Patient Perspectives on Treatment Outcomes and Priorities in Autoimmune Bullous Diseases: An Exploratory Survey among Dutch Patients

Abstract

Introduction

Autoimmune bullous diseases (AIBDs), comprising pemphigoid and pemphigus diseases, have seen limited therapeutic advances beyond rituximab for pemphigus vulgaris. As novel therapies are evaluated in clinical trials, well-defined, uniform, and relevant outcomes with patient involvement are essential. To date, however, patient-reported outcomes remain underrepresented, leaving uncertainty about whether trial results genuinely reflect patients’ expectations. This study examines perspectives of patients with AIBD on treatment outcomes and priorities to ensure that future research focuses on what matters most to them.

Methods

A cross-sectional study was conducted among Dutch patients with AIBD between October 2023 and January 2024, using a self-developed questionnaire with both closed- and open-ended questions to assess patient perspectives on treatment outcomes and priorities, key factors in choosing a treatment, and indicators of treatment success.

Results

Regarding skin and/or mucous membrane complaints, “the formation of new blisters and wounds” emerged as the most important complaint a treatment should address for both pemphigoid (43%) and pemphigus (86%) patients. In open-ended questions, patients with pemphigoid most frequently prioritized “pruritus” (44%), while patients with pemphigus emphasized “pain” (39%). Most important concerns regarding physical and daily functioning were “vision problems” (20%), “sleep disturbances” (20%), and “self-care difficulties” (23%) for patients with pemphigoid, whereas patients with pemphigus most commonly cited “eating and/or swallowing difficulties” (57%) and “daily activity limitations” (41%). Regarding emotional/psychological functioning, both subgroups prioritized “anxiety and/or worry” as most important concern (pemphigoid: 28%, pemphigus: 43%). Side effects were identified as the most important factor in choosing a treatment (pemphigoid: 41%, pemphigus: 39%). The absence of one or more symptoms and clinical signs (i.e., “no blisters”) was mentioned as the most important indicator of treatment success (pemphigoid: 88%, pemphigus: 91%), although minimal clinical signs (i.e., “minimal blisters”) were also considered acceptable (pemphigoid: 25%, pemphigus: 13%).

Conclusion

Patients with pemphigoid and pemphigus exhibit some distinct treatment priorities, reflecting their distinct pathomechanisms. Nonetheless, both subgroups consistently prioritize not only the resolution of disease-specific clinical signs but also preservation of physical and psychological well-being, underscoring the need for more holistic and patient-centered outcome measurement to ensure the establishment of meaningful, AIBD subgroup specific treatment outcomes.

Introduction

Autoimmune bullous disease (AIBD) represents a heterogeneous group of rare blistering disorders, caused by autoantibodies against adhesion molecules of the skin and/or mucous membranes [1]. AIBDs can be categorized into two main groups: pemphigoid diseases, comprising subepidermal bullous dermatoses like bullous pemphigoid (BP), and mucous membrane pemphigoid (MMP) and pemphigus diseases, comprising intraepidermal bullous dermatoses such as pemphigus vulgaris and pemphigus foliaceus [2, 3].

AIBDs are associated with substantial morbidity and impaired quality of life [4–6]. Additionally, patients have higher mortality rates compared to age-matched controls, mainly due to secondary infections [7, 8]. The main complaints of patients include pruritus, pain, and stigmatization from visible blisters and scarring [8].

The development of international guidelines for the most common AIBDs has significantly enhanced patient care [9–11]. Nevertheless, the limited availability of treatment options remains a significant challenge. Aside from rituximab, which is approved for pemphigus vulgaris, no new therapies have been approved for AIBDs in recent decades [9]. For other AIBDs, such as BP, high-dose corticosteroids remain the cornerstone of treatment, despite their considerable and often burdensome side effects [11–13]. However, emerging treatments such as dupilumab show potential to enhance the current therapeutic landscape [14, 15].

The field of AIBD is increasingly focused on developing and accessing novel therapies in clinical trials [16–18]. However, for effective interpretation and comparison of trial data, it is crucial to have well-defined, relevant and uniform trial outcomes. Recent studies on outcome measurement in BP, MMP, and pemphigus clinical research revealed that most studies reported outcomes related to clinical response, such as complete remission, while patient-reported outcomes remain significantly underrepresented [19–21]. As a result, it remains unclear whether trial results genuinely reflect patients’ expectations and perceived satisfaction with treatment.

Incorporating patient perspectives in defining relevant outcomes is crucial for gaining insights into the real-world impact of their disease and ensuring clinical relevance [22, 23]. Therefore, this study aims to examine the views of patients with AIBD, specifically for those with pemphigoid and pemphigus, on treatment outcomes and priorities, highlighting shared and distinct concerns between the 2 patient groups to ensure that future research addresses what matters most to patients with AIBD.

Methods

This study is reported following the STROBE guideline for observational studies [24].

Study Design, Setting, and Participants

We conducted a cross-sectional study among Dutch patients with AIBD between October 7th, 2023 and January 10th, 2024. Patients attending the annual Dutch AIBD Patient Day at the Center of Expertise for Blistering Diseases of the University Medical Center Groningen (UMCG) were invited to participate through an information letter and an accompanying oral explanation. Subsequently, members of the Dutch AIBD patient organization who did not attend the annual Dutch AIBD patient day were invited to participate by email using the organization’s mailing list.

Patients were eligible to participate if they (1) had a confirmed diagnosis of an AIBD, (2) were >16 years old, and (3) were proficient in the Dutch language to understand and answers the survey questions correctly. After signing informed consent, patients received the survey either by email or by mail, depending on the patient’s preference. Reminder emails and letters were sent after 2 weeks.

Data Collection

The questionnaire was developed by the research team based on previous surveys and existing quality of life questionnaires (online suppl. Appendix S1; for all online suppl. material, see https://doi.org/10.1159/000552182) [25–28]. Responses were collected and stored using REDCap v12.4.31 (Vanderbilt University).

The questionnaire contained closed-end questions in the first and second section and open-ended questions in the third section. The first section focused on respondent characteristics. Active disease was defined as the continued presence of blisters and/or wounds, pruritus, and/or pain, whereas inactive disease was defined as the complete absence of these symptoms.

The second section covered questions regarding “treatment outcomes,” defined as “what is being measured to assess the effect of a treatment” [29]. These close-end questions were divided into four categories of treatment outcomes: (1) skin and/or mucosa complaints, (2) physical functioning, (3) daily functioning, and (4) emotional/psychological functioning. Within each category, patients were asked to select their treatment priority, defined as the most important complaint or problem a treatment should primarily address. Only one answer option was allowed per category. If patients experienced a complaint or problem that was not listed among the predefined answer options, they could describe it in free text under “other.” Patients who indicated a combination of answer options were excluded from the subsequent analysis. The third section contained open-ended questions, allowing unlimited responses, to identify their most important treatment priorities, factors guiding their treatment choices, and their personal indicators of treatment success.

Statistical Analysis

Descriptive statistical analyses were performed using SPSS Statistics 26.0 (IBM Corporation, Armonk, NY, USA). Categorical data were presented using frequency distribution. Microsoft Excel (2018) (Microsoft, Redmond, WA, USA) was used to create figures and perform content analysis of qualitative data from open-ended questions.

Results

A total of 117 patients were invited for study participation, of which 55 successfully completed the survey and were eligible for inclusion (response rate: 47%). The inclusion and exclusion process is shown in Figure 1.

Fig. 1.

Flowchart showing the inclusion and exclusion process of study participants. *Netwerk voor blaarziekten. AIBD, autoimmune bullous disease; BP, bullous pemphigoid; MMP, mucous membrane pemphigoid; NBP, nonbullous pemphigoid; EBA, epidermolysis bullosa acquisita; PV, pemphigus vulgaris; PF, pemphigus foliaceus.

Patient Characteristics

Among the 55 patients, 32 were diagnosed with pemphigoid and 23 with pemphigus. The characteristics of these patients are shown in Table 1. The mean age was higher in the pemphigoid group (73.3 years) compared to the pemphigus group (65.0 years), and most patients in both groups were female (pemphigoid: 53%, pemphigus: 57%). At the time of the survey, patients with pemphigoid had more active disease (pemphigoid: 63% vs. pemphigus: 30%).

Table 1.

Participants characteristics (n = 55)

​	Pemphigoid (n = 32)	Pemphigus (n = 23)
Mean age, years (range)	73.3 (52–86)	65.0 (36–80)
Sex, male (%)	47	43
Disease duration, n (%)
0 to <5 years	16 (50)	11 (48)
5 to <10 years	9 (28)	5 (22)
10 to <15 years	2 (6)	1 (4)
15 to <20 years	2 (6)	4 (17)
≥20 years	3 (9)	2 (2)
Disease status, n (%)
Active1	20 (63)	7 (30)
Not active2	11 (34)	13 (57)
Unknown	1 (3)	3 (13)
Actual treatment, n (%)
None	9 (28)	9 (39)
Systemic	21 (66)	10 (43)
Prednisone	5 (16)	2 (9)
Rituximab	3 (9)	8 (35)
Dapson	2 (6)	0 (0)
Methotrexate	7 (22)	0 (0)
Antihistamines	4 (13)	0 (0)
Local	17 (53)	8 (35)
Corticosteroids	17 (53)	8 (35)
Work situation, n (%)
Retired	26 (81)	13 (57)
Paid work	5 (16)	10 (43)
Volunteer	4 (13)	1 (4)
Job seeker	1 (3)	1 (4)
Householder	1 (3)	1 (4)
Disabled	1 (3)	1 (4)
Not further specified	0 (0)	1 (4)

¹Active: continued presence of blisters and/or wounds, pruritus, and/or pain.

²Not active: the complete absence of blisters and/or wounds, pruritus, and/or pain.

Most patients with pemphigoid and a substantial number of patients with pemphigus were receiving systemic therapy (pemphigoid: 66%, pemphigus: 43%). Patients with pemphigoid were primarily treated with methotrexate (n = 7/32; 22%), while patients with pemphigus were mainly treated with rituximab (n = 8/23; 35%). Moreover, a notable number of both groups received additional topical therapy with corticosteroids (pemphigoid: 53%, pemphigus: 35%). Forty-three percent of treated patients had more than one therapy. Regarding their work situation, a higher percentage of patients with pemphigoid was retired (n = 26/32; 81%) compared to those with pemphigus (n = 13/23; 57%).

Treatment Priorities: Most Important Complaints and Problems to Address

Among the close-end questions with a single answer option regarding “complaints of skin and/or mucous membranes,” “the formation of new blisters and wounds” emerged as the most important complaint that a treatment should address for both subgroups. However, this was reported more frequently by patients with pemphigus (n = 19/22; 86%) compared to those with pemphigoid (n = 12/28; 43%). In the pemphigoid group, pruritus was also a commonly reported complaint (n = 8/28; 29%) (Table 2).

Table 2.

Responses to the question, “which complaint or problem should a treatment primarily address?”, divided into four distinct outcome categories

Which complaint or problem should a treatment primarily address?	Pemphigoid	Pemphigus
1. Complaints of skin and/or mucous membranes, n (%)	n = 28	n = 22
Formation of new blisters and wounds	12 (43)	19 (86)
Pruritus	8 (29)	0 (0)
Bleeding blisters and wounds	1 (4)	2 (9)
Pain	2 (7)	0 (0)
Discoloration and scars	1 (4)	0 (0)
Burning sensation	1 (4)	0 (0)
Other	1 (4)	0 (0)
Not applicable	2 (7)	1 (5)
2. Physical functioning, n (%)	n = 30	n = 23
Problem with eating and/or swallowing	5 (17)	13 (57)
Problem with vision	6 (20)	1 (4)
Problem with sleeping	6 (20)	2 (9)
Problem with intimacy	2 (7)	0 (0)
Problem with speaking	1 (3)	1 (4)
Other	2 (7)	1 (4)
Not applicable	8 (27)	5 (22)
3. Daily functioning, n (%)	n = 30	n = 22
Problem in performing daily activities	4 (13)	9 (41)
Problem in self-care	7 (23)	3 (14)
Problem in engaging is social activities	4 (13)	2 (9)
Problem in the professional domain	1 (3)	0 (0)
Other	3 (10)	0 (0)
Not applicable	11 (37)	8 (36)
4. Emotional/psychological functioning, n (%)	n = 30	n = 23
Anxiety and/or worry	10 (33)	10 (43)
Feelings of depression and/or sadness	6 (20)	3 (13)
Shame or discomfort due to my blistering disease	5 (17)	2 (9)
Forgetfulness and inability to think clearly	1 (3)	3 (13)
Frustration and/or anger	1 (3)	0 (0)
Other	0 (0)	0 (0)
Not applicable	7 (23)	5 (22)

All included AIBD patients (n = 55) could select only one answer option per category. Patients who indicated a combination of answer options were excluded from the subsequent analysis.

When asked in open-ended questions about their most important treatment priorities, “pruritus” was the most frequently reported priority in the pemphigoid group (n = 14/32; 44%), while “pain” was the most commonly reported in the pemphigus group (n = 9/23; 39%) (online suppl. Fig. S1). For the categories “physical functioning,” “daily functioning,” and “emotional/psychological functioning,” a substantial proportion of the complaints or problems were “not applicable” to either patients with pemphigoid or pemphigus. For example, 37% (n = 11/30) of the patients with pemphigoid and 36% (n = 8/22) of the patients with pemphigus had no problems with “daily functioning,” as indicated in Table 2. Nevertheless, concerning “physical functioning,” patients with pemphigoid rather frequently reported “problems with vision” (n = 6/30; 20%) and “problems with sleeping” (n = 6/30; 20%) as their most important concerns. Conversely, most patients with pemphigus reported “problems with eating and/or swallowing” (n = 13/23; 57%) as their most important problems.

Regarding “daily functioning,” several patients with pemphigoid reported “problems in self-care” (i.e., personal hygiene) (n = 7/30; 23%), and a substantial portion of patients with pemphigus reported “problems in performing daily activities” (n = 9/22; 41%). Concerning “emotional/psychological functioning,” “anxiety and/or worry” was the primary concern in both subgroups (pemphigoid: 33%, pemphigus: 43%). “Feelings of depression and/or sadness” (n = 6/30; 20%) and “shame or discomfort due to my blistering disease” (n = 5/30; 17%) were also quite frequently reported by patients with pemphigoid.

In total, 6 patients with pemphigoid and 1 with pemphigus provided free text under “other.” These included more general concerns such as “difficulties with learning to live with this autoimmune disease” and with “moving forward in a positive direction, processing, and accepting,” as well as personal concerns, such as “I can function normally, but the excessive tearing impacts my vision and because I am constantly preoccupied with it, it makes me upset.”

Decisive Factors for Choosing a Treatment

When patients were asked in an open-ended question about “what factors are decisive for you when choosing a treatment?”, patients from both subgroups most frequently mentioned treatment-related side effects (pemphigoid: 41%, pemphigus: 39%). Expected treatment efficacy was less frequently reported (pemphigoid: 28%, pemphigus: 22%) (Fig. 2). Efficacy was described by the patients as “proven scientific effectiveness,” “achieving remission,” and “that blisters disappear and will not reappear.” Some patients reported the specific characteristics of side effects, such as “severity of side effects,” “long term side effects,” and “side effects of corticosteroids.”

Fig. 2.

Overview of decisive factors for choosing a treatment according to patients with pemphigoid (n = 32) and pemphigus (n = 23).

Indicators of Treatment Success

When patients were asked in an open-ended question about “the most important indicator of treatment success,” both subgroups reported the absence of one or more symptoms and clinical signs, such as “no blisters,” “no eye complaints,” “no wounds,” “no lesions,” “no itching,” “no pain,” “no burning sensation,” or “no swallowing problems” (pemphigoid: 88%, pemphigus: 91%). Among these, “no blisters” was mentioned most frequently (pemphigoid: 25%, pemphigus: 35%). Additionally, in the pemphigoid group, “no itching” was mentioned most commonly (n = 7/32; 22%).

Minimal clinical signs, such as “minimal blisters,” “minimal eye complaints,” “two-third decrease in lesions,” “minimal clinical signs in general,” and “minimal itching or minimal pain” were also relatively frequently considered important indicators of treatment success (pemphigoid: 25%, pemphigus: 13%). Other indicators of treatment success were “no or acceptable therapy usage,” “acceptable side effects,” and “clear insight into disease mechanisms” (Fig. 3).

Fig. 3.

Overview of indicators of treatment success according to patients with pemphigoid (n = 32) and pemphigus (n = 23).

Discussion

This survey provides insights into the perspectives of patients with AIBDs regarding treatment outcomes and priorities, key factors in choosing a treatment, and indicators of treatment success, highlighting both shared and subtype-specific differences, particularly between pemphigoid and pemphigus. Regarding complaints involving the skin and/or mucous membranes, both patients with pemphigoid and pemphigus mentioned the formation of new blisters and wounds as their primary treatment priority, consistent with AIBD research priorities defined by the James Lind Alliance [30]. However, this priority was reported twice as frequently by patients with pemphigus compared to those with pemphigoid. Open-ended responses revealed further distinctions: in pemphigoid pruritus was most frequently prioritized (44%), while in pemphigus pain was the leading concern (39%). This aligns with their distinct pathomechanisms: pemphigoid, particularly BP, typically presents with severe pruritus, tense blisters, erythema, and urticarial plaques, while pemphigus often begins with painful erosions in the oral cavity, followed by flaccid blisters and erosions on the skin that rupture easily [2, 3].

This pathomechanistic difference was also reflected in problems regarding physical and daily functioning. Patients with pemphigoid commonly reported vision problems (20%), which may arise from disease manifestations such as ocular scarring in MMP [3, 31]. In contrast, patients with pemphigus frequently reported problems with eating and/or swallowing (57%) and performing daily activities (41%), consistent with painful oral erosions that may restrict food intake and contribute to weight loss [32]. Since eating is a key daily activity, this may account for the relatively high proportion of patients with pemphigus reporting limitations in performing daily activities.

In pemphigoid, sleep disturbances (20%) were also reported relatively frequently, possibly due to the frequent use of corticosteroids in this group, which are known to induce a range of side effects, including insomnia [31]. The differences observed between the two AIBD subgroups highlight the need to refine current consensus-based outcomes, as these do not equally capture the disease specific treatment priorities of both subgroups [33–35]. Whereas the consensus statement for BP encompasses both the “absence of new or established lesions” and the “absence of pruritus,” the corresponding statements for MMP and pemphigus are restricted to the “absence of new or established lesions,” without consideration of pain or scarring. Such refinement is essential to enable a more personalized approach to AIBD management, ensuring that patient needs of each subgroup are properly addressed. Importantly, these components should be reported separately so that each retains equal importance.

Moreover, commonly used outcome measurement instruments such as the Dermatology Life Quality Index (DLQI) [28], the Autoimmune Bullous Disease Quality of Life (ABQOL) [36], and the Treatment of Autoimmune Bullous Disease Quality of Life (TABQOL) questionnaire [27], address many aspects of AIBD patients’ quality of life but do not fully capture all these subgroup-specific quality-of-life concerns, underscoring the need for ongoing validation and refinement of these questionnaires.

Regarding emotional/psychological functioning, both subgroups identified “anxiety and/or worry” as their primary concern. A recent survey of the International Pemphigoid and Pemphigus Foundation (IPPF) reported that this concern is mainly related to fear of disease recurrence and loss of treatment efficacy [37]. A subset of patients with pemphigoid additionally reported “feelings of depression and/or sadness” and “shame or discomfort,” which were less frequently reported by patients with pemphigus. This may, as mentioned earlier, reflect the more common use of oral corticosteroids in pemphigoid, which can induce mood disorders and visible changes like weight gain and “moon face,” potentially exacerbating body image concerns, leading to poor self-image, stigmatization, and social isolation [31]. Despite growing evidence of the substantial psychosocial burden of these conditions, current guidelines focus on managing disease flares without addressing these psychosocial impacts [38]. This highlights the need for treatment outcomes that extend beyond clinical response and underscores the importance of integrating psychological support in AIBD treatment to addresses overall patient well-being.

When asked about decisive factors in treatment choice, “side effects” were most frequently reported in both subgroups (pemphigoid: 41%, pemphigus: 39%), exceeding the importance of “expected efficacy” (pemphigoid: 28%, pemphigus: 22%). These findings align with previous studies and likely reflect patients’ negative experiences with multiple therapies, particularly the frequent use of high-dose corticosteroids and cumulative side effects [8, 37]. As observed in other diseases, prior disease and treatment experiences influence their acceptance of side effects [39]. This underscores the need for steroid-sparing therapies that are more targeted, effective, and better tolerated. Furthermore, appropriate patient counselling about the short- and long-term effects of AIBD treatments is crucial to manage expectations and enhance treatment satisfaction.

Both patients subgroups considered the absence of one or more symptoms and clinical signs as indicator of treatment success, consistent with consensus-based AIBD treatment goals, which typically prioritize complete remission on/off therapy [33, 34]. However, minimal clinical signs were also relatively frequently considered as satisfactory, suggesting that the strict definition of complete remission (no new or established lesions for at least 2 months) [33, 34] may require a more nuanced interpretation. Similarly, the IPPF survey found that patients often viewed “resolving lesions” or periods of “minimal disease activity” as acceptable treatment outcomes [37]. These findings highlight the importance of early patient and advocacy group involvement in trial design and consensus-building to develop relevant and less stringent treatment outcomes that more accurately reflect patients priorities.

This survey provides the first nuanced view of Dutch AIBD patients’ perspectives. Limitations include potential selection bias from recruiting patients at an annual event, which may underrepresent those with more severe disease. The frequent “not applicable” responses concerning functioning in the close-ended questions suggested mild disease. The small sample size, partly due to the rarity of the disease, and the single-center design limits generalizability and scientific rigor. Furthermore, the survey lacked some specific details, such as the timeframe within an outcome should be achieved, and as the survey relied on patients’ perceptions, recall bias may be present.

A thorough understanding of patient preferences and perceptions regarding treatment priorities and therapy-related risks is essential for identifying unmet needs. This study shows that patients with pemphigoid and pemphigus exhibit some distinct treatment priorities, reflecting their distinct pathomechanisms and disease characteristics. Nonetheless, both groups consistently prioritize not only the resolution of disease-specific clinical signs but also preservation of physical and psychological well-being. These findings underscore the need for a more holistic approach to outcome measurement, with early and systematic involvement of patients and advocacy groups in trial design and consensus development to ensure the establishment of meaningful, AIBD subgroup specific treatment outcomes.

Acknowledgments

We extend our gratitude to José Duipmans and Birthe Ruiter for their support in disseminating the survey.

Statement of Ethics

This study was performed in accordance with the Declaration of Helsinki. This human study was approved by the Medical Research Ethics Committee of the University Medical Center Groningen (UMCG), which confirmed that the Dutch Medical Research Involving Human Subject Act (WMO) does not apply to this study and that Institutional Review Board review is not necessary – approval: 2023/453. Written informed consent was obtained from all participants.

Conflict of Interest Statement

M.A.J.H., E.W.H.K., and M.C.B. have no relevant conflicts to declare. J.M.M. reports fees from Argenx (Advisory Boards, Consultations), Sanofi (Consultation), Janssen-Cilag (Consultation). All contracts were reviewed by the Board of Directors of the UMCG, and all fees were paid to the institution. B.H. reports fees from Janssen-Cilag (Advisory Boards, Educational grants, Consultations, Investigator Initiated Studies), AbbVie (Advisory Boards, Educational grants, Consultations, Investigator Initiated Studies), Novartis Pharma (Advisory Boards, Consultations, Investigator Initiated Studies), UCB Pharma (Advisory Boards, Consultations), Leo Pharma (Consultations), Solenne B.V. (Investigator Initiative Studies), Celgene (Consultations, Investigator Initiated Studies), Akari therapeutics (Consultations, Investigator Initiated Studies), Philips (Consultation), Roche (Consultation), Regeneron (Consultation), and Sanofi (Consultation), Argenx (Advisory Boards, Consultations) and Boehringer-Ingelheim (Advisory Boards, Consultations). All contracts were reviewed by the Board of Directors of the UMCG, and all fees were paid to the institution.

Funding Sources

The authors did not receive support from any organization for the submitted work.

Author Contributions

Marjolein A.J. Hiel: conceptualization, data curation, formal analysis, investigation, methodology, project administration, visualization, writing – original draft, review, and editing. Eva W.H. Korte: conceptualization, data curation, formal analysis, investigation, methodology, visualization, writing – original draft, review, and editing. Maria C. Bolling AND Joost M. Meijer: conceptualization, data curation, methodology, visualization, supervision (supporting), writing – review and editing. Barbara Horváth: conceptualization, data curation, methodology, supervision (main), visualization, writing – original draft, review, and editing. All authors have provided approval for the submission of this manuscript and agree to be accountable for all aspects of the work.

Funding Statement

The authors did not receive support from any organization for the submitted work.

Data Availability Statement

The data supporting the findings of this study are not publicly available due to ethical restrictions but are available upon reasonable request to the corresponding author.