Research ArticleNEUROSCIENCE

A plasma protein classifier for predicting amyloid burden for preclinical Alzheimer’s disease

See allHide authors and affiliations

Science Advances  06 Feb 2019:
Vol. 5, no. 2, eaau7220
DOI: 10.1126/sciadv.aau7220
  • Fig. 1 Pyramid plot to display the effect sizes (Cohen’s d) of protein significantly (P = <0.05) associated with Aβ burden (Aβ− versus Aβ+).

    On the right are proteins associated with cognitively unimpaired individuals and the association with the addition of individuals with MCI and AD on the left. Gray bars illustrate a nonsignificant effect size.

  • Fig. 2 Protein classifier to predict Aβ positivity in cognitively unimpaired individuals.

    (A) Graph showing the AUC statistic of the 50 classifier models produced using the “cognitively unimpaired cohort” training dataset. The AUC when testing each classifier model in the training dataset is in black, and the AUC when testing the classifier model in the testing dataset (KARVIAH) is in orange. On the x axis is the number of features used in each classifier model. For the classifier with the best AUC in the testing dataset (this was the classifier that used 12 features; Table 5), three graphs access the classifier’s performance: (B) ROC curve, (C) sensitivity and specificity plotted in black and orange, respectively, and (D) PPV and NPV plotted in black and orange, respectively.

  • Fig. 3 Protein classifier to predict Aβ positivity that includes participants with MCI and AD.

    (A) Graph showing the AUC statistic of the 50 classifier models produced using the “mixed diagnosis cohort” training dataset. The AUC when testing each classifier model in the training dataset is in black, and the AUC when testing the classifier model in the testing dataset (KARVIAH) is in orange. On the x axis is the number of features used in each classifier model. For the classifier with the best AUC in the testing dataset (this was the classifier that used 10 features; Table 6), three graphs access the classifier’s performance: (B) ROC curve, (C) sensitivity and specificity plotted in black and orange, respectively, (D) PPV and NPV plotted in black and orange, respectively.

  • Table 1 Subject demographics for cognitively unimpaired individuals (AIBL, n = 144; KARVIAH, n = 94).

    MMSE, Mini Mental State Examination; ns, not significant.

    AIBL discovery cohort (11C-PiB)KARVIAH replication cohort (18FBB)
    Aβ−Aβ+P valueAβ−Aβ+P value
    Number of subjects (n)100445935
    Aβ PET SUVR
    [means (SD)]
    1.16 (0.1)1.90 (0.3)1.11 × 10−541.16 (0.1)1.70 (0.2)8.35 × 10−23
    Gender; females
    [n (%)]
    51 (50.0)21 (50.0)ns40 (67.8)21 (60.0)ns
    Age in years
    [means (SD)]
    70.8 (7.1)75.5 (6.9)0.00177.5 (5.5)79.6 (5.3)ns
    APOE ε4 carrier [n (%)]27 (27)27 (61.4)7.70 × 10−46 (10.2)15 (42.9)0.001
    MMSE [means (SD)]28.9 (1.2)28.5 (1.3)ns29.4 (1.6)29.1 (1.4)ns
  • Table 2 Subject demographics for the mixed diagnosis cohort (AIBL, n = 190; KARVIAH, n = 94).

    n/a, not available.

    AIBL discovery cohort (11C-PiB)KARVIAH replication cohort (18FBB)
    Aβ−Aβ+P valueAβ−Aβ+P
    Number of subjects (n)108825935
    Aβ PET SUVR
    [means (SD)]
    1.16 (0.1)2.09 (0.4)7.44 × 10−601.16 (0.1)1.70 (0.2)8.35 × 10−23
    Gender; females
    [n (%)]
    54 (50.0)40 (48.7)ns40 (67.8)21 (60.0)ns
    Age in years
    [means (SD)]
    71.17 (7.2)74.04 (7.7)0.01177.5 (5.5)79.6 (5.3)ns
    Clinical diagnosis [n (%)]
    Cognitively unimpaired
    MCI
    AD
    100 (92.6)
    7 (6.5)
    1 (0.9)
    44 (53.7)
    14 (17.1)
    24 (29.2)
    3.38 × 10−1159 (100.0)
    n/a
    n/a
    35 (100.0)
    n/a
    n/a
    n/a
    APOE ε4 carrier [n (%)]28 (25.9)54 (65.9)5.91 × 10−56 (10.2)15 (42.9)0.001
    MMSE [means (SD)]28.7 (1.4)26.1 (4.0)5.96 × 10−729.4 (1.6)29.1 (1.4)ns
  • Table 3 GLM-adjusted protein groups significantly associated with Aβ SUVR in cognitively unimpaired participants stratified by Aβ+/− classification after multiple testing correction.

    Protein groups were also associated with Aβ SUVR with an adjustment for APOE genotype. All protein groups that are nominally associated with Aβ in cognitively unimpaired (P > 0.05) are shown in table S1.

    UniProt IDProtein group
    description
    Aβ+ groupAdjusted for APOE ε4 status
    tMean
    difference
    P valueQtMean
    difference
    P valueQ value
    P05067APP−5.651−0.2904.57 × 10−082.08x10−05−4.686−0.2414.71 × 10−060.001
    Q9H2A3NGN2−5.556−0.3197.43 × 10−082.08x10−05−4.787−0.2762.99 × 10−060.001
    P07196NfL−4.639−0.2295.80 × 10−060.001−3.716−0.1842.53 × 10−040.047
    O95704APBB3−4.389−0.2741.71 × 10−050.002−3.374−0.2100.001ns
    Q13127REST3.5700.1424.33 × 10−040.0483.3850.1350.001ns
  • Table 4 GLM-adjusted protein groups significantly associated with Aβ SUVR in all subjects stratified by Aβ+/− classification after Benjamini-Hochberg multiple testing corrections.

    Protein groups were also associated with Aβ SUVR with an adjustment for APOE genotype. All protein groups that are nominally associated with Aβ (P > 0.05) are shown in table S2 (A and B).

    UniProt IDProtein group
    description
    Aβ+ groupAdjusted for APOE ε4 status
    tMean differenceP valueQtMean differenceP valueQ value
    P05067Aβ A4 protein−6.419−0.2795.79 × 10−103.24 × 10−07−5.019−0.2209.22 × 10−070.001
    Q9H2A3NGN2−5.702−0.2832.98 × 10−088.34 × 10−06−4.624−0.2315.75 × 10−060.002
    P07196NfL−5.258−0.2192.88 × 10−075.38 × 10−05−3.928−0.1641.08 × 10−040.015
    O95704APBB3−5.189−0.2724.05 × 10−075.67 × 10−05−3.710−0.1942.49 × 10−040.028
    Q13127REST4.2470.1432.94 × 10−053.29 × 10−033.9840.1358.64 × 10−050.015
    P81274GPSM24.0160.1247.61 × 10−050.0073.5640.1114.28 × 10−040.040
    Q13103SPP23.8330.1081.56 × 10−040.0133.3100.0940.0010.084
    Q8IVF4DNAH10
    (axonemal)
    −3.548−0.0904.54 × 10−040.032−3.165−0.0810.0020.120
  • Table 5 Feature list for multianalyte classifier predicting elevated Aβ burden in cognitively unimpaired cohort.

    The classifier was training in the AIBL cohort (n = 144), achieving a testing AUC of 0.891 in the KARVIAH cohort (n = 94). GPCR, G protein–coupled receptor; sens, sensitivity; spec, specificity.

    Feature
    no.
    Cognitively unimpaired classifier for Aβ burden
    (Testing AUC = 0.891, sens = 0.78, spec = 0.77, NPV = 0.68, PPV = 0.85)
    UniProt IDFeature
    description
    Tryptic amino acid peptide sequences contributing to protein scores
    1P00734Prothrombin87–94, 98–116, 125–133, 178–198, 199–224, 217–224, 225–243, 248–263, 315–327, 328–344, 345–363,
    354–363, 441–447, 447–452, 467–468, 518–537, 544–560, and 560–575
    2Q8IZF3Adhesion GPCR
    F4
    100–107, 315–328, 657–689, and 673–698
    3P05067Aβ A4 protein677–687 and 688–699
    4Q9H2A3NGN2114–120
    5n/aAPOE ε4 countn/a
    6Q8IVF4DNAH10
    (axonemal)
    207–223, 528–540, 879–915, 1133–1148, 1524–1528, 1870–1927, 2404–2436, 2906–2941, 2923–2945, and
    4100–4139
    7Q13127REST353–369, 727–742, and 882–896
    8P07196NfL137–144, 154–164, 318–331, and 447–462
    9P51812RPS6KA380–100, 109–132, 193–216, 226–242, 274–282, 508–525, and 570–591
    10P81274GPSM2267–278, 318–347, and 422–433
    11B1AJZ9FHAD1861–868, 876–887, 897–905, 931–942, 1030–1052, 1059–1070, and 1104–1139
    12n/aAge of
    participant
    n/a
  • Table 6 Feature list for multianalyte classifier predicting elevated Aβ burden in a mixed diagnosis cohort.

    The classifier was training in the AIBL cohort (n = 169), achieving a testing AUC of 0.905 in the KARVIAH cohort (n = 94).

    Feature no.Mixed diagnosis classifier for Aβ burden
    (Testing AUC = 0.904, sens = 0.81, spec = 0.80, NPV = 0.72, PPV = 0.87)
    UniProt IDFeature
    description
    Tryptic amino acid peptide sequences contributing to protein scores
    1n/aAPOE ε4 count*n/a
    2P05067Aβ A4 protein*677–687 and 688–699
    3P07196NfL*137–144, 154–164, 318–331, and 447–462
    4Q9H2A3NGN2*114-120
    5Q8IVF4DNAH10
    (axonemal)*
    207–223, 528–540, 879–915, 1133–1148, 1524–1528, 1870–1927, 2404–2436, 2906–2941,
    2923–2945, and 4100–4139
    6Q13127REST*353–369, 727–742, and 882–896
    7O95704APBB3155–175 and 414–430
    8P81274GPSM2*267–278, 318–347, and 422–433
    9P00734Prothrombin*87-94, 98–116, 125–133, 178–198, 199–224, 217–224, 225–243, 248–263, 315–327, 328–344,
    345–363, 354–363, 441–447, 447–452, 467–468, 518–537, 544–560, and 560–575
    10B1AJZ9FHAD1*861–868, 876–887, 897–905, 931–942, 1030–1052, 1059–1070, and 1104–1139

    *Proteins also included in the classifier for predicting Aβ in the cognitively unimpaired individuals.

    Supplementary Materials

    • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/5/2/eaau7220/DC1

      Table S1. Residual scores for all 560 protein groups.

      Table S2. The association of plasma protein groups with Aβ SUVR in cognitively unimpaired subjects.

      Table S3. The association of plasma protein groups with Aβ SUVR in all subjects.

    • Supplementary Materials

      Download PDF

      Other Supplementary Material for this manuscript includes the following:

      • Table S1 (Microsoft Excel format). Residual scores for all 560 protein groups.
      • Table S2 (Microsoft Excel format). The association of plasma protein groups with Aβ SUVR in cognitively unimpaired subjects.
      • Table S3 (Microsoft Excel format). The association of plasma protein groups with Aβ SUVR in all subjects.

      Files in this Data Supplement:

    Navigate This Article