Case

A 70-year-old woman presented with white stool and the patient was referred to our hospital for further investigation of jaundice. The patient had undergone cholecystectomy for the diagnosis of a non-dilated biliary tract with PBM (P-C type) approximately 30 years prior at another hospital. On physical examination, the patient’s abdomen was soft, and no mass was palpated. Laboratory data on admission revealed high levels of carcinoembryonic antigen (6.1 ng/mL), while carbohydrate antigen 19 − 9 was within normal ranges. Esophagogastroduodenoscopy revealed a tumor in the papilla of Vater, and the histological examination of the biopsy specimens revealed adenocarcinoma. (Fig. 1A). Endoscopic ultrasound and intraductal ultrasonography showed that the tumor was located in the common channel with no invasion to the sphincter of Oddi, duodenal muscular layer, or pancreas (Fig. 1B and C). Abdominal enhanced computed tomography (CT) revealed a 14 × 14 mm tumor in the duodenum. No enlarged lymph nodes or distant metastases were observed (Fig. 2). Magnetic resonance cholangiopancreatography demonstrated dilatation of the extra/intrahepatic bile duct and main pancreatic duct; the length of the common channel was 23 mm (Fig. 3). We finally diagnosed her with PVca with a non-biliary dilated PBM (cT1aN0M0, cStage IA, according to the Union for International Cancer Control [UICC], 8th edition), and subtotal stomach-preserving pancreaticoduodenectomy was performed.

Fig. 1

Endoscopy and echography. (A) Esophagogastroduodenoscopy revealed a tumor of the papilla of Vater. (B, C) Endoscopic ultrasound and intraductal ultrasonography revealed that the tumor was located in the common channel and demonstrated no invasion of the duodenal muscular layer and pancreas. CBD, Common bile duct; MPD, Main pancreatic duct

Fig. 2

Abdominal enhanced computed tomography (CT). CT revealed a tumor in the duodenum, with a size of 14 × 14 mm (arrow). CBD, Common bile duct; MPD, Main pancreatic duct

Fig. 3

Magnetic resonance choledochopancreatography (MRCP). MRCP revealed dilatation of the extra/intrahepatic bile duct and main pancreatic duct, and the length of the common channel was 23 mm

Surgical procedures

A median incision was placed in the upper abdomen. The pancreas was dissected at the anterior surface of the superior mesenteric vein. The modified Child’s reconstruction procedures were performed, and the remnant pancreas was anastomosed with the jejunal limb using the modified Blumgart method. Neither peritoneal dissemination nor lymph node metastases were detected during surgery. The regional lymph nodes of the papillary carcinoma were dissected. The operation lasted 370 min, and the estimated blood loss was 15 mL. No intraoperative blood transfusions were required.

Postoperative course

The postoperative course was uneventful, and the patient was discharged on the 24th postoperative day. The patient was recurrence-free for 4 years after surgery.

Macroscopic and pathological findings of the resected specimen

This was a PBM case without biliary dilation (P-C type), the tumor was diagnosed as PVca developing from the epithelium of the common channel, and the tumor diameter was 9 × 8 mm. The tumor invaded the Oddi sphincter and submucosa but did not invade the muscularis propria of the duodenum (No. 1, Fig. 4). The pathological diagnosis was pT1bN1M0 pStage IIIA, according to the UICC, because of the presence of a positive lymph node (2/37 lymph nodes). A front was observed between hyperplasia and dysplasia areas within the mucosal epithelium of the common channel, slightly upstream of the main tumor (No. 2, Fig. 4). At the hyperplastic area, there was no evidence of increased nuclear-to-cytoplasmic ratio, increased nuclear chromatin and loss of nuclear polarity, or cell overlap (No. 2 and No.3, Fig. 4, Supplement Fig. 1). In addition, at the dysplastic area, the findings of disturbed polarity, increased nuclear chromatin, and increased nuclear-to-cytoplasmic ratio suggested that the tumor was equivalent to BilIN-3 (high grade dysplasia) (No. 2, Fig. 4). A hyperplastic mucosa was found throughout the common bile duct (No. 3, Fig. 4). Immunostaining revealed positivity for CEA, COX-2, HER2, and IL-33 in the carcinoma (Fig. 5). CK7 and MUC1 positivity; MUC5 partial positivity; CDX2 and MUC2 negativity; and mostly CD20 negativity indicated that this PVca was of the pancreatobiliary type, not gastric type. In addition to MUC6 negativity in the carcinoma area, CDX2 was also negative, thereby we did not determine the lesion to be of the intestinal type [8]. p53 was wild-type immunostaining pattern (Supplement Fig. 2).

Fig. 4

Morphological evaluation derived from pathological findings. No adenomas were observed, suggesting a hyperplasia–dysplasia–carcinoma sequence carcinogenic mechanism. No. 1: Main part of the carcinoma (C) in the common channel. No. 2: The depiction of front between hyperplasia (H) area, dysplasia (D) area, and carcinoma (C) area is observed in the common channel. No. 3: Hyperplasia (H) in the common bile duct. There was no evidence of increased nuclear-to-cytoplasmic ratio, increased nuclear chromatin and loss of nuclear polarity, or cell overlap. MP, Muscularis propria; Panc., Pancreas

Fig. 5

Immunohistological findings of the resected specimen. CEA was highly expressed at all sites of the hyperplasia, dysplasia, and carcinoma. COX-2, HER2, and IL-33 expression were positive in the carcinoma tissues

Genetic analysis using FFPE and LB1–1 Cell-free total nucleic acid (cfTNA) and genomic DNA extraction

Thirteen plasma samples were collected between July 2019 and September 2020. Cell-free total nucleic acid (cfTNA) was extracted using the MagMAX™ Cell-Free Total Nucleic Acid Isolation Kit (Thermo Fisher Scientific) or the NextPrep-Mag™ cfDNA Automated Isolation Kit (PerkinElmer), according to the manufacturers’ protocols. Genomic DNA was extracted from the buffy coat using the FlexiGene DNA Kit (Qiagen).

Ten 5-µm slices of FFPE slides were used to extract genomic deoxyribonucleic acid (DNA). Genomic DNA from the tumor tissue, common bile duct, and normal tissue was extracted using a GeneRead™ DNA FFPE Kit (Qiagen), according to the manufacturer’s protocol. The extracted cfTNA and genomic DNA were quantified using the Qubit™ DNA (High Sensitivity) Assay Kit and Qubit DNA (Broad Range) Assay Kit (Thermo Fisher Scientific), respectively. The quality and size of the extracted cfTNA were evaluated with the High-Sensitivity D5000 ScreenTape Assay (Agilent) and the quality of genomic DNA was evaluated with the Genomic DNA ScreenTape Assay (Agilent), using TapeStation (Agilent).

1–2 Library construction

The NGS library was constructed using the Oncomine™ Pan-Cancer Cell-Free Assay (Thermo Fisher Scientific), according to the manufacturer’s protocol. Libraries were constructed using 12.3–20 ng of cfTNA, and 30 ng of genomic DNA from buffy coat.

Regarding tumor tissue, libraries were prepared using 40 ng of extracted genomic DNA from FFPE using the Ion AmpliSeq™ Comprehensive Cancer Panel (Thermo Fisher Scientific) with Ion Xpress™ Barcode Adapters (Thermo Fisher Scientific), according to the manufacturer’s protocol (Thermo Fisher Scientific). The quality of all constructed libraries was evaluated with a High-Sensitivity D1000 ScreenTape (Agilent), using TapeStation (Agilent).

1–3 Targeted NGS

The constructed libraries were subjected to template preparation using the Ion Chef™ System (Thermo Fisher Scientific) with either the Ion 540 Chef Kit (Thermo Fisher Scientific) or the Ion 550 Chef Kit (Thermo Fisher Scientific). Thereafter, sequencing was performed using the Ion GeneStudio™ S5 Prime System (Thermo Fisher Scientific).

1–4 Sequencing data analysis

Sequence alignment with hg19 as the reference genome and variant calling were performed using the Torrent Suite Software v5.16 and Ion Reporter v5.16 and v5.18. The workflows used for analyses included Oncomine TagSeq Pan-Cancer Liquid Biopsy w2.5 for cfDNA as well as buffy coat, and AmpliSeq CCP w1.2 Tumor–Normal pair for tumor genomic DNA with default parameters. The cutoff for variant calling in cfDNA was 0.065%. Regarding tumor-tissue alterations, mutations with a mutant allele frequency ≥ 5% were considered positive after excluding variants (single-nucleotide polymorphisms) detected in normal tissue. Mutations detected in the buffy coat that were also detected in the plasma cfTNA were evaluated as clonal hematopoiesis-associated mutations [9, 10].

1–5 Results of genetic analysis

Genetic mutations with single nucleotide variants exclusively detected in tumors and not in normal tissue and bile ducts from FFPE specimens included ERBB2 (Mutant allele frequency; MAF, 81.95%), POU5F1 (MAF, 12.43%), FLT1 (MAF, 9.91%), NCOA2 (MAF, 8.00%), and KMT2D (MAF, 7.14%). ERBB2 was also detected as a genetic mutation with copy number variant present exclusively in tumors compared to normal tissue, with a copy number of nine (Supplement table). The genetic mutations with single nucleotide variants detected in bile ducts included KIT (MAF, 11.03%) (Table 1). NGS identified no genetic abnormalities in p53. Considering the immunostaining results, this case was considered wild-type for p53. The cell-free DNA (cfDNA) (baseline) obtained from preoperative plasma was ERBB2 (MAF, 0.24%) and ERBB2 was never detected after surgery (Supplement Fig. 3).

Table 1 Genetic mutations with single nucleotide variant of mutant allele frequency > 5% detected only in tumors and bile ducts compared to normal tissue