Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disorder characterized by motor neuron death in the primary motor cortex, brain stem, and spinal cord, leading to muscular weakness and atrophy [1,2]. ALS can lead to progressive, variable weakness of the skeletal muscles involved in swallowing, speaking, coughing, limb movement, trunk control, and breathing. Respiratory muscle denervation causes diaphragm and accessory muscle weakness, hypoventilation, dyspnea, and orthopnea [3,4]. Ineffective airway clearance and aspiration result from bulbar denervation. Respiratory and bulbar dysfunction lead to the need for mechanical ventilation [5], shortened survival, and diminished quality of life [6]. Typical survival is 2-5 years post diagnosis [7]. Treatments that preserve or improve breathing would have a considerable impact on the duration and quality of life.

Acute intermittent hypoxia (AIH) consists of short, repeated exposures to mild or moderate hypoxia (9%-12% O₂), separated by normoxic episodes (21% O₂). Hypoxic episodes activate the carotid body chemoreceptors, sending chemoafferent signals to stimulate medullary respiratory neurons [8] as well as brain stem raphe neurons that contain the neurochemical serotonin. When this neural network is repeatedly stimulated during AIH [9], serotonin release from the raphe neurons triggers cell signaling pathways resulting in new synthesis of brain-derived neurotrophic factor protein. In turn, brain-derived neurotrophic factor strengthens synapses onto respiratory (and nonrespiratory) motor neurons [10-12], increasing respiratory motor output. AIH-induced motor plasticity was first described as a persistent enhancement of phrenic neural activity lasting hours after AIH has ended, an effect known as phrenic long-term facilitation (pLTF); a similar effect occurs in hypoglossal motor neurons [13,14]. pLTF is a form of activity-independent, neuromodulator-induced neuroplasticity that exhibits sensitivity to the pattern of hypoxic episodes and time of day [8,15,16]. Repeated AIH appears to amplify and prolong pLTF without detectable pathology [17-20], Thus, AIH may induce respiratory rehabilitation in ALS as well as other neurological conditions that compromise breathing [21].

Our group recently completed the first clinical trial of AIH administration in patients with ALS compared with unaffected controls. In this study, AIH enhanced tidal volume, minute ventilation, and collective respiratory muscle activation of participants living with ALS versus age- and sex-matched controls. Importantly, AIH was well-tolerated and free of treatment-related adverse effects; the participants typically could not distinguish between AIH versus sham AIH (ie, normoxia) [22].

While the modest AIH used in rehabilitation typically elicits motor facilitation via a serotonin-dependent mechanism (Gq protein–coupled receptor signaling [termed the “Q” pathway]) [8,14,23,24], an alternative AIH-induced signaling pathway results from adenosine (Gs protein–coupled receptor signaling [termed the “S” pathway]) [25-28], acting on adenosine 2A (A₂A) receptors [29]. Serotonin and adenosine signaling oppose one another through powerful cross-talk inhibition; in extreme cases, this “battle” cancels phrenic motor plasticity [28,30-32]. Although both serotonin and adenosine receptors can be activated during mild hypoxia, the serotonergic “Q” pathway predominates. However, multiple personal characteristics, disease states, and treatment factors upregulate the adenosinergic “S” pathway, sufficiently to inhibit plasticity (Figure 1), including aging [33], low estrogen [33-35], inflammation [36], active phase delivery [37], genetic factors [33], and traumatic or neurodegenerative diseases [38].

In rodent models of spinal cord injury, an increased pLTF was observed after the administration of paired AIH and the A₂A receptor inhibitor istradefylline [31]. In a subsequent clinical study, the impact of combined AIH and caffeine, a nonselective adenosine antagonist, found the combined approach further improved walking speed and distance in people with chronic spinal injuries [39]. Thus, combined AIH and A₂A receptor antagonism may amplify respiratory motor plasticity [40]. However, since caffeine is a nonselective adenosine antagonist that also inhibits phosphodiesterase action, off-target effects such as nausea and elevated heart rate (HR) could limit its use in the doses needed to augment AIH plasticity. This study uses istradefylline for its greater affinity for the A₂A receptor and its tolerance in clinical studies of older adults and those with Parkinson disease (PD).

Elevated spinal adenosine is a characteristic of neurodegenerative diseases, such as ALS [41,42]. In the SOD1^G93A (copper zinc superoxide dismutase 1) mouse model of ALS, A₂A receptor signaling contributed to motor neuron degeneration, while A₂A receptor antagonism delayed disease progression [42]. A₂A antagonism can also have neuroprotective and anti-inflammatory effects in motor neurons [43]. NOURIANZ (istradefylline) is a highly selective A₂A receptor antagonist currently approved for use in adults with PD to offset inhibitory effects of upregulated A₂A receptors on D2 dopamine receptors of the basal ganglia [44]. Istradefylline reduces daily “off” time in people with PD, improving periods of mobility, and is reported safe and well tolerated [45]. Dyskinesia (5%) and hallucinations (3.4%) were the most common side effects associated with longer-term use [46], but cognition, sedation, or risk of cardiovascular events were not impacted [47]. Considering the demonstrated safety of chronic istradefylline in PD and healthy controls, we hypothesize that this brief, temporary treatment will be safe in ALS.

We further hypothesize that the combination of istradefylline and AIH, in patients with ALS, will elicit greater gains in respiratory and nonrespiratory motor function than either intervention alone, and that will contribute to improvements in breathing and motor function. In a rodent model of phrenic motor neuron death, significant A₂A receptor upregulation preceded motor neuron loss, but the coadministration of istradefylline improved motor neuron survival and maintained diaphragm function [43]. The information gained in this investigation will help us optimize intervention approaches for future studies of repeated, daily AIH as a rehabilitation tool. Thus, this study represents an important step in the iterative process of moving AIH protocols toward clinical practice.

The objective of this study is to evaluate single and combined effects of selective A₂A receptor antagonism and AIH on respiratory strength, resting breathing, and participant-reported symptoms in adults living with ALS and age-matched unaffected adults. The following two aims are proposed to test the hypotheses:

Aim 1: a single 20 mg istradefylline dose prior to AIH is safe and feasible. Dependent variables will include rate of adverse events (primary outcome), acute changes in vital signs (HR, respiratory rate, blood pressure [BP], etc), patient-reported symptoms (dyskinesias and dizziness), and study completion.

Aim 2: a single 20 mg istradefylline dose prior to AIH improves respiratory motor function when compared to either intervention alone. Dependent variables will include resting tidal volume (primary outcome), breathing rate and minute ventilation, respiratory muscle assessments via multichannel surface electromyography (sEMG), cough force, maximal inspiratory pressure (MIP), and maximal finger pinch force.

This is a repeated-measure, double-blinded, placebo-controlled, randomized trial, administered at a single site, the University of Florida (UF). This clinical trial was registered in the United States National Library of Medicine through ClinicalTrials.gov (NCT05377424).

Patients living with ALS will be recruited from the UF ALS clinic, as well as through flyers, word of mouth, and the study registry website. Patients who meet the study criteria will be identified by the study physicians through the UF adult neuromuscular clinic. Healthy age-matched controls will be recruited via advertising for the study on ClinicalTrials.gov, StudyConnect, social media for the UF Clinical and Translational Science Institute, and flyers throughout the local community. Additional potential healthy controls will be identified from the Pepper Center Registry maintained by the UF Institute on Aging, and institutional review board (IRB)–approved flyers will be sent to their address. Study interventions will occur at a university-based clinical research center located in Gainesville, FL.

Eligible participants include nonsmoking adults diagnosed with ALS or age-matched unaffected controls, aged 21-80 years. Eligible participants will have El Escorial diagnostic classification of probable or definite ALS and vital capacity (VC) ≥60% of predicted value; we seek to treat participants before they approach respiratory failure and need for mechanical ventilation, which typically occurs when VC falls below 50% predicted [48]. The ALS Functional Rating Scale-Revised (ALSFRS-R) consists of 12 ordinal items of 5 points (rated 0-4) each, which describe impairments in different systems functionally, where a score of “4” reflects normal function and “0” reflects profound dysfunction [49]. Scores below 2 for respiratory and bulbar items typically reflect severe functional impairment (ie, need for supplemental tube feeding and continuous use of respiratory support). Clinically prescribed ALS medications must be at a stable dose for 30 days.

Participants (both affected and controls) are ineligible if they are pregnant, have an active respiratory infection, took antibiotics within 4 weeks of recruitment, are diagnosed with another neurodegenerative disease, have symptomatic cardiovascular disease or dysrhythmias, have a BMI of >35 kg/m², have a seizure disorder, use respiratory inhalers daily for airway disease, or require external respiratory support while awake and upright. In addition, the following conditions are exclusionary for the use of istradefylline: routine use of CYP3A4 inducers (eg, carbamazepine, rifampin, phenytoin, and St John’s Wort), history of moderate renal impairment or severe hepatic impairment, and history of hallucinations or psychosis.

AIH has shown to improve cognitive and exercise performance by improving hemodynamics and work capacity in healthy older adults [50,51]. While effects of AIH and istradefylline have been tested separately in healthy older adults, there is a lack of evidence regarding the additive effects of pairing istradefylline with AIH. Since older adults are the individuals who typically have the greatest need for rehabilitation, we are interested to test whether combined istradefylline and AIH can augment respiratory motor plasticity in healthy older adults.

Healthy older adults have been previously included in clinical trials to evaluate the exposure-response relationship and pharmacokinetic disposition of istradefylline in patients with PD [45,52]. By including healthy older adults in our study, we can determine whether any observed differences with paired AIH and istradefylline among individuals with ALS are specific to ALS or are potentially more generalizable to the older adult population.

The SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials) flow chart with all study procedures can be found in Table 1.

Ethical approval was granted by the UF IRB (IRB 202101568) on May 16, 2022. The US Food and Drug Administration provided clearance for off-label investigational use of istradefylline as described in the protocol (investigational new drug application, approved April 8, 2022; 156267). Informed consent will be obtained in writing from all participants prior to enrollment by study team members. See Multimedia Appendices 1 and 2 for informed consent forms. Participants were informed they would be able to withdraw from the study at any time for any reason. This work is supported by the ALS Association. Outside of the human research protection review, the funding sponsor does not have a direct role in directing the study design; collection, management, analysis, and interpretation of data; writing of the report; or the decision to submit the report for publication.

Health information will be collected, used, and shared with others (ie, clinical and research staff and the UF Research Participant Payment team) to determine if interested individuals are eligible to participate, and then as part of the participation in the study. The following information may be collected, used, and shared with others: medical history, including past conditions and surgeries, medications, and information related to diagnosis of ALS; social security number (for reimbursement purposes); demographic information; contact information; and the results of the research tests done for this study. The research team may collect this information from other health care providers, such as laboratories, which are a part of this research, as well as health care providers that are not part of this research (other doctors, hospitals, or clinics). Other professionals at the UF or UF Health Hospital who provide study-related care, and the UF IRB, may also collect health information. Participants will be informed of the collection and protection of health information collected as part of the informed consent process. Health information and study data will be stored in locked filing cabinets or on computer servers with secure passwords or encrypted electronic storage devices, as required by UF policy.

Participants will be compensated US $20 for the screening visit and US $120 for each treatment visit, via prepaid Visa gift cards. Additionally, overnight hotel expenses will be covered by the study for participants traveling from out of town, and lunch will be provided during treatment visits.

Interested individuals who meet the basic study criteria, agree to participate, and sign the IRB-approved informed consent form will undergo further screening procedures prior to the first interventional visit. Prospective participants will be informed of the eligibility criteria, study procedures, participation risks, privacy and safety precautions in place, alternatives to study participation, and the right to withdraw from the study at any time. The study coordinator will schedule visits and communicate with participants by phone and email. The study coordinator or a designated investigator will complete consenting and screening tests (Figure 2).

Screening tests shall include:

After providing written consent and completing screening, participants will complete 4 individual study visits separated by 2 weeks (±5 days). Visits will include (1) istradefylline + AIH, (2) placebo + AIH, (3) istradefylline + sham AIH, and (4) placebo + sham AIH (Figure 3), in randomized order. Participants will undergo single sessions of each treatment condition. They will be instructed to avoid exercise, caffeine, and nicotine products prior to testing for a minimum of 6 hours and to avoid eating within 1 hour of sessions.

Session order was determined by a random number table generated in advance of the enrollment of the first participant and maintained by the investigational pharmacist. Participants and investigators will remain blinded to istradefylline versus placebo for study administration and data analysis. Participants as well as the investigator collecting test outcomes will be blinded to AIH versus sham AIH. Prior to each study visit, the investigational pharmacist will inform the study coordinator which gas to administer during intervention (ie, hypoxic air for AIH vs normoxic air for sham AIH). Study team members conducting gas treatment cannot be fully blinded to AIH versus sham AIH due to monitoring of pulse oximetry oxygen saturation (SpO₂), which is necessary for safety; thus, the investigator collecting testing outcomes will not be present during gas administration.

Participants will sit upright in a chair with the head and trunk supported while breathing through a sealed face mask. Resting ventilation, respiratory muscle activation, maximal respiratory pressures, maximal pinch force, and blood draws will be completed at the start of each study visit, then participants will receive study medication to be taken orally. After a 4-hour break, all outcome-measure testing will be repeated, followed by gas delivery. Postintervention testing will occur at 60 and 120 minutes following gas delivery. Table 2 delineates outcome measures taken at each time point.

Study visits are expected to last 7-8 hours, including equipment setup, repeated outcome measurements, oral administration of study medication, waiting periods, gas delivery, patient positioning and pressure relief, rest, and snacks or meal breaks.

During each study visit, participants will take a single, 20 mg NOURIANZ pill or an identical-appearing placebo. The study agent will be taken after the completion of baseline breathing tests. Study gas delivery will begin 4 hours later. During the 4-hour waiting period, participants will have a private room with a bed and recliner and access to water, snacks, and a light meal, as well as bathroom facilities. Participants will also be free to move about the research unit.

Following the 4-hour break and second round of testing, participants will undergo a 45-minute session of either AIH (up to 1 minute of 9%-10% O₂ with 2 minutes of 21% O₂ for 15 cycles; target SpO₂ nadir 82%-85%) or sham AIH (Figure 4). During sham AIH administration, the 1-minute periods of study gas will use a gas bag containing 21% O₂, with 2-minute room-air periods, so that participants are breathing normoxic air for the entire session. Outcome measures will be repeated 60 and 120 minutes after gas delivery.

A schematic of the respiratory circuit is shown in Figure 5. Participants will sit upright in a chair with the head and trunk supported while breathing through a sealed face mask (Hans Rudolph 7450). The face mask will be connected to a 2-way nonrebreathing valve (Hans Rudolph 2700) that allows the exhaled breath to escape to room air. A pressure transducer (MK1S, GM Instruments) and gas analyzer (Gemini CO₂ & O₂ monitor; CWE Inc) will be connected directly to the face mask connector. A pneumotachograph (Hans Rudolph 4813) will be connected to the inspiratory port of the nonrebreathing valve. Gas bags will be connected to a 3-way stopcock valve (Hans Rudolph 2100), and the direction of the valve will be manually alternated between the study gas (1-minute intervals) and normoxic room air (2-minute intervals). Respiratory data will be recorded during gas delivery, using a data acquisition system (PowerLab 16/35 and PowerLab C; ADInstruments).

Respiratory rate, targeted O₂ value, end-tidal CO₂, HR/rhythm, and SpO₂ will be continuously monitored during the testing sessions. Multichannel sEMG (Trigno Centro; Delsys) will also be recorded from respiratory muscles (bilateral submental, scalene, sternocleidomastoid, second parasternal, and diaphragm) as well as abductor pollicis muscle during testing and captured in real time in LabChart (AD Instruments). At the end of each 1-minute hypoxic (or sham-hypoxic) presentation, participants will report their rating of dyspnea (10-point modified Borg scale). For every fifth presentation, noninvasive BP will be taken. We will establish an individualized communication plan (eg, hand signal, head tilt, call switch, etc) prior to study gas administration to ensure that all participants can alert the investigator of any discomfort, shortness of breath, or requests to remove the face mask.

The target SpO₂ during AIH is 82%-85% O₂. If the SpO₂ reaches the lower limit of the target range, gas delivery will be switched to the normoxic recovery time. Subsequent hypoxia durations will also be reduced by 5 seconds. AIH will be stopped in the unlikely event the HR increases or decreases >30% from the resting value, or systolic BP changes >30% from the resting value. The expected acute physiological response to hypoxia is a stable or decreased CO₂ due to increased ventilation. However, if end-tidal CO₂ increases >3 mmHg above baseline, we will shorten AIH bouts using the same process described above for SpO_2. Additional stopping criteria include a respiratory rate of >30 that fails to normalize within 1 minute of normoxia, SpO₂ <82% that fails to normalize within 1 minute of normoxia, or SpO₂ <82% occurring within 30 seconds of hypoxia initiation.

In the unlikely event that a participant is injured as a direct result of participation in this study, professional services rendered by any UF Health Science Center health care provider will be provided without charge. These health care providers include physicians, physician assistants, nurse practitioners, dentists, and psychologists. Any other expenses, including UF Health hospital expenses, will be billed to the participant or his or her insurance provider.

Safety oversight will occur internally through the local IRB and a data safety monitoring board (DSMB). The DSMB was convened before enrollment of the first patient, after completion of the first 3 participants, and at least annually thereafter. The DSMB includes 3 members—1 expert in intermittent hypoxia, 1 clinical medicine expert who interprets or prescribes respiratory testing and treatment, and 1 rehabilitation specialist.

Testing and enrollment of participants will be paused, and DSMB will convene for any of the following events: (1) any clinically significant respiratory event, (2) a clinically significant cardiovascular event, or (3) a serious adverse event as defined by the UF IRB. Should any of the events occur, research staff shall inform the IRB and the DSMB. Any subsequent study activities would only be resumed when cleared or the protocol modified by the IRB and DSMB. As of January 2025, these serious events have not occurred, and the DSMB has only been convened for planned early and annual meetings.

Secondary outcome measures will be interpreted as exploratory in nature.

The sample size calculation was based on pilot data from 8 adults with ALS showing ventilation changes in AIH versus a sham condition. The preliminary data showed 10.3% (SD 23%) larger tidal volume and 10% (SD 29%) greater minute ventilation 60 minutes post AIH, as compared to sham. Published work in rodent spinal cord injury models indicates that istradefylline paired with AIH greatly enhances improvement in diaphragm activity versus AIH or drug alone (effect size of combined approach: d=1.92).

Additional preliminary data using caffeine as a nonspecific adenosine receptor antagonist conservatively suggested a capacity for a 10% additional gain in ventilation. With 16 crossover participants, a repeated measures ANOVA comparing 4 conditions across 4 time points, with a pooled SD of 15.3, will be sensitive to detect a difference of power of >0.8 (f=0.31; F_{3, 36}=2.86; ρ=0.5; α=.05). A target sample of 24 participants will account for potential attrition. The design includes age-matched unaffected adults as a comparator group.

Descriptive statistics, including frequencies, means, SDs, and medians, will be used when appropriate. Continuous variables will be summarized with descriptive statistics, and categorical variables will be summarized with counts and percentages. Demographics and disease-progression variables will be reported descriptively; while these are not included in a priori analyses of study aims, they may be used in an exploratory manner to interpret findings as appropriate. Change scores and percent change scores will be calculated from baseline for follow-ups at postdrug or placebo administration, 60 and 120 minutes after gas intervention. We will attempt to minimize missing data by planning the study procedures and proper biosignal acquisition, thoroughly documenting the study administration and data collection, and collecting the essential information to complete the study aims. Should missing data points still occur, statistically valid analyses that appropriately address the missing values are maximum likelihood, expectation maximization, and multiple imputation [64]. Any statistical code or algorithms developed to analyze acquired data will be available and shared upon request.