Experiences and hope in caregivers of children with Aicardi Goutières Syndrome (AGS)

Abstract

Aicardi Goutières Syndrome (AGS) is a type I interferonopathy that results in a spectrum of neurologic impairment. In general, neurodegenerative disorders of childhood strongly affect the quality of life (QoL) and perspective of the whole caregiver network. This paper aims to define the impact of AGS on affected individuals and families through a multi-method approach using QoL surveys and qualitative interviews.

Qualitative interviews (N=45) and QoL questionnaires, including the Pediatric Quality of Life-Generic Core (PedsQL-GC, N=46) and Family Impact (PedsQL-FI, N=45) modules, Caregiver Priorities and Child Health Index of Life with Disabilities (CPCHILD, N=30), and Caregiver TBI-CareQoL (N=37), were administered to caregivers of children with AGS. The open-ended interviews with guided questions addressed the caregivers’ perception of AGS and determinants of QoL. Interviews were conducted until thematic saturation and qualitative analyses generated salient themes.

Per QoL questionnaires motor abilities and activities of daily living were more impaired compared to other domains (Comfort and Emotion, Psychosocial Health, [mixed effect analysis with Bonferroni’s correction, p<0.0001; Wilcoxon signed-rank test, p<0.0001, respectively).

In the qualitative interviews, families described stress surrounding symptoms of AGS, poor neurologic development with communication and motor impairment, and the overall impact of the disease on the family.

This study shows the profound effects of AGS on families, which have global implications for family life. CPCHILD and PedsQL-FI are the preferred tools for routinely assessing the impact of disease in the AGS population.

INTRODUCTION

Aicardi Goutières Syndrome (AGS), one of the most common leukodystrophies, is characterized by a broad spectrum of progressive neurologic and systemic autoinflammatory complications.[1–7] While AGS is associated with a significant disease burden,[5–7] its impact on quality of life remains unexplored. Broadly, the caregivers of children with severe medical conditions report the importance of support from family, friends, and advocacy groups.[8–11] Social isolation and feelings of disconnection are common,[12, 13] as is the strain on personal relationships.[14, 15] Caregivers have been reported to express uncertainty and insecurity in many realms, including future health and finances.[13–16] Feelings of helplessness were also commonly reported.[17] All these constructs contribute to the overall perceived quality of life (QoL) impacts on affected individuals and their families.[18–20]

A multidisciplinary and tailored clinical care approach is critical in the context of a multisystemic disease such as AGS. However, this necessitates characterization of the disease-specific burden on affected individuals and their caregivers.[21]

QoL and the impact of disease in a pediatric population have been historically captured through the use of questionnaires.[22–24] In addition, qualitative interviews exploring the burden of disease have proved to be successful strategies for identifying caregivers’ priorities required to determine potential clinical endpoints.[23]

Recently, the Food and Drug Administration (FDA) published a guidance series on Patient-Focused Drug Development (PFDD), indicating a prioritization of patient-centric approaches in clinical trial design and conduct. This includes the selection and fit-for-purpose evaluation of Clinical Outcome Assessments (COAs). [25–30] One of the key topics is “Health Concepts” (HCs), which can be defined as aspects of health that are important to patients. The identification of these HCs is critical to derive Concepts of Interest (COI) required for COA selection and fit-for-purpose evaluation, per the FDA. Given forthcoming clinical trials for AGS,[31, 32] there is an urgent need to understand caregivers’ expectations and priorities for improvements in functional and emotional domains related to patient’s QoL. This process will inform processes to identify HCs, which is a key first step towards a patient-centric approach to clinical trial design.

The aim of this paper is to define the impact of AGS on affected individuals and their families through a multi-method approach using patient proxy reported quality of life/impact of disease surveys and qualitative interviews. This is critical to patient-centric clinical trial design and clinical endpoint selection in imminent therapeutic interventions in the form of clinical trials for AGS.

METHODS

Participants and Recruitment.

Children and their caregivers were recruited as part of the Myelin Disorders Biorepository Project (MDBP). MDBP is approved by the Institutional Review Board at the Children’s Hospital of Philadelphia (CHOP): # 14–011236. Written informed consent (consent for research) was obtained from all participants (or their legally authorized representatives if unable to provide consent due to age or diminished capacities) in the study.

Materials.

Impact of disease and quality of life questionnaires.

Quantitative assessments were selected by a panel of experts (FG, SL, SD, LA, AV) through a process of expert consensus (requiring an agreement of 80%).[33, 34] This included two patient-centric questionnaires (the Pediatric Quality of Life Inventory^™ Generic Core [PedsQL-GC] parent-report and the Caregiver Priorities and Child Health Index of Life with Disabilities [CPCHILD]) and two caregiver-centric questionnaires the Pediatric Quality of Life Inventory^™ Family Impact Module [PedsQL-FI], and the caregiver specific Traumatic Brain Injury (TBI)-CareQoL), previously applied in the context of another pediatric rare neurodegenerative disease.[33] . All measures are well-established, with evidence of reliability and validity.

The Pediatric Quality of Life Inventory (PedsQL) is a brief, standardized, generic assessment instrument that systematically assesses patients’ and parents’ perception of Health-Related Quality of Life (HRQOL) in pediatric patients with chronic health conditions.[35, 36] We administered two modules of the PedsQL: the PedsQL-GC (parent-report) and the PedsQL-FI. The PedsQL-GC (parent-report) explores the impact of the HRQOL in four functional dimensions: Physical, Emotional, Social and School/Work.[35] The PedsQL-FI explores HRQOL impact on the family network due to the child’s disease on the family network through 8 functional dimensions: Physical, Emotional, Social, Cognitive, Communication, Worry, Daily Activities, and Family Relationships.[37] Both modules are comprised of a series of items scored on a Likert scale ranging from 0–4, which are then converted into a reversed percentage (0=100%, 1=75%, 2=50%, 3=25%, 4=0%). For both PedsQL-GC and PedsQL-FI, the floor effect in this study was defined as scores below 20%.

The Caregiver Priorities & Child Health Index of Life with Disabilities (CPCHILD) is a validated measure of HRQOL exploring the impact on the health, comfort, and well-being of individuals with severe disability.[38, 39] It is comprised of six sections exploring key themes impacting the quality of life of affected children, such as Personal Care and Activities of Daily Living, Positioning, Transferring and Mobility, Comfort and Emotions, Communication and Social Interaction, Health, and Overall Quality of Life. For the purpose of this study, floor effect was defined as scores below 20%.

The Traumatic Brain Injury (TBI)-CareQoL (CareQoL) was created to assess the quality of life in individuals who experienced traumatic brain injuries and their caregivers.[40] This tool has shown the ability to capture caregiver strain,[41] feelings of being trapped,[42] vigilance,[43] anxiety,[44] emotional suppression,[45] and feelings of loss.[46] The CareQoL questionnaires were adapted in this context; the wording was modified from “TBI” to “disease”. This tool was selected as it was designed for a population that suffers an acute injury/decline, which resembles the acute phase in AGS.[7, 47]

The questionnaires described above were completed by one parent for each AGS affected individual, and were estimated to take up to two hours to complete.

Qualitative interview.

A semi-structured interview guide was developed by an expert panel of child neurologists and neonatologists (FG, SL, SD, LA, AV) based on clinical experience and the existing data on natural history of AGS. The interview included questions exploring disease impact on children based on the caregiver’s perspective, disease symptoms with the highest impact, symptoms yielding a perception of meaningful improvement, and the overall impact of the disease on caregivers and families in a domestic and healthcare environment (Supplemental Figure 1).[4, 7] The final interview guide was designed specifically for this study after repeated iterations among the research team and consensus among disease experts, including clinicians with substantial experience caring for children with AGS. All study visits took place in person, and interviews were audio recorded and transcribed verbatim into a text document by a professional medical transcription service (ADA Transcription). The transcribed document was then de-identified by the study personnel prior to further analysis. (Figure 1)

Figure 1.

Performance of the AGS cohort at the PedsQL-Generic Core. Overall, the assessed population showed lower scores in the Physical Health domain Standard Scores when compared to the Psychosocial Health domain (paired t-test, p<0.0001). When exploring the subdomains, Physical, Social, and School/Work are to be more impacted than the Emotional subdomain at the Mixed effect analysis with multiple comparisons with Bonferroni’s correction, p<0.0001 for all comparisons.

Data collection.

RedCAP surveys were issued via email, and qualitative interviews were conducted in the context of research encounters. When multiple caregivers were available for an affected individual, each caregiver was interviewed separately. Caregivers were recruited until thematic saturation (i.e., absence of new themes observed during interview administration)[48] was achieved in the interim analysis. There was a partial overlap between the cohort that completed surveys and the cohort that completed qualitative interviews: 6 caregivers completed both the CPCHILD and the interview, 9 caregivers completed the PedsQL modules and the interview, while 5 subjects completed both the CareQoL surveys and the interview. Baseline demographic information of affected individuals, such as disease onset, age at evaluation, race, and ethnicity, were collected from medical records. Demographic data on caregivers was not collected.

Data storage.

Study data was managed on the Research Electronic Data Capture (REDCap) database hosted at the Children’s Hospital of Philadelphia.[49, 50] REDCap is a secure, web-based software platform designed to support data capture for research studies, providing 1) an intuitive interface for validated data capture; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for data integration and interoperability with external sources. De-identified interview transcripts were stored and managed in encrypted folders in secure systems within the CHOP server, accessible only to study personnel.

Data analysis.

Quantitative analysis.

Descriptive statistics included frequency and percentages for categorical variables and computation of means, standard deviations (SD), medians, ranges (minimum to maximum values), and Interquartile range (IQR) for continuous variables. Mixed effect analysis with multiple comparisons with Bonferroni’s correction and Wilcoxon test and paired t-test were used to determine differences within survey domain and subdomain performance. Spearman correlation was used to assess the correlation between survey subdomain scores for pairwise complete pairs. The corrplot package in R was used to create correlation matrix figures.

Qualitative analysis.

All interview transcripts were uploaded to Atlas.ti v7 qualitative data analysis software for management and coding. This study used a template analysis (TA) approach to guide the data analysis. [51]We (AM and FG) developed a codebook based on a priori interests and our initial engagement with the data. This codebook was utilized to code the qualitative data through a consensus process. Interim analyses were conducted to ensure the codebook’s appropriateness. The final analysis followed the six steps of template analysis as outlined in previous research[52] (1) review of the collected dataset in its entirety to build familiarity, (2) identification of codes and broad themes through preliminary coding of a subset of data, (3) organization of initial codes and themes into meaningful clusters, (4) development of an initial codebook, (5) application of the codebook to additional interviews and revision of the codebook as necessary, and (6) finalization of the codebook for application to the full set of interviews and interpretation of findings. Members of the research team (AM, FG) discussed coding and revisions to themes throughout the analysis process. Code reports were reflected independently and collaboratively, and memos were used to document discussions and thoughts about the data. Analytical decisions were presented and defended to the entire research team, including disease experts. An overall code primary documents matrix for all codes across the primary documents (interview transcripts) was conducted to identify the most prominent codes and extract the top six absolute totals. Upon analysis of the code reports, the final codes were determined. Except for minor edits to enhance clarity and eliminate identifiable information or obscenities, all quotes are presented unchanged. The frequencies at which the themes were mentioned were described using quantifiers such as “many”, “some”, “few”.[53]

Data availability.

Anonymized data not published within this article will be made available by request from any qualified investigator.

RESULTS

Participating cohort

Questionnaire responders were as follows: 46 caregivers completed the PedsQL-GC, 45 completed the PedsQL-FI, 30 caregivers completed the CPCHILD, and 37 completed the CareQoL. The surveys were made available to the families in both in-person and online formats to be completed as per the family’s preference. The questionnaires were completed by one parent only for each AGS affected individual.

The demographic characteristics of the cohort are described in Table 1. The majority of caregivers completed surveys when the patient was a median of 7.40 to 7.60 years and when the disease length was a median of 7.14 to 7.32 years (Table 1). All genotypes were represented across survey types, and most subjects whose caregivers completed the surveys and/or interviews were Caucasian/White (63.3% to 75.8%). However, other races were also represented in the cohort (Table 1). A subset of caregivers (“Core-cohort” n=29) completed all quality-of-life surveys. When compared to the additional subjects not part of the core-cohort, the demographic distributions were overall not significantly different between cohorts (Wilcoxon rank sum or chi-square test: p>0.05) except for the disease length in CareQoL; subjects with only CareQoL (median=4.53 years) were at an earlier disease state than then core-cohort (median=7.89 years; Wilcoxon rank sum test: p=0.0427); Supplemental Table 1–3).

Table 1.

Cohort Demographics

	Caregiver Interviews	CPCHILD	PedsQL	PedsQL FIM	Care-QoL
	(N=33)	(N=30)	(N=46)	(N=45)	(N=37)
Sex
Female	14 (42.4%)	11 (36.7%)	15 (32.6%)	15 (33.3%)	12 (32.4%)
Male	19 (57.6%)	19 (63.3%)	31 (67.4%)	30 (66.7%)	25 (67.6%)
Race
Asian	1 (3.0%)	3 (10.0%)	3 (6.5%)	3 (6.7%)	3 (8.1%)
Bi-/Multiracial	2 (6.1%)	3 (10.0%)	4 (8.7%)	4 (8.9%)	4 (10.8%)
Black/African-American	2 (6.1%)	1 (3.3%)	1 (2.2%)	1 (2.2%)	1 (2.7%)
Caucasian/White	25 (75.8%)	17 (56.7%)	28 (60.9%)	28 (62.2%)	22 (59.5%)
Hispanic/Latino	1 (3.0%)	3 (10.0%)	4 (8.7%)	4 (8.9%)	3 (8.1%)
Unknown/Not Reported	2 (6.1%)	3 (10.0%)	6 (13.0%)	5 (11.1%)	4 (10.8%)
Genotype
ADAR1	5 (15.2%)	5 (16.7%)	7 (15.2%)	7 (15.6%)	7 (18.9%)
IFIH1	3 (9.1%)	4 (13.3%)	5 (10.9%)	5 (11.1%)	5 (13.5%)
RNASEH2A	2 (6.1%)	1 (3.3%)	1 (2.2%)	1 (2.2%)	1 (2.7%)
RNASEH2B	10 (30.3%)	10 (27.0%)	13 (28.3%)	13 (28.9%)	10 (27.0%)
RNASEH2C	0 (0.0%)	1 (2.7%)	1 (2.2%)	1 (2.2%)	1 (2.7%)
RNU7–1	0 (0.0%)	2 (5.4%)	3 (6.5%)	3 (6.7%)	2 (5.4%)
SAMHD1	8 (24.2%)	7 (18.9%)	9 (19.6%)	9 (20.0%)	7 (18.9%)
TREX1	5 (15.2%)	4 (10.8%)	7 (15.2%)	6 (13.3%)	4 (10.8%)
Age Symptom Onset (years)
Mean (SD)	0.32 (0.60)	0.574 (1.64)	0.54 (1.49)	0.55 (1.50)	0.57 (1.64)
Median (IQR)	0.08 (0.27)	0.170 (0.420)	0.16 (0.54)	0.17 (0.58)	0.17 (0.42)
[Min, Max]	[0, 3.00]	[0, 10.0]	[0, 10.0]	[0, 10.0]	[0, 10.0]
[Q1, Q3]	[0, 0.27]	[0, 0.42]	[0, 0.54]	[0, 0.58]	[0, 0.42]
Age at Assessment (years)
Mean (SD)	9.15 (7.02)	10.1 (8.30)	9.62 (7.80)	9.68 (7.87)	10.1 (8.30)
Median (IQR)	6.08 (11.1)	7.60 (7.90)	7.50 (7.73)	7.40 (7.90)	7.60 (7.90)
[Min, Max]	[1.50, 23.1]	[0.20, 31.1]	[0.20, 31.1]	[0.20, 31.1]	[0.20, 31.1]
[Q1, Q3]	[3.58, 14.7]	[4.20, 12.1]	[4.33, 12.1]	[4.20, 12.1]	[4.20, 12.1]
Disease Length (years)
Mean (SD)	8.83 (6.90)	9.52 (7.93)	9.08 (7.44)	9.13 (7.51)	9.52 (7.93)
Median (IQR)	5.68 (11.3)	7.32 (8.15)	7.23 (7.89)	7.14 (7.97)	7.32 (8.15)
[Min, Max]	[1.25, 23.0]	[0.20, 30.9]	[0.20, 30.9]	[0.20, 30.9]	[0.20, 30.9]
[Q1, Q3]	[3.41, 14.7]	[3.95, 12.1]	[3.84, 11.7]	[3.80, 11.8]	[3.95, 12.1]
Age Difference from Caregiver Interview (years)
n	33	6	9	9	5
Mean (SD)		3.50 (0.22)	3.51 (0.24)	3.52 (0.24)	3.42 (0.03)
Median (IQR)		3.41 (0.03)	3.42 (0.06)	3.43 (0.08)	3.43 (0.01)
[Min, Max]		[3.38, 3.95]	[3.35, 4.00]	[3.37, 4.00]	[3.37, 3.45]
[Q1, Q3]		[3.39, 3.43]	[3.37, 3.43]	[3.37, 3.45]	[3.42, 3.43]

Quality of life questionnaires

PedsQL-GC (n=46).

The Physical Health domain demonstrated the lower standard scores (median (IQR): 18.8 (39.1)) compared to the Psychosocial Health domain (median (IQR): 50.0 (21.7)) (n=46; Wilcoxon signed-rank test, p<0.0001) on the PedsQL-GC. When exploring the subdomains, Physical (median (IQR): 18.8 (39.1)), Social (median (IQR): 45.0 (20.0)), and School/Work (median (IQR): 41.7 (30.0)) demonstrated lower scores compared to the Emotional subdomain (median (IQR): 70.0 (20.0); mixed effect analysis with Bonferroni’s correction for multiple comparisons, p<0.0001 for all comparisons). Caregivers rated the Physical Health Domain as being severely affected, with floor effect identified in 61% of the subjects (28/46 individuals).

CPCHILD (n=30).

Activities of Daily Living and Mobility (median (IQR): 38.4 (12.3) and 33.3 (19.4), respectively) were rated as more impacted compared to the Comfort and Emotions Domain (median (IQR): 80.2 (10.3)); mixed effect analysis with Bonferroni’s correction, p<0.0001 for all comparisons) (Figure 2A). No performance below the floor was observed in the cohort. Caregiver’s Priorities showed a higher priority for items belonging to the themes of Comfort and Emotions, Overall Health, and Communication and Social Interaction (Figure 2B). When considering the top 10 items, all domains were represented.

Figure 2.

Performance of the AGS cohort at the CPCHILD. A. Overall, the assessed population showed lower scores in the domains exploring Activities of Daily Living and Mobility when compared to the Comfort and Emotions Domain at the Mixed effect analysis with multiple comparisons with Bonferroni’s correction, p<0.0001 for all comparisons. B. Section 7 of the CPCHILD showed a priority of responders for items belonging to the categories of Health (blue color), Comfort (grey color), and Communication and Social Interaction (green color). The dotted line was added to delineate the top 10 priorities.

PedsQL-FI (n=45).

As measured by the PedsQL-FI, there were comparable scores in the Parent Health-Related Quality of life (HRQOL) Summary Score (median (IQR): 53.8 (28.8)) when compared to the Family Functioning Summary Score (median (IQR): 50.0 (34.4); paired t-test, p=0.93; Figure 3). No floor effect was observed in the explored domains.

Figure 3.

Performance of the AGS cohort at the PedsQL-Family Impact. Overall, the assessed population showed comparable scores in the Parent Health-Related Quality of Life (HRQOL) Summary Score when compared to the Family Functioning Summary Score (paired t-test, p=0.9254).

CareQoL (n=37).

Of the 37 caregivers who completed the CareQoL (Figure 4), 4 responders did not complete the full set of surveys. The assessed population showed high levels of Caregiver Strain, Caregiver Emotional Suppression, and Caregiver Sense of Feeling Trapped when compared to Specific Anxiety and Feelings of Loss (mixed effect analysis with Bonferroni’s correction, p<0.0001 for all comparisons). No floor effect was observed in the explored domains.

Figure 4.

Performance of the AGS cohort at the CareQoL. Overall, the assessed population showed high levels of Caregiver Strain, Caregiver Emotional Suppression, and Caregiver sense of Feeling trapped when compared to Specific Anxiety and Feelings of Loss (Mixed effect analysis with multiple comparisons with Bonferroni’s correction, p<0.01 for all comparisons)

Correlations between patient-HRQOL questionnaires.

Correlation between performance within patient centric-HRQOL questionnaires (PedsQL-GC and CPCHILD) was then performed. Within domains from PedsQL-GC, the highest correlation was observed between the following sections: Social Functioning and School/Work functioning (Spearman Correlation r=0.53, p=0.0002, 95% CI: [0.26, 0.72]). In the CPCHILD, the highest between-domain correlation was between sections: (i) Communication and Social Interaction, and Positioning, Transferring, and Mobility (Spearman r=0.74, p<0.0001, 95% CI: [0.48, 0.88]), and (ii) Positioning, Transferring, and Mobility, and Personal Care and Activities of Daily Living (Spearman r=0.63, p=0.0002, 95% CI: [0.31, 0.82]). Between the PedsQL-GC and CPCHILD assessments, the highest correlation was observed between the CPCHILD Health and the PedsQL-GC School/Work functioning sections (Spearman r=0.52, p=0.0040, 95% CI: [0.12, 0.74]), and between the CPCHILD Comfort and Emotion and the PedsQL-GC Physical Functioning (Spearman r=0.50, p=0.0067, 95% CI: [0.13, 0.75]). All correlation analyses are included in Figure 5.

Figure 5.

Correlation Matrix of the CPCHILD and PedsQL-Generic Core. Within the same tool, the highest correlation was observed between the PedsQL-GC Social Functioning and School/Work functioning sections (Spearman Correlation r=0.53, p=0.0002, 95% CI: [0.26, 0.72]) and the CPCHILD Communication and Social Interaction and Positioning, Transferring, and Mobility sections (Spearman r=0.74, p<0.0001, 95% CI: [0.48, 0.88]), and Positioning, Transferring, and Mobility and Personal Care, Activities of Daily Living sections (Spearman r=0.63, p=0.0002, 95% CI: [0.31, 0.82]). Between the two assessments, the highest correlation was observed between the CPCHILD Health and the PedsQL-GC School/Work functioning sections (Spearman r=0.52, p=0.0040, 95% CI: [0.12, 0.74]), and between the CPCHILD Comfort and Emotion and the PedsQL-GC Physical Functioning (Spearman r=0.50, p=0.0067, 95% CI: [0.13, 0.75]).

Qualitative interviews

Overall, 45 caregivers of 33 subjects completed the interview (27 mothers, 18 fathers). There were no significant differences in the responses of the mother and father within each family. There was minimal overlap between the participants in the quantitative and qualitative components of this study (Table 1), and there were no significant differences in the demographic characteristics of the children cohorts (Supplemental Tables 4–7). One caregiver declined participation in the qualitative interview. Five major themes emerged from our analysis and are represented using quotes as described by the caregivers (Table 2). There were no differences in themes by genotype, severity, or time from disease onset. The results are presented in aggregate.

Table 2.

Themes and example quotes

Theme	Example quotes
Theme 1: AGS impacts the entire child. Motor, cognitive, and social limitations hinder both physical and mental function and transform a child’s life. (N=189)
Global neurologic impact	“[AGS] has affected his whole development “ “He’s still not really met any milestones up until now other than smiling and cooing and laughing ”
Cognitive and communication abilities	“Not being able to communicate back what he knows, is trapped in his mind” “I think she tries to speak. I mean, when I ask a question, she will go - and it doesn’t come out.” “I think that there is a whole world in here and - of all these emotions and all these frustrations and all these joys and all of these things that she wants to share, and she can’t”
Motor abilities	“You can tell he tries and wants to do something, but his body just doesn’t follow”
Social interactions	“I feel like she - a lot of times she sees her restrictions almost more than I do. And so, she won’t be eager to participate unless she knows that she will be welcomed in an activity... in other situations with other adults then she might not naturally try to include herself in an activity because she’s nervous of how she’ll be able to do it, how she’ll be able to accomplish it”.
General wellbeing and independence	“There’re experiences that he hasn’t had that he should have had by now, and it’s just completely just transformed his day- to-day life” “He can’t run around or have fun” “It’s impacted every part of her life - from development to learning to everyday activities.”
Theme 2: Families greatest hopes and priorities for their children include improvement of functional abilities and a focus on happiness and independence to improve quality of life. (N=98)
Hope for improved communication	“If he could speak, and I could know what he was thinking so that he could communicate with me and be included in the world” “I wish I knew what he was thinking. When he was younger, when he was in therapies, they used to ask me, so what is it that you think you would love for him to do? And I used to say walk. I wish he could walk. But now as I’ve gotten older and he’s gotten older, I really wish he could speak. Walking to me is not important anymore but knowing what he’s thinking is more important”
Hope for improved motor skills	“If he could walk, I’m telling you that would mean I could get him a job. I could - it would be different for us if he was mobile.” “One big thing, miracle this, for him just to move his arm is so huge for me already, for him to just prop himself a little would be huge ”
Hope for happiness and comfort	“I want him comfortable - no pain, less discomfort” “I think mainly just him being settled and happy is the first thing to improving his quality of life.
Hope for independence	“I think that that would be a huge thing for her to have her independence” “To be able to feed herself and prepare her own food” “Being able to drive a wheelchair...”
Theme 3: The abrupt and often severe onset of AGS symptoms caused significant stress and uncertainty _for_families. (N=298)
Mental status and systemic symptoms	“The first two years of his life were complete [obscenity], complete misery” “It was just constant crying, like all day crying, not just certain periods. It was all day and night crying” “She had 35 fevers in 14 months” “She stopped eating” “Sleepy all the time” “She stopped sleeping” “Major, major skin rashes. Constantly in the emergency room, draining puss from his fingers.”
Developmental delay and regression	“Apparent right away that he was severely affected, disabled” “He stopped pulling himself up, he wouldn’t stand up on his own anymore...And then, he started having a little bit of difficulty crawling...It wasn’t until he actually had problems sitting himself upright on his own...that we started getting concerned.” “Within three weeks she couldn’t sit, stand, walk or crawl, and within a month and a half she couldn’t use her hands”
Theme 4: AGS impacts the entire family emotionally, socially, and financially. (N=165)
Emotional impact	“It is physically demanding and emotionally and mentally” “[It has been] a roller coaster of emotions” “survival is the best word to use. We are trying to survive ” “It’s emotionally exhausting for a mother and father to have to sit and retell that story over and over and over again... People don’t listen well to parents of kids with special needs and It’s very frustrating” “There are times really that I’m so tired and still so sleepy. So, sometimes maybe I feel guilty of not really doing my part” “I don’t even have the luxury of sleep anymore. I know it’s affected me [...]. It’s so tiring, I wasn’t prepared for this”
Career and financial impact	“I’ve switched my job positions from days to nights to try to accommodate for him”. “I had to stop working. I have to stop working to take care of her”. “I need to do my job, and I know I need to do it well so I can continue to provide for my family and then we can have the medical coverage we have ” “It’s almost not worth me going to work when we have to pay [for childcare]”
Marriage impact	“I would give anything for my child not to have AGS, but it has brought my family closer” “We’re a somewhat dysfunctional family because that’s now part of it. It causes emotional problems within us ”
Theme 5: Families cope with the impact of AGS by reaching a place of acceptance and relying on both medical and faith-based support systems. (N=139)
Acceptance	“It took a few months after his diagnosis...to just, I guess, psychologically accept that my life was going to be very different, and you have to be content with that”
Faith	“My faith is very strong, and people always say, God doesn’t give you more than you can handle”
Medical system	“It’s very comforting to be here with a team ofpeople that understand this disease”

Theme 1: AGS impacts the entire child. Motor, cognitive, and social limitations hinder both physical and mental function and transform a child’s life.

The multi-systemic nature of AGS and its association with variable degrees of disability[4, 5, 7] was reported by caregivers. One caregiver reported that AGS has “impacted every part of her life”. Stress regarding motor limitations was commonly reported by caregivers as the feature of AGS that had the largest impact. The dissociation between motor and cognitive function was noted as a source of concern: “[he] wants to do something, but his body just doesn’t follow”. Families noted the negative impact of AGS on peer interactions and self-esteem. One family described the impact of AGS on their child’s self-perception: “A lot of times she sees her restrictions almost more than I do. And so, she won’t be eager to participate unless she knows that she will be welcomed in an activity.”

Theme 2: Families’ greatest hopes and priorities for their children include improvement of functional abilities and a focus on happiness and independence to improve quality of life.

Most caregivers described a desire for an improvement in neurologic function. Most families voiced the desire for their children to be able to better communicate, such as “I would love […] to be able to express when he is hurting or uncomfortable or not feeling well or happy”. Caregivers expressed hope for improved motor skills: “It would be different for us if he was mobile.” Even small gains were important to families: “One big thing, miracle […] for him just to move his arm is so huge for me already, for him to just prop himself a little would be huge”. Common goals were improved comfort, happiness, and independence for their children: “I think mainly just him being settled and happy is the first thing to improving his quality of life”.

Theme 3: The abrupt and often severe onset of AGS symptoms caused significant stress and uncertainty for families.

Universally, the families described stress associated with the early features of AGS. As described by one participant, AGS can result in the acute onset of severe symptoms: “… he started screaming one night nonstop out of nowhere and before that he never cried [...]. And from that moment on for about three and a half months, he would not eat. He would not sleep. And he screamed 24/7”. Many families noted distress surrounding the early symptoms: “He would just cry and cry, and we couldn’t figure out what it was”. Other common sources of stress were feeding difficulties, recurrent fevers, skin inflammation, and abnormal sleep patterns.

Theme 4: AGS impacts the entire family emotionally, socially, and financially.

Caregivers described AGS as having major effects on the entire family. Many caregivers noted an emotional burden from AGS, which included feelings of inadequacy and challenges in sharing their children’s condition with others. One caregiver reported feels of guilt for “not really doing my part” because of fatigue. Families described the stress on marriages, with both positive and negative effects. The economic burden of AGS on families was reported, with specific concerns related to the costs of treatments and insurance coverage. Several caregivers reported changes in their employment after the diagnosis of AGS: “I had to stop working. I have to stop working to take care of her”.

Theme 5: Families cope with the impact of AGS by reaching a place of acceptance and relying on both medical and faith-based support systems.

Caregivers reported a spectrum of overlapping approaches to coping with AGS: faith, reliance on the medical system, self-care, and formal psychotherapy approaches. Several caregivers noted the importance of having a support system that understands AGS: “The problem is people are uncomfortable around him, and they want to help, but they don’t know how to help. So, even though you feel that there is support, it’s still different.”

DISCUSSION

This study is the first to describe the QoL impacts of AGS on the well-being of the affected children and their families. The evaluated patient-centric HRQOL questionnaires confirmed the severe impact of AGS on motor skills and activities of daily living and are in alignment with prior reports based on natural history studies and case reports in the literature.[4, 7, 47] Despite covering similar themes, the HRQOL questionnaires demonstrated different strengths and weaknesses in this unique population. In the CPCHILD, we observed discordance between domains of function most impacted by the disease (motor and activities of daily living) and most domains important for overall quality of life (comfort and health-related domains). This apparent discrepancy might be explained by changes over time in family priorities based on disease duration and progression, interventions, and caregiver’s coping strategies, as previously described in the literature.^[54]

We observed how caregivers are severely affected by their child’s condition from a practical and emotional perspective. The output from the PedsQL-FI was in alignment with other leukodystrophy populations.[55] We observed that the CareQoL tools demonstrated lower compliance with some dropouts. Based on feedback from families, we hypothesize that decreased completion of questionnaires may correlate with the time and emotional burden of the surveys.

The themes that arose in the qualitative interviews were consistent with the questionnaire findings. The caregiver’s concerns about the impact of early AGS-related multi-systemic manifestations aligned with published medical reports [4, 5, 7] necessitating early diagnosis.[56]

We observed that functional impairment was a source of stress regardless of severity. While motor function impairment in AGS is typically severe, a subset of children exhibits milder impairments. [57] Nevertheless, motor impairment was consistently reported as a significant source of stress across the entire population, even among those with an attenuated phenotype. This aligns with previous studies highlighting the association between chronic disability, mobility impairment, and poor self-image. [58] Furthermore, many families described a notable disconnection between receptive and expressive language abilities. These observations underscore the need for approaches that better assess receptive language function to capture cognitive abilities in the AGS population accurately.

As already reported in other contexts,[59–62] caregivers of children with AGS described the severe emotional and economic impact of the disease, as well as the strain on marital relationships. Despite this, many families described their desires and hopes for the future, stressing the importance of small changes in neurologic function. This suggests that symptomatic relief, maximizing neurologic function, and providing family support should be critical components of the comprehensive care of children with AGS, as well as targets of therapeutic interventions.

Our study has several limitations. As it is based at a large US-based specialty care center, it only includes families who consented to participate in research. As such, our cohort could be biased to preferentially include families with greater clinical knowledge of AGS and improved access to medical resources. In order to mitigate this potential concern, we offered remote access to the impact of disease surveys, reducing the potential enrollment bias within our cohort. Another related limitation of the study is its inability to assess differences in demographic characteristics of caregivers participating in the study.

The qualitative interview guide was developed, and consensus was determined only among disease experts. Caregivers’ feedback was not solicited, which could impact the caregivers’ interpretation of the question and could impact responses during the interviews. While we did not observe thematic differences based on the duration of illness, our numbers were insufficient for analysis. The study population included in this project had a long disease history, and further studies comparing a younger population with an older population or long-term longitudinal studies within a cohort would be required to address this issue. It is likely that both the stress from a diagnosis and family priorities change over time.

Additionally, the cohorts included in the quantitative and qualitative approaches of our study overlap only minimally. As the presentation of AGS is heterogeneous, we may not have equal representations of the disease phenotypes across both sections of the study. Finally, while we were able to explore the existing correlation between HRQOL questionnaires, we were unable to determine the internal consistency validity (ICV) or intercorrelation reliability (ICR) of the COAs due to the limited sample size,[63, 64] warranting future evaluation in the context of a larger study.

This is the first study to evaluate stressors and quality of life of children with Aicardi Goutières Syndrome and their caregivers through the use of a multi-method approach. We demonstrate that validated quality of life questionnaires are feasible in AGS. While we hypothesize that these COAs are applicable in longitudinal settings to explore the impact of potential therapeutic interventions, future research is warranted. Additionally, through descriptive correlation between quality of life questionnaires, we were able to get a more granular understanding of the multidimensional impact of AGS.

Our data support several unique aspects of AGS that are common sources of stress: early irritability, profound motor limitations, and the disconnect between receptive and expressive language. This work will facilitate the identification of AGS-specific Health Concepts that, in alignment with recent Food and Drug Administration (FDA) guidelines on Patient-Focused Drug Development (PFDD),²⁶ will inform the design of future patient-centric clinical trials. By applying a multi-methodology, including a qualitative and quantitative component to explore the impact of disease in rare conditions, the study enabled us to provide a more accurate impression of the daily impact of the disease on affected children and their families.