A recent study sponsored by the National Cancer Institute has demonstrated that the particular patterns of genes that are active in tumor cells can be used to predict the odds that patients with diffuse large B-cell lymphoma (DLBCL) will survive after treatment with chemotherapy. Known as gene expression profiling, this method could help physicians to make more informed treatment decisions, armed with an idea of how likely standard therapy is to be effective or whether other options, such as bone marrow transplant, should be considered.

Researchers involved in the study identified 17 genes associated with survival from DLBCL, the most common type of non-Hodgkin’s lymphoma among adults, that could be used to develop a molecular predictor of risk. Researchers expect that this number is small enough to allow this method to be easily used in routine practice. The study involved examining tumor biopsies from 240 patients with DLBCL and comparing over 600 genes for differences between those who responded well to treatment and those who did not. The result was that the most marked differences were seen in 17 genes, which were then used to create a formula to break down predicted survival into four groups. The five-year survival rates for those groups were 73%, 71%, 34%, and 15%.

Researchers hope that their findings will offer an improved method of evaluating patients with this form of cancer. Currently, the International Prognostic Index (IPI), which is based on clinical factors like age and tumor stage, is used but has not been very useful for predicting patient response to particular therapies. In this study, 32 of the patients were classified according to IPI as having the worst prognosis; however, 4 of these patients were able to be cured by standard chemotherapy – and the gene expression profiling was successful in identifying each patient.