DISCUSSION
The outcome of patients with HB changed dramatically after the establishment of Qtx protocols associated with liver resection, pioneered by the SIOPEL11 and COG12 study groups. The 10-y expected overall survival with primary LT was 85%, and 40% in patients who underwent LT as a “rescue therapy”13 in the SIOPEL-1 study. After the initial studies, and others that followed, the absolute indication for LT, and the basis for considering a tumor unresectable after neoadjuvant Qtx, is POST-TEXT IV or POST-TEXT III with major vascular tumor involvement14, 15. Five-year patient survival in the present report is similar to event-free survival, since recurrence meant fatality in almost all patients.
Determining the risk factors associated with patient survival is paramount to an improved selection of patients for LT, but efforts to identify such prognostic factors in the setting of a rare tumor have been hampered by extensive fragmentation resulting in relatively small patient cohorts 16. In the report of the Children’s Hepatic tumors International Collaboration (CHIC), advanced PRETEXT group, macrovascular venous or portal involvement, contiguous extrahepatic disease, primary tumor multifocality and tumor rupture at enrollment, higher age (>8 years), low AFP (<100 ng/ml), and metastatic disease were associated with the worst outcome; however, patients in all stages of HB were included.
Studies in children with HB undergoing LT, despite the limited number of patients enrolled in each cohort, seem to point to the same direction. Umeda et al. 6, in a retrospective analysis of 24 children with unresectable HB submitted to LT, showed that the response to Qtx at LT, evaluated by the decline (>95%) in serum AFP levels, could predict post-LT relapse for patients receiving both primary and rescue LT. Browne et al.17, in a cohort with 14 patients, demonstrated that a drop of 99% in peak AFP levels was associated with 100% survival following LT. Other authors also have correlated the trends in AFP levels18 or pretransplant AFP values with outcome 2, 19. In our study, a drop ≥70% in AFP levels was associated with improved survival at 5-years (81.3% versus 40%). It showed that it is still safe to indicate LT for patients with a more modest, but still significant drop in AFP levels post-Qtx. The ROC curve analysis (Figure 3) presented an AUC of 0.8, demonstrating a good performance/correlation between the studied variable and event-free (event-No group) outcome.
Salvage or rescue LT has been associated with tumor recurrence and worse patient survival since the SIOPEL-1 report13. Many other studies replicated those results7, 17, 20-22, and the present study corroborates worst patient survival with rescue transplantation, where all primary tumors were assumed to be resected during the first operation. Time between diagnosis and LT > 12 months was also associated with decreased patient survival (Figure 2b, 77% <12 mo. vs. 40% >12 mo., P=0.01). Indeed, lengthier Qtx regimens are often required in high-risk subgroups, and are characterized by marked chemoresistance and poor outcome23, 24.
The Cox-Regression analysis and the respective hazard ratios of the three variables (AFP reduction < 70%, time from diagnosis to treatment ≥ 12 months, and rescue LT) associated with increased risk of tumor recurrence/death in this series are shown in Figure 4. Placed together, the observed event-free survival was 84.6% (no factors), 66.7% (1 risk factor), and 32.3% (≥ 2 risk factors). Most importantly, these are pre-operative risk factors that may be able to help families and physicians during the decision-making process, especially in the context of live donation. The advantage of timely LT must be weighted at all times against the risks of the surgical procedure in the living donor and the chances of cure of the children with unresectable HB. In our own experience, Candido et al.25 reported a rate of post-operative complication of 4.8% (29/601) in left liver segment donors used in pediatric LDLT with no patient mortality; however, the estimated rate of donor death “definitely” related to donor surgery has been reported to be 0.15%26.
Vascular invasion has been associated with increased risk of recurrence27-29, and the explant analysis with presence of vascular invasion in the present study was also associated with recurrence/death [patient survival with no vascular invasion (82.4%) vs. microvascular invasion (50%) vs. macrovascular invasion (20%) (p=0.04)]. Most probably, if vascular invasion is detected in the pre-transplant staging (POSTTEXT), it might be useful to include this variable so as to compose a risk stratification to be used with the ones described in this publication; however, the present retrospective analysis did not allow such inclusion/comparison.
The present report shows the experience of a transplant center. The majority of the patients were referred to us from different oncology groups in the country, after the pretransplant chemotherapy regimens had already been defined. Despite the limitation in sample size, one should keep in mind that the study was designed to determine the risk factors for recurrence in a rare disease scenario, over a relatively long period of time. Also, chemotherapy regimens have changed over the study period, even the one used in the PLADO group. The authors recognize these shortcomings but they are inherent of the retrospective nature of this series.
In conclusion, LDLT for HB is a treatment option for unresectable HB, with no distant metastasis and adequate response to Qtx. The following pre-transplant factors - AFP reduction < 70%, time from diagnosis to LT > 12 months and rescue LT - were associated with higher recurrence/death risk and should be critically evaluated and discussed with the patient’s family and the multidisciplinary team in order to move forward with the LDLT in a timely fashion in order to avoid drug toxicity and chemoresistance. However, due to the limited number of patients enrolled in this study, a larger number of patients is required to corroborate these findings.