revankar

, et al. Alterations in gastric acidity in patients infected with human immunodeficiency virus. Clin Infect Dis 1995; 21: 1431–1438.