I-DSD: The First 10 Years

I-DSD was initially conceived as a registry following the Consensus Workshop on DSD which was jointly hosted by the European Society of Paediatric Endocrinology (ESPE) and the Lawson Wilkins Pediatric Endocrine Society of North America in 2005 when the need for pooling standardized information in cross-border registries was identified as a priority [1]. While databases and registers did exist at a regional and national level, they lacked key features that would be considered to be desirable when promoting research in a rare group of conditions, such as international uniformity and transparent structures for data governance.

With the help of a small project grant from the ESPE Research Unit in 2006, a group that consisted of Faisal Ahmed (Glasgow), Silvano Bertelloni (Pisa), Sten Drop (Rotterdam), Olaf Hiort (Luebeck), Ieuan Hughes (Cambridge), and Richard Sinnott (Glasgow) developed an initial ESPE DSD (Differences/Disorders of Sex Development) Register. The prototype ESPE DSD Register became the EuroDSD Registry when it was supported by EUFP7 between 2008 and 2011 as part of the EuroDSD project. In this project, it was anticipated that the EuroDSD Registry would facilitate the development of a “virtual research environment” which would accommodate the needs of the scientific work packages and the clinical contributors [2]. By the end of the EuroDSD project, the EuroDSD Registry was supporting several centres in countries beyond Europe and in 2011 it became the International DSD (I-DSD) Registry when it secured an international partnership grant from the UK Medical Research Council (MRC). During this period, the I-DSD Registry was also including cases of congenital adrenal hyperplasia (CAH) and to facilitate this further, under the guidance of a consortium led by Richard Ross (Sheffield), the Registry developed the I-CAH Registry, a mirror site which used the same registry platform as I-DSD, and which was launched in 2014.

Over the last few years, the I-DSD Registry has also included cases of Turner Syndrome (TS) and in 2020, the project started work on developing another mirror site (I-TS) which could use the same registry platform as I-DSD and I-CAH. This work has been led by a consortium led by Janielle van der Velden (Nijmegen). Following the end of the MRC anchor funding in 2017, the I-DSD and I-CAH Registries have been supported from funding from a variety of sources and by 2021, the registries I- were supporting a wide range of activities including research, education, and quality care improvement. Although it may seem that the project has taken on more functions than its original aim of creating “a virtual research environment,” the evolution of I-DSD can also be viewed as an acknowledgment that successful clinical research is closely linked to training and education as well as improvement in the quality of care. This review will provide some examples of these activities of the registries and their network.

All of the activities of I-DSD/I-CAH/I-TS are managed and administered by a joint Steering Committee which meets twice a year and is responsible for the overall direction of the initiative [3]. Initially, the composition of the Steering Committee was guided by the original needs of the EuroDSD project. However, as the project spread further across the globe and covered more conditions, the Steering Committee sought to represent as many stakeholders as possible. In addition to patient representatives, the committee also has representatives of cross-border professional societies that have a dedicated working group for DSD. By pursuing this policy, I-DSD has also sent a clear signal that international societies which represent professionals or scientists in the field of DSD should consider having a dedicated working group for DSD within their society. The project also has a Data Access Committee [4] which reviews new research proposals and a Care Quality Improvement Committee [5], both of which report to the Steering Committee. Within its committees, the project has ensured that the interests of its three constituents (DSD, CAH, and TS) are represented, but there are no specific representatives for any of these conditions given the wide phenotypic diversity and overlap across these three broad groups of conditions. Day-to-day management of the project is undertaken by the Project Management Group [6] which is based at the Office for Rare Conditions at the University of Glasgow.

As of May 2021, the I-DSD/I-CAH/I-TS had developed a network that reached 260 centres in 63 countries on all 6 habitable continents. Of these, 115 centres from 40 countries use the registries and have entered over 5,500 cases where information can be shared for a range of purposes that have the ultimate aim of improving the health of people with these rare conditions. The remaining 145 centres receive newsletters and participate in the other activities such as education events and surveys (Fig. 1). The current patterns suggest that although the rate of increase in the number of participating countries is now slowing down, new centres within the countries continue to join. The pattern of participation in the registries varies from one country to another, such that in some countries, such as Switzerland, all participating centres contribute to the registries, while in other countries such as the UK, there is a mix of centres with a proportion contributing actively to the registries while others choose to remain within the I-DSD network but do not actively participate in the actual registries (Fig. 2).

Of the cases, the current age of the individuals ranges from newborn to 72 years, with a median of 17 years. A total of 54% of these individuals are male and the most commonly registered diagnoses are CAH (n, 1,507, 28.4%), disorders of gonadal development (n, 1,419, 26.7%), and disorders of androgen action (n, 674, 12.7%) (Fig. 3).

Participation in research is a vital component of the project and since 2010, over 20 original data communications containing data from 86 centres have been published in peer-reviewed journals with several ongoing projects that are expected to bear fruit in the future [7]. Details of all active, as well as previous research, and related projects are available on the project website [8] and it is anticipated that this level of visibility will encourage more centres to join the collaboration and will also encourage other researchers to collaborate on existing projects. It also allows prospective researchers to identify gaps in the research activity. Lastly, this research visibility facilitates engagement with the wider public as well as patients who may have contributed to the project. The success of these projects depends heavily on the extent of participation from the clinical centres that provide the high-quality data that are necessary for a scientific publication. In addition to data, these centres also provide scientific input into the resulting publications. Some recent publications have had contributions from over 30 centres from almost 20 countries. While this level of partnership bodes well for the future success of the project, it is also clear that this level of attainment attracts more projects which in return increases the burden on participating centres. On average, it takes approximately 2 years from the date of project approval to a publication in a peer-reviewed journal, and prospective researchers need to be aware of this timeline.

Of the 46 projects that were approved by May 2021, 29 (63%) had patient level data supplied to the investigators, 4 (9%) were surveys of practice, 2 (4%) focussed on quality of practice, and 11 (24%) had not been activated. Of the 35 projects that had been activated, 31 (89%) were publicly or university funded, 3 (9%) were funded by industry, and one was funded by all three sources. Of the cases on the registry, 3,383 (59%) have been included in at least one published study, with a range of 0–8 studies per case. The most commonly studied conditions to date have been CAH (1,399 [41%] of cases in published studies) and disorders of androgen action (508 [15%] of cases in published studies). As such, there remains a wealth of data on individuals with other conditions such as Leydig cell defects and cloacal abnormalities, who have not yet been included in studies, but which will no doubt offer new directions and collaborations for future work, with the overall aim of improving the current understanding about these conditions.

A survey performed by I-DSD and the EU COST Action, DSDnet, in 2014 had revealed that of the 75 specialist DSD centres surveyed, 26 (35%) reported that they only kept a local DSD registry, 40 (53%) shared their data in a multicentre, national or the International DSD registry, and 9 centres (12%) did not record any data [9]. Another more recent survey performed to assess the knowledge and participation of specialist centres that were joining the European Reference Network for rare endocrine conditions in 2016 (Endo-ERN) showed that of 37 centres that had expertise in the field of sex development, 70% reported an awareness of a registry and only 52% reported participation in international registries for DSD [10]. The 2014 survey had also revealed that the main hurdles for participation in a registry were reported as lack of personnel in 64% of centres, lack of available time by 56% centres, and difficulties in obtaining consent by 27% centres [9]. However, it is increasingly becoming recognized that participation in registries and related activities is an essential feature of complex and specialist clinical services that deliver care in the field of rare diseases [11]. Not only do registries allow these specialist centres to participate in research, but involvement in their related network also facilitates professional development training and improvement in quality of care. Patient registries can also enable the development of local circles of patients and parents with similar conditions who can support each other.

The case for participating in standardized data collection and exchange for DSD has now been made at several levels by patients and professionals and it is likely that this will become standard practice in centres that care for people with DSD [11, 12]. Measures to reduce the burden on data entry include limiting the frequency of data entry for each patient. The number of core data fields that must be completed at time of patient registration have been limited to promote user participation and include consent, year of birth, underlying diagnosis, sex at birth, current gender, and karyotype. More detailed descriptive information including those collected at subsequent clinical encounters can be added at a later stage. It is unlikely that busy clinical centres will enter comprehensive information on all cases and they are advised to complete only those fields that are required by the active studies in which the centre is participating and which are listed at https://home.i-dsd.org/ongoing-studies/. Some ongoing projects that rely on long-term continuous monitoring such as the I-CAH acute adrenal insufficiency quality improvement project described at https://home.i-dsd.org/i-cah-registry-care-quality-report/ have developed their own solutions for promoting collection of relevant data by combining the centre-specific care quality report with a report on data quality.

In the future, it is possible that there may be other solutions such as automated extraction of data from local electronic health records using methods that rely on machine learning, but this will require further development and quality assessment. However, the need for dedicated personnel that assist with data entry and quality control at source will continue to be an important requirement for the foreseeable future. The project also offers other incentives to those who participate in research such as discounted registration to its scientific meetings as well as postgraduate training courses. Engagement with health care providers and patients through workshops and surveys performed through the COST Action DSDnet have identified several areas that are important to these two important groups of stakeholders and a challenge for the future is to encourage prospective as well as current investigators to use the I-DSD Registry to address these key priority areas [12‒14].

The International DSD symposium was held in Luebeck (Host, Olaf Hiort) in 2004, 2006, and 2011 and since then has been held biennially in Glasgow (Host, Faisal Ahmed) in 2013, Ghent (Host, Martine Cools) in 2015, Copenhagen (Host, Anders Juul) in 2017, and Sao Paulo (Host, Berenice Mendonca) in 2019. In 2021, the meeting was due to be held in Bern, Switzerland, but was replaced by a virtual meeting and to be followed by the original venue in 2022 (Host, Christa Flueck). The virtual meeting in 2021 was such a success that it is anticipated that in the future the biennial face-to-face meeting will alternate with a short virtual meeting in the gap year. For the first time, in 2022, a 2-day postgraduate course will also be held prior to the scientific meeting. This course has been structured on the highly popular training schools that were held annually between 2015 and 2017 by DSDnet, an EU COST Action [15], and the long-standing training schools of ESPE.

A number of studies that have been performed using data from the registries can be used to directly improve care by understanding the reasons for variation and providing feedback to individual centres through individualized centre-specific reports. An example of this is the study that was performed to examine the rates of sick day episodes and adrenal crises among children with CAH [16]. In addition to highlighting clinical determinants that altered the risk of these acute adrenal insufficiency-related adverse events, the study also identified differences between centres and countries [16]. In a similar vein, a study by Bacila et al. [17] identified systematic differences between countries on the dose of hydrocortisone used in CAH. In DSD, two studies have revealed differences in practice of gonadectomy in a wide range of conditions and provide a benchmark of practice [18, 19]. In addition to providing participating centres with centre-specific reports, studies such as these need to be repeated at regular intervals to examine whether regular feedback is associated with any change in practice. To increase the focus on care quality improvement, in 2021, the project created a new committee for care quality improvement with the role of increasing the focus on activities aimed at improving the quality of care of people with conditions such as DSD, CAH, or TS.

Although the initial scope of the project was to develop a registry for conditions associated with DSD, it rapidly became clear that the platform was suitable to collect information on several overlapping conditions. Given that DSD is an umbrella term that covers several conditions that affect sexual development and maturation, it was logical to extend to CAH and more recently TS, especially as several centres were already including cases with these conditions. While over a decade ago, there was relatively little interest in detailed disease registries for rare conditions, there are now over 800 rare disease registries in Europe alone. The usefulness and long-term sustainability of a detailed disease registry can increase significantly if the data that are stored within the registry adhere to the principles of FAIR data (findable, accessible, interoperable, and reusable). This involves designing it to comply with international standards of quality, structure and content, access control, and also adopting common methods and processes for information/patient discoverability and sharing [20]. Thereby, users can more easily compare, pool, and analyse patient datasets, using sufficient numbers of cases for meaningful clinical research and public health purposes in areas identified as high priority by the DSD community. A high-quality disease registry also requires clear governance structures and scrutiny and the effort involved in providing this is often underestimated. The I-DSD Registry has previously undergone a quality assessment [21] and has most features that experts would identify as essential features of a rare disease registry [22]. Recently, it has also created a detailed data dictionary that paves the way to making the data discoverable [23]. To improve quality of the data further, the I-CAH Registry has started to provide its users with feedback on the quality of data they enter through a centre-specific report (Fig. 4). In a recent review of existing registries in the field of rare endocrine conditions, the I-DSD Registry and its sister registry I-CAH were identified as commonly used registries for conditions affecting sex development and maturation. However, this survey also highlighted that several expert centres were unaware of these registries, so there is a need to continue raising their awareness [10]. Other conditions that may affect sexual development but may not be traditionally considered as DSD such as hypogonadotrophic hypogonadism and Klinefelter Syndrome are already included in the I-DSD Registry. In fact, of the 5,771 cases in May 2021, 479 (8%) had a karyotype of 47,XXY. Thus, it would seem appropriate to extend the registry to these important groups of conditions which would benefit from both standardized data collection and the high level of data governance afforded by the I-DSD Registry.

In summary, the I-DSD, I-CAH, and I-TS registries are powerful tools for improving our knowledge of a group of rare conditions where research, particularly in the field of outcome, has been hampered by the inherent rarity and heterogeneity of the conditions. The registries have also led to the development of a strong clinical and research network and it is anticipated that the wider user community will continue to use the registries to develop new studies and quality improvement exercises within all related disciplines.

The authors would like to thank past and current members of the I-DSD/I-CAH/I-TS Steering Committee, the Data Access Committee, the Care Quality Improvement Committee, and the Organizing Committee of the DSD Postgraduate Course. They would also like to thank the patients, clinicians, and the researchers who have contributed or used the resources to improve our knowledge of DSD. The authors would also like to thank Silvano Bertelloni, Sten Drop, Olaf Hiort, and Ieuan Hughes who along with Faisal Ahmed were the members of the ESPE DSD Working Group who developed the prototype ESPE DSD Registry.

A.K.L.H., S.R.A., C.M., M.E.R., M.M., and J.B. have no conflicts of interest to declare. S.F.A. is an associate editor of Hormone Research in Paediatrics.

A.K.L.H. is supported by the Joint NES/CSO Postdoctoral Clinical Lectureship Scheme. S.R.A. is supported by the Gardiner Lectureship at the University of Glasgow and an unrestricted education grant from Diurnal Ltd and Neurocrine Biosciences. M.E.R. is supported by Glasgow Children’s Hospital Charity. S.F.A. and J.B. are supported by the European Union’s Health Programme (2014–2020) on the EuRRECa project “777215/EuRRECa” and S.F.A. is also supported by the European Union’s Health Programme (2014–2020) on the EuRR-Bone project “946831/EuRR-Bone.” The I-DSD Registry was developed using support from the Medical Research Council (G1100236), the Seventh European Union Framework Program (201,444), and the European Society for Paediatric Endocrinology Research Unit and is maintained through fees applied for data access and through its biennial symposiums.

Angela K. Lucas-Herald, Salma R. Ali, Craig McMillan, Martina E. Rodie, Martin McMillan, Jillian Bryce, and S. Faisal Ahmed wrote sections of the review, redrafted versions, and read and approved the final draft before submission.