Article Text

Download PDFPDF

Letter
Increased [11C]PIB-PET levels in inclusion body myositis are indicative of amyloid β deposition
  1. Walter Maetzler1,2,
  2. Matthias Reimold3,
  3. Jens Schittenhelm4,
  4. Matthias Vorgerd5,
  5. Antje Bornemann4,
  6. Ina Kötter6,
  7. Christina Pfannenberg7,
  8. Gerald Reischl8,
  9. Ludger Schöls1
  1. 1Hertie Institute for Clinical Brain Research, Department of Neurodegenerative Diseases, University of Tuebingen, Germany
  2. 2DZNE, German Center for Neurodegenerative Diseases, University of Tuebingen, Germany
  3. 3Department of Nuclear Medicine and PET Centre, University of Tuebingen, Germany
  4. 4Institute of Brain Research, Neuropathology, University of Tuebingen, Germany
  5. 5Department of Neurology, Bergmannsheil, Ruhr-University Bochum, Germany
  6. 6Department of Internal Medicine II, University of Tuebingen, Germany
  7. 7Department of Diagnostic Radiology and PET Centre, University of Tuebingen, Germany
  8. 8Radiopharmacy, PET Centre, University of Tuebingen, Germany
  1. Correspondence to Dr G Reischl, Radiopharmacy, PET Centre, University of Tuebingen, Roentgenweg 15, Tuebingen 72076, Germany; gerald.reischl{at}uni-tuebingen.de

Statistics from Altmetric.com

Introduction

Pittsburgh Compound B ([11C]PIB), a 11C-labelled substituted benzothiazole, is a positron emission tomography (PET) marker recently described to detect amyloid β in vivo in the brains of patients with Alzheimer's disease.1 As amyloid β is also accumulated and misfolded in sporadic inclusion body myositis (IBM), and PIB has been demonstrated to penetrate the cell membrane,2 it is assumed that [11C]PIB may have the potential to depict amyloid β in the skeletal muscles of IBM patients.

Methods

Thirteen subjects with clinical symptoms suggestive of IBM underwent complete clinical workup, including electrophysiology, muscle biopsy and [11C]PIB-PET. Biopsy specimens underwent standard histopathological staining. Based on clinical, electrophysiological and standard histopathological criteria, seven subjects were diagnosed as having IBM, two as having polymyositis, two as having neurogenic muscle atrophy, one as having peripheral neuropathy and one as having myalgia of unclear aetiology.

Demographic and clinical details as well as biopsy results can be provided by the authors on request. The analysis was approved by the ethics committee of the University of Tuebingen. All subjects gave written informed consent.

Twenty minutes after intravenous bolus injection of 740MBq [11C]PIB, whole body distribution of radioactivity was measured with a Hi-Rez Biograph 16 (Siemens Medical Solutions, Knoxville, USA), consisting of a high resolution three dimensional LSO PET and a 16 row multi slice CT. PET acquisition time was 3 min per field of view. Patients were measured from the lower leg to the neck to enable interindividual comparison of [11C]PIB uptake. Whole body CT was performed with low radiation dose used for attenuation correction and anatomical co-registration. Reading of [11C]PIB-PET was done by at least two nuclear medicine physicians, looking for asymmetry and focally enhanced uptake within skeletal muscles in colour coded whole body [11C]PIB-PET images. For retrospective analysis, three-dimensional elliptoid regions of interest were placed over four different skeletal muscles per side—that is, a proximal and distal muscle of the upper and lower extremity (see supplementary table available online). The average [11C]PIB concentration was expressed in standardised uptake values (SUV, local radioactivity concentration divided by administered radioactivity per body mass (g/ml) or unitless). PET analysis was performed blind to clinical diagnosis and to PIB histology. The inter-rater variability (MR, WM) was <10 %.

From paraffin sections of the biopsy specimens, PIB and amyloid β staining (6E10, 1:1000 dilution; Abcam) was performed. For PIB staining, non-radioactive PIB was synthesised and muscle sections were treated as previously described.3 In patient IBM-1, only routine diagnostics were performed as no tissue was available for additional staining. Negative and positive control slides (frontal cortex of healthy and Alzheimer diseased brain) were processed in parallel. All analyses were made blind to the [11C]PIB-PET results and to routine diagnosis.

Results

All non-IBM subjects presented with [11C]PIB-SUV levels below 0.5 (highest 0.48—the deltoid muscles of subject non-IBM-4). All patients that were classified as ‘IBM’ after the clinical workup, including standard histopathological parameters, presented at least in one investigated muscle with [11C]PIB-SUV levels above 0.5. Six of seven IBM patients showed PIB-SUV levels of 0.6 and higher in at least one gastrocnemius muscle, and the median [11C]PIB-SUV of the gastrocnemius muscles was significantly higher in IBM patients than in non-IBM subjects (Wilcoxon rank sum test, p=0.004). In three IBM patients, [11C]PIB-SUV levels ≥0.5 were also found in additional muscles such as the vastus lateralis, deltoid and long finger flexor muscles. Details are supplied in the supplementary table (available online).

Figure 1 shows PET/CT images (SUV) of the lower legs of patient non-IBM-6 (this subject had the highest gastrocnemius [11C]PIB-SUV levels among the non-IBM subjects; SUVgastrocnemius=0.47/0.45, left/right) and an IBM subject (IBM-6, SUVgastrocnemius=0.68/0.73).

Figure 1

Pittsburgh Compound B (PIB) uptake and histological findings in crural muscles. Patient non-inclusion body myositis (IBM)-6 had the highest gastrocnemius [11C]PIB-standardised uptake values (SUV) within the non-IBM group (0.45) and displayed increased [11C]PIB binding in medium sized and large blood vessels but not in muscular regions of interest. Muscle biopsy (indicated by the red square) revealed no PIB and no amyloid β staining, and normal routine staining. Patient IBM-5 showed a control-like [11C]PIB-SUV level in the muscle that was chosen for biopsy (right gastrocnemius, 0.40) and no specific PIB staining in the biopsy specimen. Amyloid β staining displayed diffuse deposition. Singular muscle fibres were positive for ubiquitin and tau. Haematoxylin–eosin (HE) staining revealed mononuclear cell invasion. Patient IBM-6 showed high [11C]PIB-SUV (0.73) with corresponding histopathology of the right gastrocnemius muscle (red square). PIB staining revealed dense inclusions in muscle fibres, and staining for amyloid β also displayed dense deposition (areas indicated by rectangles are shown at higher magnifications at the very right border of the figure). Comparable with subject IBM-5, singular muscle fibres were positive for ubiquitin and tau, and HE staining revealed mononuclear cell invasion. Sections from the frontal cortex of a subject with Alzheimer's disease served as (positive) controls. Scale bar=40 μm.

Biopsies were made in different muscles (gastrocnemius, vastus lateralis, deltoid). In two patients, muscle biopsies were available for muscles with a [11C]PIB-SUV of 0.7 or higher. Gastrocnemius muscles of patient IBM-3 (SUVgastrocnemius=0.70) and patient IBM-6 (SUVgastrocnemius=0.73) showed several fibres of small diameter with dense amyloid β and PIB positive inclusions (figure 1).

Biopsies of muscles with [11C]PIB-SUV below 0.7 showed no dense or demarcated inclusions in either staining. The only exception was the right deltoid muscle of patient IBM-7 where amyloid β positive muscle fibres (but not PIB positive inclusions) were detectable.

Discussion

Here we report significantly increased PIB uptake levels in the gastrocnemius muscle of IBM patients and dense PIB positive muscle fibre inclusions in muscles with [11C]PIB-SUV levels of 0.7 or higher. This suggests that [11C]PIB-PET may depict muscular amyloid β deposits in vivo. Conversely, clinically severely affected muscles did not show increased [11C]PIB binding on PET/CT and no PIB staining within muscle fibres. The reason for this phenomenon is unclear. It may be argued that dense amyloid β containing inclusions in IBM are rather protective, which would fit with recent findings in Alzheimer's disease research.4 In addition, as shown in figure 1 (adjacent sections of IBM-6), PIB may have a high affinity only to a subset of amyloid fibril conformations or complexes. This raises the question of whether the PIB positive structures are those which are related to clinical outcome.

In our approach, PET/CT was used for whole body imaging to investigate four different muscles according to typical IBM associated clinical signs. This is of relevance as misfolded protein deposition in IBM may be patchy and restricted, and not be displayed in every muscle biopsy.5 The downside of this approach is that the PET protocol is not optimised with respect to quantitative imaging. Therefore, due to different uptake intervals, a comparison of uptake values between different regions of a subject (eg, arms, legs) should, if at all, be done with care. In our eyes, this does not question the described group differences in selective muscles as all subjects were investigated with the same PET protocol (ie, consecutive acquisition with 3 min per field of view from the lower leg to the neck). In future studies, one may choose to investigate selected regions with both imaging and quantitative (immuno)histochemical methods (eg, of gastrocnemius muscles) and in the PET studies focus on the pharmacokinetic analysis of the full time course of tracer delivery and washout. In addition, due to the number of included subjects, our findings must be considered as preliminary and further studies in a larger number of participants is needed to further examine and confirm these findings.

Acknowledgments

The authors wish to thank all of the patients who took part in the study.

References

View Abstract

Footnotes

  • Competing interests None.

  • Ethics approval The study was conducted with the approval of the ethical commission of the Medical Faculty of the University of Tuebingen.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.