Abstract
Background
Radiation necrosis (RN) is a potential complication after radiation therapy for brain tumors. It is hypothesized that VEGF plays an important role in the pathophysiology of RN. Bevacizumab, a monoclonal antibody against VEGF-A, is often successful in the management of RN. The objective of this study is to assess whether VEGF receptor (VEGFR) inhibitors, a group of oral tyrosine kinase inhibitors (TKIs), can prevent or reverse RN.
Methods
We retrospectively studied a cohort of 102 patients with renal cell carcinoma and brain metastases seen at The Ohio State University James Cancer Center between January 1, 2011 and April 30, 2019. We identified those who developed RN and analyzed the temporal relationship between the use of VEGFR TKIs and the development of RN.
Results
The cumulative incidence of RN is 13.7% after radiation treatments that included LINAC-based stereotactic radiosurgery, fractionated stereotactic radiotherapy, or Gamma Knife radiosurgery. There was no statistically significant difference in the cumulative incidence of RN between patients taking TKIs and patients who were off TKIs (9.9% and 11.5% respectively, P = .741). The median time to development of RN was only numerically shorter in patients taking TKIs (151 vs 315 days, P = .315). One patient developed RN after stopping cabozantinib. Eight patients developed RN while on cabozantinib, pazopanib, or sunitinib. One patient was started on axitinib during active RN without significant improvement subsequently.
Conclusions
VEGFR TKIs do not consistently prevent RN. The therapeutic effects of VEGFR TKIs against RN warrant further research.
Keywords: bevacizumab, radiation necrosis, tyrosine kinase inhibitors, vascular endothelial growth factor
As patients with primary brain tumors and brain metastases live longer thanks to better systemic therapies, and with the increased use of concomitant chemotherapy and radiation therapy (RT) with newer radiation modalities (ie, LINAC-based stereotactic radiosurgery (SRS) or Gamma Knife radiosurgery), late toxicities of radiation treatments have become more relevant. More specifically, it is estimated that the risk of cerebral radiation necrosis (RN) ranges mostly from 5% to 10%.1,2 Certain tumor histological subtypes such as renal cell carcinoma (RCC), lung adenocarcinoma, and BRAF-negative melanoma carry an increased risk of RN.2 RN can cause serious morbidity as patients frequently present with headaches, focal neurological symptoms, and seizures. The pathophysiology of RN is not well understood but it is hypothesized that VEGF plays an important role.3 It remains unclear what patient/tumor characteristics lead to the development of RN in some patients but not others.
The management of RN includes the use of corticosteroids. However, corticosteroids often fail to control RN and are not optimal for long-term management because of multiple side effects. Surgical resection when feasible provides excellent control. Given the presumed role of VEGF in RN, bevacizumab (a monoclonal antibody against VEGF-A) has been tried with excellent success rates.4 VEGF-A, a member of the VEGF family, activates VEGF receptors (VEGFR) 1 and 2. VEGF-A/VEGFR2 (encoded by kinase insert domain receptor) are predominantly responsible for tumor angiogenesis.5 Bevacizumab is an intravenous infusion that is administered every 2 weeks at 7.5 mg/kg for a total of 4 doses in the setting of RN. However, from our experience and the experience of others, relapses of RN after bevacizumab do occur.6 Moreover, bevacizumab can have significant side effects and is often contraindicated in patients with thromboembolic disease or in pathologies with high tendencies for intracranial bleed such as squamous cell carcinomas. Other treatment options include hyperbaric oxygen,7Boswellia serrata,8 pentoxyifylline (Trental) and vitamin E (NCT01508221). However, the evidence for their use to treat RN is not well established and is limited to small studies.
Despite the success of bevacizumab, the role of oral VEGFR inhibitors in RN remains unclear. VEGFR inhibitors are a group of oral tyrosine kinase inhibitors (TKIs: sunitinib, sorafenib, axitinib, pazopanib, and cabozantinib). Most of these TKIs inhibit VEGFR1 to 3 and are generally well tolerated with mainly gastrointestinal and less often hematologic toxicity. Given the presumed role of VEGF in the pathogenesis of RN and the success of VEGF inhibition with bevacizumab in treating RN, our study aimed to assess whether VEGFR TKIs can prevent or improve the radiographic appearance of RN and can be used as alternative oral therapies for the management of RN. We looked into the temporal relationship between development of RN and the use of VEGFR TKIs. VEGFR TKIs are routinely used for renal cell carcinoma (RCC), therefore, our study population consisted of patients with RCC and brain metastases.
Methods
Patient Selection
Under an IRB-approved protocol, The Ohio State University Data Warehouse provided a list of 348 patients with RCC and brain metastases between January 1, 2011 and April 30, 2019, based on the International Classification of Diseases, ninth revision (ICD9) codes (C79.31) for brain metastases and (C64.9) for malignant neoplasm of the kidneys corresponding to ICD10 codes (198.4) and (189), respectively.
We then went through the individual patient charts and identified those who truly had brain metastases from RCC histology who were treated with RT and had at least one brain MRI following RT. We confirmed those who subsequently developed RN and applying the Response Assessment in Neuro-Oncology Brain Metastases criteria for progression.9 Radiation necrosis was defined as a greater than 20% increase in the larger diameter of a lesion—on T1-postcontrast sequences—that was favored to be treatment effect by the multidisciplinary team caring for patients and via biopsy when applicable.
Data pertaining to patients’ treatment history, RT modality and dose, duration between RT and development of RN, method of diagnosis of RN, and management of RN was collected. In addition, for each patient, the number of days while on or off TKIs to event (death, loss of follow-up, or development of RN) was calculated. Patients off TKIs were on no systemic treatment, mTOR (mammalian target of rapamycin) inhibitors, or immunotherapy at the time.
Statistical Analysis
Cumulative incidence was calculated as the number of RN events/number of patients at risk (following RT) in the beginning of the follow-up period. Chi-square test was performed to compare the incidence of RN between the cohort of patients who developed RN while on TKIs vs while off TKIs. A nonparametric test was used to compare the median time to development of RN between the 2 groups. A Cox-regression model was used to graph a hazard plot of time to event for both groups. IBM SPSS Statistics 26 was used.
Results
Overall and Per-Group Cumulative Incidence Rates
Of the 348 patients identified by the Data Warehouse, 102 patients with RCC and brain metastases treated with RT with sufficient follow-up were included. We identified 14 patients out of 102 with RCC and brain metastases who developed RN following RT corresponding to an overall cumulative incidence of 13.7% over a median follow-up of 195 days (range, 12-2861 days). Two of the 14 patients had 2 lesions each that developed RN. These were calculated separately when taking into account temporal relationship to TKIs. The median time to development of RN in the entire cohort was 244.5 days.
Seven out of 64 patients developed RN at the time of taking TKIs, corresponding to a cumulative incidence of 9.9% and an incidence rate of 2.5/10 000 person/days. On the other hand, 9 out of 69 patients developed RN while off TKIs, corresponding to a cumulative incidence of 11.5% and an incidence rate of 3/10 000 person/days. The chi-square test yielded a P value of .741, suggesting no significant difference between the cumulative incidences between the 2 groups. The median time to development of RN was 151 days (range, 18-428 days) and 315 days (range, 11-587 days) while on or off TKIs, respectively. Although numerically shorter while on TKIs, this test was not statistically significant (P = .315).
Figure 1 illustrates a hazard plot of the risk of developing RN in each group over time.
Figure 1.
Hazard Plot of Risk of Developing Radiation Necrosis Over Time in Each Group. TKI, tyrosine kinase inhibitor.
Characteristics of Patients Who Developed Radiation Necrosis
Fourteen patients developed cerebral RN. All patients who developed RN had metastatic RCC with clear cell histology. One patient had clear cell RCC with 80% sarcomatoid differentiation. Patients received LINAC-based stereotactic radiosurgery (SRS), Gamma Knife radiosurgery, or fractionated stereotactic radiotherapy. None of the patients had received whole-brain radiation therapy. Cumulative doses of RT preceding RN ranged from 15 to 49 Gy. RN developed around irradiated surgical beds or naive lesions. Five patients underwent surgical resections for diagnostic or therapeutic purposes (3 off TKIs and 2 on TKIs). On the other hand, 9 patients were diagnosed with RN based on MRI with perfusion along with the clinical course and subsequent imaging. Two patients developed acute RN 11 and 18 days after RT, but the majority developed late-injury/RN more than 100 days after RT. Corticosteroids were used for those who did not have surgery with tapering schedules over 4 to 6 weeks with stable imaging or partial/complete improvement in scans in all patients as defined using T1-postcontrast images. Cerebral edema invariably and proportionally improved with improvement in the size of enhancing lesions. None of the patients were on dexamethasone prior developing RN. Six of 9 patients in the non-TKI cohort and 5 of 7 patients in the TKI cohort received steroids for RN. Bevacizumab was not used for RN in this cohort. Table 1 summarizes the patients’ characteristics, radiation modality used, doses of radiation, and the associated morbidity using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0.
Table 1.
Characteristics of Patients Who Developed Radiation Necrosis
| No. | Age at RN, y/sex | Pathology | Location | Surgery prior to RT? | RT modality | RT dose, Gy | RT to RN, d | RN diagnosis | CTCAE |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 61 M | Clear cell | L temporal | Naive | SRS | 20 × 1 | 281 | Pathology | 3 |
| L frontal | Resected | FSRT | 8 × 3 | 587 | Imaging | 2 | |||
| 2 | 56 M | Clear cell | L parietal | Naive | SRS | 15 × 1 | 174 | Imaging | 3 |
| 3 | 43 M | Mixed | L temporal | Resected | FSRT | 8 × 3 | 431 | Pathology | 3 |
| 4 | 63 F | Clear cell | L parietal– R cerebellar | Naive | SRS | 20 × 1 | 315 | Imaging | 2 |
| 4 | R frontal | Naive | SRS | 20 × 1 | 126 | Imaging | 2 | ||
| 5 | 47 M | Clear cell | L frontal | Resected | SRS | 20 × 1 | 208 | Pathology | 3 |
| 6 | 64 F | Clear cell | R parietal | Naive | SRS | 20 × 1 | 343 | Imaging | 3 |
| 7 | 59 F | Clear cell | L parietal | Resected | GK | 24 | 409 | Pathology | 3 |
| 8 | 75 M | Clear cell | R occipital | Both | GK/FSRT | 24/5 × 5 | 538 | Imaging | 2 |
| 9 | 71 F | Clear cell | L temporal | Resected | FSRT | 8 × 3 | 428 | Imaging | 3 |
| 10 | 60 M | Clear cell | R parietal | Naive | FSRT | 8 × 3 | 101 | Imaging | 2 |
| 11 | 68 F | Clear cell | L basal ganglia | Naive | FSRT | 6 × 5 | 35 | Imaging | 2 |
| 12 | 72 M | Clear cell | L occipital | Naive | FSRT | 8.5 × 3 | 151 | Pathology | 3 |
| 13 | 41 M | Clear cell | L temporal | Naive | FSRT | 8 × 3 | 115 | Imaging | 2 |
| 14 | 52 M | Clear cell | L occipital–parietal | Resected | FSRT | 8 × 3 | 392 | Imaging | 2 |
Abbreviations: CTCAE: Common Terminology Criteria for Adverse Events (CTCAE) v5.0; F, female; FSRT, fractionated stereotactic radiotherapy; GK, Gamma Knife; L, left; M, male; R, right; RN, radiation necrosis; RT, radiation therapy; SRS, stereotactic radiosurgery.
Patient 3 had a resection of a metastatic lesion in the left temporal lobe followed by fractionated stereotactic radiotherapy. The pathology was consistent with clear cell RCC with 80% sarcomatoid differentiation. He was then started on cabozantinib for systemic treatment 5 months after his brain RT. He remained on cabozantinib for 8 months before it had to be discontinued because of gastrointestinal side effects. His brain MRI 11 days after discontinuation of the drug and more clearly 72 days later showed a significant increase in the enhancement and fluid-attenuated inversion recovery changes (Fig. 2). He underwent re-resection to rule out recurrent metastatic disease, and the pathology confirmed areas of necrosis with no neoplastic cells. RN stabilized thereafter.
Figure 2.
MRI images for patient 3: (Left) Postcontrast T1 image at the time of irradiation to the surgical bed in the left temporal lobe. (Middle) Postcontrast T1- (right) fluid-attenuated inversion recovery image from MRI of the brain 72 days after stopping cabozantinib (431 days following radiation therapy) showing significant increase in contrast enhancement and vasogenic edema that was later proven on pathology to represent radiation necrosis.
Three other patients developed RN while on cabozantinib (one patient had multiple interruptions in treatment, however, due to gastrointestinal toxicity). Two patients developed RN while on pazopanib. Three patients developed RN while on sunitinib. One patient started axitinib during active RN without significant improvement subsequently and was the only patient on active immunotherapy with nivolumab at the time of development of RN. Finally, 4 patients developed RN while having been on no systemic VEGFR inhibitors. Table 2 summarizes the timeline between treatment with VEGFR TKIs and development of RN.
Table 2.
Temporal Relationship Between Treatment With VEGF Receptor Tyrosine Kinase Inhibitors and Development of Radiation Necrosis
| ID | RN and VEGFR TKI treatment | TKI dose |
|---|---|---|
| 1 | Off TKIs | |
| 2 | Developed RN while on pazopanib | 800 mg daily |
| 3 | Developed RN after stopping cabozantinib | 40 mg daily |
| 4 | MRI did not improve after axitinib | 5 mg twice a day |
| 5 | Developed RN while on cabozantinib | 40 mg daily |
| 6 | Developed RN while on cabozantinib | 40 mg daily |
| 7 | Off TKIs | |
| 8 | Off TKIs | |
| 9 | Developed RN while on sunitinib | 50 mg daily (2 wks on/1 wk off) |
| 10 | Developed RN while on sunitinib | 50 mg daily (2 wks on/1 wk off) |
| 11 | Developed RN while on pazopanib | 800 mg daily |
| 12 | Developed RN while on sunitinib | 25-37.5 mg daily (4 wks on/2 wks off) |
| 13 | Off TKIs | |
| 14 | Off TKIs |
Abbreviations: ID, identification; RN, radiation necrosis; TKI, tyrosine kinase inhibitor; VEGFR, VEGF receptor.
Discussion
Cerebral RN presents with significant morbidity and is often challenging to diagnose and manage. VEGF is thought to play a role in the pathogenesis of RN. Bevacizumab, an anti–VEGF-A monoclonal antibody, is a widely used treatment for RN with excellent response rates. However, relapses of RN do occur after bevacizumab. The role of VEGFR TKIs in RN has not been studied. We conducted this retrospective cohort study to assess whether VEGFR TKIs can prevent or improve the radiographic appearance of RN. We chose a cohort of patients with RCC and brain metastases because multiple VEGFR TKIs have been approved by the FDA for the management of RCC.
The overall cumulative incidence of cerebral RN in our cohort was 13.7% over a median follow-up of 195 days (range, 12-2861 days). There was no statistically significant difference in the cumulative incidence of RN between patients taking TKIs and patients who were off TKIs (9.9% and 11.5%, respectively). Likewise, there was no statistically significant difference in median time to development of RN between the aforementioned 2 groups (151 vs 315 days, respectively). The previously reported range of RN for patients with RCC and brain metastases appears to be 8% to 10%.2 Furthermore, in our cohort, one patient developed RN only shortly after stopping cabozantinib and hence it was thought that cabozantinib may have prevented RN in his case. Most of the other patients developed RN while on cabozantinib, pazopanib, or sunitinib. One patient started axitinib during active RN without subsequent improvement.
The literature varies as to results of previous similar reports. A retrospective study of 326 patients with RCC and brain metastases reported an incidence of RN up to 9%.10 TKI use within 30 days of SRS was associated with a significantly increased 12-month incidence of radiation necrosis (10.9% vs 6.4%, P = .040). However, the duration of TKI treatment was not taken into account. Another study also reported an increased 12-month cumulative incidence of RN in patients with RCC with brain metastases treated with SRS concurrently with VEGFR TKIs vs SRS alone (14.3% vs 6.6%, respectively, P = .04).11 On the other hand, a retrospective study of 106 patients with RCC and brain metastases who were treated with SRS plus sorafenib or sunitinib noted no incidents of RN after a median follow-up of 14.7 months.12
It is unclear why a VEGF monoclonal antibody would have better effects than VEGFR TKIs. The efficacy of bevacizumab for RN suggests that VEGF is a major player in the pathogenesis of RN. Other etiologies include overexpression of inflammatory markers and cytokines (TNF-α, IL-1, and IL-6).13 It is believed that bevacizumab does not cross the blood-brain barrier (BBB)14 and perhaps exhibits its effect on the brain by sequestering VEGF in the serum.
Furthermore, there are no robust clinical pharmacodynamic studies for the TKI agents and their ability to cross the BBB. Small studies have suggested restricted transport of sunitinib across the BBB,15 but the ability of axitinib and pazopanib to cross the BBB is unknown.16,17 Similarly, there are no reliable clinical data regarding cabozantinib and its ability to penetrate the BBB.18 On the other hand, it is hypothesized that VEGFR TKIs are radiosensitizers and perhaps lead to an increased risk of RN given the improvement of survival in this patient population.10
In summary, our study suggests that TKIs do not consistently prevent the occurrence of RN. However, it is important to note that the role of bevacizumab in RN prophylaxis is not known and a direct comparison is not possible. Moreover, no conclusion can be made regarding the potential therapeutic activity of VEGFR TKIs when used in TKI-naive patients, as only one RN patient in our cohort started a TKI after developing RN. Follow-up studies are needed to further delineate the role of therapeutic effects of VEGFR TKIs in RN.
Acknowledgments
This study has not been previously presented or published.
Author contributions include the following: data mining: Iyad Alnahhas and Appaji Rayi; data analysis: Iyad Alnahhas; and data interpretation and manuscript writing/editing: all authors.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Conflict of interest statement. None declared.
References
- 1. Vellayappan B, Tan CL, Yong C, et al. Diagnosis and management of radiation necrosis in patients with brain metastases. Front Oncol. 2018;8:395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Miller JA, Bennett EE, Xiao R, et al. Association between radiation necrosis and tumor biology after stereotactic radiosurgery for brain metastasis. Int J Radiat Oncol Biol Phys. 2016;96(5):1060–1069. [DOI] [PubMed] [Google Scholar]
- 3. Nordal RA, Nagy A, Pintilie M, et al. Hypoxia and hypoxia-inducible factor-1 target genes in central nervous system radiation injury: a role for vascular endothelial growth factor. Clin Cancer Res. 2004;10(10):3342–3353. [DOI] [PubMed] [Google Scholar]
- 4. Levin VA, Bidaut L, Hou P, et al. Randomized double-blind placebo-controlled trial of bevacizumab therapy for radiation necrosis of the central nervous system. Int J Radiat Oncol Biol Phys. 2011;79(5):1487–1495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Takahashi S. Vascular endothelial growth factor (VEGF), VEGF receptors and their inhibitors for antiangiogenic tumor therapy. Biol Pharm Bull. 2011;34(12):1785–1788. [DOI] [PubMed] [Google Scholar]
- 6. Sadraei NH, Dahiya S, Chao ST, et al. Treatment of cerebral radiation necrosis with bevacizumab: the Cleveland Clinic experience. Am J Clin Oncol. 2015;38(3):304–310. [DOI] [PubMed] [Google Scholar]
- 7. Chuba PJ, Aronin P, Bhambhani K, et al. Hyperbaric oxygen therapy for radiation-induced brain injury in children. Cancer. 1997;80(10):2005–2012. [DOI] [PubMed] [Google Scholar]
- 8. Kirste S, Treier M, Wehrle SJ, et al. Boswellia serrata acts on cerebral edema in patients irradiated for brain tumors: a prospective, randomized, placebo-controlled, double-blind pilot trial. Cancer. 2011;117(16):3788–3795. [DOI] [PubMed] [Google Scholar]
- 9. Lin NU, Lee EQ, Aoyama H, et al. ; Response Assessment in Neuro-Oncology (RANO) group . Response assessment criteria for brain metastases: proposal from the RANO group. Lancet Oncol. 2015;16(6):e270–e278. [DOI] [PubMed] [Google Scholar]
- 10. Juloori A, Miller JA, Parsai S, et al. Overall survival and response to radiation and targeted therapies among patients with renal cell carcinoma brain metastases [published online ahead of print January 18, 2019]. J Neurosurg. doi:10.3171/2018.8.JNS182100. [DOI] [PubMed] [Google Scholar]
- 11. Kim JM, Miller JA, Kotecha R, et al. The risk of radiation necrosis following stereotactic radiosurgery with concurrent systemic therapies. J Neurooncol. 2017;133(2):357–368. [DOI] [PubMed] [Google Scholar]
- 12. Staehler M, Haseke N, Nuhn P, et al. Simultaneous anti-angiogenic therapy and single-fraction radiosurgery in clinically relevant metastases from renal cell carcinoma. BJU Int. 2011;108(5):673–678. [DOI] [PubMed] [Google Scholar]
- 13. Yoshii Y, Sugimoto K, Fujiwara K. Progressive enlargement of a mass lesion in late cerebral radionecrosis. J Clin Neurosci. 2011;18(6):853–855. [DOI] [PubMed] [Google Scholar]
- 14. Verduin M, Zindler JD, Martinussen HM, et al. Use of systemic therapy concurrent with cranial radiotherapy for cerebral metastases of solid tumors. Oncologist. 2017;22(2):222–235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Dudek AZ, Raza A, Chi M, et al. Brain metastases from renal cell carcinoma in the era of tyrosine kinase inhibitors. Clin Genitourin Cancer. 2013;11(2):155–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Poller B, Iusuf D, Sparidans RW, et al. Differential impact of P-glycoprotein (ABCB1) and breast cancer resistance protein (ABCG2) on axitinib brain accumulation and oral plasma pharmacokinetics. Drug Metab Dispos. 2011;39(5):729–735. [DOI] [PubMed] [Google Scholar]
- 17. Minocha M, Khurana V, Mitra AK. Determination of pazopanib (GW-786034) in mouse plasma and brain tissue by liquid chromatography–tandem mass spectrometry (LC/MS–MS). J Chromatogr B Analyt Technol Biomed Life Sci. 2012;901:85–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Abdelaziz A, Vaishampayan U. Cabozantinib for the treatment of kidney cancer. Expert Rev Anticancer Ther. 2017;17(7):577–584. [DOI] [PMC free article] [PubMed] [Google Scholar]