Neuro-regulatory drug discovery (NRGDD): a neural-circuit framework for systemic therapeutic modulation

1 Introduction: the brain as an underutilized regulator in drug discovery

1.1 Historical foundations of neural regulatory biology

The idea that the nervous system governs peripheral physiology has a long history. Walter Cannon showed in 1932 that the body maintains stable internal conditions through coordinated autonomic output rather than passive equilibrium (Cannon, 1932). Hans Selye demonstrated that the brain’s stress response produces systemic hormonal and immune changes with direct consequences for health (Selye, 1936). Claude Bernard earlier established that organisms actively defend their internal environment through regulatory mechanisms, meaning the brain continuously adjusts organ function through dedicated circuits (Thayer and Lane, 2009). McEwen showed that prolonged disruption of neural-peripheral signaling contributes directly to disease (McEwen, 2007). Each of these findings pointed to the same conclusion: the nervous system does not merely respond to disease. It actively regulates the physiological processes that determine whether disease develops or resolves (Kandel et al., 2014).

The most direct therapeutic evidence came from Tracey’s 2002 discovery that vagal nerve signals suppress TNF production in the spleen through cholinergic receptors on macrophages (Tracey, 2002). This identified a specific neural circuit with pharmacologically relevant anti-inflammatory effects. Pavlov and Tracey later defined the molecular mechanism: the alpha-7 nicotinic acetylcholine receptor (alpha7 nAChR) on macrophages acts as the terminal effector of this pathway, and the mechanism is conserved across species and disease settings (Pavlov and Tracey, 2017). Clinical validation through VNS trials in inflammatory arthritis demonstrated that engaging a defined neural circuit produced sustained TNF suppression and clinical improvement in patients who had not responded to conventional therapy (Koopman et al., 2016).

1.2 The translational gap and the case for NRGDD

Despite this evidence, neural regulatory mechanisms are not routinely used as starting points in drug discovery. Most pharmaceutical research still begins with an isolated molecular target. Animal models used in early development often do not reflect the neural-peripheral coupling seen in humans. Around 90% of drug candidates that show efficacy in preclinical studies fail to produce benefit in clinical trials (Scannell et al., 2012). Part of this failure reflects a mismatch between reductionist discovery logic and the integrative regulatory architecture of human physiology.

NRGDD addresses this mismatch by repositioning the brain as the initiating therapeutic organ. It draws on established neuroscience and pharmacology and applies them within a discovery structure defined by specific circuit eligibility criteria, causal proof requirements, and biomarker validation standards that have not previously been combined into a unified development framework. The working hypothesis is that drugs targeting specific neural circuits will produce consistent, measurable effects in peripheral tissues. These effects can be tracked and used to make go/no-go decisions throughout development.

Neural circuits also offer a meaningful advantage in the durability of therapeutic response. Molecular inhibitors typically require continuous exposure to maintain target suppression, which creates compliance challenges and can allow resistance mechanisms to emerge. Neural circuits, by contrast, may be amenable to resetting or recalibration through appropriately timed interventions, potentially yielding benefits that persist beyond the active treatment period. Evidence from VNS studies indicates that engaging these circuits can induce lasting shifts in inflammatory set points and autonomic balance (Koopman et al., 2016; Tesser et al., 2025). This durability argument strengthens the case for neural circuits as primary targets in complex chronic disease.

1.3 Relationship to established fields

NRGDD builds on, but is distinct from, three adjacent research traditions. Bioelectronic medicine, as defined by Famm et al. (2013) and advanced through the work of Tracey and colleagues, uses implantable devices to modulate organ-specific nerve fibers and has produced the clinical evidence base on which NRGDD draws most directly. Neuroimmune regulation encompasses the mechanistic biology of neural control over immune function, including the cholinergic anti-inflammatory pathway, HPA axis immunomodulation, and sympathetic adrenomedullary effects on lymphocyte trafficking (Pavlov and Tracey, 2017; Wrona, 2006). Systems pharmacology integrates network-level models of drug action with emergent physiological properties rather than isolated molecular targets (Sorger et al., 2011). NRGDD does not supersede these fields. It draws on their findings and reconfigures them within a drug-discovery framework that specifies operational decision gates, causal proof standards, and translational bridging requirements that none of these fields, individually, provides as a development logic. The novelty of NRGDD is organizational and operational: it defines what must be demonstrated, in what sequence, before a neural-circuit-based drug program can advance.

2 Conceptual definition and boundaries of neuro-regulatory drug discovery

2.1 Formal definition

NRGDD is a drug-discovery framework proposed by the author in which treatment works by engaging a specific neural circuit. The effect of that engagement must be measurable in peripheral tissues through validated biomarkers. The neural circuit is the primary therapeutic target in the most direct sense: the neural engagement event is the primary target. It precedes and governs the therapeutic outcome. It is not an incidental consequence of drug exposure or a secondary modifier of disease expression. Interventions are designed, optimized, and evaluated based on their capacity to reproducibly activate specific neural pathways and produce quantifiable downstream biological effects. This positioning represents a fundamental departure from conventional pharmacology, which has historically treated molecular receptors, enzymes, and signaling proteins as targets for therapeutic intervention (Table 1).

TABLE 1. NRGDD Classification and intervention class.

NRGDD, compared with related intervention classes; NRGDD, uniquely requires both defined neural circuit engagement and demonstrated peripheral biomarker causality. Device-based NRGDD, serves as a validation platform for subsequent pharmacological development.

Peripheral changes in cytokines, metabolic markers, hormones, or organ function are treated as outputs of neural control, not as independent pharmacodynamic effects arising from direct drug action at peripheral sites. This distinction has direct consequences for assay design, biomarker selection, and clinical development planning. Without this causal link, an intervention may produce symptomatic benefit through some other route but would not qualify as NRGDD. Causality must be demonstrated experimentally, not inferred from temporal association. The framework requires three converging lines of evidence: temporal precedence (neural activation before biomarker change), dose-response concordance (graded engagement producing proportional effects), and disruption attenuation (pathway blockade markedly reducing the peripheral response).

The NRGDD framework recognizes two distinct modes of pharmacological engagement with a neural regulatory circuit. The first is proximal engagement: the drug acts at a neural node within the circuit (a ganglion, a pre- or postganglionic fiber, or a central nucleus), and the downstream peripheral effects arise from neural signal propagation. The second is terminal receptor engagement: the drug acts at the circuit’s defined peripheral effector receptor—a receptor whose identity, location, and role within the circuit have been established by prior neuroanatomical and pharmacological work—and produces effects mechanistically equivalent to those arising from upstream neural activation. Terminal receptor engagement qualifies as NRGDD only when the receptor has been formally characterized as the terminal effector of a defined neural circuit, when the peripheral effects mirror those produced by upstream neural activation in disruption experiments, and when the drug does not produce those effects through an independent non-neural mechanism. This distinction is applied explicitly to Case Study III in Section 6.4.

NRGDD encompasses both pharmacological agents and device-based interventions, provided the initiating mechanism is defined as neural circuit engagement with demonstrable peripheral consequences. Pharmacological agents, including acetylcholinesterase inhibitors, alpha7 nAChR agonists, CRH receptor modulators, and sympathomimetics, represent the primary modality for drug discovery. Device-based neuromodulation is included as a mechanistic validation strategy to identify druggable circuits that can subsequently be accessed pharmacologically. The term 'Drug Discovery’ in NRGDD reflects this primary pharmacological orientation. Table 1 compares NRGDD with related intervention classes.

2.2 What does not qualify as NRGDD

Defining what NRGDD excludes is necessary to prevent the concept from becoming so broad as to be operationally useless. Four categories do not qualify.

2.2.1 CNS symptom drugs

Drugs such as SSRIs, antipsychotics, and analgesics act at central receptors to change how symptoms are perceived. They do not produce defined causal effects on peripheral physiology, as required by NRGDD.

2.2.2 Non-specific autonomic drugs

Non-selective beta-blockers or general sympatholytic agents broadly alter autonomic tone. Without a defined target circuit and measurable peripheral biomarker response, they do not qualify.

2.2.3 Placebo responses

Context and expectancy can activate neural pathways and produce some physiological changes. These responses are inconsistent, show high individual variability, and cannot be replicated by disruption testing. The neural networks involved differ substantially from the autonomic and hypothalamic circuits targeted in NRGDD.

2.2.4 Structural and genetic repair therapies

Gene therapy for inherited enzyme deficiencies, cell replacement for end-stage tissue loss, and other approaches that reconstruct rather than regulate physiology fall outside NRGDD. These are discussed further in Section 7.1 and summarized in Figure 1.

3 Scientific rationale: why neural circuits offer advantages in complex disease

3.1 Hierarchical control and simultaneous multi-effector regulation

A single molecular target controls a single step in a single pathway. A neural circuit controls multiple pathways simultaneously. A single vagal signal simultaneously reduces TNF, IL-1beta, IL-6, and HMGB1 while also shifting macrophage behavior, altering lymphocyte trafficking, and altering how metabolic substrates are distributed. No molecular inhibitor can replicate this. This multi-effector coordination explains why neural-initiated interventions may outperform molecular inhibitors in diseases where multiple redundant pathways sustain inflammation, and where inhibiting any one target allows compensatory restoration of pathological activity.

Three well-characterized neural regulatory circuits illustrate this with specific molecular detail. Their properties, measured biomarkers, and disease applications are summarized in Table 2.

TABLE 2. Neural circuit classes for NRGDD development.

NTS, nucleus tractus solitarius; DMV, dorsal motor nucleus of vagus; NPY, neuropeptide Y; CAP, cholinergic anti-inflammatory pathway; GR, glucocorticoid receptor; HRV, heart rate variability; Epi = epinephrine; NE, norepinephrine.

3.1.1 Cholinergic anti-inflammatory pathway (CAP)

Vagal efferent fibers release acetylcholine (ACh) at the celiac ganglion, activating the splenic nerve through norepinephrine. Splenic nerve terminals stimulate T lymphocytes expressing choline acetyltransferase (ChAT) to produce ACh. That ACh binds alpha7 nAChRs on splenic macrophages, activates JAK2-STAT3, blocks NF-kappaB translocation, and suppresses TNF, IL-1beta, IL-6, and HMGB1 (Tracey, 2002; Pavlov and Tracey, 2017). Drug access points include acetylcholinesterase inhibitors, such as pyridostigmine, alpha7 nAChR agonists, such as GTS-21 and PNU-282987, and partial agonists, such as transdermal nicotine. Circuit activity is tracked by serum TNF, IL-6, IL-1beta, HMGB1, and HRV spectral analysis (Pullan et al., 1994).

3.1.2 Hypothalamic-pituitary-adrenal (HPA) axis

Corticotropin-releasing hormone (CRH) from the paraventricular nucleus drives pituitary ACTH secretion, which triggers cortisol release from the adrenal cortex. Cortisol binds glucocorticoid receptors (GR) on immune cells, raises IkappaB-alpha, suppresses NF-kappaB, and reduces pro-inflammatory cytokines, COX-2, and adhesion molecules. Drug access includes CRH receptor-1 blockers such as antalarmin and pexacerfont, GR partial agonists with separated anti-inflammatory profiles, and ACTH analogues. Tracked by cortisol, ACTH, and salivary cortisol. Relevant to autoimmune disease, metabolic syndrome, and post-traumatic stress disorder (Sapolsky et al., 2000).

3.1.3 Sympathetic adrenomedullary axis

Hypothalamic and brainstem signals drive preganglionic sympathetic fibers to the adrenal medulla, triggering release of epinephrine (80%) and norepinephrine (20%). These catecholamines act on beta-1 and beta-2 adrenergic receptors on immune cells, heart tissue, and metabolic organs, suppressing TNF transcription through the cAMP-PKA-CREB pathway. Drug access includes selective beta-2 AR agonists, alpha-2 adrenergic agonists such as clonidine, and centrally acting sympathomimetics. Tracked by plasma catecholamines, HRV, and blood pressure. Relevant to septic shock, acute respiratory distress, and metabolic syndrome (Sterling, 2012). (Table 2).

3.2 Quantitative clinical evidence

Because NRGDD prioritizes anatomically conserved circuits and human-measurable biomarkers, translational bridging is more direct than in conventional programs. Focusing on circuits that are functionally conserved across species aligns development with human biology from the start (Kandel et al., 2014; Scannell et al., 2012). The four trials below provide the quantitative clinical foundation. Full comparative data are in Table 3.

TABLE 3. Controlled study outcomes supporting NRGDD-relevant neural circuits.

CAP, cholinergic anti-inflammatory pathway; AChE = acetylcholinesterase; VNS, vagus nerve stimulation; DAS28 = Disease Activity Score-28; ACR, american college of rheumatology response; RCT, randomized controlled trial.

Koopman et al. (2016) tested implantable VNS in 17 patients with RA. Active stimulation reduced serum TNF by 39% versus 16% in the sham group at week 12 (p < 0.05). DAS28 improved 1.78 points versus 0.48 in sham. Stopping stimulation at week 12 reversed TNF suppression within 4 weeks, satisfying the disruption criterion.

The RESET-RA trial (Tesser et al., 2025) enrolled 140 patients. ACR20 response was 54.9% in the active VNS arm versus 36.9% in sham (p = 0.021). ACR50 was 31.0% versus 14.1% (p = 0.013). DAS28-CRP fell 1.48 versus 0.72 (p = 0.003). IL-6 dropped 38% versus 12%. These results are comparable to biologic DMARDs in patients who had failed prior treatment.

Fragoso-Saavedra et al. (2022) randomized 188 patients with severe COVID-19 to pyridostigmine or placebo. Twenty-eight-day mortality was 15.8% versus 24.3% (p = 0.048). IL-6 fell 41% versus 12% (p < 0.01) and ferritin fell 38% versus 8% (p < 0.01) at day 14.

Pullan et al. (1994) tested transdermal nicotine in 72 patients with active ulcerative colitis. Clinical remission occurred in 39% versus 9% with placebo (p < 0.001). Mucosal IL-1beta and TNF were significantly reduced on biopsy (p < 0.05), providing direct tissue-level evidence of alpha7 nAChR-mediated anti-inflammatory effects. (Table 3).

4 Formal framework of neuro-regulatory drug discovery

4.1 Distinguishing NRGDD from neuromodulation and placebo effects

A central requirement of NRGDD is clear separation between neuro-initiated regulation and other forms of neural involvement in therapy. Failure to make this distinction has historically blurred the interpretation of neural effects and contributed to skepticism about brain-centered therapeutic strategies. Neuro-initiated regulation refers specifically to therapeutic effects arising from a defined neural circuit actively governing downstream peripheral biology. The neural pathway is the initiating mechanism, not a modifier of effects arising from other causes. General neuromodulation alters neural activity without demonstrating systemic physiological regulation, and many neuromodulation approaches target symptom control without establishing a causal link to peripheral biological changes. Placebo responses can produce measurable physiological changes but are inconsistent across subjects, exhibit high individual variability, and cannot yield reproducible peripheral biomarker profiles. The neural networks activated by placebo phenomena differ substantially from the autonomic and hypothalamic circuits targeted in NRGDD. Table 4 compares the operational criteria distinguishing these categories.

TABLE 4. Operational criteria distinguishing NRGDD from general neuromodulation and placebo effects.

NRGDD, requires demonstrated causality through disruption testing and reproducible peripheral biomarker standards. HRV, heart rate variability.

4.2 Eligibility criteria for neural circuits

Not all neural circuits are suitable for NRGDD development. To qualify, a circuit must meet four criteria. First, it must be anatomically definable: its nodes, connections, and neurotransmitter systems must be identifiable by established neuroanatomical or imaging methods. Second, it must produce measurable output to peripheral systems: activation or inhibition must produce quantifiable effects in immune, metabolic, endocrine, or cardiovascular compartments. Third, it must be pharmacologically accessible through drugs or clinically usable devices. Fourth, both circuit engagement and its downstream effects must be confirmable in human clinical studies using validated biomarkers. Circuits meeting all four criteria are NRGDD-eligible. Table 2 summarizes suitability by circuit class. Figure 2 illustrates the overall causal architecture required by NRGDD.

4.3 Causal inference requirements

Demonstrating causality is the defining scientific requirement of NRGDD. Statistical associations between neural activity and peripheral changes are insufficient. Three converging lines of evidence are required: temporal precedence (neural activation must demonstrably precede the peripheral biomarker change), dose-response concordance (graded engagement must produce proportional peripheral effects), and disruption attenuation (pharmacological blockade, surgical interruption, or receptor antagonism of the neural pathway must abolish or significantly reduce the peripheral response). These requirements mirror established causal inference frameworks in physiology and align with regulatory expectations for mechanism-based claims. Figure 2 illustrates how these requirements fit within the NRGDD causal architecture (Figure 2).

5 Discovery and development workflows

5.1 Translating neural control biology into a discovery logic

A defining strength of NRGDD is converting established principles of neural control biology into an operational discovery workflow. Unlike conventional target-centric programs, which move from molecular screening to phenotypic validation with mechanism inferred after the fact, NRGDD programs proceed through an iterative, causality-driven sequence in which neural engagement and peripheral regulation are evaluated in parallel from the earliest stages. The starting point is the identification of a neural circuit with demonstrated regulatory influence over a disease-relevant peripheral process. This comes from convergent evidence in neuroanatomy, autonomic physiology, and clinical observation, not from molecular target novelty alone. Figure 3 shows the six-phase pipeline and the decision gates at each stage (Figure 3).

5.2 Assay cascades and decision gates

Operationalizing NRGDD requires structured assay cascades that explicitly link neural engagement to peripheral biological response. These cascades increase evidentiary confidence progressively while filtering out candidates that fail causal criteria. Table 5 shows the assay stages, measurements, and gate criteria. Initial assays confirm that the proposed intervention engages the intended neural circuit through electrophysiological recording, imaging-based activation patterns, or autonomic indices such as HRV spectral analysis. Secondary assays measure peripheral biomarkers mechanistically linked to the circuit. Decision gates are embedded at each stage; candidates who fail to demonstrate concordance are stopped, regardless of symptomatic effects. Efficacy signals alone are not sufficient for advancement.

TABLE 5. NRGDD assay cascade and gate criteria.

Each stage must be cleared before the next is initiated. LF/HF, low frequency/high frequency power ratio; fMRI, functional magnetic resonance imaging; CV, coefficient of variation; MLA, methyllycaconitine.

5.3 Causal inference in early development

Causal inference must be built into development early to avoid downstream ambiguity. Conventional pharmacology relies on correlative improvements in disease markers, leaving causal mechanisms for post hoc interpretation. This has contributed to late-stage failures in which the presumed mechanism fails to explain the observed benefit or cannot be reproduced outside the original trial conditions. NRGDD requires explicit demonstration that neural activity is both necessary and sufficient to drive peripheral biological change before advancing to human efficacy trials. This is achieved through study designs that integrate neural engagement assessment, peripheral biomarker measurement, and pathway disruption within a single experimental framework. Such designs mirror classical physiological approaches to establishing causality and align with regulatory expectations for mechanism-based development claims. Figure 4 shows the detailed causal validation sequence (Figure 4).

5.4 Translational bridging

NRGDD is inherently compatible with early human validation. Because the framework prioritizes anatomically conserved circuits and human-measurable biomarkers, translational bridging occurs earlier and more directly than in conventional programs. Preclinical studies focus on confirming the conservation of neural-peripheral coupling rather than exhaustively modeling disease phenotypes. Human exploratory studies then confirm circuit engagement and peripheral regulation using the same biomarker framework established in animals. This approach aligns with regulatory emphasis on human-relevant mechanistic evidence and reduces the risk of late-stage surprises when animal pharmacology does not transfer to human neural architecture (FDA, 2022a). Figure 5 shows how the preclinical, translational, and clinical tracks relate to one another and where gates apply.

5.5 A worked example: causal proof for the vagal-splenic circuit

A worked example illustrates the causal proof sequence. An alpha7 nAChR agonist, such as GTS-21, is administered to animals with adjuvant-induced arthritis at 4 dose levels. HRV spectral analysis shows increased high-frequency power (reflecting parasympathetic activation) within 30 min. Serum TNF falls within 60–90 min of the HRV change, establishing temporal precedence. TNF suppression is proportional across all four dose levels, establishing dose-response concordance. Co-administration of methyllycaconitine (MLA), a selective alpha7 nAChR blocker, restores TNF to vehicle levels despite continued GTS-21 dosing, establishing disruption attenuation. All three criteria are satisfied. The program advances to IND-enabling studies.

6 Clinical development and case studies

6.1 Clinical development strategy

Clinical trials under NRGDD begin with mechanistic confirmation, not symptom endpoints. First-in-human studies must demonstrate both circuit engagement and downstream peripheral biomarker changes using endpoints defined before the trial starts. Crossover designs, graded exposure studies, and disruption protocols (where ethically feasible) are used in early phases. Clinical symptom outcomes serve as supporting rather than primary evidence. This reduces susceptibility to placebo effects and provides mechanistic data to guide dose selection and patient stratification in later phases.

6.2 Case study I: vagus nerve stimulation in rheumatoid arthritis (device-based NRGDD)

VNS in RA is the most complete clinical demonstration of NRGDD principles and establishes proof of concept for the vagal-splenic circuit as a pharmacological target. VNS activates efferent vagal fibers, which release ACh at the celiac ganglion, activating the splenic nerve through norepinephrine. Splenic nerve activity stimulates ChAT-expressing T cells to produce ACh, which binds alpha7 nAChRs on macrophages, blocks NF-kappaB, and reduces TNF, IL-6, and HMGB1. Because the circuit, the molecular target, and the peripheral readout are all defined, VNS data provide the mechanistic template for pharmacological NRGDD programs that access the same circuit through drugs. Table 6 summarizes the NRGDD validation evidence for this case.

TABLE 6. VNS in RA: device-based NRGDD with full mechanistic validation.

ACh, acetylcholine; NE, norepinephrine; ChAT, choline acetyltransferase; JAK2 = Janus kinase 2; STAT3 = signal transducer and activator of transcription 3; NF-kappaB = nuclear factor kappa-light-chain-enhancer of activated B cells.

Koopman et al. (2016): TNF fell 39% versus 16% in sham (p < 0.05); DAS28 improved 1.78 versus 0.48 points. Tesser et al. (2025) RESET-RA: ACR20 was 54.9% versus 36.9% (p = 0.021); DAS28-CRP fell 1.48 versus 0.72 (p = 0.003); IL-6 dropped 38% versus 12%. Stopping stimulation reversed TNF suppression within 4 weeks, satisfying the disruption criterion (Table 6).

6.3 Case study II: Pyridostigmine in COVID-19 (Pharmacologic NRGDD)

Pharmacological interventions qualify as NRGDD-aligned when their primary effect is to augment endogenous neural regulatory pathways. Pyridostigmine is an acetylcholinesterase (AChE) inhibitor. It prevents ACh breakdown at cholinergic synapses, increases vagal tone, and prolongs alpha7 nAChR activation on macrophages. This amplifies the endogenous CAP without surgery or device implantation. By inhibiting AChE at vagal terminals and ganglionic synapses, pyridostigmine raises synaptic ACh throughout the vagal-splenic axis, activates alpha7 nAChRs on splenic and pulmonary macrophages, and reduces NF-kappaB-driven cytokine production in the context of COVID-19 hyperinflammation.

Fragoso-Saavedra et al. (2022) RCT, n = 188: 28-day mortality was 15.8% versus 24.3% in controls (p = 0.048). IL-6 at day 14 fell 41% versus 12% (p < 0.01). Ferritin fell 38% versus 8% (p < 0.01). Circuit specificity is moderate, as systemic AChE inhibition also enhances muscarinic signaling at non-immune sites. This limitation is acknowledged in the NRGDD framework and can be resolved in future programs using selective α7 nAChR agonists (Fragoso-Saavedra et al., 2020).

6.4 Case study III: Transdermal nicotine in ulcerative colitis (peripheral Alpha7 nAChR NRGDD)

Transdermal nicotine presents the most operationally complex of the three case studies and requires the definitional clarification introduced in Section 2.1. Nicotine activates alpha7 nAChRs on colonic epithelial cells and lamina propria macrophages directly, without requiring intact efferent vagal signaling. The alpha7 nAChR at this location is the established terminal effector of the cholinergic anti-inflammatory pathway in the gut: it has been characterized neuroanatomically as the endpoint of vagal efferent signaling, and its pharmacological blockade with methyllycaconitine abolishes the mucosal anti-inflammatory effects of both vagal stimulation and cholinergic agonists in preclinical models (Ghia et al., 2006). Nicotine therefore engages the circuit at its terminal receptor using a mechanism that is mechanistically equivalent to upstream neural activation, not through an independent peripheral pharmacological pathway unrelated to neural control. This distinguishes it from a direct cytokine inhibitor that happens to reduce TNF through an entirely different mechanism. The case qualifies as terminal receptor NRGDD but not as proximal circuit NRGDD. Although the mechanism bypasses the central efferent vagal component, this case represents pharmacological engagement at the defined terminal effector of a characterized neural control pathway.

Pullan et al. (1994) RCT, n = 72: clinical remission occurred in 39% versus 9% with placebo (p < 0.001). Mucosal IL-1beta and TNF were significantly reduced on biopsy in the nicotine arm (p < 0.05). This provides direct tissue-level evidence of the mechanism. Circuit engagement was not directly measured at the central neural level in this trial, which the framework records as a limitation of this case study. Future studies in this indication should include HRV monitoring and, where feasible, vagal engagement confirmation to establish whether central circuit contribution is present alongside the peripheral receptor effect. Table 7 compares all three case studies across the key NRGDD validation dimensions.

TABLE 7. Comparative analysis of NRGDD case studies.

RCT, randomized controlled trial; VNS, vagus nerve stimulation; CAP, cholinergic anti-inflammatory pathway; AChE = acetylcholinesterase; DAS28 = Disease Activity Score-28.

6.5 Evidential breadth: NRGDD across circuit classes

The three primary case studies draw on a common biological substrate—the cholinergic anti-inflammatory pathway and its terminal effector, the alpha7 nAChR. The concentration of examples within this circuit class reflects the maturity and rigor of available clinical evidence in that domain, not a limitation of NRGDD’s scope. The framework specifies three eligible circuit classes beyond the CAP (Table 2): the HPA axis, the sympathetic adrenomedullary axis, and brainstem autonomic nuclei. Each has independent clinical and translational evidence consistent with NRGDD criteria, though none has yet been developed through the full NRGDD pipeline.

HPA Axis Evidence. CRH receptor-1 antagonists have been evaluated in clinical trials for major depression and stress-related disorders. Pexacerfont reduced ACTH and cortisol responses to social stress in healthy volunteers in a dose-dependent manner (Coric et al., 2010). The quantitative cortisol-ACTH biomarker response and dose-response concordance satisfy NRGDD causal criteria, even though those programs were not framed in circuit terms. In metabolic syndrome, Dallman et al. (2003) demonstrated that chronic HPA axis dysregulation drives visceral adiposity, dyslipidemia, and insulin resistance through glucocorticoid receptor signaling on adipose and hepatic tissue—a peripheral regulatory output measurable by standard clinical biomarkers (cortisol, insulin, HOMA-IR). HPA-targeting NRGDD programs have a fully defined circuit, validated human biomarkers, and pharmacologically accessible nodes (NIH, 2020).

Sympathetic Adrenomedullary Evidence. Beta-2 adrenergic agonists suppress TNF transcription via the cAMP-PKA-CREB pathway in macrophages and reduce IL-6 in sepsis models (Severn et al., 1992). Clonidine, a centrally acting alpha-2 adrenergic agonist, reduces sympathetic outflow and has demonstrated anti-inflammatory and metabolic effects in surgical stress and critical illness (Inzucchi and Sherwin, 2005). These interventions engage the sympathetic adrenomedullary circuit at defined nodes and produce peripheral biomarker changes consistent with NRGDD criteria, supporting the framework’s generalizability beyond the vagal-cholinergic axis. Taken together, the HPA and sympathetic adrenomedullary evidence confirms that NRGDD is not a reformulation of cholinergic pharmacology but a general discovery logic applicable across the neural regulatory architecture wherever a circuit meets the four eligibility criteria defined in Section 4.2.

7 Boundary conditions, limitations, and failure modes

7.1 Diseases outside NRGDD scope

NRGDD modulates dysregulated physiological control systems. It is not designed for conditions in which tissue is irreversibly destroyed, a cell population is absent, or a fixed genetic defect prevents the expression of an essential enzyme. End-stage neurodegeneration, congenital metabolic disorders, and advanced solid tumors requiring cytotoxic therapy fall outside its core scope. Neural redundancy also creates a practical boundary: parallel pathways may restore pathological states even when the primary circuit is successfully modulated, which can buffer therapeutic effects and requires robust biomarker strategies and adaptive trial designs.

In oncology, NRGDD may serve as an adjunct through neural regulation of immune surveillance, tumor microenvironment metabolism, cancer-associated inflammation, cachexia, and treatment-related immune dysfunction. It cannot substitute for primary antitumor therapy in established malignancies. The boundary conditions are summarized visually in Figure 1 at the end of Section 10.

7.2 Limitations of the NRGDD framework

Three limitations warrant substantive treatment because they bear directly on the framework’s translational credibility: biomarker variability, circuit redundancy, and translational limitations. Additional failure modes and their mitigations are catalogued in Table 8.

TABLE 8. Common failure modes in NRGDD programs and mitigation strategies. Most arise from methodological gaps rather than conceptual limitations.

. IND, investigational new drug application; HRV, heart rate variability.

Biomarker Variability. The peripheral biomarkers central to NRGDD—serum TNF, IL-6, HMGB1, HRV spectral indices, and cortisol—exhibit substantial intra-individual and inter-individual variability that can obscure circuit-level effects. Serum TNF has a coefficient of variation exceeding 40% in healthy subjects across repeated same-day measurements, and HRV indices show circadian fluctuations of 15%–30% depending on the measurement protocol (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). In inflammatory disease, baseline cytokine levels vary further depending on disease activity, comorbidities, concomitant medications, and the time of sampling. This variability has two direct consequences: it inflates the sample sizes required to detect meaningful circuit-level changes, and it risks misclassifying non-responders as biomarker failures. The NRGDD framework addresses this through two design requirements. First, biomarker assays must be prespecified as mechanistically linked to the target circuit, rather than selected from broad inflammatory panels post hoc. Second, the gate criteria in Table 5—requiring a minimum 30% change from vehicle at the peripheral response stage and a coefficient of variation below 30% in the translational bridging cohort—enforce quantitative standards that discourage underpowered or loosely defined biomarker strategies. Stratification by baseline cytokine tertile and baseline HRV phenotype at enrollment further reduces within-group variability. These requirements structure development around circuits whose signal-to-noise characteristics are known and manageable, but they do not eliminate biomarker noise as a practical challenge in early-phase programs.

Circuit Redundancy. Peripheral physiological states are regulated by overlapping neural circuits with partially redundant outputs. The gut receives cholinergic anti-inflammatory input via vagal efferents and enteric neurons, as well as via adrenergic sympathetic fibers and local neuropeptide-mediated pathways. Modulating a single circuit node may be buffered by compensatory upregulation via parallel routes, attenuating or masking the peripheral response. Redundancy is most problematic in chronic disease, where long-standing pathology may have reorganized regulatory architecture and shifted reliance toward compensatory circuits (Sterling, 2012). The practical consequence is that successful circuit engagement—confirmed by electrophysiology or HRV—may produce a smaller-than-predicted peripheral biomarker change because redundant circuits partially restore the pathological state. NRGDD does not yet have a systematic solution to circuit redundancy, and this represents a genuine conceptual limitation rather than a methodological gap alone. Near-term strategies include multi-modal circuit engagement that targets redundant nodes in parallel, adaptive trial designs that allow dose titration guided by biomarker response, and patient selection based on autonomic phenotyping to identify individuals with limited compensatory capacity. Multi-circuit NRGDD programs targeting the CAP and sympathetic adrenomedullary axis simultaneously represent a longer-term developmental strategy warranting prospective investigation.

Translational Limitations. The conservation of neural-peripheral coupling across species is a stated assumption of the NRGDD framework, but the degree of conservation varies by circuit and by disease context. The alpha7 nAChR-macrophage interface is well conserved between rodents and humans (Pavlov and Tracey, 2017), but HPA axis responsiveness to CRH receptor modulators shows species differences in receptor subtype distribution and feedback kinetics that have contributed to failed translation of CRH-R1 antagonists from animal efficacy models to human clinical trials (Binneman et al., 2008). The sympathetic adrenomedullary circuit shows differences in catecholamine ratios and adrenergic receptor density between rodent and human immune cells that may affect the magnitude of anti-inflammatory effects. NRGDD requires translational bridging studies in human exploratory cohorts before Phase 2, specifically to confirm whether the circuit-biomarker relationship established in animals is reproduced in the human context. This requirement is built into Phase 5 of the pipeline (Table 5) and is a structural response to translational uncertainty rather than an assumption of automatic cross-species conservation. Where species differences in circuit architecture are known, the framework requires that they be addressed before IND submission rather than encountered at Phase 2 failure. Beyond species differences, patient population heterogeneity in autonomic tone—measured by resting HRV—poses a distinct translational challenge: patients with very low basal HRV, common in metabolic syndrome and established inflammatory disease, may exhibit attenuated circuit responses even with confirmed receptor engagement, requiring adaptation of dosing strategies and stratified enrollment criteria.

Table 8 lists the most common failure modes.

8 Future directions and technological integration

The future of NRGDD depends on several enabling technologies. Advances in bioelectronic medicine, including miniaturized implantable devices and closed-loop stimulation systems, expand the precision with which neural circuits can be modulated and enable adaptive dosing based on real-time physiological feedback. Computational modeling and AI offer opportunities to integrate multivariate datasets from autonomic monitoring, blood biomarkers, and imaging, though such tools must remain anchored to biological plausibility and causal validation standards. Figure 6 illustrates how these technologies relate to the NRGDD framework.

Functional neuroimaging can show in real time whether a drug has activated the intended circuit. Wearable devices that track HRV over days or weeks allow circuit engagement monitoring in ambulatory settings. Closed-loop bioelectronic devices that adjust stimulation based on measured biomarker feedback serve as precision research tools for mapping circuit-biomarker relationships and guiding drug development programs.

The most promising near-term pharmacological NRGDD targets are selective alpha7 nAChR partial agonists that avoid receptor desensitization (TC-5619, ABT-107), CRH receptor-1 blockers that normalize HPA axis tone without broadly suppressing immunity, and beta-2 AR agonists with defined circuit-level access. Bispecific agents that engage two nodes of the same neural-peripheral circuit may offer additive effects. AI-assisted multiomics integration may identify novel neuro-regulatory axes and predict individual circuit responsiveness from baseline autonomic and inflammatory phenotypes, enabling precision patient selection for NRGDD clinical trials (Figure 6).

9 Regulatory and ethical considerations

9.1 Regulatory pathways

NRGDD does not require a new regulatory pathway. Pharmacological NRGDD drugs follow the standard FDA IND and NDA routes for small molecules or biologics. Device-based NRGDD programs use the FDA’s De Novo or PMA pathways, or combination product pathways when a drug and device are used together. For combination programs, the FDA’s Office of Combination Products (OCP) should be engaged to define the primary mode of action and the appropriate review division. The precedent established by approved implantable 5NS devices for epilepsy and depression provides a regulatory framework that pharmacological NRGDD programs can reference.

NRGDD’s biomarker-anchored mechanistic framework aligns directly with the FDA’s emphasis on drug development tools (DDTs), biomarker qualification programs, and pharmacodynamic biomarkers that support mechanism-of-action claims. The requirement that a drug’s therapeutic effect be shown to arise from neural circuit engagement rather than direct peripheral receptor binding poses a new challenge for characterization. Sponsors should engage the FDA’s Division of Pharmacometrics and Biomarker Qualification Program early to establish qualification plans for circuit engagement endpoints, such as HRV as a pharmacodynamic measure of vagal tone, and for peripheral biomarkers, such as serum TNF, as a surrogate for CAP activation (FDA, 2022b; 2023). Clear articulation of causal pathways and evidentiary sufficiency will be central to regulatory acceptance.

9.2 Ethical considerations

Drugs that act on autonomic circuits or the HPA axis may produce changes in mood, cognition, or behavior in addition to anti-inflammatory or metabolic effects. Informed consent in NRGDD trials must describe the neural mechanism and explain the possibility of central nervous system effects. Participant selection criteria should account for individual differences in autonomic tone and baseline neuroimmune function. Long-term safety monitoring should include standardized assessments of autonomic function, mood, and cognitive performance alongside traditional safety endpoints. As neuro-regulatory therapies mature, regulatory and ethical frameworks must evolve accordingly, with particular attention to autonomy, long-term behavioral safety, and equitable access.

10 Conclusion

Drug development has achieved extraordinary molecular precision while systematically underusing the brain’s capacity to regulate peripheral physiology. The 90% attrition rate in clinical development reflects, in part, a mismatch between reductionist, target-based discovery logic and the integrative regulatory architecture that governs peripheral organ function. NRGDD addresses this mismatch directly. Neural circuits serve as the primary therapeutic targets, peripheral biomarkers serve as causal validation outputs, and circuit engagement rather than receptor occupancy serves as the primary pharmacodynamic anchor.

Three lines of convergent clinical evidence support the scientific foundation: device-based VNS producing quantifiable, causally validated suppression of TNF and IL-6 in treatment-refractory RA (TNF -39%, ACR20 55%, disruption-attenuation confirmed); pharmacologic AChE inhibition with pyridostigmine reducing COVID-19 mortality by 35% relative and IL-6 by 41%; and peripheral alpha7 nAChR modulation with nicotine achieving 39% remission in ulcerative colitis with mucosal cytokine reduction. All three work through the same fundamental neural regulatory circuit, demonstrating that NRGDD principles can be implemented with clinically meaningful benefit across pharmacological and device-based modalities in multiple disease contexts.

NRGDD complements rather than replaces conventional pharmacology. It identifies contexts where neural initiation provides superior access to complex systemic disease processes. When applied with circuit-specific assay cascades, causal disruption experiments enforced as decision gates, quantitative biomarker standards, and human-relevant translational bridging, NRGDD has the potential to identify a class of therapeutics that operates in harmony with the brain’s endogenous regulatory architecture, producing durable, multi-effector benefits that individual molecular inhibitors cannot replicate. Future research should test whether pharmacological precision engagement of alpha7 nAChR, CRH receptor-1, and sympathetic adrenomedullary circuits yields drug candidates that satisfy the full NRGDD causal criteria in Phase 1 experimental medicine studies. That is a directly testable prediction of the framework proposed here.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Mark Dallas, University of Reading, United Kingdom

Reviewed by: Rie Hasebe, National Institute for Physiological Sciences (NIPS), Japan

Joelson Crispim, Universidade Federal de Pernambuco, Brazil