Not medical advice. A living research archive of historical and continually changing sources, for research and personal reference only. Summaries are paraphrases — verify against the originals. Full disclaimer
A research map, not medical advice
Dr Powers’ current model is one of neurosteroid excess, with THDOC and allopregnanolone held high by feedback loops, a reversal of the older deficiency framing. These papers supply the receptor-level background: both steroids as GABA-A modulators, paradoxical and tolerance effects, steroid synthesis inside the brain, and why 5α-reductase type 1 versus type 2 matters, since finasteride targets type 2. The pathway question is whether central symptoms such as anhedonia, brain fog and low mood run through GABA-A dysregulation.
Where interventions would act (a research question, not guidance): Anything that shifts GABA-A neurosteroid tone has to reckon with sign-flip and tolerance effects; these papers describe that risk.
MECH-001 · Trends in Pharmacological Sciences · 2003
Powers' current theorizing centers on THDOC/allopregnanolone excess held in place by feedback loops; this review is the canonical statement that THDOC reaches physiologically active levels during stress and modulates the HPA circuits implicated in post-drug depressive phenotypes. It also documents that 5α-reductase inhibition blocks the DOC-to-THDOC conversion — the exact enzymatic step his model implicates. Any causal-pathway search for a neurosteroid-excess syndrome must start from the evidence base this review consolidates.
MECH-002 · Proceedings of the National Academy of Sciences · 1991
This is the empirical root of the stress–THDOC story Powers leans on: without stress-induced neurosteroid surges, there is no substrate for an "excess held in place by feedback loops" model. It serves as the founding observation that later work (Reddy 2003; Cadeddu 2025) builds on, and it defines the normal physiology that any pathological-excess hypothesis must be measured against.
MECH-003 · Nature Neuroscience · 2007
Powers' reversal from a deficiency to an excess theory requires a mechanism by which more GABAergic neurosteroid produces worse outcomes; this paper supplies exactly that — a sign-flip at α4β2δ receptors driven by subunit composition. It explains how elevated allopregnanolone/THDOC could be anxiogenic rather than calming, mapping onto anxiety phenotypes in PFS/PSSD. It also implicates receptor plasticity as a candidate mechanism for persistence after drug withdrawal.
MECH-004 · Neuroscience · 2006
A neurosteroid-excess model needs evidence that sustained high neurosteroid exposure produces lasting adaptation rather than just acute sedation; this tolerance study provides it at the behavioral level. The persistence of tolerance beyond drug clearance parallels the clinical puzzle of symptoms outlasting exposure. It motivates the receptor-level studies (MECH-005) that reveal the adaptation's substrate.
MECH-005 · European Journal of Pharmacology · 2006
This paper gives the molecular face of neurosteroid tolerance — downregulated α4 subunits — which is the kind of self-sustaining receptor remodeling Powers' feedback-loop theorizing invokes. If chronic excess remodels GABA-A composition, then removing the drug does not restore the original circuit, offering a persistence mechanism independent of continued exposure. It is the strongest preclinical precedent for "the system does not reset."
MECH-006 · European Journal of Neuroscience · 2004
If excess neurosteroid signaling is causal, then blocking it is the logical therapeutic probe; UC1011 is the founding demonstration that allopregnanolone's effects can be selectively antagonized. Its lineage runs to isoallopregnanolone/sepranolone, the clinical-stage neurosteroid antagonist — a genuine treatment-candidate anchor recorded here as mechanistic research only. It also shows the field already accepts that allopregnanolone excess can be pathogenic (in PMDD), a precedent for extending the logic to post-drug syndromes.
MECH-007 · Proceedings of the National Academy of Sciences · 2006
Powers' model and the Melcangi group's CSF findings both turn on where neurosteroids are made; this study proves the brain synthesizes allopregnanolone and THDOC locally via 5α-reductase type I, independent of gonads or adrenals. That means finasteride's CNS effects cannot be dismissed as secondary to peripheral DHT loss — and it explains how neurosteroid disruption could survive castration, a specific claim in Powers' theorizing. The type I anatomical map is the essential background for the isoenzyme division of labor shown in MECH-008.
MECH-008 · Science Advances · 2025
This is among the most PFS-relevant isoenzyme papers in recent years: it shows 5αR2 — finasteride's primary target — is the stress-recruited, male-specific driver of prefrontal allopregnanolone synthesis, while 5αR1 maintains baseline. That division maps directly onto what finasteride does and does not disrupt in brain neurosteroidogenesis, and why effects might be sex-specific and stress-gated. Any causal-pathway model involving 5α-reductase inhibition must reckon with this 2025 result.
MECH-009 · Molecular Psychiatry · 2011
Post-drug syndromes present with prominent depressive and anhedonic features; this review provides the best-supported framework linking those symptoms to GABAergic and neurosteroid dysfunction rather than monoamine deficiency. It legitimizes searching the causal pathway in GABA-A plasticity and neurosteroid tone instead of serotonin, and Powers' excess model is in effect a special case of the dysregulated-inhibition story told here.
MECH-010 · Psychopharmacology · 2014
The human-focused counterpart to the rodent THDOC literature: it documents that allopregnanolone/allo-THDOC normally brake the HPA axis and that chronic stress degrades this brake — the feedback-loop architecture Powers' model invokes. For a corpus built partly on patient-reported phenotypes, having the human regulatory physiology in the record matters more than another rodent study. It ties neurosteroid dysregulation to the exact mood-disorder outcomes the corpus tracks.
MECH-011 · Toxicology Letters · 2006
Community members inject DHB, and Powers theorizes it stacks as glucuronidated metabolites when UGT2BXX function is defective; this paper is the only rigorous pharmacology of the parent compound — potency, tissue effects, hepatic activity, AR agonism without metabolic activation. Without it, the corpus would discuss DHB entirely from forum anecdote. It also establishes DHB as a powerful non-aromatizing androgen, relevant to androgen-trapping hypotheses.
MECH-012 · Biological Mass Spectrometry · 1992
Powers' glucuronidation-defect model is specifically about metabolite clearance: DHB "stacks as metabolites" only if reductive metabolism plus conjugation is the clearance route, and this study proves exactly that for the boldenone series in humans. It identifies which reduced metabolites form and how they are excreted, giving the corpus a concrete metabolic map against which UGT2BXX-deletion claims can be evaluated. It is the metabolic companion to Friedel 2006's pharmacology.
The exit route in Dr Powers’ model: if UGT- and SULT-driven clearance fails, androgens stay trapped inside cells as weak metabolites. These papers cover the transporters that move testosterone glucuronides into bile, how UGT2B17 deletion reroutes clearance, SULT2A1 copy-number effects on androgen-sulfate excretion, and the β-glucuronidase rationale behind the calcium-D-glucarate discussion. The pathway question is whether clearance failure is what makes symptoms persist.
Where interventions would act (a research question, not guidance): Anything that changes glucuronidation or deconjugation acts here. The records are mechanism-level grounding, not an endorsement, and there is no human data on calcium-D-glucarate and androgen excretion.
MECH-013 · The Journal of Steroid Biochemistry and Molecular Biology · 2019
Powers' model hinges on how much androgen is actually eliminated versus trapped or recycled, yet clearance is usually discussed only at the UGT-conjugation step. This study fills the missing next step: the transporters that move androgen glucuronides into bile, which is the direct precondition for gut deconjugation and reabsorption. It belongs in the causal-pathway search as the mechanistic bridge between glucuronidation and enterohepatic androgen recycling.
MECH-014 · The FASEB Journal (Experimental Biology meeting abstract) · 2021
This is the closest published work to Powers' UGT2BXX-as-risk-modifier theorizing: it asks what the body does differently when UGT2B17 is absent and finds a metabolic rerouting strategy rather than collapse. The same study then shows the gut microbiome can undo that clearance by regenerating testosterone from its glucuronide. It belongs in the corpus because it simultaneously advances the "risk modifier" genetics angle and the gut-deconjugation angle Powers threads together.
MECH-018 · Cancer Letters · 1990
Calcium-D-glucarate is the treatment candidate Powers threads through his beta-glucuronidase theory: block the enzyme, and glucuronidated steroids get excreted instead of deconjugated and reabsorbed. This review is the canonical peer-reviewed summary of that mechanism, including the animal evidence that glucarate administration lowers circulating estradiol and androgen-derived steroid precursors. Recorded here strictly as a mechanistic research record, not a recommendation.
MECH-019 · Frontiers in Endocrinology · 2013
Powers' glucuronidation discussion tends to center on UGT enzymes, but sulfation (SULT2A1) is the parallel clearance route for androgens — and it too varies by gene copy number. This human in-vivo study is the exact counterpart of the UGT2B17-deletion logic in the corpus: clearance genotype predicts how fast steroid metabolites leave the body. It belongs here because a complete causal-pathway map must include both conjugation systems, and it surfaces another genetically variable clearance node.
MECH-049 · Journal of Biological Chemistry · 2007
This is the clearance-focused core of the axis: it identifies the exact enzymes (UGT2B15/UGT2B17) that terminate androsterone signaling by glucuronidation, which is precisely the pathway Powers theorizes is defective in the buildup model (and the gene family of his hypothesized UGT2B17 "risk modifier" deletion). It makes the mechanistic chain explicit — defective UGT2B17 → reduced androsterone glucuronidation → androsterone accumulation — and gives the treatment-candidate search a concrete enzymatic target. Classification is "steroid clearance / glucuronidation" because the paper is about termination of signaling, not systemic hormone levels.
Gut bacteria can undo hormone clearance by deconjugating glucuronides so the hormones are reabsorbed, and some can make neurosteroids, including allopregnanolone, themselves. These papers cover bacterial β-glucuronidase, the “testobolome”, gut-derived progestin synthesis and how gut transit time relates to blood hormone levels. The pathway question is whether the microbiome closes the clearance loop, or acts as a neurosteroid source that outlasts the drug.
Where interventions would act (a research question, not guidance): Microbiome-directed approaches would act downstream of this mechanism.
MECH-015 · npj Biofilms and Microbiomes · 2026
The corpus has the estrobolome concept for estrogens; this 2026 review is the androgen-side counterpart and therefore the more relevant term for PFS, which centers on androgen signaling. It gives the corpus a single citable anchor for "gut microbes recycle testosterone," a mechanism Powers' metabolite-pileup framing implicitly requires. Open-access (CC-BY).
MECH-016 · Journal of Biological Chemistry · 2019
This is the primary-data paper behind the estrobolome claim that the corpus's reviews cite. Powers' background model asserts that gut bacterial enzymes "undo hormone clearance" — this study is the concrete experimental demonstration of exactly that, at the enzyme level, for estrogen glucuronides. It earns its place as the mechanistic anchor for all enterohepatic-recycling claims in the corpus. Open-access (CC-BY).
MECH-017 · Proceedings of the National Academy of Sciences · 2026
This paper quantifies the recycling loop in human populations and shows that the gut's deconjugation capacity is a variable trait — one that differs between people by lifestyle history. For a corpus concerned with individual susceptibility in post-drug syndromes, that variability matters: the same biliary steroid load can yield different systemic exposure depending on a person's microbiome. It also supplies the bile-versus-feces recovery numbers that make the recycling argument quantitative. Open-access (CC-BY-NC-ND).
MECH-020 · Cell · 2024
This is the strongest primary-data paper for the microbiome–neurosteroid link in the corpus: gut bacteria were shown to manufacture allopregnanolone, the same neurosteroid the Melcangi group has repeatedly implicated in finasteride's long-term CNS effects. It reframes the gut not merely as a steroid recycler but as a net producer of neuroactive steroids, which belongs in both the causal-pathway search (a second neurosteroid source that finasteride withdrawal might perturb) and the treatment-candidate landscape.
MECH-021 · British Journal of Cancer · 1997
Powers' background material treats transit time as a lever on steroid exposure, and the corpus already holds a transit-time review — this 1997 human intervention is the primary study behind it and makes the link experimental rather than theoretical. It earns its place because it shows that a mundane gut parameter (transit speed) measurably changes circulating steroid levels, exactly the kind of individually variable exposure factor the causal-pathway search needs.
MECH-022 · International Journal of Molecular Sciences · 2021
The Melcangi group is the corpus's central PFS neurosteroid laboratory; this paper extends their work from drug effects to the microbiome itself, showing that removing gut microbes reshapes brain allopregnanolone, DHT, and 3alpha-diol — the exact steroids implicated in PFS and PSSD. It is the in-corpus group's own evidence that the gut-brain axis is a neurosteroid axis, tying the axis to the corpus's existing Melcangi cluster. Open-access.
How the HPG and HPA axes behave after androgen manipulation: what medical castration, the closest human analogue to the relugolix trial Dr Powers describes, does to brain structure and cognition; how the HPG axis restarts over time; where androgen and estrogen receptors sit in the brain; and androsterone, the DHT metabolite he reports building up. The pathway question is how peripheral hormone recovery separates from recovery in the brain.
Where interventions would act (a research question, not guidance): GnRH-axis and HPA-axis interventions act here. There are no relugolix neuropsychiatric trial data.
MECH-025 · Brain Research · 2013
This is the most direct mechanistic bridge between finasteride and HPA-axis dysregulation: central 5α-reductase blockade is sufficient to disinhibit the stress axis in a mammalian brain. It gives the hormonal-axes axis a concrete published mechanism for how 5AR inhibition can leave a lasting neuroendocrine scar — the kind of axis-level injury consistent with Powers' HPA-shutdown protocol theorizing and with Melcangi-group reports of persistent neurosteroid disruption after drug withdrawal. It also explains why restoring circulating testosterone alone might not normalize a stress axis whose central DHT-dependent brake has been altered.
MECH-026 · Cancer Medicine · 2022
Powers' unpublished theorizing places GnRH-axis manipulation (relugolix as an oral GnRH antagonist "castration trial," HPA-shutdown protocol) at the center of his current model, and this paper gives that move published company: it argues GnRH analogs act directly in the brain via GnRH/LH receptor signaling, with cognitive consequences beyond testosterone suppression. That is the closest peer-reviewed analog to the claim that manipulating the HPG axis is itself a neurological intervention. It also predicts the relugolix-trial pattern Powers describes — restoring androgenic signaling may not rescue symptoms if the GnRH/LH signaling disruption is a separate pathogenic axis.
MECH-027 · Medicina · 2023
ADT is the cleanest human parallel to pharmacologic androgen silencing, and this review documents its central phenotype: cognitive decline plus mood disturbance, with structural and molecular correlates — a direct precedent for treating PFS/PSSD brain fog and anhedonia as consequences of androgen signaling loss rather than as separate psychiatric entities. The shared-gene signal with Alzheimer's disease also gives the corpus a mechanistic hook for why androgen-axis disruption converges on neurocognitive pathology. It is the anchor record for the "androgen-silencing parallel" in this axis.
MECH-028 · PLoS ONE · 2013
This is the structural-brain counterpart to the ADT cognitive review: it shows androgen silencing producing gray-matter loss in prefrontal regions — a physical substrate for the brain-fog/cognitive-slowing phenotype. It directly supports the task's framing that central symptoms demand a central explanation, and it gives the corpus an imaging-level precedent for androgen signaling as a maintenance factor for adult cortical structure. Exploratory and small, but the morphometry-behavior correlation (motor cortex volume vs. response time) is the kind of objective finding the post-drug-syndrome field lacks.
MECH-029 · Asian Journal of Andrology · 2013
This is the canonical quantitative record of HPG-axis restart kinetics after medical castration: even with only 9 months of ADT, a meaningful fraction of men took a full year to normalize testosterone. It sets the recovery timeline against which post-drug-syndrome persistence can be judged — symptoms that outlast this restart window point to something beyond simple testosterone recovery, which is exactly the logic of Powers' theorizing that persistent central symptoms survive restored peripheral androgen signaling. It also documents baseline testosterone as a recovery predictor, relevant to any treatment-candidate protocol that involves HPG-axis restart.
MECH-030 · Current Opinion in Supportive & Palliative Care · 2017
This is the axis's receptor-level record: androgen and estrogen receptors in the hippocampus and cortex make the brain a direct target of androgen-axis disruption, not just a downstream casualty of low circulating testosterone. Its explicit discussion of testosterone's dual brain fate — conversion to DHT by 5α-reductase or to estradiol by aromatase — ties the hormonal-axes axis back to 5AR (the finasteride target) and to the Melcangi-group theme of altered neuroactive-steroid signaling in the brain. It also provides the rationale for why manipulating steroid synthesis enzymes, not just replacing testosterone, is the mechanistically interesting treatment-candidate space.
MECH-048 · Epilepsia · 2005
Powers' androsterone-buildup thread treats androsterone as pharmacologically meaningful, not inert waste. This paper is the peer-reviewed anchor for that claim: androsterone is a GABA-A positive allosteric modulator with demonstrated anticonvulsant activity, so its accumulation or depletion has plausible neuropsychiatric consequences. That turns androsterone from a clearance-pathway footnote into a candidate mediator in the causal-pathway search, and it links the steroid-clearance axis (MECH-049) to functional CNS effects.
MECH-050 · The Journal of Steroid Biochemistry and Molecular Biology · 2006
This paper reframes androsterone glucuronide from waste product to the field's best window on total androgen exposure — exactly the quantity the buildup hypothesis claims is disturbed when glucuronidation fails. For the causal-pathway search it supplies the biomarker logic: if UGT-mediated clearance is defective, ADT-G dynamics are where the lesion shows up, and serum testosterone will miss it. It also connects the corpus's hormonal-axes records to measurable clinical chemistry (LC-MS/MS ADT-G), which is what a future treatment-candidate evaluation would need to monitor.
If symptoms in the brain persist after blood hormone levels normalize, failed brain clearance is a natural next suspect. These papers cover the glymphatic system (fluid exchange that depends on aquaporin-4), sleep’s role in clearance, and the cycle between neuroinflammation and glymphatic failure. The pathway question is why the brain might not follow the bloodwork.
Where interventions would act (a research question, not guidance): Sleep, aquaporin-4 and neuroinflammation are a separate research target from hormonal approaches.
MECH-031 · Science Translational Medicine · 2012
The foundational record for the entire brain-clearance axis: without the glymphatic discovery, the hypothesis that central symptoms persist because of failed brain waste clearance has no mechanism. It identifies AQP4 — an astrocyte protein — as the molecular linchpin of clearance, which makes astroglial dysfunction a legitimate causal-pathway candidate for persistent brain fog and anhedonia after drug withdrawal. Every downstream glymphatic claim in this axis rests on this paper.
MECH-032 · Science · 2013
Sleep disturbance is a near-universal feature of post-drug syndromes, and this paper turns that complaint into a clearance-failure mechanism: poor sleep means the brain's interstitial space never opens for its nightly waste wash. If glymphatic failure is a persistence mechanism for brain fog and anhedonia, then the sleep disruption seen in PFS/PSSD is not just a symptom but part of the causal loop — which also makes sleep quality a mechanistically grounded treatment-candidate target (research-only framing). It is the second pillar of the axis alongside Iliff 2012.
MECH-033 · International Journal of Molecular Sciences · 2021
This paper supplies the persistence mechanism the axis needs: a self-sustaining neuroinflammation–glymphatic loop that can keep running after the triggering drug is long gone, producing exactly the chronic central symptoms (brain fog, anhedonia) that outlast hormonal recovery. It dovetails with Powers' unpublished theorizing that persistent central symptoms require a central explanation beyond circulating hormone levels, and with the Melcangi-group theme of chronic neuroinflammation as a PFS/PSSD feature. The emphasis on AQP4 polarization as reversible also keeps a therapeutic door open (research-only) for clearance-restoring interventions.
MECH-034 · CNS Neuroscience & Therapeutics · 2026
The most current synthesis available, and the axis's only record that links glymphatic dysfunction explicitly to mood and cognition outcomes — the two symptom domains (brain fog, anhedonia) the axis is meant to explain. It also catalogues the measurable biomarkers (DTI-ALPS, enlarged perivascular spaces) and the clearance-restoring strategy space, giving the corpus both a way to test the brain-clearance hypothesis in post-drug patients and a set of mechanistically grounded treatment-candidate directions to track. It closes the axis from discovery (Iliff, Xie) through mechanism (Mogensen) to clinical translation.
Which genetic variants actually have human association data, the test any gene-level theory has to pass. These papers cover CYP2C19, ABCB1 and HTR2A variants and SSRI sexual side effects, human UGT2B17-deletion data including a null result that cuts the other way, SULT2A1 copy number, androgen-receptor CAG-repeat population data, and the only human finasteride pharmacogenomics (CYP3A4/5). The pathway question is why these drugs harm some people and not others.
Where interventions would act (a research question, not guidance): None directly: this area is about who may be vulnerable, which informs risk rather than treatment.
MECH-035 · The Canadian Journal of Psychiatry · 2023
This is the most directly on-point susceptibility-marker study for the PSSD axis yet found outside the corpus: it links measured CYP2C19 metabolizer status to on-treatment changes in sexual arousal, with serum drug levels as an exposure bridge. It complements the corpus's CYP2D6-focused Discord records by adding CYP2C19 and the blood-brain-barrier transporter ABCB1 to the candidate susceptibility panel, and its unexpected direction of effect is a cautionary data point for any treatment-candidate search that would stratify patients by metabolizer status.
MECH-036 · The Pharmacogenomics Journal · 2013
ABCB1 sits squarely in the causal-pathway logic Powers applies to drug clearance: a transporter variant that changes brain exposure to an SSRI is a mechanistically credible susceptibility factor for who develops severe or persistent serotonergic side effects, sexual effects included. It extends the corpus's pharmacogenomic coverage beyond metabolizing enzymes (CYPs) to the blood-brain-barrier efflux step, and pairs naturally with MECH-035, which tested the same gene in the same phenotype.
MECH-037 · Neuropsychopharmacology · 2006
HTR2A is the receptor most directly implicated in the serotonin-brake model of SSRI sexual dysfunction — the same model invoked in PSSD discussions — and this is the primary-outcome study tying its promoter variant to the phenotype, with a sex-specific pattern that echoes the corpus's interest in hormonal context (oral contraceptives as mediator). Together with MECH-035 and MECH-036 it builds a three-gene pharmacodynamic plus pharmacokinetic susceptibility panel (HTR2A/GNB3, CYP2C19, ABCB1) for the treatment-candidate search.
MECH-038 · The Pharmacogenomics Journal · 2008
UGT2B17 is the single gene Powers names most prominently — his theorized "base, core defect," later reframed as a risk modifier — and this is human association data showing that its deletion changes androgen handling inside a steroid target tissue and tracks with an androgen-driven disease outcome. It is the natural companion to the corpus's Yang 2008 (UGT2B17 copy number and circulating testosterone/estradiol) and belongs in any causal-pathway diagram of how impaired steroid clearance could leave a vulnerable endocrine milieu after 5-ARI exposure.
MECH-040 · PLoS Genetics · 2011
DHEA/DHEAS sits at the top of the neurosteroid cascade that the Melcangi group places at the center of PFS pathophysiology, and this GWAS pins two corpus-relevant genes — SULT2A1 (Powers-named) and CYP2C9 (a drug/xenobiotic-metabolizing CYP) — to circulating DHEAS in humans at genome-wide significance. It gives the corpus its first genome-wide (hypothesis-free) anchor for the sulfation axis and adds CYP2C9, relevant to retinoid and drug metabolism, to the marker list.
MECH-041 · Progress in Neuro-Psychopharmacology and Biological Psychiatry · 2013
MTHFR is Powers-named and one-carbon metabolism is his recurring explanatory thread for persistent neuropsychiatric symptoms; this meta-analysis supplies the human association data — modest but real — that a susceptibility model needs. It also motivates the treatment-candidate side of the search: if impaired folate-pathway function raises depression risk, then folate-pathway interventions (e.g., the L-methylfolate augmentation RCT of Papakostas et al. 2012) become testable candidates rather than speculation.
MECH-042 · International Journal of Endocrinology · 2016
The corpus already covers AR repeats in PFS patients (Cauci 2017, Cecchin 2014); this adds the general-population complement — AR CAG length tracking sexual function in eugonadal men, which is exactly the hormonal context of most PFS/PSSD patients (normal testosterone, persistent symptoms). It sharpens the androgen-sensitivity node of the causal model: the same androgen level can mean different receptor signaling depending on CAG length, a candidate explanation for why only a subset of exposed men develop persistent dysfunction.
MECH-043 · PLoS ONE · 2015
This is the only human study found that directly ties common genetic variants to finasteride blood levels — the exposure variable at the very start of any PFS causal chain — with effect sizes large enough to matter (halving to 1.5x-plus). It converts "CYP3A4 metabolizes finasteride" from textbook fact into a quantified susceptibility factor and belongs at the intake end of the corpus's causal-pathway map, upstream of the neurosteroid and epigenetic findings.
One bridging paper: the histone demethylase LSD1 (KDM1A) acting as an androgen-receptor coactivator, the histone–receptor link the epigenetics supplement lacked. It is cell-line mechanistic background, not association data; the rest of this area is covered by the epigenetics supplement.
MECH-044 · Nature · 2005
The corpus's epigenetic supplement (EPI-001...015) covers DNA methylation and miRNA but has no histone-modifier entry; this landmark paper supplies the missing mechanism by which chromatin state directly gates AR output. It is the conceptual bridge for the epigenetic-persistence hypothesis: if histone-modifying enzymes set the transcriptional competence of androgen-responsive genes, then a drug-induced shift in their activity could lock in altered AR signaling after the drug is gone — exactly the kind of durable state change the Traish epigenetic model of PFS proposes.
Zimelidine, the first SSRI to be marketed (1982), was withdrawn in 1983 after reports of Guillain–Barré syndrome, probably immune-mediated, with a hypersensitivity and liver signal alongside. It is a precedent that serotonergic drugs can trigger severe immune-mediated harm, context for any immune or inflammatory theory of PSSD, and the start of a regulatory arc that reaches the EMA’s 2019 recognition of PSSD.
MECH-045 · Social Science & Medicine · 2015
The corpus's PSSD-recognition arc (EMA 2019, FDA labeling, class-wide signal reviews) needs its historical bookend: the SSRI class has already had a member pulled from the market for severe post-marketing harms. A peer-reviewed history of Zelmid's rise and fall supplies exactly that precedent and documents how early post-marketing neurological toxicity was handled by the manufacturer and regulators, giving readers a baseline for judging whether PSSD recognition has been proportionate. It also anchors the androgen-clearance axis: Zelmid's Astra/Carlsson lineage is the same pharmacological era Powers discusses.
MECH-046 · Journal of Neurology, Neurosurgery & Psychiatry · 1985
This is the primary clinical evidence behind the only SSRI-market withdrawal in history, so it earns its place as the documented anchor of the corpus's drug-safety-history strand. The ~25-fold risk estimate and the explicitly immunological framing matter for the user's stated interest in immune-system-related mechanisms of post-drug syndromes. It also illustrates the corpus's honest-status ethos: rare, severe, mechanistically unexplained harms were what ended Zelmid, and the paper is explicit about the immunological inference being probable rather than proven.
MECH-047 · Acta Psychiatrica Scandinavica. Supplementum · 1983
This is the peer-reviewed source for the hypersensitivity/liver signal the user asked to verify rather than assume. It shows the corpus's immune-related adverse-event thread for SSRIs is grounded in contemporary clinical-use data, not hindsight. Paired with MECH-046, it gives the axis a complete safety-profile picture: a frequent immune-type hypersensitivity reaction plus a rare immune-type neuropathy, matching the two immune-system-related claims the user wanted checked.