Acute Corticotropin-Releasing Factor Receptor Type 2 Agonism Results in Sustained Symptom Improvement in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome

Background Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a complex multi-symptom disease with widespread evidence of disrupted systems. The authors hypothesize that it is caused by the upregulation of the corticotropin-releasing factor receptor type 2 (CRFR2) in the raphé nuclei and limbic system, which impairs the ability to maintain homeostasis. The authors propose utilizing agonist-mediated receptor endocytosis to downregulate CRFR2. Materials and Methods This open-label trial tested the safety, tolerability and efficacy of an acute dose of CT38s (a short-lived, CRFR2-selective agonist, with no known off-target activity) in 14 ME/CFS patients. CT38s was subcutaneously-infused at one of four dose-levels (i.e., infusion rates of 0.01, 0.03, 0.06, and 0.20 μg/kg/h), for a maximum of 10.5 h. Effect was measured as the pre-/post-treatment change in the mean 28-day total daily symptom score (TDSS), which aggregated 13 individual patient-reported symptoms. Results ME/CFS patients were significantly more sensitive to the transient hemodynamic effects of CRFR2 stimulation than healthy subjects in a prior trial, supporting the hypothesized CRFR2 upregulation. Adverse events were generally mild, resolved without intervention, and difficult to distinguish from ME/CFS symptoms, supporting a CRFR2 role in the disease. The acute dose of CT38s was associated with an improvement in mean TDSS that was sustained (over at least 28 days post-treatment) and correlated with both total exposure and pre-treatment symptom severity. At an infusion rate of 0.03 μg/kg/h, mean TDSS improved by −7.5 ± 1.9 (or −25.7%, p = 0.009), with all monitored symptoms improving. Conclusion The trial supports the hypothesis that CRFR2 is upregulated in ME/CFS, and that acute CRFR2 agonism may be a viable treatment approach warranting further study. Clinical Trial Registration ClinicalTrials.gov, identifier NCT03613129.


Etiological Hypothesis
The authors hypothesize that these abnormalities could all originate from a single pathway, involving the corticotropinreleasing factor (CRF, also known as corticotropin-releasing hormone or CRH) system and its regulation of serotonin (5HT) in the limbic system, which controls the response to homeostatic threat (Dedic et al., 2018;Deussing and Chen, 2018;Godoy et al., 2018), usually termed "stress, " but avoided here to preclude narrow connotations. This hypothesis synthesizes numerous independent in vivo studies to propose the mechanisms of homeostasis. It rests on three constructs (Figure 1).

Therapeutic Approach
If upregulated CRFR2 causes ME/CFS, the resulting inability to maintain homeostasis under dynamic threat cannot be repaired by static approaches (e.g., fixed-doses of CRF/5HT antibodies, CRFR2 antagonists or GABA agonists). This leaves CRFR2 downregulation as the most reasonable approach.
In vivo, subcutaneous CT38s, a short-lived CRFR2-selective agonist with no known off-target activity (Supplementary Material), induces dose-dependent changes in norepinephrine and corticosterone release, spontaneous movement (possibly motor effect), gastrointestinal transit, urine volume, respiratory minute volume, core body temperature, heart rate (HR) and mean arterial pressure (MAP) (Supplementary Figures 1A-I).
That is, CRFR2 stimulation in healthy rats produces signs analogous to complaints of ME/CFS. These data also suggest that CT38s enters the central nervous system as: norepinephrine/corticosterone involve the hypothalamus (Dedic et al., 2018;Deussing and Chen, 2018;Godoy et al., 2018); respiration and core body temperature involve 5HT in the medullary respiratory neurons (Hilaire et al., 2010) and preoptic area of the anterior hypothalamus (Lin et al., 1998;Boulant, 2000), and consistent with CRFR2 stimulation elevating 5HT to decrease respiratory function and temperature (Supplementary Figures 1F-G); and HR involves CRFR1 and CRFR2 in the BNST (Oliveira et al., 2015). Escalating doses cause the HR and MAP responses to peak and then decrease, notably at lower concentrations or exposures by infusion than by bolus ( Supplementary Figures 2A-D). Since agonist-mediated receptor endocytosis increases with agonist concentration and the duration of stimulation (Markovic et al., 2008(Markovic et al., , 2011Hauger et al., 2013), this apparent loss of HR and MAP sensitivity plausibly resulted from CRFR2 endocytosis in limbic neurons, occurring at lower concentrations by infusion due to the additive effect of time. Thus, the authors propose utilizing agonist-mediated CRFR2 endocytosis to treat ME/CFS.

InTiME
InTiME Investigated the safety and efficacy of subcutaneouslydosed CT38s in ME/CFS patients. This open-label trial was conducted at the Bateman Horne Center, under a physiciansponsored investigational new drug application filed with the United States Food and Drug Administration (FDA), registered at ClinicalTrials.gov (NCT03613129), approved by Aspire IRB, in compliance with the Declaration of Helsinki and Good Clinical Practice, with all patients providing informed consent.
Aims InTiME sought to show that an acute dose of subcutaneous CT38s in ME/CFS patients, could safely induce sustained symptom improvement, determined by comparing symptoms (see below) in the 28-day pre-treatment assessment period and the 28-day post-treatment assessment period (Figure 2). CRFR2 expression in the raphé nuclei and limbic system is neuronally-specific and adapts in realtime so cannot be measured. However, the observation of dose-dependent effect would link CRFR2 with ME/CFS, as CT38 is CRFR2-selective and has no off-target activity. Sustained effect would suggest CRFR2 endocytosis had occurred, as CT38 does not persist in rats or healthy humans, evidenced by the rapid normalization of induced HR increases (Supplementary Material), which are CRFR1/CRFR2-mediated in the BNST (Oliveira et al., 2015).

Patients
InTiME included 18-60 year-old, male or female patients, meeting the Fukuda, Canadian and National Academy of Medicine criteria for ME/CFS, living between 3,500 and 5,500 feet above sea level (related to cardio-pulmonary exercise testing, CPET), with a stable state of illness in the prior 3 months, i.e., absence of active or uncontrolled co-morbidities including infections or depression. InTiME excluded patients with untreated endocrine diagnoses, tachycardia, severe hypotension, renal impairment, or who were taking anti-retrovirals, short-term antivirals/antibiotics, rituximab, or medications interfering with 5HT, norepinephrine, dopamine or cortisol.

Intervention
The drug substance, CT38s, is the acetate salt of CT38 (free base), a custom, 40-amino acid peptide. It is a potent CRFR2-selective agonist (EC50 nmol/% of Emax: 17.1/100), with no known offtarget activity. The drug product was supplied in active (CT38s: 1 mg/ml) and diluent (vehicle: 0.05 M TRIS buffer, 0.67% NaCl in sterile H 2 O for injection, USP, pH 7.5-7.7) vials, requiring on-site dilution. The drug was delivered subcutaneously via programmable syringe pump (McKinley TM T34) utilizing a soft cannula infusion set (Neria TM Soft 90). Pharmacokinetic (PK) data are expressed in terms of CT38. CT38s has been studied in animals and healthy human subjects in a prior Phase 1 clinical trial (Supplementary Material), but data is not publicly available as it is part of an ongoing drug development program.

Dosing
The acute dose was intended to reproduce the reduced HR sensitivity noted in healthy rats by infusion, where apparent endocytosis occurred at a total exposure (i.e., area under the plasma concentration-time curve or AUC) of ∼40 ng h/ml and a plasma concentration of at least ∼1.50 ng/ml, human equivalents of 7 ng h/ml and 1.40 ng/ml, respectively FIGURE 2 | InTiME schema. Figure 2). In an infusion, the maximum plasma concentration (Cmax) and AUC are governed by the rate and duration of infusion. Thus, the starting dose-level was planned as a 3-h treatment, involving a priming bolus of 0.15 µg/kg, and a continuous infusion at a rate (in µg/kg/h) of 0.20 for 45 min, escalating to 0.22 for 45 min, then escalating to 0.24 for 90 min-dose: 0.825 µg/kg (below the maximum tolerated bolus dose in the prior Phase 1 trial of 0.833 µg/kg, Supplementary Material); projected Cmax: 1.37 ng/ml (below the maximum tolerated concentration of 1.56 ng/ml in the prior Phase 1 trial, Supplementary Material); projected AUC: 4.16 ng h/ml. This was to be repeated at a second treatment to provide a total AUC of 8.33 ng h/ml. It was assumed that any achieved CRFR2 endocytosis would persist, so the number of treatments and their separation were not critical (generally planned as two treatments separated by at least two days, but varied due to dosing changes, see below). Assuming safety at this starting dose-level, InTiME planned two higher dose-levels in subsequent patients.

Dosing Changes
At the first treatment, the patient (ID24) experienced higher than anticipated hemodynamic changes. Since these changes were Cmax-related in rats and healthy subjects (Supplementary Material), the priming bolus and infusion rate escalations were eliminated, and the infusion duration increased to 3.5 h (referred to as D20, meaning 0.20 µg/kg/h), then utilized for ID24's second treatment and both of ID23's treatments. Both patients exhibited poor tolerability to D20. This necessitated reducing the dose-level to D03 (0.03 µg/kg/h) to lower the Cmax and adding a third treatment to increase the total AUC. To characterize the doseresponse more fully, two additional dose-levels, D06 and D01, were also tested (Figure 3)-all approved by Aspire IRB.

Blind Dose
Though open-label, patients were unaware of their relative doses. The first two patients (D20) expected the lowest dose, but received the highest. The next three patients (D03) knew that for safety reasons their dose was lower than the first two patients. The remaining patients did not know their relative dose-levels, and these were not administered sequentially. Patients had no contact with one another.

Outcomes
The primary endpoint was the change in the mean total daily symptom score (TDSS), averaged over 28 days before the first treatment (TDSS pre ) and 28 days before exit from the trial (TDSS post ). The TDSS summed 13 individual symptoms (specifically, fatigue, muscle/joint pain, sleep issues, cognitive impairment, OI, abnormal temperature sensations, flu-like symptoms, headaches or sensory sensitivities, shortness of breath, gastrointestinal function, urogenital function, anxiety and depression), each patient-reported daily throughout the trial, on a 0-5 scale (0 = none, 1 = very mild, 2 = mild, 3 = moderate, 4 = severe, and 5 = very severe). The TDSS modifies the CFS Symptom Inventory (Wagner et al., 2005), expanding some symptoms (e.g., "sensory sensitivities" instead of "sensitivity to light"), collapsing others (e.g.,"abnormal temperature sensations" instead of "fever" and "chills") and utilizing a single 24-h score (instead of intensity and frequency, considered too cumbersome for daily use).
Secondary endpoints included: (i) general health, assessed via the 36-Item Short Form Survey Instrument (SF-36, 0-100 scale, 0 = maximum disability, 100 = no disability) (Ware et al., 1994), completed at enrollment, prior to the start of treatment, and at exit, each referencing the preceding 4 weeks; (ii) Fitbit TM metrics (activity, HR, and sleep), continuously monitored throughout the trial; and (iii) patient-reported daily assessments of: (a) completion of activities of daily living (ADL, 0 = not at all, 1 = a little, 2 = some, 3 = a lot, and 4 = completely); (b) avoidance of physical/mental exertion over PEM concerns (same as ADL scale); and (c) perceived level of physical/mental exertion in the prior 24 h (same as individual symptom scale). The original protocol included daily cognitive testing and CPET (both pre-and post-treatment). Both were eliminated, as the former showed evidence of patient learning before treatment, and the latter because seven patients had already undergone their post-treatment CPET before the third treatment was approved (i.e., after receiving only a fraction of the target AUC). Safety measures included a blood test at enrollment (for blood chemistry, complete blood count and estimated glomerular filtration rate), and urine tests at enrollment and before treatment (for illicit drugs and pregnancy). During treatment, HR, systolic (sBP) and diastolic (dBP) blood pressure were recorded just before dosing (baseline), every 15-min thereafter, and for at least 90 min post-treatment. Dosing was to be stopped in a given patient, if HR > 120 bpm (or < 45 bpm), sBP < 90 mmHg or dBP < 50 mmHg; or if sBP or dBP decreased by more than 20 mmHg or 15 mmHg from baseline on three consecutive readings, respectively. PK blood samples were obtained at intervals before, during and after treatment. The principal investigator (PI) and site staff were responsible for soliciting, recording and reporting events qualifying as adverse events (AEs) and serious adverse events (SAEs), which were followed until resolution/stabilization.

Statistics
InTiME was an exploratory study. All data are reported as mean and standard deviation, unless otherwise noted. Pre-/post-treatment data were compared by Student's t-test. Relative sensitivity to CT38 among patients and healthy subjects were assessed by Kolmogorov-Smirnov test. Correlations were assessed by the Pearson product-moment correlation coefficient. No correction was applied for missing data, which was minimal (93.5% compliance). Study data were collected and managed using REDCap R electronic data capture tools (Harris et al., 2009(Harris et al., , 2019. Data collection was completed in April 2019.

Patient Disposition
Between July 2018 and April 2019, 17 patients were consented and enrolled. Of these, two were screen failures, one voluntarily withdrew, and 14 received CT38 treatment at one of four doselevels (Figure 3).
There were no study discontinuations. Two patients discontinued study drug, but remained in the study until exit: (i) ID35 (D01) received the first treatment, but experienced headache, facial numbness, body flushing, dyspnea, dizziness and swollen lymph nodes in the days following, so the PI decided to avoid further treatment; and (ii) ID34 (D06) received two treatments, but was noted to have poor venous access (necessary for blood sampling and safety, in the event of hypotension), so the PI decided to forego the third treatment.
There were three dose-related protocol deviations. ID23's first treatment and ID29's second treatment leaked at the cannula (observed by site staff and confirmed by PK). In addition, ID29's first treatment was only 2.5 h (daylight savings time error), so ID29 was given a fourth treatment. Age, Age_Onset and Age_Diagnosis indicate mean ± standard deviation.

Patient Demographics
The trial population was reasonably represented in sex (six male, eight female), age (mean: 43.9 years, range: 29.4-59.7 years), disease onset (eight gradual, six sudden), triggers (13 infectious, 4 toxins, 2 over-exertion, 4 emotional-some patients recorded multiple triggers), and disease duration (mean: 13.0 years, range: 2.1-25.0 years) ( Table 1). Patients' symptoms were heterogeneous (with 6 of 13 individual symptoms indicated as the worst, and 11 indicated among the worst 3), and of mild to moderate severity (TDSS pre range: 13.8 to 44.7, Table 2). Actual CT38s dose varied due to changes in individual treatments (Figure 3) and drug preparation (which added a small drawn volume of active solution to a diluent). Thus, PK parameters were calculated for individual treatments, yielding mean Cmax and total AUC ( Table 2).

CRFR2 Sensitivity
Relative to healthy subjects in the prior Phase 1 trial (Supplementary Material), ME/CFS patients were objectively more sensitive to the hemodynamic effects of CRFR2 stimulation during treatment. For a given CT38 concentration, the hemodynamic effects were greater in ME/CFS patients than in healthy subjects, and this differential diminished with increasing concentration, e.g., CT38 Cmax of 0.3 and 0.8 ng/ml result in mean HR increases in patients that are, respectively, 1.8× and 1.6× the corresponding change in healthy subjects Mean Cmax averages across individual treatments; Total AUC sums individual treatments; = pre-/post-treatment change in 28-day means of: total daily symptom score (TDSS); ADL = the extent to which activities of daily living were completed; PEM = the extent to which activities were avoided for fear of inducing PEM; Exertion = the perceived level of exertion; and Steps = Fitbit TM -recorded level of activity (steps). TDSS data indicated as mean ± standard deviation. *Ignores leaked treatment.
Frontiers in Systems Neuroscience | www.frontiersin.org ( Figures 4A-D). Hemodynamic sensitivity was significantly different between ME/CFS patients and healthy controls, with their cumulative distributions for maximum hemodynamic change (HR or dBP) per unit of PK parameter (Cmax or AUC) differing between individual patient treatments (n = 35, InTiME) and individual healthy subject doses (n = 47, prior Phase 1) by Kolmogorov-Smirnov test (Figure 4).

Biphasic Dose-Response
The pre-/post-treatment change in 28-day mean TDSS for individual patients was statistically significant in 11 of 14 patients ( Figure 5). CT38 effect appeared to be biphasic, with symptoms improving at D01 and D03, but worsening at D20. At D06, symptoms were between D03 and D20. The AUC for ID27 (D03) exceeded that of ID34 and ID36 (D06), suggesting that this biphasic character might be driven by concentration rather than AUC.
Subgroup analysis for sex, age, illness duration, triggers and Fitbit TM -recorded sleep and HR data, did not yield treatment-related insights, but this study was not powered to determine such effects.

Safety
There were no deaths. There was one study drug discontinuation (ID35, D01) due to symptom worsening (headache, facial numbness, dyspnea, dizziness and swollen lymph nodes) in the days following treatment, which did not require intervention.
There was one SAE during ID23's (D20) second treatment (first treatment leaked). The patient experienced tachycardia and hypotension (baseline: 92 bpm, 108/74 mmHg; peak: 125 bpm, 89/50 mmHg), recoded as two severe treatmentemergent AEs (TEAEs). The patient required rescue saline, but recovered and remained in the trial until planned exit. The SAE was reported to the FDA and IRB. It resulted from a poor prediction of the starting dose (derived from healthy animals/humans without CRFR2 upregulation), and an inadvertent continuation of dosing after the dose-stopping criteria were met. The patient also recorded severe fatigue after each treatment, so four severe TEAEs in total.

DISCUSSION
InTiME is the first study to identify that a CRFR2-selective agonist may provide therapeutic benefit in ME/CFS patientsnine of 10 patients where Cmax did not exceed 0.25 ng/ml, showed significant, sustained TDSS improvement ranging from −1.6 to −16.0, dependent on both AUC and pre-treatment severity. This work hypothesizes that ME/CFS is caused by CRFR2 upregulation in the raphé nuclei and limbic system, observed in vivo (Waselus et al., 2009;Lebow et al., 2012;Wood et al., 2013), with symptoms explained by the known effects of CRFR2 activation on this subset of 5HT neurons (Waselus et al., 2005;Kirby et al., 2008;Lukkes et al., 2008). It proposes that since CRFR2 undergoes agonist-mediated endocytosis (Markovic et al., 2008(Markovic et al., , 2011Reyes et al., 2008Reyes et al., , 2014Hauger et al., 2013), treatment with an agonist may downregulate membrane-bound CRFR2. The trial results support these ideas.

Biphasic Dose-Response
InTiME sought to invoke agonist-mediated CRFR2 endocytosis, which is considered protective against overstimulation and known to increase with agonist concentration and duration of stimulation (Markovic et al., 2008(Markovic et al., , 2011Hauger et al., 2013). The sustained symptom improvement over at least 28 days, with a peptide whose half-life is 1.5 h, suggests that endocytosis occurred, but surprisingly, only at low CT38 dose-levels (D01 and D03). Why might this be the case? The threat response is likely terminated by UCN1mediated CRFR2 endocytosis (Markovic et al., 2008(Markovic et al., , 2011Hauger et al., 2013). Like other G protein-coupled receptors, CRFR2 endocytosis is mediated by β-arrestin, which was thought to be recruited by activated G proteins, so requiring agonist concentrations above the threshold at which the particular agonist activates the G proteins. Recently, however, G protein-independent β-arrestin recruitment has been observed with another G protein-coupled receptor (Pack et al., 2018), and may even accelerate β-arrestin recruitment (Markovic et al., 2011).
This notion may explain the biphasic dose-response. CT38 displaces CRF (respective CRFR2 binding affinities: 1.1 and 44.5 nmol), and activates G proteins at a threshold concentration of ∼0.25 ng/ml, i.e., the lowest Cmax at which HR increased in healthy subjects in the prior Phase 1 trial (Figure 4A and Supplementary Material). Thus, for mean Cmax < 0.25 ng/ml (D01 and D03), G proteins did not activate, and symptom improvement (Figure 5) likely resulted from G proteinindependent CRFR2 endocytosis, as effect was: (i) sustained long after drug clearance; (ii) AUC-dependent, consistent with endocytosis preventing overstimulation ( Figure 7A); and (iii) indirectly dependent on symptom severity, also consistent with overstimulation, since if CRFR2 drives symptoms, then mild patients with only mild CRFR2 stimulation, will require relatively more treatment to achieve CRFR2 overstimulation ( Figure 7A). For mean Cmax > 0.25 ng/ml, G proteins activated during treatment (early for D20; late for D06), and sustained symptom worsening (Figure 5) suggests CRFR2 upregulation (Figure 7B), arguably demonstrating PEM ( Figure 6B). These data suggest that overstimulation, and resulting endocytosis, can be achieved by extended durations at concentrations below the stimulatory threshold.

Total Dose
InTiME tested whether an acute CT38 exposure could have a sustained effect. Tolerability and symptom worsening at concentrations above 0.25 ng/ml necessitated a reduced infusion rate, so limiting the AUC that could be delivered in the allotted timeframes. However, provided Cmax remained below 0.25 ng/ml, safety concerns were absent, and thus longer/additional infusions, to bring exposure closer to target (respectively, 3.7 and 4.9 ng h/ml for moderate and mild symptoms by extrapolation, Figure 7A), should increase efficacy. This point is evident in five patients (D03), who received 3-4 treatments sufficiently-separated to assess the effect of each treatment (Figure 9), resulting in mean TDSS decreases following treatments 1 and 2 combined (too close to separate), 3 and 4 (ID29). Overall, the see data support the notion that the AUC can be delivered as a single or multiple treatments, and increasing AUC increases effect.

CRFR2 Sensitivity
This is the first demonstration that the CRFR2 pathway is sensitized in ME/CFS patients, so consistent with the hypothesis. As noted above, CT38 plasma concentrations below 0.25 ng/ml do not activate G proteins in healthy subjects, and thus the elevated hemodynamic response that occurred below 0.25 ng/ml in ME/CFS patients (Figures 4A,C), likely represents elevated constitutive (agonist-independent) activity, putatively due to increased receptor expression (Black et al., 2016;Berg and Clarke, 2018). This increased sensitivity to low-level CRFR2 stimulation may aid diagnosis.

TEAEs Versus Symptoms
As CT38s is CRFR2-selective and has no off-target activity, the TEAEs resulted from CRFR2 stimulation. They included fatigue, myalgia, aches, sleep disturbance, forgetfulness, cognitive disturbance, dizziness, dysequilibrium, chills, influenza-like illness, sore throat, swollen lymph nodes, headache, paresthesia, shortness of breath, constipation, diarrhea, anxiety, emotional liability, etc. (and headache, dyspnea, sore throat and pain, in healthy subjects in the Phase 1, Supplementary Material). Such TEAEs are also known symptoms of ME/CFS. In fact, of the 108 TEAEs (excluding treatment day hemodynamics and flushing), 92 were well recognized ME/CFS symptoms and can be classified under the headings used in the TDSS endpoint (Table 3). This overlap of ME/CFS symptoms and TEAEs resulting from CRFR2 stimulation, including those in healthy subjects, supports the involvement of CRFR2 in ME/CFS.
The highest InTiME dose (D20: 0.795-1.620 µg/kg, Cmax = 1.15-1.32 ng/ml, AUC = 3.91-9.85 ng h/ml) was associated with sustained effects (e.g., headache, dyspnea) in ME/CFS patients that were transient at a comparable dose in the Phase 1 healthy subjects (1.667 µg/kg, Cmax = 2.46 ng/ml, AUC = 7.11 ng h/ml, Supplementary Material). That is, where healthy subjects reversed the effects of high-dose CT38s administration, ME/CFS patients did not, suggesting an impaired FIGURE 9 | Effect of CT38 individual treatment AUC on mean TDSS pre-treatment (purple bar) and post successive treatments (green bars), with standard deviations (error bars) and the level of AUC delivered in each successive treatment, by patient.
ability to reverse the effects of intense CRFR2 stimulation and therefore a susceptibility to CRFR2 maladaptation.

Long-Term Data
InTiME showed at least 28-day effect with a drug that clears in hours. Long-term follow-up in nine (of 14) patients, who are the PI's patients (with medical chart history), show that the effects are holding over a year from treatment (close to 2 years in the earlier-treated patients). Patients noted subtle improvements in sleep, brain fog, appetite, activity and PEM (crashed less often, recovered more rapidly, but it took weeks to appreciate these changes). While anecdotal, these data support the hypothesis/treatment approach.
First, if CRFR1/CRFR2 are pivotal in homeostasis as proposed, the very nature of homeostatic threat dictates that CRFR1/CRFR2 adaptations are neuronally-specific, and thus maladaptations are also neuronally-specific. This ties individual signs/symptoms to specific neurons, e.g., CRFR2-induced 5HT elevations in the motor pathway could inhibit motor neuron firing (Perrier et al., 2013;Perrier and Cotel, 2015) manifesting as fatigue (Supplementary Figures 1B,C), while CRFR2-induced 5HT elevations in medullary respiratory neurons (Hilaire et al., 2010) could diminish breathing capacity (Supplementary Figure 1F). This may explain how the same symptom can present in different diseases (e.g., fatigue in ME/CFS, fibromyalgia, multiple sclerosis), and how symptoms can vary within a given disease.
Second, for limbic response to be precise, neuronal CRFR1/CRFR2 expression can only depend on the threat and its resolution. This suggests the absence of other influences on receptor levels, and thus treatment-induced CRFR2 endocytosis (mimicking threat resolution) should persist, thereby removing the impetus for elevated 5HT and allowing the 5HT 1A autoreceptors to normalize and properly inhibit 5HT. Importantly, such endocytosis does not alter the adaptive nature Data indicate actual number of recorded events; T, total number of drug treatments, by group.
of the system; it only restores the set points, reducing the level of 5HT release for a given threat. Future threats will continue to modulate CRFR1/CRFR2, albeit releasing less 5HT, but nothing prevents future severe threats from provoking maladaptations that might accumulate and eventually cause dysfunction. Third, many acquired chronic diseases (Kotas and Medzhitov, 2015;Furman et al., 2019), including ME/CFS (Nacul et al., 2020), have been theorized to involve lost homeostasis brought on by chronic low-grade inflammation, possibly arising from chronic infection, diet, gut dysbiosis, environment, etc., but explaining how such inflammation leads to specific symptoms is challenging. CRFR1/CRFR2 maladaptation offers an alternative explanation for lost homeostasis and specific symptoms. This pathway controls autonomic and endocrine function (Dedic et al., 2018;Deussing and Chen, 2018;Godoy et al., 2018), and CRFR2-selective agonists modulate metabolic activity in obesity models (Jamieson et al., 2011;Chen et al., 2013;Paruthiyil et al., 2018) and immune response in models of sepsis (Gonzalez-Rey et al., 2006a), Crohn's disease (Gonzalez-Rey et al., 2006b), rheumatoid arthritis (Gonzalez-Rey et al., 2007) and cancer (Argilés et al., 2008), suggesting that CRFR2 plays a fundamental role. By implication, CRFR2 maladaptation could induce widespread dysfunction.
Fourth, maladaptations in limbic CRFR1/CRFR2 have no direct sequelae in bodily fluids. This is important because symptoms in the absence of bodily fluid abnormalities often lead to psychiatric diagnoses (common in ME/CFS), when the problem could be CRFR2 maladaptation. Equally, peripheral abnormalities need not imply peripheral dysfunction, e.g., poor lung function by spirometry might be diagnosed and treated as lung obstruction, yet could result from CRFR2induced changes in respiratory rate and tidal volume (Supplementary Figure 1F). Thus, an understanding of this pathway could have important diagnostic and treatment implications.
Fifth, given the proposed connection between individual neurons and symptoms, and the overlap of triggers, signs/symptoms with those of ME/CFS, the authors postulate that the pathway and treatment approach may apply to post-acute sequelae of SARS-CoV-2, chronic Lyme disease, fibromyalgia, post-traumatic stress disorder and multiple chemical sensitivities.
In sum, the authors propose that the CRFR1/CRFR2-5HT pathway controls homeostasis and that its disruption leads to lost homeostasis. If validated, this could fundamentally alter current conceptions and treatment of many acquired chronic diseases.

Limitations
InTiME had several limitations. It was small (n = 14) and open-label as treatment causes flushing. Tolerability (mild at D06, severe at D20) limited 2 dosing groups, and necessitated CT38 concentration reductions that were only partially offset by longer infusions, so target AUCs were low relative to the target dose. The concentrations of interest were close to the PK limit of quantitation (0.20 ng/ml). TDSS is not a validated endpoint in ME/CFS.

Conclusion
This study hypothesizes that ME/CFS is caused by CRFR2 upregulation in the raphé nuclei and limbic system, and it tests agonist-mediated CRFR2 endocytosis as a novel treatment approach. The results support CRFR2 involvement in ME/CFS, and identify a treatment paradigm that is Cmaxlimited and both AUC-and severity-dependent, leading to sustained symptom improvement. The PK-dependence of this response argues against a chance effect. These findings warrant further study.

DATA AVAILABILITY STATEMENT
The datasets presented in this article are not publicly available because they are part of an ongoing submission to the United States FDA. Requests to access the datasets should be directed to GP, gpereira@corteneinc.com.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Aspire Institutional Review Board and Independent Investigational Review Board Inc. The patients/participants provided their written informed consent to participate in this study. The animal studies were reviewed and approved by P&G's Institutional Animal Care and Use Committee and/or attending veterinarian in full compliance with the Animal Welfare Act.

AUTHOR CONTRIBUTIONS
GP conceived the hypothesis and treatment approach, and drafted the article. All authors developed the protocol. LB and HG served as the study PI and Medical Monitor, respectively. GP, HG, SC, and MC proposed dosing and changes thereto, which LB approved. LB, TM, and SV collected the data. MC and GP assembled and analyzed the data. All authors revised and approved the final version.

FUNDING
This research was funded by Cortene Inc., which owns the rights to CT38.