In patients with diffuse large cell lymphoma treated with chemotherapy the presence of high levels of serum beta 2 microglobulin has been considered as a bad prognostic factor. Until now, attempts to detect early relapse in patients with diffuse large cell lymphoma have been sparse. To address this issue we began a prospective clinical trial to evaluate the role of different clinical, laboratory and radiographic tests in the detection of early relapse in non-Hodgkin's lymphoma (NHL). Only serum beta 2 microglobulin levels had clinical significance and 26 of 53 patients (49%) had abnormal levels, 3 to 23.1 months (mean 8.5 months) before evident relapse. Elevated serum lactic dehydrogenase (LDH) levels and beta 2 microglobulin were observed in six patients and all relapsed, suggesting that the combination of these two tests should be considered in future prospective clinical trials in order to define the utility of both tests to detect early relapse. This information may allow us to begin chemotherapy when the tumor mass is still low thereby making the probability of achieving a long second remission more likely.

The optimal management of primary gastric lymphoma (PGL) remains undecided because a definitive classification for therapeutic decision is not available. The International Index Project has proved to be useful in patients with nodal disease, but in extranodal presentation it has not been tested. We reviewed 297 patients with early stage PGL. They were initially classified according to the prognostic features of the International Index Project. No influence on duration of time to treatment failure (TTF) or overall survival was observed. For this reason we developed a logistical model to identify prognostic factors in patients with early stage PGL. Levels of beta-2 microglobulin and lactic dehydrogenase were observed to have prognostic significance in both univariate and multivariate analysis. With these parameters we constructed a logistical model to identify patients at low risk (TTF = 76%; at 7 years overall survival was 96%), statistically different to patients at high risk (TTF = 34% and overall survival = 22%). The number of patients at intermediate risk were too small to compare with the other groups. Because pathological or other clinical or laboratory prognostic features cannot help in the identification of a prognostic model, we propose that the use beta-2 microglobulin and lactic dehydrogenase can define different groups at risk and develop a prognostic system to define the best therapeutic approach in this patients.