Methods
Data collection
The
Italian Register for HIV Infection in Paediatrics is a database concerning cases of children with HIV-1 infection, established in 1985 on the initiative of the Immunology Study Group, within the Italian Society of Paediatrics. Data have been recorded since June 1, 1985. Data of children born before this period, about 13% of the data, have been collected retrospectively [
7].
The source of the Register is represented by a network of 106 paediatric clinical centres distributed throughout Italy. These centres, which participate voluntarily, register all children with HIV-1 infection. Two coordination centres are located at the Paediatric Clinic of Infectious Diseases of the Department of Paediatrics of the University of Florence and the Department of Paediatric and Adolescent Sciences of the University of Turin. The Registry includes all children with HIV-1 infection, either born to a mother with HIV-1, or identified at birth or reported later. The data are representative of the entire population of children born to mothers with HIV-1 infection in Italy [
7].
Data on mother–child couples are collected using specific reporting and follow-up forms and are processed anonymously, in accordance with the relevant guidelines and regulations.
In particular, basic information regarding the mother include: the mode of infection such as sexual, drug addiction, transfusions, endemic area origin; number of pregnancies; mode of delivery; gestational age and clinical condition of the mother at the time of delivery reported according to the CDC classification system. Since 1989, data concerning the mother include a classification of women treated and untreated, since 1993 there is information on the type of therapy, gestational age at the beginning and at the end of the therapy and since 1994, with the introduction of PACTG 076 protocol, it is also indicated whether the treatment has been administered intrapartum or to the newborn. If the treatment has been administered during pregnancy, all information concerning the combination of drugs used is reported in a special section. Information on possible resistance to treatment, concomitant maternal infections and the conformity of therapy is not collected. The immunological condition of the mother, indicated by the CD4+ T lymphocyte count, and the viral load (VL) at the time of delivery have been reported since 1996 and are only known in a minority of cases. The information concerning the child reported in the Register, are the following: birth weight, age at first observation, type and duration of breastfeeding and state of infection. Whereas, in the follow-up records there are information regarding the values of anti-HIV-1 antibodies, viral load, T CD4+ lymphocyte count, stage of disease according to CDC classification criteria, age at diagnosis of AIDS, laboratory investigations, therapy used, date of start and discontinuation of antiretroviral therapy, start and type of Pneumocystis jirovecii pneumonia prophylaxis.
Through the use of reporting forms, the referral physician of each center in the Registry provides to transfer to the two coordination centers basic information regarding those who have acquired HIV-1 infection within 13 years of life or who are at risk of becoming infected because they were born to HIV-1 positive mothers; while the follow-up forms are filled in and updated every six months by a paediatrician in charge in each center. In fact, once an infection has been diagnosed, the children are visited every two months. In addition, any changes in clinical and immunological status, together with vaccinations, tests performed and treatment used, are recorded on the follow-up form. Children born to HIV-1 positive mothers who have not acquired the infection are examined at least once a year. Through a double cross-check in the two coordination centers, the quality of the information forwarded is checked. In addition, other control mechanisms are used, while maintaining the anonymity of children, to avoid double reporting. Finally, at least once a year, an audit of the referring physicians is organized in order to verify the results obtained, to plan and encourage research lines and to standardize procedures.
At the moment of patient’s enrolment, written informed consent is obtained from a patient’s parent or legal guardian; data are treated anonymously as each patient is identified by an alphanumerical code. The study was approved by the review boards and ethics committees of each participating institution; in particular, the ultimate form of the study was approved by the Ethics Committee for Human Investigation at Meyer Children’s University Hospital.
Study design
We performed a cross-sectional study. Children enrolled into the Italian Register for HIV Infection in Paediatrics were selected whether they: have perinatal HIV-1 infection, were receiving cART and were in follow-up at the December 31, 2018. Children were excluded from the study whether they were: uninfected; with horizontally acquired HIV-1 infection (through transfusions, drug addiction, other); not receiving cART at the December 31, 2018, or aged > 18 years.
The following information has been collected for each patient: age (years); sex (male/female); body weight (Kg); clinical class (class N, A, B, C according to CDC classification); immunological category (category 1, 2, 3 according to CDC classification); CD4+ T lymphocytes count (cell/μL); birthplace of the child and his/her mother (Italian nationality/immigrant), viral load (VL, copies/mL).
The therapy children were receiving was categorized by distinguishing between off-label and on-label use. We have considered off-label use when the drug was prescribed under the age or weight it was licensed for according to the European Medicine Agency (EMA) [
5]. Unfortunately, due to a large portion of missing data regarding the children’s weight, it was possible to evaluate off-label prescription by weight only relatively to few drugs.
For each therapeutic regimen the following data have been extracted from the Register: type of therapeutic regimen, number of drugs and specific drugs included in the regimen, year in which the therapeutic regimen was started, duration of therapy, age at the beginning of the regimen, clinical status, viral load and count and percentage of CD4+T lymphocytes at the beginning of therapy and at the last follow-up and the age at the last control and reported adverse events.
Immunological success was defined as a percentage of CD4+T lymphocytes > 25% or an increase > 10% between the beginning of the regimen and the last observation if the regimen was in progress or at the end of the regimen if it was discontinued. When the percentage of CD4+ T lymphocytes at the baseline and the last observations were both > 25%, we considered it as an immunological success. Virological success was defined as a VL not detectable at the last observation during therapy. Due to changes in the detection limits of the HIV RNA test over time, the undetectable VL was defined as < 400 copies/mL.
Definitions
Definition of perinatal HIV-1 infection:
Perinatal infection is defined as the transmission of the infection in utero (congenital infection), during labour, or immediately after birth in breastfed children [
8].
HIV-1 infection was diagnosed on the basis of the persistence of anti-HIV-1 antibodies after 18 months of life or, before 18 months, by detecting the positivity of viral markers on at least two occasions, excluding any investigations carried out on umbilical cord blood. Children for whom the detection of antibodies to HIV-1 and the PCR investigation were negative on at least two occasions are defined as non-infected [
9].
Definition of cART
cART was defined as the combination of three or more antiretroviral drugs of at least two different classes [
10,
11].
Definition of off-label use of drugs in paediatric age
Off-label use of a drug was defined as the use of pharmaceutical drugs for an unapproved indication or in an unapproved age group, dosage, or route of administration, according to the package leaflet of medicines authorized by EMA [
5]. Detailed data are reported in Additional file
1: Table S1.
Statistical analysis
The continuous variables were expressed as median and interquartile range (IQR). The categorical variables have been expressed as numbers and percentages.
For the categorical variables we performed the χ2 or Fisher test; for the continuous variables the medians between groups were compared using the Wilcoxon Mann–Whitney U test. Univariate logistic regression analysis was used to evaluate possible risk factors for receiving an off-label therapy, and correspondent odds ratios (ORs) and 95% confidence intervals were calculated. Statistical analyses were performed using the STATA/SE version 10.0 software package (Stata Corporation, College Station, TX). p < 0.05 was considered statistically significant.
Discussion
Off-label drugs in our cohort of perinatally HIV-1 infected children were prescribed to 22.2% (50/225) of children at their last check in 2018. At univariate logistic regression analysis, risk factors for receiving an off-label regimen were age < 2 years, being of first HAART regimen and a detectable VL before starting that regimen. At multivariate analysis only younger age was confirmed to be an independent risk factor for receiving an off-label therapy. However, we also noticed that several children (56.52%, 13/23), aged between 11 and 12 years of age were receiving an off-label cART at last check. In this sub-group of children physicians probably considered minimal the potential risks related to an off-label use of the ARV drug since 12 years was the age for which the drug was licensed.
The most commonly antiretroviral drugs prescribed off -label were lopinavir-ritonavir (Kaletra®) in younger children, and three fixed dose combinations in older children: elvitegravir/cobicistat/emtricitabine/tenofovir alafenamide (Genvoya®), abacavir/dolutegravir/lamivudine (Triumeq®) and tenofovir disoproxil/emtricitabine (Truvada®).
Comparing our results to a 2016 Spanish multicentre study, conducted by Cooke et al. on a cohort of 318 perinatally HIV-1-infected children and adolescents between 1988 and 2012, it was found that our off-label therapy rates in children are lower than the ones of the Spanish sample, which reveals an off-label use of antiretroviral therapy in 69% of included children (221/318) [
6]. This difference could reflect the increase in approval coverage for the clusters under 2 years of age, which occurred after 2012 and consequently reduced the off-label incidence. For example, in 2012 atazanavir was approved in children over 6 years of age, while in 2014 the approval was brought forward to children over 3 months of age; raltegravir in 2012 was approved only in children over 2 years of age, while by 2018 it was approved from birth; efavirenz, which until 2013 was approved only for children over 3 years of age, is now approved in children over 3 months of age [
5]. Considering that efavirenz was used by the 24.5% of our population and that among the non-nucleoside reverse transcriptase inhibitors (NNRTIs) it was the most widely used after nevirapine, this fact may have contributed to reduce the off-label percentage in our population. However, in our cohort, off-label use was still widespread. The 10% of our off-label patients have reached the fifth therapeutic regimen; this confirms that off-label therapies are chosen in some multi-experienced children, failing previous therapy and probably with no in-label therapeutic options available.
In our study, before the current cART, the 48% of the off-label children had a detectable VL, while, at the last check, the 80% of the patients had an undetectable VL and the 90% a CD4
+ lymphocyteS count > 25%. Similar to our results, Cooke et al
. found that the main reason for the off-label use of cART was the virological failure of the previous therapy. Moreover, the onset of adverse events and problems related to the administration of drugs, such as formulations unsuitable for age, respectively led in 12% and 5% of cases to the discontinuation of treatment. Adherence to the therapy is essential for its effectiveness and this is in agreement with our data, since the main drugs used off-label are fixed-combination drugs. Clay et al. meta-analysis, comparing adult patients’ adherence of single tablet regimens (STR) to multiple tablet regimens (MTR), revealed that patients on STR cART were significantly more adherent and had a better VL suppression when compared to patients on MTR [
12]. Indeed, in our study, we observed that four MTR (lopinavir/ritonavir; emtricitabine/tenofovir disoproxil; elvitegravir/cobicistat/emtricitabine/tenofovir alafenamide and abacavir/dolutegravir/lamivudine) were the most commonly used off-label drug formulations.
Since 2012 lopinavir/ritonavir is licensed for the use in children aged > 14 days, while since 2002 it was licensed for children aged > 2 years. In our dataset, the age group receiving off-label lopinavir/ritonavir was mostly represented by children aged < 12 months. Previous studies found that lopinavir/ritonavir is more effective than nevirapine, which may be one of the reasons why physician decided to start this therapy [
13]. However, since cases of lopinavir/ritonavir cardiotoxicity (complete heart block and cardiomyopathy) have been reported in infants, caution is recommended when using this drug [
13]. No cardiac adverse event in young children was reported in our dataset. What we can observe instead, is a good therapeutic success, demonstrated by the fact that 100% of children receiving off label lopinavir/ritonavir had a CD4
+ T lymphocytes percentage at last check-up > 25% and 90% had an undetectable viremia.
For what concerns elvitegravir/cobicistat/emtricitabine/tenofovir alafenamide, the age most involved by their off-label use was 8–11 years; in our cohort, 100% of off-label patients the last VL was undetectable and the CD4+ T lymphocytes percentage > 25%. In the same age group, we also find off-label use for abacavir/dolutegravir/lamivudine. In these children an undetectable VL at the last check was reached in the 87.7% of patients and undetectable and CD4+ T lymphocytes percentage > 25% in the 87.5% of them.
Regarding emtricitabine/tenofovir disoproxil, the age most involved by off-label use was 11–14 years, (the drug has to be considered off label when used in children aged 14 years since when these children started the therapy it was approved > 18 years). At the last check the VL was undetectable in the 81.8% and CD4+ T lymphocytes percentage > 25% in the 81.8%.
In our sample, a 10 years old child received an off-label treatment with rilpivirine, emtricitabine and tenofovir disoproxil. Before the current therapy the patient had a detectable viral load, while at the last check the VL was undetectable. No adverse event was reported for this patient. A previous multicentre study that enrolled 17 children and adolescents with perinatal HIV-1 infection from 2013 to 2015 demonstrated the efficacy of off-label use of rilpivirine in combination with emtricitabine and tenofovir disoproxil, as children showed good control of the infection, improved CD4
+ T lymphocytes count and CD4
+/CD8
+ T lymphocytes ratio [
14].
Our data suggest similar proportion of virological and immunological success at last check among children receiving off- or on-label cART. No adverse event to off-label ARV drug was reported in our dataset.
The potential limitations of our study lie in its retrospective design. In addition, one limitation of the study is the scarcity of data regarding the collected adverse events. Even if no adverse event related to the use of off-label ARV drug was reported we may not exclude an underreporting bias. Other studies are needed to better elucidate this issue. Another possible limitation of our study is that in our dataset young children (aged less than 2 years) were more likely to receive off label drugs than older patients. However, the great majority of young children received lopinar/ritonavir, which has been subsequently licenced for the use in children aged 14 days or more. More recent studies, considering recent updates in paediatric drug licences, might lead to a different conclusion. Similarly, since patients’ information was collected up to 2018, a trend toward a general reduced off label use of ARV drugs is likely in recent years since new approvals have occurred over time.
Finally, we could not explore the risk of prescription of not-corrected dose in children receiving an off-label regimen, due to a large proportion of missing data regarding body weights.
In a previous study more than 10% of children treated with an off-label prescription received an overdose and a further 10% received an underdose, defined as the administration of a 25% dose respectively above or below the recommended dose [
6]. A previous multi-centre cohort study from Ireland and UK, including 615 children aged 2–12 years and treated with cART, showed that children receiving a dose less than 90% of the recommended dose accounted for 6–62% of the study population [
15].
Acknowledgements
All principal investigators and sites who participate in the “Italian Register for HIV infection in children” are listed below.
Maurizio Ruggeri (Bergamo), Francesco Baldi (Imola), Giacomo Faldella (Bologna), Piergiorgio Chiriacò (Brindisi) Carlo Dessì (Cagliari), Maria Grazia Pantò (Catania), Elisa Anastasio (Catanzaro), Luisa Abbagnato (Como), Maria Rita Govoni (Ferrara), Maurizio Bigi (Forlì), Elisabetta Bondi (Genova), Riccardo Borea, Giovanni Cenderello (Imperia), Donato Tommasi (Lecce), Raffaella Pinzani (Milano) Ernesto Renato Dalle Nogare, Marcello Saitta (Palermo), Leonardo Felici (Pesaro), Rita Consolini (Pisa), Angelo Antonellini (Ravenna), Gianfranco Anzidei, Orazio Genovese, Salvatore Catania, Fabio Natale (Roma), Paolina Olmeo (Sassari) Letizia Cristiano (Taranto), Vincenzo Portelli (Trapani), Marco Rabusin (Trieste), Sara Parigi (Firenze), Francesca Orlandi (Firenze), Alessandra de Martino (Firenze), Giada Maria DiPietro (Milano), Leone Fabrizio (Roma).
Other members of the Italian Register for HIV Infection in Children
Sara Parigi1, Francesca Orlandi1, Alessandra de Martino1, Raffaella Pinzani2, Luisa Abbagnato3, Maurizio Ruggeri4, Francesco Baldi5, Giacomo Faldella6, Piergiorgio Chiriacò7, Carlo Dessì8, Maria Grazia Pantò9, Elisa Anastasio10, Maria Rita Govoni11, Maurizio Bigi12, Elisabetta Bondi13, Riccardo Borea14, Giovanni Cenderello15, Donato Tommasi16, Ernesto Renato Dalle Nogare17, Marcello Saitta18, Leonardo Felici19, Rita Consolini20, Angelo Antonellini21, Gianfranco Anzidei22, Orazio Genovese23, Salvatore Catania24, Fabio Natale24, Paolina Olmeo25, Letizia Cristiano26, Vincenzo Portelli27, Marco Rabusin28, Giada Maria Di Pietro29, Leone Fabrizio30
1Paediatric Infectious Diseases Unit, Department of Health Sciences, Anna Meyer Children University Hospital, Florence, Italy
2Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy.
3Paediatric Unit, Sant’Anna Hospital, Como, Italy
4Paediatric Department, Hospital Papa Giovanni XXIII, Bergamo, Italy.
5Azienda Sanitaria Locale Imola, Italy
6Neonatal Intensive Care Unit, S. Orsola-Malpighi University Hospital, Department of Medical and Surgical Sciences (DIMEC), University of Bologna, Bologna, Italy.
7Infectious Diseases Unit, A. Perrino Hospital, Brindisi, Italy
8Azienda Sanitaria Locale 8, Cagliari, Italy.
9Infectious Disease Unit, Cannizzaro Hospital, Catania, Italy
10Department of Medical and Surgical Sciences, Pediatrics Unit, University “Magna Graecia”, Catanzaro, Italy
11Pediatric Unit, University Hospital ‘S. Anna’, Ferrara, Italy.
12Pediatric Community Unit, Azienda Sanitaria Locale Romagna, Rimini, Italy.
13Pediatric Infectious Disease Division, IRCCS Gaslini, Genoa, Italy.
14Maternal and child health department, ASL1, Imperia, Italy
15Infectious Disease Unit, Ospedale Sanremo, Azienda Sanitaria Locale 1, Genoa, Italy.
16Unità Operativa di Malattie Infettive Ospedale Fazzi, Lecce, Italy
17Unit of Infectious Diseases and Medicine of Migrant and Vulnerable Populations, National Relevance Hospital Trust (ARNAS), Civico-Benfratelli Hospital, Palermo, Italy.
18BioMorf Department, University of Messina, Messina, Italy.
19Pediatric Unit, Pesaro, Italy
20Division of Pediatrics, Department of Clinical and Experimental Medicine, University of Pisa, Pisa, Italy.
21Pediatric Unit, Hospital of Ravenna, Ravenna, Italy
22Division of Pediatrics, Spallanzani Hospital, Rome, Italy
23Pediatric Intensive Care Unit, Catholic University and Policlinico Gemelli, Fondazione Policlinico Universitario Agostino Gemelli IRCSS, Rome, Italy.
24Department of Maternal and Child Health, Sapienza University of Rome, Rome, Italy.
25Pediatric Clinic, Department of Surgical, Microsurgical and Medical Sciences, University of Sassari, Sassari, Italy.
26Infectious Disease Unit, SS Annunziata Hospital, Taranto, Italy.
27Infectious Disease Unit, Hospital of Trapani, Trapani, Italy.
28Pediatric Hematology-Oncology, Institute for Maternal and Child Health IRCCS “Burlo Garofolo”, Trieste, Italy.
29Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy.
30Department of Pediatrics, Sapienza University, Rome, Italy.