Buspirone Challenge Replication: A Modern Test of the Exaggerated Prolactin Response
The historical literature reviewed in Chapter Endocrine and Metabolic Dysfunction (The Prolactin Axis and the Buspirone Challenge) describes a replicated but abandoned finding: ME/CFS patients show an exaggerated prolactin response to an oral buspirone challenge despite normal baseline prolactin. Because this line was never re-tested with modern diagnostic criteria — and because the ME/CFS-versus-depression comparison has never been performed with buspirone at all — an adequately powered replication is a high-value, low-cost next study: feasible for an interested research physician with access to an endocrine clinic and a handful of otherwise-standard blood draws.
The replication is designed to test the core, falsifiable prediction that motivated this topic (A Modern Buspirone Challenge Will Show Exaggerated Prolactin in ME/CFS But Not Depression Exaggerated Prolactin Response to Buspirone in ME/CFS, Replicated by an Independent Group): under a standardized oral buspirone challenge, the prolactin area-under-curve (AUC) will be significantly larger in ME/CFS than in major depression, with healthy controls intermediate, at normal baseline prolactin. This ME/CFS-versus-depression contrast has never been measured with buspirone — the historical depression data come from a d-fenfluramine probe — so the prediction extends, rather than merely re-tests, the historical finding. It is falsified if a modern IOM-criteria ME/CFS cohort shows no difference from depression or from healthy controls, if the historical effect disappears once medication washout, fasting, time-of-day, sex, and menstrual phase are controlled, or if it proves to be an epiphenomenon of differential nausea after adjusting for emetic symptoms.
Consequence: If confirmed, this one cheap study would establish a candidate biological test that distinguishes ME/CFS from depression and justify the follow-up mechanistic work — though an independent-group replication would still be required before any clinical use; if it fails, it would retire a notable but unverified lead.
1 Rationale
The finding has three properties that make it unusually worth replicating: it was reproduced by an independent group (Bakheit 1992 (Bakheit et al. 1992); Sharpe 1996 (Sharpe et al. 1996) from separate research teams), a related serotonergic-probe study found prolactin responses contrasting across diagnoses — lowest in depression, highest in CFS ((Cleare et al. 1995); from an author cluster overlapping the fenfluramine-null group, so not independent corroborations of each other) — and, per secondary accounts of the era, the finding was not anticipated in advance, which reduces (though cannot eliminate) concern that it is a confirmatory-bias artifact. Yet every historical study was small (n = 11–20 per arm), used Fukuda or pre-Fukuda criteria, and the line was abandoned around 2001. No post-2010 replication exists. The null fenfluramine results (Yatham et al. 1995) (Bearn et al. 1995) argue against a purely global serotonergic mechanism and constrain the signal toward pathways buspirone engages but fenfluramine does not — without localizing it (What Explains the Exaggerated Prolactin Response — Dopamine, Serotonin, Estradiol, or Nausea?).
2 Design Overview
- Design:: Cross-sectional neuroendocrine challenge, three arms — ME/CFS (IOM/CCC criteria), major depression (DSM-5), healthy controls.
- Intervention:: Single oral buspirone dose (0.5 mg/kg, as in the Sharpe protocol (Sharpe et al. 1996)) administered on a standardized morning after overnight fasting; serial plasma prolactin sampling (baseline and post-dose timepoints to capture AUC).
- Sample:: n ≈ 25–30 per arm. The historical effect was large but its estimate is highly uncertain (small n, single era, nausea-confound risk), so the power calculation must use conservative effect-size assumptions and pre-specify whether a subset-only effect would be interpretable.
- Confounder control:: same hour of day, overnight fasting, ≥4–6 week medication washout, sex- and menstrual-phase-matched analysis (men and women analyzed separately; premenopausal women tested in the same phase).
- Co-primary measures:: prolactin AUC and peak; secondary: baseline prolactin, cortisol (since cortisol modulates the buspirone-prolactin response (Dinan et al. 2001)), growth hormone, and rated nausea/emetic symptoms at each timepoint — so the hormonal signal can be tested against, and adjusted for, differential drug tolerability (the key unexamined confound; Sharpe’s CFS group reported more nausea (Sharpe et al. 1996)).
- Feasibility:: relatively quick and cheap — one visit, standard endocrine assays, a drug already in clinical use; within scope for a research physician without a large grant. The challenge is acutely unpleasant for some participants (nausea in the historical studies), so observation and participant-support provisions belong in the protocol.
3 If Confirmed
A positive replication would (a) establish a candidate objective challenge-test discriminator between ME/CFS and depression — with the nausea-adjusted analysis showing the signal is hormonal rather than a tolerability artifact — (b) revive the mechanistic question of whether the exaggerated response reflects reduced dopaminergic inhibitory tone, serotonergic or estradiol modulation, or none of these (What Explains the Exaggerated Prolactin Response — Dopamine, Serotonin, Estradiol, or Nausea?), and (c) justify the follow-up mechanistic and biomarker work, after independent-group replication. It would also argue for sex-stratifying the test, potentially connecting to the ~80% female predominance (Estradiol Modulation of the Prolactin Response as a Female-Predominance Link).
4 If Null
A null result — no difference between ME/CFS and depression or healthy controls once modern criteria and confounders are controlled — would retire a notable but abandoned line and prevent decades more of interest in a lead that may reflect pre-Fukuda diagnostic heterogeneity, uncontrolled confounders, an era-specific effect, or differential nausea. Either outcome is scientifically informative.
5 Comparator Probes to Isolate the Dopamine Component
Because buspirone is multi-receptor (5-HT1A agonist + D2 blocker), a pure D2-antagonist comparator (e.g., metoclopramide or domperidone, which block D2 without meaningful 5-HT1A agonism) and a pure 5-HT1A comparator would help localize whether an exaggerated response traces to the dopaminergic or serotonergic component (Prolactin as a Peripheral Window on Hypothalamic Dopamine Tone Medium-Spiny-Neuron Dopaminergic Link as a Broader Dopamine-Axis Sign). Haloperidol has also been suggested as a central D2-antagonist probe, but it is a first-generation antipsychotic with QT-prolongation, extrapyramidal, and neuroleptic-malignant-syndrome risk — a research-protocol option only under specialist supervision, never for off-label or self-initiated use. This is a diagnostic/research-design refinement, not a treatment: these agents would be used as single controlled challenge probes by a research physician. Interaction/safety note: all comparator probes named here — metoclopramide, domperidone, and haloperidol — are proposed strictly for a supervised single-dose challenge in a research protocol; none is recommended as an ME/CFS treatment. They carry their own acute hazards (QT prolongation with domperidone and haloperidol; extrapyramidal effects and hypotension) that are of particular concern in a population with documented dysautonomia, and their interaction profile with common ME/CFS co-prescriptions has not been checked in this integration — any clinical use requires prior specialist review of drug interactions and contraindications. (Origin: brainstorm ideas 3.1/3.2 — metoclopramide/domperidone and haloperidol as D2-antagonist probes.)
6 Limitations
The replication tests a small historical effect from a single diagnostic era; a null result cannot fully exclude a real effect present only in a subset (e.g., a particular severity or onset subgroup — the historical cohorts were ambulatory and not severity-stratified, so nothing is known about applicability to severe/very-severe patients, for whom a fasting, serial-venipuncture challenge is itself a significant burden), and a positive result would require replication by an independent group before clinical use. Buspirone is multi-receptor, so the probe does not isolate a single pathway, and the peripheral prolactin read-out is an indirect measure of one dopaminergic subsystem only.