V. Endoscopic Treatment

Endoscopic treatments for early gastric cancers are divided into two groups: mucosal resection methods and coagulation methods (e.g., laser therapies). Mucosal resection methods are usually preferred because through these we can obtain pathological specimens to evaluate. This chapter will present an overview of the current techniques and their future directions.

2. Indication

The indication of endoscopic treatment is limited to gastric cancer without lymph node metastasis. The presence and absence of lymph node metastasis is closely related to the histology, size, and depth of the lesion.

The Japan Gastroenterological Endoscopy Society recommends endoscopic treatment for:

1.Elevated type, differentiated adenocarcinoma of less than 20 mm in size, that does not reach deeper than the mucosal layer

2.Depressed type, differentiated adenocarcinoma of less than 10 mm in size, that does not reach deeper than the mucosal layer and does not have an ulcer or ulcer scar

EMR for early gastric cancer can be performed using the method of Tada et al. [2], which requires a doublechannel endoscopy that has two forceps channels (Fig. 1). There are two types of endoscopy: front-viewing endoscopy and oblique-viewing endoscopy. Obliqueviewing endoscopy is used for lesions on the posterior wall or in the cardiac region, which are difficult to treat with usual endoscopy. The disadvantage of the doublechannel endoscopy method is its uniqueness. This type of endoscopy is rarely available in a usual setting, thus we perform this complex procedure only in highly equipped facilities.

3.2 Single-Channel Endoscopy Method

A more usual endocopic method is cup-fitted panendoscopy that is equipped with a transparent plastic hood at the tip of the endoscope (EMRC) (Fig. 2) [3]. With this method, we are able to use a regular endoscope. After suctioning the lesion into the transparent hood, we cut it off with the snare. The advantages of this method are that we do not need a special endoscope and we can easily reach sites that are less approachable by other EMR methods.

3. Choice of Procedure:

Endoscopic Mucosal Resection

Endoscopic mucosal resection (EMR) is similar to surgical resection in that we can obtain specimens for pathological examination, while laser therapy and microwave coagulation methods coagulate or vaporize the lesion, leaving no specimens for further examination. Since the latter methods do not allow us to know whether any malignant cells remain, they are not suitable as first choice for curative treatment. The EMR method [1]) is thus preferred, and various techniques have been developed in order to improve the procedure.

3.3 Endoscopic Submucosal

Dissection (ESD)

Current methods such as double-channel endoscopy methods or EMRC have been shown to be effective in treating early gastric cancers, but may not achieve en bloc resection in all cases.

The ESD technique, which makes en bloc resection possible, has been developed in Japan. The actual steps of this procedure are: (a) burn the lesion boundary, marking with the tip of electrocoagulation devices; (b) inject saline or mucinous substances mixed with indigo carmine into the submucosal layer beneath the lesion;

(c) precut the rim of the marked lesion; (d) separate the submucosal layer and mucosa with an insulated tip knife

(IT knife) [4]) and/or other devices (hook knife, flex knife, snare, etc.); and (e) remove the entire tumor in one piece (en bloc resection). New equipment has been designed to make the ESD technique much safer. An IT

Fig. 2. Endoscopic mucosal resection using cup-fitted panendoscopy: EMRC

knife is a knife that has a ball made of ceramic material at the tip of the high-frequency cutting knife to prevent the knife from penetrating too deeply (Fig. 3).

4. Evaluation of the Procedure

Specimens obtained by EMR should be handled in the following manner. The specimen is spread out, pinned on a flat cork, and fixed in formalin solution. The size of specimen, the size and shape of the tumor, and the margins should be recorded on a schematic diagram. The fixed materials should be sectioned serially at 2-mm

intervals parallel to a line that includes the closest resection margin of the specimen (Fig. 4).

The key point of the evaluation is whether the whole lesion has been resected completely. It is easy to evaluate the en bloc specimen in which the resected specimen has enough distance between the resected margin and the border of the lesion, but the evaluation is rather difficult if we have resected the lesion in separate pieces. The Japanese guidelines for gastric cancer treatments [5] recommend that the results of resections should be evaluated in the following format:

Resection EA:

Depth M (mucosa), histologically pap or tub, no ulcer or ulcer scar in the tumor, vertical margin (VM)(-), no tumor cells within 1 mm of lateral margin (LM), neither lymphatic nor venous invasion

Resection EB:

No margin involvement but not fulfilling criteria for “EA”

Resection EC: VM(+) and / or LM(+) (LM, lateral margin: VM, vertical margin)

When the lesion is resected in separate pieces, we have usually marked the lesion using a high-frequency knife and/or biopsy scars. However, we are apt to lose track of the site in those cases where a piecemeal resection has been necessary. The use of clips can help us work out the delineation and enables resection of the entire lesion without leaving cancer cells behind [6]. By fixing the colored clips around the lesion before resection, we can deduce the margin of the lesion and whether any residue is left after the procedure. Tani et al. [7] reported that they were able to resect an entire lesion of more than 15 mm in diameter by means of several well-planned partial resections.

5. Results

In a review of 1832 cases of early gastric cancer treated by EMR in 12 facilities [8–11], total removal of cancer was successful in 1353 cases (74%), but in the remaining 479 cases (41%) the tumors could be removed only partially. Of these, 195 patients subsequently underwent surgical resection.

5.1 Partially Resected Cases

The partially resected cases were treated as follows: repeated EMR in 12.7% (61 cases), laser therapy in 12.9% (62 cases), both repeat EMR and laser therapy in 0.8% (4 cases), ethanol injection plus heater probe coagulation in 15.2% (73 cases), and heater probe coagulation in 1.5% (7 cases).

6. Complications

Major complications of EMR are bleeding and perforation. The reported risk of bleeding varies from 1.2% to 11.8% while the risk of perforation is considered to be 0.4%–2.4% [11].

7. Extended Indications for Endoscopic Treatment

Cancers limited to the mucosal layer can be resected by endoscopy, but when we find invasion deeper than the mucosa in the resected specimen, we usually treat them surgically in our facility. However, when the patient’s

condition does not permit surgery, we can achieve a good outcome by using Nd-YAG laser therapy [12]. Fujisaki et al. [13] reported finding no lymph node metastasis in lesions invading only less than 200 mm into the submucosa. Further research may enable us to better predict the risk of lymph node metastases in various cases, possibly extending the indications for EMR.

8. Problems of Endoscopic

Treatment

The most important point we have to be careful about in performing EMR is the possibility of synchronous lesions and metachronous lesions. Synchronous lesions are those that occur in multiple sites at the same time and metachronous lesions are those that occur after a successful resection of the original lesion during the long-term follow-up. Iwanaga et al. [14] reviewed a series of patients who underwent surgical resection and found by thorough pathological examination that 12.8% had synchronous lesions while only 8.6% were recognized before the surgery. In an independent review, Takagi et al. [15] reported that as many as 18.7% of early gastric cancer patients had synchronous lesions. Takekoshi [16] followed up post-EMR cases and reported the frequency of synchronous and/or metachronous lesions as 14.3%, 67% of which were synchronous lesions. In conclusion, even for small cancers that could be successfully removed by EMR, we have to sufficiently explore synchronous lesions and closely follow these up for possible metachronous lesions.

9. Conclusion

Endoscopic mucosal resection has been recognized as a less invasive procedure than surgical resection. We consider that it is important to further evaluate the long-term outcomes.

References

1.Takahashi H, Fujita R, Sugiyama K, et al (1991) Endoscopic therapy of gastric cancer: comparison of endoscopic mucosal coagulation and resection. Dig Endosc 3:215–221

2.Tada M, Shimada M, Yanai H, et al (1984) New technique of biopsy [in Japanese with English abstract]. Stomach Intest 19:1107–1116

3.Takeshita K, Tani M, Inoue H, et al (1997) Endoscopic treatment of early esophageal or gastric cancer. Gut 40:123–127

194 V. Endoscopic Treatment

4.Ono H, Kondo H, Gatoda T, et al (2001) Endoscopic mucosal resection for treatment of early gastric cancer. Gut 48:151–152

5.Japanese Gastric Cancer Association (1998) Japanese classification of gastric carcinoma, 2nd English edition. Kanehara, Tokyo

6.Inatsuchi S, Tanaka M (1994) Clinical evaluation of an improved technique in strip biopsy for gastric lesion [in Japanese with English abstract]. Gastroenterol Endosc 36: 939–948

7.Tani M, Takeshita K, Saeki I, et al (1997) Protection of residue or recurrence following endoscopic mucosal resection for gastric tumorous lesion [in Japanese with English abstract]. Prog Digest Endosc 50:74–78

8.Takekoshi T, Baba Y, Ota H, et al (1994) Endoscopic resection of early gastric carcinoma: result of a retrospective analysis of 308 cases. Endoscopy 26:352–358

9.Tada M, Matsumoto Y, Murakami A, et al (1993) Problems and their solution in curative endoscopic resection of early gastric carcinomas. Endosc Digest 5:1169–1174

10.Takahashi H, Kojima T, Parra A, et al (1997) Clinical evaluation of endoscopic therapy for early gastric cancer. Endoscopy 29:E21

11.Kojima T, Parra-Blanco A, Takahashi H, et al (1998) Outcome of endoscopic mucosal resection for gastric cancer; review of the Japanese literature. Gastrointest Endosc 48:550–555

12.Yasuda K, Miguma Y, Nakajima K, et al (1993) Endoscopic laser treatment for early gastric cancer endoscopy 25:451–454

13.Fujisaki J, Ikegami M, Oota Y, et al (1997) Endoscopic mucosal resection for early gastric cancers: its follow up and management for problematic cases [in Japanese with English abstract]. Prog Digest Endosc 50:70–73.

14.Iwanaga T, Koyama H, Imaoka S, et al (1988) Occurrence of a heterochromous cancer in the remnant stomach following partial gastrectomy in gastric cancer. Gan No Rinsho 34:442–446

15.Takagi K, Ohashi I, Ohta T, et al (1980) Histological transefiguration in gastric cancer [in Japanese with English abstract]. Stomach Intest 15:11–19

16.Takekoshi T (1992) Prognosis of endoscopic resection in early gastric cancer—establishment for radical treatment [in Japanese with English abstract]. Prog Digest Endosc 41:139–143