Clonality in Langerhans' cell histiocytosis

BMJ 1995; 310 doi: https://doi.org/10.1136/bmj.310.6972.74 (Published 14 January 1995) Cite this as: BMJ 1995;310:74
  1. Finbarr E Cotter,
  2. Jon Pritchard
  1. Senior lecturer in haematology and oncology Haematology and Oncology Unit, Institute of Child Health, London WC1N 1EH
  2. Consultant in paediatric oncology Great Ormond Street Hospital, London WC1N 3JH

    Women are more informative than men

    Most human tumours are monoclonal, which suggests that they originate from a single cell. “Clonality” can be investigated by several techniques. In plasma a dominant monoclonal class of immunoglobulin suggests a neoplasm derived from a single altered plasma cell, which gives it a survival and growth advantage over non-neoplastic cells. In lymphoid malignancies, studies on rearrangements of genes coding for antigen receptors provide an excellent system of clonal markers, with consistent rearrangement of clonal immunoglobulin heavy chains (in B cell leukaemias and lymphomas)1 and T cell receptor ß chains (in T cell tumours). In other tumours careful cytogenetic and molecular analysis has shown consistent reciprocal chromosomal translocations and partial and complete chromosomal deletions or duplications which reflect tumour clonality as well as indicating stretches of altered sequences of DNA that may have a role in the initial neoplastic change. Cytogenetic and DNA analysis has not yet, however, provided clonal markers for many solid tumours.2

    An alternative approach, applicable only to females, is to take advantage of the phenomenon of “Lyonisation” (X inactivation mosaicism).3 4 Early in the development of female embryos one X chromosome of each pair is randomly inactivated in each somatic cell. Some enzymes occur in two forms (isoenzymes)—if they are encoded by the X chromosome one form but not the other is represented in each cell. The result of clonal expansion of a tumour cell reflects this event—only one form of the isoenzyme is expressed, as shown by Fialkow by using isoenzymes of glucose-6-phosphate dehydrogenase.3 These X inactivation studies have now been extended by molecular techniques, including polymerase chain reaction,5 and examination of very small samples of tissue is now possible.

    Applying this approach, Willman and colleagues have clearly shown the clonal nature of Langerhans' cell histiocytosis (histiocytosis X).6 The disorder, first described over 100 years ago,7 had previously been known as Hand-Schuller-Christian disease, eosinophilic granuloma of bone, Letterer-Siwe disease, and self healing reticulocytosis. Abnormal Langerhans-like cells are pathognomonic of the condition, and they and their normal counterpart (the Langerhans histiocyte, which presents antigen) are rare in expressing the CD1a antigen. The clinical presentation and course of Langerhans cell histiocytosis vary from a spontaneously regressing solitary lytic lesion of bone to an intermediate form with lesions in bone, skin, mucous membranes, liver, and spleen with variable organ dysfunction to a lethal leukaemia-like disorder most often seen in infants.8

    This variation has led to much speculation about the nature of the condition. A viral cause has been suggested, but ultrastructural and serological studies yield negative results,9 10 and no consistent evidence for an immune deficiency or an abnormal immunological response has been found. The unexpected and intriguing finding of Willman and colleagues was that all 10 lesions from patients with Langerhans' cell histiocytosis whom they studied contained clonal populations of cells, with the proportion of clonal cells corresponding to the proportion of lesional Langerhans-like cells, whether from solitary lesions or extensive multisystem disease. The clonal nature of Langerhans' cell histiocytosis from the initiation of the disease process suggests that this is a true tumour and not a non-specific condition that predisposes to the development of neoplasia. The initial lesion may provide the necessary clues to the primary molecular event leading to tumorigenesis. Yu et al last year reported similar findings in tissues enriched by flow sorting of CD la positive cells from three patients.11 Sequential studies, to show that the “clonality” pattern is constant throughout the course of the disease, are needed, but these observations still provide the first solid clue to defining the aetiology of the condition in 100 years.

    Somatic mutation

    The hypothesis that Langerhans' cell histiocytosis arises from somatic mutation of DNA in a normal Langerhans' cell or precursor cell, leading to a neoplastic phenotype, must now be seriously considered, even though studies of clonality have not shown mutation. A single mutation that provides a growth advantage does not necessarily equate with malignancy, but the increased proliferation of cells may predispose to additional somatic mutations12 and evolution in some cases to cancer.

    Throughout this process the abnormal Langerhans'-like cells would retain the same X linked clonal marker. Varying clinical outcomes would depend on the number of mutations acquired, possibly on the role of immune surveillance and on the micro environment where the mutated cells are sited or to which they migrate. Additional gene mutations might cause clonal evolution of the disease from a benign to a more malignant invasive form.

    Alone, monoclonality is necessary but not sufficient to define neoplasia. Benign parathyroid adenomas may be clonal,13 and lymphoproliferation associated with immunodeficiency may move through a benign polyclonal proliferative disease to a monoclonal process, as shown by rearrangements of immunoglobulin heavy chains, eventually progressing to a malignant phenotype.14

    Molecular technologies for examining complex genome changes, such as fluorescent in situ hybridisation and comparative genome mapping,15 are now at a stage at which they could be applied to Langerhans' cell histiocytosis and provide clues to the molecular pathology of this disorder. Willman and her colleagues have presented a powerful demonstration of the use of clonal analysis and an important direction for future research into Langerhans' cell histiocytosis.6 Application of X linked analysis to other disorders of histiocytes will be informative. In research to improve the prediction of prognosis and to suggest treatments samples of female tissue will be at a premium.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    View Abstract

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
    Sign up for a free trial