Growth-associated protein 43 and tensor-based morphometry indices in mild cognitive impairment

doi:10.21203/rs.3.rs-3805148/v1

Background

Growth-associated protein 43 (GAP-43) is found in the axonal terminal of neurons in the limbic system, which is affected in people with Alzheimer's disease (AD). We assumed GAP-43 may contribute to AD progression and serve as a biomarker. So, in a two-year follow-up study, we assessed GAP-43 changes and whether they are correlated with tensor-based morphometry (TBM) findings in patients with mild cognitive impairment (MCI).

Methods

We included MCI and cognitively normal (CN) people with available baseline and follow-up cerebrospinal fluid (CSF) GAP-43 and TBM findings from the ADNI database. We assessed the difference between the two groups and correlations in each group at each time point.

Results

CSF GAP-43 and TBM measures were similar in the two study groups in all time points, except for the accelerated anatomical region of interest (ROI) of CN subjects that were significantly greater than those of MCI. The only significant correlations with GAP-43 observed were those inverse correlations with accelerated and non-accelerated anatomical ROI in MCI subjects at baseline. Plus, all TBM metrics decreased significantly in all study groups during the follow-up in contrast to CSF GAP-43 levels.

Conclusions

Our study revealed significant associations between CSF GAP-43 levels and TBM indices among people of the AD spectrum.

Growth-associated protein 43

Alzheimer’s disease

tensor-based morphometry

mild cognitive impairment

region of interest

Alzheimer’s disease (AD) is a globally widespread neurodegenerative disorder expected to rise drastically within the upcoming years (Esquerda-Canals, Montoliu-Gaya et al. 2017). AD primarily manifests with deficiencies in forming recent memories and progress to behavioral and cognitive changes (Soria Lopez, Gonzalez et al. 2019). Various areas of the brain can be damaged through considerable neuronal loss and neurological injuries by the time of diagnosis (Mantzavinos and Alexiou 2017). AD is also characterized by degenerative changes in a variety of neurotransmitter systems in various brain regions. These regions include the temporal and parietal lobes and restricted regions within the frontal cortex and cingulate gyrus (Wenk 2003). The underlying mechanisms in AD are associated with the amorphous aggregates of Aβ and hyperphosphorylated tau in brain regions involved in memory such as the hippocampus, entorhinal cortex, amygdala and basal forebrain, which lead to reduced numbers of synapses and damaged neurites (West, Coleman et al. 1994, Mattson 2004). Other depositions such as the growth-associated protein 43 (GAP-43) may also contribute to AD progression (Kumar, Sharma et al. 2020), which can be used for differential diagnosis (Sandelius, Portelius et al. 2019). Preclinical research supports the hypothesis of increased GAP-43 level in association with memory decline in the AD spectrum (Holahan, Honegger et al. 2007).

GAP-43 is found in the axonal terminal of neurons in the limbic system especially mossy fibers, which are affected in AD patients (Kiktenko, Uranova et al. 1995, Denny 2006). GAP-43 physiologically binds to phosphatidylinositol 4,5-bisphosphate (PIP₂) in the cell membrane, leading to the removal of PIP₂'s inhibition on actin filaments (Denny 2006). Therefore, GAP-43 increases and interacts with protein kinase, leading to neuroplasticity and long-term potentiation (Ramakers, McNamara et al. 1999, Denny 2006). Increased levels of the GAP-43 in the cerebrospinal fluid (CSF) of people with AD propose its relation to synaptic dysfunction and disease stage (Denny 2006, Zhang, Lyu et al. 2021). Previous studies of other neurodegenerative diseases have also suggested GAP-43 as a potential biomarker in these diseases. This protein's mRNA is reported to increase by two- to four-fold in the anterior horn cells of people with amyotrophic lateral sclerosis compared to healthy controls (Parhad, Oishi et al. 1992). Moreover, nigrostriatal systems in Parkinson's disease express more GAP-43 when compared with matched controls, although this increase is not significant (Saal, Galter et al. 2017).

Novel neuroimaging modalities help find the characteristic changes in the brain of people with neurologic and psychiatric disorders (Teipel, Drzezga et al. 2015). Tensor-based morphometry (TBM) is an imaging analysis technique in which a series of three-dimensional images are obtained to evaluate brain structural changes under different conditions. These images can be compared with standardized images in order to find out any deviations from standard group (Hua, Gutman et al. 2011). TBM uses warping process in which a group of images are non-rigidly registered into a template image and determines deformation fields to discover any volumetric changes (Ashburner, Good et al. 2000). There are other structural imaging modalities such as diffusion tensor imaging (DTI), which analyzes diffusion anisotropy effects in MR images to track fibers in the brain (Le Bihan, Mangin et al. 2001), and voxel based morphometry (VBM), which uses voxel-wise comparisons to detect subtle and gross brain structural changes (Mechelli, Price et al. 2005). Compared to the mentioned modalities, TBM can provide a less biased method as it can handle low-quality scans, resulting in less exclusion of data in clinical trials due to their quality (Hua, Hibar et al. 2013). Moreover, TBM can deliver regional assessments varying from the voxel level to whole-brain analysis. Also, it estimates substructure volumes by integrating Jacobian determinant values over a candidate region of interest (Hua, Leow et al. 2008). TBM scans have shown a variety of structural and morphologic brain changes in AD patients such as ventricular/CSF expansion and brain tissue loss (atrophy) (Gutman, Wang et al. 2015).

Previous studies have shown that there might be associations between changes in the molecular and neuroimaging findings of nervous system disorders (Femminella, Thayanandan et al. 2018, Simrén, Leuzy et al. 2021) Given the information above, we hypothesized that there might be such an association in patients with mild cognitive impairment (MCI). MCI represents a transitional phase between healthy aging and clinical AD (Anderson 2019). Therefore, we aimed to investigate how GAP-43 changes and whether it is correlated with TBM findings in patients with MCI of Alzheimer’s dementia spectrum in two years of follow-up.

2.1. Participants and Design

Alzheimer's Disease Neuroimaging Initiative (ADNI) is a longitudinal multicenter study initially launched in 2004 as a public-private partnership funded primarily by the National Institute on Aging (NIA) and the contribution of various organizations such as the National Institute of Biomedical Imaging and Bioengineering (NIBIB), the Food and Drug Administration (FDA), industrial companies and foundations (Mueller, Weiner et al. 2005, Mueller, Weiner et al. 2005). The primary aim of ADNI is to develop clinical, imaging, genetic, and biochemical biomarkers for the early detection and tracking of AD. The imaging and clinical data used in this study were obtained from ADNI public database (www.loni.ucla.edu/ADNI). Since its launch in 2004, ADNI has collected data regarding people with AD and MCI and cognitively normal (CN) controls. In this study, we only included MCI and CN people with available CSF GAP-43 and TBM metrics in baseline and 24-month follow-up. In addition to the cross-sectional assessment of the participants, we longitudinally evaluated subjects after two years regarding the study outcomes. All participants were examined with the proper clinical and cognitive assessment tools at the time of scanning to make sure that each subject is in the proper category. The cognitive assessments of each category are discussed below in detail. The patient exclusion and inclusion criteria were based on criteria from ADNI protocol (Mueller, Weiner et al. 2005, Mueller, Weiner et al. 2005). We made sure that Good Clinical Practice guidelines and the declaration of Helsinki (PP 1964) were applied to each step of our study.

2.2. Cognitive Assessments

The ADNI subjects were divided into groups of AD, MCI, and CN and we only included MCI and CN subjects. CN people had no memory complaints, but MCI subjects reported memory complaints, either by themselves or their verified study partners. In relation to MMSE (Mini-Mental State Examination) (Folstein, Folstein et al. 1975) results, ranging from 24 to 30 are consistent with CN and MCI. On the CDR (Clinical Dementia Rating) (Hughes, Berg et al. 1982), the score for the CN group was 0 and for subjects with MCI was 0.5, with a mandatory requirement being 0.5 or greater in the memory box score. For the memory criterion, the Logical Memory II subscale (delayed recall) of the Wechsler Memory Scale-Revised (Powel 1988) (maximum score of 25) was used, and cutoff scores as follows were applied based on education: CN ≥ 9 for 16 years of education, ≥ 5 for 8–15 years of education, and ≥ 3 for 0–7 years of education. For subjects with MCI, these scores were ≤ 8, ≤ 4, and ≤ 2, respectively. In brief, CN participants had no significant impairment in cognitive functions or everyday activities; the general cognition and function of MCI subjects were preserved that they did not meet the AD diagnosis by an ADNI physician.

2.3. MRI Acquisition

All participants were scanned according to the standard ADNI MRI protocol (Jack, Bernstein et al. 2008). Using 1.5 Tesla MRI scanners from General Electric Healthcare and Siemens Medical Solutions, high-resolution structural brain MRI scans were acquired at multiple ADNI sites. For each participant, two T1-weighted MRI scans were collected using a sagittal 3D MP-RAGE scanning protocol in a 1.5 Tesla magnetic field strength, with a repetition time (TR) of 2400 ms, minimum full echo time (TE) inversion time (TI) of 1000 ms, flip angle of 8°, 24 cm field of view, with a 192×192×166 acquisition matrix in the x-, y-, and z-dimensions, yielding a voxel size of 1.25×1.25×1.2 mm3. In-plane, zero-filled reconstruction yielded a 256×256 matrix for a reconstructed voxel size of 0.9375× 0.9375×1.2 mm3. To ensure consistency among scans acquired at different sites, the images were calibrated with phantom-based geometric corrections (Gunter, Bernstein et al. 2005, Jack, Bernstein et al. 2008).

More image corrections were also applied, using a processing pipeline at the Mayo Clinic, consisting of a procedure for correction of geometric distortion due to gradient non-linearity (Jovicich, Czanner et al. 2006), known as GradWarp, adjustment of image intensity inhomogeneity due to B1 nonuniformity, B1-correction, using calibration scans (Jack, Bernstein et al. 2008), N3 bias field correction for reducing residual intensity inhomogeneity (Sled, Zijdenbos et al. 1998), and a scale factor, according to a phantom scan acquired for each participant, to adjust for scanner- and possible session-specific calibration errors and geometrical scaling (Gunter, Bernstein et al. 2006, Jack, Bernstein et al. 2008). Both original uncorrected image files and images with all of these corrections are available to the scientific community at http://www.loni.ucla.edu/ADNI/Data/.

2.4. MRI Preprocessing

To adjust for global differences in scale and position of the brain across all individuals, all scans were linearly registered to International Consortium for Brain Mapping (ICBM-53) (Mazziotta, Toga et al. 2001). With the use of a 9-parameter (9P) transformation (3 translations, 3 rotations, and 3 scales) by use of the Minctracc algorithm (Collins, Neelin et al. 1994). The aligned images were resampled in an isotropic space of 220 voxels along each axis (x, y, and z) with a final voxel size of 1 mm3. For comparison between groups, a Minimal deformation Target (MDT) was created based on 40 randomly selected normal subjects with a mean age of 40-year-olds.To quantify a 3D model of volumetric tissue changes, all individual brain images were aligned nonlinearly with the MDT for the normal group (Leow, Thompson et al. 2005). For each subject, separate Jacobian matrix fields were obtained from the stem field gradients, where the individual brains were aligned with the MDT model. Color-coded Jacobin determinants were used to indicate regions of volume expansion, i.e., det J (r) > 1, or contraction regions, i.e., regions where det J (r) < 1 compared to the general group model(Freeborough and Fox 1998, Toga 1999).

Illustrating the systematic differences and changes between groups in this paragraph using statistical maps, we created Jacobian maps with TBM that show individual shifts from the normal brain temple. To show systematic changes in the groups, we built a statistical map according to the Z statistic. We calculated the total importance of group differences using Jacobian methods and double-checked for various comparisons with permutation tests (Bullmore, Suckling et al. 1999). Briefly, a null distribution for group differences in the Jacobian map (brain atrophy rate) was calculated using 5000 random relabelling, and for each time, new Z statistics were considered. At each voxel, a re-randomized P-value was calculated by the ratio of times relabelled groups' Z statistics were more than original groups' Z statistics. In order to calculate the amount of brain atrophy at the regional level in each group, binary masks were developed for each lobe (frontal, temporal, peritoneal, or occipital) by an upskilled anatomist utilizing the partial volume classification (PVC) algorithm of the Brainsuit software program (Shattuck and Leahy 2002). Furthermore, an alternative statistically-defined ROI (stat-ROI) was defined based on voxels with significant atrophic rates over time (p < 0.001) within the bounds temporal lobes, in a non-overlapping training complex of 22 AD patients accessible at the time of image download, i.e., independent of the testing set used to compute the power numbers. In principle, ROI can be selected specifically to provide maximum power. A separate stat-ROI was generated and applied to each TBM pattern with different algorithms (3DMI vs. sKL-MI) and different framework settings (6 versus 9P linear registration and different S and L). In a nutshell, the average atrophy rate for all ROI-containing vasculature - was calculated for each individual to summarize the annual changes in ROI (Hua, Lee et al. 2009).

2.5. GAP-43 Measurement

A board-certified laboratory technician measured cerebrospinal fluid GAP-43 levels using enzyme-linked immunoassay (ELISA) technology by an in-house ELISA method performed at the Clinical Neurochemistry Lab, University of Gothenburg, Sweden. The ELISA was developed by using a combination of monoclonal GAP-43 antibody NM4 (coating antibody) and polyclonal GAP-43 antibody ABB-135 (detector antibody), capable of recognizing the C-terminal of GAP-43. Values are reported as pg/mL. Quality control CSF samples were from the Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, Mölndal, Sweden. The ELISA assay of CSF GAP-43 is described previously in detail (Sandelius, Portelius et al. 2019).

2.6. Statistical Analysis

Data analysis was conducted using the SPSS software version 29 (IBM Corp, Armonk, New York, USA) and the figures were prepared via GraphPad Prism version 10 (GraphPad Prism Software Inc., San Diego, CA). We presented continuous variables as mean ± SD, and categorical variables as frequencies with percentages. Shapiro-Wilk test was used to test the compliance of variables for normal distribution. We used the independent sample t-tests (with Levene test for variance equality) to compare normally-distributed continuous variables and the Mann-Whitney U test was used for skewed-distributed variables. Pearson correlation analysis was used to assess the associations between CSF Gap-43 levels and TBM-derived metrics at each timepoint. In order to longitudinally assess the changes of these variables after 24 months, we also performed paired t-tests or Wilcoxon signed-rank according to the normality of the data. Finally, we performed a general linear regression model for each group (CN/MCI) at each timepoint to find out if CSF GAP-43 values can predict TBM-derived metrics, covarying age and sex. P-values less than 0.05 were considered significant.

3.1. Demographic characteristics of participants

We investigated CSF GAP-43 levels and TBM data of 70 participants, including 33 CN and 37 MCI subjects. Table 1 shows a summary of participants’ demographic data, handedness, education, and ApoE4 status. As shown, subjects of the two groups were statistically comparable regarding sex (p-value: 0.64, handedness (p-value: 0.48), education (p-value: 0.30), and ApoE4 status (p-value: 0.11). However, the independent t-test results showed that CN subjects were statistically older than those with MCI (p-value: 0.02). The 24-month follow-up data of all participants were available.

Table 1. Characteristic data of participants at baseline and follow-up

	CN	MCI	P-value*
N	33	37	-
Age (mean ± SD)	75.27 ± 6.04	71.81 ± 6.44	0.02
Sex (f/m)	19/15	18/19	0.64
Handedness (right/left)	30/3	31/6	0.48
Education (years)	16.73 ± 2.63	16.08 ± 2.63	0.30
ApoE4 (+/-) **	6/27	14/23	0.11

Abbreviations: CN: cognitively normal; f/m: female-to-male ratio; MCI: mild cognitive impairment; SD: standard deviation.
* The p-values of the Chi-square test for categorical variables, the independent t-test for parametric continuous variables, and the Mann-Whitney U test for non-parametric continuous variables are shown.
** The presence of at least one allele of ApoE4 was considered positive.

3.2. Comparison of TBM and GAP-43 data at each time point

We compared TBM metrics and CSF GAP-43 levels of CN and MCI groups at each time point. As presented in Table 2, the two study groups had statistically similar values of GAP-43 and all TBM metrics, except for accelerated anatomical ROI in the baseline; accelerated anatomical ROI values of CN subjects were significantly greater than those of MCI (t (68): 2.22, p-value: 0.03). However, the observed difference was no longer present at the follow-up.

Table 2. Imaging and cerebrospinal fluid markers of the participants at baseline and follow-up

	CN	MCI	P-value*
Baseline
GAP-43 (pg/ml)	5342.38 ± 2596.87	4935.26 ± 2117.59	0.66
Accelerated statistical ROI	987.16 ± 8.82	984.33 ± 12.25	0.59
Non-accelerated statistical ROI	984.73 ± 8.99	983.16 ± 12.47	0.88
Accelerated anatomical ROI	996.65 ± 6.83	992.90 ± 7.23	0.03
Non-accelerated anatomical ROI	994.52 ± 6.42	992.73 ± 7.56	0.29
Follow-up at 24 months
GAP-43 (pg/ml)	5358.02 ± 2336.90	5174.25 ± 2148.94	0.91
Accelerated statistical ROI	973.24 ± 13.35	973.25 ± 16.95	0.99
Non-accelerated statistical ROI	972.05 ± 12.47	970.23 ± 17.14	0.61
Accelerated anatomical ROI	989.57 ± 9.38	989.78 ± 9.07	0.92
Non-accelerated anatomical ROI	989.78 ± 9.13	989.19 ± 10.19	0.79

Abbreviations: CN: cognitively normal; GAP-43: Growth-associated protein 43; MCI: mild cognitive impairment; ROI: region of interest. All the continuous variables are presented as mean ± standard deviation.
* The p-values of the independent t-test for parametric continuous variables and the Mann-Whitney U test for non-parametric continuous variables are shown.

3.3. Correlation of TBM indices and GAP-43 levels

To find out the possible correlations between CSF GAP-43 and imaging markers, we also performed a Pearson correlation for each pair of GAP-43 level and TBM metric for both groups at baseline and follow-up. The results are shown in Table 3. As shown, in MCI subjects, values of accelerated and non-accelerated anatomical ROI were inversely associated with GAP-43 levels at baseline (r(35): -0.39, p-value: 0.01 and r(35): -0.37, p-value: 0.02, respectively). No significant correlation was observed at follow-up.

Table 3. Pearson correlations between Growth-associated protein-43 (GAP-43) levels and tensor-based morphometry (TBM) metrics at baseline and follow-up

			Accelerated statistical ROI	Non-accelerated statistical ROI	Accelerated anatomical ROI	Non-accelerated anatomical ROI
Baseline
GAP-43	CN	Correlation Coefficient	-0.14	-0.08	-0.04	0.05
	CN	P-value	0.42	0.63	0.82	0.75
	MCI	Correlation Coefficient	-0.12	-0.18	-0.39	-0.37
	MCI	P-value	0.44	0.28	0.01	0.02
Follow-up at 24 months
GAP-43	CN	Correlation Coefficient	-0.07	-0.21	-0.03	-0.10
	CN	P-value	0.68	0.22	0.85	0.55
	MCI	Correlation Coefficient	-0.02	-0.02	-0.30	-0.06
	MCI	P-value	0.86	0.88	0.07	0.69

Abbreviations: CN: cognitively normal; GAP-43: Growth-associated protein 43; MCI: mild cognitive impairment; ROI: region of interest.

3.4. Longitudinal changes of TBM indices and GAP-43 levels over the follow-up period

To assess the longitudinal changes of CSF GAP-43 and TBM metrics of each study group during 24 months of follow-up, we used paired t-tests and Wilcoxon signed-rank tests. Figure 1 shows the trajectory of each variable during follow-up. All TBM metrics showed significant reductions in both CN and MCI groups (p-values < 0.05). However, CSF GAP-43 levels did not show any significant change. Performing linear regression models for each study group, we also assessed if CSF GAP-43 values can predict TBM-derived metrics, covarying age and sex of participants. None of the regression models showed statistically significant results (data not shown).

In the present study, we assessed profiles of CSF GAP-43 and TMB in CN and MCI groups during a 24-month follow-up. Our findings indicated that all these CSF and imaging measures were comparable in the two study groups in all time points, except for the accelerated anatomical ROI of CN subjects that was significantly greater than those of MCI. The only significant correlations with GAP-43 observed were those with accelerated and non-accelerated anatomical ROI in MCI subjects at baseline, which were inverse. Moreover, although all TBM metrics decreased significantly in all study groups during the follow-up, CSF GAP-43 levels did not change statistically.

To explore brain changes in this study, we used an ROI confined to the temporal lobe, which is described as one of the brain regions that play a key role in episodic memory and is hit at the early stages of AD by neurofibrillary tangles (Chauveau, Kuhn et al. 2021). Evidence demonstrates the medial temporal atrophy (MTA) as a predictor of MCI progression to AD, as it is correlated with the disease progression and the rate of cognitive impairment in affected people (Dubois, Feldman et al. 2014, Weiler, Agosta et al. 2015). Moreover, whole brain analyses have revealed that not only the volume reduction of the hippocampus, amygdala, and other subregions of the medial temporal lobe (MTL) but also their intensity of shape asymmetry is of great value for disease staging and predicting the progression of cognitive impairment towards advanced stages (Wachinger, Salat et al. 2016). Some studies focused on other brain structures rather than the temporal lobe, such as brain ventricles, in this regard. Nonetheless, they found similar atrophy patterns in MCI converters to AD with the most prominent ventricular subregional atrophy, including the areas near the MTL structures and posterior cingulate gyrus (Shi, Stonnington et al. 2015).

The MTL of the human brain contains structures like the hippocampus and dentate, entorhinal, and parahippocampus gyri, which are essential for memory formation and are therefore commonly identified as an AD signature region (Lech and Suchan 2013). The hippocampus is mostly believed to play a role in converting short-term memory into long-term memory (Anand and Dhikav 2012) while the entorhinal cortex, connects the hippocampus with the rest of the neocortex and makes up the perforant path (Granger, Colon-Perez et al. 2023). In the early stages of AD, the atrophy of the hippocampus and the entorhinal cortex leads to functional disconnection from other parts of the brain and memory impairment (Rao, Ganaraja et al. 2022). There are other brain regions in MTL, such as the amygdala, the atrophy of which is associated with global cognitive function decline in people with early AD (Poulin, Dautoff et al. 2011). All this evidence implies how temporal lobe atrophy indicates the disease progression. TBM measures significantly dropped in all diagnostic groups after two years in our study.

As shown in the results, the ADNI database carried out a parallel MRI approach when assessing people with Alzheimer's disease spectrum. This method accelerates the MRI acquisition by acquiring reduced k-space data with an array of receiver coils (Deshmane, Gulani et al. 2012). Although acceleration through parallel MRI leads to savings in time and cost alongside fewer motion artifacts, it can cause a decreased signal-to-noise ratio and potentially increased image artifacts due to the reconstruction of images in less time (Blaimer, Breuer et al. 2004). In another ADNI study, Vemuri et al. compared the performance of unaccelerated and accelerated structural MRI in CN, MCI, and AD people (Vemuri, Senjem et al. 2015). They observed that accelerated MRI metrics had higher estimated values than unaccelerated ones; however, both scans had similar performance regarding discrimination among the three clinical groups (Vemuri, Senjem et al. 2015). In our study, accelerated and non-accelerated scans were not comparable in discriminating CN and MCI subjects in case of anatomical ROI, but had similar correlations regarding CSF GAP-43 in MCI. Furthermore, we reported values of both statistical and anatomical maps of the participants. The differences between these two maps are provided in the methods and materials section. In our study, only anatomical ROI values showed significant differences and correlations among groups. Therefore, our findings indicate different behavior of anatomical and statistical values in TBM.

As explained in the introduction, AD has long been considered a disease of tauopathy and abnormal amyloid-β accumulation (Chong, Ng et al. 2018). However, in this study, we explored CSF GAP-43 levels as a new biomarker in the AD spectrum and its possible associated temporal lobe atrophy. However, we found that GAP-43 levels do not discriminate between MCI and CN, which is in contrast to the previous reports (Qiang, Skudder-Hill et al. 2022). The Qiang et al. study has reported that CSF GAP-43 is also correlated with CSF phosphorylated tau 181, and its high levels are associated with longitudinal deterioration of cognitive scores in affected patients (Qiang, Skudder-Hill et al. 2022). Moreover, in a longitudinal study, Lu et al. reported that people with progressive MCI tend to have higher CSF GAP-43 than people with stable MCI. CSF GAP-43 levels were also a significant predictor of cognitive decline and brain glucose hypometabolism over time (Lu and for the Alzheimer’s Disease Neuroimaging 2022). All of this evidence suggests GAP-43 as a possible biomarker of AD.

Herein, we assessed the potential utility of CSF GAP-43 in relation to changes in TBM measures too. We reported various inverse correlations between its metrics and CSF GAP-43 in people with MCI. However, there is more evidence of the correlations between imaging findings and this biomarker. For example, increased GAP43 levels are reported to be significantly associated with faster hippocampal atrophy rates in structural MRI settings (Lan, Li et al. 2022). Even in the healthy relatives of AD patients, there are reports that higher CSF levels are associated with lower cortical thickness in AD-related brain regions (entorhinal, inferior temporal, middle temporal, and fusiform) (Milà-Alomà, Brinkmalm et al. 2021).

The results of our study should be interpreted in light of some limitations. These include the small sample size, as we only included subjects with available baseline and follow-up data. We did not also recruit people with clinical AD, due to very low sample size of this subgroup., i.e., four. Additionally, any TBM investigation is restricted by the precision with which deformable registration can match anatomical boundaries between individual brains and corresponding areas on the template (Leow, Yanovsky et al. 2009). We also only investigated the TBM measures in the temporal lobes and whole brain investigations are required to further expand findings in this regard. Moreover, other imaging parameters such as morphometrical measures should be evaluated in addition to the volumetric measures.

In conclusion, our study revealed significant associations between CSF GAP-43 levels and TBM indices among patients of AD spectrum. More studies with higher sample sizes, longer follow-ups, and optimized modalities like multivariate tensor-based morphometry (mTBM) might add up to the findings of our study. These studies help unravel the underlying mechanisms of this disease and, subsequently, can lead to modified treatments and prognostic tools for AD.

Statements and Declarations

6.1. Competing Interests

Authors declare that they have no competing interests to disclose regarding this work.

6.2. Funding

No funding was provided for this work by government or non-government organizations.

Anand KS, Dhikav V (2012) Hippocampus in health and disease: An overview. Ann Indian Acad Neurol 15(4):239–246
Anderson ND (2019) State of the science on mild cognitive impairment (MCI). CNS Spectr 24(1):78–87
Ashburner J, Good C, Friston KJ (2000) Tensor based morphometry NeuroImage 11(5):S465
Blaimer M, Breuer F, Mueller M, Heidemann RM, Griswold MA, Jakob PM (2004) SMASH, SENSE, PILS, GRAPPA: how to choose the optimal method. Top Magn Reson Imaging 15(4):223–236
Bullmore ET, Suckling J, Overmeyer S, Rabe-Hesketh S, Taylor E, Brammer MJ (1999) Global, voxel, and cluster tests, by theory and permutation, for a difference between two groups of structural MR images of the brain. IEEE Trans Med Imaging 18(1):32–42
Chauveau L, Kuhn E, Palix C, Felisatti F, Ourry V, de La Sayette V, Chételat G, de Flores R (2021) Medial temporal lobe subregional atrophy in aging and Alzheimer's disease: A longitudinal study. Frontiers in aging neuroscience 13
Chong FP, Ng KY, Koh RY, Chye SM (2018) Tau Proteins and Tauopathies in Alzheimer's Disease. Cell Mol Neurobiol 38(5):965–980
Collins DL, Neelin P, Peters TM, Evans AC (1994) Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space. J Comput Assist Tomogr 18(2):192–205
Denny JB (2006) Molecular mechanisms, biological actions, and neuropharmacology of the growth-associated protein GAP-43. Curr Neuropharmacol 4(4):293–304
Deshmane A, Gulani V, Griswold MA, Seiberlich N (2012) Parallel MR imaging. J Magn Reson Imaging 36(1):55–72
Dubois B, Feldman HH, Jacova C, Hampel H, Molinuevo JL, Blennow K, DeKosky ST, Gauthier S, Selkoe D, Bateman R (2014) Advancing research diagnostic criteria for Alzheimer's disease: the IWG-2 criteria. Lancet Neurol 13(6):614–629
Esquerda-Canals G, Montoliu-Gaya L, Güell-Bosch J, Villegas S (2017) Mouse Models of Alzheimer’s Disease. J Alzheimers Dis 57:1171–1183
Femminella GD, Thayanandan T, Calsolaro V, Komici K, Rengo G, Corbi G, Ferrara N (2018) Imaging and molecular mechanisms of Alzheimer’s disease: a review. Int J Mol Sci 19(12):3702
Folstein MF, Folstein SE, McHugh PR (1975) Mini-mental state: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12(3):189–198
Freeborough PA, Fox NC (1998) Modeling brain deformations in Alzheimer disease by fluid registration of serial 3D MR images. J Comput Assist Tomogr 22(5):838–843
Granger SJ, Colon-Perez L, Larson MS, Bennett IJ, Phelan M, Keator DB, Janecek JT, Sathishkumar MT, Smith AP, McMillan L, Greenia D, Corrada MM, Kawas CH, Yassa MA (2023) Reduced structural connectivity of the medial temporal lobe including the perforant path is associated with aging and verbal memory impairment. Neurobiol Aging 121:119–128
Gunter J, Bernstein M, Borowski B, Felmlee J, Blezek D, Mallozzi R, Levy J, Schuff N, Jack C (2005) Validation Testing of the MRI Calibration Phantom for the Alzheimer's Disease Neuroimaging Initiative Study. Proceedings on International Society for Magnetic Resonance in Medicine 14
Gunter J, Bernstein M, Borowski B, Felmlee J, Blezek D, Mallozzi R, Levy J, Schuff N, Jack C (2006) Validation testing of the MRI calibration phantom for the Alzheimer’s Disease Neuroimaging Initiative Study. Proceedings on International Society for Magnetic Resonance in Medicine 14: 1088–1093
Gutman BA, Wang Y, Yanovsky I, Hua X, Toga AW, Jack CR Jr., Weiner MW, Thompson PM (2015) and I. Alzheimer's Disease Neuroimaging Empowering imaging biomarkers of Alzheimer's disease. Neurobiol Aging 36 Suppl 1: S69-80
Holahan MR, Honegger KS, Tabatadze N, Routtenberg A (2007) GAP-43 gene expression regulates information storage. Learn Mem 14(6):407–415
Hua X, Gutman B, Boyle CP, Rajagopalan P, Leow AD, Yanovsky I, Kumar AR, Toga AW, Jack CR Jr, Schuff N (2011) Accurate measurement of brain changes in longitudinal MRI scans using tensor-based morphometry. NeuroImage 57(1):5–14
Hua X, Hibar DP, Ching CR, Boyle CP, Rajagopalan P, Gutman BA, Leow AD, Toga AW, Jack CR Jr, Harvey D (2013) Unbiased tensor-based morphometry: improved robustness and sample size estimates for Alzheimer's disease clinical trials. NeuroImage 66:648–661
Hua X, Lee S, Yanovsky I, Leow AD, Chou Y-Y, Ho AJ, Gutman B, Toga AW, Jack CR Jr, Bernstein MA (2009) Optimizing power to track brain degeneration in Alzheimer's disease and mild cognitive impairment with tensor-based morphometry: an ADNI study of 515 subjects. NeuroImage 48(4):668–681
Hua X, Leow AD, Parikshak N, Lee S, Chiang MC, Toga AW, Jack CR Jr., Weiner MW, Thompson PM (2008) Tensor-based morphometry as a neuroimaging biomarker for Alzheimer's disease: an MRI study of 676 AD, MCI, and normal subjects. Neuroimage 43(3): 458–469
Hughes CP, Berg L, Danziger W, Coben LA, Martin RL (1982) A new clinical scale for the staging of dementia. Br J psychiatry 140(6):566–572
Jack CR Jr., Bernstein MA, Fox NC, Thompson P, Alexander G, Harvey D, Borowski B, Britson PJ, J LW, Ward C, Dale AM, Felmlee JP, Gunter JL, Hill DL, Killiany R, Schuff N, Fox-Bosetti S, Lin C, Studholme C, DeCarli CS, Krueger G, Ward HA, Metzger GJ, Scott KT, Mallozzi R, Blezek D, Levy J, Debbins JP, Fleisher AS, Albert M, Green R, Bartzokis G, Glover G, Mugler J, Weiner MW (2008) The Alzheimer's Disease Neuroimaging Initiative (ADNI): MRI methods. J Magn Reson Imaging 27(4):685–691
Jovicich J, Czanner S, Greve D, Haley E, van der Kouwe A, Gollub R, Kennedy D, Schmitt F, Brown G, Macfall J, Fischl B, Dale A (2006) Reliability in multi-site structural MRI studies: effects of gradient non-linearity correction on phantom and human data. NeuroImage 30(2):436–443
Kiktenko AI, Uranova NA, Orlovskaia DD (1995) [Mossy fibers of the hippocampus in Alzheimer's disease]. Zh Nevrol Psikhiatr Im S S Korsakova 95(1):43–46
Kumar D, Sharma A, Sharma L (2020) A comprehensive review of Alzheimer’s association with related proteins: Pathological role and therapeutic significance. Curr Neuropharmacol 18(8):674–695
Lan G, Li A, Liu Z, Ma S, Guo T (2022) Presynaptic membrane protein dysfunction occurs prior to neurodegeneration and predicts faster cognitive decline. Alzheimers Dement
Le Bihan D, Mangin JF, Poupon C, Clark CA, Pappata S, Molko N, Chabriat H (2001) Diffusion tensor imaging: concepts and applications. J Magn Reson Imaging: Official J Int Soc Magn Reson Med 13(4):534–546
Lech RK, Suchan B (2013) The medial temporal lobe: memory and beyond. Behav Brain Res 254:45–49
Leow A, Thompson P, Hayashi K, Bearden C, Nicoletti M, Monkul S, Brambilla P, Sassi R, Mallinger A, Soares J (2005) Lithium effects on human brain structure mapped using longitudinal MRI. Society for Neuroscience, Washington, DC
Leow AD, Yanovsky I, Parikshak N, Hua X, Lee S, Toga AW, Jack CR Jr., Bernstein MA, Britson PJ, Gunter JL, Ward CP, Borowski B, Shaw LM, Trojanowski JQ, Fleisher AS, Harvey D, Kornak J, Schuff N, Alexander GE, Weiner MW, Thompson PM (2009) Alzheimer's disease neuroimaging initiative: a one-year follow up study using tensor-based morphometry correlating degenerative rates. biomarkers and cognition Neuroimage 45(3):645–655
Lu Y, I. for the Alzheimer’s Disease Neuroimaging (2022) Cerebrospinal fluid growth-associated protein 43 levels in patients with progressive and stable mild cognitive impairment. Aging Clin Exp Res 34(10):2399–2406
Mantzavinos V, Alexiou A (2017) Biomarkers for Alzheimer's Disease Diagnosis. Curr Alzheimer Res 14(11):1149–1154
Mattson MP (2004) Pathways towards and away from Alzheimer's disease. Nature 430(7000):631–639
Mazziotta J, Toga A, Evans A, Fox P, Lancaster J, Zilles K, Woods R, Paus T, Simpson G, Pike B (2001) A probabilistic atlas and reference system for the human brain: International Consortium for Brain Mapping (ICBM). Philosophical Transactions of the Royal Society of London. Ser B: Biol Sci 356(1412):1293–1322
Mechelli A, Price CJ, Friston KJ, Ashburner J (2005) Voxel-based morphometry of the human brain: methods and applications. Curr Med Imaging 1(2):105–113
Milà-Alomà M, Brinkmalm A, Ashton NJ, Kvartsberg H, Shekari M, Operto G, Salvadó G, Falcon C, Gispert JD, Vilor-Tejedor N, Arenaza-Urquijo EM, Grau-Rivera O, Sala-Vila A, Sanchez-Benavides G, González-de-Echávarri JM, Minguillon C, Fauria K, Niñerola-Baizán A, Perissinotti A, Kollmorgen G, Suridjan I, Zetterberg H, Molinuevo JL, Blennow K, Suárez-Calvet M (2021) CSF Synaptic Biomarkers in the Preclinical Stage of Alzheimer Disease and Their Association With MRI and PET: A Cross-sectional Study. Neurology 97(21):e2065–e2078
Mueller SG, Weiner MW, Thal LJ, Petersen RC, Jack C, Jagust W, Trojanowski JQ, Toga AW, Beckett L (2005) The Alzheimer's disease neuroimaging initiative. Neuroimaging Clin 15(4):869–877
Mueller SG, Weiner MW, Thal LJ, Petersen RC, Jack CR, Jagust W, Trojanowski JQ, Toga AW, Beckett L (2005) Ways toward an early diagnosis in Alzheimer’s disease: the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Alzheimer's Dement 1(1):55–66
Parhad IM, Oishi R, Clark AW (1992) GAP-43 gene expression is increased in anterior horn cells of amyotrophic lateral sclerosis. Ann Neurol 31(6):593–597
Poulin SP, Dautoff R, Morris JC, Barrett LF, Dickerson BC (2011) Amygdala atrophy is prominent in early Alzheimer's disease and relates to symptom severity. Psychiatry Res 194(1):7–13
Powel J Wechsler memory scale-revised: David A., Wechsler (1988) New York: The Psychological Corporation. Harcourt Brace Jovanovich, Inc, 1987. 150 pp. Archives of Clinical Neuropsychology 3(4): 397–403
PP R (1964) Human experimentation. Code of ethics of the world medical association. Declaration of Helsinki British medical journal 2(5402):177–177
Qiang Q, Skudder-Hill L, Toyota T, Wei W, Adachi H (2022) CSF GAP-43 as a biomarker of synaptic dysfunction is associated with tau pathology in Alzheimer’s disease. Sci Rep 12(1):17392
Ramakers G, McNamara RK, Lenox RH, De Graan P (1999) Differential changes in the phosphorylation of the protein kinase C substrates myristoylated alanine-rich C kinase substrate and growth-associated protein-43/B-50 following Schaffer collateral long-term potentiation and long-term depression. J Neurochem 73(5):2175–2183
Rao YL, Ganaraja B, Murlimanju BV, Joy T, Krishnamurthy A, Agrawal A (2022) Hippocampus and its involvement in Alzheimer's disease: a review. 3 Biotech 12(2): 55
Saal KA, Galter D, Roeber S, Bähr M, Tönges L, Lingor P (2017) Altered Expression of Growth Associated Protein-43 and Rho Kinase in Human Patients with Parkinson's Disease. Brain Pathol 27(1):13–25
Sandelius Å, Portelius E, Källén Å, Zetterberg H, Rot U, Olsson B, Toledo JB, Shaw LM, Lee VM, Irwin DJ (2019) Elevated CSF GAP-43 is Alzheimer's disease specific and associated with tau and amyloid pathology. Alzheimer's Dement 15(1):55–64
Shattuck DW, Leahy RM (2002) BrainSuite: an automated cortical surface identification tool. Med Image Anal 6(2):129–142
Shi J, Stonnington CM, Thompson PM, Chen K, Gutman B, Reschke C, Baxter LC, Reiman EM, Caselli RJ, Wang Y (2015) Studying ventricular abnormalities in mild cognitive impairment with hyperbolic Ricci flow and tensor-based morphometry. NeuroImage. 104:1–20
Simrén J, Leuzy A, Karikari TK, Hye A, Benedet AL, Lantero-Rodriguez J, Mattsson-Carlgren N, Schöll M, Mecocci P, Vellas B, Tsolaki M, Kloszewska I, Soininen H, Lovestone S, Aarsland D, Hansson O, Rosa-Neto P, Westman E, Blennow K, Zetterberg H, Ashton NJ (2021) The diagnostic and prognostic capabilities of plasma biomarkers in Alzheimer's disease. Alzheimers Dement 17(7):1145–1156
Sled JG, Zijdenbos AP, Evans AC (1998) A nonparametric method for automatic correction of intensity nonuniformity in MRI data. IEEE Trans Med Imaging 17(1):87–97
Soria Lopez JA, Gonzalez HM, Leger GC (2019) Alzheimer's disease. Handb Clin Neurol 167:231–255
Teipel S, Drzezga A, Grothe MJ, Barthel H, Chételat G, Schuff N, Skudlarski P, Cavedo E, Frisoni GB, Hoffmann W (2015) Multimodal imaging in Alzheimer's disease: validity and usefulness for early detection. Lancet Neurol 14(10):1037–1053
Toga A (1999) Brain Warping. Academic Press, San Diego
Vemuri P, Senjem ML, Gunter JL, Lundt ES, Tosakulwong N, Weigand SD, Borowski BJ, Bernstein MA, Zuk SM, Lowe VJ, Knopman DS, Petersen RC, Fox NC, Thompson PM, Weiner MW, Jack CR (2015) Accelerated vs. unaccelerated serial MRI based TBM-SyN measurements for clinical trials in Alzheimer's disease. NeuroImage. 113:61–69
Wachinger C, Salat DH, Weiner M, Reuter M, Initiative A (2016) Whole-brain analysis reveals increased neuroanatomical asymmetries in dementia for hippocampus and amygdala. Brain 139(12): 3253–3266
Weiler M, Agosta F, Canu E, Copetti M, Magnani G, Marcone A, Pagani E, Balthazar MLF, Comi G, Falini A (2015) Following the spreading of brain structural changes in alzheimer’s disease: a longitudinal, multimodal MRI study. J Alzheimers Dis 47(4):995–1007
Wenk G (2003) Neuropathologic Changes in Alzheimer’s Disease. J Clin psychiatry 64 Suppl 9:7–10
West MJ, Coleman PD, Flood DG, Troncoso JC (1994) Differences in the pattern of hippocampal neuronal loss in normal ageing and Alzheimer's disease. Lancet 344(8925):769–772
Zhang H, Lyu D, Jia J, Initiative A (2021) The Trajectory of Cerebrospinal Fluid Growth-Associated Protein 43 in the Alzheimer’s Disease Continuum: A Longitudinal Study. Journal of Alzheimer's Disease(Preprint): 1–12

No competing interests reported.

Growth-associated protein 43 and tensor-based morphometry indices in mild cognitive impairment

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

1. Introduction

2. Materials and Methods

2.1. Participants and Design

2.2. Cognitive Assessments

2.3. MRI Acquisition

2.4. MRI Preprocessing

2.5. GAP-43 Measurement

2.6. Statistical Analysis

3. Results

4. Discussion

5. Conclusion

Declarations

References

Additional Declarations

Status:

Version 1