Brain metastasis usually carries poor prognostic foresight. Radiation therapy is a key component in the management of brain metastases. The challenge for radiation oncologists treating brain metastases is to individualize the treatment approach depending on patient and tumour factors. This study reports favourable outcomes following HSRT for intact brain metastases using 5-fraction schedules. We report a high 1-year local control rate of 84.4% for all lesions, with an acceptable ARE rate of 27%. Delivery of a higher radiation dose was associated with higher rates of local control. These findings may have implications for the implementation of HSRT in the clinic as well as the design of prospective HSRT trials.
HSRT for brain metastases is now common practice. Previously reported 1-year local control estimates of HSRT have ranged from 52%-95% [11–15]. The 1-year local control in our study appears comparable. One important feature of our study was that we excluded patients who had previously undergone brain radiotherapy or surgery, thereby resulting in a more realistic local control rate of HSRT.
Currently, 30 Gy in five fractions is the recommended treatment [16] and is commonly used in clinical practice. However, we observed an improvement in LC when treating with > 30 Gy/5 fractions, with a 1-year cumulative incidence of local failure of 11.2%. Similar results have also been reported [17–19]. Sten Myrehaug’s study showed that local failure at 1 year was 33% for patients treated with < 30 Gy/5 fractions versus 19% for patients treated with ≥ 30 Gy/5 fractions, with a total dose of < 30 Gy as opposed to ≥ 30 Gy (HR, 1.68, P = 0.03) as a significant predictor of local failure [18]. After retrospective analysis of 525 patients with 720 brain metastases treated with HSRT, Baliga Sujith et al. found that the tumour control probability increased with the BED10 with a hazard ratio equal to 0.77 for every 10 Gy10. A 1-year local control rate > 70% seemed to be achieved with a BED10 of 40 to 50 Gy. A BED10 of 50 to 60 Gy seemed to achieve a 1-year local control rate of at least 80% at 12 months [20]. Our study confirmed a 1-year tumour control estimate with dose escalation. Therefore, our results further support the recognition that the radiation dose is dependent on local tumour control.
Our data did not find that the tumour diameter affected the rate of 1-year local control after HSRT (P = 0.39. Figure 4). For small lesions treated with HSRT, the 1-year local control rate was comparable to that for small lesions treated with single-fraction SRS. The high proportion of local control observed in the present study supports the growing employment of HSRT according to the literature, which confirmed small tumour size as a prediction index of improved local tumour control, and HSRT appears to be a reasonable alternative for small brain metastases [21–22]. Studies have shown that SRS in the treatment of large lesions compared with small lesions can significantly reduce the local control rate of large lesions. However, when we set the tumour diameter to ≥ 1.5 cm, we did not find that the local control rate decreased. Due to the high ratio of hypoxic cells in large brain metastases, HSRT may have an advantage with respect to LC versus single-fraction SRS, especially for large metastases. Given that surviving hypoxic tumour cells reoxygenate during fractionation and become more sensitive to subsequent irradiation, reoxygenation between dose fractions is a pivotal phenomenon that should be fully utilized in HSRT [23].
Generally, higher iLC was observed in patients receiving targeted therapy or immunotherapy combined with HSRT than in those receiving chemotherapy combined with HSRT. Common chemotherapies have been evaluated in clinical trials but have failed to demonstrate a significant benefit in patients with brain metastases [24–25]. For targeted therapy combined with HSRT, a good response has recently been reported in non-small cell lung cancer patients with brain metastases treated with tyrosine kinase inhibitors (TKIs). Many targeted therapies, such as EGFR-TKIs or ALK-TKIs, have good central nervous system (CNS) penetration with an intracranial response rate of more than 70% [26]. HSRT can disrupt the BBB, thereby increasing the CNS penetration of TKIs [27]. Therefore, combinatorial HSRT and TKIs may prove to be useful. A retrospective study showed that the local control rate of patients with NSCLC brain metastasis treated with HSRT in combination with TKIs was higher than that of patients treated with HSRT alone [28]. For patients receiving targeted therapy, a lower dose of brain HSRT can also result in a higher local control rate of brain metastases. In patients with negative driver genes, a higher prescription dose may be required to improve intracranial lesion control (Fig. 3).
For immunotherapy combined with HSRT, it has been reported that radiotherapy may be able to counteract the immunosuppressive tumour microenvironment through a variety of mechanisms [29–31]. HSRT is known to increase both innate and adaptive immune responses, making tumour cells more susceptible to T-cell-mediated killing that will boost local effects [32].
Our study showed that none of the patients receiving HSRT had severe CNS toxicity. According to previous reports, the risk of radiation-induced CNS toxicity generally increases with higher doses, prior history of radiation therapy, and larger target volumes [33–35]. HSRT does not damage surrounding tissues to the same extent as it better enables cellular reoxygenation and target volume redistribution, thus better preserving normal tissues than single-fraction SRS [36]. Minniti et al. published a retrospective study that compared the incidence of radionecrosis after single-fraction SRS with multifraction SRS. The results showed that the probability of radionecrosis after treatment with multifraction SRS was significantly lower than that of single-fraction SRS [37]. Due to the small sample size and the difficulty in identifying whether CNS toxicity is caused by brain metastasis or HSRT, we have not found any predictive factors for toxicity.
There are several limitations to this preliminary study. First, the current study was retrospective in design and based on a single institutional experience. There may be a bias in the selection of patients. Second, our sample size was relatively small, and some patients were followed for a shorter time. This led to some numerical differences in the results, but the differences did not reach statistical significance. However, robust data support the safety and efficacy of HSRT in the treatment of metastases.