A case of colon cancer implanted on endoscopic resection ulcer certified by cancer genomic testing

We present the case of a 90 year-old man diagnosed with rectal cancer. In December 2018, he presented with back pain, and computed tomography (CT) findings revealed wall thickening at the splenic flexure. He was referred to our hospital for a colonoscopy, which showed a circumferential neoplastic lesion in the descending colon and multiple colorectal tumors. Colonoscopy revealed a 30 mm type 2 advanced cancer in the descending colon (Lesion B), a 10 mm 0-Is lesion in the descending colon, a 6 mm 0-Is lesion in the sigmoid colon, and a 20 mm 0-IIa lesion in the rectum above the peritoneal reflection (Ra) (Lesion A). Before surgery for advanced cancer of the descending colon, endoscopic mucosal resection (EMR) was performed on the 10 mm 0-Is lesion in the descending colon and 6 mm 0-Is lesion in the sigmoid colon, while ESD was performed on the 20 mm 0-IIa lesion on the Ra (Lesion A, Fig. 1). No clipping or other suture treatments were performed at the bottom of the ulcer. Pathological examinations of these three lesions revealed tubular adenomas with negative vertical and horizontal margins. One month after endoscopic treatment for colorectal neoplasms, he underwent laparoscopic resection of the descending colon and D3 dissection for advanced cancer. The pathological findings were as follows: tub2 > tub1, pT3(SS-A) INF b, Ly 1a, V1a, Budding grade (BD) 1, Pn1a, pN1b, pM0, and pStage IIIb (Lesion B, Fig. 2). One year after surgery for descending colon cancer, surveillance colonoscopy revealed type 2 advanced cancer in the rectum. Colonoscopy revealed a semi-peripheral ulcerated carcinoma, approximately 40 mm in size, with sharply demarcated and raised margins on the Ra (Lesion C, Fig. 3). Barium colonography revealed a shadow defect on the lateral image on the right lateral wall of the Ra (Fig. 3). This lesion on the Ra was located on the scar resulting from the previous ESD of the rectal tumor. Contrast-enhanced CT of the thorax and abdomen and PET-CT showed no evidence of metastasis to the lymph nodes or other organs. Therefore, the lesion was surgically resected. The pathological findings of this lesion revealed tumor tissue consisting of moderately large or small irregular ducts infiltrating into the mucosa of the outer membrane. The tumor cells were large and highly columnar, consistent with those observed in moderately differentiated ductal adenocarcinomas. The infiltrated lesion of the carcinoma was poorly differentiated and non-metastatic. There was no metastasis of cancer to the regional lymph nodes; however, venous invasion was observed. Based on these findings, the lesion was diagnosed as a tub2 > por2, pT3 (SS/A), INFb, Ly0, V1b, BD 1, Pn0, pN0, pM0, and pStage IIa (Fig. 3).

Fig. 1figure 1

Clinicopathological findings of lesion A. a Lesion A was located in the rectum above the peritoneal reflection. This lesion was observed as a flat, elevated-type lesion measuring 20 mm in size on white-light imaging. b Ulcer immediately after endoscopic submucosal dissection. c Resected specimen. d Frontal view of lesion A on barium colonography. e Lateral view of lesion A on barium colonography. f Tumor tissue observation reveals tubular adenoma with severe atypia, high grade

Fig. 2figure 2

Clinicopathological findings of lesion B. a Lesion B was located in the descending colon. This lesion was 30 mm in size in type 2 advanced cancer. b Barium colonography of lesion B. The lesion shows the apple core sign. c Macroscopic view of the surgically resected specimen of lesion B. d Tumor tissue observation reveals fusion and infiltration of small and large atypical glandular ducts from the superficial layers to the sub-serosal tissues. Venous invasion, lymphatic invasion, and intramural nerve invasion are observed. Tumor clusters are mild. Tumor metastasis is also observed in the lymph nodes. e High-magnification view of the pathology for lesion B. Atypical gland ducts corresponding to moderately differentiated adenocarcinoma are seen

Fig. 3figure 3

Clinicopathological findings of lesion C. a One year later, a 30-mm type 2 advanced cancer was observed on the ESD scar of lesion A. b Frontal view of lesion C on barium colonography. c Barium colonography showing a shadow defect on the lateral image of the right lateral wall of the Ra. d Sections show tumor tissue with moderately large or small irregular ducts fused in some areas and infiltrating from the mucosa to the adventitia. Most of the ductal adenocarcinomas are well-differentiated ductal adenocarcinomas with a poorly differentiated noncomplex component in the advanced areas. Venous invasion is also observed. Fibrosis is present in the deeper layers of tumor cells. e High-magnification view of the pathology for lesion C. Atypical gland ducts similar to those in lesion B are seen

We performed cancer multigene panel testing to investigate the correlation between lesions A, B, and C. Formalin-fixed, paraffin-embedded (FFPE) specimens of the three lesions were dissected into 10-µm-thick sections and placed on microscopic slides. DNA extraction was performed using a QIAamp DNA FFPE Tissue kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The quality and quantity of DNA from the three lesions and blood cells were confirmed for next-generation sequence library calculations using a Qubit 1.0 Fluorometer (Life Technologies, Grand Island, NY, USA) and Genomic DNA ScreenTape Analysis (Agilent Technologies, Santa Clara, CA, USA). Subsequently, sequence libraries targeting 468 cancer-related genes from the MSK-IMPACT Clinical Sequencing Cohort (Supplementary Table 1) were prepared from the DNA of the three lesions and blood cells using our sequence library method (see Supplementary Text). The resulting pooled libraries were sequenced via paired-end reads using the HiSeq X platform (Illumina, San Diego, CA, USA). Sequencing reads were analyzed and annotated as described in the Supplementary Text.

Deep sequencing via cancer multigene panel testing revealed that the sequence reads of all samples aligned almost 100% to the reference, with a mean depth of over 300. The numbers of somatic single nucleotide variants (SNVs) in lesions A, B, and C were 73, 31, and 10, respectively. Moreover, the numbers of pathogenic variants in lesions A, B, and C were 8, 5, and 4, respectively. No SNVs were detected in any of the three lesions. However, two SNVs were commonly identified between lesions A and C, and three SNVs were commonly detected between lesions A and C (Fig. 4, Supplementary Table 2). The two shared SNVs between lesions A and C were classified as variants of uncertain significance. In contrast, the three shared SNVs between lesions B and C were BRAF, PIK3R1, and TP53, all of which were determined to be pathogenic (Supplementary Table 2).

Fig. 4figure 4

Genomic landscape for lesions A, B, and C. a The association of clonal evolution among lesions B and C is shown using a phylogenetic tree. The green bar shows the common variants between A and C (shared lesions A and C). The yellow bar indicates the common variants between B and C (shared lesions B and C). Red, purple, and brown bars indicate variants found only in A, B, and C, respectively. b The genomic alterations in lesions A, B, and C are shown as bar graphs. Each line shows each variant in lesions A, B, or C. Shared variants on the right bar indicate common variants between A and C or B and C

This study was approved by the Institutional Review Board of Hiroshima University Hospital (approval number: E2012-9990) and conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from the patient.

留言 (0)

沒有登入
gif