Studies Reveal New Hope, Old Problems With AD Biomarkers

Brain imaging and cognitive tests may be the gold standards for tracking Alzheimer disease progression, but clinical trials using these procedures are expensive, risky, and time-consuming, making the development of reliable go-betweens a pressing goal for AD researchers. Three recent papers highlight new promise, and expose nagging pitfalls, of one of the field’s most widely used surrogate measures: cerebrospinal fluid biomarkers. CSF samples offer researchers an indirect peek at signature molecules in AD and other neurodegenerative diseases without having to probe the brain itself.

In one study, scientists use a new method for quantifying protein turnover in the central nervous system to show that a candidate AD drug lowers CNS Aβ production in healthy people. This approach has the potential to streamline drug development, by indicating early on if and to what extent experimental compounds reach their CNS target. More generally, biomarker analysis presents an ongoing challenge in that their measurement varies markedly in different assays, even among different labs using the same assays. Researchers have tried to get a better handle on this problem, and the second study reported in this story finds sobering variability among 20 labs worldwide that measured Aβ42, tau, and phospho-tau in the same CSF samples. Meanwhile, a third report describes a potential new AD biomarker in the form of reduced CSF levels of the sortilin-related receptor SORLA.

Several years ago, Randy Bateman, David Holtzman, and colleagues at Washington University School of Medicine in St. Louis, Missouri, devised a method that combines stable isotope labeling with CSF sampling to measure production and clearance of Aβ peptides in real time in human CNS (Bateman et al., 2006 and ARF related news story). Through a lumbar catheter, the researchers collect hourly CSF and blood samples from patients receiving intravenous infusion of 13C-labeled leucine, which gets incorporated into newly made Aβ. Tracking the rise and fall of the percentage of labeled Aβ over time reveals how quickly Aβ gets produced and cleared, respectively.

In a study that appeared online 10 April in the Annals of Neurology, these scientists applied the method, called stable isotope-linked kinetics (SILK), to gauge how effectively an experimental AD drug could decrease Aβ production. Twenty healthy male volunteers (ages 21-50) received Eli Lilly’s γ-secretase inhibitor LY450139 once at one of three oral doses (100 mg, 140 mg, 280 mg), or placebo. Using ELISA, the researchers measured Aβ1-x, Aβ1-40, and Aβ1-42 concentrations in CSF and blood samples collected hourly from each participant over a 36-hour period. During the 12 hours of labeled leucine administration, patients on the highest drug dose had an 84 percent drop in CNS Aβ generation, followed by 52 percent and 47 percent decreases for the middle and lowest doses, respectively. Aβ clearance rates did not differ among the groups.

“The study showed for the first time that the Lilly drug actually inhibited Aβ production in the human CNS,” wrote Holtzman in an e-mail to ARF. He noted that previous human studies (Siemers et al., 2006; Siemers et al., 2007) with this drug were unable to demonstrate this for several possible reasons, including timing of CSF sampling, which likely did not account for hour-by-hour fluctuations of CSF Aβ levels seen routinely in healthy people (Bateman et al., 2007). Though the study showed that a single dose of Lilly’s γ-secretase inhibitor reduced new CNS Aβ synthesis by about 50 to 90 percent over a 12-hour window, “whether this is enough to influence AD pathogenesis is not clear,” Holtzman wrote.

In the meantime, C2N Diagnostics, a St. Louis company he founded with Bateman and others, is using the SILK-Aβ technology to test other AD drug candidates, such as those that inhibit Aβ production. Bateman told ARF in an e-mail that he thinks the SILK method could be used in early (Phase 1 or 2) clinical trials to determine if drugs are hitting their CNS targets. Such information could help sponsors decide how to proceed with longer, costlier trials that test their drug’s effect on cognition—or, in some cases, to scrap further development of the drug.

Alongside this ray of hope comes news reminding the field how unruly these biomarkers can be when it comes to reproducing their measurements reliably across labs and protocols. Though CSF levels of Aβ42, tau, and phospho-tau can distinguish patients with AD from those without symptoms or with other types of dementia in the hands of several independent groups by now, absolute biomarker measurements differ significantly among labs, complicating data analysis in multicenter studies. A multi-institutional collaboration led by Marinus Blankenstein and Niek Verwey of VU University Medical Center, Amsterdam, has tried to address this issue more formally by comparing AD biomarker measurements made by 20 centers worldwide. The researchers sent three CSF samples—one with high tau and phospho-tau levels, one with low Aβ1-42, and one with a normal biomarker profile—to 13 labs in 2004. In 2008, they distributed the same samples to 18 labs, 11 of which had also participated in 2004.

The main gist of the findings, reported online 2 April in the Annals of Clinical Biochemistry, is that the sites differed considerably in their measurements of all three biomarkers tested—particularly Aβ1-42, which had more than 22 percent variability among centers. The team found substantial variation even when comparing biomarker measurements made by the same lab in 2004 and 2008. Variability must drop to less than 10 percent before scientists can reliably compare biomarker measurements from different sites to determine a reference range for AD that all centers can use, Verwey wrote in an e-mail to ARF (see full comment below).

For their part, Alzheimer’s Disease Neuroimaging Initiative (ADNI) scientists have analyzed CSF samples from several hundred subjects in the U.S. and Canada and defined, in a recent paper (Shaw et al., 2009 and ARF related news story), threshold values for CSF Aβ and tau, as well as standardized protocols for measuring protein levels and handling samples. (See also in-depth ARF ADNI series). Standardization may become even more crucial as new protein detection methods—for example, an ELISA that differentiates between oligomeric and monomeric forms of Aβ (Xia et al., 2009 and ARF related news story)—come onto the scene.

The kind of assay validation done by ADNI has not yet been done for blood Aβ measurements, as a recent study shows (Okereke et al., 2009). Researchers led by Francine Grodstein at Brigham and Women’s Hospital, Boston, spiked plasma samples with known amounts of Aβ. They sent the spiked samples to various U.S. labs that used five different protocols to measure plasma Aβ40 and Aβ42. Though assay reliability and Aβ stability after processing delay was encouraging, the study was disappointing in its recovery data. Recovery rates for Aβ40 ranged from -24.3 to 44.2 percent, and for Aβ42 from 17.1 to 60.7 percent. These findings make “comparisons of absolute Aβ values across studies inaccessible at this time,” wrote lead author Olivia Okereke, also of Brigham and Women’s Hospital, in an e-mail to ARF. “Clearly, an important next step for the field is assay standardization work, as has been accomplished recently for some of the lipid and inflammatory biomarkers in cardiovascular disease.”

While studies of the older CSF markers Aβ and tau proceed apace, scientists have identified what could turn out to be a new one. In this month’s Archives of Neurology, a team led by Greg Cole at the University of California, Los Angeles, has detected reduced levels of SORLA in CSF of AD patients.

SORLA, a transmembrane neuronal sorting protein that reduces Aβ production, emerged several years ago as a genetic risk factor for late-onset AD (Rogaeva et al., 2007 and ARF related news story). Scientists have detected decreased SORLA expression in LOAD brain tissue (Scherzer et al., 2004; Zhao et al., 2007), and a more recent study has linked SORLA gene variants with reduced CSF Aβ42 in AD (Kölsch et al., 2008). For his part, Cole has shown that putting mice on a diet rich in omega-3 fatty acids leads to increased SORLA expression, suggesting that SORLA might be involved in mediating the Aβ-lowering effects of this special diet (Ma et al., 2007).

Whether this increase in SORLA expression occurs in people has not been tested. “But if you could see the levels of the protein in CSF, then you would be able to ask that question,” Cole told ARF. Given that SORLA gets cleaved near its membrane C-terminus, his team reasoned that the soluble N-terminal piece is secreted into the CSF and could be detected there. Analyzing postmortem human CSF samples, first author Qiu-Lan Ma and colleagues confirmed this hunch. They were able to detect SORLA in the CSF samples, and found that its expression was reduced in autopsy-confirmed AD cases. Furthermore, they showed that SORLA levels in CSF correlated strongly with Aβ42 concentrations.

“Taken together, these observations support the hypothesis that [SORLA] is directly involved in the pathogenesis of Alzheimer disease,” wrote Richard Mayeux, Columbia University, New York, and Peter St. George-Hyslop, University of Toronto, Canada, in an editorial accompanying the study. “More importantly, it is clear that a better understanding of subcellular trafficking of APP as well as the various functional roles of SORL1 may point to a novel therapeutic strategy that has not yet been considered.”—Esther Landhuis

Comments

  1. CSF levels of Aβ1-42, tau and p-tau can discriminate Alzheimer disease patients from controls or from patients with other types of dementia, and can identify incipient AD among patients with mild cognitive impairment (Hansson et al., 2006; Blennow et al., 2003). However, large differences are reported in absolute biomarker levels between centers (Sunderland et al., 2003), making it difficult to set up multi-center studies including multicenter treatment trials. To overcome this problem, external quality control assessment schemes (EQAS) are needed (Libeer et al., 2001). Our study describes the first worldwide EQAS for CSF biomarkers (20 different centres measured the same CSF samples), reporting relatively high inter-center variations, especially for Aβ1-42 (>22 percent). Lower inter-center variability (

    References:

    Hansson O, Zetterberg H, Buchhave P, Londos E, Blennow K, Minthon L. Association between CSF biomarkers and incipient Alzheimer's disease in patients with mild cognitive impairment: a follow-up study. Lancet Neurol. 2006 Mar;5(3):228-34. PubMed.

    Blennow K, Hampel H. CSF markers for incipient Alzheimer's disease. Lancet Neurol. 2003 Oct;2(10):605-13. PubMed.

    Sunderland T, Linker G, Mirza N, Putnam KT, Friedman DL, Kimmel LH, Bergeson J, Manetti GJ, Zimmermann M, Tang B, Bartko JJ, Cohen RM. Decreased beta-amyloid1-42 and increased tau levels in cerebrospinal fluid of patients with Alzheimer disease. JAMA. 2003 Apr 23-30;289(16):2094-103. PubMed.

    Libeer JC. Role of external quality assurance schemes in assessing and improving quality in medical laboratories. Clin Chim Acta. 2001 Jul 20;309(2):173-7. PubMed.

    View all comments by Niek Verwey

  2. This paper nicely describes the use of an elegant method for assessing brain-relevant biomarkers of protein metabolism in humans.

    View all comments by Lary Walker

  3. This paper presents interesting new data on a possible mechanism responsible for amyloidβ abnormality in patients with Alzheimer disease.

    View all comments by Richard C. Mohs

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. CSF Aβ—New Approach Shows Rapid Flux, May Help Evaluate Therapeutics
  2. ADNI: GWA Nearly Complete, Biomarker Analysis Update
  3. Research Brief: New Methods for Aβ Detection, Production
  4. SORLA Soars—Large Study Links Gene to Late-onset AD

Therapeutics Citations

  1. Semagacestat

Paper Citations

  1. Bateman RJ, Munsell LY, Morris JC, Swarm R, Yarasheski KE, Holtzman DM. Human amyloid-beta synthesis and clearance rates as measured in cerebrospinal fluid in vivo. Nat Med. 2006 Jul;12(7):856-61. PubMed.
  2. Siemers ER, Quinn JF, Kaye J, Farlow MR, Porsteinsson A, Tariot P, Zoulnouni P, Galvin JE, Holtzman DM, Knopman DS, Satterwhite J, Gonzales C, Dean RA, May PC. Effects of a gamma-secretase inhibitor in a randomized study of patients with Alzheimer disease. Neurology. 2006 Feb 28;66(4):602-4. PubMed.
  3. Siemers ER, Dean RA, Friedrich S, Ferguson-Sells L, Gonzales C, Farlow MR, May PC. Safety, tolerability, and effects on plasma and cerebrospinal fluid amyloid-beta after inhibition of gamma-secretase. Clin Neuropharmacol. 2007 Nov-Dec;30(6):317-25. PubMed.
  4. Bateman RJ, Wen G, Morris JC, Holtzman DM. Fluctuations of CSF amyloid-beta levels: implications for a diagnostic and therapeutic biomarker. Neurology. 2007 Feb 27;68(9):666-9. PubMed.
  5. Shaw LM, Vanderstichele H, Knapik-Czajka M, Clark CM, Aisen PS, Petersen RC, Blennow K, Soares H, Simon A, Lewczuk P, Dean R, Siemers E, Potter W, Lee VM, Trojanowski JQ, Alzheimer's Disease Neuroimaging Initiative. Cerebrospinal fluid biomarker signature in Alzheimer's disease neuroimaging initiative subjects. Ann Neurol. 2009 Apr;65(4):403-13. PubMed.
  6. Xia W, Yang T, Shankar G, Smith IM, Shen Y, Walsh DM, Selkoe DJ. A specific enzyme-linked immunosorbent assay for measuring beta-amyloid protein oligomers in human plasma and brain tissue of patients with Alzheimer disease. Arch Neurol. 2009 Feb;66(2):190-9. PubMed.
  7. Okereke OI, Xia W, Irizarry MC, Sun X, Qiu WQ, Fagan AM, Mehta PD, Hyman BT, Selkoe DJ, Grodstein F. Performance characteristics of plasma amyloid-beta 40 and 42 assays. J Alzheimers Dis. 2009;16(2):277-85. PubMed.
  8. Rogaeva E, Meng Y, Lee JH, Gu Y, Kawarai T, Zou F, Katayama T, Baldwin CT, Cheng R, Hasegawa H, Chen F, Shibata N, Lunetta KL, Pardossi-Piquard R, Bohm C, Wakutani Y, Cupples LA, Cuenco KT, Green RC, Pinessi L, Rainero I, Sorbi S, Bruni A, Duara R, Friedland RP, Inzelberg R, Hampe W, Bujo H, Song YQ, Andersen OM, Willnow TE, Graff-Radford N, Petersen RC, Dickson D, Der SD, Fraser PE, Schmitt-Ulms G, Younkin S, Mayeux R, Farrer LA, St George-Hyslop P. The neuronal sortilin-related receptor SORL1 is genetically associated with Alzheimer disease. Nat Genet. 2007 Feb;39(2):168-77. PubMed.
  9. Scherzer CR, Offe K, Gearing M, Rees HD, Fang G, Heilman CJ, Schaller C, Bujo H, Levey AI, Lah JJ. Loss of apolipoprotein E receptor LR11 in Alzheimer disease. Arch Neurol. 2004 Aug;61(8):1200-5. PubMed.
  10. Zhao Y, Cui JG, Lukiw WJ. Reduction of sortilin-1 in Alzheimer hippocampus and in cytokine-stressed human brain cells. Neuroreport. 2007 Jul 16;18(11):1187-91. PubMed.
  11. Kölsch H, Jessen F, Wiltfang J, Lewczuk P, Dichgans M, Kornhuber J, Frölich L, Heuser I, Peters O, Schulz JB, Schwab SG, Maier W. Influence of SORL1 gene variants: association with CSF amyloid-beta products in probable Alzheimer's disease. Neurosci Lett. 2008 Jul 25;440(1):68-71. PubMed.
  12. Ma QL, Teter B, Ubeda OJ, Morihara T, Dhoot D, Nyby MD, Tuck ML, Frautschy SA, Cole GM. Omega-3 fatty acid docosahexaenoic acid increases SorLA/LR11, a sorting protein with reduced expression in sporadic Alzheimer's disease (AD): relevance to AD prevention. J Neurosci. 2007 Dec 26;27(52):14299-307. PubMed.

Other Citations

  1. biomarkers

External Citations

  1. C2N Diagnostics

Further Reading

Papers

  1. Okereke OI, Xia W, Irizarry MC, Sun X, Qiu WQ, Fagan AM, Mehta PD, Hyman BT, Selkoe DJ, Grodstein F. Performance characteristics of plasma amyloid-beta 40 and 42 assays. J Alzheimers Dis. 2009;16(2):277-85. PubMed.
  2. Kölsch H, Jessen F, Wiltfang J, Lewczuk P, Dichgans M, Kornhuber J, Frölich L, Heuser I, Peters O, Schulz JB, Schwab SG, Maier W. Influence of SORL1 gene variants: association with CSF amyloid-beta products in probable Alzheimer's disease. Neurosci Lett. 2008 Jul 25;440(1):68-71. PubMed.
  3. Bateman RJ, Munsell LY, Morris JC, Swarm R, Yarasheski KE, Holtzman DM. Human amyloid-beta synthesis and clearance rates as measured in cerebrospinal fluid in vivo. Nat Med. 2006 Jul;12(7):856-61. PubMed.
  4. Fagan AM, Roe CM, Xiong C, Mintun MA, Morris JC, Holtzman DM. Cerebrospinal fluid tau/beta-amyloid(42) ratio as a prediction of cognitive decline in nondemented older adults. Arch Neurol. 2007 Mar;64(3):343-9. Epub 2007 Jan 8 PubMed.

Primary Papers

  1. Bateman RJ, Siemers ER, Mawuenyega KG, Wen G, Browning KR, Sigurdson WC, Yarasheski KE, Friedrich SW, Demattos RB, May PC, Paul SM, Holtzman DM. A gamma-secretase inhibitor decreases amyloid-beta production in the central nervous system. Ann Neurol. 2009 Jul;66(1):48-54. PubMed.
  2. Mayeux R, St George-Hyslop P. Brain traffic: subcellular transport of the amyloid precursor protein. Arch Neurol. 2009 Apr;66(4):433-4. PubMed.
  3. Verwey NA, van der Flier WM, Blennow K, Clark C, Sokolow S, De Deyn PP, Galasko D, Hampel H, Hartmann T, Kapaki E, Lannfelt L, Mehta PD, Parnetti L, Petzold A, Pirttila T, Saleh L, Skinningsrud A, Swieten JC, Verbeek MM, Wiltfang J, Younkin S, Scheltens P, Blankenstein MA. A worldwide multicentre comparison of assays for cerebrospinal fluid biomarkers in Alzheimer's disease. Ann Clin Biochem. 2009 May;46(Pt 3):235-40. PubMed.
  4. Ma QL, Galasko DR, Ringman JM, Vinters HV, Edland SD, Pomakian J, Ubeda OJ, Rosario ER, Teter B, Frautschy SA, Cole GM. Reduction of SorLA/LR11, a sorting protein limiting beta-amyloid production, in Alzheimer disease cerebrospinal fluid. Arch Neurol. 2009 Apr;66(4):448-57. PubMed.

Annotate

To make an annotation you must Login or Register.