The clinical application of angiostatic therapy in combination with radiotherapy: past, present, future

In 2014, Gilbert et al. and Chinot et al. published the results of two phase III trials that evaluated whether bevacizumab improves the efficacy of standard chemoradiotherapy for patients with newly diagnosed glioblastoma [64, 65]. Both studies were initiated based on previous observations in phase I/II trials that suggested a potential benefit of this combination treatment [66‐69]. In the trial of Gilbert et al., bevacizumab (or placebo) was added in the fourth treatment week of concurrent radiotherapy (30 × 2 Gy) plus temozolomide, whereas Chinot et al. started bevacizumab (or placebo) already in the first treatment week of radiotherapy (30 × 2 Gy) plus temozolomide. Gilbert et al. found no benefit of bevacizumab in terms of overall survival (OS; 15.7 vs. 16.1 months; HR 1.13) and progression-free survival (PFS; 10.7 vs. 7.3 months; HR 0.79), but reported a worse quality of life and a decline in neurocognitive function in the bevacizumab group. Patients treated with bevacizumab experienced grade 3 or higher adverse events more frequently [65]. Chinot et al. did observe a prolonged PFS in the bevacizumab group as compared to placebo (10.6 vs. 6.2 months), but also failed to show a significant difference in OS. Again, grade 3 or higher adverse events were more common in the bevacizumab cohort (66.8 vs. 51.3% in the placebo group) [64]. Of note, since the statistical design of the two trials was not comparable, a direct comparison cannot be made. However, both studies point toward a favorable PFS with the addition of bevacizumab to radiotherapy plus temozolomide in newly diagnosed glioblastoma, especially when combination treatment is initiated at the start of radiotherapy. Unfortunately, the favorable PFS is accompanied with increased toxicity. Thus, it remains to be seen if bevacizumab is really a beneficial addition to the first-line treatment in glioblastoma patients.

Overall, the clinical value of combining bevacizumab with radiotherapy is open for debate, especially since most current trials either report no clinical benefit [70‐76] or only a (minor) benefit in PFS or pCR (pathological complete response) [77, 78]. For example, several phase II studies have been performed in rectal cancer patients based on promising results in phase I trials [79, 80]. Borg et al. [71] added bevacizumab to neoadjuvant 5-FU and RT for 46 patients with stage III rectal cancer before total mesorectal excision, evaluating the proportion of patients achieving a pathological complete response (pCR; ypT0N0). The study did not reach a significant difference from expected pCR (10.0%) with a rate of 11.4%. Salazar et al. also failed to show a significant difference in pCR with the addition of bevacizumab to capecitabine-based neoadjuvant chemoradiotherapy (CRT) in 44 patients with stage II/III rectal cancer as compared to 46 patients undergoing only CRT (16 vs. 11%, p = 0.54) [75]. Comparable observations were reported by Dellas et al. [81]. More recently, Landry et al. reported on the 5-year clinical outcomes of a phase II trial in patients with locally advanced rectal cancer that received preoperative chemoradiation with bevacizumab followed by postoperative chemotherapy (FOLFOX) plus bevacizumab. Despite excellent 5-year OS and DFS, the primary endpoint (30% pCR) was not reached. Moreover, the treatment schedule was associated with substantial neoadjuvant and surgical toxicity. This resulted in low compliance to adjuvant treatment, and therefore, it was recommended to not further explore this combination treatment [73]. Also Kennecke et al. [82], who evaluated preoperative bevacizumab treatment added to oxaliplatin, capecitabine and radiation in 42 locally advanced or low rectal cancer patients, suggested that the results of their study did not justify a phase III trial aimed at exploring the benefit of neoadjuvant bevacizumab in rectal cancer. Of note, Spigel et al. [83] observed an improved pCR rate of 29% in patients with stage II/III rectal cancer who were treated prior to surgery with 5-fluorouracil, bevacizumab and radiotherapy. Also Xiao et al. [84] reported that sandwich-like neoadjuvant therapy with bevacizumab was safe and effective for locally advanced rectal cancer. Possibly, differences in the timing of surgery, i.e., 2–8 versus 6–8 weeks following chemoradiation could underlie the different observations [85]. This should be taken into account when further optimizing dose-scheduling of both treatments in rectal cancer. Regarding scheduling, the patient accrual in a single-arm phase II study by Resch et al. [86] in rectal cancer patients that received bevacizumab concurrent with chemoradiation was terminated due to toxicity. Based on these current results, it can be argued whether the combination of radiotherapy with bevacizumab will provide a clinical benefit to rectal cancer patients.

A similar conclusion can be claimed regarding the combination of radiotherapy and bevacizumab in nasopharyngeal cancer. In a phase II clinical trial by Lee et al. [74], 44 patients with stage IIB–IVB nasopharyngeal cancer received radiotherapy (33 × 1.2 Gy) in combination with three cycles of bevacizumab and cisplatin, followed by standard adjuvant treatment consisting of fluorouracil in combination with bevacizumab. While the study was designed to test toxicity, the estimated PFS was lower than previously reported in standard therapy (75 vs. 86%). The estimated PFS was also not reached in the study by Yao et al. in patients with locally advanced squamous cell carcinoma of the head and neck that received bevacizumab concurrent with chemoradiation followed by maintenance bevacizumab treatment. Nevertheless, they concluded that the regimen could be further studied in appropriately selected patients, especially those not eligible for cisplatin administration [76]. Of note, a study combining both bevacizumab and erlotinib with concurrent chemoradiation in head and neck cancer patients appeared to give favorable locoregional control and OS compared to historical controls. It was suggested to further study this in a randomized study [87].

Altogether, most current studies combining bevacizumab with radiotherapy only show moderate to no clinical benefit at all. In addition, in most studies the combination treatment is associated with increased albeit manageable toxicities. All this is in line with previous observations [24].

As mentioned previously, most trials on combining angiostatic drugs with irradiation that were published between 2012 and 2015 involved the addition of bevacizumab to radiotherapy. However, a few trials assessed the combination with other angiostatic drugs. For example, several studies evaluated the addition of the tyrosine kinase inhibitor sorafenib to radiation therapy. Brade et al. [88] tested the safety of combining sorafenib with SBRT (up to 51 Gy) in liver cancer patients. They observed a high rate of adverse events and DLTs, predominantly in patients in which a high volume of liver was irradiated. This is in line with reports from Goody et al. and Dawson et al. [89, 90]. Based on these observations, it appears not advisable to combine concurrent sorafenib with SBRT in patients with locally advanced HCC [88]. Interestingly, Chen et al. [91] observed more acceptable toxicities when combining concurrent sorafenib with conventional fractionated radiotherapy (2.0–2.5 Gy/fraction up to 60 Gy). Since the response rate appeared similar compared to historical studies of radiotherapy alone, it was concluded that the schedule could be further investigated, albeit with caution [91]. A comparable conclusion was drawn by Hainsworth et al. [92] who applied maintenance sorafenib plus temozolomide treatment following fractionated radiotherapy (30 × 2 Gy) plus temozolomide in newly diagnosed glioblastoma patients. Of note, the necessity to combine radiotherapy with sorafenib with caution was further exemplified by the observation that sorafenib prior to radiotherapy can result in reduced liver volumes which might require adjustment of the radiation dose [93]. In addition, sorafenib treatment was associated with gastrointestinal perforation after radiotherapy in advanced renal cell carcinoma patients [94]. These findings again illustrate the necessity to gain more insight in the effects of alternative dose-scheduling regimes when combining radiotherapy with angiostatic drugs.

The opportunities and challenges of this combination treatment were previously reviewed by Kleibeuker et al. [63]. Here, we briefly discuss some of the latest insights. Recently, Jakob et al. published the results of two phase I trials that explored the feasibility of concurrent sunitinib and radiotherapy for treatment of locally advanced soft tissue sarcoma (STS) prior to surgery [95, 96]. They observed acceptable toxicity which was comparable to other studies that evaluated the combination of radiotherapy with angiostatic drugs in STS, including pazopanib [97] and bevacizumab [98]. Based on favorable responses, all these studies recommended further investigation of the combination treatment in future trials [95‐98]. However, a single phase Ib/II study of sunitinib with radiotherapy in soft tissue sarcoma reported unacceptable toxicities as well as increased local relapse rates [99]. Interestingly, the initial dose of sunitinib used in this study was higher compared to the other studies (50 vs. 25–37.5 mg), which might explain the observed increased toxicity. Horgan et al. explored the feasibility, tolerability and efficacy of sunitinib adjuvant to surgery in locally advanced esophageal cancer patients that received neoadjuvant chemoradiation (irinotecan/cisplatin + 25 × 2 Gy). This regime appeared feasible although it was poorly tolerated. In addition, the treatment did not show any clinical benefit compared to (historical) controls [100]. All these findings support the previous conclusions by Kleibeuker et al. [63] that effective combination of radiotherapy with sunitinib relies on better insights in the optimal dosing and scheduling of the combination treatment.

Besides the trials discussed above, a handful of studies described the combination of additional angiostatic drugs with radiotherapy. For some of these, e.g., SU5416 (semaxanib) and vandetanib, no trial results have been published, possibly because the combination treatment is no longer of interest. For other inhibitors, some information is available. As mentioned above, pazopanib was combined with radiotherapy in soft tissue sarcoma patients [97]. Based on a dose–escalation study, Haas et al. [97] concluded that neoadjuvant pazopanib (daily dose of 800 mg daily for 6 weeks) in combination with 50 Gy (25 × 2 Gy) appeared safe, albeit that toxicity should be carefully monitored in future studies. Of note, a case study reported complete remission of gastric and esophageal metastases in a renal cancer patient after treatment with radiotherapy (10 × 3 Gy) and neoadjuvant as well as adjuvant pazopanib [101]. The study by Haas et al. also reported favorable responses which warrants further studies on the clinical benefit of radiotherapy combined with pazopanib.

Two recent studies evaluated the combination of endostar/endostatin with radiotherapy in NSCLC patients. In the study by Bao et al. [102], patients with unresectable stage III NSCLC were treated with endostar combined with concurrent chemoradiation (docetaxel/cisplatin + 30–33 × 2 Gy). They reported promising short-term efficacy and local control rates, and the treatment regimen was generally well tolerated [102]. These findings are in agreement with a previous study in NSCLC patients and warrant future evaluation of this treatment [103]. On the other hand, Sun et al. evaluated the addition of endostatin to concurrent chemoradiation (carboplatin/paclitaxel + 30–33 × 2 Gy) followed by maintenance chemotherapy + endostatin in patients with unresectable stage III NSCLC. This study was closed early because of unacceptable toxicity, i.e., four out of ten patients presented with grade III pulmonary toxicity [104].

Collectively, the results of the current studies that evaluate the feasibility to combine angiostatic drugs with radiotherapy have not provided remarkable novel insights regarding this treatment regimen. The occurrence of toxicities remains an issue of concern [105]. Combining radiotherapy with bevacizumab generally shows limited efficacy while the combination with other angiostatic drugs has shown favorable responses, but still awaits further confirmation in (randomized) clinical studies. Optimization of dosing and scheduling of both treatment modalities remains one of the key future challenges.