4. Discussion
In this retrospective multicenter study of 130 patients with HB in the CHIC-HS, the very low- and low-risk groups underwent surgery with curative intent and were followed up for >5 years. We investigated the effect of NACT on the 5-year OS and EFS. According to our results, there was no survival benefit to using NACT before surgery in terms of EFS and OS. Pathological classification was a risk factor for adverse events and mortality; three (30%) patients with MEM pathology experienced tumor recurrence and one (10%) patient died. One (4%) patient with EMEF pathology had tumor recurrence and remains alive to date, while two (8%) patients with EMEF pathology died from other factors. All SCU patients died of progressive disease after relapse, and no adverse events or deaths were observed in patients with PFH pathology. Moreover, pathological alterations were observed after NACT.
Because of the rarity of HB and the fact that only 30% of patients had resectable tumors at the time of diagnosis, it was difficult to perform a prospective study, and few studies have addressed which patients benefit the most from upfront surgery. Currently, surgeons mainly judge whether the tumor is resectable based on subjective evaluation and experience. For PRETEXT stage III patients, most surgeons usually choose to receive NACT to make surgery easier, even when tumors are resectable at the time of diagnosis. Therefore, very few patients with PRETEXT stage III HB were included in this study. The SIOPEL group reported that patients with resectable tumors using NACT before surgery had less surgical complications and better outcomes. Therefore, they recommended that all patients receive NACT before surgery 4, resulting in further reduction in the number of patients undergoing upfront surgery. Establishing objective patient selection criteria for US is paramount, and the COG argues for decreasing the total cumulative administered dose of cisplatin to protect patients from hearing loss6,12. The JPLT-2 study showed PRETEXT stage I patients and some PRETEXT stage II patients without positive annotation factors who underwent US and had good outcomes (5-year EFS and OS: 74.2% and 89.9%, respectively) 5. The COG study achieved similar outcomes; the 5-year EFS and OS of patients with PFH pathology were 100% 12, and in PRETEXT stage I or II patients without PFH and SCU pathology, the 5-year EFS and OS were 88% and 91%, respectively 6. However, there was no comparison between the two approaches in these studies, and differences in the inclusion criteria made it difficult to compare these results.
To standardize surgical decision-making, an objective and comparable evaluation method was used to determine the optimal patients for upfront surgery. We used the risk-stratified staging developed by the CHIC to explore whether patients with HB in the very low-risk and low-risk groups benefited from upfront surgery. In this study we verify whether the CHIC-HS can be used to screen patients with HB for tumors that can be resected at the time of diagnosis. According to our data, both patient groups had a relatively favorable prognosis, and there was no significant difference in the 5-year OS and EFS between the groups. This suggests that US can achieve long-term disease control in these patients, which can decrease cisplatin chemoresistance and reduce the total chemotherapy dose.
NC has some effects on surgical operations, which can lead to tumor shrinkage and downstaging and the tumor shrinking further away from the blood vessels 13. This approach can reduce the risk of intraoperative bleeding and other complications, making surgery easier to perform. However, after the tumor shrinks, the tissues around the tumor often show an abnormal shape when relieving compression from the tumor. Pathological examination showed that these tissues are usually liver parenchyma, and some surgeons choose to retain this part of the tissue to obtain a larger residual liver volume. Our results showed that this did not affect the surgical outcome. There was no statistical difference in the rate of R1 resection between the groups, and R1 resection had no effect on the 5-year OS and EFS (Table 3), which are consistent with the findings of previous studies14,15.
Notably, we found that, among the pathologic changes induced by NACT, some EMEF tumors presented mature mesenchymal tissues (Table S1). This finding is consistent with those reported by Stephen et al.16, and “maturation” of malignant clones or the result of selective ablation of immature clones were thought to predict a better prognosis 17. However, a recent study showed that patients with MEM pathology were more likely to experience recurrence and metastasis because they had low sensitivity to chemotherapy, but no prechemotherapy pathology was available in that study 18. Consequently, it was difficult to compare these two studies. In our research, the rate of recurrence and metastasis in patients with MEM and SCU pathology was higher than that in other patients, but no conclusion could be drawn due to the small sample size. The prognostic effect of pathological classification changes due to NC and effect of pathological classification on prognosis in different risk groups merit further investigation.
The study limitations include its retrospective nature and small sample size, that NACT regimens and the number of cycles of NACT varied across institutions, and our failure to evaluate tumor response to chemotherapy and acute and long-term toxicities of platinum-based chemotherapy, especially ototoxicity.
In the future, it will be necessary to explore whether more patients with HB can benefit from upfront resection, similar to our study, and the impact of preoperative chemotherapy in altering pathologic type upon prognosis. Experiments need to be designed to evaluate the chemotherapy resistance of tumors and the degree of hearing loss in patients. Owing to the rarity of HB, further studies with a larger sample size and multicentric samples are needed.
In conclusion, our findings suggested that upfront resection can achieve long-term disease control in patients with HB and resectable tumors at diagnosis in the CHIC-HS very low- and low-risk groups. This treatment approach can reduce the cumulative toxicity of platinum-based chemotherapy drugs, including in PRETEXT stage III patients. SCU patients had poorly differentiated tumors and a poor prognosis; treatment of these patients is controversial, and the optimal therapies need further investigation. This HB risk stratification system provides an objective criterion to evaluate whether patients are suitable for upfront resection. Our findings may help future clinical studies to explore whether more patients with HB can benefit from upfront resection.