PURPOSE We report here the outcomes and late effects of the Japanese Study Group for Pediatric Liver Tumors (JPLT)-2 protocol, on the basis of cisplatin-tetrahydropyranyl-adriamycin (CITA) with risk stratification according to the pretreatment extent of disease (PRETEXT) classification for hepatoblastoma (HB). PATIENTS AND METHODS From 1999 to 2012, 361 patients with untreated HB were enrolled. PRETEXT I/II patients were treated with up-front resection, followed by low-dose CITA (stratum 1) or received low-dose CITA, followed by surgery and postoperative chemotherapy (stratum 2). In the remaining patients, after 2 cycles of CITA, responders received the CITA regimen before resection (stratum 3), and nonresponders were switched to ifosfamide, pirarubicin, etoposide, and carboplatin (ITEC; stratum 4). Intensified chemotherapeutic regimens with autologous hematopoietic stem-cell transplantation (SCT) after resection were an optional treatment for patients with refractory/metastatic disease. RESULTS The 5-year event-free and overall survival rates of HB patients were 74.2% and 89.9%, respectively, for stratum 1, 84.8% and 90.8%%, respectively, for stratum 2, 71.6% and 85.9%%, respectively, for stratum 3, and 59.1% and 67.3%%, respectively, for stratum 4. The outcomes for CITA responders were significantly better than those for nonresponders, whose outcomes remained poor despite salvage therapy with a second-line ITEC regimen or SCT. The late effects, ototoxicity, cardiotoxicity, and delayed growth, occurred in 61, 18, and 47 patients, respectively. Thirteen secondary malignant neoplasms (SMNs), including 10 leukemia, occurred, correlating with higher exposure to pirarubicin and younger age at diagnosis. CONCLUSION The JPLT-2 protocol achieved up-front resectability in PRETEXT I/II patients with no annotation factors, and satisfactory survival in patients who were CITA responders in the remaining patients. However, outcomes for CITA nonresponders were unsatisfactory, despite therapy intensification with ITEC regimens and SCT. JPLT-2 had a relatively low incidence of cardiotoxicity but high rates of SMNs.
