In France, considering potential greater efficacy of such treatment, current guidelines for management of postexposure prophylaxis (PEP)  preferentially recommend the use of a 4-week course of triple antiretroviral therapy with two nucleoside reverse transcriptase inhibitors (NRTIs) and one protease inhibitor/ritonavir in cases of both occupational and nonoccupational exposures. To improve the efficacy of such PEP, several other factors must be considered, especially tolerability and toxicity profile to decrease the dropout rate. The tolerability of a drug regimen in the setting of prophylaxis cannot be extrapolated from data available from HIV-infected patients taking the same drugs and one author has suggested that treatment interruption is eight times higher in HIV-negative individuals compared to HIV-seropositive patients, and that the incidence of adverse events is approximately six times higher . So, specific studies must be performed to evaluate PEP tolerability.
For 10 years, we have successively evaluated the tolerability of several antiretroviral combinations used as PEP in order to find the best-tolerated PEP regimen: zidovudine/lamivudine+nelfinavir (ZDV/3TC+NFV) twice daily (b.i.d.) ; ZDV/3TC+lopinavir/ritonavir soft gel capsules (ZDV/3TC+LPV/r SGC) b.i.d. ; and lamivudine+tenofovir+atazanavir boosted by ritonavir (3TC+TDF+ATV/r) once a day . These studies were performed by using the same methodology to assess discontinuation and adverse events, permitting us to compare results. Although the dropout rate due to side effects appeared substantially higher in the ZDV/3TC+NFV PEP study compared to the other regimens, the treatment interruption remained significant in all of them.
To find a PEP regimen with a lower rate of treatment interruption due to adverse events, we then decided to test, by using the same study condition, the combination of tenofovir/emtricitabine+lopinavir/ritonavir tablet formulation (TDF/FTC+LPV/r). This choice was based on the proven efficacy of TDF when used as PEP in monkeys. Several studies have shown that 4-week, uninterrupted treatment with TDF completely prevented simian immunodeficiency virus (SIVmne) or HIV-2 infection in cynomolgus macaques if treatment was initiated within 24 h after vaginal inoculation [6–8]. Moreover, in a study realized in the largest caring center for HIV-infected and high-risk group of men having sex with men (MSM) in New England, authors found that TDF-containing regimens for nonoccupational PEP were generally well tolerated with high completion rates when compared to ZDV-containing regimens (historical controls) . Concerning LPV/r, the recent new tablet formulation showed good record in the clinical practice and advantage due to the absence of need of a separated booster. Furthermore, compared to the older soft gel formulation, the recently available LPV/r tablets yield several advantages: reduced pill burden (from 6 to 4 daily), less variability of LPV blood concentrations, and no requirement of food restriction.
We conducted a multicenter observational prospective single arm study. Duration planned for the study was 18 months. The selection of center of inclusion was on voluntary basis. In each of these centers, all adult participants (≥18 years of age) presenting to take a medical advice after an exposure associated with HIV transmission risk (e.g., occupational, accidental, sexual, needle share among intravenous drug users) in the past 48 h and who were able to give informed consent were eligible (inclusion on voluntary basis too).
Initially, the physician prescribed 3 days of medication. If the PEP was prescribed in hospitals' emergency departments, further advice was given on day 2 by a physician specializing in HIV/AIDS, who decided to stop or to complete a 4-week course of treatment after risk re-evaluation.
The study treatment was selected for first-line PEP when resistance to these drugs was neither suspected nor proven in the source patient. If the source patient was known to be infected by HIV, an AIDS specialist was immediately consulted and helped to choose a regimen taking into account the source patient's medication history and any known drug resistance.
Prior to therapy, baseline sociodemographic characteristics, description of the exposure event, and HIV serostatus of the source patient were collected on a standardized form. First, biological assessment was performed before first drug intake at day 0 and included HIV serology, complete blood count, renal (blood urea, creatinine, and phosphate) and hepatic [alanine aminotransferase (ALT), aspartate transaminase (AST)] function test, metabolic (glucose, triglycerides, cholesterol), and pregnancy test (for women of child-bearing potential).
On days 14 and 28, a laboratory monitoring for toxicity (including complete blood count, metabolic test, and renal and hepatic function tests) and a clinical evaluation were also performed; and self-reported tolerability, adherence, and side effects were recorded using a standardized form.
At 4 weeks and 3 months after treatment discontinuation, HIV p24Ag and HIV serology were, respectively, assessed.
Laboratory and questionnaire data were entered into and analyzed by EPI-INFO v 6.04. Factors for noncompletion of PEP and factors associated with the occurrence of adverse events were assessed using χ2 statistics and Fisher's exact test, as appropriate. Logistic regression model was then used to identify variables significantly and independently linked to the next binary-dependent variables: noncompletion of PEP and report of adverse event occurrence (Intercooled Stata 5.0 for Windows 95; Stata Corporation, College Station, Texas, USA).
Between November 2006 and June 2008, 249 participants in 10 French hospitals were included in the study.
These participants (details shown in Table 1) were mostly men (n = 165) and the mean age was 31.5 ± 10 years with no significant difference between men and women. None of the women were pregnant.
One hundred and sixty-eight participants had previously checked their HIV serostatus at least once.
Forty participants (16%) presented after blood exposure in a healthcare setting, 204 (82%) after sexual exposure (14 cases of rape), and five (2%) after exposure to blood during an assault.
Forty-seven participants had previously sought medical advice after exposure and half among them (n = 23) had then received PEP (two healthcare workers, 20 with sexual exposure, and one person with another exposure event).
The median duration of time between exposure and first-drug intake was 2 h (min = 0; max = 43) for occupational exposure and 15 h (min = 1; max = 48) for the other exposures.
In 83 (33.6%) source patients, HIV serostatus could be determined: 60 (73.2%) were known as HIV-infected at the time of the first presentation to medical services (25 cases of occupational exposures and 35 cases of sexual exposures). Among source patients screened after the introduction of PEP, one (1.2%) tested positive and 22 (26.8%) negative.
During this study, 27 (11%) participants were lost to follow-up (Fig. 1.).
In 34 cases (13.5%), PEP was discontinued shortly after treatment initiation because the source patient subsequently tested HIV-seronegative or because the injury was reassessed as ‘low-risk’ by a specialized physician, or because of a unilateral participant decision not related to tolerance. In this group, the median treatment period was 3 days (min = 1; max = 18). Poor tolerability to PEP was reported in 11 (32.3%) of these 34 persons: the most frequently reported adverse effects were diarrhea (80%), asthenia (66%), and abdominal pain (44.4%).
Among the remaining 188 participants (75.5%) who should have completed 4 weeks of PEP:
1. 22 (12%) of these participants discontinued PEP before day 28 due to adverse events (median treatment period 7 days); these side effects included two cases of skin rashes, one renal lithiasis, and one rhabdomyolysis.
2. and the other 166 (88%) participants completed the 28 days of TDF/FTC+LPV/r tablet formulation PEP with good tolerability in 96 cases, moderate in 59 cases, and poor in 11 cases. In this population, full adherence to PEP was declared in 69.8% of cases; one or two missing doses was declared in 26.5% of cases and more than three missing doses were declared in 3.7% of cases.
Among the 92 (22 + 59+ 11) participants who experienced at least one side effect, 78% reported diarrhea, 78% asthenia, 59% nausea and/or vomiting, and 38% headache.
Gastric disturbance (nausea or vomiting) and insomnia were more frequent in women than in men (P < 0.01 and P < 0.05, respectively). No significant differences were found between sexes concerning diarrhea, asthenia, or headache.
During PEP, the most frequent laboratory abnormalities were as follows:
1. grade 1 or grade 2 hypertriglyceridemia in 35% of cases with only 4.4% of grade 2 (2.5–5 × upper limit of normal, ULN) at 4 weeks;
2. grade 1 hypercholesterolemia (ULN – 7.75 mmol/l) in 15.7% of cases;
3. grade 3 hypophosphatemia (0.3–0.6 mmol/l) in three cases;
4. and grade 3 ALT elevation (5.0–20 × ULN) in three cases.
No grade 4 biological abnormalities were observed in this study except for creatinine phosphokinase (CPK) in the rhabdomyolysis case.
No significant abnormalities were found in creatinine or blood cell counts.
No HIV seroconversions were recorded during the study.
Using a univariate model on the whole cohort, considering factors associated with either PEP discontinuation or better tolerability, we observed the following points (Table 2):
1. less PEP discontinuation and a better tolerability in men than in women (P < 0.02 and P < 0.001, respectively).
2. less PEP dropouts in participants who have taken medical advice after exposure for the first time during this study than in participants who have previously taken medical advice after exposure (P < 0.03).
No significant association was found between PEP dropouts or tolerance and age, type of exposure, or knowledge of source's HIV serology.
When we focus only on sexual exposure, the univariate model analyzed showed the following findings:
1. less PEP discontinuation and better tolerability in MSM compared with participants consulting after heterosexual intercourse (P < 0.012 and P < 0.002, respectively);
2. and less PEP discontinuation in men than in women (P < 0.006), when analysis only considered participants with a heterosexual intercourse exposure.
In the logistic regression model on whole cohort, only female sex (P < 0.05) appeared as an independent predictor of poorer PEP tolerability and higher discontinuation rate.
Finally, if we compare the participants who took TDF/FTC+LPV/r tablet formulation PEP regimen to historical controls who have received one of the three protease inhibitor-based PEP regimens previously tested in the same study conditions, the rate of treatment interruption was significantly lower with TDF/FTC+LPV/r tablet formulation PEP regimen (P < 0.05; Table 3).
In this study, analysis to identify factors for noncompletion of PEP and adverse event occurrence showed that female sex is associated with a higher risk of PEP interruption and with poorer tolerability. Specifically, we found that gastric adverse events (such as nausea or vomiting) were more frequent in women. This was previously observed in a study evaluating a PEP regimen, including a combination of ZDV/3TC and TDF  and in another study among HIV-infected patients treated by TDF/FTC+LPV/r SGC . These findings are important because a preventive treatment of these specific adverse events in this selected population could be necessary and help to increase HIV PEP completion rates.
Concerning adverse events observed during TDF/FTC+LPV/r SGC PEP, two cases of skin rashes leading to PEP discontinuation were reported in our study. Skin rash due to TDF/FTC has been described in at least 5% of patients in clinical trials of FTC or TDF with other antiretroviral drugs. In the cases reported during our study, skin rashes cannot be related to other drugs. Therefore, taking into account the semiology of the demonstrations observed and the chronology, the responsibility of TDF/FTC in the occurrence of these skin rashes is considered as plausible, according to the criteria of imputability of the French method . In our study, skin rashes were benign in all cases (grade 1) and promptly resolved with drug (TDF/FTC) discontinuation.
A case of renal lithiasis was reported during our study. At least four cases of renal lithiasis due to LPV/r are published in the literature among HIV-infected patients . However, in all these cases, the delay between the initiation of LPV/r and the occurrence of lithiasis was greater than 4 weeks (8–16 months in reported cases versus 7 days in our study) and all patients but two recovered without treatment modification. On the other hand, one case of biliary lithiasis occurring during a PEP regimen, including LPV/r has been previously published . Therefore, a connection between LPV/r and the occurrence of this renal lithiasis must be considered as plausible according to the criteria of the imputability of the French method .
We report also a case of rhabdomyolysis documented in a 32-year-old woman who presented with myalgia and CPK elevation after 10 days of TDF/FTC+LPV/r PEP initiation. No other casual factors were identified and a complete normalization in the CPK level was observed a week after the substitution of TDF/FTC+LPV/r by ZDV/3TC+didanosine PEP regimen.
Nevertheless, when compared with our historical controls who received one of three other protease inhibitor-based PEP regimens, the rate of treatment interruption was significantly lower when TDF/FTC+LPV/r tablet formulation regimen is used. And finally, in this study, only 12% (22/188) of participants who should have completed 4 weeks of PEP did not complete PEP due to an adverse event and 51% (96/188) of participants who should have completed 4 weeks of PEP, completed 4 weeks of PEP and tolerated it well.
Another recent study evaluating the tolerability of PEP with combination of ZDV/3TC and saquinavir boosted by ritonavir (SQV/r) found a PEP discontinuation rate due to adverse effects lower than the rate found in our study (9 versus 12%) . Adverse events were reported in only 25% of participants taking ZDV/3TC+SQV/r PEP regimens. However, tolerability and noncompletion rate of PEP for this SQV/r PEP-containing regimen cannot be directly compared to our findings due to different study designs. Nevertheless, we consider that SQV/r could present some disadvantages compared to LPV/r: higher pill burden (3 versus 2 b.i.d.), food restriction, and necessity of a separate booster requiring storage in a refrigerator that could have an impact on PEP kit storage for the medical centers.
In conclusion, due to this good tolerability, the low discontinuation rate, and easiness of use, TDF/FTC+LPV/r tablet formulation should today be considered as the standard of care concerning HIV PEP.
W.T. contributed to data collection, analysis, data interpretation, literature search, and writing. P.M., T.P., G.B., B.C, Y.Q and E.P. contributed to data collection. E.B. contributed to data collection and writing. C.R. contributed to study design, data collection, analysis, data interpretation, literature search, and writing.
2. Quirino T, Niero F, Ricci E, Pusterla L, Carradori S, Gabbuti A, et al
. HAART tolerability: postexposure prophylaxis in healthcare workers versus treatment in HIV-infected patients. Antivir Ther 2000; 5:195–197.
3. Rabaud C, Bevilacqua S, Beguinot I, Dorvaux V, Schuhmacher H, May T, Canton P. CISIH Nancy Lorraine Champagne-Ardenne, France. Tolerability of postexposure zidovudine + lamivudine + nelfinavir prophylaxis of HIV infection. Clin Infect Dis 2001; 32:1494–1495.
4. Rabaud C, Burty C, Grandidier M, Christian B, Penalba C, Béguinot I, et al
. Tolerability of postexposure prophylaxis of HIV infection with the combination of zidovudine/lamivudine and lopinavir/ritonavir (Combivir + Kaletra). Clin Infect Dis 2005; 40:303–305.
5. Burty C, Christian B, Pouaha J, Penalba CH, Chavanet P, Prazuck TH, et al. Tolerability of post-exposure prophylaxis of HIV infection with the combination of lamivudine + tenofovir + atazanavir boosted with by ritonavir (LTA/r)
[abstract 92/23O]. Presented at 25
th Reunion Interdisciplinaire de Chimiotherapie anti Infectieuse
; 2005; Paris.
6. Tsai CC, Follis KE, Sabo A, Beck TW, Grant RF, Bischofberger N, et al
. Prevention of SIV infection in macaques by (R)-9-(2-phosphonylmethoxypropyl) adenine. Science 1995; 270:1197–1199 [PMID: 7502044].
7. Tsai CC, Emau P, Follis KE, Beck TW, Benveniste RE, Bischofberger N, et al
. Effectiveness of postinoculation (R)-9-(2-phosphonylmethoxypropyl) adenine treatment for prevention of persistent simian immunodeficiency virus SIVmne infection depends critically on timing of initiation and duration of treatment. J Virol 1998; 72:4265–4273 [PMID: 9557716].
8. Otten RA, Smith DK, Adams DR, Pullium JK, Jackson E, Kim CN, et al
. Efficacy of postexposure prophylaxis after intravaginal exposure of pig-tailed macaques to a human-derived retrovirus (human immunodeficiency virus type 2). J Virol 2000; 74:9771–9775 [PMID: 11000253].
9. Mayer K, Mimiaga M, Cohen D, Grasso C, Bill R, Vanderwarker R, Fisher A. Tenofovir-based regimens for nonoccupational postexposure prophylaxis (NPEP): improved tolerability and adherence compared to ZDV-based regimens
[abstract No. TuPE0432]. XVI International AIDS Conference
; 13–18 August 2006; Toronto, Canada.
10. Lugue A, Hulse S, Wang D, Shahzad U, Tanzman E, Antenozzi S, Smith B. Assessment of adverse events associated with antiretroviral regimens for postexposure prophylaxis for occupational and nonoccupational exposures to prevent transmission of human immunodeficiency virus. Infect Control Hosp Epidemiol 2007; 28:695–701.
11. Absalon J, Uy J, Yang R, Mancini M, McGrath D, for the CASTLE Study Group. Gender-based differences in ARV-naïve patients treated with boosted protease inhibitors: results from the CASTLE study (AI424138) [abstract TUPE0062]. XVII International AIDS Conference (IAC); 3–8 August 2008; Mexico City
12. Moore N, Biour M, Paux G, Loupi E, Begaud B, Boismare F, et al
. Adverse drug reaction monitoring: doing it the French way. Lancet 1985; 2:1056–1058.
13. DocoLecompte TH, Garrec A, Thomas L, Trechot PH, May TH, Rabaud CH. Lopinavir-ritonavir (Kaletra) and lithiasis: 7 cases. AIDS 2004; 18:705–706.
14. Lescure FX, Lebrette MG, Han Y, Vincensini JP, Bonnard P, Guessant S, et al
. Tolerability of postexposure prophylaxis with the combination of Combivir (CBV)–saquinavir 500-ritonavir (SQV-r) [abstract no. WEPEC006]. Presented at 4th IAS Conference; 22–25 July 2007; Sydney, Australia