Abstract 1690: A histology based approach to spatial single cell analysis of colorectal cancers

Abstract Background: Single cell sequencing is a powerful tool for the evaluation of intratumoral heterogeneity and the investigation of cancer evolution. Aims: By combining laser microdissection and single cell sequencing, we aimed to link tissue morphology and spatial information with sequencing data of the isolated cells. Materials & methods: In our preliminary study, we used fresh frozen tissue specimen of surgically resected material from a colorectal cancer (CRC) patient containing both cancerous and normal adjacent tissue (NAT). From part of the normal and cancerous tissue exome sequencing was performed in bulk (to assess somatic variants), while the other part was subjected to single cell sequencing. Fresh frozen tissues from both CRC and NAT were cryosectioned at -20°C with section thickness ranging from 16 to 25 µm to ensure that a layer of whole cells are present in the slides. Tissue slides were then scanned using a PANNORAMIC 1000 scanner (3DHISTECH Ltd.). After morphologic evaluation single normal colonocytes and cancerous cells were laser microdissected from the NAT and multiple CRC areas (invasive front, differentiated, non-differentiated cells) by using a CellCut Laser Microdissection system (MMI). The isolated cells were subjected to Repli-G Single Cell WGA Kit (Qiagen) and library preparation followed by whole exome sequencing (WES) on NextSeq 550 (Illumina). Blood sample was also collected before surgical treatment, cell-free DNA was isolated and exome sequencing was completed. Bioinformatic analysis was conducted using BaseSpace and GATK4 best practices. Common and unique variants were identified between cells and also compared with the bulk exom and cell free DNA sequencing results of the patient. Identified variants were further investigated using the oncoKB annotator. Results: Both healthy (1) and cancerous epithelial cells (3) were dissected and sequenced successfully. A median depth of coverage of 192 was achieved with a median of 43.3% of coverage of 50x or above in the target region compared to 73.9 and 52.8% 208 and 97.6% in tissue and plasma samples, respectively. Overall, we identified 105, likely oncogenic” and 2, predicted oncogenic” unique variants in the single cells using the oncoKB annotator. Among the cancerous single cells, we identified 10, likely oncogenic” and 1, predicted oncogenic” common variants (such as ARID4B, DNMT3B, MSH6 and HNF1A), while 15, 25 and 7, likely oncogenic” (such as CTLA4, MLH1, MSH2, CDK12, CDKN1B) variants were identified uniquely in the 3 cancerous cells. Conclusion: We were able to dissect, isolate and sequence single cells from CRC and NAT thus combining valuable morphologic information with sequencing data on a single cellular level with maintained spatial information. The distribution of variants among the single cells shows that it is a viable approach to investigate tumor heterogeneity and to link the morphologic phenotypes and genotypes of cancerous cells. Citation Format: William Jayasekara Kothalawala, Alexandra Kalmár, Gitta Szabó, Barbara Kinga Barták, Sára Zsigrai, Zsófia Brigitta Nagy, Ildikó Felletár, Krisztina Andrea Szigeti, István Takács, Béla Molnár. A histology based approach to spatial single cell analysis of colorectal cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1690.