HIV: Emerging antiretroviral drug interactions

Emerging antiretroviral drug interactions
.
J. Antimicrob. Chemother. (2011) 66 (2): 235-239. doi: 10.1093/jac/dkq448 First published online: December 5, 2010

Emerging antiretroviral drug interactions
Journal of Antimicrobial Chemotherapy, 11-Jan-2011
.
Full Text (PDF)

Paul A. Pham1,* and
Charles Flexner2
+ Author Affiliations

1Division of Infectious Diseases, Johns Hopkins University, School of Medicine, Baltimore, MD 21287, USA

2Division of Clinical Pharmacology, Johns Hopkins University, School of Medicine, Baltimore, MD 21287, USA
*Corresponding author.

Abstract

With HIV-infected patients living longer and recommendations to initiate antiretrovirals (ARVs) being made earlier, the likelihood for potential drug–drug interactions between ARVs and concurrent medications used to manage co-morbid conditions will increase. In order to maximize the clinical benefit and minimize potential toxicity of ARVs and co-administered medications, it is important for clinicians to recognize significant drug–drug interactions. This article highlights clinically significant drug–drug interactions with antituberculosis agents, antimalarials, anticoagulants, chemotherapeutic agents and pulmonary antihypertensive agents when they are co-administered with newer ARVs (e.g. darunavir, raltegravir, maraviroc and etravirine).
Key words

Introduction
Management of antiretroviral (ARV) drug–drug interactions remains a critical component of HIV care. The US Department of Health and Human Services (DHHS) guidelines now recommend treating HIV-infected patients with CD4 counts less than 500 cells/mm3 and to consider treatment in patients with CD4 counts greater than 500 cells/mm3.1 This decision was driven by findings from large observational cohort studies showing mortality benefit when treatment is started before CD4 counts drop below 350 cells/mm3.2 With more patients starting ARVs earlier, the likelihood for potential drug–drug interactions between ARVs and concurrent medications used to manage co-morbid conditions will increase. In order to maximize the clinical benefit and minimize potential toxicity of ARVs and co-administered medications, it is important for clinicians to recognize significant drug–drug interactions. Here we highlight clinically significant drug–drug interaction involving newer ARVs (e.g. darunavir, raltegravir, maraviroc and etravirine) and several classes of drugs now commonly combined with ARVs.

Drug interactions with drugs used to treat tuberculosis and malaria
Tuberculosis remains one of the most common co-morbid conditions affecting HIV-infected patients, especially in sub-Saharan Africa. A randomized study comparing early highly active ARV therapy (HAART) initiation (within 2 months of tuberculosis therapy) versus deferred treatment (starting HAART after completion of antituberculosis agents) found a 55% reduction in mortality in patients treated with HAART within 2 months of starting tuberculosis therapy.3 Although there is clinical benefit in starting HAART concurrently with tuberculosis therapy, the cytochrome P450 induction properties of rifamycins, especially rifampicin, may lead to significant decreases in concentrations of HIV protease inhibitors (PIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs).


Rifampicin decreases PI concentrations by as much as 80%, and therefore co-administration is contraindicated.4 Attempts to overcome rifampicin induction by doubling the dose of some PIs such as lopinavir/ritonavir resulted in significant gastrointestinal intolerance and hepatotoxicity.5 With lopinavir/ritonavir co-administration, dose-adjusted rifabutin (150 mg once every other day) has been recommended; however, this approach may lead to sub-therapeutic rifabutin concentrations in some patients.6 Although serum concentrations of dose-adjusted rifabutin when co-administered with darunavir/ritonavir appear adequate, monitoring of rifabutin concentrations should be considered in patients with a high body mass index (BMI) and in patients with advanced tuberculosis to prevent rifamycin resistance. Rifampicin decreases the maraviroc area under the concentration–time curve (AUC) by 78%; although the maraviroc dose could be increased to 600 mg twice daily to overcome this interaction, there are no clinical data, and this combination should be avoided if possible.4 Rifabutin, a less potent CYP3A4 inducer, is likely to cause only modest decreases in maraviroc concentrations.
Rifampicin induction of glucuronidation results in raltegravir AUC and trough decreases of 40% and 61%, respectively. Doubling the dose of raltegravir increases the AUC, but trough concentrations are still decreased.4 Since there is no clinical correlation between raltegravir trough concentrations and virological outcome, the clinical significance of this interaction is not known. More clinical data are needed before raltegravir can be recommended with rifampicin. Rifabutin does not significantly interact with raltegravir and is preferred over rifampicin in HIV/tuberculosis co-infected patients.7

Malaria and HIV co-infections are prevalent in many resource-limited countries. Since many antimalarial drugs are metabolized by cytochrome P450 isoenzymes, there are potential drug–drug interactions with PIs and NNRTIs. Clinical data with ARVs and antimalarial drugs are limited to several small pharmacokinetic studies. Mefloquine, lumefantrine and artemisinin derivatives are CYP3A4 substrates, therefore serum concentrations are expected to increase with darunavir/ritonavir and decrease with etravirine co-administration. However, co-administration of mefloquine with ritonavir did not significantly affect mefloquine serum concentrations.8 This finding suggests the important role of other mefloquine metabolic pathways in vivo and/or ritonavir mixed CYP3A4 inhibitory and induction properties. On the other hand, the lumefantrine AUC was increased by 193% with lopinavir/ritonavir co-administration.8 Although there were concerns for QTc prolongation with high lumefantrine concentrations, the risk is low in patients without underlying cardiovascular risk factors.9 Significant drug–drug interactions between antimalarial drugs and maraviroc or raltegravir are unlikely due to the lack of significant inhibition or induction of drug metabolizing enzymes or drug transport proteins. Additional pharmacokinetic and clinical data are needed to guide dosing recommendations when antimalarial drugs are co-administered with ARVs.

Interactions with anticoagulants
An older HIV-infected population and more patients with thrombotic events means a greater need for anticoagulation. Warfarin exists in a racemic mixture of two isomers that differ in potency and route of metabolism. S-warfarin is 2 to 5 times more potent than R-warfarin. Although the more potent S-isomer is a CYP2C9 substrate, the R-isomer is metabolized via CYP1A2 and CYP3A4.10,11 Although the majority of significant warfarin drug interactions have been attributed to inhibition or induction of CYP2C9, affecting CYP1A2 and CYP3A4 can also lead to significant changes in warfarin concentrations. Several case reports of significant drug–drug interactions between PIs, NNRTIs and warfarin have been described. Increased warfarin requirements have been reported with nevirapine and lopinavir/ritonavir co-administration.12 Since ritonavir is a mixed CYP3A4 inhibitor and inducer and a mild CYP2C9 inhibitor, single-dose co-administration may result in supra-therapeutic international normalized ratio (INR). However, at steady state, sub-therapeutic warfarin concentrations as a result of CYP3A4 enzyme induction may occur.13–15 It is important to note that these reports used higher than currently recommended doses of ritonavir. Darunavir/ritonavir (600/100 mg twice daily) also decreases the mean S-warfarin AUC by 29%.16 Although both efavirenz and etravirine inhibit CYP2C9 and 2C19 in vitro, at steady state efavirenz mainly induces CYP2C9 and CYP2C19 substrates (e.g. voriconazole) in humans.17 This may lead to an increase in warfarin requirements. On the other hand, etravirine is a mild inhibitor of CYP2C9 in vivo.18 Although S-warfarin concentrations are not significantly increased with etravirine co-administration, the ratio of S-warfarin parent drug to 7-OH-S-warfarin metabolites increased by 82%.18 The clinical significance of this interaction remains to be determined. Significant interactions with warfarin are unlikely to occur with raltegravir and maraviroc since they do not affect CYP2C9, 1A2 or 3A4. Due to the mixed cytochrome P450 enzyme induction and inhibition properties of PIs and NNRTIs, predictions of warfarin requirements are difficult. Close INR monitoring should be performed when initiating warfarin in patients on PI- and NNRTI-based regimens and when a PI or NNRTI are discontinued in patients on a stable warfarin dose.
Omeprazole, a CYP2C19 inhibitor, decreases the antiplatelet activity of clopidogrel by inhibiting the biotransformation of the clopidogrel prodrug into its active metabolite.19 In patients who have been hospitalized for acute coronary syndrome, this interaction is associated with a 27% increased risk of death or rehospitalization.20 By analogy, inhibition of CYP2C19 by etravirine may also inhibit clopidogrel antiplatelet activity. Until more data become available, the co-administration of CYP2C19 inhibitors (e.g. etravirine and azoles) and clopidogrel is not recommended.21


Interactions with chemotherapeutic agents
With effective HAART, the incidence of non-Hodgkin's lymphoma (NHL) and Kaposi's sarcoma has significantly declined; however, AIDS patients remain at increased risk for other cancers, including Hodgkin's lymphoma.22 Before the availability of HAART, response to chemotherapy and prognosis of NHL were poor. A retrospective study comparing survival rates between patients diagnosed with NHL pre-HAART (1988–1996) and post-HAART (after 1996) found a 47% improvement in lymphoma-free survival in patients treated with chemotherapy plus HAART.23 Similarly, a multi-cohort European study identified 1176 patients with NHL and found a 1 year survival rate of 66% for systemic NHL in the HAART era.24 For Hodgkin's lymphoma, the combination of chemotherapy plus HAART produced a significantly better 2 year overall survival compared with patients not receiving HAART (74% versus 30%, P < 0.001).25
Since many chemotherapeutic agents are CYP3A4 substrates, there is a potential for drug–drug interactions with HIV PIs and NNRTIs. One retrospective study found a higher incidence of anaemia, leucopenia and peripheral neuropathy in patients treated with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) plus PI-based HAART.26 An observational study evaluated 65 patients with intermediate- to high-grade HIV-associated NHL treated with full-dose CHOP or modified CHOP (half-dose cyclophosphamide and doxorubicin, combined with full-dose vincristine and prednisone); although cyclophosphamide clearance was decreased by 50%, no significant difference in toxicity between the two regimens was observed. However, complete response was 30% versus 48% in the modified CHOP and full-dose CHOP, respectively.27 The German AIDS-related Lymphoma Study Group investigated PI-based HAART administered concomitantly with CHOP in 72 patients who were stratified into standard- and high-risk groups.

Atazanavir and indinavir are also potent inhibitors of glucuronidation as well as CYP3A4 metabolism. This may result in significant elevation of irinotecan concentrations and increase the risk of severe neutropenia. Conversely, CYP3A4 enzyme inducers such as NNRTIs (e.g. efavirenz, nevirapine and etravirine) can potentially decrease concentrations of chemotherapeutic agents that are 3A4 substrates (e.g. doxorubicin, cyclophosphamide and vincristine) and potentially decrease chemotherapy efficacy. A list of potential drug–drug interactions between ARVs and chemotherapy is provided in Table 1.


View this table:
In this window
In a new window
Table 1.

Potential drug–drug interactions with chemotherapeutic agents
Clinicians are frequently faced with a clinical dilemma of switching to an alternative HAART regimen or stopping HAART during chemotherapy cycles. Based on improved survival rates associated with continuing HAART in combination with chemotherapy, stopping HAART is generally not recommended. Since raltegravir and maraviroc are not inhibitors or inducers of cytochrome P450, they are unlikely to interact with chemotherapeutic agents. If the ARV regimen contains zidovudine, prevention of additive bone marrow suppression by switching to an alternative nucleoside reverse transcriptase inhibitor (NRTI) is recommended. Guidelines on empirical dose adjustment of chemotherapy when co-administered with interacting ARVs need to be developed. Close monitoring for potential reduced chemotherapy efficacy or increased toxicity when NNRTIs (e.g. efavirenz, nevirapine and etravirine) and PIs (e.g. darunavir, atazanavir, lopinavir and ritonavir) are co-administered is critical.


Interactions with agents for pulmonary hypertension
Pulmonary hypertension may be associated with HIV infection. In a case series, 8.1% of patients presenting with cardiopulmonary complaints were diagnosed with primary pulmonary hypertension.29 In a retrospective analysis, treated patients with both HIV and pulmonary hypertension had favourable survival rates, with an overall survival of 88% at 1 year and 72% at 3 years.30 Drug interactions with agents used to treat pulmonary hypertension are possible (Table 2).

View this table:
In this window
In a new window
Table 2.

Drug–drug interactions with agents for pulmonary hypertension
The co-administration of bosentan (a CYP3A4 substrate) and lopinavir/ritonavir resulted in a 48-fold increase in bosentan serum concentrations on day 4, but a 5-fold increase at steady state on day 10. Therefore co-administration is only recommended when lopinavir/ritonavir has reached steady state. In patients who have been treated with lopinavir/ritonavir for more than 10 days, low-dose bosentan (62.5 mg once daily or every other day) can be co-administered. In patients on bosentan who need to be started on lopinavir/ritonavir, discontinuation of bosentan for at least 36 h prior to initiating lopinavir/ritonavir, then reinstituting low-dose bosentan after lopinavir/ritonavir has reached steady state (10 days) is recommended.1 Although there are no data with other PIs, a similar dosing approach should be considered, with close monitoring. In order to prevent rebound pulmonary hypertension, an alternative agent should be given during the transition period. With the proper dose adjustment, bosentan co-administration with HAART resulted in symptomatic improvements (e.g. better 6 min walk distance), decreased pulmonary vascular resistance and favourable overall survival.31 Ambrisentan, an alternative endothelin-1 receptor antagonist, is metabolized by CYP3A, CYP2C19 and glucuronosyltransferases. There is a 35% increase in ambrisentan AUC with co-administration of ketoconazole, a moderate inhibitor of CYP3A4.32 Since there is only a modest increase in ambrisentan concentrations when co-administered with CYP3A4 inhibitors, co-administration with PI-based HAART can be considered with close monitoring in patients unable to tolerate bosentan.

PIs increase sildenafil serum concentrations by up to 11-fold.33 Although low-dose sildenafil (25 mg every 48 h) can be co-administered with PI for the treatment of erectile dysfunction, high-dose sildenafil is not recommended with PI co-administration due to an increased risk of severe hypotension.1 For the treatment of pulmonary hypertension in patients on PIs, clinicians can consider reducing the dose of sildenafil to 10 mg/day with slow titration. In severe pulmonary hypertension, epoprostenol should be considered. Epoprostenol is hydrolysed non-enzymatically in plasma to 6-keto-prostaglandin F-1 alpha, with low potential for drug–drug interactions.34 The standard epoprostenol dose can be given with PIs and NNRTIs.

Conclusions
The management of HIV infection continues to evolve and drug–drug interactions are becoming increasingly complex. Although drug interaction databases (e.g. http://www.hopkins-hivguide.org/ and http://www.clinicalcareoptions/) are available to assist clinicians with the management of these interactions, clinicians should also recognize drugs with the highest potential for interactions. Close monitoring is critical when potent CYP3A4 inducers (e.g. rifampicin and most NNRTIs) or potent inhibitors (e.g. PIs and azole antifungals) are added or removed from a medical regimen. As HIV-infected patients grow older, drug interactions arising from the management of other health problems will be more common in the future.


Transparency declarations
P. A. P. does not have any conflicts of interest. C. F. reports having served as a consultant to Bristol-Myers Squibb, GlaxoSmithKline, Merck, Roche, Schering-Plough and Tibotec, and has received honoraria for presentations at meetings sponsored in part by Abbott Laboratories.

© The Author 2010. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com
Previous Section

References
↵
Department of Health and Human Services. Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. December 1, 2009. http://aidsinfo.nih.gov/contentfiles/AdultandAdolescentGL.pdf (18 July 2010, date last accessed).
↵
Kitahata MM,
Gange SJ,
Abraham AG,
et al. Effect of early versus deferred antiretroviral therapy for HIV on survival. N Engl J Med 2009;360:1815-26. doi:10.1056/NEJMoa0807252.
CrossRefMedline
↵
Abdool Karim SS,
Naidoo K,
Grobler A,
et al. Timing of initiation of antiretroviral drugs during tuberculosis therapy. N Engl J Med. 2010.362 p. 697-706.
↵
Sterling TR,
Pham PA,
Chaisson RE. HIV infection-related tuberculosis: clinical manifestations and treatment. Clin Infect Dis 2010;50(Suppl 3):S223-30. doi:10.1086/651495.
Abstract/FREE Full Text
↵
Centers for Disease Control and Prevention. Managing Drug Interactions in the Treatment of HIV-Related Tuberculosis. http://www.cdc.gov/tb/publications/guidelines/TB_HIV_Drugs/default.htm (25 March 2010, date last accessed).
↵
Boulanger C,
Hollender E,
Farrell K,
et al. Pharmacokinetic evaluation of rifabutin in combination with lopinavir-ritonavir in patients with HIV infection and active tuberculosis. Clin Infect Dis 2009;49:1305-11. doi:10.1086/606056.
Abstract/FREE Full Text
↵
Brainard DM,
Petry AS,
Fang L,
et al. Lack of a clinically important effect of rifabutin on raltegravir pharmacokinetics. Abstracts of the Forty-Ninth Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, CA, 2009. Washington, DC, USA: American Society for Microbiology; Abstract A1-1296.
↵
Dooley KE,
Flexner C,
Andrade AS. Drug interactions involving combination antiretroviral therapy and other anti-infective agents: repercussions for resource-limited countries. J Infect Dis 2008;198:948-61. doi:10.1086/591459.
FREE Full Text
↵
van Vugt M,
Ezzet F,
Nosten F,
et al. No evidence of cardiotoxicity during antimalarial treatment with artemether-lumefantrine. Am J Trop Med Hyg 1999;61:964-7.
Abstract
↵
Rettie AE,
Korzekwa KR,
Kunze KL,
et al. Hydroxylation of warfarin by human cDNA-expressed cytochrome P-450: a role for P-4502C9 in the etiology of (S)-warfarin-drug interactions. Chem Res Toxicol 1992;5:54-9. doi:10.1021/tx00025a009.
CrossRefMedlineWeb of Science
↵
Kaminsky LS,
Zhang ZY. Human P450 metabolism of warfarin. Pharmacol Ther 1997;73:67-74. doi:10.1016/S0163-7258(96)00140-4.
CrossRefMedlineWeb of Science
↵
Liedtke MD,
Rathbun RC. Warfarin–antiretroviral interactions. Ann Pharmacother 2009;43:322-8. doi:10.1345/aph.1L497.
Abstract/FREE Full Text
↵
Kumar GN,
Rodrigues AD,
Buko AM,
et al. Cytochrome P450-mediated metabolism of the HIV-1 protease inhibitor ritonavir (ABT-538) in human liver microsomes. J Pharmacol Exp Ther 1996;277:423-31.
Abstract/FREE Full Text
Newshan G,
Tsang P. Ritonavir and warfarin interaction. AIDS 1999;13:1788-9. doi:10.1097/00002030-199909100-00028.
CrossRefMedlineWeb of Science
↵
Knoell KR,
Young TM,
Cousins ES. Potential interaction involving warfarin and ritonavir. Ann Pharmacother 1998;32:1299-302. doi:10.1345/aph.17456.
Abstract
↵
Darunavir [package insert]. East Bridgewater, NJ: Tibotec Therapeutics; 2010. April.
↵
Liu P,
Foste G,
LaBadie RR,
et al. Pharmacokinetic interaction between voriconazole and efavirenz at steady state in healthy male subjects. J Clin Pharmacol 2008;48:73-84. doi:10.1177/0091270007309703.
Abstract/FREE Full Text
↵
Schöller-Gyüre M,
Kakuda TN,
Stevens T,
et al. Effect of etravirine on cytochrome P450 isozymes assessed by the Cooperstown 5 + 1 cocktail. Abstracts of the Forty-Eighth Interscience Conference on Antimicrobial Agents, Washington, DC, USA, 2008. Washington, DC, USA: American Society for Microbiology; Abstract A-955.
↵
Gilard M,
Arnaud B,
Cornily J,
et al. Influence of omeprazole on the antiplatelet action of clopidogrel associated with aspirin. J Am Coll Cardiol 2008;51:256-60. doi:10.1016/j.jacc.2007.06.064.
Abstract/FREE Full Text
↵
Ho MP,
Maddox TM,
Wang L,
et al. Risk of adverse outcomes associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome. JAMA 2009;301:937-44. doi:10.1001/jama.2009.261.
Abstract/FREE Full Text
↵
Clopidogrel [package insert]. Bridgewater, NJ: Bristol-Myers Squibb/Sanofi Pharmaceuticals Partnership; 2009. May.
↵
Engels EA,
Pfeiffer RM,
Goedert JJ,
et al. Trends in cancer risk among people with AIDS in the United States 1980–2002. AIDS 2006;20:1645-54. doi:10.1097/01.aids.0000238411.75324.59.
MedlineWeb of Science
↵
Navarro JT,
Ribera JM,
Oriol A,
et al. Influence of highly active anti-retroviral therapy on response to treatment and survival in patients with acquired immunodeficiency syndrome-related non-Hodgkin's lymphoma treated with cyclophosphamide, hydroxydoxorubicin, vincristine and prednisone. Br J Haematol 2001;112:909-15. doi:10.1046/j.1365-2141.2001.02656.x.
CrossRefMedlineWeb of Science
↵
Bohlius J,
Schmidlin K,
Costagliola D,
et al. Prognosis of HIV-associated non-Hodgkin lymphoma in patients starting combination antiretroviral therapy. AIDS 2009;23:2029-37. doi:10.1097/QAD.0b013e32832e531c.
CrossRefMedlineWeb of Science
↵
Hentrich M,
Maretta L,
Chow KU,
et al. Highly active antiretroviral therapy (HAART) improves survival in HIV-associated Hodgkin's disease: results of a multicenter study. Ann Oncol 2006;17:914-9. doi:10.1093/annonc/mdl063.
Abstract/FREE Full Text
↵
Vaccher E,
Spina M,
di Gennaro G,
et al. Concomitant cyclophosphamide, doxorubicin, vincristine, and prednisone chemotherapy plus highly active antiretroviral therapy in patients with human immunodeficiency virus-related, non-Hodgkin lymphoma. Cancer 2001;91:155-63. doi:10.1002/1097-0142(20010101)91:1<155::AID-CNCR20>3.0.CO;2-B.
CrossRefMedline
↵
Ratner L,
Lee J,
Tang S,
et al. Chemotherapy for human immunodeficiency virus-associated non-Hodgkin's lymphoma in combination with highly active antiretroviral therapy. J Clin Oncol 2001;19:2171-8.
Abstract/FREE Full Text
↵
Weiss R,
Mitrou P,
Arasteh K,
et al. Acquired immunodeficiency syndrome-related lymphoma: simultaneous treatment with combined cyclophosphamide, doxorubicin, vincristine, and prednisone chemotherapy and highly active antiretroviral therapy is safe and improves survival-results of the German Multicenter Trial. Cancer 2006;106:1560-8. doi:10.1002/cncr.21759.
Medline
↵
Speich R,
Jenni R,
Opravil M,
et al. Primary pulmonary hypertension in HIV infection. Chest 1991;100:1268-71. doi:10.1378/chest.100.5.1268.
Abstract/FREE Full Text
↵
Degano B,
Guillaume M,
Savale L,
et al. HIV-associated pulmonary arterial hypertension: survival and prognostic factors in the modern therapeutic era. AIDS 2010;24:67-75. doi:10.1097/QAD.0b013e328331c65e.
CrossRefMedlineWeb of Science
↵
Degano B,
Yaïci A,
Le Pavec J,
et al. Long-term effects of bosentan in patients with HIV-associated pulmonary arterial hypertension. Eur Respir J 2009;33:92-8. doi:10.1183/09031936.00094808.
Abstract/FREE Full Text
↵
Richards DB,
Walker GA,
Mandagere A,
et al. Effect of ketoconazole on the pharmacokinetic profile of ambrisentan. J Clin Pharmacol 2009;49:719-24. doi:10.1177/0091270009335870.
Abstract/FREE Full Text
↵
Muirhead GJ,
Wulff MB,
Fielding A,
et al. Pharmocokinetic interactions between sildenafil and saquinavir/ritonavir. Br J Clin Pharmacol 2000;50:99-107. doi:10.1046/j.1365-2125.2000.00245.x.
CrossRefMedlineWeb of Science
↵
Rosenkranz B,
Fischer C,
Frolich JC. Prostacyclin metabolites in human plasma. Clin Pharmacol Ther 1981;29:420-4. doi:10.1038/clpt.1981.58.
CrossRefMedlineWeb of Science
↵
Tulpule A,
Groopman J,
Saville MW,
et al. Multicenter trial of low-dose paclitaxel in patients with advanced AIDS-related Kaposi sarcoma. Cancer 2002;95:147-54. doi:10.1002/cncr.10634.
CrossRefMedlineWeb of Science