Point/Counterpoint: Stereotactic radiosurgery without whole-brain radiation for patients with a limited number of brain metastases: the current standard of care?

We are now in the era of stereotactic radiosurgery (SRS) alone as the accepted upfront therapy for patients presenting with limited brain metastases. The routine use of upfront whole-brain radiation (WBRT), either on its own or with SRS, should no longer be considered first line. To show why this transition in practice has occurred, please consider the evolution of the available high-level evidence.

The first 2 randomized controlled trials (RCTs) evaluating SRS with WBRT versus WBRT alone in patients with limited brain metastases provided critical evidence that local control with WBRT alone is suboptimal^1,2 and that survival can be improved with more aggressive management using SRS² (analogous to the surgical trial of surgery plus WBRT vs WBRT alone).³ These trials achieved a major change in practice, and it was accepted that WBRT on its own should not be considered standard first-line therapy for patients presenting with limited metastases.

Second, WBRT plus SRS was then compared with SRS alone in 3 RCTs.^4–6 These trials were similar in terms of inclusion criteria, with patients presenting with up to 3–4 SRS-eligible metastases confirmed by MRI and a KPS ≥70. What was not consistent were the primary endpoints which included brain tumor recurrence, neurocognitive status, and the proportion of patients with a World Health Organization performance status of 0–2.⁷ Importantly, no trial was powered to evaluate survival differences, and what we learned most from these pivotal trials can be summarized according to the following statements: These trials were successful in providing evidence-based pros and cons of adjuvant WBRT for both patients and physicians to consider in terms of decision-making; these trials were unsuccessful in answering the question of which treatment option was the best—and left the profession divided.

• Adjuvant WBRT reduces the risk of distant relapse and increases local control.⁷

• Adjuvant WBRT increases the risk of neurocognitive decline.⁵

• Adjuvant WBRT increases the risk of serious toxicities (eg, leukoencephalopathy and radiation necrosis).⁴

• Adjuvant WBRT can harm quality of life (QOL) outcomes.⁸

• Adjuvant WBRT does not improve survival.⁷

One camp interpreted the data to support SRS with WBRT as the standard of care, arguing that the benefits of intracerebral tumor control outweigh the potential cons of worse subjective patient-reported outcomes, especially when survival is not adversely impacted. The other camp interpreted the same data to support SRS alone on the basis that the cons of worse patient-reported outcomes^5,8 (eg, QOL and neurocognitive functioning⁸) attributable to adjuvant WBRT are of paramount importance in the palliative patient at the end of life, and outweigh the higher risk of intracranial relapse. Furthermore, recurrent metastases can always be salvaged with WBRT or further SRS, particularly when survival is not adversely impacted.

Further high-level evidence has recently been reported specific to the question of the toxicities of WBRT. These trials include the evaluation of hippocampal-sparing WBRT,⁹ memantine with WBRT,¹⁰ and prophylactic cranial irradiation in patients with non–small cell lung cancer.¹¹ As a result, we now can confirm with evidence that: With the causal relationship now established between normal brain tissue radiation exposure and compromised neurocognitive function, I argue that the best way to mitigate the risks of radiation-induced adverse effects is to limit radiation exposure to the normal brain; hence, SRS alone without adjuvant WBRT is the best option for patients presenting with a limited number of brain metastases. However, there are many who still argue that reducing the risk of brain relapse is of primary importance and stubbornly continue to support the use of adjuvant WBRT because survival has previously been shown not to be adversely effected. However, even this dogma has now been challenged.

• Radiation damages critical areas in the brain involved with memory, and sparing these areas may reduce the harmful effects of WBRT.⁹

• Approximately half of all patients treated with WBRT suffer cognitive-function failure following WBRT, and drug therapies may be a strategy to mitigate the risk, but the gains are minor and need further evaluation.¹⁰

• It is the WBRT and not the brain tumor recurrence that compromises neurocognitive function.¹¹

A recent individual-patient-data (IPD) meta-analysis of the 3 RCTs evaluating SRS with or without WBRT has concluded a survival advantage in patients aged ≤50 years treated with SRS alone.¹² Beyond age 50, no survival differences were observed. The IPD meta-analysis also observed that in those patients aged ≤50 years, WBRT did not add any significant benefits with respect to distant brain failure; however, WBRT did reduce the rate of distant relapse in older patients. As a result of these observations, the authors hypothesized that if exposed to WBRT without realizing the benefit of reducing the rate of distant brain failure, the adverse effects of WBRT may be impacting survival negatively. In other words, getting treatment without benefit can cause harm, and the harms in this case compromise survival. It is an interesting and provocative finding that needs validation; however, those who still support adjuvant WBRT on the basis of no survival detriment are now faced with an evidence-based rebuttal to be disproven.

When considering all the high level evidence it is now clear to most, including large professional organizations, that SRS alone is the preferred upfront treatment option for these patients. In fact, the American Society for Radiation Oncology (ASTRO) recently made a bold recommendation in their Choose Wisely campaign and stated, “Don't routinely add adjuvant whole brain radiation therapy to stereotactic radiosurgery for limited brain metastases.”¹³ Moreover, the recently revised National Comprehensive Cancer Network (NCCN) recommendations now support SRS alone even beyond those with “limited” metastases and do not specify an upper limit as to number of metastases.¹⁴ The latter is forward thinking; if the argument is to spare patients from the toxic effects of WBRT, then metastasis counting is irrelevant, and what matters is the suitability of the lesions for SRS.

At the time of writing this point/counterpoint, an abstract on NCCTG N0574 was presented at ASCO in May 2015,¹⁵ describing 213 patients with 1 to 3 brain metastases who were randomized to SRS vs WBRT plus SRS. The primary endpoint was neurocognition at 3 months post-treatment based on HVLT assessment. The abstract data confirm that WBRT plus SRS is more harmful than SRS alone with respect to cognition and quality of life at 3 months and at all time points out to and including 1 year post-treatment. Although not significant, the median OS was 10.7 months in those treated with SRS alone vs. 7.5 months in those treated with WBRT plus SRS. This trial not only validates the earlier randomized study by Chang et al.⁵ but solidifies our understanding that the harm caused by WBRT is greater than the harm resulting from brain progression or recurrence. The N0574 authors favor SRS as treatment of choice. It is important to note that this point/counterpoint is based mainly on the interpretation of the several randomized studies such as this latest trial, and we caution readers to avoid over-interpreting the several important but less reliable retrospective studies.

Lastly, SRS alone has been shown to result in more cost-effective care than SRS plus WBRT, despite the increased need for salvage therapy.¹⁶ Therefore, I support the stance that patients with limited brain metastases should not get WBRT and that the current standard of care is treatment with SRS alone.¹⁷