| Chromosome
Arm 11q Deletion Predicts for Neuroblastoma Outcome: A Children's
Oncology Group Study
Authors: E. F. Attiyeh, Y. P. Mosse, Q. Wang, C.
Winter, D. Khazi, G. Hii, P. W. McGrady, K. K. Matthay, W. B. London,
J. M. Maris
Background: Chromosome arm 11q loss of heterozygosity
(LOH) is a frequent occurrence in neuroblastoma. Previous studies
have shown an association between 11q LOH and high-risk disease.
However, these studies lacked the statistical power to conclusively
analyze the prognostic impact of 11q LOH in multivariable analyses
after adjusting for the currently used prognostic factors.
Methods: We screened 917 primary neuroblastomas
for chromosome 11 LOH with a panel of 4-8 microsatellite markers.
Further mapping with up to 52 markers was performed on selected
samples. The only inclusion criteria were availability of a constitutional
DNA sample and outcome data. The sample set was representative of
the overall neuroblastoma population. The median follow-up was 2.7
years.
Results: LOH at chromosome arm 11q and unbalanced
11q LOH (loss of 11q material with retention or gain of 11p material)
were present in 302 (33%) and 148 (16%) cases, respectively. The
pattern of LOH for each tumor sample was consistent with the presence
of a single region of deletion; all but two deletions involved 11q23.
The majority of tumors with 11q LOH did not have MYCN amplification.
Despite this, both 11q LOH and unbalanced 11q LOH were associated
with high-risk disease (p<0.0001). Patients whose tumors showed
unbalanced 11q LOH had 3-year EFS and OS of 50%±6% and 65%±5%, compared
to 74%±2% (p<0.0001) and 83%±2% (p<0.0001) in those cases that did
not (log-rank test). In a Cox proportional hazards model for multivariable
analysis, unbalanced 11q LOH was independently prognostic for EFS
after adjusting for stage 4, MYCN amplification, and Shimada histology
(p=0.0068).
Conclusions: Unbalanced 11q LOH is associated with
aggressive neuroblastoma behavior and is independently prognostic
for decreased EFS. Unbalanced 11q LOH status could be especially
useful in determining whether a patient with locoregional neuroblastoma
without MYCN amplification would benefit from more adjuvant chemotherapy.
| Multivariable analysis |
| Variable |
Relative Risk |
p-value |
| INSS Stage |
4 2.23 |
<0.0001 |
| MYCN amplification |
2.42 |
<0.0001 |
| Unfavorable Shimada histology |
2.50 |
<0.0001 |
| Unbalanced 11q LOH |
1.60 |
0.0068 |
| Age > 1 year |
Not significant |
| DNA index = 1 |
Not significant |
| 1p LOH |
Not significant |
|
