Introduction
Pancreatic cystic lesions (PCLs) are being increasingly recognized on imaging studies,
with prevalence increasing with age and ranging from 2 % to 37 % [1]. Mucinous cysts, including mucinous cystic neoplasms (MCNs) and intraductal papillary
neoplasms (IPMNs), are the most prevalent type of pancreatic cysts and are at risk
of progressing into invasive carcinoma. These cysts require further management, yet
current strategies including radiographic surveillance and surgical resection present
their own set of difficulties. Radiographic surveillance is associated with high cost,
radiation exposure, and offers no definitive treatment, and surgical resection carries
substantial risk for serious adverse events (AEs) (20 %–40 %) and mortality (1 %–3 %)
[1]. Endoscopic ultrasound (EUS)-guided fine needle infusion is an evolving and promising
method for chemically ablating appropriately selected mucinous type pancreatic cysts
with an endoscopic, minimally invasive technique that carries complete ablation rates
ranging from 50 % to 79 % [2]
[3]
[4]. Rates of associated serious AEs for EUS-guided chemoablation range from 0 % using
an alcohol-free technique to as high as 10 % when alcohol is used in the ablation
process [1]
[2]
[3]
[4]. The most common AE is pancreatitis, which is believed to be due to the inflammatory
and toxic effects of the dehydrated alcohol [1].
Interestingly, although pancreatitis is considered an AE, we note that in randomized
trials where the treatment results of patients who suffered acute pancreatitis (AP)
were reported, the cyst ablation rates of patients with AP were as high as 100 % [1]
[5]. There was complete ablation in the two trials in which these metrics were reported,
yet when reexamining another randomized trial with the investigator, this phenomenon
may not have been observed [6]. This evidence, however, suggests that localized recruitment of the immune system
may represent an alternative treatment approach – the addition an immune stimulating
agent to the chemoablation cocktail – which could stand to significantly increase
the efficacy of EUS-guided chemoablation. Here we present a case at our institution
that impressively illustrates the pancreatitis kills cysts phenomena.
Case report
A 53-year-old female with no significant medical history noticed a palpable abdominal
mass, and subsequent abdominal computed tomography (CT) and magnetic resonance imaging
(MRI) showed an approximate 8 × 7 cm cystic mass ([Fig. 1] and [Fig. 2]) in the head of the pancreas with internal septations and thin calcifications in
the cyst wall. Fluid cytology showed pigment-laden macrophages and glandular debris.
Carcinoembryonic antigen could not be obtained due to high viscosity of the cyst fluid.
The cyst was most consistent with a high-risk MCN given the presentation in a middle-aged
female with no prior history of pancreatitis and characteristic cross-sectional imaging
with lack of a competing diagnosis.
Fig. 1 Coronal abdominal CT showing a 7.7 cm × 6.7-cm cystic pancreatic head lesion displacing
the gallbladder (red arrow) portal and superior mesenteric veins (yellow arrow).
Fig. 2 T2-weighted MRI showing a 7.2 cm × 6.6-cm cystic mass lesion of the pancreatic head
with mass effect displacing the gallbladder (red arrow) and portal venous system (yellow
arrow).
The patient had undergone a technically challenging EUS-guided fine-needle aspiration
(FNA) of the cyst at an outside institution, leading to multiple FNA needle exchanges.
She subsequently developed acute pancreatitis leading to a 7-day hospitalization.
After being offered a Whipple procedure (panceaticoduodenectomy), she presented to
our institution for consideration of alcohol-free, EUS-guided chemoablation of her
pancreatic cyst. However, magnetic resonance imaging (MRI) and magnetic resonance
cholangiopancreatography (MRCP) 5 weeks after her hospitalization for pancreatitis
showed a significantly smaller cyst of 3 cm without fistulization, fluid collections,
or development of walled-off pancreatic necrosis (WOPN), and when the EUS-guided ablation
procedure was performed at 11 weeks, no cyst could be found. Follow-up MRI-MRCP at
9 and 18 months ([Fig. 3]) showed complete cyst resolution without development of further complications or
other intraabdominal findings.
Fig. 3 T2-weighted MRI at 18 months post FNA-induced pancreatitis showing complete resolution
of the MCN. Cystic duct is shown by the red arrow and a non-dilated section of the
pancreatic duct is shown by the yellow arrow.
Discussion
Cyst ablation rates in patients who develop acute pancreatitis as a result of cyst
chemoablation are reported near 100 % in the literature versus 50 % to 79 % with EUS-chemoablation
without development of pancreatitis. Here we illustrate the inflammatory phenomena
of acute pancreatitis that leads to higher rates of cyst ablation, and speculate on
the potential role of an immune-stimulating agent as an additive to the chemoablation
cocktail which may stand to improve the efficacy of premalignant pancreatic cyst chemoablation.
Acute pancreatitis involves a complex cascade of immunologic events with recruitment
of leukocytes including macrophages, monocytes, and T cells, and as such, the addition
of an immune-activating agent to the chemoablation admixture by EUS-guided infusion
may carry independent efficacy and have a synergistic effect on ablation rates.
Immunotherapy represents a recent paradigm shift in the treatment of malignancies
and has had notable clinical success regarding sustained treatment response for certain
types of cancers. Immunotherapy has not been shown to be effective for treatment of
pancreatic cancer, primarily due to the existence of a unique tumor microenvironment
consisting of an abundant desmoplastic stroma and immunosuppressive immune cells that
renders the immune system unable to mount an effective antitumor response. The desmoplastic
stroma is thought to act as a physical barrier for effective delivery of cancer therapies,
and the tumor microenvironment also contains immunosuppressive cell types and lacks
effector T cells that lead to suppressed immune response against the tumor. Pancreatic
tumor stroma also contain carcinoma-associated fibroblasts, which express fibroblast
activation protein-α (FAP) that further contribute to an immunosuppressive environment
[7]. These barriers intrinsic to pancreatic cancer are major contributors to the poor
treatment response seen with immunotherapy. Yet, if premalignant mucinous pancreatic
cysts lack the tumor microenvironment and immunosuppressive immune cells of pancreatic
cancer, they may represent a viable target for immunotherapy. A pilot study evaluating
EUS-guided chemoablation with the addition of an immunotherapy agent in patients with
premalignant pancreatic cysts would be warranted to validate this approach.
Conclusion
High rates of complete cyst ablation are noted when pancreatitis is associated with
pancreatic cyst ablation. This is evident both in the literature and by the representative
case at our institution. Localized activation of the immune system likely plays a
role in tumor destruction, and so these lesions may show a robust response to immunotherapy.
We propose that further studies are warranted to investigate the utility of adding
an immune-activating agent to the chemoablation admixture in an effort to increase
rates of complete ablation over that noted currently with chemoablation alone (50 %–79 %).
A pilot study involving EUS-guided infusion of an immune-activating agent with one
of the currently used chemoablation agents (paclitaxel ± gemcitabine) for premalignant
pancreatic cyst ablation would be warranted to validate this approach.
