Stage III—IV

Stage III—IV

significance. These include age >60 years, tumor stage, B symptoms, serum LDH level, and the presence of more than one extranodal site of involvement. The IPI is essentially a scoring system in which these five clinical features are tallied to categorize patients into one of four prognostic groups that correlate with both relapse-free survival (RFS) and overall survival (OS) (Table 54.2). Moreover, patients with low-risk disease consistently attain complete response more frequently than the patients with higher risk disease. In addition to providing prognostic guidance at an individual patient level, the IPI provides a common language for clinical trials to allow homogeneity when comparing patient groups. For now, the IPI remains the most clinically useful tool to predict prognosis.

Recent research highlights the clinical and molecular heterogeneity of this most common form of NHL even further, via gene expression profiles and microar-ray technology. Microarray analysis of gene expression involves the simultaneous evaluation of thousands of genes by hybridizing complementary DNA to the mRNA in a cell of interest. In an elegant pilot study by Alizadeh and colleagues,16 normal and malignant lymphoid cells were analyzed on a "lymphochip" microar-ray that included nearly 18,000 genes of interest. These genes were all of putative importance in normal B-cell development and/or in the progression to neoplasia. The authors identified two molecular signatures, germinal center B cell like (GC) and activated B cell like (ABC). They then applied the lymphochip to 40 patients with previously untreated DLBCL and compared the genetic signature with the clinical outcome. The OS for all patients at 5 years was 52%; however, the OS for patients with the GC pattern of gene expression was 76% as compared to only 16% for patients with the ABC pattern (p < 0.01). Even more powerful is their finding that patients within the low-risk IPI group could be further separated, in terms of prognosis, by the gene expression pattern. In summary, their data suggest that DLBCL comprises at least two distinct diseases differentiated by the expression pattern of hundreds of genes. A larger multicenter study confirms these findings, validates the significance of these two molecular signatures, and identifies a probable third signature (labeled "Type 3") of genetic importance.17

Most recently, Lossos and colleagues evaluated 36 genes identified by microarray analysis that predicted survival.18 Using quantitative real-time polymerase chain reaction (RT-PCR) in 66 frozen patient samples, they ranked the genes by their ability to predict survival and identified the six most powerful genes: LMO2, BCL6, FN1, CCND2, SCYA3, and BCL2. They then developed a prognostic model based on the weighted expression of these six genes, which is potentially a simpler approach to determine prognosis than is the use of microarrays. It is likely that similar reports will eventually allow further ability to predict survival at an individual patient level using readily available techniques such as immunohistochemistry or flow cytometry. In addition to its potential clinical utility, microarray technology also aids in the understanding of lymphoma biology by identifying genes and genetic pathways that determine growth rate, responsiveness to therapy, or other important clinical parameters.

0 0

Post a comment