The utility of PET/CT in monitoring treatment response and detecting unusual metastases in esophageal cancer

The examination of an uncommon presentation of a common disease.

A 57-year-old man was referred to the oncology clinic with a 1.5-year history of cough without hemoptysis and progressive dysphagia, resulting in a 65-lb weight loss, approximately one-third of his baseline weight.

Imaging studies performed before consultation included a barium swallow that revealed a long irregular stricture in the distal esophagus involving the cardioesophageal junction. A CT scan of the chest and abdomen also revealed a 9 cm × 7 cm × 4 cm mass involving the lower esophagus extending to the gastroesophageal junction projecting into the stomach, but without evidence of mediastinal lymphadenopathy or metastatic disease.

Munir Ghesani

An esophagogastroduodenoscopy (EGD) was performed and pathology revealed a moderately differentiated adenocarcinoma with Barrett’s epithelium with high-grade dysplasia. His physical examination was notable for a non-tender 6 cm × 6 cm firm mass just distal to the right elbow on the dorsal aspect of the forearm, which the patient noticed to have increased in size over the last few months.

An MRI of the right arm confirmed the presence of a 4.4 cm × 5.0 cm × 6.2 cm heterogeneous mass in the anterior and lateral soft tissues of the proximal forearm interposed between the brachioradialis and lateral extensor musculature, surrounding the proximal radius.

A staging PET/CT was performed that demonstrated a large circumferential mass involving the distal esophagus and the gastroesophageal junction spanning more than 8 cm in length, with a maximum SUV of up to 18, as well as a 1.5 cm soft tissue mass protruding into the lumen of the midthoracic esophagus, with a maximum SUV 6.8.

Unfortunately, he was found to have widely metastatic disease, including:

• Confluent right axillary lymphadenopathy measuring 5.4 cm × 2.8 cm × 4.8 cm with a maximum SUV 20.4.
• A 2.4 cm × 2.1 cm aortopulmonary window lymph node, with a maximum SUV 12.8.
• Confluent lymphadenopathy in the gastroesophageal junction, gastrohepatic ligament, gastropancreatic and celiac regions, with a representative nodal mass measuring 3.3 cm × 1.9 cm, with a maximum SUV 16.8.
• Retroperitoneal lymphadenopathy with individual lymph nodes measuring as large as 3.3 cm in diameter with an associated maximum SUV of up to 16.6.

The area corresponding to the anterior and lateral soft tissues of the right proximal forearm demonstrated a 6-cm mass with a maximum SUV of 13.9. Also notable were a tumor implant in the left paracolic gutter with a maximum SUV of 6.7, and two tumor implants anterior to and to the left of the rectum with maximum SUVs of 14.5 and 16.5, respectively.

Figure 1. Each figure has four rows of images. The top two are from the initial exam; the bottom two from the follow-up. In each examination, the upper left image is axial CT, upper right image is corresponding PET, lower left image is fusion image containing PET images displayed on a color scale and CT images displayed on a gray scale. Lower right image is maximum intensity project (MIP) image of whole body PET study. Initial exam demonstrates hypermetabolic activity corresponding to the mural thickening of the esophagus, corresponding to the primary esophageal carcinoma. Follow-up reveals considerable decrease in mural thickening with concordant degree and extent of associated metabolic activity. Initial MIP image demonstrates extensive metastatic disease with the follow-up MIP image demonstrating remarkable positive response to therapy. Note improvement in soft tissue metastases to the right upper extremity. Photos courtesy of M Ghesani, MD

To confirm the suspicion of metastatic disease, a fine needle aspiration biopsy of the right axillary lymph node was performed and pathology was consistent with a poorly differentiated carcinoma.

The patient subsequently underwent palliative chemotherapy with a combination of epirubicin, oxaliplatin and capecitabine (EOX regimen) every three weeks.

The patient tolerated three treatment cycles well, complicated only by mild anorexia, nausea and vomiting. Surveillance PET/CT done after the third cycle showed an excellent response to systemic chemotherapy.

The circumferential mass involving the distal esophagus and the gastroesophageal junction significantly decreased in size and metabolic activity from the maximum SUV of 8 to 5.7. Likewise, the previously identified 1.5-cm hypermetabolic soft tissue mass in the mid-thoracic esophagus with a maximum SUV of 6.8 was no longer appreciated on the restaging scans.

The previously identified confluent right axillary lymphadenopathy measuring 5.4 cm × 2.8 cm × 4.8 cm decreased in size to 4 cm × 1.7 cm × 3.6 cm, with an associated decrease from a maximum SUV of 20.4 to 2.8.

The 2.4 cm × 2.1 cm aortopulmonary window lymph node with a maximum SUV of 12.8 decreased in size to 1 cm × 0.9 cm and was no longer hypermetabolic.

The confluent adenopathy at the gastroesophageal junction, gastrohepatic ligament, gastropancreatic and celiac regions decreased in size and metabolic activity from a maximum SUV of 16.8 to 4.8.

The retroperitoneal lymphadenopathy also significantly decreased in size and metabolic activity with a maximum SUV of not more than 3 as compared with 16.6 on the prior examination.

The bulky mass that was observed in the right forearm significantly decreased in size and metabolic activity from a maximum SUV of 13.9 to 2.1. The tumor implants noted on previous imaging also had a marked reduction in size and associated hypermetabolic activity.

Discussion

Esophageal cancer ranks among the 10 most common cancers worldwide and is the fourth most common gastrointestinal cancer after colon, gastric and liver cancers. Cancers of the esophagus are histologically classified as squamous cell carcinoma or adenocarcinoma.

Squamous cell tumors are more likely to occur in black patients and are associated with achalasia, tylosis and Plummer-Vinson syndrome, while adenocarcinomas are diagnosed predominantly in white men, are associated with gastroesophageal reflux disease and typically arise in Barrett’s esophagus. Although both are linked to smoking and alcohol abuse, the association is much stronger in the former.

Staging of esophageal cancer is based on the tumor, node, metastasis (TNM) classification developed by the American Joint Committee on Cancer. Essential to accurate staging are an EGD, CT scan of the chest, abdomen and pelvis, and biopsy confirmation of suspected metastatic disease. Bronchoscopy, endoscopic ultrasound and PET/CT are indicated in staging if there is no evidence of M1 disease.

Figure 2. Comparison of initial with the follow-up exam demonstrate remarkable positive treatment response to the right axillary and aortopulmonary window lymphadenopathy. Photo by M Ghesani

Surgery is the gold standard for resectable disease. However, the results of surgical resection alone have been discouraging, bringing forth a series of clinical trials to determine the efficacy of chemotherapy and radiation. Preoperative chemoradiation followed by surgery is the most common approach for patients with resectable cancer.

For advanced unresectable disease, chemoradiation may be appropriate and can occasionally facilitate surgical resection in select cases. Definitive chemoradiation is an appropriate option for patients with resectable disease who are “unfit” for surgery.

The most common sites for distant metastases are to nonregional lymph nodes, liver, lungs, bones and adrenal glands. As these patients do not benefit from surgical resection, detection of metastases is important in patient management. PET/CT has been reported to be useful in the initial staging of patients and may be more accurate than conventional imaging studies in the detection of metastatic disease.

In fact, PET/CT has been reported to detect distant nodal and organ metastases in 20% of patients who were initially considered to have resectable disease based on conventional staging. In a series by Bruzzi et al, imaging with PET/CT detected sites of metastatic disease in uncommon anatomical locations or had an unusual appearance. These uncommon sites of metastases included skeletal muscle, brain, peritoneum, subcutaneous soft tissues, pleura, pancreas and thyroid gland.

To date, there are only a few published reports of intramuscular metastases from esophageal cancer. All of the intramuscular metastases in Bruzzi’s series were located within skeletal muscle, were asymptomatic and caused little or no anatomic abnormality on CT images.

However, the metastases manifested as intensely avid foci of FDG uptake on the corresponding PET/CT images. Subcutaneous metastases were also demonstrated in their series, manifesting as small, painless hypermetabolic nodules on PET/CT.

In terms of the utility of PET/CT in monitoring of treatment response, Brucher et al demonstrated that the sensitivity for detecting a response was 100%, with a specificity of 55%, while the positive and negative predictive values were 72% and 100%, respectively.

The perfect negative predictive value of PET indicates that serial PET scanning is particularly well suited for the identification of nonresponding patients. An SUV reduction of 35% or greater predicts an ultimate histopathological response to therapy with a sensitivity of 93% and a specificity of 95%.

Figure 3. Comparison of initial with the follow-up exam demonstrate remarkable positive treatment response to the upper abdominal lymphadenopathy. Photo by M Ghesani

The intent of treatment for metastatic esophageal cancer is palliation. Many agents, including 5-FU and cisplatin, have demonstrated activity in esophageal cancer, but there is no true consensus as to the best regimen in the first-line setting. The combination of epirubicin, oxaliplatin and capecitabine is widely used based on a trial that concluded that capecitabine is noninferior to 5-FU, that oxaliplatin is noninferior to cisplatin and that overall survival was longer with EOX than the combination of epirubicin, cisplatin and 5-FU (ECF regimen).

Eric Gamboa, MD, is an Oncology Fellow at St Luke’s-Roosevelt Hospital Center.

Seth Cohen, MD, is an Attending Oncologist at St Luke’s-Roosevelt Hospital.

Munir Ghesani, MD, is Associate Clinical Professor of Radiology at Columbia University College of Physicians and Surgeons and Attending Radiologist at St.Luke’s-Roosevelt Medical Center.

For more information:

• Brucher BLDM. Ann Surg. 2001;233:300-309.
• Bruzzi JF. Curr Probl Diagn Radiol. 2007;36:21-29.
• Cunningham D. N Engl J Med. 2008;358:36-46.
• Devesa SS. Cancer. 1998; 83: 2049-2053.
• Heffernan E. Clin Nucl Med. 2006;31:810-811.
• Salavati A. Nucl Med Commun. 2009;30;95-116.