Lack of bone lesions at diagnosis is associated with inferior outcome in multisystem langerhans cell histiocytosis of childhood

Br J Haematol. 2015 Apr;169(2):241-8. doi: 10.1111/bjh.13271. Epub 2014 Dec 18.

Abstract

Skeletal involvement is generally, but not universally, characteristic of Langerhans cell histiocytosis (LCH). We investigated whether the presence of bone lesions at diagnosis is a prognostic factor for survival in LCH. Nine hundred and thirty-eight children with multisystem (MS) LCH, both high (386 RO+) and low (RO-) risk, were evaluated for bone lesions at diagnosis. Risk organ (RO+) involvement was defined as: haematopoietic system (haemoglobin <100 g/l, and/or white blood cell count <4·0 × 10(9) /l and/or platelet count <100 × 10(9) /l), spleen (>2 cm below the costal margin), liver (>3 cm and/or hypoproteinaemia, hypoalbuminaemia, hyperbilirubinaemia, and/or increased aspartate transaminase/alanine transaminase). Given the general view that prognosis in LCH worsens with increasing extent of disease, the surprising finding was that in MS+RO+ LCH the probability of survival with bone involvement 74 ± 3% (n = 230, 56 events) was reduced to 62 ± 4% (n = 156, 55 events) if this was absent (P = 0·007). An even greater difference was seen in the subgroup of patients with both liver and either haematopoiesis or spleen involvement: 61 ± 5% survival (n = 105; 52 events) if patients had bony lesions, versus 47 ± 5% (n = 111; 39 events) if they did not (P = 0·014). This difference was retained in multivariate analysis (P = 0·048). Although as yet unexplained, we conclude that bone involvement at diagnosis is a previously unrecognized favourable prognostic factor in MS+RO+ LCH.

Keywords: bone lesion; langerhans cell histiocytosis; multisystem disease; risk organ.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Bone and Bones / pathology*
  • Child
  • Child, Preschool
  • Histiocytosis, Langerhans-Cell / mortality*
  • Histiocytosis, Langerhans-Cell / pathology*
  • Humans
  • Infant
  • Prognosis
  • Proportional Hazards Models