Metformin and Dementia Risk: A Systematic Review with Respect to Time Related Biases

Abstract

Background: When studying drug effects using observational data, time-related biases may exist and result in spurious associations. Numerous observational studies have investigated metformin and dementia risk, but have reported inconsistent findings, some of which might be caused by unaddressed time-related biases. Immortal time bias biases the results toward a “protective” effect, whereas time-lag and time-window biases can lead to either a “detrimental” or “protective” effect. Objective: To conduct a systematic review examining time-related biases in the literature on metformin and dementia. Methods: The electronic databases PubMed, Web of Science, and ProQuest were searched for the terms “Metformin” AND (“dementia” OR “Alzheimer’s Disease” OR “cognitive impairment"). These databases were searched from inception through 09/24/2021. Only English language articles and human research were eligible. Results: Seventeen studies were identified: thirteen cohort studies, two case-control studies, and two nested case-control studies. Eleven (64.7%) studies reported a reduced risk of dementia associated with metformin use; two (11.8%) suggested metformin increased dementia risk, while four (23.5%) concluded no significant associations. Eight (61.5%) of thirteen cohort studies had immortal time bias or did not clearly address it. Fifteen (88.2%) of seventeen reviewed studies had time-lag bias or did not clearly address it. Two (50.0%) of four case-control studies did not explicitly address time-window bias. The studies that addressed most biases concluded no associations between metformin and dementia risk. Conclusion: None of the reviewed studies clearly addressed relevant time-related biases, illustrating time-related biases are common in observational studies investigating the impact of anti-diabetic medications on dementia risk.

Diabetes, antidiabetic medications and risk of dementia: A systematic umbrella review and meta‐analysis

“…5‐6). Of the 100 included reviews, 32 related diabetes to cognitive function in mixed diabetic populations, 2,6,8,26–54 seven examined the effect of type 1 diabetes, 3,55–60 19 focused on type 2 diabetes, 5,7,61–77 14 observed brain structure in diabetes, 1,4,78–89 two examined genetic and other biomarkers of dementia 90,91 and 26 focused on cognitive effects of antidiabetic treatment 14,15,20–22,92–112 . Of the 27 studies included in the meta‐analyses, 15 related metformin use to risk of dementia, 113–127 10 were used in the thiazolidinedione meta‐analysis, 113,122–130 four for pioglitazone, 126,129–131 seven for DPP‐4is, 123,125–127,132–134 five for α‐glucosidase inhibitors, 123,126,127,129,135 two for meglitinides, 126,129 seven for insulin, 122,123,125,127,129,136,137 11 for sulphonylureas,…”

Metformin Cessation and Dementia Incidence

“…Second, all individuals were metformin initiators, mitigating the potential for cohort effects and confounding by indication. Additionally, beginning follow-up at the age of early termination of metformin use prevents immortal person-time bias …”

“…Additionally, increasing metformin use and decreasing age-specific dementia incidence in the US complicate observational comparisons between metformin and other agents . Furthermore, some prior studies compared prevalent users with nonusers, which can lead to immortal person-time and confounding biases …”

“…10 Furthermore, some prior studies compared prevalent users with nonusers, which can lead to immortal person-time and confounding biases. 17 Individuals may terminate metformin for several reasons. 18,19 Gastrointestinal adverse effects are more common with metformin than with other antidiabetes agents, 20 leading to lower adherence 18 and replacement with other antidiabetes agents in one-fifth of early users.…”

Linking the Amyloid, Tau, and Mitochondrial Hypotheses of Alzheimer’s Disease and Identifying Promising Drug Targets

“…The largest group of drugs tested in phase 3 of AD drug development consists of disease-modifying small molecules [ 105 ], which include substances modulating (i) metabolism and bioenergetics, e.g., metformin [ 354 , 355 ], semaglutide [ 356 ], and tricaprylin [ 357 , 358 ]; (ii) oxidative stress and blood flow in the brain, e.g., omega-3 polyunsaturated fatty acids [ 359 ] and ethyl eicosapentaenoate (icosapent ethyl) [ 360 ]; (iii) synaptic plasticity and neuroprotection, e.g., blarcamesine [ 361 , 362 ], atuzaginstat [ 363 ], AGB101 (levetiracetam) [ 364 , 365 ], and simufilam [ 366 ]; (iv) Aβ pathology, e.g., ALZ-801 (valiltramiprosate), an oral prodrug of homotaurine that blocks the formation of toxic Aβ oligomers [ 334 , 367 , 368 ]; (v) neuroinflammation, e.g., NE3107 [ 369 ] and curcumin [ 337 , 370 , 371 , 372 , 373 ]; and more. Disease-modifying drugs are also the monoclonal antibodies directed at Aβ; besides aducanumab [ 374 ] it is also gantenerumab [ 374 , 375 ], lecanemab [ 375 ], donanemab [ 376 ], and solanezumab [ 374 , 377 ], which have shown some efficiency and are tested in phase 3 clinical trials [ 317 ].…”