| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CLINICAL CASE SEMINAR |
MyelomaDepartment of Clinical Biochemistry (D.J.H., M.S., S.O.), Addenbrooke’s Hospital, Cambridge CB2 2QR, United Kingdom; Departments of Haematology-Oncology (M.M.) and Clinical Biochemistry (R.M.-B.), Queen Mary’s Hospital, Sidcup, DA14 6LT, United Kingdom; Department of Clinical Biochemistry (M.N.F.-W.), Southend Hospital, Essex SS0 0RY, United Kingdom; and Clinical Laboratory (R.M.-B., G.W.), Royal Surrey County Hospital, Guildford GU2 7XX, United Kingdom
Address all correspondence and requests for reprints to: David J. Halsall, Department of Clinical Biochemistry, Addenbrooke’s Hospital, Cambridge CB2 2QR, United Kingdom. E-mail: djh44{at}hermes.cam.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
myeloma in whom frequent hypoglycemia resulted from the ability of the monoclonal IgA- to bind insulin.
Objectives: The aim of this study was to describe the occurrence of profound hypoglycemia in a patient with IgA- myeloma, characterize biochemically the nature of the IgA:insulin complex present, and place this case in the context of the published literature on hypoglycemia resulting from autoantibodies to insulin.
Design: A case study was performed.
Patients: A single case of profound hypoglycemia associated with IgA- myeloma was studied.
Intervention: There were no interventions.
Main Outcome Measures: A case study was performed.
Results: Polyethylene glycol precipitation and gel filtration chromatography were used to demonstrate high-molecular weight insulin immunoreactivity in the patient’s plasma. This was characterized as an insulin binding IgA- paraprotein present at 4200 mg/dl (42 g/liter) with a relatively high insulin dissociation constant of 0.32 µM/liter using radiolabelled insulin binding studies.
Conclusions: We present the first case of hypoglycemia due to IgA binding insulin antibodies in a patient with an IgA- paraprotein myeloma. The hypoglycemia was associated with high-plasma insulin levels and relatively low C-peptide levels. A plausible mechanism for the hypoglycemia is the delayed clearance of insulin. This case broadens the spectrum of monoclonal gammopathies that have been associated with anti-insulin reactivity and spontaneous hypoglycemia.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To date all of these cases have been associated with IgG myeloma. Here, we present the first case of hypoglycemia due to an IgA myeloma.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Patient A was a 72-yr-old Caucasian male presented to the Emergency Department of his local hospital having had a grand mal seizure. He was confused and unresponsive on admission, and was found to be hypoglycemic. The hypoglycemia could only be reversed by iv 50% dextrose infusion. Computed tomography (CT) of the head and cerebrospinal fluid examination were normal. Renal function was normal, but plasma total protein concentration was elevated at 9.7 g/dl (albumin 2.9 g/dl). IgA was elevated at 5050 mg/dl, with IgG and IgM 240 and 50 mg/dl, respectively. Serum electrophoresis showed a discrete dense paraprotein band in the ß zone with a virtual absence of background 
globulins. The paraprotein was identified as IgA- by immunofixation. Plasma viscosity was elevated at 2.83 cP, and a bone marrow examination showed a large excess of plasma cells (
80%). Corrected serum calcium was slightly elevated at 10.8 mg/dl (2.57 mmol/liter) (reference range 8.8–10.2 mg/dl), and skeletal radiography showed a small lucent area in the skull close to frontal suture and presumed lytic lesion in the head of the left humerus. Abdominal, pelvic, and head CT did not reveal evidence of metastatic disease. A diagnosis of IgA myeloma was made, and the patient was commenced on plasma exchange and combination chemotherapy (cyclophosphamide, vincristine, adriamycin, and dexamethasone) for myeloma. Over the succeeding 2 months, the patient had five episodes of severe symptomatic hypoglycemia requiring medical help. During all of these episodes, the plasma glucose was between 25 and 43 mg/dl (1.4–2.4 mmol/liter). During one of these episodes, his plasma insulin and C-peptide were measured at 15,300 and 174 pmol/liter, respectively (fasting ranges: insulin <60 and C-peptide <960 pmol/liter). CT of the abdomen did not show any pancreatic mass.
After six cycles of cyclophosphamide, vincristine, adriamycin, and dexamethasone chemotherapy, the paraprotein band was undetectable, and the patient had become normoglycemic. Four months later the patient had relapsed with a paraprotein level of 5.8 g/dl and episodes of hypoglycemia. His further treatment included intermediate dose melphalan, plasma exchange, dexamethasone, and thalidomide. Unfortunately, he did not respond to this treatment and died 6 wk later.
Biochemical analysis
Insulin and C-peptide were assayed using Perkin-Elmer Wallac reagents (Beaconsfield, Bucks, UK) and performed using an AutoDELFIA 1235 automatic immunoassay system, with protocols provided by the same manufacturer and Mercodia Iso-insulin ELISA and Mercodia C-peptide ELISA specific assays. IgA analysis was performed using a Dade Behring BNII nephelometer (Dade Behring Inc., Deerfield, IL).
Polyethylene glycol (PEG) precipitation. Patient’s plasma was mixed with an equal volume of PEG 6000 (25% weight-volume) and centrifuged for 30 min at 1400 x g, before reassaying the supernatant. Insulin recovery from the plasma of 10 insulin-resistant subjects after PEG precipitation was used as a control.
Heterophile blocking tubes. Heterophile blocking tubes were provided by Skybio Limited (Bedfordshire, UK) and used as directed by the manufacturer.
Gel filtration chromatography
Plasma was eluted from a 40 x 1.6-cm column of Sephacryl S-100 (GE Healthcare UK Ltd., Buckinghamshire, UK) with 0.02 mol/liter Tris, 0.26 mol/liter sodium chloride, and 0.006 mol/liter sodium azide (pH 7.4) at a flow rate of 0.3 ml/min. Fractions of 1.5 ml were assayed for insulin, IgA, and albumin.
Radiolabeled insulin binding studies
All incubations were performed in binding buffer [50 mM Tris (pH 7.4), 120 mM NaCl, 15 mM Na acetate, 1.2 mM Mg2SO4, and 1% BSA].
Insulin autoantibody Ig isotyping. Plasma (5 µl) was incubated with 50 µl anti-IgG, anti-IgA, or anti-IgM agarose for 4 h at 4 C in 100 µl buffer with mixing. Antibody bound to the isotype-specific agaroses was washed three times with cold PBS and resuspended in 250 µl 125I-insulin ± 10 µM cold insulin overnight at 4 C with mixing. Agarose pellets were washed twice with cold PBS, and bound radioactivity was determined.
Quantitative insulin autoantibody binding studies.
Plasma was stripped of endogenous insulin using acid dextran-coated charcoal (10). Stripped plasma (100 µl, final dilution 1/6250) was incubated with 125I insulin and increasing concentrations of unlabeled insulin for 2 h at 37 C in 250 µl final volume of binding buffer. Protein-bound insulin was precipitated on ice by the addition of ice-cold carrier bovine globulin (0.09% final concentration) and PEG (12.5% final concentration) for 15 min, followed by centrifugation at 10,000 rpm, for 5 min at 4 C. The pellet was washed once with 10% PEG before determination of protein-bound radioactivity.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The patient’s plasma insulin concentration during a hypoglycemic event was estimated at 15,300 pmol/liter (normal fasting level <60). At this time plasma C-peptide was disproportionately low at 174 pmol/liter (normal fasting level <80). Results were essentially similar using the Mercodia insulin and C-peptide assays.
Proprietary heterophile antibody blocking tubes had no effect on insulin recovery (data not shown). Recovery of plasma immunoreactive insulin after PEG precipitation of the plasma was low (5%) compared with control samples (>70%), suggesting that a high-molecular weight insulin immunoreactivity was present in the patient sample.
Gel filtration chromatography
Gel filtration chromatography showed the majority of insulin immunoreactivity in the patient’s plasma to be of higher molecular mass than expected for insulin. The high-molecular mass insulin immunoreactivity co-eluted with the IgA paraprotein in the patient’s plasma (Fig. 1
). Two plausible mechanisms for the high-molecular mass insulin immunoreactivity are an immunoreactive insulin:IgA complex or a noninsulin containing "heterophilic" IgA antibody that is capable of cross-linking the capture and detection antibodies used in the insulin immunoassay (11). The mass resolution of the column is insufficient to distinguish these two mechanisms.
|
The IgA precipitable fraction from the patient’s plasma can bind radiolabeled insulin (Fig. 2). This is reversible because it can exchange with unlabeled insulin. It shows the high-molecular weight insulin immunoreactivity in this patient to be due to insulin:IgA complex rather than a cross-linking antibody.
|
) Scatchard analysis showed a single class of sites with relatively low affinity for insulin (Kd = 0.32 µM/liter) but at high concentration (0.567-mM binding sites/liter plasma) (Fig. 3B). Assuming two antigen binding sites per IgA molecule, and that plasma IgA is monomeric with a molecular mass of 150 kDa, this is equivalent to 4.2 g/dl. This is in agreement with the serum paraprotein concentration of 5.0 g/dl found in this patient.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The presence of hypoglycemia in this patient suggests the presence of circulating bioactive insulin. Despite the vast excess of IgA in this patient (insulin will be >99.9% bound using the Kd derived from the Scatchard plots), it is likely that the relatively weak binding constant of the IgA compared with the insulin receptor will allow insulin to dissociate from the IgA and bind to the receptor. High-capacity, low-affinity antibodies are likely to be a key feature of this syndrome. In previous reports on myeloma-related hypoglycemia due to insulin binding Ig, where the antibody binding has been characterized, this type of binding has also been found (5, 6). Similar low-affinity monoclonal sites have also been described in IAS (12), although this syndrome more commonly presents with polyclonal insulin binding antibodies (13, 14). The polyclonal and monoclonal anti-insulin antibodies found in IAS may represent a continuum of disease states, which switch in individual patients as the disease progresses. A monoclonal picture is typical of severe hypoglycemia, with a more polyclonal state in remission (15). Because anti-insulin antibodies can be associated with both hyperglycemia associated with treatment failure or hypoglycemia due to the mechanisms described previously, we postulate that low-affinity sites are more likely to cause hypoglycemia and that high-affinity sites will cause hyperglycemia because insulin will be bound, but not active.
Anti-insulin antibodies are a very common feature of insulin therapy, particularly with animal insulin, but also with recombinant human insulin. Their presence, which can be reliably detected using competitive radiobinding assays (16), must raise awareness of interference with the insulin assay because the effect of the anti-insulin antibody on the assay is not predictable. Repeating the assay using different antibodies increases confidence that the assay is correct, but interference in both assays is possible (16).
In conclusion, we present the first case of hypoglycemia due to IgA binding insulin antibodies in a patient with an IgA- paraprotein myeloma. Anti-insulin IgA was present in the patient’s plasma at high concentration, but relatively low affinity. The hypoglycemia was associated with high-plasma insulin levels and relatively low C-peptide levels. A plausible mechanism for the hypoglycemia is delayed clearance of insulin, which is still available to bind its receptor due to the relatively weak affinity of the IgA for insulin.
|
Footnotes |
|---|
First Published Online April 3, 2007
Abbreviations: CT, Computed tomography; IAS, insulin autoimmune syndrome; PEG, polyethylene glycol.
Received January 11, 2007.
Accepted March 26, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
