NPC-0501 trial on the value of changing chemoradiotherapy sequence, replacing 5-fluorouracil with capecitabine, and altering fractionation for patients with advanced nasopharyngeal carcinoma.

BACKGROUND: A current recommendation for the treatment of patients with locoregionally advanced nasopharyngeal carcinoma (NPC) is conventional fractionated radiotherapy (RT) with concurrent cisplatin followed by adjuvant cisplatin and 5-fluorouracil (PF). This randomized NPC-0501 trial evaluated the therapeutic effect of changing to an induction-concurrent sequence or accelerated-fractionation sequence, and/or replacing 5-fluorouracil with capecitabine (X). METHODS: Patients with American Joint Committee on Cancer/International Union Against Cancer stage III to stage IVB NPC initially were randomly allocated to 1 of 6 treatment arms (6-arm full-randomization cohort). The protocol was amended in 2009 to permit centers to opt out of randomization regarding fractionation (3-arm chemotherapy cohort). RESULTS: A total of 803 patients were accrued (1 of whom was nonevaluable) from 2006 to 2012. Based on the overall comparisons, neither changing the chemotherapy sequence nor accelerated fractionation improved treatment outcome. However, secondary analyses demonstrated that when adjusted for RT parameters and other significant factors, the induction-concurrent sequence, especially the induction-PX regimen, achieved significant improvements in progression-free survival (PFS) and overall survival. Efficacy varied among different RT groups: although no impact was observed in the accelerated-fractionation group and the 3-arm chemotherapy cohort, a comparison of the induction-concurrent versus concurrent-adjuvant sequence in the conventional-fractionation group demonstrated a significant benefit in PFS (78% vs 62% at 5 years; P = .015) and a marginal benefit in overall survival (84% vs 72%; P = .042) after adjusting for multiple comparisons. Comparison of the induction-PX versus the adjuvant-PF regimen demonstrated better PFS (78% vs 62%; P = .027) without an increase in overall late toxicity. CONCLUSIONS: For patients irradiated using conventional fractionation, changing the chemotherapy sequence from a concurrent-adjuvant to an induction-concurrent sequence, particularly using induction cisplatin and capecitabine, potentially could improve efficacy without an adverse impact on late toxicity. However, further validation is needed for confirmation of these findings.

Lipiodol versus diaphragm in 4D-CBCT-guided stereotactic radiotherapy of hepatocellular carcinomas.

PURPOSE: The purpose of this work was to investigate the potential of lipiodol as a direct tumor surrogate alternative to the diaphragm surrogate on four-dimensional cone-beam computed tomography (4D-CBCT) image guidance for stereotactic radiotherapy of hepatocellular carcinomas. METHODS: A total of 29 hepatocellular carcinomas (HCC) patients treated by stereotactic radiotherapy following transarterial chemoembolization (TACE) with homogeneous or partial defective lipiodol retention were included. In all, 4-7 pretreatment 4D-CBCT scans were selected for each patient. For each scan, either lipiodol or the diaphragm was used for 4D registration. Resulting lipiodol/diaphragm motion ranges and position errors relative to the reconstructed midventilation images were analyzed to obtain the motion variations, and group mean (ΔM), systematic (Σ), and random (σ) errors of the treatment setup. RESULTS: Of the lipiodolized tumors, 55 % qualified for direct localization on the 4D-CBCT. Significant correlations of lipiodol and diaphragm positions were found in the left-right (LR), craniocaudal (CC), and anteroposterior (AP) directions. ΔM and σ obtained with lipiodol and diaphragm were similar, agreed to within 0.5 mm in the LR and AP, and 0.3 mm in the CC directions, and Σ differed by 1.4 (LR), 1.1 (CC), and 0.6 (AP) mm. Variations of diaphragm motion range > 5 mm were not observed with lipiodol and in one patient with diaphragm. The margin required for the tumor prediction error using the diaphragm surrogate was 6.7 (LR), 11.7 (CC), and 4.1 (AP) mm. CONCLUSION: Image-guidance combining lipiodol with 4D-CBCT enabled accurate localization of HCC and thus margin reduction. A major limitation was the degraded lipiodol contrast on 4D-CBCT.