Background
Due to their nature and the non-specific symptoms at presentation and during the early phases of the disease, rare diseases are often un- or misdiagnosed for extended periods, leading to a long diagnostic delay [
1‐
4]. Patients may visit many different healthcare professionals and undergo multiple unnecessary investigations before the correct diagnosis is finally achieved [
1‐
4]. This diagnostic odyssey may significantly add to the burden of the disease [
1,
2,
4]. An early diagnosis is particularly essential if a disease-modifying treatment is available because the patients’ outcome often depends on the timely initiation of treatment [
5‐
7]. Finally, because approximately 80% of rare diseases are inherited, an early diagnosis may allow genetic counseling and informed decision-making in family planning (
https://www.eurordis.org/sites/default/files/publications/Fact_Sheet_RD.pdf).
To prevent unnecessarily delayed diagnoses, numerous campaigns have been launched to increase awareness of rare diseases. Many campaigns, such as the ‘rare diseases day’ initiative, which has become a yearly event in many countries worldwide, are of a general nature, raising awareness of the existence of ‘rare diseases’. Other initiatives focus on specific diseases and promoting an early diagnosis, thereby allowing the timely initiation of treatment [
8‐
10] (
http://www.rarediseaseday.org/events/world). These campaigns are organized by patient advocacy groups, health care providers and pharmaceutical companies.
However, to the best of our knowledge, no studies have specifically investigated whether these campaigns have reduced the diagnostic delay. We investigated the time to diagnosis of two very rare, invariable progressive and severe, inborn errors of metabolism: mucopolysaccharidosis type I (MPS I; estimated birth prevalence 1:100,000) for which treatment has been available for more than 15 years, and mucopolysaccharidosis type III (MPS III; estimated birth prevalence 1:60,000) for which treatment is under study (Table
1). Both disorders belong to the group of lysosomal storage disorders. We assessed whether the diagnostic delay has decreased over recent decades.
Table 1
Symptoms frequently observed in MPS I and MPS III patients and information regarding the different phenotypes and enzymatic subtypes
Mucopolysaccharidosis type 1 (MPS I) |
MPS I – Hurler (MPS I-H) | 607,014 | α-L-iduronidase (IDUA) | Dermatan sulfate (DS) and heparan sulfate (HS) | Progressive neurocognitive decline, hernias, facial dysmorphisms, corneal clouding, stiff joints, dysostosis multiplex, cardiac problems and hepatosplenomegaly. Death in childhood if untreated. | HSCT | 1.07/1.19 per 100.000 newborns |
MPS I – Hurler-Scheie (MPS I-H/S) | 607,015 | | | Phenotype intermediate between MPS I-H and MPS I-S. Can present with or without neuronopathic disease. | HSCT or ERT | |
MPS I – Scheie (MPS I-S) | 607,016 | | | Corneal clouding, stiff joints, mild dysostosis multiplex. Normal intelligence en life expectancy. | ERT | |
Mucopolysaccharidosis type 3 (MPS III) |
MPS IIIA | 252,900 | Heparan N-sulfatase (SGSH) | Heparan sulfate (HS) | Progressive neurocognitive decline, behavioral problems, sleep disturbances, progressive loss of motor functions. Death in second or third decade of life. Broad spectrum of disease severity. | Not available | 1.52/1.89 per 100.000 newborns |
MPS IIIB | 252,920 | N-acetyl-α-glucosaminidase (NAGLU) | | | | |
MPS IIIC | 252,930 | Acetyl CoA:α-glucosaminide N-acetyltransferase (HGSNAT) | | | | |
MPS IIID | 252,940 | N-acetylglucosamine 6-sulfatase (GNS) | | | | |
Methods
Patients
This single center study was conducted at the Academic Medical Center (AMC) in Amsterdam and involved interviews with patients and/or parents or legal guardian(s) of patients with MPS I and MPS III with a confirmed diagnosis since 1988. Before 1988, reliable data were unavailable. The data were verified and/or supplemented with chart reviews or data inquiries from the general practitioner (GP) and the medical specialist(s) visited prior to diagnosis. Our center is a center of expertise for MPS I and MPS III in the Netherlands.
All MPS I and MPS III patients were included regardless of the phenotype. Table
1 presents the symptoms frequently observed in MPS I and MPS III patients and information regarding the different phenotypes and enzymatic subtypes [
11‐
15]. The phenotypes were assessed by an experienced clinician (FAW) based on the available clinical data. Only patients with a diagnosis confirmed by enzymatic testing and/or a mutation analysis were included. Patients were only included if the diagnostic studies leading to the final diagnosis were based on the clinical symptoms. Patients who underwent diagnostic studies because of an affected family member were excluded. All patients and/or their parents or legal guardians provided informed consent for this study. The study proposal was reviewed by the Medical Ethics Committee of the AMC, who deemed that formal ethical approval was not necessary for this study.
Data collection
The data were collected using structured telephone interviews with patients and/or the patients’ parents or legal guardian(s). The following variables were recorded:
- Year/month of first visit to the GP for a symptom that was, in hindsight, likely related to MPS I/MPS III.
- Year/month of first referral visit to a medical specialist for a symptom that was, in hindsight, likely related to MPS I/MPS III.
- Year/month of the confirmatory diagnosis, which was defined by the first demonstration of deficient enzyme activity or the presence of disease causing mutations.
From each of these visits, the following data were recorded:
- MPS I/MPS III-related symptom leading to the visit.
- Other MPS I/MPS III-related symptoms present at that time point.
- Type of medical specialist visited at first referral for a disease-related symptom.
- Type of medical specialist who made the diagnosis.
MPS I and MPS III disease-related symptoms are presented in Table
2.
Table 2
Disease-related symptoms for MPS I and MPS III
Hernias
• Inguinal hernia • Umbilical hernia |
Developmental delay or decline
• Neurocognitive functions • Motor functions |
Ear, nose, throat problems
• Frequent upper airway infections • Obstructive sleep apneas or excessive snoring during sleep • Tympanostomy tubes • Adenoidectomy • Tonsillectomy |
Behavioral problems
• Hyperactivity/restlessness • Aggression • Anxiety • Autistic behaviors • Other |
Gastro-intestinal problems
• Hepatosplenomegaly |
Dysmorphic features
• Coarse facial features • Coarse hair • Hirsutism • Other |
Cardiac problems
• Cardiomyopathy • Valvular dysfunction |
Skeletal and joint problems
• Joint stiffness • Skeletal deformities • Kyphosis • Hip dysplasia • Bullet shaped metacarpals • Stunted growth of the long bones • Broad oar shaped ribs • Short stature • Carpal tunnel syndrome • Trigger fingers • Tendon shortening • Early arthrosis |
Ear, nose, throat problems
• Frequent upper airway infections • Frequent ear infections • Hearing problems • Tympanostomy tubes • Adenoidectomy • Tonsillectomy |
Gastro-intestinal problems
• Frequent diarrhea • Hepatomegaly • Other |
Hydrocephalus
|
Sleeping problems
|
Corneal clouding
|
Seizures
|
Dysmorphic features
• Frontal bossing • Depressed nasal bridge • Full lips • Macroglossia |
Hernias
• Inguinal hernia • Umbilical hernia |
Developmental delay
|
Statistical analyses
The statistical analyses were performed using SPSS software for Windows (version 23.0, SPSS Inc., Chicago, Illinois, USA). Non-parametric Mann-Whitney U tests were performed to assess the significant differences in the time between the first visit to the GP and diagnosis and the time between the first visit to a medical specialist and the final diagnosis within the cohort of MPS I patients and between the Hurler and non-Hurler patients. The same analyses were performed for the RP and SP MPS III patients.
To assess whether the diagnostic delay changed over time, the MPS I and MPS III patients were divided into different groups based on the year of diagnosis using a 5-year time interval. Non-parametric Kruskall-Wallis tests were performed to assess the significant differences among these groups.
Discussion
This study is the first to report the diagnostic odyssey in MPS I and MPS III patients in the Netherlands. We demonstrate the presence of a substantial diagnostic delay in both MPS I and MPS III patients without a reduction in the time between the first consultation with a medical doctor (GP or medical specialist) for disease-related symptoms and the time of the final diagnosis over a 20-year period.
In the Dutch healthcare system, patients, including children, are typically first seen by a GP, who may refer the patient to a medical specialist. Thus, the time to diagnosis after the visit to the GP was longer than the time between the visit to a medical specialist and the diagnosis. Remarkably, the longest diagnostic delay was observed after the first visit to a medical specialist, particularly in the MPS III patients.
The MPS I patients were diagnosed at a significantly younger age than the MPS III patients, which is most likely due to the early manifestation of the somatic symptoms [
11,
16], leading to earlier medical attention and referral. MPS I patients with the severe Hurler phenotype were diagnosed at a significantly younger age than the non-Hurler patients. The median age at diagnosis in the Hurler patients was comparable to that reported in previous studies [
2,
17‐
19]. However, the more attenuated non-Hurler patients in our cohort were diagnosed at an earlier age than that reported in other studies [
14,
17‐
20]. This finding may be due to the relatively small sample size of non-Hurler patients in our cohort. The lack of a decrease in the time to diagnosis over the previous two decades is disappointing and worrisome for two reasons. First, an early diagnosis allows for the early initiation of treatment and better disease outcomes. Treatment with hematopoietic cell transplantation (HCT) for MPS I Hurler was first shown to be effective in halting or preventing the cognitive decline in the early 1980s and is currently the treatment of choice for this group of patients. Earlier HCT leads to better outcomes [
5,
21,
22]. In addition, intravenous enzyme replacement therapy (ERT) is the treatment of choice for MPS I patients with a non-Hurler phenotype, and studies have demonstrated that an early start of treatment is beneficial [
6,
7,
23,
24]. Second, to reduce the diagnostic delay and promote early diagnosis, numerous MPS I awareness campaigns have been launched, particularly after the introduction of ERT for the treatment of the somatic symptoms in 2003. These campaigns included direct mailings to health care professionals in the Netherlands presenting the typical features of MPS I patients, expert lectures on early symptoms of MPS I at scientific meetings of relevant medical specialists (including pediatricians, ENT specialists, pediatric rheumatologists and pediatric neurologists) and exhibit booths of a pharmaceutical company commercially marketing ERT for MPS I (Genzyme Sanofi) providing educational material on lysosomal storage disorders, including MPS I, at major relevant medical conferences in the Netherlands. Our data indicate that these efforts have not led to a significant reduction in the time to an MPS I diagnosis.
In our cohort of MPS III patients, the diagnosis was established at a significantly younger age in the severe RP patients (age 54 months; 4 years and 6 months) than in the SP patients (age 71 months; 5 years and 11 months). However, the diagnostic process preceding the diagnosis did not differ between the two groups, and the age at final diagnosis is comparable to observations reported in other studies [
25‐
27]. Although no disease-modifying treatment is currently available, several clinical trials, including intrathecal ERT and gene therapy, have recently been initiated for MPS III types A and B [
28,
29]. An early diagnosis and early start of treatment before the onset of progressive cognitive deterioration are considered essential. Given that patients with the RP phenotype plateau in development by 30 months and exhibit rapid cognitive decline at 40 – 50 months, a diagnosis should be made before the age of 3 years to allow the initiation of therapy at the optimal timing [
13]. This goal, however, was only achieved in 9% of the patients in this study, and no decrease in age at diagnosis was observed over the previous 20 years.
Our study has some limitations. First, we defined diagnostic delay as the time between the first visit to a GP or medical specialist for a potential disease-related symptom and the final diagnosis, whereas diagnostic delay generally refers to the time between the onset of symptoms and diagnosis in other studies [
14,
30,
31]. However, we consider the use of the time of symptom onset susceptible to a significant recall bias, whereas the time of the first visit to a medical doctor can be verified, thereby providing more reliable data. Second, our study has a retrospective design. Nevertheless, the amount of missing data was small, and the data could be verified in the medical records. In addition, due to the rarity of both disorders, a prospective design is not feasible. Third, the number of patients included in our study was small. Given that we were able to recruit almost all patients from the Netherlands diagnosed with MPS I and MPS III between 1988 and 2017, we assume that our data reliably represent the situation in our country. Larger scale, multi-national, studies on the diagnostic delay in patients with MPS or other rare or ultra-rare diseases are needed to corroborate our findings. In Europe, such studies may be initiated by the recently established European Reference Networks for rare diseases (ERNs) (
https://ec.europa.eu/health/ern_en). Finally, MPS I and MPS III are ultra-rare (ultra-orphan) diseases because they affect less than one person per 50,000 people (
http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32014R0536). The results of our study might not be applicable to relatively more common rare diseases affecting one person per 2000 – 50,000 people.
The lack of a reduction in the diagnostic delay over time was previously reported for MPS I by d’Aco et al.
, based on data from an observational international MPS I registry [
18]. In addition, a study investigating the time of diagnosis in Pompe disease, which is a lysosomal storage disease in which the timing of the start of therapy (ERT) is essential, to the surprise of the authors, also failed to demonstrate a reduction in the diagnostic delay despite improved diagnostic laboratory techniques allowing for a rapid diagnosis [
32]. Multiple efforts to increase awareness of Pompe disease and expedite its diagnosis have been exerted globally over recent decades.
Determining why awareness campaigns for rare diseases fail to reduce the diagnostic delay in MPS I and III and Pompe disease is challenging. Due to the very low birth prevalence of these disorders, many specialists, including GPs, general pediatricians, orthopedic surgeons and ENT specialists, may visit with no or only one undiagnosed patient during their entire career. Awareness of specific (combinations of) symptoms of a (ultra) rare disease may be lacking when confronted with a patient (many) years after exposure to an awareness campaign. Long-lasting knowledge regarding the symptoms of (ultra) rare diseases can likely only be achieved by intensive repetitive learning, which is not a feasible option for all medical specialists. Furthermore, because most symptoms at presentation are not specific, considerable time is generally spent excluding more common disorders.
Several alternative strategies are possible. One strategy involves the selective screening of groups of patients with certain symptoms but without a diagnosis of the rare disease of interest. Such studies have been performed for MPS I and included studies investigating MPS screening in patients with previous surgical repair or the presence of inguinal and/or umbilical hernia in combination with pediatric ENT surgery and children visiting rheumatology, hand or skeletal dysplasia clinics (
clinicaltrials.gov identifiers: NCT02095015, NCT01675674). Both trials have been terminated. To the best of our knowledge, these results have not been published, suggesting a failure to identify significant numbers of otherwise unrecognized patients. A study investigating screening patients under the age of 18 years with carpal tunnel syndrome for MPS also failed to detect patients with MPS [
33]. The extremely low yield of screening certain groups of patients for an ultra-rare disorder likely discourages participation, leading to the discontinuation of these programs. The yields of selective screening may improve when groups of patients are screened for a multitude of disorders, thus obviating the need of knowledge regarding specific rare disorders. Because the diagnostic approach in children with impaired cognitive development may significantly differ among health care systems in different regions of the world and obtaining an early diagnosis in patients with MPS III is very difficult, screening of children with an intellectual developmental disorder for several rare diseases may significantly reduce the diagnostic delay. A diagnostic algorithm for the identification of treatable causes of cognitive impairment has been proposed [
34], and several publications have demonstrated the importance of an early metabolic screening in all patients with unexplained developmental delay [
35,
36]. In addition, a review by Cleary and Green [
37] provided a guideline for the metabolic screening of patients with a developmental delay. The authors emphasize that IEMs can present with isolated developmental delay and that any regression of skills is suggestive of an IEM and warrants an intensive metabolic investigation. The slowing of cognitive development with a speech delay is one of the first symptoms of MPS III and often occurs before the age of 2.5 years; these symptoms could lead to an early diagnosis if these guidelines are followed. However, as the median age at diagnosis of patients with the most common RP phenotype is 54 months (range 34 – 79 months) in our study, it is clear that these guidelines are not used in the Netherlands. Indeed, the current guideline by the Dutch Society for Pediatrics (NvK, 2005) recommends screening for IEMs only if additional symptoms are present and not in in the presence of isolated cognitive delay (
https://www.nvk.nl/Portals/0/richtlijnen/mentale%20retardatie/mentaleretardatie.pdf). Fortunately, a new guideline is currently under development.
An interesting option for the (near) future is computer-assisted diagnosis, which can expedite the diagnosis of rare diseases. Artificial intelligence, deep learning and even a 3D facial analysis may assist clinicians during the diagnostic process, suggesting both diagnoses and appropriate investigations based on information in the electronic patient records [
38‐
40].
Finally, newborn population screening (NBS) may ensure very early diagnosis in patients with rare diseases and should be considered if a disease meets at least the following criteria (first proposed by Wilson and Jungner in 1968) [
41]: (a) the condition is an important health problem; (b) a suitable test for diagnosis is available; (c) a latent or early symptomatic state is recognizable; (d) the understanding of the condition’s natural history is adequate; and (e) an acceptable treatment for patients with a recognized disease is available. Because MPS I is considered to meet these criteria, this disorder has been introduced in NBS programs in the USA and Taiwan [
42] and will be introduced in the NBS panel in the Netherlands (
https://zoek.officielebekendmakingen.nl/blg-775624.pdf). However, this will lead to new challenges, including the detection of pseudo deficiencies for MPS I, as well the challenges often associated with newborn screening such as uncertain diagnoses and the inability to predict the phenotype, which may lead to significant emotional burden [
43‐
45]. MPS III is currently not considered eligible for NBS because no disease-modifying therapy is yet available.