Safety and efficacy of droxidopa for dysautonomia in adults with Menkes disease and occipital horn syndrome in the USA: a randomised phase 1/2a crossover trial

Summary

Background

Menkes disease is an X-linked recessive disorder of human copper metabolism. Droxidopa is a synthetic amino acid effective in reversing neurogenic orthostatic hypotension and correcting neurochemical abnormalities in congenital absence of dopamine-beta-hydroxylase (DBH), a copper-dependent enzyme that influences autonomic function. Individuals with disorders associated with variants in the copper transport gene ATP7A may manifest symptoms of dysautonomia, due to deficient DBH activity. We aimed to evaluate the safety and efficacy of droxidopa for dysautonomia in adults with Menkes disease or with occipital horn syndrome (an ATP7A allelic variant).

Methods

We conducted a phase 1/2a, randomised, double-blind, placebo-controlled, crossover trial at one academic medical centre in Columbus, OH, USA. We compared placebo versus droxidopa treatment in adults with Menkes disease or occipital horn syndrome who manifested symptoms of dysautonomia (including orthostatic hypotension). Participants were recruited by invitation, screened, and randomly assigned to receive droxidopa or placebo for 6 weeks (Arm 1). Following a 7–10 day washout period, participants received the opposite treatment for 6 weeks (Arm 2/Crossover treatment). An open-label dose titration was utilised in advance to determine each participant's maximally tolerated dose (100, 200, or 300 mg) of droxidopa. The primary outcome of this trial was safety and tolerability assessed at 6 weeks, as reflected in the type and incidence of adverse events in the droxidopa treatment versus placebo groups. This trial is registered with ClinicalTrials.gov, NCT04977388.

Findings

Between July 12, 2021 and Oct 30, 2023, three male participants were enrolled: two individuals with Menkes disease (19 and 26 years old) and one individual with occipital horn syndrome (age 35). We found significant improvements in norepinephrine levels (P < 0.01) and in a critical parameter of orthostatic hypotension, diastolic blood pressure drop during tilt table testing, while receiving droxidopa (−8.6, 95% CI −15.5 to −1.7; P = 0.018). There was no substantial difference in adverse events between the droxidopa and placebo groups.

Interpretation

In this early phase trial, droxidopa was well tolerated in adults with Menkes disease and occipital horn syndrome and was associated with correction of orthostatic hypotension. These preliminary findings suggest that droxidopa at doses adjusted for patient tolerance is likely to be efficacious for treatment of dysautonomia in adults with ATP7A-related disorders. Further research is required, including in younger individuals with these conditions.

Funding

The Menkes Disease Foundation UK, Associazione Angeli Per La Vita, and the Abigail Wexner Research Institute at Nationwide Children's Hospital.

Research in context.

Evidence before this study

Evidence before undertaking this study was identified through PubMed searches for relevant work, published in English, using terms such as “ATP7A”, “Menkes disease”, “occipital horn syndrome”, “dopamine-beta-hydroxylase deficiency”, “autonomic failure”, “droxidopa”, and “L-DOPS” covering the period 1999 to 2023. Our search yielded 77 studies. Articles were also identified for review from the author's collection of scientific literature, and papers presented at the Kaler laboratory journal club. Some prior studies of autonomic disorders have included neuroimaging, which was not a component of this trial.

Added value of this study

Menkes disease is an ultra-rare neurogenetic disorder with lengthening life expectancy related to recent improvements in early diagnosis and treatment. No previous study has addressed the debilitating dysautonomia that can complicate the day-to-day lives of Menkes survivors, individuals who escaped the premature mortality typically associated with this illness, through early copper histidinate treatment. Classic signs and symptoms of autonomic insufficiency are among the lingering unaddressed medical issues in such individuals, as well as in participants with less severe variants in ATP7A, including occipital horn syndrome. Their dysautonomia is caused by deficiency of dopamine-beta-hydroxylase, a copper-dependent enzyme localised to a neurochemical pathway that droxidopa can bypass. This clinical trial evaluated the safety, tolerability, dosing, and efficacy of droxidopa in this patient population.

Implications of all the available evidence

Droxidopa appears safe and effective in restoring autonomic function in participants with deficient activity of dopamine-beta-hydroxylase from varied causes, through restoration of circulating norepinephrine levels. Individuals with variants in the Menkes disease copper transporter gene, ATP7A, are at risk for autonomic dysfunction potentially responsive to droxidopa treatment. The present findings suggest that droxidopa at doses adjusted for patient tolerance is efficacious for treatment of dysautonomia in adults with ATP7A-related disorders and may be appropriate for younger individuals with these conditions. If feasible, neuroimaging in future studies of ATP7A-related dysautonomia could add value.

Introduction

Menkes disease is an X-linked recessive disorder of human copper metabolism with a predicted minimum birth prevalence of 1 in 34,810 live male births based on loss-of-function variant frequencies in the Genome Aggregation Database (gnomAD) for ATP7A, which encodes an essential copper-transporting ATPase.1, 2, 3, 4 Abnormal plasma neurochemical levels due to deficiency of a copper-requiring enzyme, dopamine-β-hydroxylase, are diagnostic in affected Menkes newborns.³ A brief window of therapeutic opportunity exists in the newborn period during which medical intervention can prevent the inexorable downhill course otherwise expected for this illness, as demonstrated with Copper Histidinate (CuHis) treatment.^4,5 Another emerging treatment approach, adeno-associated virus-mediated ATP7A gene therapy, also appears highly promising, demonstrating a synergistic effect with CuHis in a Menkes disease mouse model.6, 7, 8

Survivors of the profound central nervous system (CNS) effects of Menkes disease, spared by early diagnosis and successful responses to early subcutaneous CuHis treatment (daily injections for the first three years of life), often develop symptoms of dysautonomia beginning in late childhood. This reflects a persistent defect in Cu delivery into the secretory pathway of cells, including sympathetic (dopaminergic) neurons that regulate the autonomic nervous system, resulting in diminished DBH activity and deficiency of norepinephrine (Figure S1A).

The synthetic amino acid droxidopa (L-threo-3,4-dihydroxyphenylserine or L-DOPS, Northera^TM) is effective in reversing orthostatic hypotension and correcting neurochemical abnormalities in patients with autosomal recessive congenital absence of (DBH).⁹ Since droxidopa (half-life approximately 2.5 h) is metabolised by the enzyme dihydroxyphenylalanine (DOPA) decarboxylase, which does not require copper to produce norepinephrine, the DBH enzymatic defect is bypassed (Figure S1B).

Individuals with Menkes disease and occipital horn syndrome (OHS) manifest partial deficiencies of DBH and show distinctive neurochemical abnormalities and clinical symptoms of dysautonomia, such as syncope, dizziness, orthostatic hypotension, abnormal sinoatrial conduction, nocturnal bradycardia, and bowel or bladder dysfunction.^2,10,11 As in patients with congenital absence of DBH, these problems in autonomic nervous system function seemed potentially responsive to restoration of normal neurochemical levels and sympathetic nerve function by treatment with droxidopa.¹² We evaluated this hypothesis in a double-blind placebo controlled crossover clinical trial of adult participants with ATP7A variants who suffered symptoms of dysautonomia.

Methods

Ethics

The study was approved by the US Food and Drug Administration (IND #151326, SG Kaler, sponsor-investigator) and the Nationwide Children's Hospital Institutional Review Board (IRB) and overseen by a Data Safety Monitoring Board (DSMB) comprised of biostatistical and relevant topical expertise. External monitoring for GCP compliance was performed by IQVIA (Durham, NC). Written informed consent was obtained from all participants.

Study design

This investigator-initiated Phase 1/2a clinical trial with a double-blind placebo-controlled randomised crossover design was conducted at Nationwide Children's Hospital (NCH) [Columbus, OH USA] between July 12, 2021 and Oct 30, 2023. The study was registered on ClinicalTrials.gov (NCT04977388). The clinical trial protocol is provided in Supplementary Materials. The clinical trial profile is shown in Fig. 1 and detailed study design in Fig. 2A.

Fig. 1.

Trial profile.

Fig. 2.

Clinical trial design and droxidopa dose titration strategy. (A) Clinical trial design. (B) Algorithm for determining droxidopa dose during open-label dose titration period. Blood pressure greater than 140 mm Hg systolic or 90 mm Hg diastolic, or headache longer than 2 h was grounds to lower or discontinue droxidopa dose. If hypertension persisted despite discontinuation of droxidopa, conventional medical treatment would be instituted or arranged by the research team in a timely fashion. Signs or symptoms of allergic-type reactions or malignant hyperthermia in response to droxidopa were grounds for withdrawing participants from the study.

Each participant underwent a screening evaluation and, if eligible, was blindly randomised to droxidopa or placebo for the first arm of the study. An open label ascending dose titration strategy (Fig. 2B) was used to identify a safe, tolerable, and potentially effective dose of droxidopa for the individual participants. After a 7–10 day washout period, either droxidopa or placebo was initiated as per prior randomisation and continued for 6 weeks, followed by crossover to the alternative treatment for 6 weeks after another 7–10 day washout period (Fig. 2A).

Participants

Eligible participants were at least 18 years of age with a documented pathogenic variant in ATP7A and either classic severe Menkes disease or Occipital Horn Syndrome, a milder allelic variant (Table 1).^11,13 The Menkes disease participants had survived beyond the expected natural history after receiving early CuHis treatment for three years and had attained normal or near-normal neurodevelopment including independent ambulation.^3,5

Table 1.

Demographic data and baseline physical findings.

Subject ID	Droxi-01	Droxi-02	Droxi-03
Age (years)	25	19	35
Diagnosis	Menkes	Menkes	OHS
ATP7A variant	R201X	G666R	A1362D
Enrolled	7/12/21	8/16/21	7/17/23
End of trial	25-Oct-21	29-Nov-21	30-Nov-23
Weight (kg)	68	59	46.2
Height (cm)	166.2	176.7	173.0
Supine SBP (mm Hg)	103	81	98
Supine DBP (mm Hg)	61	52	61
Supine HR (beats/min)	74	50	70
Upright SBP (Tilt)	68	48	63
Upright DBP (Tilt)	44	33	46
Upright HR (Tilt)	103	80	94
SBP Dropa	35	33	35
DBP Dropa	17	19	15
HR Increasea	29	30	24

SBP = systolic blood pressure. DBP = diastolic blood pressure. HR = heart rate.

^aThese values indicate profound orthostatic hypotension in these subjects, defined as a drop in SBP ≥20 mm Hg or in DBP ≥10 mm Hg, and is typically also associated with higher pulse (HR).

Participants were recruited by invitation based on the Principal Investigator (PI)'s awareness of their medical condition. They all manifested clinical signs and symptoms of dysautonomia, e.g., orthostatic hypotension: specifically, a decrease in systolic or diastolic blood pressure of at least 20 or 10 mm Hg, respectively, within 3 min after standing (Table 1), and/or chronic diarrhoea: production of loose stools with or without increased stool frequency, for more than four weeks immediately preceding enrolment. Additionally, they reported a history of at least thrice weekly occurrence of dizziness/feeling lightheaded while standing upright and/or thrice weekly episodes of diarrhoea or an urgent need to defecate after food ingestion for more than four weeks immediately preceding enrolment. Exclusion criteria included pre-existing liver or kidney disease, hypertension, anti-hypertensive therapy, heart failure, cardiac arrhythmia or bleeding diatheses or concomitant medications including alpha-1 adrenoreceptor agonists, beta-blockers, DOPA decarboxylase inhibitors, midodrine, ephedrine, and triptan medications.

Randomisation and masking

Randomisation occurred at the conclusion of the screening visit (Visit 1). The NCH Investigational Drug Service (IDS) Pharmacy established the schedule as to which of the two treatment arms (first or second) would include droxidopa or placebo. To ensure a balanced allocation of participants to each randomisation sequence, the study employed a permuted block randomisation scheme with block sizes randomly alternating between 1 and 2. Block length was unknown to the clinic personnel. The permuted randomisation scheme was designed to effectively conceal the allocation and reduce the chance that testing personnel would be able to discern the next intervention group assignment.

The study involved an initial inpatient open-label dose titration period of 4 days (Visit 2) to identify the ideal droxidopa dose for each adult participant, followed by assignment to droxidopa or placebo. Droxidopa was provided as a tasteless white crystalline powder in capsules of 100 mg, 200 mg, or 300 mg inside unlabeled gelatin color capsules (sky blue and white, size 0), which accommodated the size 1, 2, and 3 droxidopa capsules plus a variable quantity of microcrystalline cellulose to complete the fill. The placebo control contained approximately 500 mg microcrystalline cellulose in unlabeled gelatin color capsules (sky blue and white, size 0). The ProFill Capsule Filling System (Torpac; Fairfield, NJ) was utilised by the NCH IDS Pharmacy to generate the treatment capsules. Placebo capsules were indistinguishable in weight, taste, texture, consistency, and visible characteristics from droxidopa capsules.

This was a double-blind crossover study, in which both participants and the clinical staff conducting the study were blinded to the treatment administered. Only the NCH Investigational Drug Service (IDS) Pharmacy staff were unblinded as to the contents of the treatment dispensed. The IDS Pharmacy maintained the codes in sealed envelopes for each participant. At the conclusion of the study, identification of droxidopa and placebo arms for each participant was revealed to the PI, biostatistician, protocol clinical staff, the participants, and their families. A pre-specified provision to reveal the code sooner if requested by the Data Safety Monitoring Board or for an emergency indication was not needed. The PI designated a 24hr/7 d unblinding contact person who had access to the sealed unblinding codes in case of such an emergency.

Procedures

At visit 1, participants were consented, screened and, if determined to be eligible based on the study's inclusion and exclusion criteria, were randomly assigned (blindly) to either droxidopa or placebo by the NCH IDS. After informed consent, the participant's medical history and concomitant medications were reviewed, vital signs (including orthostatics) and physical exam performed, and the Orthostatic Hypotension System Assessment (OHSA Questionnaire [appendix 1]), FDA Clinical outcomes Assessment (based on the Patient Global Impression of Severity [PGI-S] (Appendix 2)), and Irritable Bowel Syndrome-Diarrhea (IBS-D [appendix 3]) report forms and a walk-through of physical therapy exercise tolerance tests completed.

Visit 2 consisted of a hospital inpatient stay to assess droxidopa tolerance and establish the participant's dose for the clinical trial. This consisted of an open label dose titration to determine a maximally tolerated dose of droxidopa (Fig. 2) and included four days of testing, followed by assignment to droxidopa versus placebo based on prior randomisation. Testing for the first day included history and physical exam, electrocardiogram, echocardiogram, placement of intravenous saline lock (for safety and blood draws), baseline laboratory work: CBC with differential/platelets, comprehensive metabolic panel, serum copper and ceruloplasmin, urinalysis, and plasma catecholamines (baseline sample), review of concomitant medications and/or adverse events and baseline physical therapy exercise tolerance tests: standing time, 6-Minute Walk Test, Timed Up and Go.

During the remaining inpatient days (days two to four), the open label dose titration plan was initiated to determine each participant's maximally tolerated dose. Ascending doses of droxidopa of 100 mg, 200 mg, 300 mg (maximum) were administered as tolerated based on the algorithm for determining droxidopa dose (Fig. 2B). The droxidopa dose ascension was dependent on blood pressure less than 140 mm Hg systolic and 90 mm Hg diastolic and absence of symptoms of headache and/or nausea longer than 2 h. Tilt table testing (WR Medical S1 model) with continuous blood pressure and heart rate monitoring (DINAMAP PRO 100) was performed to quantify orthostatic hypotension at baseline and after each droxidopa dose. The tilt table testing involved gently strapping a patient onto the tilt table with feet resting on footplates. Continuous heart rate and blood pressure monitors were then attached to the participant. The entire table was tilted to move from a 0° to a 60° angle in approximately 45 s, allowing determination of changes in pulse (P) and blood pressure (BP) based on supine (0°) and upright/tilted (60°) positions.

On the mornings of days 2 and 3 at protocol visit 2, each participant underwent tilt table testing at baseline and 75–90 min after droxidopa doses (100 and 200 mg respectively as noted in Fig. 2), orthostatic blood pressure/pulse measurements, and plasma catecholamine levels, drawn from saline lock. Participants were supine (0°) for 5 min and in head tilt (60°) position for 3 min. Participants then rested for 10 min off the tilt table, and 2 h post dose plasma catecholamine levels were drawn, followed by performance of the physical therapy exercise tolerance tests with subsequent 4- and 6-h post dose plasma catecholamine levels.

If dose #1 (100 mg) was well-tolerated, the participant was given dose #2 (200 mg) droxidopa in the afternoon, with subsequent orthostatic blood pressures performed hourly for the first 4 h post dose and plasma catecholamine levels drawn 2 h post dose. Overnight supine BP/P was measured every 3 h, and a 10-h post dose catecholamine level was also drawn.

If the participants tolerated the prior evening 200 mg droxidopa dose, on the morning of day 3 they underwent the same testing as performed on day 2 after taking dose #3 (200 mg) droxidopa. If that 200 mg dose was tolerated, they received dose #4 (300 mg) droxidopa that afternoon and were monitored overnight, as during the previous evening/night.

If dose #4 was tolerated, the participant received dose #5 (300 mg) droxidopa on the morning of day 4 followed by orthostatic BP/P measurements and plasma catecholamine levels. Participants were discharged with a 1-month supply of droxidopa or placebo. After completing a 7–10 day washout period to eliminate any carryover effects of droxidopa when transitioning to the clinical trial phase, participants were instructed to take the study medication twice daily (early morning, late afternoon) at doses determined during the open label dose titration phase. The droxidopa dose for each participant was their maximum tolerated dose (100, 200 or 300 mg). Droxidopa was taken without meals during the open-label titration phase and participants were instructed to be consistent as to taking the study medication at home, either with or without meals.

Visits 3 to 6 represented outpatient follow-up evaluations (two per treatment arm, Fig. 2A) which included tilt table testing, safety laboratory tests, serial blood collections for plasma neurochemicals, physical therapy exercise tolerance tests, and review of concomitant medications and adverse events.

The study also included daily and weekly home monitoring between Visits 3 to 6. During Visit 2, participants and families were taught and practiced taking and recording their blood pressure and pulse utilising the Omron 3 Series Home BP monitor. At home they recorded their blood pressure and pulse (supine, after 3 min sitting, and after 3 min standing) within 2 h of taking the drug/placebo twice a day (before and after school or work) via the home monitoring device. In addition, they completed daily drug administration and headache/other side effects diaries, and IBS-D reports. The Orthostatic Hypotension Symptom Assessment (OHSA Questionnaire) and FDA Clinical Outcomes Assessment (based on the Patient Global Impression of Severity [PGI-S]) were completed at home weekly.

For monitoring and assessments, the participants received weekly telephone follow-up by a member of the study team which included review of daily and weekly monitoring activities, as well as review of concomitant medication and adverse events.

Outcomes

The primary outcome of this trial was safety and tolerability, as reflected in the type and incidence of adverse events in the 6-week droxidopa treatment versus 6-week placebo arm, using the CTCAE (v5) scoring tool. Tolerability was assessed based on the parameters shown in Fig. 2B. Briefly, these included prolonged headache (>2 h), nausea, hypertensive crisis (systolic BP > 180 mm Hg/diastolic >110 mm Hg), allergic-type reactions (e.g., skin hives, bronchospasm), and malignant hypothermia (high fever, muscle rigidity, mental status changes).

Secondary outcomes were also evaluated for the 6-week droxidopa treatment versus the 6-week placebo arm, including (1) change in level of plasma norepinephrine and dihydroxyphenylglycol (DHPG) after droxidopa compared to baseline and placebo; (2) change in standing systolic blood pressure (mm Hg), and the change between supine and head-up tilt table positions after droxidopa compared to baseline and placebo; (3) improved gastrointestinal symptoms (via daily IBS-D reports detailing bowel movement number and Bristol stool consistency); and (4) improved performance on physical exertion tests (standing time, 6-Minute Walk test, and the timed up and go test).

Change in scores on the Orthostatic Hypotension Symptom Assessment (OHSA) questionnaire after droxidopa, compared to baseline and placebo, was an exploratory outcome. Scores range from zero to 10 with 0 meaning no symptoms and 10 meaning the worst possible symptoms. All outcomes are prespecified unless indicated otherwise.

Statistical analysis

We initially aimed for a sample size of 6–10 participants; however, recruitment of participants both eligible and willing to participate proved difficult. When this became evident, the DSMB was consulted and agreed to unblinding after three participants with data clean-up and data lock in advance of unblinding.

We used GraphPad Prism and R version 4.5.0 software to analyse differences in outcomes between treatment with droxidopa and placebo. Paired measurements were analysed using paired t-tests or the non-parametric Wilcoxon signed-rank test. Linear mixed models (LMEs) including a participant-level random effect assessed differences between treatment groups in blood pressure and heart rate, accounting for the multiple measurements on each participant and using a Satterthwaite approximation to the degrees of freedom. Residual plots and graphical displays (boxplots, density plots) assessed modelling assumptions including homoscedasticity and normality. LMEs also evaluated differences in individual OHSA items and OHSI and PGI-S mean scores between treatment groups, while differences on individual PGI-S items were evaluated using the negative binomial generalised linear mixed model (NB-GLMM) to account for the integer valued outcome. Differences in number of spontaneous BMs and stool scores between study groups were also evaluated using the NB-GLMM. Statistical significance was defined as P ≤ 0.05. Safety assessments for participants treated with droxidopa and placebo included recording of treatment-emergent adverse events, which were compared as a percent of total.

Post-hoc analyses included diastolic blood pressures, the ratios of dihydroxyphenylacetic acid (DOPAC) to DHPG, and PGI-S results. Specifically, we evaluated participants' change in standing diastolic blood pressure (mm Hg), and the change in diastolic blood pressure between supine and head-up tilt table positions after droxidopa compared to baseline and placebo. We also examined the change in DOPAC:DHPG after droxidopa compared to baseline and placebo, as well as participants’ weekly responses to the Patient Global Impression of Severity (PGI-S) scale. The PGI-S seeks to capture the severity and degree of improvement of specific symptoms since starting the study drug/placebo and is based on the FDA Patient-Focused Drug Development Guidance Public Workshop: “Methods to Identify What is Important to Patients & Select, Develop or Modify Fit-for-Purpose Clinical Outcomes Assessments” (October 15–16, 2018). It uses a five-letter/point scale ranging from 1 (a) (none/much improved), to 5 (e) (very severe/much worse). Lower scores indicate less severe symptoms or greater improvement. No sensitivity analyses were performed.

Role of the funding source

The funders were not involved in study design, or data collection, analyses, and interpretation or in writing the report.

Results

Dose titration

Based on the dose titration phase of the trial, Participant Droxi-01 reached the 200 mg dose but was ultimately assigned the 100 mg dose due to intolerable side effect (vomiting) noted after the second 200 mg dose. Participant Droxi-02 tolerated 300 mg during dose titration phase and was assigned to the 300 mg dose; at Visit 5, however, mid-way through Arm 2/Crossover treatment of the study, this patient showed higher than normal blood pressures and his drug/placebo dose was lowered to 200 mg for the second half of Arm 2/Crossover treatment. Participant Droxi-03 reached 300 mg in the dose titration phase but was noted with supine hypertension (140–164 mmHg) in the post-300 mg dose follow-up inpatient checks (2–4 h s post-dose). Consequently, he was assigned 200 mg as a safe, tolerable, and potentially effective dose of droxidopa, in accordance with the dose titration decision tree (Fig. 2B). The drug/placebo in this participant was stopped in Arm 2/Crossover treatment after 9 doses due to headache and transient supine hypertension noted at home, held for 3 days (6 doses) and resumed at 100 mg. These dose titration results for all three participants are summarised in Table S1.

In the open-label dose titration phase, participants showed statistically significant differences in systolic BP, diastolic BP, and heart rate at baseline before droxidopa administration (Fig. 3A). These orthostatic differences were reversed by droxidopa in a dose-dependent fashion (Fig. 3B and C). Comparisons of baseline versus 200 mg droxidopa by position indicated statistically significant differences for tilt position systolic BP (difference = 56.2, 95% confidence interval (CI) (29.1, 83.2), P < 0.001) and tilt position diastolic BP (difference = 27.2, 95% CI (16.3, 38.1), P < 0.001) (Fig. 3D).

Fig. 3.

Orthostatic hypotension in Menkes and OHS adults (n = 3) is corrected by droxidopa. Tilt table tests were performed 75–90 min after assigned droxidopa dose. Tilt table test: 5 min supine; 3 min tilted at 60°; blood pressure and heart rate were measured at the end of each phase. (A) Baseline. (B) After 100 mg Droxidopa. (C) After 200 mg Droxidopa. (D) Comparisons of Baseline and 200 mg Droxidopa by position (supine and tilt). SBP = systolic BP; DBP = Diastolic BP; HR = heart rate. Bars show mean values ± SD. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗∗P < 0.0001; ns, nonsignificant.

Clinical laboratory testing results at baseline revealed the appearance of droxidopa (L-DOPS) in plasma after oral administration (Figure S2A), a rise in plasma norepinephrine levels following droxidopa administration (Figure S2B) and lowering of the DOPAC:DHPG ratio in the blood after droxidopa (Figure S2C).

Primary outcomes

Unblinding of the participants' codes to identify the droxidopa treatment and placebo arms was performed after agreement by the Data Safety Monitoring Board overseeing this double blind, placebo-controlled crossover trial. In terms of the primary outcome measure, safety and tolerability, there was no substantial difference in adverse events between droxidopa and placebo although the percent of adverse events was higher for droxidopa 62.5% versus 37.5% (Table 2). Most of the adverse events occurred while on droxidopa, including the drug's most commonly reported side effects (headache, nausea, hypertension). All three participants experienced episodes of elevated blood pressure as described in the Results Section. Participant Droxi-01 also had an episode of vomiting while on the tilt table, and mild nausea on a car ride home from study visit. Additionally, asymptomatic bacteriuria was found on a protocol-associated urinalysis attributed to his history of bladder diverticula. Participant Droxi-02 experienced two transient episodes of mild headache and episode of vomiting while on droxidopa. As previously described Droxi-03 experienced an episode of headache while on droxidopa.

Table 2.

Adverse events.

#	Subject ID	AE code	Major/Minor	Description of event	Treatment
1	Droxi-01	2	Minor	Vomited during Tilt Table test	D
2	Droxi-01	2	Minor	Sustained hypertension (systolic BP: 171, 148, 143 mm Hg)	D
3	Droxi-01	1	Minor	Brief episode of nausea on drive home	D
4	Droxi-01	1	Minor	Potential Urinary Tract Infection	D
5	Droxi-02	2	Major	Pre-syncopal episode on baseline tilt table test	P
6	Droxi-02	1	Minor	Proteinuria (30 mg/dL)	P
7	Droxi-02	1	Minor	Slightly elevated serum creatinine (1.43 mg/dL) (Normal range: 0.5–1.2)	P
8	Droxi-02	1	Minor	Moderate occult blood noted on urinalysis	P
9	Droxi-02	2	Major	Seizure while walking to bathroom	P
10	Droxi-02	1	Minor	Vomiting and headache post-seizure	P
11	Droxi-02	1	Minor	Headache	D
12	Droxi-02	1	Minor	Elevated BP	D
13	Droxi-02	1	Minor	Headache (less intense)	D
14	Droxi-02	1	Minor	Vomiting	D
15	Droxi-03	2	Minor	Headache	D
16	Drox-03	2	Minor	Elevated BP	D

CTCAE Codes: 1 = Mild; 2 = Moderate; 3 = Severe; 4 = Life-threatening; disabling; 5 = Death. D = Droxidopa; P = Placebo/no treatment.

Secondary outcomes

During the 12-week crossover clinical trial, mean systolic blood pressure was higher in all positions tested while on droxidopa, with the increase statistically significant for supine systolic BP (P < 0.05, Fig. 4A). Droxidopa treatment was also associated with lower mean drop in systolic blood pressure on tilt table testing but the difference was not statistically significant (Fig. 4B, left panel). The change in heart rate with tilt between drug and placebo was also not significantly different (Fig. 4B, right panel).

Fig. 4.

Blood pressure and orthostatic responses to Droxidopa in adult participants (n = 3) with ATP7A-related disorders. (A) Mean systolic and diastolic blood pressures increased in both supine and tilted positions with droxidopa. (B) Droxidopa treatment improved orthostatic hypotension. The change in systolic BP after 60° tilt up from supine position was less when participants received droxidopa, but not statistically significant (left panel). In contrast, the change in diastolic BP was statistically significant (middle panel). Change in heart rate after 60° tilt was not significantly different (right panel). Error bars = SEM. ∗P < 0.05; ns, nonsignificant.

Analysis of the unblinded cohort data also identified notable effects of droxidopa on increasing plasma levels of norepinephrine and DHPG in all positions (Fig. 5A and B).

Fig. 5.

Droxidopa raises plasma norepinephrine and DHPG levels in participants with ATP7A variants and symptoms of dysautonomia. (A) Increased plasma norepinephrine (NE) in response to droxidopa. (B) Increased plasma dihydroxyphenylglycol (DHPG) in response to droxidopa. (C) Droxidopa treatment is associated with normalisation (lowering) of the ratios of proximal: distal metabolites in catecholamine biosynthetic pathway (see Figure S1B). These graphs reflect unblinded data for all three participants collected during tilt table testing at follow-up visits three weeks apart during each six-week treatment arm (Droxidopa or Placebo). Error bars = SEM. ∗P < 0.05; ∗∗P < 0.01; ns, nonsignificant.

To assess changes in gastrointestinal symptoms, we evaluated IBS-D reports daily during the study to evaluate bowel movement number and Bristol stool consistency during the droxidopa and placebo treatment arms (Figure S3C). Droxidopa treatment was associated with a slight reduction in spontaneous bowel movements, but the result was not significant (mean ratio = 0.94, 95% CI (0.83, 1.06), P = 0.3). There was no significant difference in stool consistency (P = 0.5) reported by the participants (Figure S3).

Regarding tests of physical exertion, assessed via standing time, the Timed Up and Go, and 6-Minute Walk test, the test results showed variable improvements in the participants while receiving droxidopa compared to the placebo control (Figure S4).

Exploratory outcome

We analysed the effects of droxidopa on participants’ symptoms of dysautonomia by comparison to placebo (Figure S3). Figure S3A summarises OHSA questionnaire data from the three participants based on treatment arm (droxidopa versus placebo). We found statistically significant improvements (reductions) in symptoms of dizziness (Q1: Mean difference = −2.1, 95% CI (−3.6, −0.9), P = 0.001), weakness (Q3: Mean difference = −2.0, 95% CI (−3.1, −0.9), P = 0.001), and capacity for short walk (Q9: Mean difference = −1.9, 95% CI (−3.5, −0.3), P = 0.02) with droxidopa treatment. The average score across the 10 OHSA questions was lower for droxidopa treatment but not statistically significant (Mean difference = −0.67, 95% CI (−1.55, 0.22), P = 0.13).

Post-hoc analyses

During the 12-week crossover clinical trial, mean diastolic blood pressure was higher in all positions tested while on droxidopa, with the increase statistically significant for upright diastolic BP (P < 0.05, Fig. 4A). There was also statistically significant improvement in a critical parameter of orthostatic hypotension, the drop in diastolic blood pressure during tilt table testing while receiving droxidopa compared to placebo or no treatment (Fig. 4B, middle panel; difference = −8.6 [95% CI −15.5 to −1.7]; P = 0.018).

Consistent with the changes in plasma NE and DHPG (Fig. 5A and B), the ratio of DOPAC to DHPG (see catecholamine biosynthetic pathway, Figure S1B) in plasma was also significantly different while receiving droxidopa (Fig. 5C), commensurate with normal or near-normal values,¹⁴ and in contrast to untreated Menkes disease or occipital horn syndrome.^3,10,11

PGI-S scores were generally lower during droxidopa treatment compared to placebo (Figure S3B), with statistically significant results for question 1 (mean ratio = 0.62, 95% CI (0.41, 0.95), P = 0.03) and for overall average score (mean difference = −0.5, 95% CI (−0.78, −0.25), P < 0.001), indicating a beneficial effect of droxidopa in this patient population.

Discussion

In this double-blind placebo-controlled phase 1/2a clinical trial, we documented the safety, tolerability, and efficacy of droxidopa in two adult Menkes disease survivors and an adult with occipital horn syndrome. The patients enrolled were well-known to us and have been characterised previously in terms of the clinical, biochemical, and molecular aspects of their copper transport defect. Participant Droxi-01 was reported as participant II-06 in a previous paper on response to copper histidinate treatment⁵ and as a case report on his molecular variant.¹⁵ Participant Droxi-002 was described previously as patient 8 in a study on neonatal diagnosis of Menkes disease³ and participant Droxi-03 was described previously as patient A-1 in a study related to ATP7A gene expression.¹³ The Menkes disease participants were both identified as affected in the newborn period based on 50% risk for the disorder based on family history. The diagnosis was confirmed by plasma catecholamine analysis, a rapid and reliable diagnostic test for Menkes disease in the newborn period, an interval when other biochemical biomarkers are unreliable and molecular diagnostic studies may require several weeks for results.¹⁶ The same abnormal pattern of blood neurochemicals and catecholamine ratios that enabled their early diagnosis and enrolment in a clinical trial of CuHis persisted in them as indicators of dysautonomia in adulthood.

Participant Droxi-03 was not treated with CuHis based on his low normal level of serum copper and mild clinical abnormalities when initially evaluated at (age 19 years).¹³ He was noted at that time to have walked at 2 years of age, to have attended normal schools, and to have suffered from chronic diarrhoea and occasional dizziness upon rising. Intermittent episodes of dizziness and chronic diarrhoea involving 12–15 loose stools daily have impacted his ability to work and participate in sports such as pick-up soccer.

Participant Droxi-01 graduated from high school and attends community college part-time; he obtained his driver's license and drives a car. Despite these noted accomplishments, his symptoms of dysautonomia including low blood pressure, dizzy spells, limited duration of standing and loose stools continue to provide a barrier to more comfortable quality of life.

Participant Droxi-02 graduated from high school with his normal peer group and is a gifted artist. His numerous episodes of dizziness, weakness/fatigue, severe difficulty standing and walking, and numerous episodes of spontaneous bowel movements have extremely limited his daily activities.

The symptoms of postural hypotension (dizziness and lightheadedness) in all participants in this trial were first noted in late childhood/early adolescence when their stature exceeded approximately 50 inches. The precise onset of chronic diarrhoea was more difficult to ascertain. The dysautonomia symptoms that these individuals endure represent significant impediments to their social, academic, and athletic pursuits, and troublesome concerns for their parents. The failure to document more notable improvements in tests of physical exertion and gastrointestinal function in this trial may reflect the relatively short period of droxidopa treatment. Longer exposure to droxidopa may further improve these aspects.

Similar signs and symptoms of dysautonomia can occur in younger Menkes disease survivors whose smaller body mass and potentially exquisite sensitivity to droxidopa will likely require smaller doses than 100 mg. For this purpose, development of a stable oral suspension of droxidopa would be ideal.

In previous clinical trials of CuHis treatment, correction of neurochemical abnormalities was not evident. That copper replacement treatment alone does not resolve the neurochemical abnormalities in participants with Menkes disease highlights the strict requirement for ATP7A to deliver copper into the secretory pathway for metalation of DBH (Figure S1). In contrast to CuHis, we show here that droxidopa clearly corrects the abnormal pattern in adult survivors of Menkes disease, as predicted by previous work in a mouse model of Menkes disease.¹⁷ Plasma norepinephrine (Fig. 5A) and DHPG (Fig. 5B) levels increased during droxidopa exposure (P < 0.01, P < 0.05, respectively), which correlated with measurable improvements in blood pressure and orthostatic hypotension (Fig. 4). As in our murine pre-clinical work, we detected the aldehyde and potential neurotoxin DOPAL in plasma of participants receiving droxidopa in this study. Quantitatively however, this chemical species was present at approximately 10,000-fold lower amounts in comparison to levels of L-DOPS (droxidopa) in the same blood specimens.

With three participants, the current study was underpowered. In addition, this study did not include neuroimaging (e.g., functional MRI, arterial spin mRI, diffusion tensor imaging, PET/SPECT) that could have been informative for linking autonomic improvements with central nervous system integrity. In future studies of dysautonomia in this population, brain imaging studies would afford the opportunity to document baseline brain atrophy, white matter lesions, and/or vascular anomalies known to occur in Menkes disease² and assess whether droxidopa use correlates with stability or change.

It will be of considerable interest to evaluate neurochemical correction in a future first-in-human gene therapy trial that provides some working copies of ATP7A to human participants with Menkes disease.⁸ The evidence of reduced dysautonomia symptom severity in adults with ATP7A variants studied in this trial warrants investigation of the risk:benefit profile of droxidopa in younger patients with ATP7A-related disorders associated with dysautonomia.

Contributors

All authors read and approved the final version of the manuscript. Maryann M. Kaler contributed to conceptualisation, data curation, funding acquisition, investigation, methodology, project administration, writing-original draft, and writing-review and editing. Guy Brock contributed to conceptualisation, data curation, formal analysis, investigation, methodology, validation, visualisation, writing-original draft, and writing-review and editing. Christopher J. Jeanty contributed to data collection, curation, and analysis of the reported study data alongside other co-authors. Megan A. Iammarino contributed to investigation, methodology, writing–review & editing. Kimberly A. Hampton contributed to data collection.

Minh T. Pham contributed to data collection, curation, and analysis of the reported study alongside other co-authors. Brenna T. Sabo contributed to data collection. Linda P. Lowes contributed to investigation, methodology, supervision, validation, writing review and editing. Patricia Sullivan contributed data collection and analysis. David S. Goldstein contributed to data collection and analysis, conceptualisation, data curation, formal analysis, resources, and validation. Stephen G. Kaler contributed to conceptualisation, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, supervision, validation, visualisation, writing-original draft, and writing-review and editing. SGK and MMK directly accessed and verified all data reported in this manuscript.

Data sharing statement

The clinical trial protocol is provided in Supplementary Materials. The statistical analysis plan, and de-identified study data will be made available to appropriate parties upon request to the corresponding author and a mutually acceptable data transfer agreement.

Declaration of interests

SGK reports funding from Nationwide Children's Hospital Research Institute, Menkes Disease Gift Fund, (Menkes Disease Foundation UK, and Associazione Angeli Per La Vita); and participation on a Data Safety Monitoring Board for Vivet Therapeutics, Inc. DSG reports receipt of royalties for published books (The Johns Hopkins University Press; Kindle Direct Publishing); consulting fees from the Autonomic and Catecholamine Healthspan Institute, LLC (ACHI) and Private Health Management under ACHI; honoraria for service as faculty at The Residents Course in Autonomic Medicine and The Clinician Program in Autonomic Medicine, The Dysautonomia Project; payment for manuscript writing (Journal of Clinical Neurology); and travel reimbursements from the Residents Course in Autonomic Medicine, the Clinician Program in Autonomic Medicine, and the Korean Neurological Society. All other authors declare no competing interests.