|Values are valid only on day of printing.|
Click CC to turn on closed captioning.
Published: December 2009Print Record of Viewing
Dr. Dogan reviews the diagnosis of amyloidosis and describes a highly specific and sensitive novel test for typing of amyloidosis in routine clinical biopsy specimens.
Presenter: Ahmet Dogan, MD, PhD
Welcome to Mayo Medical Laboratories' Hot Topics. These presentations provide short discussion of current topics and may be helpful to you in your practice.
Our presenter for this program is Dr. Ahmet Dogan, Medical Director of the Immunostains Laboratory in the Division of Anatomic Pathology at Mayo Clinic. Dr. Dogan will review the diagnosis of amyloidosis and describe a highly specific and sensitive novel test for typing of amyloidosis in routine clinical biopsy specimens.
Amyloidoses are a group of diseases characterized by abnormal extracellular deposition of proteins in a beta-pleated sheet format. The term amyloidosis is derived from “amylum,” starch in Latin. Although molecular pathogenesis of amyloidosis is not known, it is believed that amyloidosis is caused by misfolding of proteins into a structure that cannot be cleared by physiological scavenging mechanisms. Amyloidosis can be localized to a single organ site or can be systemic involving multiple organs. Systemic amyloidosis is, clinically, most important as vital organs such as the heart or the kidneys are frequently affected.
Diagnosis of amyloidosis is made by examination of biopsy specimen with special stains which give a characteristic reaction based on the physical structure of the amyloid. Of these, Congo red histochemical staining is believed to be the most sensitive and specific method. For best results, Congo red staining should be performed on 10 micron sections and viewed under a strong light source using either polarizing or fluorescent filters. Other, somewhat less specific stains or methods widely used for diagnosis of amyloidosis include thioflavin and sulphated alcian blue histochemical stains, immunohistochemistry against serum amyloid P component (SAP) and electron microscopy.
This is a duodenal biopsy specimen showing subtle amyloid deposition in the basal lamina of duodenal glands. In the upper left panel, Congo red stain gives pale orange-red reactivity with the amyloid deposits. In the upper right panel, under polarized light the deposits show characteristic apple-green birefringence. Similarly immunohistochemistry for serum amyloid P component (SAP) labels the amyloid deposited in the basal lamina of the glands. By electron microscopy (EM), the amyloid is composed of nonbranching fibrils.
Before the protein constituents of amyloidosis were known, the amyloidosis was classified according to clinical presentation. Those cases with no apparent cause were called primary amyloidosis. We now know that these were mostly AL (light chain) amyloidosis caused by an underlying plasma cell proliferative disorder. The amyloidoses developing secondary to underlying chronic inflammatory disorder such as tuberculosis or rheumatoid arthritis was called secondary amyloidosis. We now know that most of these were caused by abnormal deposition of an acute phase reactant, serum amyloid A (SAA) protein, and are now termed as AA amyloidosis. The other important systemic amyloid type is caused by accumulation by transthyretin or prealbumin, now called ATTR. ATTR amyloidosis can be seen sporadically in advanced age, so-called senile amyloidosis, or as part of a germline mutation affecting the TTR gene, so-called hereditary amyloidosis.
Now we know that at least 25 proteins cause amyloidosis. AL, ATTR and AA-type are the most common amyloidosis and account for approximately 80%-85% of all amyloid cases. However, the remaining 10%-15% is caused by much rarer types. Some of these are systemic in nature and are familial/hereditary. Others are localized, caused by abnormal accumulation of locally produced proteins such as amyloid beta protein, precursor in Alzheimer’s disease, or calcitonin in medullary thyroid cancer.
Given the diversity of proteins that could cause amyloidosis and the clinical syndromes that can be associated with amyloidosis, subtyping of amyloidosis is essential for accurate and effective management. For example AL amyloidosis is treated with high-risk strategies such as peripheral blood stem cell transplantation whereas hereditary ATTR amyloidosis may be treated with liver transplantation.
To identify the protein causing the amyloidosis, many practices use clinical parameters and surrogate laboratory studies. These may include serum protein electrophoresis, urine protein electrophoresis, immunofixation, assessment of serum free light chain levels, bone marrow examination and mutation analysis for hereditary amyloidosis. However, a monoclonal paraprotein may be absent in AL amyloidosis up to 15% of cases; a monoclonal paraprotein may be present up to 25% of hereditary amyloidosis, and the presence of a germline mutation in an amyloid-associated gene does not necessarily indicate that it is pathogenic. Therefore, tissue- based typing of amyloidosis is essential for disease management.
For typing of amyloidosis in paraffin-embedded tissues, antibody-based assays using immunohistochemistry have been the gold standard but they have low specificity and sensitivity. This is due to a number of reasons including epitope loss in the physical structure of the amyloid and increased background to nonspecific reactivity. Moreover, immunohistochemistry tests are often not validated specifically for diagnosis of amyloidosis. The interpretation is based on comparison of intensity between independent tests. The tests are not quantitative and, importantly, antibody assays require prior knowledge of possible targets.
To overcome these difficulties, we developed a new methodology to type amyloid deposits in paraffin-embedded tissues. The method uses the precision of laser microdissection and analytical sensitivity and specificity of tandem mass spectrometry (MS). In this method, the amyloid plaques are microdissected by laser from Congo red-stained paraffin sections. Often a single section of a clinical biopsy material (heart, kidney or bone marrow) is enough to obtain results.
For each test, approximately the area of 2 glomeruli would give enough peptide yield to type the amyloid. The microdissected amyloid plaques are digested into peptides using trypsin. The peptide solution is separated by HPLC and sprayed into mass spectrometry using a method called electrospray ionization (ESI). This method ionizes the peptides so that mass spectrometry could detect them. Tandem mass spectrometry then detects mass/charge ratio of the peptides as well as their fragmentation products. These measurements are then interrogated against public human protein databases using software algorithms. The algorithms identify the best peptide/protein matches for a given peptide and its daughter ions.
This is how the method works. The proteins extracted from paraffin-embedded tissue are digested into peptides by trypsin. Trypsin cuts peptides at arginine and lycine residues generating a unique peptide library for each protein. The peptide solution is then separated by HPLC.
The most abundant peptide peaks at a given retention time are then selected and fragmented by collision.
Mass/charge ratio of daughter ions generated by fragmentation of the parent peptide is then measured.
The resulting fragmentation pattern is correlated to theoretical fragmentation patterns of all possible human tryptic peptide sequences from the Swissprot database using specialized search algorithm, and the amino acid sequence of the parent peptide is predicted.
I will now discuss two cases where mass spectrometry-based proteomic analysis was diagnostic. The first case was a 74-year-old male who presented with a long history of CLL and symptoms and signs such as carpal tunnel syndrome and restrictive cardiomyopathy, suggestive of systemic amyloidosis. In 2009, a bone marrow biopsy was performed and at that point, the patient had elevated lambda free light chain levels.
The bone marrow biopsy showed a normocellular marrow with interstitial aggregates of small B cells.
Flow cytometric immunophenotyping showed an abnormal B-cell population expressing CD19, CD20, dim monotypic lambda light chains, CD23, CD5, consistent with CLL.
In addition, the bone marrow periosteum contained eosinophilic amorphous deposits.
These were Congo red-positive and gave apple-green birefringence under polarized light consistent with amyloidosis.
Immunohistochemistry performed for typing amyloid was equivocal and showed that the deposits were positive for SAP, TTR and immunoglobulin lambda light chain but appeared negative for SAA and immunoglobulin kappa light chain.
Mass spectrometry-based proteomic analysis was performed to identify the subtype of amyloid deposits. Congo red-stained sections were visualized under fluorescent light and areas giving characteristic red color were microdissected. Right panel shows the tiny fragment microdissected for mass spectrometry analysis.
This slide shows the results of mass spectrometry analysis performed on tryptic peptides obtained from four different microdissection samples (indicated as samples 1-4) and a negative blank control. The left column shows the list of proteins identified, and the percentages indicate the probability of accurate identification of each protein in each sample.
The most abundant protein identified was serum amyloid P component (SAP), followed by transthyretin (TTR). No immunoglobulin lambda light chains were present. This result indicates ATTR type amyloidosis.
Therefore this patient has CLL and ATTR amyloidosis. No mutations were identified in the TTR gene consistent with age-related or senile ATTR. The patient is being observed with regards to CLL and receiving supportive treatment for cardiac amyloidosis. If a diagnosis of AL amyloidosis was made based on equivocal immunohistochemistry results, the patient may have received unnecessary treatment for the underlying lymphoproliferative disorder.
The second case I will discuss is a 54-year-old female who presented with edema and proteinuria. A renal biopsy was performed to investigate the cause of proteinuria.
The renal biopsy showed amorphous material deposition in the glomeruli which was Congophilic and had characteristic features of amyloid under polarized light as shown in left upper and lower panels. As there were no glomeruli available in the frozen tissue sample for amyloid typing by immunofluorescence, mass spectrometry-based proteomic analysis was performed to identify the subtype of amyloid deposits. Congo red-stained sections were visualized under fluorescent light and areas giving characteristic red color were microdissected as shown in upper right panel. Lower right panel shows a single glomerulus microdissected for mass spectrometry analysis.
This slide shows the results of mass spectrometry analysis performed on tryptic peptides obtained from 4 different microdissection samples (indicated as samples 1-4) and a negative blank control. The left column shows the list of proteins identified, and the numbers in each column indicate the number of spectra identified for each protein in each sample. In addition to proteins present in the glomerular microenvironment, the most abundant protein identified included serum amyloid P component (SAP), immunoglobulin lambda light chain constant region, and immunoglobulin lambda light chain variable region. This result indicates AL (lambda)-type amyloidosis.
Therefore, the patient was diagnosed to have renal amyloidosis, AL-lambda type. Additional clinical investigations showed the presence of an IgG/lambda monoclonal protein and a small population of IgG/lambda restricted plasma cells in the bone marrow. The patient was treated for the underlying plasma cell proliferative disorder to stop progression of amyloidosis.
To establish the clinical sensitivity and specificity of mass spectrometry based proteomic analysis in amyloid classification, we examined 50 cases of systemic amyloidosis. In each case the type of amyloid was characterized by the current gold standard approach including an extensive clinical investigation for plasma cell disorders, serum and genetic testing for amyloidogenic variants of TTR, and immunohistochemistry for TTR, SAA, immunoglobulin kappa and lambda light chains, and SAP. The study included 16 TTR, 9 SAA and 25 AL amyloidosis. In this validation set, mass spectrometry identified the amyloid type with 100% specificity and sensitivity. The details of the study have been published and the citation for further reading is provided in this slide.
In summary, Congo red histochemical stain is the gold standard for diagnosis of amyloidosis. It is important to ensure that Congo red staining performed in the laboratory is sensitive and specific by using a number of appropriate positive and negative controls. The pathologist should have a low threshold for requesting a Congo red staining as early diagnosis and treatment are important for preventing irreversible organ damage. However, the pathologist should have a high threshold for a positive call on Congo red stain to prevent unnecessary further clinical interventions without a firm diagnosis of amyloidosis.
Once the diagnosis of amyloidosis is established, it is essential to facilitate tissue-based typing. As we discussed, this could be done by a number of methods. Immunofluorescence/antibody based assays on frozen tissue provide good sensitivity and specificity for amyloid typing but require fresh/frozen tissue restricting their use in routine diagnosis beyond renal pathology.
In contrast, immunohistochemistry-based assays are readily applicable on most routine paraffin-embedded biopsy specimens but provide low sensitivity and specificity in even the most experienced laboratories. For these reasons, in our practice we recommend mass spectrometry- based proteomic analysis as the main method for amyloid typing.
As always, it is also very important to correlate the mass spectrometry findings with the clinical features to establish the best management approach.