Highly active antiretroviral therapy (HAART) has changed the clinical course of HIV infection, reducing mortality and lowering the incidence of opportunistic infections. The wide range of drugs now available has boosted the likelihood of success but also the risk of drug interactions and adverse events. Some of these adverse events are unforeseen or have a delayed onset.
Controlled trials are primarily designed to report on efficacy and ensure that toxicity is manageable. Very rarely is the focus in these studies on toxicity, and their ability to impart information on toxicity is generally limited.1-3 Toxicity assessments are poor in clinical trials because of small sample sizes, specific patient types, and short duration. The short duration is particularly relevant for HIV infection, as many receive accelerated approval based on effects on surrogate clinical outcomes. Postmarketing monitoring systems are needed to identify unexpected adverse events and to clarify the setting and characteristics of patients who develop unexpected toxicities. However, these systems are unlikely to provide information on incidence of adverse events, and only if active data collection efforts are made will incidence information become available.
The SCOLTA project (Surveillance Cohort Long-Term Toxicity of Antiretrovirals) was set up as an active pharmacovigilance system for any antiretroviral therapy and a “sentinel” for unexpected and late adverse reactions arising during any antiretroviral treatment. This online recording system was established by the CISAI group (Italian Coordinators for the Study of Allergies and HIV Infection; http://www.cisai.info), especially for antiretroviral drugs. To date, 25 Italian infectious disease centers have participated, with the possibility of observing about 12,000 HIV-positive patients. In each center, a physician was trained ad hoc to survey adverse drug reactions.
Design and Objectives
This pharmacovigilance program is divided into 2 sections. The first is the New Drugs Project: this is a prospective, multicenter, observational pharmacovigilance study involving 1 cohort of patients for each new drug. Adverse reactions are defined as any response that is harmful, not wanted, or unexpected, arising at the doses used for therapy in men. This definition excludes intentional or accidental poisoning and therapeutic failures.4 The objective is to establish the incidence of known severe, rare, or unknown adverse events, risk factors, and their possible mechanisms.
Patients who start taking a newly introduced drug are consecutively asked to give written informed consent and are then enrolled and included in the cohort for that drug. Patients taking ≥1 new drug are enrolled in ≥1 observation cohort. All patients are required to give informed consent: to date no one has refused consent. As this is an observational study, the choice of therapy is entirely up to the individual physicians and patients in each center.
Adverse events are notified when they are clinically observed. Patients are followed up at 6-month intervals. Patients who do not present to the physician in >6 months leave the study and are considered lost to follow-up. If a patient stops using a “new drug,” she or he leaves the study. Severe adverse reactions or unexpected events appearing within 6 months of leaving the study are to be recorded.
Forms are available on the site for recording all AIDS Clinical Trials Group (ACTG) grade III and IV reactions and all unexpected events (ie, those not mentioned in the package insert of the drug or not reported in the clinical trials on which registration was based). Events are classified for severity using the ACTG classification system,5 and the Uppsala Monitoring Center causality categories are used to establish causal relations.6 The study is scheduled to last 10 years.
The second section is the Unexpected Events Project. This project is based on notification by participating centers for all patients taking antiretroviral therapy. The aim is to identify unexpected adverse reactions during treatment with any antiretroviral drug. Adverse reactions are defined as any response to therapy that differs in nature or severity from those described in the package insert or the marketing authorization or is unexpected in relation to the drug's characteristics.4
Data Collection and Treatment
Patients' data are recorded and entered into the computer using the dedicated site. There are forms for enrollment, follow-up, and recording adverse events. All information concerning patients, whether held temporarily or filed online, is encrypted as a guarantee of privacy.
For the New Drugs Project the usual descriptive statistics are compiled (ie, mean, standard deviation, median, range for continuous variables, absolute and relative frequencies for categorical variables). For adverse events we will calculate the incidence as the number of events in relation to persons/observation time. Incidence rates and their confidence limits will be calculated on the total population treated with each single drug and by strata of potential risk factors such as age, duration of disease, liver disease, and comorbidities. Patients lost to follow-up will be included for their period of observation. For the Unexpected Events Project, events will be collected in a database, listing each patient's main features, the severity, and the causal link with antiretroviral drugs taken.
Reported here are the preliminary findings for the New Drugs Project. So far, all patients have been enrolled consecutively when they start treatment with lopinavir/ritonavir (LPV/r), tenofovir (TDF), pegylated interferon (IFN), atazanavir (ATZ), enfuvirtide (T-20), or tipranavir (TPV). Six cohorts of patients are therefore being followed until they stop using the drug.
Between October 1, 2002, and March 30, 2004, 1184 patients were enrolled; 837 of them are men (70.7%); the mean age is 40.5 years (SD ± 7.77); and mean CD4 lymphocyte count at enrollment is 314 cells/μL (SD ± 249 cells/μL). Of these, 1255 (87.4%) were treatment experienced, and 181 were naive to antiretroviral treatment. Mean follow-up so far is 10.7 months (SD ± 6.22).
There are 703 patients in the LPV/r cohort, 585 in the TDF cohort, 35 in IFN, 95 in ATZ, 10 in T-20, and 8 in TPV. Thus far, 100 grade III and IV adverse events have been reported, 73 in the LPV/r group and 15, 7, 1, and 4 in the TDF, IFN, ATZ, and T-20 groups, respectively. A total of 41 patients (3.4%) have stopped treatment because of toxicity, 30 with LPV/r, 8 TDF, 1 IFN, 1 ATZ, and 1 with T-20.
The data presented here refer to the LPV/r cohort, which was big enough (703 patients) and followed up long enough (12.4 months) to permit a preliminary analysis. Table 1 summarizes the patients' details at enrollment. Of the 703 enrolled, 493 (70.1%) are still receiving treatment; 166 (23.6%) have stopped, and 44 (6.3%) have been lost to follow-up. Treatment was stopped in 30 cases (4.3%) because of grade III or IV toxicity; in 28 (4%) because of failure; and in 17 (2.4%) because of grade I or II events. In 47 cases (6.7%) it was stopped to simplify therapy, in 40 (5.7%) because the patient wanted to stop, and in 4 cases because of death due to HIV-related pathologies. For LPV/r, 73 grade III or IV events were reported. Four subjects (5.5%) developed lipodystrophy, and 30 (41%) developed metabolic alterations (ie, in 10 cases an increase of cholesterol levels, all degree III, and in 20 hypertriglyceridemia, 12 of degree III and 8 of degree IV, respectively). Twenty-one patients (28.8%) had gastrointestinal complaints; 11 had nausea or vomiting (grade III) and 10 had diarrhea (grade III). Two patients (2.8%) had liver toxicity with grade III alanine aminotransferase increase, and 1 (1.4%) had acute renal failure. Fifteen patients (20.5%) presented reactions that did not fall under any of the above headings.
The rate of adverse events per 100 person-years was 14.2 (Table 1) on the basis of all patients treated, 9.8 (95% CI 9.1-10.5) for treatment-naive patients, and 15.0 (95% CI 14.6-15.4) for experienced patients. As the causal relation with treatment had to be indicated for each event according to the Uppsala Monitoring Center algorithm,6 we took this into account when working out the incidence of adverse events. The incidence of events for which there was a positive correlation with the drug(s) was 12.4 (95% CI 12.1-12.7).
In European countries, pharmacovigilance studies are considered essential to cover the gaps left by the artificial situation of controlled clinical trials, to gain a better picture of drug toxicity. The SCOLTA project was designed to assist with pharmacovigilance for newly introduced antiretroviral drugs and to act as a “sentinel” system for unexpected or delayed adverse events related to any such treatment. The online recording system for adverse events designed specifically for antiretroviral drugs helps optimize patient enrollment; adverse events can be reported in real time and communication is easy between centers.
One of the study's limitations is that the data are not formally representative of the entire HIV population (ie, at the national level), because the 25 centers enrolled are representative of only the northern and central part of Italy. The 2nd limitation is that adverse events were not confirmed by an events validation board. Our online system has been tested now for about 18 months in 25 infectious disease departments throughout Italy, following a total of about 12,000 patients. We have therefore already been able to enroll >1000 patients with homogeneous data and can start investigating new drugs as soon as they enter clinical practice. The first drug to be assessed was LPV/r, for which the sample and follow-up are now adequate to provide initial information on adverse reactions. The incidence of adverse events per 100 person-years is 14.2, assessed on all the patients receiving this treatment. The incidence is lower, however, if we take only patients receiving their first antiretroviral treatment, in which case it is 9.8 (95% CI 9.1-10.5) compared with 15.0 (95% CI 14.6-15.4) for experienced patients. This observation supports the report by Murphy et al.7
We also compared the data for the LPV/r cohort with the findings of a previous study in which we assessed the toxicity of regimens including protease inhibitors.8 The incidence of adverse events with LPV/r was lower than the figures we already had for ritonavir (n = 35.26) and indinavir (n = 16.32) and higher than for saquinavir (n = 9.67) and nelfinavir (n = 6).
The findings reported here are, as we have said, preliminary, but they do show that this data collection method gives timely real-life information from which to assess the impact of short- and long-term toxicity of new antiretroviral drugs.
The SCOLTA system may be most useful in clarifying the clinical setting where expected but incompletely evaluated adverse events occur shortly after antiretroviral treatment initiation.
1. Cameron DW, Hearth-Chiozzi M, Danner S, et al. Randomized placebo-controlled trial of ritonavir in advanced HIV-1 disease. Lancet
2. Hammer SM, Kathleen ES, Hughes MD, et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4+ cell counts of 200 per cubic millimeters or less. N Engl J Med
3. Collier AC, Coombs RW, Schoenfeld DA, et al. Treatment of human immunodeficiency virus infection with saquinavir, zidovudine and zalcitabine. N Engl J Med
4. Mariani L, Minora T, Ventresca GP. Drug surveillance and adverse reactions to drugs: the literature and importance of historical data. Clin Ter
5. AIDS Clinical Trials Group. Table of Grading Severity of Adult Adverse Experiences
. Rockville, MD: Division of AIDS, National Institute of Allergy and Infectious Diseases; 1996.
6. Olsson S. National Pharmacovigilance Systems
, 2nd ed. Uppsala, Sweden: World Health Organization Collaborating Centre for International Drug Monitoring; 1999.
7. Murphy R, Brun S, Hicks C, et al. ABT-378/ritonavir plus stavudine and lamivudine for the treatment of antiretroviral-naive adults with HIV-1 infection: 48-week results. AIDS
8. Bonfanti P, Valsecchi L, Parazzini F, et al. Incidence of adverse reactions in HIV patients treated with protease inhibitors: a cohort study. J Acquir Immune Defic Syndr
The Coordinamento Italiano Studio Allergia e Infezione da HIV (CISAI) Study Group comprises the following members:
Coordination: T. Quirino (Busto Arsizio), P. Bonfanti, G. M. Vigevani (Milano).
Statistical Analysis: F. Parazzini, E. Ricci (Milano).
Recruitment Sites and Investigators: R. Cinelli, U. Tirelli (Aviano), G. Cocca, G. Rizzardini (Busto Arsizio); C. Grosso, A. Stagno (Cesena); L. Pusterla, D. Santoro (Como); C. Magnani, P. Viganò (Cuggiono); S. Carradori, F. Ghinelli (Ferrara); A. Gabbuti, F. Mazzotta (Firenze); C. Martinelli, F. Leoncini (Firenze); G. Penco, G. Cassola (Genova); S. Miccolis, A. Scalzini (Mantova); L. Valsecchi, A. Cargnel (Milano); T. Bini, S. Melzi, M. Moroni (Milano); E. Rosella, G. Fioni (Milano); M. Gargiulo, A. Chirianni (Napoli); M. Franzetti, P. Cadrobbi (Padova); C. Sfara, G. Stagni (Perugia); G. Parruti, G. Marani Toro (Pescara); B. Adriani, A. Paladini (Prato); G. Madeddu, M. S. Mura (Sassari); P. Marconi, A. Antinori (Roma); G. Orofino, D. Zeme, P. Caramello (Torino); G. Cristina, F. Carcò (Vercelli); and D. Migliorini, O. Armignacco (Viterbo).