Synchronous Primary Neoplasms Detected on ¹⁸F-FDG PET in Staging of Patients with Esophageal Cancer

MATERIALS AND METHODS

RESULTS

In this series of 366 patients who underwent ¹⁸F-FDG PET during initial staging of esophageal cancer, synchronous primary lesions were identified in 20 patients (5.5%). Except for the 5 renal tumors, the described neoplasms were not detected by routine staging procedures. Details of these 20 patients are summarized in Table 1. Sixteen unexpected foci of uptake could be pathologically confirmed. Ten patients (50%) had a malignant synchronous tumor, and the remaining 10 patients had a benign or premalignant neoplasm.

Malignant Lesions

¹⁸F-FDG PET suggested a synchronous neoplasm in the ascending colon in patients 1 and 2. Colonoscopy revealed adenocarcinoma in both patients, and they underwent transthoracic esophagectomy (Ivor–Lewis method (14)) and right hemicolectomy in a single surgical session.

A synchronous primary tumor was found in the kidney in 5 patients (patients 3, 4, 5, 6, and 7). In all these patients, initial abdominal CT also revealed a mass in the renal region. Patient 3 underwent a diagnostic laparoscopy revealing peritonitis carcinomatosa, and therefore, no surgical therapy was indicated. In patient 4, cytology revealed a Grawitz’s tumor; however, the tumor of the esophagus was not eligible for curative therapy. Patient 5 underwent a curative esophageal resection followed by a curative nephrectomy 3 mo later. Patients 6 and 7 were considered eligible for esophagectomy and nephrectomy in a single surgical session. However, in patient 7 (Fig. 1), after a successful nephrectomy the tumor of the esophagus was considered unresectable because of unexpected invasion of the thoracic aorta. Pathologic examination revealed a Grawitz’s tumor in both patients.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

¹⁸F-FDG PET scan (coronal view) of patient 7, with synchronous carcinoma of mid esophagus (top arrow) and Grawitz’s tumor of left kidney (bottom arrow).

In patient 8, with a squamous cell carcinoma of the esophagus, ¹⁸F-FDG PET revealed a lesion in the right upper lobe suggestive of lung metastasis or synchronous lung carcinoma. Additional bronchoscopic examination provided cytologic evidence of a primary squamous cell carcinoma of the lung. The patient underwent right upper lobectomy and esophagectomy in a single session.

In patient 9, a supraclavicular metastasis was suspected on the basis of ¹⁸F-FDG PET findings (Fig. 2). However, neither initial CT nor sonography of the cervical region showed any lymphatic abnormalities, nor did additional MRI of the cervical region. A second sonographic examination with fine-needle aspiration demonstrated a Hürthle cell thyroid tumor, for which a thyroidectomy was performed several months after the initial esophagectomy.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

¹⁸F-FDG PET scan (coronal view) of patient 9, with tumor in mid esophagus (bottom arrow) and synchronous Hürthle cell thyroid tumor (top arrow) as confirmed by fine-needle aspiration.

¹⁸F-FDG PET showed in patient 10 a focus suggestive of a synchronous neoplasm in the oral cavity. Examination by a maxillofacial surgeon revealed a small squamous cell carcinoma of the gingiva. The gingival carcinoma in this patient was resected, but because of severe comorbidity the esophageal tumor was treated with combined intraluminal and extraluminal irradiation with curative intention.

Benign Lesions

Synchronous neoplasms were suspected in the ascending colon of patients 11 and 12 and in the descending colon of patients 13–15. Colonoscopy revealed tubulovillous adenoma in patients 13–15, which was resected endoscopically. In patients 11 and 12, no histologic examination was performed before surgery. Patient 11 underwent transthoracic esophagectomy with curative intent. However, because locoregional recurrence developed shortly after resection, follow-up of this synchronous lesion was not possible. In patient 12, diagnostic laparoscopy, which is standard in the diagnostic work-up of cardia carcinoma, revealed peritonitis carcinomatosa, and the lesion in the colon was therefore not analyzed further.

In 4 patients, an unexpected ¹⁸F-FDG accumulation was observed in the rectum and considered suggestive of a neoplasm. Sigmoidoscopy revealed tubulovillous adenoma in patient 16 (Fig. 3) and tubular adenoma in patients 17 and 18. All these adenomas were resected endoscopically. In patient 19, no histologic examination was performed. All the patients underwent esophagectomy with curative intention. Patient 19 died postoperatively; therefore, no histologic examination of the rectal lesion was performed.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

¹⁸F-FDG PET scan (sagittal view) of patient 16, with distal esophageal tumor (top arrow) and synchronous rectal adenoma (bottom arrow).

In patient 20, ¹⁸F-FDG PET showed a lesion suggestive of a neoplasm in the thyroid gland. Additional sonography of the neck showed a hypoechogenic lesion suggestive of thyroid adenoma, and the lesion did not progress during 6 mo of follow-up.

DISCUSSION

The rate of synchronous neoplasms seen on ¹⁸F-FDG PET in this unselected group of 366 patients with esophageal cancer was 5.5%. Most lesions were in the colon or rectum (11/20, or 55%). Sixteen lesions were evaluated histologically.

The rate of unexpected synchronous neoplasms in this study lies within the range of synchronous tumors for esophageal cancer as reported in the literature (4,6,7). In general, the occurrence of synchronous cancers strongly depends on the type of the initial cancer. Synchronous malignant tumors were detected in 9%–18% in patients with head and neck cancer, whereas in an inhomogeneous group of various cancer types the rate of malignant and premalignant tumors was 1.7% (15–17).

Misinterpretation of synchronous primary neoplasms may lead to incorrect upstaging of the primary tumor, as was demonstrated in patients 8 and 9. These patients were initially suspected of having distant metastases (stage IV), on the basis of the PET findings, and would have erroneously been considered ineligible for surgery if histologic conformation had not been sought. Therefore, additional investigations are mandatory to confirm the PET findings before any therapeutic decision is made. However, 4 lesions were not histologically verified in this study. In some patients, it was argued that the esophageal cancer heavily determined prognosis and that, therefore, the verification of assumed benign or premalignant lesions was not necessary. Nevertheless, the 2 lesions of the colorectum that turned out to be carcinomas are an argument in favor of verification of any positive PET finding. Another reason to verify positive PET findings is the well-known risk of false-positive results. ¹⁸F-FDG is not a tumor-specific substance, and false-positive results may occur as a result of increased glucose metabolism in benign lesions (e.g., inflammatory tissue). Therefore, positive findings on ¹⁸F-FDG PET must be confirmed by additional investigations, preferably by percutaneous or ultrasound- or CT-guided cytologic biopsy, or dedicated radiography, before patients are denied surgery with curative intent (11). Usually, physiologic colonic activity appears more tubular and diffuse than do separate colonic tumors, which appear more focal and of higher intensity. Because the renal excretion pattern of ¹⁸F-FDG is similar to that of other radiopharmaceuticals, physiologic activities in renal collecting systems are easily detectable based on the precise location of activity, intensity of distribution, shape of the calyces and pelvis, and overall pattern of both kidneys.

The detection of synchronous tumors by whole-body ¹⁸F-FDG PET poses a dilemma in the choice of the most suitable therapeutic strategy. For synchronous cancers, including esophageal cancer, the highest-priority treatment should focus on the tumor most limiting the prognosis (6). Therefore, optimal pretreatment staging of both tumors to assess their prognosis is the first step in clinical assessment. If discrimination between 2 independent primary tumors versus metastatic disease is not possible based on conventional histology, cytogenetic analysis such as determination of loss of heterozygosity and p53 aberrations may be helpful to therapeutic decision making (18).

Resection of both neoplasms with curative intention frequently offers the best long-term survival; however, even in the case of an incurable synchronous cancer (e.g., metastatic prostate cancer), esophagectomy is not always contraindicated (19). The type of treatment for such esophageal carcinomas strongly depends on the type and prognosis of the synchronous malignancy.

Evidence-based arguments about whether to perform a simultaneous or a staged operation are not available. Suzuki et al. report that simultaneous resection of both neoplasms has acceptable morbidity and mortality rates (18). However, for each patient, the risks and benefits of simultaneous surgery should be weighed against those of a second operation (20).

The incidental detection of synchronous colorectal polyps or cancers and other malignancies by ¹⁸F-FDG PET is not uncommon (17,21,22). Unfortunately, ¹⁸F-FDG PET is not able to differentiate between colorectal adenoma and carcinoma (23). A true association between adenocarcinoma of the esophagus and colonic neoplasms would suggest common causes and might indicate the existence of an inherited general genetic defect. Another possible explanation might be exposure to environmental factors such as alcohol, smoking, and a fatty diet. In addition, increased expression of the cyclooxygenase 2 enzyme is central to the predisposition of both esophageal and colorectal cancers (24,25). However, a population-based cohort study in Sweden did not demonstrate an association between colorectal cancer and adenocarcinoma of the esophagus (26).

CONCLUSION

¹⁸F-FDG PET may detect synchronous primary neoplasms in patients with esophageal cancer. Sites of suspected metastases should be confirmed histologically before treatment, because synchronous neoplasms can mimic metastatic disease.

ACKNOWLEDGMENT

This study was supported by a ZonMw program for Health Care Efficiency Research.