AIM: In a previous phase 1B study, we determined the optimal biological dose of interleukin-2 (IL-2) and 13-cis retinoic acid (RA), given as maintenance therapy to patients with a variety of solid tumors, responding to chemotherapy, with a high risk of relapse. This therapy produced a statistically significant increase of the CD4+/CD8+ ratio, natural killer (NK) and lymphocyte cell counts and a decrease of vascular endothelial growth factor (VEGF). The aim of this phase II randomized study was to verify the role of RA in this drug combination. PATIENTS AND METHODS: One hundred and twelve patients, with locally advanced or metastatic tumors responding to chemotherapy, were randomized to receive IL-2, 1.8 x 10(6) I.U. for 5 days/week for 2 consecutive cycles of 3 weeks, with a 1-week interval (arm A), or the same regimen plus oral RA, 0.5 mg/Kg (arm B). VEGF, the CD4+/CD8+ ratio, NK and tumor markers were assessed every 2 months and response every 4 months. RESULTS: The baseline characteristics were well balanced between the two treatment arms for age, performance status, type of disease, amount of previous chemotherapy and baseline values of NK, CD4+/CD8+ and VEGF. Toxicity was minor in both arms. After a median follow-up of 42 months, all immunological parameters improved in both arms with respect to the baseline values; this improvement was statistically more significant in arm B. There was no statistically significant difference in progression-free and in overall survival between the two arms. CONCLUSION: These data show that low-dose IL-2 and oral RA is more effective than IL-2 alone in improving all known prognostically significant parameters in a variety of solid tumors, including an increase of lymphocytes and a decrease of VEGF.