Histological and proteomic characterization of musculoskeletal amyloidomas

Introduction. The term amyloidoma applies to localized deposits of amyloid in the absence of systemic amyloidosis. Skeletal and soft tissue amyloidomas are very rare and the pathogenesis is usually associated with lymphoproliferative disorders (plasmacytomas or plasmacytoid lymphomas) or as a consequence of local chronic inflammation.
Methods. In this paper we report the histological and immunohistochemical features of four cases of musculoskeletal amyloidoma in association with combined laser capture microdissection (LCM) of Congo Red positive regions with a recent microproteomics workflow that improves the sensitivity of the analysis in order to confirm the nature of the protein deposit.
Results. Proteomic techniques allowed to elucidate the nature of the amyloid protein deposit, improving the results obtained by immunohistochemistry (IHC). IHC results were confirmed in two cases while LCM coupled with bottom-up microproteomics was necessary to type the other two cases, for which IHC was inconclusive.
Conclusions. In conclusion, proteomic techniques were thus confirmed as a fundamental tool for the complete investigation of protein deposits.

Introduction

The term amyloidoma applies to localized deposits of amyloid in the absence of systemic amyloidosis. Amyloidomas affecting the skeleton and the soft tissues are very rare and the clinical presentation as mass lesions frequently leads to the initial suspicion of a neoplasm. Bone involvement occurs more frequently in the axial skeleton, mainly in the thoracic spine and skull ¹, while soft tissue amyloidomas localize more frequently in the mediastinum and retroperitoneum, as well as in the extremities ². Regarding their pathogenesis, bone and soft tissue amyloidomas of the mediastinum or abdomen are frequently associated with lymphoproliferative disorders (plasmacytomas or plasmacytoid lymphomas), whereas soft tissue amyloidomas of the extremities may arise as a consequence of local chronic inflammation ^1,2. The histopathologic diagnosis requires the identification and characterization of the amyloid, and this may be difficult because amyloid deposits may be obscured by inflammation and giant cell foreign body reaction or may undergo calcification, bony metaplasia or present unusual morphology, including the formation of corpora amylacea-like structures with concentric lamination.

Antibody-based techniques for amyloid typing are widely used in clinical practice ³. However these techniques have some limitations, linked to the non-specificity of the antibodies to the mutated or truncated forms of amyloid proteins and with the non-specific interactions of the antibody with fibrils ⁴. In the last decade mass spectrometry based bottom-up proteomics, along with electron microscopy, has become one of the reference techniques for amyloid typing ^4,5 and several protocols have been developed ^6-8. Bottom-up proteomics is based on the proteolytic digestion of proteins and on the separation and analysis of the resulting proteolytic peptides by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) ^9,10. Peptides are identified by matching their MS/MS spectra to in-silico simulated peptide spectra calculated from curated protein databases. The identified peptides are then merged to provide information at the protein level, both in terms of identity and quantity ^11,12. Proteomics can be coupled with laser capture microdissection (LCM) of Congo Red (CR) stained sections to focus the analysis on amyloid-containing regions ⁶. LCM enables the precise dissection of specific tissue regions under a microscope, to isolate these regions-of-interest, and thereby to process them for proteomics analysis using liquid-based protocols ¹³. Here we report the clinico-pathologic features of 4 cases of musculo-skeletal amyloidoma that were studied by combined LCM of CR-positive regions with a recently reported microproteomics workflow ¹⁴ that further improves the sensitivity of the analysis.

Patients and methods

PATIENTS

The salient clinical features of the 4 cases of musculoskeletal amyloidoma analyzed in this study are summarized in Table I.

MATERIALS

Histology-grade solvents and Congo Red were purchased from DiaPath S.p.A. (Martinengo, Italy). SuperFrost Plus glass microscope slides were purchased from VWR International, LLC (Radnor, PA, USA). LC-MS grade water, acetonitrile (ACN) and formic acid (FA) were purchased from Thermo Fisher Scientific (Rockford, IL, USA). Phosphoric acid was purchased from Merck (Darmstadt, Germany). Adhesive Cap 200 tubes were obtained from Zeiss (Oberkochen, Germany). All the other reagents were purchased from Sigma-Aldrich (Saint Louis, MO, USA).

IMMUNOHISTOCHEMISTRY

Immunohistochemical analysis was conducted on formalin-fixed paraffin-embedded (FFPE) full tumor tissue sections. Five mm sections were deparaffinized, hydrated and after endogenous peroxidase inactivation, immunostained with BenchMark^® Ultra stainer (Ventana, Tucson, AZ, USA). The following primary antibodies were employed: CD68 (clone KP1, Roche Diagnostics, Rotkreuz, Switzerland); kappa and lambda light chain (polyclonal, Roche Diagnostics, Rotkreuz, Switzerland); MDM2 (clone IF2, Invitrogen, Rockford, IL, USA); CD138/syndecan-1 (clone B-A38, Cell Marque, Darmstadt, Germany); MUM1 (clone EP190, Cell Marque, Darmstadt, Germany). The reaction was revealed with iVIEW DAB detection kit, providing a brown reaction product. After completing the staining process, the slides were removed from the autostainer, counterstained with hematoxylin, dehydrated and mounted with a permanent medium. As negative control, we substituted primary antibody with a Ventana dispenser filled with non-immune serum at the same concentration for each immunohistochemical reaction.

LASER CAPTURE MICRODISSECTION

Tissue sections, 8 μm thickness, were cut using a Leica RM2245 microtome (Leica, Wetzlar, Germany) and mounted onto microscope slides using a Leica HI 1210 water warm bath (Leica). The tissue sections were heated at 60°C for one hour to increase adherence, then submerged in xylene twice for 15 minutes to remove paraffin, and finally stained with hematoxylin followed by Congo Red. The stained tissue sections were then mounted onto a Zeiss Axio Observer Z1 microscope equipped with a PALM Microbeam laser and a RoboMover (both Zeiss, Oberkochen, Germany) for LCM. Amyloid areas were detected in fluorescence mode (HPX 120C lamp, 43 DSRed filter, Zeiss) and confirmed under brightfield using the PALMRobo software (v 4.9, Zeiss). The microdissection laser was operated in AutoLPC mode, with a distance between spots of 10 μm and a distance from the border of 5 μm. Fragments were collected in the cap of Adhesive Cap tubes and stored at -20°C. The plaque from one sample (case 3) was manually collected from the glass slide using the tip of a pipette, since the tissue was fragile and detached from the slide. The average area isolated by LCM for each sample was approximately 0.25 mm².

PROTEOMICS SAMPLE PREPARATION

The lysis buffer, 35 μL of 10 mM Tris/ 1mM EDTA/5% SDS, was added to the tube caps containing the small pieces of tissue isolated by LCM. Blank samples were processed alongside the samples by adding the lysis buffer to empty Adhesive Cap tubes. Solutions were transferred to a clean tube. The tissue and solvent solutions were heated to 98°C for 90 min with occasional vortexing to effect heat-aided antigen retrieval. Samples were then sonicated using a Bioruptor Pico (Diagenode, Liège, Belgium) to lyse the tissues (15 cycles, 30 sec ON, 30 sec OFF), followed by heating to 95°C for 10 min. Proteins were reduced with dithiothreitol (final concentration 20 mM) and incubated at 45°C for 30 min. Protein alkylation was performed with iodoacetamide (IAA, final concentration 40mM) for 30 min at RT in the dark. Excess IAA was quenched by addition of another aliquot of DTT (final concentration 40mM). Proteins were then digested using Micro S-traps (ProtiFI, Farmingdale, NY, USA) as follows: A ratio of 1:10 v/v of 12% phosphoric acid was added to each sample, followed by 400 μL of S-trap binding buffer (90% MeOH, final TEAB concentration of 100 mM, pH 7.1). The protein extracts of the LCM samples were loaded on the S-trap columns and centrifuged at 4000xg for 1 min and the flow-through discarded. Multiple centrifugations were performed to load all the protein extract onto the S-trap column. Bound proteins were washed four times with 150 μL of S-trap binding buffer. Digestion was performed by loading 0.5 μg of trypsin/LysC (Promega, Madison, WI, USA) in 20 μL of digestion buffer (50 mM TEAB, pH 8) and incubating at 37°C for 18 hours. Peptides were eluted in three steps with 40 μL of 50 mM TEAB, 40 μL 0.2% FA and 35 μL of 50% ACN, each followed by a centrifugation at 4000xg for 1 min. Peptides were dried and resuspended in 10% formic acid and stored at -20°C until LC-MS/MS analysis.

PROTEOMICS ANALYSIS

One tenth of the peptide amount was loaded onto an EASY-nLC 1000 (Thermo Scientific) equipped with an Acclaim PepMap 100 pre-column (2 cm x 75 μm, C18, 3 μm, 100 Å; Thermo Scientific). Peptides were separated with a 60 min gradient on an EASY-Spray analytical column (ES803: 50 cm x 75 μm, C18, 2 μm, 100 Å; Thermo Scientific) and analyzed using an Orbitrap Fusion Tribrid mass spectrometer (Thermo Scientific, San Diego, USA).

Peptide ions were analyzed using the Top Speed data dependent method, with a 3 sec cycle time; MS1 scans were performed in the Orbitrap (m/z 375 to 1500 at 120K resolution with an AGC Target 5x10⁵ and 100 msec maximum injection time) and MS2 scans acquired in the ion trap using a 1.6 m/z isolation window, 30% HCD Collision Energy and an AGC target of 5x10³.

Raw data were analyzed using Proteome Discoverer (v.2.1, Thermo Scientific) and searched against the SwissProt Homo sapiens database (Uniprot, 21/10/2019, 20365 entries). An in-house contaminant database was added to the search (250 entries). Searches were performed with a precursor mass tolerance of 10 ppm and using a strict FDR of 0.01. A maximum of two missed cleavages were allowed. Methionine oxidation (+15.995 Da) and acetylation (+42.01 Da, protein N-terminus) were set as dynamic modifications while carbamidomethylation of cysteine (+57.021 Da) was set as a static modification. Protein Groups were filtered by eliminating protein identified in the contaminant database and precursor ion intensities were exported to GraphPad Prism for bar chart plotting (version 5, GraphPad Softwares, Inc, San Diego, CA). An online tool ¹⁵ was used to compare the Protein Groups between case 1 and 2. Enrichment analysis was performed in String ¹⁶ (v11.5) using the whole Homo sapiens database as background.

Results

HISTOPATHOLOGIC FINDINGS

Two lesions (cases 2 and 3) presented with the usual histological appearance of amyloidoma and consisted of diffuse depositions of amorphous eosinophilic material, with focal giant cell reaction and with a sparse focal lymphoplasmacytic infiltrate (Fig. 1B). The deposits within the medullary bone were well circumscribed and surrounded by thickened trabeculae (Fig. 1A). In both cases, a uniform population of mature appearing plasma cells was present between the amyloid deposits. These were positive for immunohistochemical staining of kappa light chain and negative for lambda light chain.

In case 2 amyloid was deposited over the bony trabeculae, sometimes surrounding them, and undergoing mineralization as well as osseous metaplasia (Fig. 2A). In other areas amyloid deposits consisted in part of rounded structures reminiscent of corpora amylacea, which frequently displayed a concentric lamination and partial calcification (Fig. 2B).

In case 1 and 4, the soft tissue lesion consisted of lobules of mature adipose tissue separated by collagenous septa and by amorphous eosinophilic extracellular matrix. This was present in small amounts among adipocytes or accumulated in bands or large deposits without intervening adipocytic elements. In addition, in case 1, several scattered atypical cells, both mono and multinucleated, were present within the hyaline extracellular matrix, as well as near mature adipocytes (Fig. 3A). These cells appeared as spindle elements with central hyperchromatic nucleus, or as larger polygonal cells often with two or three nuclei. This appearance mimicked quite closely that of an adipocytic neoplasm. Scattered clusters of foreign body giant cells and scant infiltrates of mature-appearing lymphocytes and plasma cells were also detected (Fig. 3B-C). The adipocytes showed moderate variation in shape and dimension, but no atypical nuclear features were observed. However, the possibility of atypical lipomatous tumor was considered in the histologic differential diagnosis, but was excluded based on a negative staining for MDM2.

In all cases, amyloid deposits were positive for Congo-red with apple-green birefringence. Immunohistochemistry showed positivity for kappa light chain in all cases except for case 1, that was negative for both kappa and lambda light chains (Tab. I). In addition, plasma cells were positive for CD138 and MUM1, whereas histiocytes and multinucleated foreign body-type giant cells were CD68 positive.

PROTEIN IDENTIFICATION VIA BOTTOM-UP PROTEOMICS

LCM of the amyloid regions, as defined by Congo Red positivity, was performed on FFPE patient tissues obtained from the four amyloidoma cases. Table II shows the size of the areas isolated by LCM, and the number of protein groups identified by LC-MS/MS.

Quantitative microproteomics analysis of the proteins extracted from the isolated amyloidoma regions led to the identification of 212 to 911 proteins groups (Supporting Material 1). The levels of the 30 most abundant proteins from each patient are shown in Figure 4. For cases 1-3 the most abundant protein was an immunoglobulin, while for case 4 keratin and hemoglobin chains were more abundant than the amylogenic protein, probably indicating ambient/blood contamination.

The 30 most abundant proteins in the amyloidoma region of case 1 (Fig. 4A) included 5 immunoglobulin light chains (IGLV6-57, IGLC2, IGKV3-20, IGKC, IGKV2-40) and one heavy chain (IgG-1 heavy chain). The most abundant protein group, IGLV6-57 (Immunoglobulin lambda variable 6-57), is the variable domain of the λ6 light chain immunoglobulin ¹⁷; its very high level indicates a local AL-lambda amyloidoma. IGLV6-57 was identified with 103 MS/MS spectra that matched to a single unique peptide, and which represents 17% sequence coverage. The second most abundant amyloid protein was the IgG-1 heavy chain, whose intensity was just 3.4% of the intensity of IGLV6-57.

For both cases 2 and 3 (Fig. 4B-C) the most abundant protein group is IGKC, the constant region of immunoglobulin κ light chains, indicating an AL-kappa amyloidoma. IGKC was identified in case 2 with 110 MS/MS spectra matching to 7 peptides (all unique), and which represented 80% sequence coverage. IGKC was identified in case 3 by 112 MS/MS spectra matching 8 peptides, all of them assigned either to IGKC chain or to immunoglobulin kappa light chain (P0DOX7), covering 84% of the protein sequence. In case 4 the most abundant amylogenic protein is IGKC, which was detected as the 6^th most abundant protein in the dataset, identified through 49 MS/MS spectra matching to 7 peptides (two of which unique to the protein), representing 80% sequence coverage (Fig 4D). The plaque was thus determined to be an AL-kappa amyloidoma. The sample from patient 4 contained high levels of keratin and hemoglobin chains, indicative of contamination due to handling and blood respectively, and was presumably the reason why IGKC was only the 6^th most abundant protein identified.

Discussion

Amyloidomas or amyloid tumors are nodular tumor-like masses of amyloid that may occur at various body sites, but their localization in the skeleton and soft tissues of the extremities is exceedingly rare ¹⁸. Such tumoral deposits often lead to the suspicion of a neoplastic process, both clinically and histologically, and the diagnosis may be challenging.

In the present series, two amyloidomas arising in bone represented massive amyloid deposition associated with multiple myeloma, and one localized in the subcutaneous soft tissues of the leg was associated with Sjogren syndrome and chronic hepatic disease. In the remaining case involving the soft tissues of the popliteal fossa there was no clear evidence of systemic or chronic disease. Although rare, this occurrence has been previously reported in the literature ¹⁹. In these cases, it has been hypothesized that the plasma cell clone may be obscured by the amyloid deposits and difficult to recognize, and amyloid deposits with no or sparse plasma cell infiltrate may indeed represent a ‘burnt-out’ plasmacytoma ²⁰.

Histologically, the cases reported here show remarkable similarities to those previously described and required careful evaluation to exclude a neoplasm. Amyloid deposits that assume a tumor-like clinical presentation most often are accompanied by an intense inflammatory reaction with foreign body giant cells that may obscure the amyloid substance, necessitating a distinction from giant cell rich tumors, including tenosynovial giant cell tumor, giant cell tumor of the soft tissues, or osteoclastic giant cell-rich tumors of bone. Notably, case 1 of the present series resembled quite closely an adipocytic neoplasm both clinically and histologically. Indeed, amyloid deposits were distributed mainly between the adipocytes of the subcutaneous fat, and less prominently with a solid nodular appearance. In addition, several “atypical” cells were present within the amyloid substance, thus forming a picture that mimicked quite closely the appearance of atypical lipomatous tumor/well differentiated liposarcoma. Most of these cells were CD68 positive histiocytes, the remaining negative elements likely being fibroblasts and myofibroblasts. Atypical lipomatous tumor/well differentiated liposarcoma can be further excluded with a negative staining for MDM2 and CDK4 or with a search for the amplification of the 12q13-15 chromosome region using FISH or other molecular methods.

LCM coupled with microproteomics was used to confirm the nature of the protein deposit. A comparison of the results obtained from mass spectrometry-based typing of amyloidomas with those obtained via immunohistochemistry is shown in Table III. Cases 2-4 displayed high concordance with immunohistochemistry analysis, but case 1 could only be typed by mass spectrometry. Case 1 was determined to be an AL-lambda amyloidoma by mass spectrometry, whereas IHC was negative for both kappa and lambda chains. This result might be due to the lower sensitivity of IHC compared to LC-MS/MS based proteomics. AL is the most common amyloid deposit in systemic ²¹ and localized amyloidosis ²² and is generally associated with plasma cell dyscrasia or multiple myeloma ²¹. The λ6 family is overrepresented among amyloid proteins ²³; IGLV6-57 in particular is known to form amyloid fibrils ^24-26. Constant regions of light chain immunoglobulins have been shown to deposit alongside variable domains in amyloid plaques ²⁷ and have a role in fibril aggregation ^28,29. The lack of identification of the variable domains is likely due to the somatic mutations of the variable sequences, which were not present in the FASTA database used for the database search ³⁰. IHC of cases 2 and 3, both determined to be AL-kappa amyloidomas by mass spectrometry, were positive for kappa chains and negative for lambda chains, fully confirming the MS-based typing. IHC of tissues from case 4 were positive for both kappa and lambda chains, whereas the MS-based analysis determined the amyloidoma type to be AL-kappa. This difference in typing may be due to the blood contamination of the sample from case 4, leading to false positive immunohistochemical positivity for AL-lambda.

In general, proteomics showed concordance with IHC, and elucidated the nature of the protein deposit in the dubious IHC cases 1 and 4. Several proteins known to deposit within the amyloid plaque were also among the thirty most abundant proteins, including the “amyloid signature” ^21,26,31 proteins apolipoprotein E (APOE), apolipoprotein A-IV (APOA4) and serum amyloid-P component (APSC). APOA1 is another amyloid precursor ²¹; APOA1, together with APOA4, APOE and APSC, have been detected in many cases of λ light chain amyloidomas ³².

Calcification and/or ossification may occur within the amyloid produced by plasmacytomas ^1,33,34. Radiographically, these tumors present with lytic expansile destruction of bone, often associated with an internal calcified or osseous matrix, an appearance that may simulate the appearance of chondrosarcoma ^1,33. In these cases, amyloid deposits assume the appearance of rounded bodies with a vague concentric lamination resembling corpora amylacea or undergo calcification or osseous metaplasia. To investigate possible biological pathways linked with calcification, proteins identified from case 2 (calcified femur amyloidoma) were compared with proteins from case 3 (normal appearance femur amyloidoma). Gene Ontology analysis was performed on protein identified exclusively from case 1, but no significant enrichment in bone formation or calcification-related pathways was observed (Supporting Material 2). The extremely low number of samples and the restriction of the proteomic analysis to the amyloid plaque region do not allow to draw any conclusion on the nature of the calcification detected in case 2.

In conclusion, four cases of amyloidoma were described by histopathological examination flanked by IHC and proteomics typing of the amyloid deposit. Proteomics techniques allowed to elucidate the nature of the amyloid protein deposit, improving the results obtained by immunohistochemistry. IHC results were confirmed in two cases while LCM coupled with bottom-up microproteomics was necessary to type the other two cases, for which IHC was inconclusive. Proteomic techniques were thus confirmed as a fundamental tool for the complete investigation of protein deposits.

CONFLICT OF INTEREST STATEMENT

The authors report there are no competing interests to declare.

FUNDING

This research received no external funding.

ETHICCAL CONSIDERATION

This study is covered by ethical vota of the medical faculty of the University of Pisa for retrospective translational research activities. An informed consent was collected from all patients. The study was approved by the ethics committee ‘Comitato Etico Regionale per la Sperimentazione Clinica della Regione Toscana, Sezione: Area Vasta Nord Ovest’ (protocol n. 16037). All experiments, which are routinely performed for diagnostic purposes, were performed in accordance with the principle of Good Clinical Practice (GCP) and the ethical principles contained in the current version of the Declaration of Helsinki.

AUTHOR CONTRIBUTIONS

F.G. and R.G. performed study concept and design; F.G., F.A., and L.McD. performed development of methodology; F.G., R.G., A.F. and L.McD. performed writing, review and revision of the paper; F.G., R.G., A.F., R.C. and L.A. provided acquisition, analysis and interpretation of data. All authors read and approved the final paper.

DATA AVAILABILITY

The data that support the findings of this study are openly available in PRIDE repository ³⁵ at https://www.ebi.ac.uk/pride/archive, reference number PXD031789.

ABBREVIATIONS

SUPPORTING MATERIAL

Supporting Material 1. List of identified protein groups in cases 1-4.

Supporting Material 2. List of protein groups identified exclusively in case 1, exclusively in case 2 or in common between case 1 and 2. Enrichment analysis of protein groups exclusively identified in case 1 (Cellular component and KEGG pathways).

History

Received: October 23, 2024

Accepted: November 13, 2024

Figures and tables

Figure 1. Amyloidoma of bone. The medullary spaces are occupied by deposits of amyloid substance, with marked thickening of bony trabeculae (A) (H&E, 2.5X). Foreign body-type multinucleated giant cells are detected within the amyloid (B) (H&E, 20X).

Figure 2. Amyloidoma of bone. In this case the amyloid appears both as amorphous eosinophilic and calcified masses (A) (H&E, 2,5X). In other areas, amyloid deposits present as rounded structures reminiscent of corpora amylacea (B) (H&E, 4X).

Figure 3. Amyloidoma of soft tissues. Amyloid deposits are distributed mainly between the adipocytes of the subcutaneous fat (A) (H&E, 2.5X). Several “atypical” cells are present within the amyloid substance (B) (H&E, 40X), which were further characterized as CD68 positive histiocytes. Sparse aggregates of lymphocytes are present within the amyloid substance (C) (H&E, 4X).

Figure 4. The 30 most abundant protein groups identified by LC-MS/MS based proteomics of the Congo-Red positive regions. Bar plots showing the relative intensity of the protein groups, with respect to the most abundant protein in the dataset, for cases 1-4 (A-D). Protein groups highlighted in orange are known to form amyloid plaques, protein groups highlighted in yellow are known to co-precipitate in the amyloid plaque. Hemoglobin chains are highlighted in red.

Table I. Summary of clinical data, proteomics analysis, immunohistochemical findings and histopathologic diagnosis of the 4 patients.

Case	Sex, Age	Site	Clinical background	Treatment	Abundant protein	Tentative MS-based diagnosis	Immunohistochemical findings	Histopathologic diagnosis
1	F 62	Soft tissues, leg	Sjogren syndrome, chronic liver disease	Intralesional resection	IGLV6-57	AL-λ chains	CD68+, MDM2-, CD138-, kappa-, lambda-	Soft tissue amyloidoma
2	F 83	Proximal femur and diaphysis	Multiple myeloma (monoclonal for Kappa chain)	Resection of proximal femur, prosthesis	IGKC	AL-k chains	MUM1+, CD138+, kappa+, lambda-, CD68-	Amyloidoma in multiple myeloma (kappa chain monoclonal)
3	M 58	Proximal femur	Multiple myeloma (monoclonal for Kappa chain)	Resection of proximal femur, prosthesis	IGKC	AL-k chains	kappa+, lambda-	Amyloidoma in multiple myeloma (kappa chain monoclonal)
4	F 65	Soft tissues, popliteal fossa	Type II diabetes mellitus; hypertension; no evidence of chronic inflammatory diseases or plasma cell proliferative disorders	Intralesional resection	IGKC	AL-k chains	CD68+, CD138+, kappa+, lambda+	Soft tissue amyloidoma

Table II. Areas isolated by LCM and number of protein groups identified in samples from cases 1-4.

Case	Isolated plaque area	# protein groups
1	0.208 mm2	365
2	0.131 mm2	911
3	> 0.4 mm2	267
4	0.385 mm2	212

Table III. Summary of MS-based and IHC-based amyloidoma typing for cases 1-4.

Case	MS-based diagnosis	Most abundant protein group	IHC-based diagnosis
1	AL-lambda	IGLV6-57	kappa-, lambda-
2	AL-kappa	IGKC	kappa+, lambda-
3	AL-kappa	IGKC	kappa+, lambda-
4	AL-kappa	IGKC	kappa+, lambda+