In 2018, Alzheimer’s Disease International estimated that approximately 50 million people live with dementia worldwide, a statistic predicted to triple by 2050 [1]. This was validated by the 2019 Global Burden of Disease study, which estimated that reported dementia cases could rise to 130–175 million in 2050 [2]. In the United States, the 2023 costs of AD or other dementias were estimated at USD 345 billion, with approximately 25% being patient out-of-pocket expenses [3]. In 2019, global societal costs of all-cause dementia were estimated at USD 1.3 trillion, working out to approximately USD 23,000 per patient [4]. Thus, combating global dementia prevalence is an urgent public health and economic priority.

Alzheimer’s disease (AD) is the most common neurodegenerative disease in humans, and the most common form of dementia. It displays an extensive distribution of two core pathologies: extracellular plaques and intraneuronal neurofibrillary tangles (NFTs), consisting of amyloid beta (Aβ) and the microtubule-associated protein tau, respectively. Although histological lesions were first reported over a century ago by Dr. Alois Alzheimer and Dr. Gaetano Perusini, with the support of Dr. Emil Kraepelin ^[5][6][5,6], it was not until 70 years later that these pathologies gathered widespread attention ^[7][8][9][7,8,9]. Several additional years followed before extensive clinical neuropathological studies supported both plaques and NFTs with probable AD diagnosis ^[10][11][10,11]. In the three decades since, hundreds of promising therapeutics targeted towards Aβ, tau and other pathologies have failed in clinical trials ^[12][13][12,13], resulting in the stagnation of curative research.

More recently, Biogen’s ‘EMERGE’ and ‘ENGAGE’ clinical trials of a Aβ plaque-directed monoclonal antibody, aducanumab [14], resulted in an accelerated approval in 2021 by the Food and Drug Administration (FDA). Though while plaques were reduced in patients, this may have been used as a surrogate end point, with aducanumab’s clinical efficacy in terms of cognitive decline protection being debated within the field ^[15][16][17][15,16,17]. In comparison, Eisai’s lecanemab—an Aβ-monoclonal antibody that can target protofibrils—appears to hold greater promise for patients in terms of both reduced plaques and protection against cognitive decline ^[18][19][18,19]. While also approved by the FDA via their accelerated approval pathway, similar side effects are present for both antibodies, namely amyloid-related imaging abnormalities and edema ^[14][19][14,19].

2. iPSC Technology and iPSC-Derived Microglia

As an alternative to using immortalized lines, primary cells from animal models and human microglia from brain excision surgery, recent advances in technology have facilitated a renewed interested in the generation of microglia from human-induced pluripotent stem cells (hiPSCs). Getting around the issue of tissue availability and variation in microglial phenotypes, hiPSCs can be used for both in vitro and xenographic in vivo work to understand cellular physiology in both health and disease. For example, microglial chimeric mouse models have been successfully generated, wherein hiPSC-derived macrophage progenitors are transplanted into neonatal mouse brains. Single-cell RNA sequencing reported that these xenografted cells largely retain human microglial identity and become widely dispersed by 6 months of age, including in both the hippocampus and cerebral cortex, as reported by TMEM119 staining ^[20][82]. It has been suggested that TMEM119 is both a reliable and specific marker, able to discriminate resident microglia from peripherally derived macrophages ^[21][22][83,84]. However, recent work suggests that TMEM119 is not exclusive, nor does it stain all microglia. The latter appears especially true under cellular stress conditions, including murine models of both focal stroke and Parkinson’s disease ^[23][85]. As such, supplementary staining methods will be important for future work using this model. Despite this issue though, in tandem with murine microglial depletion methods such as the pharmacological inhibition of CSF1R ^[24][25][86,87], the function of humanized microglia can be validated and compared against traditional non-chimeric animals. Pharmacological and genetic studies which report benefits in both systems will likely strengthen the translational potential of any promising future therapeutic target. Although the understanding of microglial function continues to improve, cellular dysfunction in AD remains to be fully elucidated, despite the identification of dozens of genetic risk loci (Table 1). The use of hiPSCs to decipher the pathological contributions of these risk variants could thus prove invaluable. Following the seminal publication by Muffet and colleagues in 2016 ^[26][88], many protocols to produce hiPSC-derived microglia have become available ^{[27][28][29][30][31][32]}[89,90,91,92,93,94], with others having reviewed some of these protocols previously ^[33][95]. The intricate work by Brownjohn and colleagues showcases how hiPSC-derived microglial research can elucidate both the physiological function and pathological disruption of AD risk genes ^[31][93]. In this study, fibroblasts were acquired from a patient either homozygous for the TREM2 p.T66M mutation or the TREM2 p.w50C mutation, both of which are missense mutations known to cause frontotemporal dementia-like syndrome and Nasu-Hakola disease, respectively ^[41][97]. These cells were then successfully differentiated into hiPSC-microglia, wherein aberrant processing was reported for all TREM2 mutant backgrounds and abnormalities in TREM2 trafficking were reported for homozygous mutant cells. Despite this, hiPSC-microglia displaying these missense mutations responded effectively to inflammatory challenge and remain phagocytically competent. Although the study had limitations due to the low sample number, expansion of this approach could hold importance to understanding the role of not only TREM2 risk variants in AD, but dozens of other risk loci identified in human patients (Figure 1). These approaches outweigh the simplified use of both murine microglia and immortalized cell lines. However, it remains unlikely that two-dimensional monoculturing techniques will sufficiently recapture the effects of risk loci on the diverse array of functions displayed by endogenous human microglia. Instead, the use of gene editing technology to incorporate risk loci-containing microglia into 3D organoids may be an approach with greater translational potential. Several groups have successfully developed and validated three-dimensional organoid systems from hiPSCs ^[42][43][44][98,99,100]. Although, microglia are often absent in organoids developed from currently available protocols. Despite this, several groups have reported successful integration of microglia into human brain organoids via co-culturing techniques ^[29][45][91,101]. In addition, to get around the variability that has often plagued organoid research ^[46][47][102,103], newer protocols can reliably generate a diverse array of cell types typical for the human cerebral cortex with close correlation to endogenous tissue ^[42][48][98,104], strengthening the potential use of 3D cerebral organoids for neurodegenerative disease research. However, whether this reliability can be recaptured in organoids expressing hiPSC-derived microglia with known AD risk loci remains to be determined.