Better survival with interleukin-2-based regimens? Possibly only in highly selected patients.

Abstract

In this issue of the Journal of Clinical Oncology, Atzpodien et al report interesting but surprising results of a randomized multicenter trial of chemoimmunotherapy in patients with metastatic renal carcinoma. Patients were randomly assigned to receive either a triple combination of subcutaneous interleukin-2 (IL-2) and interferon (IFN) with IV fluorouracil (FU), or the same combination combined with oral 13-cis-retinoic acid (po-13cRA); the control group was treated with IFN combined with vinblastine. The response rates, which were the primary end point of the study, were not different among the treatment groups. Both combinations of FU, IL-2, and IFN, with or without po13cRA, demonstrated a survival advantage compared with the control arm; the largest difference was seen with FU, IL-2, IFN, and po-13cRA. By the survival curves, the largest differences in survival seem to occur about 1 year after treatment starts, but these differences dramatically diminish after 4 to 5 years. These results are surprising for several reasons. First, it is not often that treatments of metastatic cancer lead to an advantage in patient survival without a significant difference in response rates. In fact, the opposite is more common, including in studies of patients with metastatic kidney cancer. This observation is important because the demonstration of a survival gain without significant improvement of response rates means that the main treatment effect is more likely due to tumor growth inhibition than to tumor reduction. Such an effect is generally expected from treatments administered over a prolonged period of time and also from treatments that are mainly active through blocking mechanisms of tumor growth factors. These are not the usual mechanisms of action identified with IL-2 and FU, which seem to be the most important components in these combinations. The median duration of treatments was 2.2 months, which is rather short. Moreover, other authors recently published in this journal a study demonstrating that prolonged exposure to a combination of cytokines failed to improve disease-free or overall survival in patients with metastatic renal cancer. As a consequence, if we analyze the results of this trial, we must consider that the inhibition of tumor growth was a delayed effect of treatment. These results are also unexpected because most previous attempts using combinations of various cytokines alone or cytokines plus chemotherapy have failed to show any survival advantage with IL-2– based regimens. Combinations of IL-2 and IFN were not superior to each cytokine alone or to tamoxifen. Finally, the addition of FU to IL-2 and IFN in another randomized trial did not improve results, compared with the combination of cytokines. Only one previous study from the same group had reported a survival gain in favor of the combination of IL-2, IFN, and FU, compared with tamoxifen. These results, however, were not definitive, because they had been obtained from a limited number of patients. One important difference between the present trial and previous studies is the selection of patients at enrollment; indeed, only favorable risk groups were recruited. The effect of this selection is evidenced by the rather high median survival observed in the control group (16 months), clearly higher than most median survival data previously reported with IFN. It is, therefore, possible that the survival advantages associated with these treatments could only be observed in a similarly favorably selected population. This hypothesis has two major consequences. First, these conclusions may not be applicable to the general population of metastatic renal cancer patients. Second, patients may need to be selected using the rather complex risk score used in this trial, not currently used in routine practice. JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 22 NUMBER 7 APRIL 1 2004