Assessing Tumorigenicity in Stem Cell-Derived Therapeutic Products: A Critical Step in Safeguarding Regenerative Medicine

Abstract

Stem cells hold promise in regenerative medicine due to their ability to proliferate and differentiate into various cell types. However, their self-renewal and multipotency also raise concerns about their tumorigenicity during and post-therapy. Indeed, multiple studies have reported the presence of stem cell-derived tumors in animal models and clinical administrations. Therefore, the assessment of tumorigenicity is crucial in evaluating the safety of stem cell-derived therapeutic products. Ideally, the assessment needs to be performed rapidly, sensitively, cost-effectively, and scalable. This article reviews various approaches for assessing tumorigenicity, including animal models, soft agar culture, PCR, flow cytometry, and microfluidics. Each method has its advantages and limitations. The selection of the assay depends on the specific needs of the study and the stage of development of the stem cell-derived therapeutic product. Combining multiple assays may provide a more comprehensive evaluation of tumorigenicity. Future developments should focus on the optimization and standardization of microfluidics-based methods, as well as the integration of multiple assays into a single platform for efficient and comprehensive evaluation of tumorigenicity.

Keywords: cell therapy; regenerative medicine; stem cells; tumorigenicity.

Figure 1

Pluripotent stem cell-derived functional cells for tissue regeneration and associated tumorigenesis risk due to rare undifferentiated stem cells. iPSC: induced pluripotent stem cells. CM: cardiomyocytes.

Figure 2

Assessment of tumorigenicity by animal models. (A) Workflow of cell xenografting and tissue quantitation. (B) Representative H&E images of teratoma containing the tissues derived from three germ layers. (C) Development of a simple grading system for visualizing growing tumors. The grading scale is as follows: 0 = no teratoma, 1 = teratoma just visible, 2 = teratoma visible, 3 = large teratoma with a typical size > 1 cm. (D) Quantitation of tumorigenicity of rare ESC spiked in non-proliferating feeder cells. Images reprinted from [65] with permission.

Figure 3

Assessment of tumorigenicity by PCR. (A) Workflow of RNA extraction, cDNA synthesis and their use in RT-PCR procedure to determine the gene expression (B) Raw data from ddPCR showing the expression of LIN28 mRNA in cardiomyocyte samples with different numbers of spiked-in ESCs. The threshold for positivity was assigned manually. FAM: 6-Carboxyfluorescein. (C) Raw data from ddPCR showing the copy numbers of LIN28 mRNA in 50 ng of total RNA. The threshold of the lower limit of detection (LLOD) is determined manually. Images reprinted from [67] with permission.

Figure 4

SCQC rapidly identifies undifferentiated hPSCs in differentiated cardiomyocytes (CMs) and definitive endodermal progenitors. (A) Overview of the SCQC approach. Cells were magnetically labelled, profiled by a microfluidic capture device, and quantified under a fluorescence microscope. (B) Representative microscope images of captured hPSCs on-chip. hPSCs were defined as DAPI⁺, OCT3/4⁺, and NANOG⁺. (C) Capture performance of rare hPSC samples spiked in 1 million derived cardiomyocytes. SCQC robustly captured the spike-in rare hPSCs in the 5–5000 cells range. Statistically, a difference was observed between 0 cell and 5 cells, which indicates the limit of detection of SCQC is 0.0005% or better. (D) Representative images of the teratomas formed from the cell samples quantified by SCQC. Images reprinted from [57] with permission.

Figure 5

Assessment of tumorigenicity by soft-agar assay. (A) Workflow of colony-based assessment of rare tumorigenic cells. (B) Representative images of Cy-5 labeled HeLa cells spiked in human mesenchymal stem cells. Due to its tumorigenic nature, HeLa cells form significantly bigger colonies over 30 days’ culture. (C) Images of automated detection of the colony formed by the single HeLa cells spiked in human mesenchymal stem cells. The mitochondria images were achieved by staining the wells with MitoTracker. The nucleus images were captured by staining the wells with Hoechst 33342. Images reprinted from [99] with permission.