Multidimensional fragmentomic profiling of cell-free DNA released from patient-derived organoids.

BACKGROUND: Fragmentomics, the investigation of fragmentation patterns of cell-free DNA (cfDNA), has emerged as a promising strategy for the early detection of multiple cancers in the field of liquid biopsy. However, the clinical application of this approach has been hindered by a limited understanding of cfDNA biology. Furthermore, the prevalence of hematopoietic cell-derived cfDNA in plasma complicates the in vivo investigation of tissue-specific cfDNA other than that of hematopoietic origin. While conventional two-dimensional cell lines have contributed to research on cfDNA biology, their limited representation of in vivo tissue contexts underscores the need for more robust models. In this study, we propose three-dimensional organoids as a novel in vitro model for studying cfDNA biology, focusing on multifaceted fragmentomic analyses. RESULTS: We established nine patient-derived organoid lines from normal lung airway, normal gastric, and gastric cancer tissues. We then extracted cfDNA from the culture medium of these organoids in both proliferative and apoptotic states. Using whole-genome sequencing data from cfDNA, we analyzed various fragmentomic features, including fragment size, footprints, end motifs, and repeat types at the end. The distribution of cfDNA fragment sizes in organoids, especially in apoptosis samples, was similar to that found in plasma, implying occupancy by mononucleosomes. The footprints determined by sequencing depth exhibited distinct patterns depending on fragment sizes, reflecting occupancy by a variety of DNA-binding proteins. Notably, we discovered that short fragments (< 118 bp) were exclusively enriched in the proliferative state and exhibited distinct fragmentomic profiles, characterized by 3 bp palindromic end motifs and specific repeats. CONCLUSIONS: In conclusion, our results highlight the utility of in vitro organoid models as a valuable tool for studying cfDNA biology and its associated fragmentation patterns. This, in turn, will pave the way for further enhancements in noninvasive cancer detection methodologies based on fragmentomics.

Identification of New Pathogenic Variants of Hereditary Diffuse Gastric Cancer.

Purpose: Hereditary diffuse gastric cancer (HDGC) presents a significant genetic predisposition, notably linked to mutations in the CDH1 and CTNNA1. However, the genetic basis for over half of HDGC cases remains unidentified. The aim of this study is to identify novel pathogenic variants in HDGC and evaluate their protein expression. Materials and Methods: Among 20 qualifying families, two were selected based on available pedigree and DNA. Whole genome sequencing (WGS) on DNA extracted from blood and whole exome sequencing (WES) on DNA from formalin-fixed paraffin-embedded tissues were performed to find potential pathogenic variants in HDGC. After selection of a candidate variant, functional validation and enrichment analysis were performed. Results: As a result of WGS, three candidate germline mutations (EPHA5, MCOA2, and RHOA) were identified in one family. After literature review and in silico analyses, the RHOA mutation (R129W) was selected as a candidate. This mutation was found in two gastric cancer patients within the family. In functional validation, it showed RhoA overexpression and a higher GTP-bound state in the RhoaR129W mutant. Decreased phosphorylation at Ser127/397 suggested altered YAP1 regulation in the Rho-ROCK pathway. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses linked RhoAR129W overexpression to changed migration/adhesion in MKN1 cell line. However, this RHOA mutation (R129W) was not found in index patients in other families. Conclusion: The RHOA mutation (R129W) emerges as a potential causative gene for HDGC, but only in one family, indicating a need for further studies to understand its role in HDGC pathogenesis fully.

Tagmentation-based analysis reveals the clonal behavior of CAR-T cells in association with lentivector integration sites.

Integration site (IS) analysis is essential in ensuring safety and efficacy of gene therapies when integrating vectors are used. Although clinical trials of gene therapy are rapidly increasing, current methods have limited use in clinical settings because of their lengthy protocols. Here, we describe a novel genome-wide IS analysis method, "detection of the integration sites in a time-efficient manner, quantifying clonal size using tagmentation sequencing" (DIStinct-seq). In DIStinct-seq, a bead-linked Tn5 transposome is used, allowing the sequencing library to be prepared within a single day. We validated the quantification performance of DIStinct-seq for measuring clonal size with clones of known IS. Using ex vivo chimeric antigen receptor (CAR)-T cells, we revealed the characteristics of lentiviral IS. We then applied it to CAR-T cells collected at various times from tumor-engrafted mice, detecting 1,034-6,233 IS. Notably, we observed that the highly expanded clones had a higher integration frequency in the transcription units and vice versa in genomic safe harbors (GSH). Also, in GSH, persistent clones had more frequent IS. Together with these findings, the new IS analysis method will help to improve the safety and efficacy of gene therapies.

Evaluation of the 1B Equation to Estimate Glomerular Filtration Rate in Pediatric Patients with Cancer.

The 1B equation is recommended for calculating the glomerular filtration rate (GFR) in children. Since few reports have evaluated the performance of the 1B equation, we investigated the performance of estimated GFR (eGFR) equations with the blood urea nitrogen (BUN) variable for pediatric cancer patients. In total, 203 children with cancer who underwent measured GFR (mGFR) assessment were enrolled. The median (range) mGFR and eGFR calculated using the updated Schwartz equation were 118 (43-241) and 135 (34-257) mL/min/1.73 m², respectively. The bias, precision (root mean square error [RMSE]), and accuracy (P30, mGFR±30%) of three eGFR equations including updated Schwartz, 1B, and full age spectrum (FAS) were compared. The median bias (mL/min/1.73 m²) was: updated Schwartz, 8.5; 1B, -9.0; and FAS, 4.2. The biases for all three eGFR equations were significantly different from zero. The P30 was: updated Schwartz, 63.5%; 1B, 66.0%; and FAS, 66.0%. The RMSE was the lowest for the 1B equation (40.4), followed by FAS (42.3), and updated Schwartz (45.5). The median eGFR/mGFR ratio for the eGFR equations decreased with age and reduced kidney functions (i.e., increased creatinine and BUN concentrations). The bias may be further reduced by using the average from two equations, such as the updated Schwartz and 1B, or FAS equation, rather than using the updated Schwartz or 1B equation alone. The use of the 1B equation may underestimate the GFR. Using creatinine and BUN variables in the eGFR equation may yield a more accurate estimate of the GFR in pediatric cancer patients.