Late recurrence of early gastric cancer 11 years after curative endoscopic submucosal dissection

Abstract

A 73-year-old man was initially diagnosed with a 50-mm, depressed-type early gastric cancer on the anterior wall of the angulus, and the lesion was curatively resected en bloc by endoscopic submucosal dissection. Pathology revealed a 54 mm × 43 mm differentiated-type predominant adenocarcinoma with focal undifferentiated-type component that was confined to the mucosa without ulceration. Eleven years after endoscopic submucosal dissection, lymph node metastasis along the lesser curve was incidentally detected on magnetic resonance imaging and diagnosed by endoscopic ultrasound-guided fine-needle aspiration biopsy. The patient received distal gastrectomy with D2 lymph node dissection. A resected lymph node revealed a well-differentiated adenocarcinoma with a poorly differentiated component; thus, the final diagnosis was late recurrence of early gastric cancer originally treated by endoscopic submucosal dissection. This report demonstrates that metastatic recurrence may occur in curative endoscopic submucosal dissection for early gastric cancer and that gastric cancer can recur even after more than 5 years.

Introduction

Endoscopic submucosal dissection (ESD) is becoming established as a minimally invasive treatment for early gastric cancer (EGC) with a negligible risk of lymph node metastasis [1]. With the development of ESD, en bloc resections can be achieved even for larger or ulcerative lesions. A Japanese multicenter prospective trail showed no post-ESD recurrence in differentiated-type mucosal cancer of any size and differentiated-type mucosal cancer measuring < 3 cm with ulceration [2]. Moreover, another Japanese multicenter prospective trial showed no post-ESD recurrence even in undifferentiated-type mucosal cancer measuring < 2 cm without ulceration [3]. As a result, differentiated-type mucosal cancers of any size, differentiated-type mucosal cancer measuring < 3 cm with ulceration and undifferentiated-type mucosal cancers measuring < 2 cm without ulceration were promoted to endoscopic curability A (eCuraA) in the 6^th Japanese Gastric Cancer Association (JGCA) guidelines [4]. However, a few reports have described recurrent EGC cases classified as eCuraA [5–8]. In addition, the follow-up period after ESD for EGC is not specified in the 6th JGCA guidelines and depends on the doctor’s discretion, although the guidelines suggest a 5-year follow-up after surgery for gastric cancer. We herein report a case of lymph node metastasis 11 years after curative ESD for EGC.

Case report

A 73-year-old man with no prior medical history except for eradication of Helicobacter Pylori underwent esophagogastroduodenoscopy (EGD) for cancer screening. EGD revealed a reddish, 50-mm sized depressed lesion with erosion on the anterior wall of the angulus, which appeared to be EGC limited to the mucosa (Fig. 1). A biopsy specimen revealed a well-differentiated adenocarcinoma. Computed tomography (CT) revealed neither lymph node nor distant metastasis. The lesion was classified under the expanded indication for ESD in the 3rd JGCA Guidelines [9] and was, therefore, resected en bloc with ESD.

Fig. 1.

Esophagogastroduodenoscopy revealed a reddish 50-mm sized depressed lesion with erosion on the anterior wall of the angular incisure which appeared to be EGC limited to the mucosa (yellow arrows)

Gross examination confirmed the resected specimen as a 50-mm depressed lesion (Fig. 2a). Microscopically, the tumor had predominantly irregularly shaped ducts with fused glands (Fig. 2b). In addition, poorly differentiated components were sparsely observed (Fig. 2c). The lesion was eventually diagnosed as tub2 > tub1 > por, 0-IIa, 54 mm × 43 mm, T1a(M), UL0, Ly0, V0, pHM0, pVM0 according to 3rd English edition of the Japanese classification of gastric carcinoma [10]. According to the 3rd JGCA Guidelines [9], which were in force at that time, the curability after ESD was judged as curative resection for tumors of expanded indication (eCuraA in the 6th JGCA Guidelines). Therefore, the patient was followed up by regular EGD and abdominal ultrasonography (US) without additional surgery, which indicated no evidence of recurrence for 5 years. In addition, annual EGD was continued after 5 years.

Fig. 2.

a Gross examination confirmed the resected specimen as a 50-mm depressed lesion. b The tumor had irregularly shaped ducts with predominantly fused glands. c Poorly differentiated components were sparsely observed (yellow arrows)

Eleven years after ESD, magnetic resonance imaging (MRI) for evaluation of adenomyomatosis of gallbladder incidentally found a 25-mm enlarged lymph node along the lesser curve (Fig. 3). Endoscopic ultrasonography (EUS) confirmed the mass at the same site detected on MRI (Fig. 4a), although EGD revealed no apparent recurrent findings at the ESD site or metachronous gastric cancer. We strongly suspected that the lymph node enlargement had been caused by lymph node recurrence of the primary EGC. EUS-guided fine-needle aspiration biopsy (EUS-FNAB) of the lymph node revealed a well-differentiated adenocarcinoma (Fig. 4b), which was similar to the primary gastric lesion. In addition, positron emission tomography revealed a high fluorodeoxyglucose uptake only in the lymph node confirmed on EUS without any other organs, suggesting that there are no primary cancers other than EGC; therefore, lymph node metastasis was diagnosed as resulting from the EGC originally treated by ESD.

Fig. 3.

Magnetic resonance imaging found a 25-mm enlarged lymph node along the lesser curve (yellow arrows)

Fig. 4.

a Endoscopic ultrasound revealed a 25-mm mass at the same site confirmed on magnetic resonance imaging. B Pathological examination of the endoscopic ultrasound-guided fine-needle aspiration biopsy revealed a well-differentiated adenocarcinoma

With the diagnosis of EGC recurrence without distant metastasis, the patient received distal gastrectomy with D2 lymph node dissection in conformity to gastric cancer with regional lymph node metastasis localized in the angulus. Pathologically, surgically resected specimen showed that the ESD site had no residual tumors and any other metachronous cancers were not detected in the stomach. In addition, closed-type atrophic gastritis was observed in the resected specimen. In a resected common hepatic artery lymph node, which corresponded to the lymph node obtained from EUS-FNAB, metastatic well-differentiated adenocarcinoma with poorly differentiated component was identified (Fig. 5). Immunohistochemical examination revealed that MUC5AC, CDX2, CK7, CK20 stain were positive both in the resected ESD specimen and the resected lymph node, which was comparable with recurrence of EGC. The final diagnosis was lymph node recurrence of EGC. The patient received adjuvant chemotherapy with capecitabine and oxaliplatin for 6 months according to the CLASSIC trial [11]. No recurrence was observed at 10 months after surgery.

Fig. 5.

Well-differentiated adenocarcinoma with poorly differentiated component (yellow arrow heads) in a resected lymph node

Discussion

Our case demonstrates two important clinical issues with regards to ESD for EGC. First, metastatic recurrence may possibly occur in ESD for EGC even those classified as eCuraA. Second, late recurrence of gastric cancer can still occur even more than 5 years after curative ESD. Furthermore, our case raises a problem about the prevailing follow-up period after ESD for EGC.

Gotoda et al. [12] reviewed the incidence of lymph node metastasis from surgically resected EGC and revealed that none of the differentiated EGC without ulcerations developed nodal metastasis regardless of tumor size. Moreover, a Japanese multicenter prospective trial showed no post-ESD recurrence in differentiated-type mucosal cancer of any size and differentiated-type mucosal cancer measuring < 3 cm with ulceration [2]. Recently, Takizawa et al. [3] reported that no recurrence was observed after curative ESD of undifferentiated-type mucosal cancer measuring < 2 cm without ulceration. Thus, recurrence of EGC is extremely rare in eCuraA cases.

However, a retrospective multicenter study showed that metastatic recurrence was observed in 0.14% (6/4202) of patients with eCuraA or eCuraB. In metastatic recurrence cases, one patient with eCuraB criteria had lymph node metastasis after 87 months from ESD, whereas five other patients with eCuraA or eCuraB had metastatic recurrence within 5 years [13]. Furthermore, four metastatic recurrent cases of EGC of eCuraA within 5 years have been reported [5–8]. Thus, notably, metastatic recurrence may possibly occur even in eCuraA cases.

The appropriate follow-up period after ESD is not specified in the current Japanese guidelines, and the length of follow-up depends on the primary doctor’s discretion. In surgical cases, the guidelines suggest that monitoring after surgery for gastric cancer should be continued for 5 years, although little evidence supporting the statements have been provided. In addition, the guidelines recommend that monitoring after more than 5 years should be continued depending on the individual cases, considering that Lee et al. [14] revealed that late recurrence after more than 5 years can occur in 2.8%.

In endoscopically resected EGCs, no cases of metastatic recurrences after more than 5 years in eCuraA have been reported. However, two cases of lymph node recurrence after more than 5 years in eCuraB criteria have been reported at 87 months and 74 months [13, 15].

As for risk factors for late recurrence, Choi et al. [16] analyzed the time until onset of recurrence after curative gastrectomy and reported that early-stage and differentiated histological type were associated with late recurrence. This finding may be applicable to EGC cases treated by ESD; therefore, follow-up for more than 5 years by CT scan and US, as well as endoscopy, may be needed, although late recurrence is extremely rare in such cases.

Lee et al. [14] reported that peritoneal seeding was the most common pattern of recurrence in surgical gastric cancer cases (76%), followed by distant metastasis (36%) and locoregional recurrence (8%). Moreover, Hanaoka [7] reported a case of lymph node recurrence and liver metastasis other than locoregional recurrence after curative ESD with eCuraA. These reports indicate that radical resection for metastatic recurrence is challenging due to distant metastasis or peritoneal seeding. In our case, MRI incidentally detected an enlarged lymph node along the lesser curve, which fortunately led to R0 resection with salvage surgery for metastatic recurrence because lymph node recurrence was localized.

Furthermore, histological examination in our case revealed a predominantly well-differentiated adenocarcinoma; however, poorly differentiated component was also sparsely identified. EGCs of mixed-type histology have been recently reported to have more malignant potential and tend to lead to lymph node metastasis compared with EGCs of pure-type histology [17–19]. Interestingly, in two of the four aforementioned curative ESD cases, histological examination revealed predominantly well-differentiated adenocarcinomas, including poorly differentiated components [6, 7]. Careful follow-up is essential after curative ESD, especially with mixed histological type.

In conclusion, metastatic recurrence may possibly occur in ESD for EGC even those classified as eCuraA, and late recurrence of gastric cancer can still occur even more than 5 years after curative ESD. Our report, thus, demonstrates that appropriate monitoring for curatively resected EGC is necessary. Furthermore, an appropriate follow-up period needs to be established for ESD for EGC.

Author contributions

KI wrote the initial draft of the manuscript. The other authors assisted in the preparation of the manuscript. All authors approved the final version of the manuscript.

Declarations

Conflicts of interest

The authors disclose no conflicts.

Research involving human participants and/or animals

This article does not contain any studies with human participants performed by any of the authors.

Informed consent

Informed consent was obtained from the patient who participated in this report.