Clinical and laboratorial prediction of bone marrow involvement in children and adolescents with Hodgkin Lymphoma.

BACKGROUND

A risk group (RG) system to predict bone marrow involvement (BMI) based on clinical and laboratory parameters was previously shown to have prognostic value in adult patients with Hodgkin lymphoma (HL). Our aim was to test the applicability of BMI RG in an independent group of childhood and adolescent HL patients.

PROCEDURE

Seventy-eight HL patients (range 3-18, median 14 years old) were retrospectively studied, including revision of histopathological diagnosis and bilateral BM biopsies. Patients were divided into BMI high-, standard- and low-RG.

RESULTS

The high-RG included 29.5% of the patients, the standard-RG, 32%, and the low-RG, 38.5%. All the patients in the high-RG had stage III or IV. The five children (6.4%) with BMI were included in the high-RG (P = 0.001, chi(2) test) and had stage IV (P < 0.00001, chi(2) test). The BMI risk group out-performed the clinico-pathologic risk assessment in predicting BMI, showing 100% sensitivity, 75% specificity and a positive predictive value of 0.88. Neither BMI nor risk of BMI was statistically associated to age groups (</=14 years old or >14 years old), suggesting that age is not a risk factor for BMI in the pediatric population.

CONCLUSIONS

The application of BMI RG score was able to efficiently foresee BMI in a pediatric group of HL patients, adding independent information to clinico-pathologic BMI risk assessment.