Skip to main content

Long-term PHARMacovigilance for Adverse effects in Childhood arthritis focussing on Immune modulatory drugs

Final Report Summary - PHARMACHILD (Long-term PHARMacovigilance for Adverse effects in Childhood arthritis focussing on Immune modulatory drugs)

Executive Summary:
With the Pharmachild project an International Pharmacovigilance Registry for children with JIA (register to follow up on safety and efficacy of medication used in JIA) has been built with already 87 actively contributing centers from 33 countries worldwide and is still growing fast. In the first step (census) the data are finalized from 10,246 patients, the data from 1448 patients are entered in the prospective cohort and the data from 6450 patients with a median follow up of 4.9 years are completed consisting of 3 groups of patients: 4007 (biologics±MTX), 1633 (MTX only) and 810 (NSAIDs only).
Till now we did not find an increased risk for malignancies in the JIA population compared to the risk in the healthy childhood population, but we will continue to watch for signals in that direction.
We found a 6.3 times higher incidence rate of inflammatory bowel disease in our JIA population compared to healthy children, but a lower rate for children using a biological than described in literature although the type of biological used will likely influence that risk.
79.9% of patients did not experience any moderate/severe/serious Adverse Events (AE) or an Event of Specific Interest (ESI) during a follow up of 4.9 years.
Only 9.7% of these patients developed an event of specific interest. This was higher in the systemic JIA group (24.4%), but this could have been caused by the definition of ESI’s as well (e.g. macrophage activation syndrome was an ESI and only occurs in systemic JIA). The events of specific interest that were most frequently seen were: Infusion related reaction (12.6%), Neutropenia (6.9%), Injection related reaction (5.4%), Macrophage activation syndrome (5.2%), Tuberculosis (3.5%), but mainly Serious/targeted infections (59.0%). Other AE's were found in the nonsystemic JIA group in 13.8% compared to the 17.2% in the systemic JIA group. Gastrointestinal, hepatobiliary, eye, and skin disorders were amongst the most frequent AE’s (7-15% per field). Also Investigations (eg. Fecal tests, endoscopy) and surgical and medical procedures were not uncommon (12.3 and 8.1% respectively).
For every center it is possible to look at their own data. In that way every center can find out in what aspect they experience more or less adverse events than normally seen. This enables them to tackle these problems and improve their care.

Another part of this project was the collection of biosamples (e.g. blood and joint fluid) in a biomaetrial bank (BMB). Over a thousand samples have been collected so far and analyses are continuously being performed.
Looking at the blood from JIA patients we found out that serum concentrations of a certain protein in SJIA correlates closely with response to treatment with biologicals and disease activity. Furthermore we introduced a novel and additional measurement for monitoring anti-inflammatory treatment of individual patients with SJIA. This novel parameter allows for the first time individual adaptation of immune-suppressive treatment in SJIA and will hereby avoid overtreatment and reduce long-term side effects.
We have further identified and defined S100-proteins as predictors for response to therapy in JIA and as novel biomarkers which allow prediction of disease flares in JIA patients on medications with biologics.

Another study showed differences in the mechanisms of effect between etanercept and adalimumab which may help our understanding of the differences in efficacy and side effects.

It was shown that the immune system of JIA patients was capable of dealing with foreign or microbial proteins and in fact the cells necessary to do so (dendritic cells) were functionally normal in JIA patients before and after starting immuosuppressive agents. However a novel population of these dendritic cells were found in the blood of JIA patients. This subset of dendritic cells in JIA patients have a suppressive effect on the immunesystem and need to be studied in more detail.

Another study revealed that, as proposed by Prof Martini, categorizing young ANA-positive JIA patients as a separate subcategory of JIA is supported by low levels of SAA1 and high levels of sIL-2R. Interestingly, blocking of sIL-2R seems to be a valid immunotherapy for some autoimmune uveitis patients. Therefore, it could be that young ANA-positive patients would benefit from such treatment. The usefulness of sIL-2R and SAA1 as biomarkers of prediction for the development of uveitis needs to be validated and might have implications for the uveitis-screening programs as well.

Many scientific presentations were given and the patientgroups have been informed internationally already and will be kept updated annually.
Already 8 scientific articles have been published but with this dynamic registry and ongoing collection of biosamples surely more will follow soon.

Project Context and Objectives:
Until the age of biotechnology, treatment options for children and adolescents with severe arthritis, inflammatory bowel diseases and other serious diseases with chronic inflammation were limited. Advances in understanding the pathophysiology of the inflammatory responses have led to the development of a new class of medications that are capable of inhibiting selectively the principal mediators of inflammation and tissue damage. The introduction of these new immunomodulators, which are collectively termed biologics, has opened a new era in the treatment of inflammatory diseases. Recent reports however suggest unexpected increases in adverse - or serious events -associated with the use of these biologics.
PHARMACHILD aims to detect, assess and understand long term and short term side effects of the use of biologics by studying the pharmacovigilance in a large international cohort of patients with Juvenile Idiopathic Arthritis (children and young adults) in order to support regulatory decisions on marketing authorizations for these products. There is a clear need for a study that will enable the promotion of more effective and safer use of biologics in JIA as children and young adolescents represent an especially vulnerable patient group, JIA is the most common chronic disease of childhood treated with biologics, and data available from adults cannot be extrapolated to children. PHARMACHILD brings together leading European scientists in the field of infectious diseases, immunity, inflammation and oncology. Our network combines resources from international organizations and existing national registries. This combination will enable us to achieve the critical mass (in terms of expertise and resources) to identify the risk factors for developing adverse events, to elucidate the mechanisms involved in the adverse events (such as latent infection reactivation) and to develop methodologies for screening and predicting patients at risk.

Specific objectives
In pursuit of our mission and within the focus set, the main scientific and technological objectives of PHARMACHILD have been set. These objectives are described below in more detail.

Research objective 1: To establish a European pharmacovigilance registry
Rare side effects are difficult to detect in small populations as evidence on the efficacy and harms of biologics for treating children with JIA is sparse and fraught with methodological issues that severely limit inferences to be drawn about the risk–benefit profiles of these agents. Therefore PHARMACHILD will establish a European registry to study pharmacovigilance of biologics in a cohort of JIA patients (children and young adults).
We will use the existing networks of PRINTO and PReS to collect the required pharmacovigilance data as well as disease outcome data. PRINTO and PReS have representatives in nearly all European countries. The existing national paediatric registries that monitor treatment outcome and/or adverse events and collect specific pharmacovigilance items have already been contacted and have committed themselves via PRINTO to be part of this initiative. The coordinators of the largest of these national registries (JuMBO, BSPAR-BNDR and BIKER-register) are a full network partner in PHARMACHILD. Initially the registry will be set up for prospective data collection of JIA patients under treatment.
Next we will retrospectively collect outcome (from the existing registries) and adverse events that occurred in the past from the local patient records.

Research objective 2: To ensure validated robust analysis of the occurrence and risk of observed serious adverse events, and in particular serious infections, malignancy and inflammatory gastrointestinal diseases, in children receiving immune modulatory drugs for JIA.
Based on the reports of the pharmacovigilance from the PHARMACHILD registry we will collect cases of adverse events in the following categories (malignancies; severe infections; IBD; other autoimmune diseases occurring under treatment). These cases will be further analyzed in detail by expert panels of: Paediatric oncologists, Paediatric infectious diseases experts, Paediatric gastro-enterologists. These panels will perform risk assessment of the use of biologics by studying event rates and the pattern of the diseases. The most important aspect when analyzing the risk of adverse events across a number of countries and a number of health care systems is to ensure that the adverse outcomes being compared all fit within a validated definition. To ensure this, the study proposes that once adverse events are reported to the registry, further information is collected by the PHARMACHILD register.
PHARMACHILD will offer a European framework for collection of data in a harmonized way, by the development of a regular reporting template. Adverse events should be coded in a uniform manner using the medical dictionary MedDRA in order to study these events properly. Presently the British and the German registries are using MedDRA. A working group will define which MedDRA terms are of specific interest and how they translate into the agreed “Manchester template”. This template is the format in which the adult registries who contribute to the EMA reporting present their semi-annual data. The template includes baseline data as well as follow up data and event rates. In the baseline form data will be included on: Personal details: Gender, Date of birth, Patient identification number, Date form completed, Date of informed consent; Drug information: Registered drug, Date drug commenced, Route, Dose, Dose Unit, Number of doses, Dose frequency, Main reason for starting registered drug; Details on the clinician: Unit/Department of treating consultant, Consultant, Person completing form if not consultant, Person obtaining consent. The follow-up form contains: Clinical rheumatological outcome (the core set parameters); Changes in medication; Adverse events reports (initial + follow-up using the MedDRA template).

Research objective 3: To analyze immunosurveillance mechanisms in a cohort of JIA patients under treatment with immuno-modulatory drugs and to identify a panel of risk factors for undesirable outcomes such as adverse events in JIA.
Juvenile idiopathic arthritis is characterized by episodic phases of inflammation substantially involving elements of the innate and adaptive immune system which frequently needs substantial long-term suppression of the immune system. On the other hand infections and even development of malignancies are severe side effects of anti-inflammatory long-term treatment especially with biologics. Patients would benefit from cost-effective surrogate markers for disease monitoring and outcome prediction. PHARMACHILD will collect serum and plasma from a small cohort of JIA patients for the evaluation of established and novel biomarkers. We will identify novel biomarkers which i) help to predict disease courses or complications in different JIA subtypes, ii) help with monitoring therapies, and iii) develop scoring systems to evaluate severity of disease. New therapeutic strategies evolving from the identification of these biomarkers, and their contribution to pathogenesis are additional long-term goals of this application. Identification of novel biomarkers, correlating to disease activity in JIA or predicting the risk of major side effects, may result in a better clinical care for patients. Such findings will be relevant for new treatment strategies and may help to individualize medication.

Research objective 4: To support regulatory decisions on marketing authorizations for the use of biologics in children by disseminating the project’s results to all relevant stakeholders, patient groups and the public at large.
PHARMACHILD will make an inventory of adverse events reported in literature. This study will perform as a baseline for the research activities. Together with the output resulting from the previous goals all data will be analyzed and combined. Based on this information draft reports will be drawn up with the conclusions, recommendations and best practices that were identified by the projects. The envisaged impact of our results will be supported by a range of dissemination activities. Our main target groups in our dissemination are: The European Commission in particular DG’s Research and Public Health (SANCO); National Public Health Authorities in Europe and beyond; The European Medicines Agency (EMEA); The US Food and Drug Administration (FDA); The Pharmaceutical industry. The scientific community at large; The general public.
The dissemination activities differ per target group. The main purpose of the dissemination activities to the EC, National Public Health Authorities, EMEA and FDA is to support regulatory decisions on marketing authorizations for medicinal products including the warnings in product information for doctors and patients. The reason for the dissemination of the project results to the scientific community at large is on the one hand to promote and protect public health by enabling more effective and safer use of medicines and on the other hand to facilitate colleagues in valuable further research on the efficacy and safety of medicines. Finally the public at large is informed on general safety issues of medicines in order to promote transparency and safeguarding of public health and to take away concerns about the safety of medicines at use.

Project Results:
Already the data from 1448 patients are entered in the prospective cohort. Since Pharmachild will not stop including data we need to exercise on intermittent time captures of the data. At this moment we captured 178 events of specific interest and 117 moderate/severe/serious adverse events in that prospective cohort. The most common ESI’s are serious and targeted infections and the most common AE’s are gastrointestinal, hepatobiliary and skin disorders.
In the more extensive retrospective database containing the data of 6450 patients with a median follow up of 4.9 years the following was found:

Malignancies:
A total of 6 patients with Malignancies were found from the Pharmachild database.
A total of 5 patients with Malignancies were retrieved from the Pharmachild database in patients exposed to MTX (N=5500).
A case of Malignancy occurred in patient with a preceding exposure to systemic steroids that had been interrupted 4 years before the onset of Malignancy; this patient received Biologic drugs after the onset of the ESI.
A total of 4 patients with Malignancies were retrieved from the Pharmachild database in patients exposed to Biologic drugs (N=4199).
Incidence Rate found in Pharmachild:
MTX usage: 1.814 /100,000 p-m= 0.022/100 patient years
Biogicals usage: 2.106/100,000 p-m = 0.025/100 patient years
This is comparable to the incidence rates we found before in literature (Milestone 18 and published in Arthritis, Research and Therapy in 2013; DOI 10.1186/ar4213) :
Healthy childhood population: 0.032/100 patient years
JIA: 0.025/100 patient years
MTX: 0.033-0,046/100 patient years
Conclusions on malignancies: Till now we did not find an increased risk for malignancies in the JIA population compared to the risk in the healthy childhood population.

Inflammatory Bowel Diseases:
A total of 13 patients with IBD were found from the Pharmachild database.
A total of 12 patients with Malignancies were retrieved from the Pharmachild database in patients exposed to MTX (N=5500).
In 1 patient the event occurred before the start of any treatment and therefore was not included in the analysis.
Incidence Rate found in Pharmachild:
MTX usage: 4.354/100,000 p-m= 0.052/100 patient years
Biogicals usage: 6.318/100,000 p-m = 0.075/100 patient years
This is higher than the incidence rates found before in literature for healthy children but lower for children using a biological [etanercept] (Milestone 18 and published in Arthritis, Research and Therapy in 2013; DOI 10.1186/ar4213) :
Healthy childhood population: 0.0083/100 patient years
JIA: unknown
MTX: unknown
Biological (etanercept): 0.31/100 patient years
Conclusions on IBD: The incidence rate of IBD in the JIA population has not before been described.
IBD was found to have a 6.3 times higher incident rate in the JIA population than in the healthy population (0.052 versus 0,0083 per 100 patient years). The use of a biological increased this risk a bit to 0.075 /100 pt years.

Other Events of Specific Interest and Adverse Events:
20.1% of the patients developed a moderate/severe/serious adverse event or an ESI.
Only 9.7% of these patients developed an event of specific interest. This was higher in the systemic JIA group (24.4%), but this could have been caused by the definition of ESI’s as well (e.g. macrophage activation syndrome was an ESI and only occurs in systemic JIA). The events of specific interest that were most frequently seen were: Infusion related reaction (12.6%), Neutropenia (6.9%), Injection related reaction (5.4%), Macrophage activation syndrome (5.2%), Tuberculosis (3.5%), but mainly Serious/targeted infections (59.0%).
Other AE's were found in the nonsystemic JIA group in 13.8% compared to the 17.2% in the systemic JIA group.
Gastrointestinal, hepatobiliary, eye, and skin disorders were amongst the most frequent AE’s (7-15% per field). Also Investigations (eg. Fecal tests, endoscopy) and surgical and medical procedures were
not uncommon (12.3 and 8.1% respectively).

For every center it is possible to export their data, analyze them and relate it to the whole consortium.
In that way every center can find out in what aspect they experience more or less adverse events than normally seen. This enables them to tackle these problems and improve their care. We have performed this analysis already for the center of Utrecht for serious infections.
Conclusions on other ESI's and AE's:
During almost 5 years of follow up the majority (79.9%) of patients did not experience moderate/severe/serious events or any event of specific interest. The majority of the events of specific interest were serious/targeted infections.

Laboratory results:
As shown for the non-systemic forms of JIA serum concentrations of S100A8/S100A9 above the normal range predicted disease flares during clinical remission accurately with high sensitivity and specificity.
We demonstrated for the first time that serum concentrations of S100A8/S100A9 in SJIA correlate closely with response to treatment with biologicals and disease activity and we introduced a novel and additional measurement for monitoring anti-inflammatory treatment of individual patients with SJIA. This novel parameter allows for the first time individual adaptation of immune-suppressive treatment in SJIA and will hereby avoid overtreatment and reduce long-term side effects.
Innate biomarkers have been found for all forms of JIA. We have identified and defined phagocytespecific S100-proteins as novel biomarkers which allow reliable and sensitive prediction of disease flares in JIA patients on medications with biologics. In addition, we have decribed S100-proteins as reliable predictors for response to therapy in JIA.

Flow cytometry confirmed previous findings that intracellular-TNFa-levels are increased in CD14+ cells after culture with etanercept compared to culture with no antibodies and rh-Fc control, while this is not the case for adalimumab. Also, the percentages TNFa+ monocytes increased while the percentage mTNFa+ monocytes decreased after culture with etanercept compared to culture with no antibodies. This is another effect that is much smaller after culture with adalimumab. Both antibodies have a limiting effect on TNFa secretion into the supernatant. Results are comparable for both HC PBMCs and JIA patient PBMCs although the patient PBMCs respond not to the same extent as HC PBMCs do. These results contribute to the knowledge about the effects of etanercept and adalimumab. Intracellular-TNFa-levels are increased in CD14+ cells after culture with etanercept, while this is not the case for adalimumab. Also, the percentages TNFa+ monocytes increased while the percentage mTNFa+ monocytes decreased much more after culture with etanercept compared to adalimumab. Both drugs have a limiting effect on TNFa secretion into the supernatant.

To study the regulation of immune responses during immunosuppressive treatment of JIA, the antigen presenting capacity of dendritic cells of JIA patients was analyzed before and during treatment with various immunemodulating drugs. We found that dendritic cells before and during treatment were able to take up antigen.
Furthermore, these cells were able to present antigen to autologous T cells. Antigen specific proliferation was observed before and after treatment with drugs in blood of JIA patients.
Proinflammatory cytokines were produced after antigenic stimulation by T cells. No significant differences were observed with healthy controls not receiving any kind of treatment.
In the Pharmachild project we also aimed to investigate whether there is a switch in phenotype of naturally occurring dendritic cells during treatment with immune-modulating drugs.
Dendritic cells can both induce immunity as well as tolerance. These functions coincide with the expression of surface molecules and the production of (pro-inflammatory) cytokines.
We extensively studied the expression of both immune-stimulating and immune-suppressing molecules on the surface of natural blood dendritic cell subsets. No significant differences were observed on plasmacytoid and myeloid dendritic cells before and after treatment with immunemodulating drugs or compared to healthy controls. Interestingly, we did observe a novel population of cells in the blood of JIA patients. These cells possessed functions resembling a population in between monocytes and BDCA-1+ myeloid dendritic cells. Characteristics of this novel cell population are: BCDA-1+, CD14+, efficient antigen uptake, expression of costimulatory molecule CD80 after stimulation, and induction of T cell proliferation (both memory as well as naïve T cell responses). However, the BDCA-1+CD14+ cells express high levels of PD-L1 and IL-10, both indicative of an inhibitory phenotype. Indeed antigen-specific proliferation of CD4+ T cells was suppressed by these cells. The presence and magnitude of the BDCA-1+CD14+ cells in JIA patients might be due to their disease and treatment status. In adults suffering from rheumatoid arthritis similar cells are observed in fluids of the inflamed joints, these cells are described to be involved in the induction and maintenance of Th17 cell responses (Segura et al Immunity 2013). These so-called inflammatory dendritic cells have a potential role in inflammation. Furthermore, these inflammatory dendritic cells are not only found in arthritic synovial fluid but also in tumor ascites. The here identified BDAC-1+CD14+ cells are also found in blood of cancer patients and therefore might be identified as an important population for further study in the blood of JIA patients.

Another study revealed that neither ANA status nor the subtype of JIA alone yields a distinct cytokine profile, but that categorising young ANA-positive JIA patients as a separate ILAR subcategory is further supported by low levels of SAA1 and high levels of sIL-2R. Interestingly, blocking of sIL-2R seems to be a valid immunotherapy for some autoimmune uveitis patients. Therefore, it could be that young ANA-positive patients would benefit from such treatment. sIL-2R and SAA1 need to be validated in a new cohort if they are to be used as markers of prediction for the development of uveitis.

Many scienitifc presentations were given and the patientgroups have been informed internationally.
Already 8 scientific articles have been published but with this dynamic registry and BMB surely more will follow soon.

Potential Impact:
1. A continuous large scale pharmacoviligance registry able to study the safety and efficacy of biologics in Juvenile Idiopathic Arthritis
2. Standardizing data collection on adverse events throughout Europe and beyond
3. Collaborating with US registries and other national registries in order to reach the maximum number of exposure years in order to get the best data available on safety.
3. Continuously new insights in the mechanisms that drive adverse events within the shortest possible time by combining knowledge and resources
4. Identification of surrogate markers for the prediction of the disease course
5. Generating new knowledge on adverse drug reactions in the use of biologics which impacts the balance of benefits and risks of these products
6. Enabling safer use of immune modulatory drugs (biologics) in children and young adults
7. Supporting regulatory decisions on marketing authorizations for biologics including the warnings in product information for doctors and patients
8. Promoting and protecting public health through more effective and safer use of biologics

List of Websites:

https://www.printo.it/project_ongoing_detail.asp?ProjectID=15
Contact email address: printo@ospedale-gaslini.ge.it