Background

Patient-derived organoids (PDOs) are regarded as reliable preclinical models for cancer research and personalized medicine [1]. They retain not only the structural complexity of the original tumor but also their biological behavior, including their responses to various therapeutic agents [2]. By faithfully recapitulating tumor-specific characteristics, PDOs offer a robust platform for testing novel therapeutics and developing individualized treatment strategies [3].

Importantly, PDOs display marked heterogeneity, and their morphology often reflects the underlying genetic and transcriptional features of the parental tumor [4–9]. For example, accumulation of mutations is frequently associated with dense or invasive organoids that can form grape-like structures, whereas organoids with lower mutational burden may maintain cystic or balloon-like morphologies resembling normal-like epithelial architecture. TP53 mutations, commonly present in head and neck, colorectal, and breast cancers, are associated with disorganized, compact structures lacking clear glandular architecture [10–12]. KRAS mutations, prevalent in pancreatic and colorectal cancers, often result in dense, multilayered organoids without lumen formation [13,14]. PIK3CA mutations lead to irregular, enlarged organoids in breast cancer [12], and CDH1 loss produces loosely cohesive or scattered organoids devoid of epithelial polarity in gastric cancer [4]. These genotype–phenotype relationships highlight the importance of preserving morphological integrity during sample preparation, as morphology serves as both a phenotypic marker and a functional readout of tumor biology.

Under laminin-rich basement membrane extract (BME), PDOs form distinct three-dimensional structures through enhanced self-renewal and self-organization, reinforcing their role in translational cancer research [15]. BME plays a critical role in maintaining polarity, lumen formation, and cystic morphology by providing a laminin- and collagen-rich extracellular scaffold [16]. However, during conventional FFPE processing, fixation often disrupts the BME, causing collapse of this structural support and leading to distortion or loss of normal-like organoid morphologies [17]. While dense organoids may partly preserve their structure through strong cell–cell adhesion, cystic or balloon-like morphologies are especially vulnerable once the BME is compromised [18].

This protocol was specifically designed to overcome the collapse of BME and distortion of fragile organoid morphologies that commonly occur during the dehydration steps of conventional FFPE processing. By functioning as a stabilizing matrix, the agarose mold immobilizes organoids within the BME, thereby preserving their native three-dimensional architecture. As a result, organoids maintain their original morphology, enabling accurate histological evaluation across a broad spectrum of organoid types, including cystic, dense, and grape-like structures.

The major strength of this approach lies in its ability to reliably preserve both structural integrity and histological detail, which are essential for downstream applications such as H&E staining, immunohistochemistry, and immunofluorescence. Although the method involves steps that may appear harsh—such as drying and the addition of warm agarose—and FFPE processing can impose certain limitations, molecular analyses that are highly sensitive to fixation stress may be restricted. Nevertheless, this protocol effectively supports morphological assessment and protein-level studies. In our validation, we confirmed the expression of cancer diagnostic markers such as CK19, pan-CK, and P53 and further demonstrated that marker expression relevant to subtype classification could be evaluated using tissue proteomics–based analysis [9,19–21]. Overall, this protocol provides a practical and robust method for generating high-quality FFPE organoid slides, making it well-suited for cancer research, biomarker evaluation, and translational applications in personalized medicine.