Article Text

Download PDFPDF

Anti-titin antibodies are not associated with a specific thymoma histology
  1. R Voltz1,2,
  2. W Albrich1,
  3. R Hohlfeld1,2,
  4. D Nagel3,
  5. M Wick3,
  6. T Kirchner4,
  7. N Sommer5,
  8. I Illa6,
  9. H Kaminski7,
  10. F Schumm8
  1. 1Institute of Clinical Neuroimmunology, Klinikum Gosshadern, München, Germany
  2. 2Department of Neurology, Klinikum Gosshadern, München
  3. 3Department of Clincial Chemistry, Klinikum Gosshadern, München
  4. 4Department of Pathology, University of Erlangen, Germany
  5. 5Department of Neurology, University of Marburg, Germany
  6. 6Servei Neurologia, Hospital Sant Pau, Universitat Autònoma Barcelona, Spain
  7. 7Department of Neurology, Case Western Reserve University, Cleveland, USA
  8. 8Department of Neurology, Christophsbad, Göppingen, Germany
  1. Correspondence to:
    Dr R Voltz;

Statistics from

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.

After the first description of antibodies to titin in patients with thymoma associated myasthenia gravis in 1990, this finding was independently confirmed and the main immunogenic region of titin (mgt30) identified.1 This 30 kDa part of titin is now also commercially available as antigen for detecting anti-titin antibodies. Furthermore, in a series of 276 patients, we could confirm the clinical usefulness of measuring anti-titin antibodies for predicting the presence of a thymic epithelial tumour in patients with myasthenia gravis that was significantly better than the conventional anti-striational antibody test.2 This has been confirmed independently, at least for patients under the age of 60.3 As Marx et al had reported a high titin epitope expression in cortical thymoma and well differentiated thymic carcinoma,4 we were now interested whether there was a correlation between anti-titin antibodies and the histology of the thymic epithelial tumour according to the Müller-Hermelink classification,5 especially whether it might be possible to identify the presence of a carcinoma.

A total of 28 myasthenia gravis patients with pairs of thymoma histology and serum were analysed, 13 from the 1997 study,2 an additional nine from the University of Barcelona (II), five from the Case Western University Cleveland (HK), and one additional patient from the University of Tübingen (NS). No thymoma patients without myasthenia gravis were analysed. As 10 of the first 14 patients were anti-mgt30 antibodies positive using ELISA2 but 11 using western blot, we used western blot for defining the antibody status. Thymoma histology was classified according to the criteria of Müller-Hermelink5 into cortical, medullary or mixed thymoma, or a well differentiated thymic carcinoma. A statistical analysis of the correlation was performed using SAS software (Fisher’s exact test).

There was no significant correlation nor a trend for an association between anti-titin antibodies and thymoma histology. Of the six well differentiated thymic carcinomas, three serum samples (50%) were anti-titin positive, as were 11 of the 16 cortical thymomas (69%). All four mixed thymomas were antibody positive.

The presence of anti-titin antibodies may point towards an underlying thymoma.1–,3 If consistent with radiology, thymectomy is performed also to exclude the presence of an infiltrating thymic carcinoma. As our data now show, titin antibodies are not correlated with thymoma histology and therefore do not add to the presurgical information on the tumour. Why there is no correlation between antibodies and thymoma histology, whether this is attributable to expression of the immunogenic titin epitope in all thymomas or elsewhere independent of the thymoma type, must remain speculation.



  • Competing interests: none declared.