The rediscovery of endoscopic ultrasound (EUS) in gastric cancer staging

Although its incidence varies greatly between Eastern and Western countries, gastric carcinoma still represents the second leading cause of cancer deaths worldwide [1]. Although modern surgical techniques allow high curative resection rates, long-term survival outcomes are still very disappointing [2]. For this reason, neoadjuvant radiochemotherapy regimens are increasingly adopted in attempts to improve the survival of patients with advanced gastric cancer [3]. To identify which patients should be allocated to preoperative treatments, reliable and accurate staging techniques are crucial.

Soon after the first descriptions of the five-layer structure of the gastric wall [4] [5] [6], endoscopic ultrasound (EUS) became a natural candidate technique for the staging of gastric cancer. The first reports in this field demonstrated good results in terms of sensitivity and specificity both for T and N staging [7] [8] [9].

Since then, gastric cancer has traditionally been a major area of research in EUS, mainly in regard to discrimination of tumor depth and detection of regional lymph node metastases, with more than 130 studies published in the period 1980 – 2010 [10]. In more recent years, the impact of EUS on patient management has also been evaluated in publications of high level of evidence, such as meta-analyses and systematic reviews [11] [12] [13].

Until a few years ago, the impact of EUS in gastric cancer was limited by the lack of therapeutic options, surgery being the only recourse either with curative or with palliative intent. The clinical arena of gastric cancer has changed substantially in recent years as treatments became more numerous. Besides the traditional surgical approach, endoscopic mucosal resection and submucosal dissection were adopted for the early stages of the disease (T1 N0) and neoadjuvant therapies (preoperative and perioperative) were introduced for the advanced stages (T3 /T4 and N + ).

As a consequence, the potential role of EUS in gastric cancer has become much more attractive to identify the patients suitable for minimally invasive treatment, those who should undergo primary surgery, and those who need neoadjuvant therapy. In this respect, the most recent guidelines of the National Comprehensive Cancer Network [14], an authority of international high repute, have introduced EUS as a preferred modality of gastric cancer staging if no evidence of M1 disease is present at computed tomography-positron emission tomography.

In this issue of Endoscopy, Kutup et al. [15] reported a large single-center retrospective study that assessed the potential influence of EUS on decision making in gastric cancer without distant metastases. The authors’ assumption was that chemotherapy would be applied only in locally advanced cancers (T3 /4 or any N + ), whereas primary surgery would be performed with T1 /2 N0 tumors. They analyzed the endosonographic and histopathological staging of 123 patients treated by primary surgery between 1993 and 2008. During this long study period, EUS was performed by six different expert endosonographers, using a radial echo endoscope at 7.5, 10, and 12 MHz. Overall T staging was correct in 45 % of the patients and overall N staging was correct in 72 %. Regarding the ability to discriminate between early and advanced stages of gastric cancer by EUS, the distinction between T1 /2 and T3 /4 tumors was made with 50 % sensitivity (correct recognition of T1 /2) and 81 % specificity (correct recognition of T3 /4). According to N staging (uN0 vs. uN + ), 41 % of tumors were understaged (being wrongly assigned to primary surgery), while 22 % were overstaged (being wrongly assigned to neoadjuvant treatment). Overall, correct decisions between primary surgery and neoadjuvant treatment based on EUS T stage results were made in only 55 /123 patients. In the event that the treatment strategy had depended solely on EUS results, 27 % of the patients would have been wrongly treated by primary surgery, because histopathology showed more advanced tumor stages, and 49 % of the patients would have been overtreated by neoadjuvant therapy because of EUS overstaging.

Based on their results, the authors concluded that the diagnostic accuracy of EUS in the clinical staging of gastric carcinoma is limited. However, the same results may also be seen in a more favorable light, i. e. that the main flaw of EUS was essentially represented by T overstaging. While EUS overstaging of other neoplasms, such as lung and pancreatic cancer, may lead to the ominous consequence of excluding a patient from potentially curative resection, overstaging gastric cancer carries the risk of patients undergoing some form of “overtreatment” but not of being excluded from surgery. Moreover, based on most recent approaches enrolling also T2 gastric cancers for preoperative regimens, what is now called “overtreatment” in many patients might no longer be considered as such.

A recent well conducted meta-analysis by Mocellin et al. [13] reported better EUS staging accuracy rates for gastric cancer than the study by Kutup et al. [15]. As far as the aim of distinguishing early vs. advanced T stages of gastric cancer was concerned, EUS showed 86 % sensitivity and 91 % specificity across 41 studies enrolling more than 3500 patients. As far as N staging was concerned, although sensitivity and specificity were lower compared with T staging, EUS could still be clinically informative as it increased the previous probability of being classified as N + from 55 % (average pre-test probability) to 84 % when positive, and it lowered the same probability to 31 % when negative.

What factors could have accounted for the results reported by Kutup et al. [15]? First, it should be mentioned that EUS was performed by six different endosonographers over the years. Although they are all respected authorities in this field, suboptimal interobserver agreement in gastric cancer staging by EUS is a well-known issue which can also occur among experienced observers [16]. Second, the TNM staging of gastric cancer was modified twice during the study period (in 1998 and in 2003); in particular, the lymph node classification was radically changed from a distance-from-tumor system to a numerical scoring system. The latter may have accounted for poor comparability among patients staged according to different classifications. Third, per protocol, only patients who had undergone radical primary surgery were enrolled, resulting in exclusion of early cancers which were treated endoscopically and of advanced cancers which were treated palliatively. As a result, a selection bias may have been introduced since 75 % of the patients belonged to the pathological T2 – T3 stages, in which EUS usually shows the lowest accuracy in contrast to the T1 and T4 stages. Fourth, it is reasonable to assume that the time elapsed between EUS staging and surgery might have been non-uniform among all the patients, thus introducing another bias in evaluating the accuracy of EUS. Lastly, N staging was based only on the morphologic evaluation of lymph nodes which suffers well-known limitations that are also pertinent to other tumors that are staged by EUS. 

What can be done to improve EUS accuracy in order to allow the best selection of therapy? First, the additional value of EUS-fine needle aspiration (EUS-FNA) over EUS alone for N and M staging of gastric cancer was recently emphasized in a study from Denmark [17]. Distant lymph node and liver metastases were detected by EUS-FNA in 42 % of the patients; interestingly, computed tomography of the abdomen or thorax had previously failed to show any abnormality in the majority of them. As a consequence EUS-FNA should be considered an integral part of the EUS staging procedure for gastric cancer in the near future. Second, new techniques of image enhancement in EUS [18] could theoretically lead to better discrimination between different tumor stages. In our experience gastric cancer imaging shows significant enhancement at contrast-enhanced harmonic EUS after intravenous injection of contrast agent. It remains to be determined whether this effect might result in more accurate tumor staging compared with fundamental B-mode imaging. Nevertheless, the continuous improvement of echo endoscopes and ultrasound scanners is likely to bring innovations that may prove useful in overcoming the current known limitations in EUS imaging of gastric cancer, such as with flat and ulcerated masses and large-volume lesions. Lastly, gastric cancer restaging after neoadjuvant treatment is likely to emerge as another clinical task for endosonographers. Although preliminary reports have yielded conflicting results in this respect [19], further research and technical improvements are warranted in order to select the most appropriate treatment for each patient in the continuous search for personalized medicine.

• References

• 1 Crew KD, Neugut AI. Epidemiology of gastric cancer. World J Gastroenterol 2006; 12: 354-362
• 2 Cardoso R, Coburn NG, Seevaratnam R et al. A systematic review of patient surveillance after curative gastrectomy for gastric cancer: a brief review. Gastric Cancer 2012; 15 [Epub ahead of print]
• 3 Lorenzen S, Blank S, Lordick F et al. Prediction of response and prognosis by a score including only pretherapeutic parameters in 410 neoadjuvant treated gastric cancer patients. Ann Surg Oncol 2012; 10 [Epub ahead of print]
• 4 Caletti G, Bolondi L, Labò G. Ultrasonic endoscopy – the gastrointestinal wall. Scand J Gastroenterol 1984; 102: 5-8
• 5 Bolondi L, Casanova P, Santi V et al. The sonographic appearance of the normal gastric wall: an in vitro study. Ultrasound Med Biol 1986; 12: 991-998
• 6 Kimmey MB, Martin RW, Haggitt RC et al. Histologic correlates of gastrointestinal ultrasound images. Gastroenterology 1989; 96: 433-441
• 7 Bolondi L, Casanova P, Caletti GC et al. Primary gastric lymphoma versus gastric carcinoma: endoscopic US evaluation. Radiology 1987; 165: 821-826
• 8 Dittler HJ, Siewert JR. Role of endoscopic ultrasonography in gastric carcinoma. Endoscopy 1993; 25: 162-166
• 9 Caletti G, Ferrari A, Brocchi E et al. Accuracy of endoscopic ultrasonography in the diagnosis and staging of gastric cancer and lymphoma. Surgery 1993; 113: 14-27
• 10 Fusaroli P, Kypreos D, Alma Petrini CA et al. Scientific publications in endoscopic ultrasonography: changing trends in the third millennium. J Clin Gastroenterol 2011; 45: 400-404
• 11 Fusaroli P, Kypraios D, Eloubeidi MA et al. Levels of evidence in endoscopic ultrasonography: a systematic review. Dig Dis Sci 2012; 57: 602-609
• 12 Puli SR, Batapati Krishna Reddy J, Bechtold ML et al. How good is endoscopic ultrasound for TNM staging of gastric cancers? A meta-analysis and systematic review. World J Gastroenterol 2008; 14: 4011-4019
• 13 Mocellin S, Marchet A, Nitti D. EUS for the staging of gastric cancer: a meta-analysis. Gastrointest Endosc 2011; 73: 1122-1134
• 14 National Comprehensive Cancer Network. NCCN Clinical practice guidelines in oncology (NCCN guidelines). Gastric cancer (including cancer in the proximal 5cm of the stomach). Version 2.2011. www.NCCN.org
• 15 Kutup A, Vashist YK, Groth S et al. Endoscopic ultrasound staging in gastric cancer: Does it help in management decisions in times of neoadjuvant treatment regimes?. Endoscopy 2012; 44
• 16 Meining A, Rösch T, Wolf A et al. High interobserver variability in endosonographic staging of upper gastrointestinal cancers. Z Gastroenterol 2003; 41: 391-394
• 17 Hassan H, Vilmann P, Sharma V. Impact of EUS-guided FNA on management of gastric carcinoma. Gastrointest Endosc 2010; 71: 500-504
• 18 Fusaroli P, Saftoiu A, Mancino MG et al. Techniques of image enhancement in EUS (with videos). Gastrointest Endosc 2011; 74: 645-655
• 19 Park SR, Lee JS, Kim CG et al. Endoscopic ultrasound and computed tomography in restaging and predicting prognosis after neoadjuvant chemotherapy in patients with locally advanced gastric cancer. Cancer 2008; 112: 2368-2376