Apoptosis in EBV-associated childhood classical Hodgkin lymphoma

MARIA VICTORIA PRECIADO; CHABAY P

INDIAN JOURNAL OF MEDICAL RESEARCH

INDIAN COUNCIL MEDICAL RES

Lugar: Nueva Dehli; Año: 2009 vol. 130 p. 504 - 505

0971-5916

In this Issue Dinand et al reported the relation of EBV and pro- and anti-apoptotic proteins, with clinical and histological factors, and patients’ survival in a cohort of Indian childhood Hodgkin lymphoma (HL). There have been several reports describing different markers which seem to be a prognostic factor for HL, but since it is a malignancy characterized by clinical, histopathologic and epidemiologic heterogeneity, none of them described an universal marker, applicable to adult and pediatric populations of different socioeconomic status and geographic regions, associated with better overall survival (OS) or failure free survival (FFS). Harris et al 1 have previously described three different epidemiologic patterns according to the level of socioeconomic development. Here Dinand et al reported a histological distribution with mixed cellularity predominance and an early childhood peak (< 10 years) coincident with distribution found by our group 2, 3 as well as by other undeveloped countries and low socioeconomic groups.4 In pediatric HL, it has been reported a male prevalence in different geographic regions, ranging from a slight male predominance in developed countries to 2-3:1 ratio in underdeveloped countries 4. However, in this report the authors have described an extremely low HL incidence among Indian girls, they authors have previously explained this fact on the basis of poor access to health care for females 5. In a previous study the authors have just found the absence of CD15 as a bad predictor for poor overall survival (OS), but other groups have failed to establish such prognostic value 6. In seems that CD15 expression is variable in the different series, which may be due perhaps to biological heterogeneity, making CD15 unsuitable as a universal marker. Epstein-Barr virus (EBV) is associated with a proportion of cases of HL, where the virus is present in the tumor cell populations. The EBV-negative and EBV-positive forms of the disease are morphologically similar but appear to differ in terms of their precise underlying molecular changes and microenvironment. There are epidemiologic differences between EBV positive and -negative HL that raises questions about the etiologic contributions to EBV-defined HL of socioeconomic and cultural exposures and of genetic predisposition. Even though the authors do not applied the gold standard EBERs in situ hybridization, they found by means of LMP1 labeling 91.6% of EBV positivity on HL. This proportion is consistent with prior observations of EBV association in undeveloped countries where most of the EBV-positive cases were besides classified as mixed cellularity 7. However, in our pediatric based studies from Argentina and Brazil we found around 50% EBV positivity resembling developed economies together with higher positivity in mixed cellularity subtype resembling developing ones 3. This discrepancy may reflect that our series may match an intermediate pattern corresponding to developing or transitional economies. Nevertheless, concerning the evaluation of EBV as a prognostic factor, in both situations EBV did not seem to worsen the overall survival or the failure free survival. This implies that EBV could not be used as a prognostic factor in pediatric population, as it was described for older adult HL8. Finally, in this report the authors assessed certain apoptosis-related proteins, since it has been described that germinal center B cells with unfavorable or crippling mutations are normally quickly eliminated by apoptosis. However, those cells that have an imbalance in pro and antiapoptotic proteins ratio could represent a transforming event rescuing preapoptotic GC B cells from apoptotic stimuli and finally could give rise to Hodgkin and Reed Sternberg cells. They described bcl-2 and bak cytoplasmic expression in 83.3% and 92% of the cases together with p53 nuclear expression 89.9% of the cases. As they mentioned, the ratio of bcl-2 family members (bcl-2/bak) rather than any one protein may better determine a cellular response to apoptotic stimuli. But they failed to find a significant relation between this ratio and any survival difference between the categories when OS and FFS were assessed. P53 inactivation is an important cofactor in neoplastic transformation. The authors reported p53 expression was associated with poorer OS in univariate analysis but was not independent prognostic factor in multivariate analysis. According to other similar studies reported in the literature, our group has found in series of 54 pediatric HL a high proportion (81%) of p53 expression, and a much lower (44%) bcl-2 expression 9. When Dinand et al assessed bcl-2, bak and p53 on the context of OS and FFS, they did not provide evidence of being useful as predictive biomarkers of disease progression or therapeutic response. Our group also failed to find bcl-2 and p53 as predictor of event free survival in pediatric HL 9. Therefore, loss of normal function of any of these oncogenes could be the first step leading to neoplastic transformation and proved to be involved in pediatric HL, but they did not provide evidence of being useful as predictive biomarkers of disease progression or therapeutic response. Finally, when the authors related EBV expression and apoptotic proteins they found that in Indian children with HL EBV is significantly associated with lower expression levels of the anti-apoptotic protein bak. Fuller understanding of the pathogenesis of HL and EBV involvement in developing economies, particularly in pediatric population who represent the first peak of the age distribution curve which is delayed to young adults in developed economies, points out the importance of creating a network that allows centers from developing countries devoted to study pediatric HL to cooperate with each other.