Article Text

Download PDFPDF
D04 Blood-CSF barrier function and CSF flow influence CSF biomarkers in huntington’s disease
  1. Jan Lewerenz1,
  2. Petra Steinacker1,
  3. Marc Suárez-Calvet2,
  4. Jens Dreyhaupt3,
  5. Estrella Morenas-Rodríguez2,
  6. Vera Lehmensiek1,
  7. Sonja Trautmann1,
  8. Christian Haass2,
  9. Markus Otto1,
  10. G Bernhard Landwehrmeyer1
  1. 1University of Ulm, Ulm, Germany
  2. 2Biomedizinisches Centrum (BMC), Biochemie, Ludwig-Maximilians-Universität München and Deutsches Zentrum für Neurodegenerative Erkrankungen (DZNE) e.V., München, Germany
  3. 3Institute of Epidemiology and Medical Biometry, University of Ulm, Ulm, Germany


Background Cerebrospinal fluid (CSF) biomarkers might be valuable tools to monitor disease progression and response to therapy in Huntington’s disease (HD). CSF flow and blood/CSF-barrier function, both of which can influence CSF biomarker levels, can be measured as CSF-to-serum albumin ratio (QAlb).

Methods We analyzed the CSF of 43 mutant Huntingtin gene carriers (HD) and 73 matched controls. In the HD group, five were pre-symptomatic, 18 had early (TFC 11–13), nine moderate (TFC 7–10), 11 advanced HD (TFC 1–6).

Results The HD group and controls did not differ in age and gender distribution, CSF cell count and lactate. However, the QAlb was significantly higher in the HD group (HD: 6.1±2.0×10–3, Ctrls: 4.8±1.5×10–3, p=0.0006), even when normalized to the age-dependent upper limit (QAlb/Qlim, HD: 0.84±0.29, Ctrls: 0.67±0.20, p=0.0018). Whereas the QAlb significantly correlated with age in controls (Pearson r=0.36), such a correlation was not observed in the HD group (r=0.15), neither did the QAlb correlate with the disease burden score (r=−0.02). QIgG, QIgA and QIgM corrected for QAlb were unchanged in the HD group.

A significant positive correlation of CSF neurofilament light chain (NFL) and soluble TREM2 (sTREM2) levels with QAlb in controls were observed (NFL: r=0.5, sTREM2: r=0.46, p<00001). Correcting NFL for QAlb, improved the performance of NFL, which became significantly upregulated even in presymptomatic HD gene carriers (Ctrl: 542±216 ng/ml, HD: 3445±3387 ng/ml, p<0.01). CSF sTREM2 levels were significantly elevated in the HD group (2.5±1.5 ng/ml vs. 1.6±1.5 ng/ml, HD vs. Ctrls, p=0.0043). This difference disappeared when correcting sTREM2 for QAlb (HD vs. Ctrl., 1.8±1.3 ng/ml vs. 1.6±1.1 ng/ml). Paired CSF/serum measurement of sTREM2 showed 1.6-fold higher levels in serum compared to CSF (n=12).

Conclusion In HD CSF, there is an early increase in QAlb that does not increase further with disease progression. Putative CSF biomarkers correlate with QAlb in control subjects. In the case of NFL, decreased CSF flow might lead to a paralleled increase of QAlb and NFL CSF levels. In case of sTREM2, a 17 kDA fragment of TREM2, both the increase in QAlb and sTREM2 might indicate low level inflammation in HD as it seems unlikely that the very low serum levels influence CSF sTREM2.

  • Cerebrospinal fluid
  • neurofilament light chain
  • albumin ratio
  • sTREM2Blood-CSF barrier function and CSF flow influence CSF biomarkers in Huntington’s disease

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.