There are insufficient data on the outcomes and their clinical predictors in pediatric
patients with lymphoblastic lymphomas (LBLs). Molecular genetics of pediatric LBLs
and acute lymphoblastic leukemia (ALL) overlap to some extent. Risk stratification-directed
treatment is the standard of care in the treatment of ALL, and such a risk stratification
model has not been developed for the clinical use in LBL patients. To bridge this
gap, a retrospective analysis of 65 consecutive patients with pediatric LBL over 15
years, treated with BFM90 protocol, was carried out and the outcomes were assessed
in terms of overall survival (OS) and event-free survival (EFS).[1]
At the end of induction, 57% of patients had a complete remission, 28% had a partial
remission, 6% had a progressive disease, and 3% had a stable disease. Treatment-related
mortality occurred in three patients overall and 20 patients relapsed. After a median
follow-up of 54.57 months, 10-year EFS was 62% and OS was 71%. These outcomes were
15%–20% lower than those reported from developed countries.[2]
[3]
On multivariate analysis, three predictors of poor OS and EFS were identified – symptom
duration <30 days, white blood cell count >12,000/cumm, and serum albumin <3.5 g/dl.
A prognostic model was devised which showed that OS was 92% if no risk factors were
present versus 37% if two or three risk factors were present. Twenty percent of patients
had two or three risk factors and had poor outcomes both as a result of increased
toxicity, higher relapses, and higher deaths (3 out of 14 in the presence of risk
factors vs. none in the other group).
Age, sex, presence of mediastinal mass, bulky disease, stage of disease, and elevated
serum lactate dehydrogenase were earlier investigated as prognostic markers with variable
results. However, the study population and the results were heterogeneous. As majority
of patients (90%) of LBL have Stage III/IV disease, stratification of patients into
localized and disseminated disease based on stage is not a useful measure in making
treatment decisions.
Among novel prognostic markers, molecular profile, minimal residual disease (MRD),
minimal disseminated disease (MDD), and use of positron emission tomography (PET)
scan were studied in LBLs. Among the molecular markers, the presence of NOTCH1 and
FBXW7 mutations was associated with favorable prognosis, while loss of heterozygosity
at 6q and PTEN mutations were associated with adverse prognosis. Based on these markers,
a prognostic genetic classifier for risk group stratification for T-cell LBL has been
proposed.[4] MRD was studied as a significant risk factor for relapse. Patients with detectable
disease in bone marrow, i.e., MDD by flow cytometry (0.01%), had varying prognosis
corresponding to the levels of MDD present.[5] Postinduction PET-computed tomography (CT) has been shown to improve the ability
to detect residual disease, thereby affecting the treatment decisions.[6]
To conclude, the current study identified useful clinical predictors, which are readily
available, inexpensive, and easy to use, that can help predict the outcomes of the
patients. This study identifies the need for tailored therapy as per the prognostic
score which can help reduce treatment-related mortality and relapses. In future, comprehensive,
multicentric, prospective trial incorporating both clinical predictors and novel markers
such as molecular profile, MRD, MDD, and use of PET-CT is required to further investigate
on a risk-oriented approach in LBL.
