This is the first report of solitary metastatic adenocarcinoma of CUP detected at the No.8a LN region. There have been three reports of perigastric or suprapancreatic LN metastasis derived from CUP, including squamous cell carcinoma, neuroendocrine tumor, and adenocarcinoma [8,9,10]. With regard to “No.8a LN” metastasis of adenocarcinoma of CUP, there was a case report of No.8a and No.3 metastasis [11], but the metastasis solitary in No.8a LN has never been reported. Furthermore, poorly differentiated abdominal LN metastasis of CUP is rare, less than well or moderately differentiated adenocarcinoma, neuroendocrine tumor, and squamous cell carcinoma [4]. Therefore, this case report is considered valuable because it suggests an extremely rare clinical condition that has never been discussed before.

We initially suspected that this tumor might be a recurrence from GC that had been treated by ESD 26 years ago. However, past GC was early stage, well-differentiated carcinoma, with no lymphovascular invasion, completely dissected, and recurrence has never observed for more than 20 years after resection. Moon et al. reported that rate of recurrence of advanced GC within 5–10 years was 8.8%, after 10 years was 2.0% [12], but in the case of early GC, no case report of recurrence has been found after more than 20 years. Furthermore, in terms of morphology, present tumor (poorly differentiated) and past GC (well-differentiated) was different. Wang et al. reported that there had been difference between preponderant histology of primary GC and histology of metastatic GC in 14.1% of cases [13], but these differences had been inferred to be due to heterogeneity of primary GC. The GC resected 26 years ago was consist of only well-differentiated adenocarcinoma, and expected not to be containing poorly differentiated adenocarcinoma. According to these results and reports, we concluded this tumor was not derived from the GC resected 26 years ago.

Immunohistochemistry findings of this case indicated that this tumor might be derived from gastrointestinal tract, mainly colon according to the NCCN guideline and ESMO guideline [3, 14], but no primary cancer was found in the abdomen or mediastinum on any of the examinations performed before or after the radical surgery. From these results, we finally concluded this LN metastasis was derived from CUP. For certain primary cancers, such as melanoma or ovarian cancer, immunohistochemistry of tumor marker is effective for diagnosis of primary regions [3, 14,15,16,17]. For gastrointestinal cancer, especially GC, however, there are few specific markers effective for diagnosis for primary regions, and it might be difficult to detect primary tumor of CUP by only histology or biological effort. Systemic and repetitive investigation (CT, upper and lower gastrointestinal endoscopy, FDG-PET CT) are essential to recognize the primary site [3, 14, 18].

Laparoscopic LN resection is standardized in laparoscopic gastrectomy, and suprapancreatic LN dissection is reported to be safe while resection is proceeded at the appropriate layer [19, 20]. In this case, tumor was in contact with pancreas, CHA and involved RGAV. However, by taking advantage of the magnification effect of laparoscopy and dissecting the layer bordering the pancreas and CHA, the tumor could be radically resected without damaging the tumor capsule. In these cases of adenocarcinoma of LN, we standardize the procedure of lymphadenectomy as the procedure of gastrectomy. Oncological and surgical safety of systemic gastrectomy contained lymphadenectomy in the case of LN metastasis is guaranteed by laparoscopic resection [7]. Laparoscopic lymphadenectomy for CUP can be considered safe to perform.