Iatrogenic Alzheimer's disease in recipients of cadaveric pituitary-derived growth hormone

Prion diseases demonstrate that misfolded protein assemblies can propagate and cause neurodegeneration, with most human cases being sporadic or inherited and a minority acquired iatrogenically. Experimental and pathological evidence indicates that misfolded proteins implicated in common neurodegenerative diseases, including Alzheimer’s disease (AD), can also spread in a prion-like manner. Prior work showed human-to-human transmission of Aβ pathology and cerebral amyloid angiopathy (CAA) in individuals exposed in childhood to cadaveric pituitary-derived growth hormone (c-hGH), and archived c-hGH batches contained Aβ (and tau) with transmissible Aβ seeding activity. Clinical reports have established iatrogenic CAA after other medical exposures (for example, dura mater grafts, neurosurgery), but there have been no premortem clinical descriptions among c-hGH recipients who did not develop iatrogenic CJD. The study investigates whether c-hGH recipients may develop AD-spectrum clinical syndromes decades after exposure, consistent with iatrogenic transmission of Aβ leading to AD.

The authors summarize evidence that mammalian prions are transmissible protein-only agents with strain diversity and long incubation periods, and that iatrogenic CJD has occurred via several medical routes, including c-hGH. Parallel concepts have emerged for AD and Parkinson’s disease, where pathological protein assemblies propagate in a prion-like manner. Human transmission of Aβ pathology has been documented: initial reports in c-hGH recipients who died of iCJD exhibited Aβ deposition and CAA; archived HWP c-hGH batches contained measurable Aβ and tau and seeded Aβ pathology in mice. Multiple autopsy-based reports confirmed iatrogenic Aβ transmission through c-hGH, dura mater grafts, and neurosurgical exposures, and iatrogenic CAA is now recognized clinically as a cause of early-onset hemorrhagic stroke. The authors note that sporadic AD typically arises later in life, familial AD results from APP/PSEN1/PSEN2 mutations, and APOE ε4 modulates risk and age at onset. They also review that HWP (Hartree-modified Wilhelmi) c-hGH preparation uniquely showed Aβ contamination versus other preparation methods.

Design: Case series assembled via the UK National Prion Clinic (NPC) within the National Prion Monitoring Cohort, which receives national referrals of suspected prion diseases and at-risk individuals, including prior c-hGH recipients. Cohort and ascertainment: Between 2017 and 2022, eight individuals with childhood c-hGH exposure were referred to or reviewed by the NPC after the authors’ prior report of iatrogenic CAA. All eight had received c-hGH prepared by the Hartree-modified Wilhelmi preparation (HWP); some received additional preparations. Archived records established batch histories. Prior experimental work had quantified Aβ-40, Aβ-42 and tau in specific HWP batches and demonstrated Aβ transmission to mice from some batches also received by three of the current patients. Exclusion of iCJD: Iatrogenic CJD was excluded in all eight cases based on clinical features, neuroimaging, fluid biomarkers and, in two cases, postmortem examination. Clinical phenotyping: Standard clinical assessments determined symptom onset, cognitive domains affected, and diagnoses according to NIA-AA and DSM-5 criteria. Latency from HWP exposure was calculated. Biomarkers and imaging: Investigations performed as part of routine care included CSF Aβ42/Aβ40 ratio and phosphorylated tau, total tau, amyloid PET (18F-Florbetaben), and structural imaging (CT/MRI) with longitudinal assessments of regional atrophy and small vessel disease markers. AT(N) categorization placed subjects on the Alzheimer’s continuum when possible. Neuropathology: For case 2, a left frontal lobe biopsy underwent H&E and immunohistochemistry for Aβ (DAKO 6F3D) and phospho-tau (AT8). For case 1, postmortem sampling of multiple brain regions used standardized high-risk autopsy protocols, including formic acid treatment and immunostaining for PrP, Aβ, and phospho-tau. Pathology staging referenced Thal phases and CERAD neuritic plaque scores. Genetics: With consent, next-generation exome sequencing (CENTOGENE) surveyed genes implicated in adult-onset neurodegeneration (including APP, PSEN1/2, PRNP, APOE, ABCA7, SORL1, TREM2 and others). APOE genotyping was reported. Three cases lacked genetic data due to unavailable samples. Ethics: Analyses used de-identified clinical data in compliance with ethical regulations; tissue use approved by local research ethics committee with next-of-kin consent for postmortem/biopsy material.

- Eight c-hGH recipients (all exposed to HWP-prepared c-hGH) were evaluated; iatrogenic CJD was excluded in all. - Five of eight developed early-onset dementia with progressive impairment in ≥2 cognitive domains: symptom onset ages 38–49 in four cases and 55 in one; two presented amnestically and three non-amnestically (dysexecutive or language), meeting NIA-AA criteria for probable AD and DSM-5 criteria for major neurocognitive disorder due to AD. - One additional individual had mild cognitive impairment (behavior/personality predominant) with onset at 42; one had subjective cognitive symptoms only; one was asymptomatic at assessment. - Biomarkers supportive of AD were present in several cases: two clinically diagnosed AD cases met AT(N) criteria (amyloid and tau positivity); one asymptomatic individual fulfilled AT(N) criteria (reduced CSF Aβ42/Aβ40 ratio 0.053; elevated CSF phospho-tau 181 at 64 pg/ml). Amyloid PET in case 3 showed diffuse cortical tracer uptake with regional prominence. - Neuroimaging showed progressive regional atrophy consistent with neurodegeneration in multiple cases (for example, mesial temporal, frontal, and parietal volume loss), with limited imaging evidence of CAA or small vessel disease. - Neuropathology: Case 1 (postmortem) exhibited widespread parenchymal Aβ deposition (Thal phase 5; CERAD score 2), severe widespread CAA (including leptomeningeal and cortical vessels), and focal Alzheimer-type tau pathology with moderately frequent neuritic plaques in insular cortex. Case 2 (biopsy and brain-restricted postmortem) showed diffuse parenchymal Aβ with a single vessel exhibiting concentric Aβ angiopathy and scant tau pathology not meeting AD criteria. - Latency from HWP exposure to symptom onset was approximately 30–44 years (three to four decades). - Genetic testing identified no pathogenic variants associated with adult-onset neurodegeneration in five tested cases; one rare APP variant of uncertain significance (likely benign) was observed. Only one patient carried an APOE ε4 allele; several were ε3/ε3, one ε2/ε3. - All affected individuals had exposure to HWP c-hGH. Prior quantified Aβ contamination and demonstrated Aβ transmissibility in mice from specific HWP batches (for example, HWP 42 and HWP 51) were received by three of the current patients, strengthening the exposure–disease link. - Alternative explanations (childhood intellectual disability, underlying diagnoses, persistent growth hormone deficiency, radiotherapy) were systematically considered and deemed unlikely to account for the clinical syndromes and biomarker profiles observed.

The convergence of clinical, biomarker, imaging, and pathological data in c-hGH recipients exposed to HWP-prepared material supports the conclusion that Aβ transmission can cause AD-spectrum disorders after prolonged incubation, analogous to acquired human prion diseases. The early-onset presentation, frequent non-amnestic phenotypes, prolonged latencies, and limited imaging evidence for clinical CAA distinguish these cases from typical sporadic and familial AD, paralleling how acquired prion diseases differ phenotypically from sporadic and inherited forms. The authors argue that Aβ strain diversity, by analogy to prion strains, may contribute to phenotypic differences and the relative paucity of overt clinical CAA despite pathological CAA in at least one case. Alternative etiologies—including intellectual disability, underlying childhood conditions, growth hormone deficiency, and cranial radiotherapy—do not plausibly explain the observed progressive dementias and AD biomarkers/pathology. Given the prior demonstration of Aβ contamination and transmissibility from archived HWP c-hGH batches, and the national referral context minimizing ascertainment bias, the findings indicate that iatrogenic AD can occur after medical exposure to Aβ seeds. Public health implications include reassessing measures to prevent iatrogenic transmission via medical/surgical procedures and considering prion-like strain dynamics when developing anti-amyloid therapeutics, as selective pressure might favor minor conformers and promote drug resistance.

This case series provides the first premortem clinical and biomarker evidence consistent with iatrogenic Alzheimer’s disease in recipients of cadaveric pituitary-derived growth hormone, extending prior autopsy and experimental findings of Aβ transmission. The work indicates that AD, like prion diseases, has sporadic, inherited, and rare acquired (iatrogenic) forms. Recognition of iatrogenic AD underscores the need for stringent decontamination and risk-reduction strategies in healthcare settings. Future research should define the full clinical and pathological spectrum of iatrogenic AD, elucidate host susceptibility factors and Aβ strain biology, quantify risks associated with various medical exposures, and evaluate how strain diversity and quasispecies dynamics might influence therapeutic response and resistance to anti-amyloid agents.

- Small case series derived from routine clinical care with heterogeneity in investigations; not all subjects underwent complete biomarker panels or standardized imaging protocols. - Genetic data were unavailable for three cases, limiting assessment of risk variants, although family histories did not suggest familial AD. - Pathological confirmation was available in only two cases (one biopsy with limited postmortem sampling and one full postmortem), constraining definitive staging across the cohort. - Case ascertainment relies on referrals to a national prion service; the true incidence and broader phenotype in c-hGH recipients may be underrecognized. - Variability in exposure intensity, age at exposure, batch composition (amount and strain of Aβ seeds), and host factors likely influenced phenotype and latency but could not be systematically quantified. - No directly comparable cohort of recipients exposed exclusively to non-HWP c-hGH was identified within referrals, limiting formal comparative risk estimation.