Lossos et al developed a prognostic model based on expression of 6 genes to predict survival in patients with diffuse large B-cell lymphoma. This can be used in conjunction with the International Prognostic Index to risk stratify patients. The authors are from Stanford University, University of Miami and Applied Biosystems (Foster City, California).
Genes in panel:
(1) LMO2 (germinal center)
(2) BCL6 (germinal center)
(3) FN1(lymph node)
(4) CCND2 (activated B-cell)
(5) SCYA3 (activated B-cell)
(6) BCL2 (activated B-cell)
NOTE: Each gene was assayed using quantitative PCR, giving a numeric value. Representative values were not given in the paper, but these should be continuous variables. For the implementation I used discrete values for absent or present, as shown in the following table.
Expression
Points
absent
0
present
1
score =
= (-0.0273 * (points for LMO2)) - (0.2103 * (points for BCL6)) - (0.1878 * (points for FN1)) + (0.0346 * (points for CCND2)) + (0.1888 * (points for SCYA3)) + (0.5527 * (points for BCL2))
Interpretation:
• minimum score: -0.4254
• maximum score: 0.7761
Score
Risk Group
Probability of 10 Year Survival
< -0.063
low risk
73%
-0.063 to 0.0929
medium risk
46%
>= 0.093
high risk
12%
from Figure 2, page 1833
To read more or access our algorithms and calculators, please log in or register.
