Endoscopic Ultrasonography-Guided Ablation Therapy and Celiac Plexus Interventions

Key Points

•

Endoscopic ultrasonography (EUS)-guided ablative therapy consists of the injection of cytotoxic agents into cystic cavities or ganglia to eliminate premalignant epithelium or to produce neurolysis.
•

Celiac plexus block or neurolysis is the most common EUS-guided intervention in current practice. Significant pain control is achieved with the injection of ethanol in the setting of pancreatic cancer. More modest results are seen in patients with abdominal pain arising from chronic pancreatitis.
•

More advanced techniques include the use of radiofrequency ablation and brachytherapy in selected cases unfit for surgery or for palliative control of locally advanced cancers. Although preliminary data are promising, most of these procedures are still under clinical investigation.
•

Although many of these EUS-based techniques are designed to be used to ablate or control pancreatic malignancies, some may facilitate the delivery of radiation therapy by placement of radiopaque markers into the tumor.

Endoscopic ultrasonography (EUS) today is not only an essential diagnostic tool for the diagnosis of gastrointestinal diseases but has also become a significant part of our therapeutic armamentarium. Using fine-needle aspiration (FNA) accessories, interventional EUS is often based on fine-needle injection (FNI) therapy. Developments in interventional EUS have also highlighted a broad range of therapies beyond FNI, including tissue ablation and cancer therapeutics. In this chapter, the current clinical and experimental applications of interventional EUS for ablation therapy and celiac plexus neurolysis are reviewed and described with technical details.

Endoscopic Ultrasound-Guided Radiotherapy ( Table 25.1 )

Radiofrequency Ablation

The principle of radiofrequency ablation (RFA) is the induction of thermal injury to the target tissue through the use of electromagnetic energy. In monopolar RFA, the patient is part of a closed-loop circuit that includes a radiofrequency (RF) generator, an electrode needle, and a large dispersive electrode (ground pad). Cells experiencing the thermal damage undergo coagulative necrosis over the course of several days.

TABLE 25.1

Endoscopic Ultrasonography-Guided Radiotherapy for Tumor Ablation

	Radiofrequency Ablation	Cryotherm (Cool-Tipped RFA)	Brachytherapy	Fiducial Placements
Device used	Needle prongs	Dedicated catheter	FNA needle	19-G or 22-G EUS FNA needle
Mechanism of action	Heat-induced necrosis	Heat-induced necrosis	DNA damage	Tattoo the tissue for RT
Target lesion	Pancreatic cancer, cystic lesions, neuroendocrine tumors, celiac ganglion	Pancreatic cancer	Pancreatic cancer	Any GI tumors that can be accessed by EUS
Human studies	Yes	Yes	Yes	Yes
Availability	Clinical trials	Clinical trials	Clinical trials	Wide

EUS, Endoscopic ultrasonography; FNA, fine-needle aspiration; NET, neuroendoscrine tumor; RFA, radiofrequency ablation; RT, radiotherapy.

The procedural technique used for RFA is based on the EUS guidance of a needle catheter into the target lesion. In RFA of liver and pancreatic lesions, this procedure requires placement of the needle across the gastric or duodenal walls. Once the needle has been successfully placed into the tissue mass, the RF current is delivered. During heating of tissue, ultrasound monitoring demonstrates a hyperechoic “cloud” surrounding the tip of the needle. EUS-guided delivery of ablative energy to localized malignant tumors has become increasingly possible through the introduction of commercial devices. In recent clinical studies, the Habib EUS-RFA probe (Habib catheter, Emcision Ltd., London) was used to apply EUS-guided RFA to the pancreas. The Habib catheter is a monopolar RFA probe with a working length of 190 cm, 3.6-Fr (1.2-mm) diameter, and 1-Fr (0.33-mm) wire, compatible with 19- and 22-G FNA needles ( Figs. 25.1 and 25.2 ). It is designed to achieve more limited injury to tissue as compared with other RFA devices.

Procedural Technique ( Video 25.1 )

The echoendoscope is inserted through the esophagus to the stomach and duodenum. After the pancreatic lesion has been located, a 19- or 22-G FNA needle is inserted through the working channel of the echoendoscope into the target lesion. The stylet is then removed from the FNA needle and the monopolar Habib catheter gently advanced through the lumen of the FNA needle. The RFA probe is connected to an electrosurgical RF generator, for which the wattage and exposure time have not yet been standardized. However, in pilot studies, RF energy with the Habib catheter was applied for 90 to 120 seconds at the 5- to 25-W setting. The ablation was repeated two to six times in each session in previous clinical studies.

Video 25.1

Animation Demonstrating the Technique of Endoscopic Ultrasonography-Guided Radiofrequency Ablation in Pancreatic Cancer Using the Habib Catheter

Clinical Outcomes

EUS RFA of pancreatic cystic neoplasms (PCNs), neuroendocrine tumors (NETs), and pancreatic ductal adenocarcinomas (PDACs) was first described in humans using the Habib EUS RFA catheter in two different studies. The PDAC study included seven patients with the lesions located in the head of pancreas in five and in the body of pancreas in two. RF was applied at 5 to 15 W over 90 seconds and the procedure was completed in all patients. The postprocedure imaging after 3 to 6 months showed a decrease in the size of the lesion in two patients, whereas the lesions were unchanged in the remaining patients. The procedure was well tolerated by all patients and no adverse event was encountered except for mild pancreatitis in one. In a study that included eight patients with neoplastic lesion (six with PCNs and two with NETs) in the head of pancreas who were treated by EUS-RFA, postprocedure imaging after 3 to 6 months revealed complete resolution of the cysts in two patients and a 48.4% reduction in size in three; only one patient had to undergo a second treatment session. Cross-sectional imaging in the two patients with NETs demonstrated a change in vascularity and central necrosis after EUS-RFA. No episodes of pancreatitis, perforation, or bleeding were reported, suggesting that the procedure is technically easy and safe.

In a recent study, an 18-G endoscopic RFA electrode was used for ablation of unresectable pancreatic cancer in six patients (four with lesions in the head and two with lesions in the body). RF was applied at 20 to 50 W ablation power for 10 seconds and was repeated to sufficiently cover the tumor mass. The procedure was performed successfully in all patients and only two experienced mild abdominal pain. In another study, a prototype 19-G internally cooled needle electrode was used for ablation in three patients with symptomatic pancreatic insulinoma at a power of 50 W. All patients had rapid symptom relief with biochemical improvement and remained symptom-free at 11 to 12 months’ follow-up. No procedure-related adverse event was reported.

Jin et al. administered EUS-RFA to the celiac ganglion for pain control in a patient with pancreatic cancer. In this procedure, the celiac ganglion was punctured with a 19-G needle ( Video 25.2 ). Then an RF probe, the Habib RF DUO 13, was advanced via the lumen of the needle to the center of the celiac ganglion; thereafter the needle was partially withdrawn to disengage contact with the active part of the probe. Ablation parameters were set at 10 W for 120 seconds and 15 W for 120 seconds. With the application of RFA, the center of the celiac ganglion gradually became hyperechoic. After the procedure, the patient’s visual analog scale (VAS) pain score decreased significantly and no opioid analgesics were needed.

Video 25.2

Endoscopic Ultrasonography-Guided Radiofrequency Ablation of the Celiac Ganglia Using the Habib Catheter

A commercial cool-tipped cryotherm device was designed and tested for pancreatic ablation ( Fig. 25.3 ). A flexible bipolar ablation probe combining RF and cryotechnology was used to induce foci of complete pancreatic ablation. The heated tip of the probe was cooled with simultaneous cryogenic carbon dioxide (650 psi). In the first human clinical trial, the flexible bipolar ablation probe was successfully applied under EUS guidance in 16 of 22 (72.8%) patients with advanced pancreatic carcinoma. Technical failure in six patients was due to excessive resistance to probe passage via the gastrointestinal wall and tumor. The median postablation survival time was 6 months.

These studies demonstrate that EUS-RFA is technically feasible and may be beneficial for selected pancreatic premalignant and malignant lesions. Although no major adverse events were observed, more prospective studies are needed to demonstrate the safety and overall survival benefit before widespread use can be recommended in clinical practice.

Brachytherapy

Brachytherapy in the form of small seeds can be used for the local control of malignant disease. Solid gastrointestinal malignant tumors often respond to the local administration of radiation therapy, and the risk of recurrence is reduced. Traditionally radiation therapy was provided intraoperatively, but precise targeting is difficult. Computed tomography (CT)-guided placement of radiation seeds adjacent to malignant gastrointestinal tumors is reportedly safe and somewhat effective. EUS-guided brachytherapy was first attempted in a pilot study of 15 patients with unresectable stage III and stage IV pancreatic adenocarcinoma. Through an 18-G EUS needle, multiple small radioactive seeds were placed into the pancreatic tissue to provide interstitial brachytherapy. Although the tumor response to brachytherapy was modest (33% of the tumors were stabilized), there was a transient clinical benefit in patients (30%), who experienced a reduction in abdominal pain. Similar results were also confirmed with iodine-125 seeds in 22 patients with unresectable pancreatic carcinoma. In another study, the long-term outcome of EUS-guided brachytherapy was prospectively evaluated in 100 cases of unresectable pancreatic cancer. Gemcitabine chemotherapy was combined with RFA in 85 patients 1 week after brachytherapy. The mean follow-up time was 7.8 ± 6.1 months. The estimated median disease progression-free survival and overall survival were 4.5 months and 7.0 months, respectively. VAS scores dropped significantly 1 week postimplantation and were maintained at significantly lower levels until the third month. Patients who underwent postimplantation chemotherapy had a longer median survival of 7.8 months versus 4 months for patients who did not receive chemotherapy. The outcome of the study suggested that EUS-guided iodine-125 seed implantation plus chemotherapy is an effective technique to prolong patient survival in pancreatic cancer. The effectiveness and safety of EUS-guided ¹²⁵ I seed brachytherapy was investigated in malignant left-sided liver tumors that were difficult to access by image-guided interventions. After localization of the tumor using a linear EUS scope, a transgastric puncture was performed with a 19-G injection needle and iodine seeds were placed directly into the lesion. Complete treatment response was achieved in 12 of 13 patients in 6 months; two patients needed a retreatment due to incomplete response. In the same study, anhydrous ethanol injections were undertaken in 10 patients with malignant left-sided liver tumors, and a complete response was achieved in 3 patients. EUS-guided ¹²⁵ I seed brachytherapy was found to be safe and highly effective for malignant liver tumors and superior to EUS-guided ethanol injection. For left-sided liver tumors, especially when located in proximity to the lesser curvature of the stomach, transgastric EUS can exclude interference by intestinal gas and provide safe access to the liver for any EUS-guided intervention.

Endoscopic Ultrasonography-Guided Fiducial Placement

Advances in radiation therapy have provided the opportunity for the real-time delivery of radiation using three-dimensional mapping guided by radiopaque markers. Respiration-dependent movement of the target lesions often results in inappropriate radiation exposure to surrounding tissue. Marking of focal malignancy allows the precise targeting by focused beams of radiation despite respiratory movements.

Although CT scanning is capable of guiding the placement of fiducials in and adjacent to pancreatic tumors, EUS guidance is likely more precise. These small radiopaque markers are placed into the periphery of a malignant lesion to facilitate better targeting by radiation therapy.

Procedural Technique

After identifying the tumor and excluding the presence of intervening vasculature, EUS-guided fiducial placement is undertaken using 19-G or 22-G FNA needles ( Videos 25.3 and 25.4 ). Commercially available sterilized gold fiducial markers are preloaded into the needle by retracting the stylet and manually backloading the fiducials into the tip of the needle. The tip of the needle is then sealed with bone wax to prevent accidental dislodgment of the fiducials. After identifying a target lesion, the tumor is punctured and the fiducial is deployed by advancing the stylet or guidewire forward. Resistance can be encountered during the deployment of fiducials if the tip of the echoendoscope is deflected. This resistance can be overcome by removing the stylet and applying hydrostatic pressure from a syringe containing sterile water attached to the needle to deposit the markers into the tumor. Depending on the size of the tumor, three to six fiducials should be deployed into the tumor to provide for ample separation of fiducials in distance, angulation, and plane. Both fluoroscopic and ultrasonographic visualization may be used to enable correct positioning of the fiducials within the tumor mass ( Figs. 25.4 and 25.5 ). Dedicated preloaded fiducial devices are now commercially available (see Video 25.4 ).

Video 25.3

Technique of Endoscopic Ultrasonography-Guided Placement of Fiducials

Video 25.4

Technique of Endoscopic Ultrasonography-Guided Placement of Fiducials Using a Dedicated Shark-Core Platform

The safety and effectiveness of EUS-guided fiducial placement in pancreatic malignancies have been shown in multiple studies, including large series, with high technical and clinical success rates (85% to 90%) and only a few minor adverse events. Recent studies have shown that fiducials can potentially be deployed into any malignant tumor that can be accessed by EUS. Also, the technique has been adopted to facilitate intraoperative localization of small neuroendocrine tumors in patients undergoing enucleation or other resection procedures ( Fig. 25.6 ). EUS-guided tattooing of pancreatic tumors with a marking solution before surgery has been attempted in six patients in a pilot study. The tattoo mark was easily detected during surgery and localized in a small area in five patients with no adverse events.

Celiac Ganglion Irradiation

EUS-guided direct celiac ganglion irradiation with iodine-125 seeds was applied in 23 patients with unresectable pancreatic carcinoma for the palliation of pain in a recent study. The mean number of seeds implanted in the celiac ganglion per patient was four. Although there was no difference in pain relief and analgesic consumption immediately after the procedure, 6 of the 12 patients (26%) reported an exacerbation of symptoms. However, the VAS score and mean analgesic consumption reduced significantly 2 weeks later. No procedure-related deaths or major adverse events were reported. This study demonstrated that EUS-guided direct celiac ganglion irradiation is feasible, but further studies are needed to prove its efficacy.

Endoscopic Ultrasound-Guided Injection Therapies

Ethanol or chemotherapy injection for solid tumors. EUS-guided ethanol injection therapy was first applied to pancreatic insulinoma. The resolution of tumor and hypoglycemic symptoms were reported in two case series. In a single-center study, five patients with pancreatic insulinoma were treated with EUS-FNI without any significant adverse effect. Patients did not report any hypoglycemia-related symptoms after the procedure during a median 13-month follow-up period. Similar results were also reported in four patients with pancreatic insulinoma in a recent study. These reports demonstrated that EUS-FNI using alcohol may be an alternative treatment option for patients with pancreatic insulinoma who are not candidates for surgery. A gastrointestinal stromal cell tumor, adrenal metastasis from lung cancer, left hepatic metastatic carcinoma, and two metastatic pelvic lymph nodes in a patient with rectal cancer were treated by EUS-guided ethanol injection without any procedure-related complications.

EUS-guided intrahepatic arterial chemotherapy (5-fluoracil or 5-fluorodeoxyuridine) has been compared with interventional radiology-guided injection in a randomized trial of 25 patients with colorectal cancer and liver metastasis. Although the overall treatment response and survival were comparable between the two groups, the median duration of hospitalization and rate of adverse events were significantly lower in EUS-FNI cohort. The study suggested that EUS-guided intraarterial chemotherapy administration is feasible and safe in a subset of patients with metastatic liver disease.

EUS-FNI of intratumoral gemcitabine was administered as one-time induction therapy prior to conventional multimodality therapy in 36 patients with locally advanced or metastatic pancreatic cancer. The primary endpoint of the study was toxicity. There were no procedure-related adverse events. Four patients (20%) with stage III unresectable tumor were downstaged and underwent a R0 resection. The study demonstrated the feasibility, safety, and effectiveness of intratumoral EUS-FNI using gemcitabine for pancreatic cancer.

In summary, although important advances have been made in recent years for the treatment of abdominal malignancies by EUS-guided injection ( Table 25.2 ), prospective trials evaluating procedural indications and adverse events are needed before the treatment can be recommended for routine clinical use.

TABLE 25.2

Endoscopic Ultrasonography-Guided Fine-Needle Injection of Gastrointestinal Tumors

Authors (Year)	Agent	Patients (n)	Target	Results	Complications
Levy et al. (2012)	Ethanol	5	Pancreatic insulinoma	Complete symptom resolution	None
Qin et al. (2014)	Ethanol	4	Pancreatic insulinoma	Complete symptom resolution	None
Artifon et al. (2013)	5-Fluoracil or 5-fluorodeoxyuridine	25	Liver metastases	Shorter hospitalization and fewer complications than interventional radiology	Minimal
Levy et al. (2016)	Gemcitabine	36	Pancreatic adenoca	20% of stage III patients were downstaged and underwent R0 resection	None
Chang et al. (2000)	Lymphocyte cytoimplants	8	Pancreatic adenoca	2 partial, 1 minor response	Low-grade fever
Hecht et al. (2003)	ONYX-015 + gemcitabine	21	Pancreatic adenoca	2 partial, 2 minor response, 6 stable, 11 progressive	Sepsis in 2 patients
Irisawa et al. (2007)	Dendritic cells	7	Pancreatic adenoca	No difference of survival	None
Hirooka et al. (2009)	Dendritic cells + gemcitabine	5	Pancreatic adenoca	1 partial response, 2 stable, survival better	None
Hecht et al. (2012)	TNFerade + chemoradio	50	Pancreatic adenoca	Locoregional control and downstaging at higher dose	Mild toxicity
Herman et al. (2013)	TNFerade + chemoradio	187	Pancreatic adenoca	No survival benefit	Minimal
Chang et al. (2012)	TNFerade + chemoradio	24	Esophageal cancer	Longer survival	Frequent adverse events, thromboembolism in higher doses