24 Squamous Neoplasia of the Esophagus
Hassan A. Siddiki and David E. Fleischer
24.1.1 Epidemiology and Risk Factors
Squamous cell carcinoma (SCC) of the esophagus and its precursor lesions are referred to as squamous cell neoplasia (SCN). SCC of the esophagus is one of the two main malignancies of the esophagus, the other being adenocarcinoma. Esophageal cancer makes up 3.7% of all non–skin-related cancers worldwide. Approximately 450,000 new cases are detected each year worldwide. It is the eighth most common cancer in both genders combined, and the sixth most common cancer in men. More than 80% of cases arise in less developed regions of the world and, of those, 90% are SCC. 1 Barrett’s esophagus (BE), the main precursor to adenocarcinoma of the esophagus, is covered in Chapter 23. adenocarcinoma of the esophagus, is covered in Chapter 23. Advanced esophageal cancer, both SCC and adenocarcinoma, is covered in Chapter 27. This chapter focuses on squamous precursors of esophageal cancer and early SCC.
Various epidemiologic associations are known for SCC. In the United States, two major associations include smoking with a population-attributable risk (PAR) of 56% (confidence interval [CI] of 36–75) and alcohol with a PAR of 72.4% (95% CI 53–86). 2 In contrast, in the economically less developed countries in eastern and central Asia and eastern and southern Africa (the so-called esophageal cancer belt) where rates (per 100,000) of SCC are the highest in the world, alcohol and tobacco exposure have little or no role (▶Fig. 24.1). 3 , 4 , 5 Studies in high-risk areas suggest that the exposure to carcinogens such as polycyclic aromatic hydrocarbons, poor oral hygiene, opium, poor aromatic hydrocarbons, poor oral hygiene, opium, poor nutrition, and thermal damages are major risks, whereas tobacco exposure plays only a minor role. In certain parts of China, coal cakes are used for both cooking and heating. The inhaled smoke has a high level of polycyclic aromatic hydrocarbons which is an important risk factor for SCC in China. 6 A report from India describes poor oral hygiene as a risk for esophageal squamous cell carcinoma (ESCC) in Kashmir. 7 Drinking hot tea, a habit common in Northern Iran, has been strongly associated with a high risk of esophageal cancer. 8 A study from Tenwek Hospital in Bomet, Kenya, describes risk factors for SCC. One of the traditions is to drink mursik which is a combination of a fermented beverage with an ash-like material. 8 There is some controversy about the role of human papilloma virus (HPV). A relationship between HPV and SCN has not been shown consistently. 9 Despite these differences in etiology, both high- and low-rate SCC share dysplasia as their precursor lesion and thus approaches to screening and treatment are common to both. ▶Fig. 24.2 shows photograph of one of the authors DF) with colleagues and patients in China (24.2a) and Kenya (24.2b).
24.1.2 Precursor Lesions for Squamous Neoplasia
Some, but not all, of these associations exist for precursor lesions for SCC. Precursor lesions may be divided histologically into (1) mild dysplasia, (2) moderate dysplasia, and (3) severe dysplasia (▶Fig. 24.3). Squamous dysplasia requires the presence of nuclear atypia (enlargement, pleomorphism, and hyperchromasia), loss of cell polarity, and abnormal tissue maturation without invasion of epithelial cells through the basement membrane.
Compared with normal, in mild dysplasia, the abnormalities are confined to the lower third of the epithelium; whereas in moderate dysplasia, these are present in the lower two-thirds of the epithelium; and in severe dysplasia, these also involve the upper third of the epithelium. Full-thickness involvement of the epithelium is called carcinoma in situ. It is considered by some to be synonymous with severe dysplasia based on their similar histologic appearance and risk of progression to invasive SCC. Risk factors for SCC and dysplasia in high-risk populations are similar in some situations: gender (no), alcohol drinking (no), family history of cancer (yes), tooth loss (yes), water source (yes), and HPV (yes/no). These are different in that tobacco seems to be a risk for SCC and not dysplasia, and serum pepsinogen levels do not appear to be a risk for SCC but risk for dysplasia. 10
The most definitive assessment of risk for squamous cell esophageal histology has come from the Linxian Dysplasia Nutrition Intervention Trial (NIT). 11 Over a 3.5-year period, mild dysplasia increased the risk of developing SCC by 2.2, moderate dysplasia by 15.8, and severe dysplasia by 72.6. On a follow-up study at 13.5 years, SCC developed in 8% of the participants who initially had a normal histology, but 24% with mild dysplasia, 50% with moderate dysplasia, and 74% with severe dysplasia. 12
24.2 Diagnostic Approaches
Since most patients with squamous dysplasia of the esophagus and early esophageal cancer will be asymptomatic, the diagnoses will generally be made by screening unless the lesion is found incidentally in a patient that has unrelated symptoms. In the United States, there are no guidelines that recommend screening for SCN. In high-incidence areas (China, Iran, Japan, East Africa) screening is increasingly employed (▶Table 24.1).
24.2.1 Nonendoscopic Techniques
A common nonendoscopic screening technique for early detection of esophageal cancer is esophageal balloon cytology developed for screening in high-risk populations. To sample the esophagus blindly, an inflatable balloon covered with netting was developed in China 13 , 14 and later an encapsulated sponge was developed in Japan (▶Fig. 24.4). 15 , 16
This technique employs “fishing” for esophageal squamous cells. A nylon or silk mesh covering a deflated balloon is swallowed by the patient and once the balloon is in the stomach, it can be inflated and pulled proximally. The compliance of the balloon is controlled with a syringe that connects to the device via tubing. On withdrawing the balloon, the device will be pulled back. Prior to that, it will be partially deflated and then pulled through the mouth where the cells which are collected in the mesh undergo cytologic staining. The method suffers from low sensitivity, less than 50%. 17 However, the ease of use and affordability are appealing. This technique was widely used in China and accepted as a diagnostic method by the World Health Organization. A recent study of 15 years of follow-up showed that balloon cytology examination remained a reliable method for early detection of esophageal cancer and has accumulated a valuable cytologic bank. 18
In this technique, a polyurethane mesh is compressed inside a gelatin capsule and attached to a string. The encapsulated sponge is swallowed by the patient and when the sponge enters the stomach the contents of the stomach dissolve the gelatin covering. The mesh devoid of its covering can now expand (▶Fig. 24.4c). As the sponge is pulled up the esophagus, mucosal cells get scraped and these exfoliated cells are then collected, processed, and stained for cellular abnormalities. There are no large studies evaluating this technique for detecting dysplasia. In a smaller study using a very early version of the sponge, the sensitivity and specificity of this technique was 24 and 92%, respectively. 19 Because of blind sampling and limited morphologic evaluation of only a fraction of the cells, the accuracy of the early sponge was not precise enough for it to be recommended as a community-based screening tool. Recent studies done using the cytosponge for dysplasia related to BE have been more encouraging, demonstrating the overall sensitivity of 79% increasing to 87.2% for patients with long-segment BE, who swallowed the device and a specificity for diagnosing BE of 92.4%. 20 In a large study conducted in the United Kingdom, where the cytosponge was combined with immunohistochemical staining for trefoil factor 3 (TFF3, a biomarker for BE) the overall sensitivity was 79.9%. Compared to endoscopy, the specificity was 92.4%. Similar studies have not been conducted for SCN. In the current form, the sponge cannot replace endoscopic screening. It is being studied in parts of the world where the incidence of SCC is high. If it is shown to be sensitive and specific, it would be a “game changer” in screening for both SCC and its precursors. One aspect of this discussion relates to whether balloon and sponge techniques differ in the evaluation of patients with squamous lesions versus BE. Barrett’s epithelium is thicker than squamous epithelium, is better vascularized, and has a mucus layer, all of which would seemingly make it more resistant to ablation than squamous tissue. On the other hand, squamous epithelium (which sloughs off more easily than Barrett’s tissue after ablation) may be more difficult to ablate because it may regenerate from the epithelial linings of ducts in the submucosal glands, and since neoplasia can extend down into the ducts, this may lead to recurrence.
24.2.2 Endoscopic Techniques
High-Resolution White Light with Magnification
White light high-definition endoscopy with magnification can detect advanced lesions as irregular mucosa, white patch, focal red area, erosions, or a plaque but many early cancers and precursors may be missed without the use of chromoendoscopy. This topic is covered in more detail in Chapter 18. According to the Paris Classification of superficial neoplastic lesions, an abnormal-appearing lesion is called “superficial” when its endoscopic appearance suggests that the depth of penetration in the digestive wall is not more than into the submucosa, that is, there is no infiltration of the muscularis propria. 21 All such superficial lesions come under the type “0” and can range from completely flat (0–IIB) to elevated (0–IIa) and depressed (0–IIc) (▶Fig. 24.5).
Chromoendoscopy with Lugol’s Solution
Iodine staining was first employed by Schiller in 1932 to detect abnormalities in the uterine cervix. The basic principle is that iodine stains normal squamous cells a brownish-green color, but that pathologic tissue depleted of glycogen does not pick up the stain and produces an unstained lesion. For example, dysplastic or malignant cells, which are devoid of glycogen, will not pick up the stain (▶Fig. 24.6). Lugol’s solution contains iodine and potassium iodide. The strength of the solution can vary, but commonly includes mixing 12 g of iodine with 24 g of potassium iodide in 1 L water. The formula gives the solution a strength of 1.2% by elemental iodine content and 3% by total iodine content. Lugol’s staining improves the sensitivity of detection of squamous neoplastic lesions, dysplasia and carcinoma, from 62 to 96% without any loss of specificity. 22
Compared with cytology balloons, its sensitivity and specificity are superior, at 96 and 63%, respectively. 23 In expert centers, specificity could reach 100% after reviewing the histology. In China, a high-risk population study showed that after staining with Lugol’s solution, 23% of the lesions containing high-grade dysplasia and 55% of the lesions containing low-grade dysplasia, which were missed by routine endoscopy, could be detected by Lugol’s solution, suggesting a strong role in the early detection of dysplastic lesions. 22
Twenty milliliters of dye is sprayed from the distal to proximal via biopsy channel using a spray catheter (Olympus America, Inc., Millville, New Jersey). The decision to spray from distal to proximal or proximal to distal is a matter of personal preference. Alternatively, an endoscopic retrograde cholangiopancreatography catheter could also be used. The margins of any lesion are better demarcated after staining, making it useful for therapeutic resections when feasible. The specificity of iodine for staining dysplastic lesions is compromised by inflamed mucosa, like esophagitis, which may also appear unstained. When examining the mucosa, special care must be taken not to overlook the upper third of the esophagus. Once Lugol’s solution is applied, there needs to be a certain urgency in obtaining biopsies since the staining lasts for only a few minutes. Both biopsies of abnormal areas and systematic sampling of normal areas should be performed and care must be taken to keep these in separate jars.
These techniques improve the visualization of mucosal morphology without needing the use of dyes. Narrow-band imaging (NBI) (Olympus, Tokyo, Japan) is the approach that has been best studied. Optical chromoendoscopy can be divided into two categories: (1) preprocessing and (2) postprocessing. As is described in Chapter 18, preprocessing techniques optimize mucosal and vascular imaging by adjusting the wavelength composition of the excitation light, generally by mainly using blue light. Blue light, by its shorter wavelength, only penetrates superficially in the tissue and causes less scattering. In addition, blue light is highly absorbed by hemoglobin resulting in optimal visualization of blood vessels. This technique is utilized by both NBI and blue laser imaging (Fujifilm, Tokyo, Japan). Postprocessing techniques use normal white light excitation and reprocessing of the reflected images by an appropriate algorithm. Examples of this are Fuji intelligent chromoendoscopy (Fujifilm, Satamia, Japan) and i-Scan (Pentax, Tokyo, Japan). All of these techniques can be combined with magnifying endoscopy. The combination enables the endoscopist to visualize intrapapillary capillary loop (IPCL) patterns which are capillary loops arising perpendicularly from smooth-branching vessels in the subepithelium. In mucosal cancer, these loops dilate causing the vascular network to be less transparent than a normal mucosa (▶Fig. 24.7). In SCC, four IPCL pattern changes have been described, including dilation, weaving, change in caliber, and variety in shape. These features may be useful in estimating atypia and depth of invasion. 24 A more current classification has been described by Inoue. 25
In a randomized-controlled trial, it was shown that NBI improved the detection rates for SCC. 26 In this study, the sensitivity of NBI for diagnosis of superficial cancer was 97.2%. The accuracy of NBI for the diagnosis of superficial esophageal cancer was 88.9%; however, no large prospective studies have shown that magnifying endoscopy with NBI is superior to Lugol’s chromoendoscopy. In another more recently conducted study, changes in the morphology of IPCLs were well correlated with the depth of SCC invasion. 27
Confocal Laser Endoscopy
This is a high-resolution imaging technique that introduced a novel endoscopic surveillance method for SCN and has the ability to detect cancer at the subcellular level (see Chapter 18). Using CLE, the mucosa can be analyzed at a magnification of 1,000 times, and with a maximum penetration depth of the scanning laser light of up to 250 microns. Changes in vessels, connective tissue, and in the cellular architecture can be evaluated during ongoing endoscopy. When confocal imaging is paired with chromoendoscopy, accuracy rates have been shown to dramatically rise, approaching more than 95%, mostly attributable to improvement of the known low specificity of chromoendoscopy alone. 28 The drawback to this technique is that existing platforms are expensive (> $150,000) and are available mainly in tertiary care centers. There is also the limitation that it is not possible to image the abnormality and perform a biopsy at the same time. CLE is not a wide-view endoscopic platform like chromoendoscopy and works best for evaluating a focal lesion.