Oral Immunotherapy in Peanut‐Allergic Adults Using Real‐World Materials

ABSTRACT

Background

Peanut oral immunotherapy (OIT) has shown effectiveness in achieving desensitization of children; however, evidence in adults is lacking.

Methods

This phase II trial evaluated peanut OIT in peanut‐allergic adults using real‐world peanut products. A Simon's minimax two‐stage design, incorporating a stop:go for futility, was employed. A separate untreated control group was also recruited for comparison of mechanistic parameters. Participants underwent baseline double‐blind placebo‐control food challenges (DBPCFC) with peanut protein doses of 0.3 to 300 mg. Reacting participants were initiated on daily OIT with 2‐weekly updosing until reaching a maintenance dose of 1000 mg (four large peanuts). The primary outcome was the proportion of OIT participants who tolerated a cumulative dose of 1.4 g peanut protein during exit DBPCFC (doses provided 0.3‐3000 mg).

Results

Twenty‐one adults (8 female; mean age 24.2 years [SD 4.9]) were enrolled in the OIT group, with 67% achieving the daily maintenance dose and meeting the primary endpoint. Three withdrew due to adverse reactions, and a further three did not complete the trial for reasons unrelated to OIT. The median tolerated dose increased from 30 mg (equivalent to approximately 1/8th of a peanut) to 3000 mg (12 peanuts) at the exit challenge, representing a 100‐fold increase (p < 0.0001). OIT was associated with an improvement in QoL measures. Suppression of peanut skin prick test sizes and induction of peanut‐specific IgG were observed in OIT but not in control participants.

Conclusions

Peanut OIT appears to be an efficacious treatment for adults with peanut allergy. Further studies are needed for confirmation and to characterize safety profiles in different adult subgroups.

Trial Registration

Grown Up Peanut Immunotherapy (GUPI) study; ClinicalTrials.gov identifier: NCT03648320

In 21 adults initiated on peanut oral immunotherapy (OIT), 67% tolerated at least 1000 mg on exit DBPCFC. Decreases in SPT size and increases in peanut‐ and Ara h 2‐specific IgG occurred in OIT participants but not in mechanistic controls (n = 9). Improvements were seen in FAQLQ‐AF and food neophobia scores. Epinephrine was used infrequently; the majority (94.5%) of adverse reactions were mild (grade 1). AF, food allergy quality of life questionnaire—adult form; DBPCFC, double‐blind, placebo‐controlled food challenge; FAQLQ‐OIT, oral immunotherapy; QoL, quality of life; SPT, skin prick test.

1. Introduction

Peanut is a major cause of food allergy in the Western world with a prevalence of approximately 1.5% [1]. Seventy‐five to 80% of peanut allergy is thought to be lifelong [2, 3] and therefore most of the life of a peanut‐allergic individual is spent as an adult. Exposure to nuts in allergic adults may be associated with more severe reactions (odds ratio of 8.9 for reduced consciousness, 3.7 for pharyngeal edema and 2.0 for bronchospasm in adults vs. children) [4].

Constant vigilance and fear of reactions place a significant burden on both the peanut‐allergic individual and their families/caregivers. Consequently, peanut allergy has a significant effect on health‐related quality of life, particularly mental/psychosocial [5]. Many aspects of day‐to‐day life can be impacted, including travel choices and job options [6].

Until recently, the sole strategy for the management of peanut allergy was avoidance, provision of epinephrine, and education on managing allergic reactions. In 2020, the FDA approved Peanut Allergen Powder (AR101, Palforzia) for use in children, the first licensed oral immunotherapy (OIT) formulation for the treatment of peanut allergy [7]. OIT involves daily ingestion of a food (in this case peanut flour) in gradually increasing doses, with the aim of inducing allergen desensitization and immunomodulation [8].

In the multi‐centre PALISADE phase III trial of AR101 in 551 peanut allergic individuals, highly significant results were seen in participants aged 17 and under, with 67.2% of the active arm tolerating 600 mg of peanut protein when challenged post treatment versus only 4% of the placebo group [9]. Results did not reach statistical significance for the 10% of participants aged 18–55 years, and AR101 has not been approved for adult treatment.

Although several trials have shown the efficacy of peanut OIT in children [8], to date there are no published studies of peanut OIT in an exclusively adult cohort. The Phase II Grown‐Up Peanut Immunotherapy (GUPI) trial was designed to test the hypothesis that adults with peanut allergy can also be effectively and safely desensitized by OIT. On the basis that peanut allergy is extremely unlikely to remit in adulthood, a novel Simon's minimax two‐stage design was employed, a feature of which is an interim analysis for futility. An intention of the study was to use real‐world peanut products (e.g., peanut flour, whole peanuts) where possible. Objective measurement of peanut reactivity before and after OIT was performed by gold standard double‐blind placebo‐controlled food challenge (DBPCFC) using GMP‐grade materials. Quality of life was also assessed, and skin prick reactivity and IgG/IgE responses to peanut were evaluated in comparison to an untreated group of peanut‐allergic adults.

2. Methods

2.1. Trial Design and Oversight

GUPI was an investigator‐led phase II single‐arm single‐centre trial using a Simon's minimax two‐stage design. The protocol (Appendix 1) was approved by London Camden and King's Cross Research Ethics Committee (17/LO/1901). The trial steering committee consisted of experts in food allergy and patient representatives. An independent data monitoring and ethics committee reviewed data quality.

2.2. Inclusion and Exclusion Criteria

OIT participants were eligible to take part in the GUPI trial if they were aged 18 to 40 years with a clinical diagnosis of peanut allergy, a positive skin prick test wheal to peanut extract (≥ 3 mm compared to negative control), a positive specific IgE to peanut Ara h 2 (> 0.35kuA/L), and a positive DBPCFC to 3–300 mg peanut protein. In addition to the cohort treated with peanut OIT, a parallel group of peanut‐allergic adults was recruited purely to control for SPT and serological changes over the course of the study. The same eligibility criteria applied to control participants, with the exception of DBPCFC, since establishing a threshold of reactivity was considered unnecessary for the purposes of controlling for SPT and antibody changes over time. For OIT participants, concomitant asthma needed to be well controlled (ACQ < 1, FEV1 > 80% predicted, no hospital admission or oral corticosteroids for 2 years, maximum 800mcg beclomethasone dipropionate equivalent) with evidence of adherence to medication from prescription records. Further exclusion criteria included a history of life‐threatening anaphylaxis to peanut, pregnancy, treatment with immunosuppression, regular use of NSAIDs, a history of eosinophilic esophagitis, and any other condition deemed incompatible with the trial.

2.3. Trial Procedures

An overview of the trial design is shown in Figure S1, appendix 2.

2.4. Double‐Blind Placebo‐Controlled Food Challenges (DBPCFCs)

OIT participants underwent DBPCFC to peanut/placebo for confirmation of eligibility and at study exit. Challenges were administered using a validated commercially available kit (Reacta Healthcare, Manchester, UK) previously used in studies conducting DBPCFCs [10]. The active material was defatted peanut flour in a dessert food matrix and the placebo food matrix alone. Active and placebo challenges were administered on two separate days at least a week apart, with the order randomized by the study statistician. Baseline challenges were performed with 0.3 mg, 1 mg, 3 mg, 10 mg, 30 mg, 100 mg, and 300 mg doses of peanut protein, with additional doses of 1000 mg and 3000 mg on the exit challenge. Challenge outcomes were scored according to PRACTALL guidelines [11].

2.5. Oral Immunotherapy

On an initial dose escalation day, participants received 0.8 mg peanut protein mixed in food vehicle and then 1.5 mg 30 min later, followed by 3 mg a further 30 min later (Figure S1). Participants who tolerated 1.5 or 3 mg continued on this dose daily at home for a minimum of 2 weeks before returning for supervised dose escalation. Dose escalations comprised 1.5 to 3 mg and thereafter 6, 12, 25, 50, 100, 250, 500, and 1000 mg.

Participants were supplied with sachets of weighed defatted peanut flour to take at home for doses up to 100 mg peanut protein (200 mg flour). The peanut flour used in this study was assayed to confirm and quantify the content of major peanut allergens by Indoor Biotechnologies (see appendix 3). The average Ara h 2 content was 10.9 mg/g flour, in line with previous data [12].

On achieving a stable dose of 50‐100 mg peanut protein, participants were switched to eating peanuts, Bamba, or peanut butter according to preference, with the first dose being taken under the supervision of the trial team. Equivalent peanut dose conversions were as per the LEAP study [13, 14]. Once participants achieved a daily dose of 1 g (4 large peanuts), they remained on this dose for at least 4 weeks before undergoing exit DBPCFC. Participants then continued daily dosing for at least 3 months post DBPCFC (study exit) as well as the option of continuing post‐study.

All participants were provided with advice on optimal timing of daily doses to ensure adequate time separation from exercise, alcohol, and other co‐factors with the potential to increase the risk of adverse reactions [15].

In the event of an adverse reaction to OIT, dosing adjustment was performed according to reaction severity as categorized by the modified WAO grading system (see Protocol, Appendix 1).

2.6. Outcomes

The primary outcome was the proportion of OIT participants who tolerated a cumulative dose of 1.4 g peanut protein without reaction assessed by DBPCFC at least 1 month after achieving maintenance OIT dosing. Participants withdrawn due to AE or other reasons did not undergo DBPCFC and were considered not to have met the primary outcome. As pre‐specified in the trial protocol, the primary efficacy population consisted of all participants who received at least one dose of OIT but did not include those withdrawn for reasons entirely unrelated to OIT. The intention to treat population included all participants given a single dose of peanut OIT.

Secondary outcomes were: tolerance of cumulative dose of 4.4 g peanut protein on DBPCFC, incidence of local and systemic reactions during updosing and maintenance, change in skin prick test reactivity following OIT, change in peanut‐specific IgG titers, change in scores recorded by the Food Allergy Quality of Life Questionnaire—Adult Form (FAQLQ‐AF), food neophobia questionnaire, and modified food situations questionnaire.

Skin prick test titrations were conducted with undiluted peanut skin prick test extract (ALK Abello) and dilutions of 1:10, 1:100, 1:1000, and 1:10,000. Antibody measurement in the sera of OIT and control participants was analyzed in blinded fashion by ImmunoCAP (Thermofisher, Oxford, UK).

2.7. Statistical Analysis

The sample size was based on the consideration that a worthwhile response rate would be at least 50% with an inadequate response rate of 20% or lower. Enrolment of 25 participants provided 90% power to detect a response rate of 50% based on a dropout rate of 20%. The Simon's minimax two‐stage design incorporated an interim analysis for futility conducted after the first 12 participants were accrued [16]. At least two of the first 12 participants were required to meet the primary endpoint for the trial to continue.

Based on the design, the primary outcome was considered to have been met if at least seven of the participants achieved the primary endpoint. Suitable summary descriptive statistics and tests were used and chosen based on population outcome distributions. Changes in skin prick test reactivity were analyzed by two‐way ANOVA, serological changes were analyzed using the Wilcoxon signed‐rank test, and questionnaires were analyzed using one‐way ANOVA.

3. Results

3.1. Participant Enrolment and Baseline Characteristics

Twenty‐one participants ranging from 19 to 39 years (mean 24 years) (Table 1) were initiated on OIT prior to the onset of the COVID‐19 pandemic in March 2020, when routine research activity was subject to mandatory suspension. When research activity resumed, a decision was taken with trial steering committee approval that it was not feasible to enroll further participants within the remaining timeline of the study.

TABLE 1.

Baseline characteristics of study participants.

	Overall		Completed protocol
	OIT	Control	OIT	Control
	(n = 21)	(n = 12)	(n = 15)	(n = 9)
Age (y), mean (SD)	24.2 (4.9)	23.0 (20.0–26.3)	23.0 (21.0–25.0)	23.0 (20.0–25.5)
Female sex, no. (%)	8 (38%)	3 (25%)	6 (40%)	3 (33%)
Race, no. (%)
White	20 (95)	9 (75%)	14 (93%)	8 (89%)
Mixed	1 (5)	0	1 (7%)	0
Asian	0 (0)	2 (17%)	0 (0)	0 (0%)
Black	0 (0)	1 (8%)	0 (0)	1 (11%)
Other	0 (0)	0 (0)	0 (0)	0 (0)
Peanut skin prick test in mm, median (IQR)	11.6 (4.7)	11.0 (4.0)	10.5 (3.7)	9.0 (1.7)
Peanut ara h 2 (kUa/L), median (IQR)	16.2 (1.9–28.5)	5.1 (1.9–19.1)	6.2 (1.2–19.0)	3.6 (1.5–15.4)
Peanut ara h 8 positive (> 0.35 kUa/L), no (%)	5 (24%)	7 (58%)	4 (27%)	5 (55%)
Asthma diagnosis	9 (43%)	8 (75%)	8 (53%)	7 (78%)

An interim analysis was performed after the first 12 participants in accordance with Simon's minimax two‐stage design. Since at least two of these 12 met the primary endpoint, the study progressed to stage two. Six out of 21 (29%) were withdrawn from the study. In three cases, this was directly OIT‐related, but one case was due to an unrelated new serious illness (Figure 1). Two further participants, who had been initiated on OIT shortly before research suspension in March 2020, were also withdrawn since the majority of updosing visits were still required but could not be performed. Once clinical trial activity resumed, the full OIT protocol could not be repeated within the remaining study timeline. Eighteen participants were included in the primary efficacy population and 21 in the intention to treat analysis.

FIGURE 1.

Flow diagram of participant recruitment.

In the mechanistic control group, 12 were recruited, with 9 completing the protocol. Demographics of these 9 are included in Table 1 for comparison with the active participants who underwent DBPCFC at study exit. There were no significant differences between the two groups.

3.2. Primary End Point

Fourteen of 18 participants (78%) met the primary endpoint of tolerating at least 1.4 g peanut protein. On intention to treat, 14 of 21 (67%) met the primary endpoint. Since the pre‐specified threshold was a minimum of 7 from 25 participants, the primary outcome of the study was therefore met. The median highest tolerated dose at baseline DBPCFC was 30 versus 3000 mg at exit DBPCFC, a 100‐fold increase (p < 0.0001, Figure 2).

FIGURE 2.

Cumulative dose of peanut protein tolerated on DBPCFC at baseline versus after 1 month of peanut OIT maintenance (1000 mg). Analyzed by Wilcoxen signed rank.

3.3. Increased Tolerance Threshold

Ten of 18 (56%) participants in the primary efficacy population met the secondary endpoint of tolerating a cumulative dose of 4.4 g peanut protein, corresponding to 48% of the intention to treat population (10/21).

3.4. Tolerability and Safety

Epinephrine was administered to one participant (4.8%) in hospital during baseline DBPCFC, who subsequently completed the OIT protocol and achieved the primary endpoint. Epinephrine was self‐administered on 4 occasions by 3 participants during home dosing (see Table S1, appendix 2). Of these, one participant (#1) went on to complete the OIT protocol and achieve the primary endpoint. Participant #5 was withdrawn per protocol after 2 anaphylactic reactions with co‐factors despite a step down in OIT dose. After missing doses on 4 consecutive days, participant #2 experienced a grade 2 reaction after taking the next OIT dose. Due to concerns regarding adherence and the effect on safety, this participant was withdrawn by the chief investigator with TSC/DMEC approval. A third participant was withdrawn due to failure to updose to 100 mg on two occasions.

A summary of adverse reactions is shown in Tables S2–S4, appendix 2. Symptoms and rescue medications during DBPCFCs are shown in Tables S5–S9. Adverse reactions most commonly occurred during dosing increments in the hospital. The majority of adverse reactions were grade 1 (95.4% of adverse reactions). All 17 grade 2 reactions (4.3%) occurred during updosing. A single grade 4 reaction (0.3%) occurred while running after a 1000 mg maintenance dose (participant subsequently withdrawn after a further grade 2 reaction despite a step down in OIT dose).

3.5. Skin Prick Test Reactivity

Follow up skin prick test titration was performed in all 15 participants who completed the trial protocol and controls. OIT was associated with significant reductions seen in wheal size elicited by undiluted peanut skin prick test extract (median reduction 2.5 mm) and with 1:10 (3.5 mm reduction), 1:100 (4 mm reduction) and 1:1000 (3 mm reduction) dilutions (p < 0.001) (Figure 3). In the control group, a small significant reduction was seen at the 1:10 (1.8 mm reduction, p < 0.05) dilution over the same time period but not at other concentrations (Figure 3).

FIGURE 3.

Skin prick test titrations at baseline and after at least 1 month of maintenance (OIT participants) or after 6–9 months (controls). Analyzed by two‐way ANOVA.

3.6. Serological Assays

Participants undergoing OIT demonstrated post‐treatment increases in peanut‐ and Ara h 2‐specific IgG, but no changes were observed in specific IgE (Figure 4). In the control group, no changes were seen during a similar time frame (Figure 4).

FIGURE 4.

Serology at baseline and after at least 1 month of maintenance (OIT participants) or after 6–9 months (controls). Analyzed by Wilcoxen signed rank.

3.7. Influence of Ara h 8 Sensitization on Outcomes

Five out of 21 participants were IgE‐sensitized to Ara h 8. No differences were apparent in rates of local or systemic adverse reactions between Ara h 8 positive or Ara h 8 negative participants. 60% (3/5) of Ara h 8 sensitized participants experienced local symptoms at any point during OIT versus 67% (10/5) of non‐sensitized participants. Ara h 8 status did not affect the likelihood of meeting primary or secondary endpoints.

3.8. Quality of Life

Quality of life scores were available before and after treatment for all 15 participants who completed OIT (Figure 5). There was a highly significant improvement (reduction) in quality of life scores at the time of exit DBPCFC (mean difference − 1.7 out of total score of 7, p < 0.001) which was sustained after a further 3 months of dosing compared to pre‐OIT. Significant improvements in food neophobia scores (mean reduction of 1.4 out of total score of 7, p < 0.001) and food situations scores (increase of 1.5 out of total score of 5, p < 0.001) were also observed. These results were also highly significant for all three parameters in the primary efficacy population, with no improvement from baseline imputed for the 3 participants withdrawn for OIT‐related reasons.

FIGURE 5.

Quality of life and food neophobia questionnaires for OIT participants (n = 15). Analyzed by one‐way ANOVA. FAQLQ‐AF: Food allergy quality of life questionnaire‐adult form (higher scores indicate poorer quality of life). FNS: Food neophobia scale (higher scores indicate greater food neophobia). FSQ: Food situations questionnaire (higher scores indicate greater willingness to try novel foods).

4. Discussion

This study represents the first clinical trial of peanut OIT performed solely in adults. We demonstrated that OIT with 2‐weekly updosing followed by a target maintenance dose of 1000 mg is effective in achieving desensitization measured under double‐blind conditions. We incorporated real‐world products into OIT, such as whole peanuts or peanut‐containing products, as used in the LEAP study [13]. This was associated with improvements in quality of life and paralleled by reduced skin prick test reactivity and enhanced IgG responses to peanut allergen. The efficacy rate (78% of primary analysis population) was broadly in line with that reported in pediatric studies. Furthermore, the rate of adverse reactions observed was also similar to those studies.

The PALISADE trial of AR101 concluded that peanut OIT was ineffective in adults, with only 42% of active participants tolerating 600 mg on exit DBPCFC versus 14% of placebo. In children aged 4–17, results were 67% (active) versus 4% (placebo). Adults comprised only 10% of the trial population, although the protocol planned for 20% [9]. Withdrawal of adult participants assigned to OIT was high at 51% (21/41). Withdrawal in children was much lower (21%, 78/372). Therefore, PALISADE appeared to experience challenges with both recruitment and retention of adults. The long‐term peanut OIT study POISED also included 22 adults out of 120 participants [17]. Peanut OIT was effective at achieving desensitization in both adults and children, but the study was not powered to address the efficacy in the adult group nor to test any difference between the two age groups. They also noted a high dropout rate of 32% in adults versus 9% in children. The management of adults on OIT may require different expertise and approaches to pediatrics and adolescents. Our centre is dedicated to Adult Allergy, and all trial participants were counseled and followed by a specialist adult allergy dietitian throughout the study.

A systematic review and meta‐analysis of peanut OIT from 12 trials and 1042 participants (median age 8.7y) concluded that OIT is effective at achieving desensitization [8]. However, there was also a significantly increased frequency of anaphylaxis, epinephrine use, serious adverse events, and non‐anaphylactic reactions compared to placebo or avoidance. There was no improvement in quality of life, which the authors attributed to “undesirable consequences” of OIT. In this trial, most reactions to OIT were mild, the most common being oropharyngeal pruritus. However, cutaneous, oral, upper, and lower respiratory symptoms were also commonplace, particularly during dose increment visits. Respiratory symptoms were continuously assessed by an allergy‐experienced physician and clinically adjudged as mild and resolved rapidly with albuterol. Epinephrine was administered in the hospital only during a DBPCFC procedure in a single patient. Nevertheless, the relatively high rate of adverse reactions per dose during hospital updosing visits highlights the risk associated with dose increments and reinforces the need for these visits to take place under appropriate supervision. Our dose increments were doubling, and smaller dose increments may have resulted in fewer adverse events, which should be considered for future OIT protocols.

Outside of hospital, three patients self‐administered a single dose of epinephrine on four separate occasions. Of these, three occasions were associated with a known co‐factor (running) or prior missed OIT doses, emphasizing the importance of patient education and adherence. The fourth occasion occurred in a participant who previously experienced a grade 4 reaction (while running) and then underwent a 2‐step dose reduction per protocol. In our view, future trials of peanut OIT in adults should consider mandatory withdrawal of a participant experiencing a grade 4 reaction and consider withdrawal as part of a shared decision‐making discussion if repeated milder systemic reactions of any severity occur during home dosing. Moreover, a low threshold should be applied for withdrawal of participants who are non‐adherent with dosing protocols and instructions relating to the avoidance of co‐factors soon after dosing. Applying consensus recommendations to prepare and consent participants for OIT is also likely to mitigate associated risks [18].

Unlike the systematic review [8], we saw highly significant improvements in quality of life scores and food neophobia. This provides evidence that the benefits associated with peanut OIT in adults relate to reduced fear around food. Informal feedback from participants included that the treatment had been “life‐changing” with increased ability to travel with less fear of reactions. In contrast to many pediatric trials of peanut OIT, GUPI was an open study, and we consider it likely that participants' knowledge of receiving active treatment was a key element leading to QoL and food neophobia score improvements.

Reductions in skin prick test wheals provided further objective evidence of peanut desensitization. A small reduction seen at the 1:10 dilution in the control group was anomalous, not occurring at other concentrations, unlike in the OIT group. Furthermore, peanut‐ and Ara h 2‐specific IgG measured under blinded conditions both clearly increased in the OIT group but not the control peanut‐allergic group. These serological changes are consistent with those reported in pediatric trials [9, 19].

Our study does have limitations. We chose a Simon's minimax design [16] without a placebo group for several reasons: firstly, spontaneous resolution of confirmed peanut allergy in adulthood is considered rare and the expected response rate without intervention during the duration of the study was estimated to be close to or 0%. Nevertheless, to minimize bias, the primary endpoint of post‐OIT peanut reactivity was measured under double‐blind placebo‐controlled conditions using a GMP‐grade oral challenge kit that comprised validated reagents. The demographics of GUPI participants showed a white predominance (Table 1). This was also seen in other studies such as PALISADE. Our centre is based in a large multicultural city and we invited peanut‐allergic individuals from a diverse range of backgrounds to participate. The reasons behind the lack of diversity in OIT trial participation, and whether this extends to the clinical practice of OIT in a non‐trial setting, are important issues that require further evaluation. Furthermore, GUPI did not address the longer effects of OIT, such as whether desensitization can be maintained without daily peanut consumption following several years of continuous OIT.

In conclusion, we have found that peanut OIT can be an efficacious treatment for desensitization of adult peanut allergic patients. Larger studies will be required to further characterize the safety profile and identify the group of adult patients most likely to benefit with the minimum risk of systemic adverse reactions, and also if OIT has the potential to lead to long‐term tolerance in this age group. These findings could also be expanded to other foods such as cow's milk, egg, and wheat, which are usually persistent when seen in adults.

Author Contributions

Research design: Stephen J. Till, Hannah Hunter, Kok Loong Ue, Victoria Cornelius, Carina Venter. Recruitment: Kok Loong Ue and Hannah Hunter. Data collection: Kok Loong Ue, Ching Ching Yung, Iason Thomas, Olympia Tsilochristou. Laboratory analysis: Janice Layhadi, Mohamed H. Shamji. Data analysis: Hannah Hunter, Kok Loong Ue, Victoria Cornelius, Stephen J. Till. Interpretation and writing: Hannah Hunter, Kok Loong Ue, Victoria Cornelius, Ching Ching Yung, Iason Thomas, Olympia Tsilochristou, Janice Layhadi, Leonard Q.C. Siew, Carina Venter, Mohamed H. Shamji, Stephen J. Till.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Appendix 1.

Appendix 2.

Appendix 3.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.