Adenomyosis is a prevalent and non-cancerous uterine disease which can significantly impaired the fertility of reproductive-age women. However, the etiology, as well as the cellular and molecular mechanism underlying adenomyosis remain largely unknown. Here, we utilized cutting-edge spatial and single-cell RNA sequencing technologies to create a comprehensive transcriptional atlas of adenomyosis pathology. Our spatial profiling clearly distinguished gland, mesenchyma and myometrium regions, recapitulating spatial transcriptome structural characteristics of uterus. Moreover, we analyzed the expression profiles of 69,115 single cells and integrated them with spatial data. The analysis of immune cells showed a distinct immune inflammatory microenvironment in the eutopic and ectopic endometrial glands of adenomyosis. Notably, we discovered an increased number of DNAH9+ ciliated cells in ectopic endometrial glands, indicating their potential role in the formation of ectopic endometrium. These findings provide cellular evidence to support the invagination theory and offer a new vision on the pathophysiology and clinical intervention of adenomyosis. Background and Aims: Adenomyosis is characterized by the presence of endometrial glands and stroma-like tissue within the myometrium 1 , resulting in abnormal uterine bleeding, dysmenorrhea, chronic pelvic pain and infertility 2 . Around 35% of reproductive-age women suffer from adenomyosis worldwide 3 , 4 . Highly variable symptoms make diagnosis challenging and commonly result in misdiagnosis. Furthermore, due to the advance of diagnostic imaging techniques such as transvaginal ultrasound scan (TVUS) and magnetic resonance imaging (MRI), there is an increasing trend of incidence rate, especially the remarkably increased proportion of young women diagnosed 4 . Although several hypotheses have been proposed to explain the pathogenesis of adenomyosis, including invagination of the endometrium into the myometrium, trauma to the endometrial-myometrial interface, and the stem cell potential of de novo ectopic tissue5, none of the these theories can fully account for all the phenotypes 6 . This ambiguous understanding of underlying pathophysiologic mechanisms hinders the development of effective early intervention strategies, emphasizing the need for further research into the biological mechanisms involved. Common symptoms of adenomyosis, such as dysmenorrhea and irregular uterine bleeding, suggest dysfunction of the endometrium, despite the absence of significant morphological change between in situ endometrium (eutopic) and the ectopic endometrial gland found within adenomyoma. Recently studies utilizing single-cell RNA sequencing (scRNA-seq) have reported functional differences within limited partial regions, offering some insight into cellular heterogeneity 7 , 8 . It is important to note that these studies have been limited in scope, either focusing on a small number of patients or lacking investigation into the spatial distribution of specific disease regions. As a result, our understanding of adenomyosis at a high-resolution level remains incomplete. Advancements in scRNA-seq and spatial transcriptomics techniques provide powerful tools to investigate rare subpopulations, cellular interactions and tissue architecture 9 , 10 , 11 , 12 . These cutting-edge technologies have the potential to enhance our understanding of adenomyosis. Method: Herein, we simultaneously combined scRNA-seq and spatial transcriptome (Geo-seq) 13 data to gain a global understanding of cell types and niches in the entire layer of adenomyosis uterus tissue. By allocating cell types to different adenomyosis niches, determining molecular mediators of intercellular interactions, and pinpointing the cellular and spatial sources of niche factors, our data shed new light on the invagination of eutopic endometrium and tissue architecture of adenomyosis. This multi-dimensional molecular profiling provides a foundation for further exploration of adenomyosis pathophysiology, enabling a better understanding of the cellular and molecular mechanisms involved. Ultimately, this study has the potential to facilitate early clinical diagnosis and medication intervention for adenomyosis. Results: Our spatial profiling clearly distinguished gland, mesenchyma and myometrium regions, recapitulating spatial transcriptome structural characteristics of uterus. Moreover, we analyzed the expression profiles of 69,115 single cells and integrated them with spatial data. The analysis of immune cells showed a distinct immune inflammatory microenvironment in the eutopic and ectopic endometrial glands of adenomyosis. Notably, we discovered an increased number of DNAH9+ ciliated cells in ectopic endometrial glands, indicating their potential role in the formation of ectopic endometrium. Conclusion: These findings provide cellular evidence to support the invagination theory and offer a new vision on the pathophysiology and clinical intervention of adenomyosis.