Is There a Benefit From Islet Autotransplantation in Patients With Type 1 Diabetes Mellitus Undergoing Total Pancreatectomy?

Abstract

Children with acute recurrent and chronic pancreatitis (CP) suffer from abdominal pain that leads to hospitalizations, opioid dependence, and poor quality of life. Total pancreatectomy with islet autotransplantation (TPIAT) is offered as a surgical option in management of debilitating pancreatitis that fails medical and endoscopic therapy to reduce or eliminate pain. Given that patients with type 1 diabetes mellitus (T1DM) lack insulin-producing β cells, the outcomes from autotransplanting islet isolates back into total pancreatectomy patients with T1DM are not fully known.

We performed TPIAT in two CP patients who also had a diagnosis of T1DM for at least 6 years prior to the operation and evaluated the clinical and laboratory outcomes before and after the operation. Post-operatively both patients’ abdominal pain had significantly subsided, they were weaned off opioid medications, and they were able to return to full-time school attendance. In addition, total daily dose of insulin in one patient was able to be slightly reduced at 12 months post-TPIAT. We observed in vitro that residual α cells and β cells in T1DM islets were able to secrete a small amount of glucagon and insulin, respectively.

Introduction

In children, acute recurrent pancreatitis (ARP) and chronic pancreatitis (CP) may be caused by different etiologies including metabolic disorders, systemic illness and infections, gallstones and biliary tract obstruction.¹ Importantly, hereditary pancreatitis risk genes variants of CFTR, SPINK1, PRSS1, and CTRC, amongst other genes, have been shown to be associated with ARP and CP in numerous pediatric and adult patient cohorts.^2–6 Children with ARP and CP may suffer from chronic abdominal pain and opioid dependence, which can lead to poor quality of life.⁶ Total pancreatectomy with islet autotransplantation (TPIAT) is one of the treatment approaches that may be considered in patients who have developed pain and debilitation from ARP or CP and have failed all medical and endoscopic treatment approaches.^7,8 Not only does the operation improve quality of life by allowing patients to become pain-free and achieve opioid independence in the majority of cases, but also exogenous insulin administration can be weaned as early as 3 months after operation, with the long term outcome of complete insulin independence in a subset of patients.^9–11

Islets in the human pancreas consist of insulin-producing β cells (<70%), glucagon-producing α cells (10–20%), somatostatin-producing δ cells (<10%), and pancreatic polypeptide-producing PP cells (<5%).^12,13 Patients suffering from type 1 diabetes mellitus (T1DM) lack insulin secretion due to autoimmune destruction of the β cells.^14,15 The benefit of islet autotranplantation in a T1DM patient undergoing total pancreatectomy for debilitating pancreatitis is not fully known.

In this study, we describe the course of two children with T1DM who presented with debilitating abdominal pain and poor quality of life from CP. In our multidisciplinary review, we made the recommendation to offer TPIAT despite positivity of autoimmune β-cell destruction biomarker, glutamic acid decarboxylase (GAD). Our hypothesis was that the presence of glucagon-producing α cells would be of benefit to these patients’ long term outcomes. In this paper, we seek to highlight the clinical outcomes and laboratory findings and in vitro testing results from these T1DM patients following TPIAT.

CASE REPORTS

Human Studies

Human tissue was collected according to standard research protocols approved by the Institutional Review Board (2014–6279) and Department of Pathology at Cincinnati Children’s Hospital Medical Center. Chart review was approved under Institutional Review Board (2020–0854).

Patient 1

A 12-year-old female originally diagnosed at 6 years of age with hemolytic uremic syndrome during which time she was found to have an insulin-dependent form of diabetes.

Pre-TPIAT

The patient’s clinical course of hemolytic uremic syndrome was complicated by the development of her first occurrence of acute pancreatitis (AP). Over the subsequent years, she had recurrent episodes of AP progressing to CP with exocrine pancreatic insufficiency. Genetic testing identified a heterozygous SPINK1 gene mutation (p.N34S). She underwent 3 incomplete attempts to deeply access her main pancreatic duct by endoscopic retrograde cholangiopancreatography (ERCP). Due to continued pain and debilitation and failed endoscopic interventions, TPIAT was recommended. Prior to TPIAT, she used an insulin infusion pump to manage her diabetes, and total daily dose (TDD) of insulin was 0.7 units/kg with a Hemoglobin A1c of 8.4%. Pre-TPIAT metabolic assessment with mixed meal testing was performed which documented minimal-to-no C-peptide secretion (range, <0.1–0.2 ng/ml) after ingestion of a Boost meal (6 mL/kg, max dose 360 mL). The islet antibody panel (performed at The Barbara Davis Center for Childhood Diabetes, University of Colorado School of Medicine, Aurora, Col) was positive for GAD antibodies at 27 units/mL (elevated >25 units/mL), and insulin autoantibodies (IAA) at 0.551 units/mL (elevated >0.013 units/mL), the latter was interpreted as due to her prior exogenous insulin exposure.

Post-TPIAT

At the time of TPIAT, patient 1 received 248 islet equivalents/kilogram body weight (IEQ/kg) of islets. The child’s postoperative course was uncomplicated, and she was discharged from the hospital within two weeks after operation. At her 3-month post-TPIAT follow-up, she reported having only minimal bouts of abdominal pain or discomfort, not similar to her pancreatitis pain, and she was back to full activity levels, off of opioids, and attending school full-time. These outcomes were sustained at the 12 month visit.

Patient 2

A 14-year-old female was diagnosed with an insulin-dependent form of diabetes at age 6 years.

Pre-TPIAT

The patient’s first occurrence of AP did not develop until 4 years after her diabetes diagnosis. She subsequently had multiple recurrent episodes of pancreatitis, often requiring hospitalization for pain control and fluid hydration. Genetic testing found her to carry heterozygous CFTR p.F508del and (TG)11–5T variants and a heterozygous SPINK1 (p.N34S) mutation. Fifteen months prior to TPIAT, she underwent a therapeutic ERCP with extraction of pancreatic intraductal stones followed by additional ERCPs for further endotherapy. She experienced only transient symptomatic relief from endotherapy, but would consistently return to her baseline level of abdominal pain and debilitation with the inability to attend school. Total pancreatectomy with islet autotransplantation was offered as a definitive therapeutic option. Prior to her operation, she was using multiple daily insulin injections to manage her diabetes, and TDD of insulin was 0.9 units/kg with a Hemoglobin A1c of 10.6%. Pre-TPIAT metabolic testing documented random C-peptide <0.1 ng/ml. The islet antibody panel (performed at The Barbara Davis Center for Childhood Diabetes, University of Colorado School of Medicine, Aurora, CO) was positive for GAD at 164 units/mL (elevated >25 units/mL), IA-2 at 52 units/mL (elevated >7 units/mL), and IAA at 0.371 units/mL (elevated >0.013 units/mL), the latter was interpreted as due to prior exposure to exogenous insulin.

Post-TPIAT

At the time of TPIAT, patient 2 received 154 IEQ/kg of islets. The patient was discharged from the hospital within two weeks following TPIAT with an uncomplicated course. She was weaned off supplemental enteral nutrition by 2.5 months post-TPIAT with removal of her feeding tube, as by that time she was taking full meals by mouth with adequate oral intake. Her abdominal pain had resolved following surgery, allowing her to return to school full-time. At 3 months postoperatively, she was off opioids. These outcomes were sustained at the 12 month visit.

For both patients, metabolic testing was performed at 3 months and 12 months post-TPIAT and documented no C-peptide secretion (all values <0.1 ng/mL) (Table 1). They had no reports of severe hypoglycemia events of less than 50 mg/dL during the 12 months post-TPIAT.

TABLE 1.

Patient Characteristics and Clinical Outcomes

	Type 1 Diabetic Patients
	Patient 1			Patient 2
Age	12 years old			14 years old
Sex	Female			Female
Mutation	SPINK1			ΔF508/SPINK1
Duration, y
T1DM	6			8
CP	4			3
ERCP	3			3
Infusion, islets, IEQ/kg	248			154
TPIAT	Pre-TPIAT	Post-TPIAT		Pre-TPIAT	Post-TPIAT
		3 mo	12 mo		3 mo	12 mo
Pancreatitis pain	Yes	No	No	Yes	No	No
No. hospitalizations	2	1 (in the first 3 mo)		9	1 (in 12 mo)
Insulin TTD, units/kg/d	0.7	0.72	0.68	0.9	1.11	0.95
C-peptide, ng/mL	<0.1	<0.1	—	<0.1	<0.1	—
HbA1c, %	8.4	7.2	7.8	10.6	8.4	10.0
Islet antibodies, units/mL	—	—	—	—	—	—
GAD (20)*	27	—	—	164	—	—
IAA (0.36)*	0.551	—	—	0.371	—	—
IA-2 (5)*	—	—	—	52	—	—

The islet antibody panel was performed at The Barbara Davis Center.

^*Numbers indicate limitation of the normal range.

T1DM indicates type 1 diabetes mellitus; CP, chronic pancreatitis; ERCP, endoscopic retrograde cholangiopancreatography; IEQ/kg, islet equivalents/kilogram body weight; TDD, total daily dose; HbA1c, Hemoglobin A1c; GAD, glutamic acid decarboxylase; IAA, Insulin autoantibodies; IA-2, islet antigen 2.

Type 1 Diabetes Mellitus Islet Morphology and Function In Vitro

We successfully isolated and cultured islets in vitro from remnant cell pellets following TPIAT and examined the cells using immunofluorescence microscopy for assessment of insulin and glucagon (see Supplemental Materials and Methods), comparing the islet cell population to that of patients without T1DM who underwent TPIAT. In a CP patient (Carboxypeptidase A1 mutation) without T1DM, we observed that the population of β cells (71%) was over 3-fold higher than that of α cells (20%) in isolated islets (Fig. 1B).¹⁷ In T1DM patients, we observed a significantly lower β-cell population (< 5% of total islets) and significantly higher α-cell distribution (> 80% of total islets) (Fig. 1A).

FIGURE 1.

Immunofluorescence microscopy of insulin and glucagon. Patient-derived pancreatic islets were isolated and cultured in vitro from pancreatitis patients at the time of TPIAT. Islet cells were examined using immunofluorescence microscopy of insulin (INS), glucagon (GCG), and DAPI in a T1DM pancreatitis patient (A: Patient 1) and in a non-TIDM chronic pancreatitis patient (B: Carboxypeptidase A1 mutation). Scale bars: 20 μm.

In addition, we observed that the mean insulin secretion from T1DM islets was extremely low (Patient 1: 9.55 ± standard error of the mean (SEM), 2.56 μIU/mL; Patient 2: 3.27 ± 0.16 μIU/mL), compared to non-T1DM islets (635.91 ± 13.33 μIU/mL; P < 0.00001), consistent with the loss of β-cell mass (Fig. 2A). Although the number of β cells in the T1DM islets was substantially reduced, we nevertheless observed both the ability to secrete detectable amounts of insulin, as well as a response to concentrated glucose (Fig. 2A). With high glucose (450 mg/dL) stimulation, insulin secretion in Patient 1 was 1.57 fold higher, compared to low glucose (100 mg/dL) conditions, although this difference did not reach statistical significance (P = 0.23). By confocal immunofluorescence microscopy, we observed that α cells were highly distributed in T1DM islets (Fig. 1A). However, although the α-cell population was abundant, the mean glucagon secretion was significantly lower (patient 2: 5.13 ± SEM, 0.94 pg/mL with 100 mg/dL glucose incubation) in T1DM islets, compared to islets from the non-TIDM patient (28.36 ± SEM, 2.22 pg/mL; P < 0.0001) (Fig. 2B).

FIGURE 2.

Endocrine function in vitro. Pancreatic islets were cultured in vitro, and insulin (A) and glucagon (B) secretion was monitored using ELISA. Endocrine function was compared between T1DM and non-TIDM patients in response to low (100 mg/dL) and high (450 mg/dL) glucose concentrations. (n = 3; data are mean ± SEM; **P < 0.002, ***P < 0.0001, ****P < 0.00001).

DISCUSSION

The primary goal of TPIAT is to reduce or eliminate pain and to improve quality of life from debilitating pancreatitis. Islet autotransplantation preserves β-cell and α-cell function to off-set a brittle form of diabetes. However, there are limited reports of CP patients with T1DM who have undergone TPIAT.¹⁸ The role of the islet autotransplantion portion of the operation in these patients is not fully understood. Herein, we present two cases of adolescent patients with CP and T1DM who underwent TPIAT. They underwent several endoscopic interventions and medical treatment for several years prior which resulted in failure to manage their debilitating pain, thereby leading to the recommendation for TPIAT as definitive intervention. The post-TPIAT clinical course was uncomplicated. Both patients showed significantly improved daily life function, their abdominal pain patterns were improved, and opioids were weaned by 3 months post-TPIAT.

Glutamic acid decarboxylase antibody was positive in both patients preoperatively, indicating that they had islet autoantibody-related diabetes. Although both patients had a paucity of β cells, we observed that the residual β cells secreted detectable amounts of insulin and were able to slightly reduce TDD for Patient 1 who received 245 IEQ/kg for islet autotransplantation at 12 months post-TPIAT. We were particularly interested in examining the α-cell population in both isolates with regard to morphology and function, and we demonstrated that both patients had the presence of glucagon-secreting cells. Importantly, both patients did not report episodes of significant hypoglycemia post operatively for the duration of follow up under this report.

The implications of a highly abundant α-cell population, massive loss of β-cells, and markedly reduced glucagon secretion in T1DM-islets are not clear, and the literature is limited. Brissova et al suggested the possibility that these findings may be associated with altered gene expression of nuclear regulators (eg, ARX, MAFB, and RFX6) that can directly or indirectly impact glucagon secretory pathways.¹⁹ Spijker et al reported that there can be β-to-α cell conversion in human islets without any genetic modification.²⁰ They observed mature human β cells rapidly converted into α cells in three-dimensional in vitro architectural cell culture condition by islet cell reaggregation. These findings may explain the altered cell populations and cell function in T1DM islets, however, the mechanism of cell-cell transition needs to be further investigated. Our findings suggest that islet autotransplantation may be beneficial in T1DM patients undergoing total pancreatectomy in order to provide residual β-cell function, but predominantly, for α-cell presence and function.

In conclusion, TPIAT is a surgical option to remove the source of debilitating abdominal pain in CP patients, with the goal of islet autotransplantation to mitigate brittle glucose homeostasis. While TPIAT has been shown to improve quality of life in patients with CP, not much is known about islet autotransplantation in the setting of underlying T1DM. Our study confirms the presence of glucagon-secreting α cells that produce detectable amounts of glucagon in the T1DM-islets. Since these patients have a one-time opportunity to receive their own islets, careful consideration should be given to offering islet autotransplantation to T1DM patients, rather than proceeding with total pancreatectomy alone. Future studies in TPIAT-related outcomes should include populations of patients with T1DM and other forms of diabetes preceding the surgical intervention.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.