A Composite Vascular Burden Score (VCING-Lite Proxy) Modifies APOE and Alzheimer’s Neuropathology Associations

Alice B. S., Nono Djotsa^1,2, Rebecca Bernal³, Obed Apochi⁴, Yannick N. Wadop^3,5, Angelique Gonzalez³, Shahroo Etemadmoghadam³, Crystal Wiedner³, Biniyam A. Ayele^6,7,8, Murali Sargurupremraj³, Xueqiu Jian³, Alfred K. Njamnshi^9,10, Cyrille Nkouonlack^11,12,Jose E. Cavazos³, Sudha Seshadri^3,13,14,15, Alfred Fonteh¹⁶, Bernard Fongang^3,5,17, Jayandra Jung Himali^{3,14,15,22,23*}, Agustin Ruiz^{3,18,19,20,21*},  and Margaret E. Flanagan^3,24* on behalf of the African Initiative on Bioinformatics Online Training in Neurodegenerative Diseases (AI-BOND)

 

* Joint senior authors

 

Authors affiliations:

¹VA HSR&D Center for Innovations in Quality, Effectiveness and Safety, Michael E. DeBakey VA Medical Center, Houston, TX 77030, USA.

²Section of Health Services Research, Department of Medicine, Baylor College of Medicine, Houston, TX 77030, USA.

³Glenn Biggs Institute for Alzheimer's & Neurodegenerative Diseases, UT Health San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78299, USA.

⁴International Graduate School of Neuroscience IGSN, Ruhr University, Bochum, Germany.

⁵Department of Biochemistry and Structural Biology, The University of Texas Health Science Center at San Antonio, San Antonio, Texas, United States of America.

⁶John P. Hussman Institute for Human Genomics, University of Miami Miller School of Medicine, 1501 NW 10th Avenue, Miami, FL 33136 USA.

⁷Department of Neurology, College of Health Science, Addis Ababa University, Yared St. Addis Ababa, Ethiopia PoBox 1171.

⁸Global Brain Health Institute (GBHI), University of California San Francisco (UCSF), 675 Nelson Rising Lane, San Francisco, CA 94143.

⁹Brain Research Africa Initiative (BRAIN), Geneva, Switzerland & Yaoundé, Cameroon.

¹⁰Neuroscience Lab, Faculty of Medicine & Biomedical Sciences, The University of Yaoundé I, Yaoundé, Cameroon.

¹¹Faculty of Health Sciences, University of Buea, Buea, Cameroon

¹²Brain Research Africa Initiative, Yaoundé, Cameroon

¹³Department of Neurology, UT Health San Antonio, 7703 Floyd Curl Drive, San Antonio, TX, USA.

¹⁴Department of Neurology, Boston University Chobanian & Avedisian School of Medicine, 72 East Concord Street, Boston, MA 02118, USA.

¹⁵The Framingham Heart Study, 73 Mt. Wayte Avenue, Framingham, MA 01702, USA.

¹⁶Neuroscience Department, Huntington Medical Research Institutes, Pasadena, USA.

¹⁷Department of Population Health Sciences, UT Health San Antonio, 7703 Floyd Curl Drive, San Antonio, TX, USA.

¹⁸Joe R. & Teresa Lozano Long School of Medicine, UT Health San Antonio, 7703 Floyd Curl Drive, San Antonio, TX, USA.

¹⁹Department of Microbiology Immunology and Molecular Genetics. UT Health San Antonio, 7703 Floyd Curl Drive, San Antonio, TX, USA.

²⁰Ace Alzheimer Center Barcelona-Universitat Internacional de Catalunya, Barcelona, Spain

²¹Networking Research Center on Neurodegenerative Diseases (CIBERNED), Instituto de Salud Carlos III, Madrid, Spain

²²Department of Biostatistics, Boston University School of Public Health, 715 Albany Street, Boston, MA 02118, USA.

²³Graduate School of Biomedical Sciences, UT Health San Antonio, 7703 Floyd Curl Drive, San Antonio, TX, USA.

²⁴Department of Pathology, UT Health San Antonio, San Antonio, Texas

 

Abstract

Background: Evidence indicates that vascular brain pathologies (VBP) may modify associations between APOE genotype and Alzheimer’s disease (AD) neuropathologic outcomes including AD neuropathologic change (ADNC), Thal phase, neuritic plaques, diffuse plaques, and Braak tangle stage. However, whether cumulative VBP burden, recently quantified using Gonzalez’s VCING-Lite Proxy, modifies APOE–AD neuropathology associations remain understudied. We therefore examined effect modification by VCING-Lite Proxy on associations between APOE genotypes and AD neuropathologic outcomes in the National Alzheimer’s Coordinating Center (NACC) dataset.

Methods: Participants were grouped as ε4-noncarriers (ε33/ε22/ε23), ε4-heterozygotes (ε24/ε34), or ε4-homozygotes (ε44). The VCING-Lite Proxy combines global ratings across three cerebrovascular domains—arteriolosclerosis, cerebral amyloid angiopathy (CAA), and gross infarcts—to classify the likelihood of vascular cognitive impairment as low, moderate, or high. We used logistic regression adjusted for age, age-squared, sex, self-reported ancestry, and education. Significance was set at FDR α=0.05, with interaction terms evaluated at p<0.10.

Results: The sample consisted of 7153 participants from the NACC database with autopsy data available (mean age 79±12 years, 48% female). We observed that VCING-Lite Proxy significantly modified the associations between APOE genotype and neuritic plaques (p<.0001); a similar but weaker pattern observed for Braak tangle stage (p = 0.076). Across VCING-Lite proxy strata, ε4-heterozygotes and ε4-homozygotes had higher odds of neuritic plaques compared with ε4 noncarriers; however, the magnitude of this association was attenuated at higher cumulative vascular pathology burden (Table 2). Specifically, odds ratios were highest in the low VCING-Lite proxy group (ε24/ε34 (OR[95% CI],p: 3.03[2.61,3.52],<.0001); ε44 (10.50[6.69,16.50],<.0001) and were lower in the high burden group (ε24/ε34 (2.17[1.72,2.74],<.0001); ε44 (4.86[3.30,7.16],<.0001)), consistent with effect modification by VCING-Lite proxy. No significant APOE × VCING-Lite proxy interaction was observed for ADNC, Thal phase, or diffuse plaques. Analyses of components of VCING-Lite proxy (Figure 1) showed that across AD neuropathologic outcomes, increasing CAA severity attenuated APOE ε4 associations and flattened the ε4 gene-dose gradient, shifting from a multiplicative pattern without CAA to near-nominal differences between ε24/ε34 and ε4/ε4 under severe CAA.   

Conclusions: Higher VCING-Lite proxy scores attenuated the association between APOE genotype and neuritic plaques, and this effect was mainly driven by CAA. Given shared etiologic pathways between APOE and CAA, these interactions warrant careful causal interpretation.

Keywords: Alzheimer’s disease, APOE genotype, neuropathology, vascular pathology lesions, VCING-Lite proxy

 

 

Table 1. Participant Characteristics by APOE ε4 genotype 

	No. (%)
	Overall (N=7153)	APOE ε4 Allele
Characteristic		ε33, ε22 ε23 (N=3998)	ε24, ε34 (N=2567)	ε44 (N=588)
Age, mean (SD), years	79 (12)	80 (12)	78 (11)	75 (9)
Systolic Blood Pressure, mean (SD)	131 (20)	131 (20)	131 (20)	130 (19)
Diastolic Blood Pressure, mean (SD)	73 (11)	73 (11)	73 (11)	74 (10)
Female	3,405 (48)	1,924 (48)	1,218 (47)	263 (45)
Education, years
Up to High School	1,629 (23)	913 (23)	592 (23)	124 (21)
Some College	1,298 (18)	738 (19)	442 (17)	118 (20)
Bachelor’s Degree	2,034 (28)	1,113 (28)	756 (30)	165 (28)
Post master’s degree	2,192 (31)	1,234 (31)	777 (30)	181 (31)
Self-reported Ancestry
White	6,758 (95)	3,800 (95)	2,414 (94)	544 (93)
Black or African American	303 (4)	134 (3)	130 (5)	39 (7)
Other	92 (1)	64 (2)	23 (1)	5 (1)
Hispanic or Latino	225 (3)	140 (4)	69 (3)	16 (3)
Smoking Status	3,207 (46)	1,835 (47)	1,127 (46)	245 (44)
Lipid Lowering Medication Use	2,425 (35)	1,296 (33)	911 (36)	218 (37)
Hypertension Treatment	193 (3)	109 (3)	70 (3)	14 (2)
Alzheimer’s pathology outcomes
Alzheimer's Disease Global Score
Alzheimer's Disease Neuropathologic Change (ADNC)
No or low ADNC	1,338 (31)	1,089 (46)	236 (15)	13 (4)
Intermediate or high ADNC	2,921 (69)	1,286 (54)	1,298 (85)	337 (96)
Parenchymal amyloid (Aβ)
Thal Phase
Phase 0	582 (14)	530 (22)	49 (3)	3 (1)
Phase 1 and Phase 2	618 (14)	458 (19)	149 (10)	11 (3)
Phase 3	510 (12)	334 (14)	158 (10)	18 (5)
Phase 4	2,602 (60)	1,094 (45)	1,189 (77)	319 (91)
Neuritic Plaques
None	1,695 (24)	1,427 (36)	258 (10)	10 (2)
Sparse	917 (13)	558 (14)	319 (12)	40 (7)
Moderate	1,345 (19)	765 (19)	496 (19)	84 (14)
Frequent	3,196 (45)	1,248 (31)	1,494 (58)	454 (77)
Diffuse Plaques
None	998 (15)	865 (24)	123 (5)	10 (2)
Sparse	854 (13)	592 (16)	233 (10)	29 (5)
Moderate	1,087 (16)	652 (18)	371 (16)	64 (12)
Frequent	3,675 (56)	1,569 (43)	1,667 (70)	439 (81)
Tau pathology
Braak Stage
Stage 0	383 (5)	319 (8)	64 (3)	0 (0)
Stage I and Stage II	1,333 (19)	1,042 (27)	272 (1)	19 (3)
Stage III and Stage IV	1,716 (24)	1,114 (29)	536 (21)	66 (11)
Stage V and Stage VI	3,606 (51)	1,431 (37)	1,675 (66)	500 (86)
Vascular brain pathologies
VCING-Lite Proxy
Low	2917 (41.03)	1840 (46.0)	964 (38.0)	113 (19.0)
moderate	2799 (39.37)	1569 (40.0)	988 (39.0)	242 (41.0)
high	1394 (19.61)	558 (14.0)	605 (24.0)	231 (39.0)
VCING-Lite Proxy components
Arteriolosclerosis
None	1,168 (18)	681 (19)	406 (17)	81 (15)
Mild	2,337 (36)	1,332 (37)	834 (36)	171 (33)
Moderate	2,071 (32)	1,132 (31)	760 (33)	179 (34)
Severe	904 (14)	485 (13)	326 (14)	93 (18)
Cerebral amyloid angiopathy (CAA)
None	2,730 (39)	2,069 (53)	611 (24)	50 (9)
Mild	2,103 (30)	1,042 (27)	944 (37)	117 (20)
Moderate	1,423 (20)	537 (14)	669 (26)	217 (37)
Severe	786 (11)	282 (7)	306 (12)	198 (34)
Gross infarcts	616 (14)	360 (15)	214 (14)	42 (12)

Abbreviations: APOE, apolipoprotein E