Pulmonary Arterial Hypertension and Intraductal Papillary Mucinous Neoplasms of the Pancreas

To the Editor:

Pulmonary arterial hypertension (PAH) is a rare, proliferative disease of the precapillary pulmonary vasculature, characterized by elevated pulmonary vascular resistance and resultant right ventricular failure.¹ Although new PAH therapies have resulted in substantially improved patient survival,² increased longevity has been accompanied by a growing incidence of other medical comorbidities. We herein report a possible association between two seemingly unrelated, rare conditions, PAH and intraductal papillary mucinous neoplasm (IPMN) of the pancreas.

Methods

We highlight key characteristics of six patients from the pulmonary hypertension clinic at the University of Colorado, all managed for ≥10 years with PAH, who were subsequently diagnosed with IPMN based on radiographic features. We also discuss unique aspects of clinical management and propose potential unifying mechanisms underlying both diseases.

Results

We identified six PAH patients treated at the University of Colorado Hospital between 1996 and 2021, four with IPAH, one with portopulmonary hypertension, and one with PAH caused by undifferentiated systemic rheumatic disease. Our clinic sees approximately 3,000 visits annually; of these, approximately 65 newly referred patients are diagnosed with PAH each year. These six patients had severe pulmonary hypertension requiring multiple medications, including IV prostacyclin (PGI₂) analog therapy for many (14 ± 4.9 years (Table 1). The most common symptoms included abdominal pain, diarrhea, nausea, vomiting, weight loss, anorexia, and fatigue. Evidence of IPMN was seen on CT or magnetic resonance cholangiopancreatography (MRCP) (Fig 1).

Table 1.

Relevant Clinical Information of Individual Patients

Characteristics	Patient No.
	1	2	3	4	5	6
Age at IPMN diagnosis	64	67	58	65	55	55
Sex	Female	Female	Male	Female	Female	Female
PH type	IPAH	IPAH	Portopulmonary hypertension	IPAH	IPAH	PAH due to undifferentiated systemic rheumatic disease
Functional class at IPMN diagnosis	III	II	III	II	II	II
Years from PAH diagnosis at time of IPMN diagnosis	16	10	14	14	12	23
Years on PGI2 at time of IPMN diagnosis	15	8	14	14	12	23
Therapy at time of IPMN diagnosis	Epoprostenol 30 ng/kg/min Sildenafil 80 mg TID	Epoprostenol 59 ng/kg/min Sildenafil 20 mg TID Bosentan 125 mg BID	Epoprostenol 58 ng/kg/min	Epoprostenol 32 ng/kg/min Sildenafil 40 mg TID	Epoprostenol 46 ng/kg/min Sildenafil 40 mg TID	Epoprostenol 37 ng/kg/min Sildenafil 40 mg TID Macitentan 10 mg daily
Tobacco history	5 pack-years	Nonsmoker	Nonsmoker	Nonsmoker	Nonsmoker	Nonsmoker
Family history of malignancy?	Lung cancer	Melanoma and colorectal cancer	Lung and colorectal cancer	Yes—unknown type	No	Lung and prostate cancer
Diagnostic modality	MRCP	CT A/P with contrast	CT A/P with contrast	CT A/P with contrast	MRCP	MRCP
Symptoms leading to abdominal imaging	Abdominal pain, vomiting, night sweats, weight loss	Abdominal pain, anorexia, vomiting	Abdominal pain	Fever, anorexia, nausea, vomiting, abdominal cramping, lightheadedness	Fatigue, anorexia, weight loss	Abdominal pain, nausea, vomiting, anorexia, weight loss, fatigue
Exocrine/ endocrine symptoms	Diarrhea	Diarrhea	None	Diarrhea	Diarrhea	None
Extrapancreatic malignancies	No	No	No	No	No	Sigmoid colon hamartoma (neuroma)
Therapy	None	Celiac plexus block and neurolysis	Celiac plexus block and neurolysis	None	None	None
Outcome	Died 3 years after IPMN diagnosis of PAH complications	Alive at 9 years. Functional class IIIb	Died 4 months after IPMN diagnosis of PAH complications	Died 7 months after IPMN diagnosis of PAH complications	Alive after 6 years	Died 1 month after IPMN diagnosis of Staphylococcus epidermidis bacteremia with mitral valve endocarditis
Right heart catheterization closest to time of IPMN diagnosis
RAP, mm Hg	10	1	8	3	1	5
mPAP, mm Hg	79	47	44	32	41	50
PCWP, mm Hg	15	7	8	13	4	12
CO Thermodilution, L/min	4.5	4.75	7.53	5.7	5.16	6.4
CI Thermodilution, L/min/m2	2.7	3.13	4.17	3.37	2.58	4.22
PVR, WU	13.6	7.0	5.4	3.4	7.9	5.9
Liver function tests at IPMN diagnosis
AST, U/L	14	32	24	13	46	22
ALT, U/L	7	20	17	6	85	18
Alk Phos, U/L	152	187	61	170	44	132
Bilirubin, total, mg/dL	0.5	0.6	0.6	0.6	0.9	0.4
Tumor markers at IPMN diagnosis
CA 19-9, U/mL	<1	<1	19.8a	25.4	Not checked	17.9
CEA, ng/mL	<1	2.0	0.7b	0.4	Not checked	0.7

A/P = abdomen and pelvis; ALT = alanine aminotransferase; AST = aspartate aminotransferase; CEA = carcinoembryonic antigen; CI = cardiac index; CO = cardiac output; IPAH = idiopathic pulmonary arterial hypertension; IPMN = intraductal papillary mucinous neoplasm; mPAP = mean pulmonary artery pressure; MRCP = magnetic resonance cholangiopancreatography; PAH = pulmonary arterial hypertension; PCWP = pulmonary capillary wedge pressure; PGI₂ = IV prostacyclin; PVR = pulmonary vascular resistance; RAP = right atrial pressure; WU = Wood units.

^aValue tested 16 months before IPMN diagnosis.

^bValue tested 19 months before IPMN diagnosis.

Figure 1.

A-1, Patient 1, MRCP, axial T2-weighted image showing pancreatic tail ductal dilation up to 9 mm (white arrow), without evidence of nodularity of abnormal enhancement. There is also marked intrahepatic biliary dilatation. A-2, Patient 1, CT A/P with pancreatic duct dilatation in the pancreatic tail (arrow). Pancreatic parenchymal enhancement is normal without inflammatory changes. B, Patient 2, CT A/P with multiple low attenuation lesions in the pancreas, measuring up to 2.0 cm (white arrows). Pancreatic parenchymal enhancement is normal. There is also extrahepatic biliary dilatation (black arrow). C, Patient 3, CT A/P, coronal reconstruction. Arrows indicating a pancreatic cystic lesion in the pancreatic neck measuring 2.3 cm. Small-volume ascites is present. D, Patient 4, CT A/P. Arrow denoting pancreatic tail cystic lesion measuring 2.1 cm. There is no evidence of abnormal enhancement. E, Patient 5, MRCP, coronal T2-weighted image. Arrow indicating a 7-mm T2 bright lesion communicating with the pancreatic duct suggestive of a branch duct IPMN. F-1, Patient 6, CT A/P. Arrow indicating a low attenuation lesion in the pancreatic body measuring up to 9 mm. F-2, Patient 6, MRCP with arrow indicating a 7-mm cystic lesion communicating with the pancreatic duct suggestive of a side branch IPMN. There is no definite nodularity or enhancement. A/P = abdomen and pelvis; IPMN = intraductal papillary mucinous neoplasm; MRCP = magnetic resonance cholangiopancreatography.

All six patients were presented at our established multidisciplinary pancreatic malignancy conference, which involves medical and surgical oncology, gastroenterology, radiology, radiation oncology, and pathology specialists. The clinical and radiographic features of the six patients were believed most consistent with IPMN. Given the PAH-related elevated peri-procedural risks, the group advised conservative, symptomatic management of their IPMNs without invasive biopsy or surgery. Although we did not observe any overt evidence of cancerous transformation of the IPMN lesions, serial radiographic imaging demonstrated progression of cystic changes in some patients, and there was morbidity attributable to ongoing abdominal pain and ensuing weight loss. Two of the six patients underwent celiac plexus blockade and neurolysis for pain management. Three patients have died a few months to years after their IPMN diagnoses of PAH; another patient died of an infectious complication (Table 1).

Discussion

To our knowledge, this is the first report describing a potential association between IPMN and PAH. IPMN, another rare condition with reported annual incidences of <5 cases per 100,000 people,³ is a mucinous tumor that originates from the epithelial cells lining exocrine ducts of the pancreas.⁴ All six patients had GI symptoms, including diarrhea, abdominal pain, and anorexia. When managing PAH, presence of such symptoms, particularly in the absence of a clear cause, should prompt consideration of IPMN. Diarrhea, abdominal pain, and anorexia can all be particularly confounding symptoms because they are both known complications of PGI₂ analog therapy and sequelae of IPMN.

The presence of concurrent severe PAH posed unique diagnostic and therapeutic challenges in the management of IPMN. The natural history of IPMN depends on the histopathologic characteristics and the ability to remove any existing cancer with clean margins.⁵ Although 5-year survival of IPMN patients without invasive carcinoma can be nearly 100%, this drops to 57.7% in those with invasive carcinoma.⁶ Reported survival rates of patients with main duct type IPMN are particularly poor, likely because of their higher frequency of cancer.⁵ Most IPMN are found incidentally and are usually followed with serial cross-sectional imaging in the absence of worrisome features or high-risk stigmata. Although surgical removal of the whole or a part of the pancreas is typically the mainstay of therapy for IPMN with high-risk features, with 30-day mortality of an invasive surgery such as pancreaticoduodenectomy as low as 1%,⁷ its precise risk in our PAH population is likely much higher and remains undefined.⁸ In patients with PAH, the potential for malignant transformation must be balanced with elevated periprocedural risks and surgical mortality,⁸ because of tenuous hemodynamic reserves and poor tolerance of anesthesia, as well as the substantial mortality associated with severe PAH.²

Whether PAH or its treatments are linked to the pathogenesis of IPMN remains unclear. Considering the many pathobiological similarities between PAH and cancer,⁹ possibly PAH and IPMN are also linked by common pathogenetic mechanisms. One plausible hypothesis concerns mutations in genes encoding SMAD proteins, which relay downstream signaling of the transforming growth factor beta superfamily. Dysregulated SMAD signaling has been implicated in both pancreatic cancers¹⁰ and PAH,¹¹ suggesting that pro-proliferative effects of imbalanced transforming growth factor beta/SMAD signaling may underlie both the vascular proliferation in PAH and growth of IPMN. More elucidating work, however, is needed because SMAD4 mutations, which are associated with PAH¹¹ and some ductal pancreatic carcinomas, appear to be particularly absent in IPMN.¹²

Another consideration is a potential impact of long-term PAH therapy, such as the chronic PGI₂ analog treatment of our cohort, on pancreatic ductal proliferation. Although PGI₂ in particular has demonstrated tumor-suppressive and anti-metastatic properties in cancer studies,¹³ eicosanoids at large have been implicated in pancreatic and other cancers.¹⁴ In addition, PGI₂ has been associated with hyperthyroid states, including goiter, suggesting a possible connection between PGI₂ and exocrine tumors.¹⁵ Testing these hypotheses and elucidating pathogenetic mechanisms will first require precisely defining the true incidence of IPMN in PAH patients through larger, registry-based studies.

Limitations of our study include the lack of pathologic confirmation of IPMN. The PAH-associated, elevated risks of invasive procedures represented a major barrier to performing diagnostic endoscopic interventions and pathology-based risk-stratification. Even in those patients who died, autopsy examination unfortunately was not obtained, for a variety of factors, including family desires. This remaining diagnostic uncertainty of IPMN and reliance on clinical and radiographic clues represent a real-world challenge that will continue to confound its diagnosis and management in PAH patients.

Successful treatment of PAH in recent decades has led to improved patient longevity and more frequent recognition of concurrent comorbidities such as IPMN. As their life expectancy continues to lengthen, we will encounter other diseases in PAH patients with increasing frequency.