[Assessment of the antitumor effect after remission induction therapy for non-Hodgkin's lymphoma--an approach to a response-adapted therapy].
We assessed the response of various tumors, following each of three consecutive courses of CHOP or CHOP-Bleo therapy in 32 untreated patients with intermediate or high-grade non-Hodgkin's lymphoma (NHL), and also retrospectively compared these results to those for a group in whom complete remission (CR) had been obtained within five courses and a non-CR group. A low CR rate was observed in these patients who showed no or only temporary antitumor effects (persistent increase, no change, or increase after a transient decrease) after two courses of therapy. Furthermore, we measured total tumor volume and tumor regression rate after each course of therapy in 21 patients who had measurable tumors, and determined their cut off values for distinguishing between CR and non-CR. When the cut off values were applied to the effects observed in the 21 patients, positive predictive values after each course of therapy were high. False negative values after the second course of the therapy were low, and showed 0% when the cut off value determined from tumor regression rate was used. These results suggest that we can accurately predict the probability of CR after the second course of CHOP or CHOP-Bleo therapy, although this should be further confirmed by prospective study. In addition, such an analysis may prove useful in setting up individualized response-adapted therapy for NHL.