Endoscopic ultrasonography for pancreatic cystic lesions: let’s enhance it

Pancreatic cysts[*] are associated with the risk of malignancy; however, recent data suggest that this risk has been overstated. As estimated in a model that takes into account the prevalence of pancreatic cyst in the US adult population and the annual number of cancers attributable to cystic precursors, the risk that an incidental cyst is malignant at the time of diagnosis is only 0.01 %. Moreover, the risk remains low (0.25 %) even under the extreme assumption that all pancreatic malignancies develop from cystic precursors [1]. This low baseline risk is modified by several factors, such as cyst size, presence of mural nodules or solid component, concomitant dilatation of the pancreatic duct, and possibly other, less well-established, predictors of malignancy [1]. Current guidelines recommend resection of cysts with risk factors and surveillance of those without [1] [2] [3]. Unfortunately, the strategies to stratify the risk have a low positive predictive value (PPV) and only 25 % of individuals who meet the criteria for resection have high grade dysplasia (HGD) or invasive carcinoma in the resected specimen (collectively referred to here as “malignancy”) [4]. The remaining 75 % undergo major surgery associated with significant mortality (2 % at high volume centers and 6 % overall), only to discover that their cyst had low malignant potential [1]. Some may find this risk acceptable because resection of a potentially malignant cyst prevents its future malignant transformation; however, the overall risk of developing carcinoma during surveillance is relatively low, between 0.24 % and 0.7 % per year, and most likely lower for cysts without HGD [1].

What can be done to reduce the number of overtreated patients? Two papers in this issue of Endoscopy focus in this context on better characterization of mural nodules using contrast harmonic endoscopic ultrasonography (CH-EUS) [5] [6]. Mural nodules occur in intraductal mucinous papillary neoplasms (IPMNs) and mucinous cystic neoplasms (MCNs) and are a well-established risk factor for malignancy and accepted indication for surgery. In a recent meta-analysis of imaging features to distinguish malignant and benign branch-duct (BD)-IPMNs the presence of mural nodules showed the highest odds ratio (6.0, 95 %CI 4.1 – 8.8) among all risk factors identified [7]. Mural nodules are considered by the 2012 international consensus guidelines to be a worrisome feature that prompts EUS evaluation and cyst resection when EUS confirms the presence of the nodule [2]. The American Gastroenterological Association guideline lists a solid component as one of the three risk factors that should be taken into account in deciding on cyst resection [1].

Despite the important role of mural nodules in the management algorithm, the term mural nodule is not well defined, overlaps with solid component, and may denote both tiny cyst wall protrusions as small as 1 – 2 mm and a mass of several centimeters in size [7] [8] [9] [10] [11]. Because the risk of malignancy increases with nodule size, this heterogeneity makes it difficult to interpret and compare results of research [9] [12]. In addition, the interobserver agreement on the presence of mural nodule is disappointing for EUS, computed tomography (CT), and magnetic resonance imaging (MRI), and these methods differ markedly in their diagnostic performance characteristics [10] [13] [14]. CT and MRI detect nodules with low sensitivity and high specificity [9] [15]. EUS is sensitive, can detect small nodules, but is less specific, because it may be difficult to distinguish on EUS a true mural nodule from mucus or debris frequently present inside the cyst [9].

A potential solution to this problem is contrast harmonic imaging [5] [6]. This method uses harmonic signals emitted by intravenous contrast agent – encapsulated microbubbles of gas that oscillate in response to the ultrasound field – to show tissue perfusion in real time and at high resolution [16] [17]. Ultrasound contrast agents have an excellent safety profile and their price is not prohibitive. Contrast harmonic imaging has been successfully used for characterization of focal lesions of the liver and other organs on percutaneous ultrasonography and for EUS characterization of solid pancreatic masses [16] [17]. Because of its capability for depicting tissue blood flow even within tiny structures, CH-EUS seems to be well suited to evaluation of cyst walls including mural nodules; however, the experience in this setting has been limited [15].

In this issue, Kamata et al. report results of CH-EUS in 70 patients with surgically verified cysts, including non-neoplastic cysts, serous cystic neoplasms, and benign and malignant MCNs and BD-IPMNs [5]. Standard EUS often diagnosed mural nodules in non-neoplastic or benign cysts, which led to misclassification of these cysts as malignant, resulting in low specificity (40 %), low PPV (55 %), and low accuracy (64 %) for diagnosis of malignancy. However, when only contrast-enhancing nodules were considered to be indicative of malignancy, the false-positive rate dropped by more than half, specificity, PPV, and accuracy rose to 75 %, 74 %, and 84 %, respectively, and sensitivity and negative predictive value (NPV) remained at about 95 %. The receiver operating characteristic (ROC) curve analysis found that the optimal threshold to discriminate between benign and malignant cysts was enhancing nodule height of ≥ 4 mm (area under the ROC curve of 0.93).

Taking a step further, the tissue blood flow in the nodule can be quantified. In another study in this issue, Yamamoto et al. analyzed time-intensity curves (TIC) – enhancement signal intensity values plotted against time – from mural nodules in 30 patients with IPMN, half of which were malignant [6]. TIC parameters differed significantly between nodules with or without malignancy and correlated well with tumor microvessel density on pathologic examination. Among various TIC parameters evaluated, the nodule-to-pancreatic parenchyma contrast ratio was the most accurate one, with sensitivity, specificity, NPV, PPV, and accuracy for detection of malignancy in the 93 % – 94 % range and area under the ROC curve of 0.89.

These promising results should be interpreted with caution. Both studies were retrospective, relatively small, and prone to significant selection and verification bias. Therefore, prospective data on diagnostic accuracy and interobserver agreement are needed before CH-EUS can be recommended for routine clinical use. Also the potential place of this method in the diagnostic algorithm has to be defined, but most likely the focus will be on patients with risk factors rather than on those with an unsuspicious cyst. Cysts without risk factors are not an indication for EUS according to current guidelines and CH-EUS is not likely to change this approach.

A scenario in which CH-EUS is most likely to be used is when a mural nodule is found on EUS in a suspected BD-IMPM or MCN. In this setting, the advantages of CH-EUS seem promising, even though the underlying evidence is still preliminary. A nonenhancing nodule most likely represents mucus or debris and hence is not a risk factor for malignancy; consequently the cyst can probably be treated as a cyst without mural nodules. In contrast to that, an enhancing nodule is composed of tissue and deserves attention, because it may be malignant itself or indicate malignancy elsewhere in the cyst. Enhancing nodule height and/or quantitative enhancement parameters are potential markers of malignant cyst which, if validated in prospective studies, may become a useful tool to select patients for surgery or surveillance. However, a word of caution is important here. As shown in a study that used contrast-enhanced EUS – a method that is similar to CH-EUS but uses the Doppler technique to depict blood flow – the malignant transformation rate for BD-IPMN with small (< 5 mm) enhancing nodules was 6.9 % and 10.7 % at 3 and 5 years, respectively [8].

In patients with main-duct IPMN the potential advantages of CH-EUS are less clear. Main-duct IPMNs are considered to be an indication for surgery regardless of the presence of nodules; hence, the differentiation between true nodules and mucus is not likely to influence management decisions. However, as the target for resection is cancer and/or HGD, a method to reliably diagnose those conditions might be useful, especially in poor surgical candidates. Most of the patients evaluated in the study by Yamamoto et al. had main-duct or mixed-type IPMN, so this study suggests that CH-EUS with TIC analysis is a potentially useful option in this setting [6].

In addition to the potential clinical implications of CH-EUS outlined above, a robust, CH-EUS-based definition of mural nodule may facilitate future research on the management of pancreatic cysts.

• References

• 1 Scheiman JM, Hwang JH, Moayyedi P. American Gastroenterological Association technical review on the diagnosis and management of asymptomatic neoplastic pancreatic cysts. Gastroenterology 2015; 148: 824-848 e22
• 2 Tanaka M, Fernandez-del Castillo C, Adsay V et al. International consensus guidelines 2012 for the management of IPMN and MCN of the pancreas. Pancreatology 2012; 12: 183-197
• 3 Del Chiaro M, Verbeke C, Salvia R et al. European experts consensus statement on cystic tumours of the pancreas. Dig Liver Dis 2013; 45: 703-711
• 4 Goh BK, Lin Z, Tan DM et al. Evaluation of the Fukuoka Consensus Guidelines for intraductal papillary mucinous neoplasms of the pancreas: Results from a systematic review of 1,382 surgically resected patients. Surgery 2015; 158: 1192-1202
• 5 Kamata K, Kitano M, Omoto S et al. Contrast-enhanced harmonic endoscopic ultrasonography for differential diagnosis of pancreatic cyst. Endoscopy 2016; 48: 35-41
• 6 Yamamoto N, Kato H, Tomoda T et al. Contrast-enhanced harmonic endoscopic ultrasonography with time-intensity curve analysis for intraductal papillary mucinous neoplasm of the pancreas. Endoscopy 2016; 48: 26-34
• 7 Kim KW, Park SH, Pyo J et al. Imaging features to distinguish malignant and benign branch-duct type intraductal papillary mucinous neoplasms of the pancreas: a meta-analysis. Ann Surg 2014; 259: 72-81
• 8 Ohno E, Itoh A, Kawashima H et al. Malignant transformation of branch duct-type intraductal papillary mucinous neoplasms of the pancreas based on contrast-enhanced endoscopic ultrasonography morphological changes: focus on malignant transformation of intraductal papillary mucinous neoplasm itself. Pancreas 2012; 41: 855-862
• 9 Zhong N, Zhang L, Takahashi N et al. Histologic and imaging features of mural nodules in mucinous pancreatic cysts. Clin Gastroenterol Hepatol 2012; 10: 192-198, 198 e1-2
• 10 Do RK, Katz SS, Gollub MJ et al. Interobserver agreement for detection of malignant features of intraductal papillary mucinous neoplasms of the pancreas on MDCT. AJR Am J Roentgenol 2014; 203: 973-979
• 11 Anand N, Sampath K, Wu BU. Cyst features and risk of malignancy in intraductal papillary mucinous neoplasms of the pancreas: a meta-analysis. Clin Gastroenterol Hepatol 2013; 11: 913-921
• 12 Ohno E, Hirooka Y, Itoh A et al. Intraductal papillary mucinous neoplasms of the pancreas: differentiation of malignant and benign tumors by endoscopic ultrasound findings of mural nodules. Ann Surg 2009; 249: 628-634
• 13 de Jong K, Nio CY, Mearadji B et al. Disappointing interobserver agreement among radiologists for a classifying diagnosis of pancreatic cysts using magnetic resonance imaging. Pancreas 2012; 41: 278-282
• 14 de Jong K, Verlaan T, Dijkgraaf MG et al. Interobserver agreement for endosonography in the diagnosis of pancreatic cysts. Endoscopy 2011; 43: 579-584
• 15 Yamashita Y, Ueda K, Itonaga M et al. Usefulness of contrast-enhanced endoscopic sonography for discriminating mural nodules from mucous clots in intraductal papillary mucinous neoplasms: a single-center prospective study. J Ultrasound Med 2013; 32: 61-68
• 16 Kitano M, Kudo M, Yamao K et al. Characterization of small solid tumors in the pancreas: the value of contrast-enhanced harmonic endoscopic ultrasonography. Am J Gastroenterol 2012; 107: 303-310
• 17 Wilson SR, Burns PN. Microbubble-enhanced US in body imaging: what role?. Radiology 2010; 257: 24-39