Previously, a phase I study determined the MTD of axitinib monotherapy was 5 mg twice daily (29). Therefore, in the phase II and phase III trials for advanced RCC with progression after first-line treatment (26, 30), the staring dose of axitinib was set as 5 mg twice daily. It was mentioned the axitinib dose could be adjusted according to individual tolerance within the range from 2 mg twice daily to 10 mg twice daily. However, the safety of axitinib in combination with RT was not yet established before our study. We had not known the potential interaction between axitinib and RT. Following a principle of the best safety, the starting dose of axitinib in our present phase I study was set as a minimal practicable dose of 1 mg twice daily, and the dose would be escalated by a relatively safe dose interval when moving to the next dose levels. In addition, we intended to deliver isotoxic and safe RT to normal liver with similar mean liver doses approaching 18 Gy for each patient. Although this caused heterogeneous prescribed RT dose in our study, this would be a necessary measure to make RT toxicities relatively constant and could enable appropriate evaluation of tolerability regarding axitinib MTD in combination with RT.
Our study successfully proved that axitinib in combination with RT is safe at least up to the dose of axitinib 3 mg twice daily, which was considered as the MTD in this study. The dose is already within the recommended dose range of axitinib: 2 mg twice daily to 10 mg twice daily. According to our data, no additional toxicities were induced by the combination of RT and axitinib. All the AEs did not exceed grade 3 and were all manageable. We did not further escalate the dose because we had only limited resources for this study. If any other study groups want to conduct another similar phase I study, a starting dose with axitinib 3 mg twice daily can be considered. A determined MTD will facilitate design of a phase II study evaluating efficacy.
HCC is known as a hypervascular cancer supplied majorly from hepatic artery, while normal liver tissue is supplied by portal vein. Therefore, transarterial therapies are relatively specific for HCC rather than normal liver. Medical anti-angiogenic agents also proved clinical therapeutic value for HCC. Combination of some anti-angiogenic agents and RT showed potential benefit at both pre-clinical and clinical level (31–34). A pre-clinical study demonstrated that axitinib can improve tumor control of RT by radiosensitizing tumor endothelial cells (35). Anti-angiogenesis may allow better maturation of cancer blood vessels, and could potentially improve tumor oxygenation and thus tumoricidal effect of RT (36, 37). In our study, the response rate of axitinib in combination with RT is encouraging compared with RT alone. High rate of complete response was observed. This could be contributed by radiosensitization effect from addition of axitinib.
Spatial cooperation may exist between local treatment (e.g. RT) and systemic therapy. In addition to extrahepatic progression, many patients with advanced HCC treated by sorafenib eventually also progressed in liver, with a median time to progression 5.5 months according to SHARP study (19). On the other hand, from the experience of RT, outfield recurrence is the most common site of first recurrence. RT in combination with effective systemic therapy possibly exert the effect of spatial cooperation which may be translated to improved PFS and even OS. Our present study showed an acceptable PFS and impressive OS for advanced HCC treated with the combination strategy. However, because the phase I trial enrolled only small numbers of patients, it is difficult to conclude that axitinib in combination with RT could offer benefits compared with RT alone based on the data presented. Further phase II or even phase III study is required to adequately evaluate the efficacy.
Clinical experiences with RT and anti-angiogenic agent are few but still exist with encouraging results. For example, one retrospective study treated advanced HCC with RT and sunitinib (another multi-kinase inhibitor), reported objective response rate of 74% and a median survival of 16 months, and concluded hypofractionated RT can be delivered safely concomitantly with sunitinib (38), which was compatible with the result of several phase I or II studies using sorafenib plus RT (25, 39). In our present study, combining axitinib and RT were safe and effective with results comparable to other study with similar combination approach.
The use of anti-HBV medicines during RT can prevent re-activation of hepatitis B virus. In a phase II clinical trial, the combination of sorafenib and RT carries fair local regional control, but still suggested that the combination therapy should be used with caution due to a relatively higher hepatic toxicity (25). Lacking routine use of anti-HBV medicine may be at least part of the reason that the study encountered the issue. In our study, we prespecified the use of anti-HBV medicine for any patient with confirmed hepatitis B. We did not observe worse liver toxicity nor re-activation of hepatitis B.
Since regorafenib and lenvatinib were both proved as effective treatment for HCC, the combination of RT with these relatively new agents could also be studied in the setting of clinical trial. Adverse effects caused by regorafenib are serious concern because a substantial portion of HCC patients cannot well tolerate even regorafenib monotherapy. Lenvatinib could be a better candidate to try a combination treatment with RT because many patients can better tolerate lenvatinib monotherapy as compared with sorafenib. Several other new treatments for advanced HCC emerge recently, including ramucirumab or immunotherapy with immune checkpoint inhibitors. Various combination treatments are worthy of further research.
This study had some limitations as stated below. Shortness of resources for further dose escalation limited the axitinib MTD at 3 mg twice daily in our study setting. The real MTD of axitinib in combination of RT may possibly be higher. The nature of phase I study escalating doses only allow determination of MTD but not for efficacy. Due to only 9 cases, we can only pool all patients together to get a mixed efficacy data involving all dose levels. The efficacy reported in our phase I study should be interpreted with caution.