It is time to reevaluate the management of patients with brain metastases.

SPECIAL ARTICLE SPECIAL ARTICLE

It Is Time to Reevaluate the Management of Patients With Brain Metastases Douglas Kondziolka, MD, MSc, FRCSC* Steven N. Kalkanis, MD‡ Minesh P. Mehta, MD§ Manmeet Ahluwalia, MD¶ Jay S. Loeffler, MDk *Departments of Neurosurgery and Radiation Oncology, NYU Langone Medical Center, New York, New York; ‡Department of Neurosurgery, Henry Ford Health System, Detroit, Michigan; §Department of Radiation Oncology, University of Maryland, Baltimore, Maryland; ¶Department of Medicine (NeuroOncology), Cleveland Clinic Foundation, Cleveland, Ohio; kDepartment of Radiation Oncology, Harvard Medical School, Boston, Massachusetts Correspondence: Douglas Kondziolka, MD, Department of Neurosurgery, NYU Langone Medical Center, 530 First Ave., Suite 8R, New York, NY 10016. E-mail: [email protected] Received, February 18, 2014. Accepted, March 7, 2014. Published Online, March 10, 2014. Copyright © 2014 by the Congress of Neurological Surgeons.

There are many elements to the science that drives the clinical care of patients with brain metastases. Although part of an understanding that continues to evolve, a number of key historical misconceptions remain that commonly drive physicians’ and researchers’ attitudes and approaches. By understanding how these relate to current practice, we can better comprehend our available science to provide both better research and care. These past misconceptions include: Misconception 1: Once a primary cancer spreads to the brain, the histology of that primary tumor does not have much impact on response to chemotherapy, sensitivity to radiation, risk of further brain relapse, development of additional metastatic lesions, or survival. All tumor primary histologies are the same once they spread to the brain. They are the same in terms of the number of tumors, radiosensitivity, chemoresponsiveness, risk of further brain relapse, and survival. Misconception 2: The number of brain metastases matters. This number matters in terms of subsequent brain relapse, survival, and cognitive dysfunction; the precise number of metastases can also be used as a limit in determining which patients might be eligible for a particular treatment option. Misconception 3: Cancer in the brain is always a diffuse problem due to the presence of micrometastases. Misconception 4: Whole-brain radiation therapy invariably causes disabling cognitive dysfunction if a patient lives long enough. Misconception 5: Most brain metastases are symptomatic. Thus, it is not worth screening patients for brain metastases, especially because the impact on survival is minimal. The conduct and findings of past clinical research have led to conceptions that affect clinical care yet appear limiting. KEY WORDS: Brain metastasis, Brain tumor, Cancer, Clinical trial, Radiosurgery, Radiotherapy, Surgery Neurosurgery 75:1–9, 2014

DOI: 10.1227/NEU.0000000000000354

O

ver the past 3 decades, considerable effort has been expended to address the clinical challenges posed by brain metastases.1-7 Unfortunately, given the large magnitude of the problem, the resources expended have not been commensurate with the scope of the problem, especially in the context of the number of prospective well-designed clinical trials. In part, this stems from the often nihilistic approach to the problem, given the relatively short survival of

SANS LifeLong Learning and NEUROSURGERY offer CME for subscribers that complete questions about featured articles. Questions are located on the SANS website (http://sans.cns.org/). Please read the featured article and then log into SANS for this educational offering.

NEUROSURGERY

ABBREVIATIONS: CI, confidence interval; HVLT, Hopkins Verbal Learning Test; SRS, stereotactic radiosurgery; WBRT, whole-brain radiotherapy Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.neurosurgery-online.com).

www.neurosurgery-online.com

many patients with metastatic disease to the brain, the inability of regulators and pharmaceutical companies to come to grips with a “compartmental solution,” ie, improving intracranial control without necessarily affecting survival, and the inability of most drugs to cross the bloodbrain barrier in sufficient concentrations to have a genuine impact on intracranial metastases. Nevertheless, a few key randomized trials, guidelines initiatives, comparative cohort studies, and modestly sized registries have been used to provide some answers to important knowledge gaps based on available resources and data.1 Some gains have clearly been made. For instance, we know that, for the most part, the whole-brain radiotherapy (WBRT) schedule and dose likely do not relate to survival, although they can affect cognition,2 that surgical resection alone is associated with a higher recurrence rate when

VOLUME 75 | NUMBER 1 | JULY 2014 | 1

Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited

KONDZIOLKA ET AL

radiotherapy does not follow resection,3 that WBRT alone is associated with a lower survival rate compared with radiation plus surgical resection in patients with a single metastasis,3 that average survival expectations remain poor,1,6 that chemotherapeutic and targeted drugs have nil to modest activity,5 and that focal high-dose radiation as exemplified by stereotactic radiosurgical techniques yields high rates of local control and improved survival.4 With early diagnosis, most patients are found to have brain metastases that are asymptomatic. It follows then that the goal of modern brain metastasis treatment is not usually to improve overall prognosis or function in such patients, but rather to prevent neurological deterioration while care of the extracranial cancer continues uninterrupted. Of the thousands of scientific articles written, most are descriptive and provide low levels of evidence on any given topic.8 Many provide conflicting conclusions that contradict previous findings. Because high-quality science should provide results that can be reproduced by others, such disparate conclusions cast a degree of suspicion on the validity of many of these findings. Could it be that we are studying separate problems collated under a single convenient label and therefore arriving at different conclusions, or could we be using incorrect or vastly different measures for the same endpoint and thus also arriving at differing conclusions?

MISCONCEPTION 1: ALL HISTOLOGIES ARE CREATED EQUAL The majority of brain metastasis randomized trials have historically pooled together different primary cancers with the assumption that the primary diagnosis was irrelevant. This simplistic approach likely developed from 2 fundamental causes. First, the early clinical series, which primarily evaluated the impact of WBRT, had combined all histologies together for many years, with the recognition that normal brain tolerance would set the dose limits, and thus a precedent was set to use the “one size fits all” approach. Second, it was easier to accrue patients to those studies by not excluding specific tumor types. A further confounding variable is that tumor diagnosis evolved from routine hematoxylin and eosin histological classification, to the inclusion of special stains, and the more recent identification of receptors and genetic/molecular characteristics that divide single histological entities into multiple different prognostic and treatment subgroups. The impact of these unique biological, genetic, and molecular features has rarely been studied in well-conducted clinical trials of patients with brain metastases.9,10 New therapeutic approaches are being developed that follow our understanding of these biological differences.11 For example, it is increasingly apparent that luminal A, luminal B, and triple negative breast cancers are different diseases in terms of the likelihood of the development of brain metastases, responding to specific therapeutic interventions, and overall survival. b-raf gene–mutated melanoma is associated with different therapeutic options and outcomes compared with melanoma without the mutation.12

2 | VOLUME 75 | NUMBER 1 | JULY 2014

Melanoma, where the brain is the first visceral metastatic site, may even be a different clinicopathological entity with reduced survival.13 Even a large breast cancer brain metastasis radiosurgery study reported in 2010 is compromised because it does not comprehensively account for these tumor subtypes.14 Medical oncologists offer different therapeutic regimens to these different patients and have different expectations as to outcome. Unfortunately, previous brain tumor clinical trials do not reflect these differences. An analysis of some of the most recognized and often-cited randomized trials in the brain metastasis literature provides insight into what may appropriately be understood from these works. In 1 study on the value of WBRT after surgical resection, primary tumor histologies (in the observation group and the radiation group) included lung cancer (n = 28 and n = 29, respectively), breast cancer (n = 4 and n = 5, respectively), unknown primary (n = 4 and n = 5, respectively), genitourinary (n = 5 and n = 3, respectively), gastrointestinal (n = 4 and n = 4, respectively), head and neck (n = 0 and n = 2, respectively), and melanoma (n = 1 and n = 1, respectively).15 This wide variety of tumor types is common to all trials. What is also curious is the relative paucity of melanoma cases, one of the most common primary cancers to spread to the brain. Essentially, non-small cell lung cancer was the only group somewhat adequately studied in this trial, and yet the study conclusions have been adopted broadly for use in multiple other tumor types. In the Patchell et al16 trial on the role of surgical resection, 11% of cases were excluded because histology did not show metastatic cancer. What is also interesting is the histology of the patients who were included. In the surgical and radiation groups, patients with the following histological subtypes were represented in widely varying degrees: lung cancer (n = 18 and n = 19, respectively), breast cancer (n = 2 and n = 1, respectively), gastrointestinal cancer (n = 2 and n = 1, respectively), genitourinary cancer (n = 1 and n = 1, respectively), and melanoma (n = 2 and n = 1, respectively). Given the major discrepancies in extracranial management and the tumor response to radiation for differing tumor types, it is not unreasonable to reject the data from the tumor types that were present in only very limited numbers of patients. Viewed through the lens of histological classification, the Patchell et al16 trial was essentially a small study on non-small cell lung cancer. However, the results have been widely extrapolated, somewhat prematurely and possibly inappropriately, to all cancer types regardless of primary histology. This is not to be critical. At the time of that study, broad histological inclusion was the norm. It remains so today. Interestingly, not all studies have reached the same conclusion as the Patchell et al trial. The 1996 study by Mintz et al,17 which showed no benefit of surgical resection in addition to WBRT, was thought less important and largely ignored. This report had 282 citations in Thomson Reuters Web of Knowledge (September 3, 2013) compared with the 1241 citations of the 1990 Patchell et al trial. Despite the fact that this study included almost twice the number of patients as the Patchell et al trial, its findings



MANAGEMENT OF PATIENTS WITH BRAIN METASTASES

were thought to be less relevant because a higher percentage of patients had active extracranial cancer at the time of management. And unlike the Patchell et al trial, magnetic resonance imaging (MRI) before treatment was not uniformly performed along with computed tomography. Indeed, active extracranial cancer is usually present when brain metastasis is identified. Recursive partitioning analysis grade 2 represents the largest patient group and is defined by a Karnofsky Performance Scale score of 70 or higher and the presence of at least 1 unfavorable prognostic factor such as age 65 years and older, extracerebral metastases, and uncontrolled primary tumor. In the Mintz et al17 study, the representation of various tumor types in the surgery plus radiation group and the radiation-alone group, respectively, included non-small cell lung cancer (n = 23 and n = 22 patients, respectively), breast cancer (n = 2 and n = 8, respectively), genitourinary cancer (n = 1 and n = 2, respectively), gastrointestinal cancer (n = 10 and n = 3, respectively), and melanoma (n = 2 and n = 2, respectively). In hindsight, this trial included more patients and evaluated a group of patients that more closely represents the population-based norm, but still included mostly patients with non-small cell lung cancer. We do not argue that this trial is any more important. It is used to explain why trials studying ostensibly the “same” problem can reach different conclusions. Mainly, these differences occurred because the factors that most strongly affected outcomes, the response to therapy with a wide array of agents used to manage vastly different diseases, were not controlled (Table 1).

MISCONCEPTION 2: ACTUAL NUMBERS OF LESIONS MATTER MOST Many brain metastasis randomized trials included the number of brain metastases identified using whatever imaging was available at that time as either a stratification or a prognostic variable.16-20 Common thresholding patterns included single lesions (often further subclassified as single or solitary), 2 to 4 tumors, fewer than 5, more than 5, or more than 10 tumors (with 1 or more of these categories being recognized as multiple lesions). This simple numerical approach arose from 4 biases. First, surgical resection was almost exclusively used in patients with 1 known tumor, although some small series in the literature reported resection of 2 to 3 lesions. Second, single-tumor patients (and. more importantly, those with solitary tumors) were believed to live longer and perhaps deserved greater attention in terms of the aggressiveness of achieving intracranial control. Third, the number of tumors was thought to be a reasonable estimate of tumor burden as well as tumor biology, leading to the concept of oligometastatic vs miliary spread of intracranial disease. Fourth, it was easy to count tumors for stratification and response analysis. In this way, an 8-mm diameter frontal lobe melanoma metastasis was given the same “weight” as a 2-cm diameter thalamic tumor from non-small cell lung cancer. Given the fact that patients with these differing lesions might present with vastly different symptomatic presentations with different degrees of

NEUROSURGERY

brain edema, potentially different radiation responses, and different forms of extracranial disease therapy, it is not surprising that clinical series containing such information could provide results that were often disparate. Even though higher quality studies attempted to match clinical criteria according to age, sex, number of patients with lung cancer, and Karnofsky Performance Scale score, they still failed to account for tumor biology and tumor volume. The number of tumors can affect how clinicians consider cognitive function. Certainly, in some patients, a higher cancer burden can affect cognition if increasing volumes of brain edema are located in brain regions important for cognition. If not, then cognition typically remains unchanged. If a decision is made to deliver WBRT simply due to the number of tumors, then that patient is placed at an increased risk of cognitive dysfunction should they survive long enough. Despite the continued interest in counting tumors, recent reports argue in a different direction. In a recent report by Likhacheva et al,21 251 patients with brain metastases underwent radiosurgery. One-year local control was 94%. The number of brain metastases (1-9) was not predictive of survival, local control, or distant brain failure. Total tumor volume greater than 2 mL was predictive of survival and local control. Similar findings were noted in 2006 by Bhatnagar et al22 in 205 patients from the University of Pittsburgh who had 4 to 18 brain tumors. Again, total tumor volume but not the number of metastases correlated with overall survival and local control. Recently, Baschnagel et al23 studied 250 patients who had 1 to 14 brain metastases initially managed with radiosurgery alone. The 1-year local control rate was 92%, and the median time to distant brain failure was 8 months. Total tumor volume was found to be a better predictor of overall survival than the number of brain metastases analyzed as a continuous variable and a better predictor of distant brain failure and even local brain tumor control. One cannot dismiss tumor number as being unimportant, but real tumor burden should be our new focus. Certainly, there are situations with high numbers of tumors in patients who have not had WBRT, where the number of tumors can help to predict distant brain failure in the WBRT-naive patient. This can help to triage the approach. There are also situations in patients with a single tumor in a resectable location where a resection may be an optimal choice, followed by local irradiation, to reduce the local recurrence risk.

MISCONCEPTION 3: THERE IS NO SUCH THING AS A SINGLE BRAIN METASTASIS Despite strong evidence suggesting that focal therapies for isolated focal metastatic lesions achieve statistically significant measures of tumor control and improved survival compared with more general treatment options like WBRT, it is still held that micrometastases (not apparent on high-resolution imaging) create a diffuse problem no matter how many tumors might be visible on an imaging study.



KONDZIOLKA ET AL


TABLE 1. Brain Metastases Clinical Trialsa Author

Year

Patchell et al Surgical resection 1 WBRT vs WBRT alone Komarnicky et al WBRT (2 doses) with or without misonidazole Mintz et al Surgical resection 1 WBRT vs WBRT alone Murray et al Accelerated hyperfractionated WBRT (54.4 Gy) vs 30 Gy/10 fractions Patchell et al

1990

Journal

No. of Citations

No. of Tumorsb

Histology

Primary Finding

Limitations

NEJM

1241

1

All

Resection beneficial

Mainly lung cancer

1991

IJROBP

132

No limit

All

Misonidazole negative

Poorly controlled

1996

Cancer

282

1

All

Resection not beneficial

Mainly lung cancer

1997

IJROBP

208

No limit

All

Hyperfractionation no benefit

Survival endpoint

1998

JAMA

579

1

All

WBRT beneficial post resection

60% lung cancer; only 2 melanoma cases

Surgical resection 1 WBRT vs surgical resection alone Kondziolka et al WBRT 1 radiosurgery vs WBRT alone Andrews et al

1999

IJROBP

432

2-4

All

SRS1WBRT beneficial

Poorly controlled

2004

Lancet

676

1-3

All

SRS1WBRT beneficial

Simplistic functional measures

WBRT 1 radiosurgery vs WBRT alone Meyers et al

2004

JCO

208

No limit

All

Motexafin gadolinium not beneficial

Lung cancer value?

WBRT with or without motexafin gadolinium Aoyama et al

2006

JAMA

445

1-4

All

WBRT did not improve survival, only local control

Mainly lung cancer

Radiosurgery 1 WBRT vs WBRT alone Suh et al

2006

JCO

73

No limit

All

No clear radiosensitization with efaproxiral

Needs further study in breast cancer

2009

Lancet Onc 247

1-3

All

WBRT worsened memory

Limited assessments

WBRT with or without efaproxiral Chang et al Radiosurgery alone vs radiosurgery 1 WBRT www.neurosurgery-online.com

a

NEJM, New England Journal of Medicine; WBRT, whole-brain radiotherapy; IJROBP, International Journal of Radiation Oncology Biology Physics; JAMA, Journal of the American Medical Association; JCO, Journal of Clinical Oncology; Lancet Onc, Lancet Oncology. Citations from Web of Knowledge (9/3/13). b Inclusion criteria.



In the 2010 brain metastases guidelines, the authors concluded that although both single-dose stereotactic radiosurgery (SRS) and WBRT were effective for treating patients with brain metastases, single-dose radiosurgery alone appeared to be superior to WBRT alone for patients with as many as 3 metastases in terms of a survival advantage.4 Thus, if deadly micrometastases frequently created a diffuse disease scenario, then WBRT populations should be associated with distinct survival advantages. But in no large study does the addition of WBRT to SRS improve survival. If micrometastases are present and not treated, they should become apparent on later imaging. Blindly managing assumed metastases is no longer best practice when such tumors can be defined with serial images. The research implications of this question are addressed in the following section.

MISCONCEPTION 4: WBRT IS ALWAYS HARMFUL EVENTUALLY The effects of WBRT on neurocognition can be viewed from several vantages. On the one hand, we know that WBRT can reduce the rate of distant metastases intracranially and can delay the time to relapse; from older data, this is believed to occur in about one-third of patients.24,25 It is possible that some of these new tumors could be symptomatic. However, when identified when small using serial imaging, they are usually asymptomatic. As noted earlier, many in this patient group could be defined with high-resolution imaging. Thus, helping to reduce overall tumor burden in this way could actually preserve neurocognitive function or slow down its decline if any new tumor appeared in a location critical to cognition. On the other hand, several studies have shown that WBRT is associated with a diffuse leukoencephalopathy, which can lead to an early decline in neurocognition and overall brain function.26 In a series of lung cancer patients surviving more than 1 year with brain tumors, 36 of 37 treated with WBRT had leukoencephalopathy compared with only 1 of 31 who had radiosurgery alone.26 So, where is the balance between the harmful effect of WBRT and any benefit in reducing tumor burden to preserve brain function? An argument can be made that cognitive decline in patients with multiple brain metastases can be due to overall disease burden rather than the effects of WBRT and that patients with large volumetric tumor burdens actually have a delay in cognitive decline when receiving WBRT compared with those who do not.27,28 However, the counterargument can be made, stating that WBRT may not even be required given the superiority of SRS in providing fast and effective tumor control and that any delayed problems after WBRT are simply avoided.20,29 Randomized studies have evaluated the role of a radiosurgery boost to WBRT, but have not evaluated SRS alone to WBRT alone. Thus, when used alone, it is not clear whether one modality improves outcome relative to another. However, the addition of WBRT does not appear to improve survival and may worsen function.20,24 The Chang et al29 study was stopped by their data monitoring committee after 64% accrual. They determined a 96% chance

NEUROSURGERY

that 52% of WBRT patients would have significant adverse cognitive effects (measured at 4 months) vs 24% of the radiosurgery alone group. The 4-month assessment provides one important measure, but more work in this area is needed, including analysis at earlier and later time points, as well as an evaluation of functional recovery. Other domains of functional independence should also be studied. A survival analysis of this trial also demonstrated significantly more early deaths in the WBRT arm, putatively due to progressive systemic disease. The impact of systemic progression of disease and associated management on cognitive decline therefore remains an “unsorted variable” and further underscores that when cognitive decline is evaluated in patients with brain metastases, all variables, including extracranial disease and the timing of any therapies, need to be accounted for. Clearly, a balanced approach that allows individualization to the patient is warranted. For instance, for high-functioning patients who are concerned about cognitive decline or for those in whom extended survivals may be expected, WBRT should be avoided.30 This is often considered to be important for diseases like melanoma, renal cell carcinoma, thyroid cancers, and most sarcomas, with the assumption that these histologies are considered universally “radioresistant.” However, in analysis of studies in which WBRT is excluded, melanoma is often identified as a high-risk feature for subsequent brain relapse.31,32 For example, in an analysis of 100 patients with brain metastases treated with radiosurgery only at the University of Alabama, the 1-year actuarial risk of relapse was only 17% in low-risk patients, but 82% in high-risk patients, defined as having 1 or more of the following 3 features: 3 or more metastases, the presence of extracerebral disease, or melanoma histology.33 Whether these patients should be managed with SRS and upfront WBRT as opposed to sequentially repeated SRS or inclusion of a drug-based therapy remains a poorly researched topic. The value of ipilumimab has been evaluated in several series, but firm conclusions cannot yet be reached.34,35 With the availability of new targeted therapies, the roles for both radiosurgery and WBRT may evolve further. Fortunately, the majority of new brain metastases are small and asymptomatic when detected early with elective, regularly scheduled MRI. In the past 2 years, 2 critical new trials have refocused attention on the role of WBRT. Brown et al36 recently reported results of RTOG 0614, a randomized trial evaluating the N-methyl-D-aspartate receptor agonist memantine as a potential neuroprotector during WBRT. Although the primary outcome of less decline in delayed recall did not reach statistical significance, overall, patients treated with memantine had better cognitive function over time in other domains. Specifically, memantine delayed time to cognitive decline and reduced the rate of decline in memory, executive function, and processing speed in patients receiving WBRT. A different approach has been to try to avoid irradiation of hippocampal subependymal stem cell niches that may be 1 anatomic site relevant in maintaining cognition. At the ASTRO 2013 Annual Meeting Plenary Session, Gondi et al37 presented data from the single-arm phase II RTOG



KONDZIOLKA ET AL

0933 trial in which the concept of WBRT with avoidance of the perihippocampal neurogenic stem cell compartment was tested in a study that enrolled 113 patients; 3 patients (4.5%) had progression in the hippocampal avoidance region, consistent with the expected event rate. Various memory domain tests represented the major endpoints, based mostly on the Hopkins Verbal Learning Test (HVLT) immediate, delayed, and total recall. Mean relative decline in HVLT delayed recall from baseline to 4 months was 7.0% (95% confidence interval [CI]: 24.7% to 18.7%, P = .0003, relative to the control group). Mean relative decline in HVLT total recall and HVLT immediate recall from baseline to 4 months was 3.6% (95% CI: 22.9% to 10.1%) and 1.6% (95% CI: 22.8% to 6.0%), respectively. The mean relative decline in HVLT delayed recall from baseline to 6 months was only 2.0% (95% CI: 29.2% to 13.1%). The results of this study were superior to the expected declines from historical controls. However, without a proper control group, the results should be interpreted cautiously and used for future clinical trial design. In 2011, the report of the European Organisation for Research and Treatment of Cancer 22952 to 26001 study (359 patients) provided evidence that adjuvant WBRT reduced intracranial relapse and neurological deaths compared with observation alone after radiosurgery or resection in patients with 1 to 3 brain tumors.38 However, it failed again to improve overall survival or the duration of functional independence. In addition, any subacute decline in performance status after WBRT may affect tolerance to further systemic therapy. These approaches clearly refocus the question “where is the appropriate balance in terms of cognitive deficits induced by tumor progression due to avoidance of WBRT vs the benefit of avoiding such progression in the brain?” Ongoing and planned phase III trials will potentially provide further data on this question. For such trials to be meaningful, we need to better understand the long- and short-term effects of various therapies on neurocognition by making neurocognitive function parameters clear, measurable endpoints in clinical trials.

MISCONCEPTION 5: MOST BRAIN METASTASES ARE SYMPTOMATIC AND SCREENING DOES NOT HAVE A MAJOR IMPACT In the past, the use of imaging for neurological screening was rare. Indeed, the majority of brain metastases were found because of headaches, seizures, or neurological deficits. Thus, clinical trials mainly evaluated larger tumors that typically had surrounding brain edema. This is not the case today. Centers that have a special interest in melanoma began to obtain screening MRI in the late 1990s, as patients who entered immunotherapy clinical trials needed to show that a brain tumor was not present or was treated and under control. Given the frequency with which melanoma metastasizes to the brain, MRI helped to identify many tumors when small and asymptomatic. Although routine brain screening is not commonplace, most oncologists now obtain an MRI at any hint of neurological


symptoms. Fine-cut MRI with enhanced-dose contrast has led to the identification of more and more small tumors.33 Whole-body positron emission tomography for staging sometimes shows a brain tumor before symptoms develop. In the University of Pittsburgh radiosurgery series, the average brain metastasis was 12 mm in diameter at diagnosis. The tumor was most commonly found on a screening scan and caused no or minimal symptoms. Thus, the goal of modern brain metastasis treatment is not usually to improve overall prognosis or function, but rather to prevent neurological deterioration while care for the extracranial cancer continues uninterrupted. This practice, now commonplace, differs greatly from the paradigm used to enter patients into the early surgical resection randomized trials in the 1990s. In the future, the inclusion of increasing numbers of asymptomatic brain metastases from screening may lead to a lead time bias for survival outcomes. The lack of symptomatic presentations was clear, even in the 2004 RTOG study of patients with 1 to 3 brain metastases.18 Of the subjects, 83% had no or only minor neurological symptoms. Again, this was mainly a lung cancer study. The number of patients in the WBRT plus radiosurgery group or the WBRT alone group with each primary tumor type was lung cancer (n = 105 and n = 106, respectively), breast cancer (n = 15 and n = 19, respectively), genitourinary cancer (n = 2 and n = 5, respectively), colon cancer (n = 4 and n = 2, respectively), melanoma (n = 7 and n = 9, respectively), other (n = 23 and n = 17, respectively), ovarian (n = 1 and n = 2, respectively), and unknown primary (n = 7 and n = 0, respectively). Recursive partitioning analysis class 2 comprised 73% of patients, and 56% had solitary brain tumors. Tsao et al39 provided a systematic review of the published evidence that related to symptoms, patients with a good or poor prognosis, and treatment aims. Of course, as noted previously, the base evidence for these recommendations is limiting.

FUTURE DIRECTIONS Current surgeons and oncologists think about cancer in ways different from how many clinical trials were conducted. Perhaps this is why even clinicians experienced in brain metastases management have great difficulty predicting the survival outcomes for individual real patients.40 To test this, we prospectively estimated survival in 150 consecutive cancer patients with brain metastases at presentation before undergoing radiosurgery. We recorded cancer type, number of brain metastases, neurological presentation, extracranial disease status, Karnofsky Performance Scale score, recursive partitioning analysis score, previous WBRT, and synchronous or metachronous presentation. We then asked 18 medical, radiation, or surgical oncologists to predict survival from the time of treatment. We found that the actual median patient survival was 10.3 months (95% CI: 6.4-14) and the median physician-predicted survival was 9.7 months. However, all physicians had individual patient survival predictions that were incorrect by as much as 12 to 18 months, and 14 of 18 physicians had individual predictions that were incorrect by more than 18 months. Of the 2700 predictions, 1226 (45%)




REFERENCES TABLE 2. Brain Metastases Clinical Trial Considerations Patient related Measures of brain location–related neurological function (ie, motor region metastases) Measures of neurocognition: serial testing over 3-24 months, account for tumors in cognitive/limbic locations Extracranial cancer burden detailed by organ system involved Serial measures of nutrition (ie, body weight) Corticosteroid use provided in dexamethasone equivalents Changes in systemic therapy that could affect brain tumor response and overall cancer control Tumor related Only 1 tumor histological type per study Tumor genetic/receptor subtypes provided and stratified Trial inclusion dependent on total brain tumor volume, not number of tumors Tumor diameters and volumes measured serially Tumor edema volume measured serially Presence or absence of leptomeningeal disease documented

were off by more than 6 months and 488 (18%) were off by more than 12 months.40 Thus, it is time for fresh thinking and new critical analyses. Table 2 details our recommendations for concepts to include in clinical trial designs. Some will require new approaches that will need validation. Outside of randomized trials, matched cohort studies can provide insight into why we choose to manage patients in certain ways.41,42 Cost-effectiveness analyses, when they contain actual cost data, also can be important science.43-45 Especially in this era of increasingly personalized medicine, one size fits all thinking is improper, especially for a diagnostic entity as wide and varied as brain metastasis. It is not surprising that results from studies aimed at exploiting tumor biology were negative, given the inherent biological differences in tumor types.46-48 Histology and overall tumor burden matter more than actual numbers and the presence or absence of lesions, but molecular and genetic subtyping will continue to unlock the real answers for why brain tumors develop in some patients with a given primary cancer, and in others, they do not.49 Ultimately, we hope that future interventions might allow all cancer patients to become more like the favorable group with as-of-yet unknown features that prevent metastatic spread in the first place. A Podcast associated with this article can be accessed online: http://links.lww.com/NEU/A622. Disclosures Drs Kondziolka and Loeffler have nothing to disclose. Dr Kalkanis is a consultant for Varian and Arbor. Dr Ahluwalia received honorarium and a travel grant from Elekta. Dr Mehta is a consultant for Abbott, BMS, Celldex, Novocure, Phillips, and Roche, owns stock in Pharmacyclics and Accuray, is on the Board of Directors of Pharmacyclics, and has received research funding from Novocure. The other authors have no personal financial or institutional interest in any of the drugs, materials, or devices described in this article.

NEUROSURGERY

1. Kalkanis SN, Linskey ME. Evidence-based clinical practice parameter guidelines for the treatment of patients with metastatic brain tumors: introduction. J Neurooncol. 2010;96(1):7-10. 2. Gaspar LE, Mehta MP, Patchell RA, et al. The role of whole brain radiation therapy in the management of newly-diagnosed brain metastases: a systematic review and evidence-based clinical practice guideline. J Neurooncol. 2010;96(1):17-32. 3. Kalkanis SN, Kondziolka D, Gaspar LE, et al. The role of surgical resection in the management of newly-diagnosed brain metastases: a systematic review and evidence-based clinical practice guideline. J Neurooncol. 2010;96(1):33-43. 4. Linskey M, Andrews D, Asher A, et al. The role of stereotactic radiosurgery in the management of newly diagnosed brain metastases: a systematic review and evidence-based clinical practice guideline. J Neurooncol. 2010;96(1):45-68. 5. Mehta MP, Paleologos NA, Mikkelsen T, et al. The role of chemotherapy in the management of newly-diagnosed brain metastases: a systematic review and evidence-based clinical practice guideline. J Neurooncol. 2010;96(1):71-83. 6. Lin NU, Lee EQ, Aoyama H, et al. Challenges relating to solid tumour brain metastases in clinical trials, part 1: patient population, response, and progression. A report from the RANO group. Lancet Oncol. 2013;14(10):e396-e406. 7. Lin NU, Wefel JS, Lee EQ, et al. Challenges relating to solid tumour brain metastases in clinical trials, part 2: neurocognitive, neurological, and quality-of-life outcomes. A report from the RANO group. Lancet Oncol. 2013;14(10):e407-e416. 8. Vogelbaum MA, Asher AL, Kondziolka D, et al. Modern treatment of cerebral metastases: Integrated Medical Learning (SM) at CNS 2007. J Neurooncol. 2009; 93(1):89-105. 9. Welsh JW, Komaki R, Amini A, et al. Phase II trial of erlotinib plus concurrent whole-brain radiation therapy for patients with brain metastases from non-smallcell lung cancer. J Clin Oncol. 2013;31(7):895-902. 10. Eichler AF, Kahle KT, Wang DL, et al. EGFR mutation status and survival after diagnosis of brain metastasis in nonsmall cell lung cancer. Neuro Oncol. 2010;12 (11):1193-1199. 11. Preusser M, Berghoff AS, Schadendorf D, Lin NU, Stupp R. Brain metastasis: opportunity for drug development? Curr Opin Neurol. 2012;25(6):786-794. 12. Narayana A, Mathew M, Tam M, et al. Vemurafenib and radiation therapy in melanoma brain metastases. J Neurooncol. 2013;113(3):411-416. 13. Ma M, Qian M, Lackaye D, et al. Challenging the current paradigm of melanoma progression: brain metastasis as isolated first visceral site. Neuro Oncol. 2012;14(7): 849-858. 14. Kondziolka D, Kano H, Harrison GL, et al. Stereotactic radiosurgery as primary and salvage management for brain metastases from breast cancer. Clinical article. J Neurosurg. 2011;114(3):792-800. 15. Patchell RA, Tibbs PA, Regine W, et al. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA. 1998;280(17):1485-1489. 16. Patchell RA, Tibbs PA, Walsh JW, et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med. 1990;322(8):494-500. 17. Mintz AH, Kestle J, Rathbone MP, et al. A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer. 1996;78(7):1470-1476. 18. Andrews DW, Scott CB, Sperduto PW. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomized trial. Lancet. 2004; 363(9422):1665-1672. 19. Kondziolka D, Patel A, Lunsford LD, Kassam A, Flickinger JC. Stereotactic radiosurgery plus whole brain radiotherapy versus radiotherapy alone for patients with multiple brain metastases. Int J Radiat Oncol Biol Phys. 1999;45(2):427-434. 20. Aoyama H, Shirato H, Tago M, et al. Stereotactic radiosurgery plus whole brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases. JAMA. 2006;295(21):2483-2491. 21. Likhacheva A, Pinnix CC, Parikh NR, et al. Predictors of survival in contemporary practice after initial radiosurgery for brain metastases. Int J Radiat Oncol Biol Phys. 2013;85(3):656-661. 22. Bhatnagar AK, Flickinger JC, Kondziolka D, Lunsford LD. Stereotactic radiosurgery for four or more intracranial metastases. Int J Radiat Oncol Biol Phys. 2006; 64(3):898-903. 23. Baschnagel AM, Meyer KD, Chen PY, et al. Tumor volume as a predictor of survival and local control in patients with brain metastases treated with gamma knife radiosurgery. J Neurosurg. 2013;119(5):1139-1144.



KONDZIOLKA ET AL

24. Varlotto JM, Flickinger JC, Niranjan A, Bhatnagar A, Kondziolka D, Lunsford LD. The impact of whole brain radiation therapy on the long-term control and morbidity of patients surviving more than one year after gamma knife radiosurgery for brain metastases. Int J Radiat Oncol Biol Phys. 2005;62(4):1125-1132. 25. Sneed PK, Lamborn KR, Forstner JM, et al. Radiosurgery for brain metastases: is whole brain radiotherapy necessary? Int J Radiat Oncol Biol Phys. 1999;43(3): 549-558. 26. Monaco EA, Faraji AH, Berkowitz O, et al. Leukoencephalopathy after whole brain radiation therapy plus radiosurgery versus radiosurgery alone for metastatic lung cancer. Cancer. 2013;119(1):226-232. 27. Li J, Bentzen SM, Renschler M, Mehta MP. Regression after whole-brain radiation therapy for brain metastases correlates with survival and improved neurocognitive function. J Clin Oncol. 2007;25(10):1260-1266. 28. Meyers CA, Smith JA, Bezjak A, et al. Neurocognitive function and progression in patients with brain metastases treated with whole-brain radiation and motexafin gadolinium: results of a randomized phase III trial. J Clin Oncol. 2004;22(1):157-165. 29. Chang EL, Wefel JS, Hess KR, et al. Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole brain irradiation: a randomized controlled trial. Lancet Oncol. 2009;10(11):1037-1044. 30. Kondziolka D, Martin JJ, Flickinger JC, et al. Long-term survivors after gamma knife radiosurgery for brain metastases. Cancer. 2005;104(12):2784-2791. 31. Mathieu D, Kondziolka D, Cooper PB, et al. Gamma knife radiosurgery in the management of malignant melanoma brain metastases. Neurosurgery. 2007;60(3): 471-481. 32. Sheehan JP, Sun MH, Kondziolka D, Flickinger J, Lunsford LD. Radiosurgery for patients with renal cell carcinoma metastatic to the brain: long-term outcomes and prognostic factors influencing survival and local tumor control. J Neurosurg. 2003; 98:342-349. 33. Sawrie SM, Guthrie BL, Spencer SA, et al. Predictors of distant brain recurrence for patients with newly diagnosed brain metastases treated with stereotactic radiosurgery alone. Int J Radiat Oncol Biol Phys. 2008;70(1):181-186. 34. Knisely JP, Yu JB, Flanigan J, Sznol M, Kluger HM, Chiang VL. Radiosurgery for melanoma brain metastases in the ipilumimab era and the possibility of longer survival. J Neurosurg. 2012;117(2):227-233. 35. Mathew M, Tam M, Ott PA, et al. Ipilumimab in melanoma with limited brain metastases treated with stereotactic radiosurgery. Melanoma Res. 2013;23(3):191-195. 36. Brown PD, Shook S, Laack NN, et al. Memantine for the prevention of cognitive dysfunction in patients receiving whole-brain radiotherapy: a randomized, doubleblind, placebo-controlled trial. Neurooncol. 2013;15(10):1429-1437. 37. Gondi V, Mehta M, Pugh S, et al. Memory preservation with conformal avoidance of the hippocampus during whole-brain radiotherapy for patients with brain metastasis: Preliminary results of RTOG 0933. Available at: https://www.astro.org/news-andmedia/news-releases/2013/avoiding-specific-region-of-the-brain-during-whole-brain. aspx. ASTRO 2013, Abstract LBA1. Accessed September 23, 2013. 38. Kocher M, Soffietti R, Abacioglu U, et al. Adjuvant whole-brain radiotherapy versus observation after radiosurgery or surgical resection of one to three cerebral metastases: results of the EORTC 22952-26001 study. J Clin Oncol. 2011;29(2):134-141. 39. Tsao MN, Rades D, Wirth A, et al. Radiotherapeutic and surgical management for newly diagnosed brain metastasis: an American Society for Radiation Oncology evidence-based guideline. Pract Radiat Oncol. 2012;2(3):210-225. 40. Kondziolka D, Parry PV, Lunsford LD, et al. The accuracy of predicting survival in individual patients with cancer. J Neurosurg. 2014;120(1):24-30. 41. Auchter RM, Lamond JP, Alexander E, et al. A multiinstitutional outcome and prognostic factor analysis of radiosurgery for resectable single brain metastasis. Int J Radiat Oncol Biol Phys. 1996;35(1):27-35. 42. Rades D, Kueter JD, Veninga T, Gliemroth J, Schild SE. Whole brain radiotherapy plus stereotactic radiosurgery versus surgery plus whole brain radiotherapy for 1-3 brain metastases: results of a matched pair analysis. Eur J Cancer. 2009;45(3):400-404. 43. Rutigliano MJ, Lunsford LD, Kondziolka D, Strauss MJ, Khanna V, Green M. The cost effectiveness of stereotactic radiosurgery versus surgical resection in the treatment of solitary metastatic brain tumors. Neurosurgery. 1995;37(3):445-453. 44. Mehta M, Noyes W, Craig B, et al. A cost-effectiveness and cost-utility analysis of radiosurgery vs. resection for single brain metastasis. Int J Radiat Oncol Biol Phys. 1997;39(2):445-454. 45. Lee WY, Cho DY, Lee HC, et al. Outcomes and cost-effectiveness of gamma knife radiosurgery and whole brain radiotherapy for multiple metastatic brain tumors. J Clin Neurosci. 2009;16(5):630-634.


46. Suh JH, Stea B, Nabid A, et al. Phase III study of efaproxiral as an adjunct to wholebrain radiation therapy for brain metastases. J Clin Oncol. 2006;24(1):106-114. 47. Murray KJ, Scott C, Greenberg HM, et al. A randomized phase III study of accelerated hyperfractionation versus standard in patients with unresected brain metastases: a report of the Radiation Therapy Oncology Group (RTOG) 9104. Int J Radiat Oncol Biol Phys. 1997;39(3):571-574. 48. Komarnicky LT, Phillips TL, Martz K, Asbell S, Isaacson S, Urtasun R. A randomized phase III protocol for the evaluation of misonidazole combined with radiation in the treatment of patients with brain metastases (RTOG-7916). Int J Radiat Oncol Biol Phys. 1991;20(1):53-58. 49. Anders CK, Deal AM, Miller CR, et al. The prognostic contribution of clinical breast cancer subtype, age, and race among patients with breast cancer brain metastases. Cancer. 2011;117(8):1602-1611.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.neurosurgery-online.com).

COMMENTS

T

his is a very interesting review that touches on important paradigm changes in brain metastasis management that have occurred over the past 2 decades. It is quite clear now from data from multiple sources that histology is paramount for brain metastases. Even subgroups of primary tumor sites that metastasize to the brain can have differing biological behaviors including distant brain failure, local control, and likelihood of tumor hemorrhage. The most common example for which histological subtype can affect management is breast cancer. Although Her2-positive breast cancers may metastasize to the brain at a higher rate, the use of antiHer2 agents has improved survival in this population such that these patients survive longer than patients with most other cancers that metastasize to the brain. Similar efforts are being made in the realm of lung cancer, and it is likely, as the authors point out, that future brain metastasis trials may be histology specific. The second 2 misconceptions that the authors address deal with the volume and number of brain metastases at the time of presentation. At the core, these misconceptions address the question of micrometastases vs reseeding of the brain by new metastases and whether burden of disease in the brain may predict outcomes. Although volume, as the authors point out, is an important variable, which likely has a large role in determining whether a patient actually dies of his or her brain metastasis, the total number of metastases in the brain should not be discounted as an important factor. Several series have demonstrated that the number of metastases at presentation are predictive of distant brain failure after radiosurgery without whole-brain radiotherapy (WBRT). Again, multiinstitutional efforts are being made at present to determine factors that may differentiate between patients who will fail early (because of presence of micrometastases) vs those who may fail later (due to reseeding). The importance of such efforts will be to help triage the proper patients to radiosurgery vs WBRT, given the expense of the former and the late cognitive toxicity of the latter. The fourth misconception addresses the cognitive toxicity causes by WBRT, which can occasionally be debilitating, but can also be quite tolerable. It has previously been shown that failure in the brain can affect cognitive status at least as much as WBRT. Also, the subacute worsening in performance status, as pointed out by the authors, should not be discounted as this can affect whether patients are eligible for systemic therapy or when they become eligible. A major focus of prospective studies has been avoidance of the toxicities of WBRT, as the authors mention the 2 prospective RTOG studies. Future studies may also focus on




cytoprotective agents that may change the biology of radiation injury before the onset of symptoms, as well as agents that help to address quality of life during and after WBRT. Finally, the authors address what has been a migration within brain metastases of asymptomatically detected metastases. Much of the traditional data generated for brain metastases are based on trials of patients with symptomatic brain metastases. Emerging data would suggest that patients with symptomatic brain metastases have worsened outcomes. This stage of migration probably plays some role in the improvement in outcomes of brain metastases over time. Further trials will likely need to address the proper populations to screen for brain metastases, the proper screening after treatment of brain metastases, and the cost-effectiveness and clinical efficacy of such screening. In summary, the authors describe several important controversies in brain metastasis management for which there have been paradigm changes over time caused by evolving data and improving therapies with a focus on prospectively gathered data. Given the impressive rate at which brain metastasis data have been gathered over the past several years, it will be important that practitioners also analyze the data carefully and consider the levels of evidence that support management. The authors should be commended on their efforts in this article to boil down the literature to several important controversies. These controversies will continue to evolve over the next several years as molecular subtyping and systemic therapies continue to improve. Michael D. Chan Winston-Salem, NC

T

his provocative article raises numerous valuable points worthy of consideration by neurosurgeons interpreting the medical literature to make treatment decisions for brain metastases. Arguably, some of these “misconceptions” may not have been as prevalent as the authors imply. For example, I have not found it widely stated that most brain metastases are asymptomatic; rational reasons for not screening for brain metastases include anticipated low yield, cost, and uncertainty of the benefit of treating presymptomatic lesions. Additionally, the notion that a belief exists that there is no such thing as a single brain metastasis is unsupported. Certainly, our understanding of brain metastasis biology has grown in the past 20 years, especially with regard to the behavior of different histologies. Indeed, the authors themselves have previously lumped brain metastases from different histologies together in prospective clinical trials. The authors do well to emphasize that lesion number may not be the be all and end all, although the relative

NEUROSURGERY

importance of lesion number vs lesion volume for various endpoints requires further analysis. Among the difficulties in performing clinical trials in this population is the lack of well-accepted response criteria. The Response Assessment in Neuro-Oncology (RANO) working group is working to address this problem, with input of neurosurgeons, neuro-oncologists, medical and radiation oncologists, among others. The final point to emphasize is that the overwhelming majority of patients with brain metastases die of systemic cancer, not their brain metastases. Thus, reliance on overall survival in some of the studies cited to support the authors’ contentions is misguided. If the volume of brain metastases correlates with overall survival, this may be because of a correlation of brain metastasis volume with burden of systemic disease. Development of better endpoints for brain metastasis studies is a complex challenge that the RANO group is undertaking. David Schiff Charlottesville, VA

CME QUESTIONS: 1. An asymptomatic patient with known melanoma is found to have a new isolated 1 cm intracranial metastatic tumor. What is the most significant factor in determining the prognosis for this patient? A. Age B. Gender C. Tumor size D. Single brain metastasis E. Extracranial melanoma 2. What factor is most predictive of response to radiotherapy for patients with brain metastases? A. Total tumor volume B. Tumor number C. Tumor volume and number D. Anatomical location 3. Single dose radiosurgery has a survival advantage over WBRT alone for patients with what number of brain metastases? A. 1-3 B. 3-5 C. 5-7 D. 7-9



Whole grains in the renal diet--is it time to reevaluate their role?

Toxic epidermal necrolysis and early transfer to a regional burn unit: is it time to reevaluate what we teach?

Hepatitis B management: it is time to change the strategy.

It is Time to Renew!

The clinical course of patients with adrenal incidentaloma: is it time to reconsider the current recommendations?

Management of newborns born to mothers with chorioamnionitis: is it time for a kinder, gentler approach?

Abusive patients: Is it time for accountability?

Follow-up of patients with early breast cancer: is it time to rewrite the story?

The Interdisciplinary Management of Brain Metastases.

Is it time for a unified definition of case management?

Capillary refill time. Is it time to fill the gaps?

Updates in the management of brain metastases.

The management of central venous catheters and infection control: is it time to change our approach?

Is it time to train neurologists in the management of mood and anxiety disorders?

Is it time to rethink cord management when resuscitation is needed?

Time to treatment of cauda equina syndrome: a time to reevaluate our clinical decision.

Capacity of patients with brain metastases to make treatment decisions.

Stereotactic radiosurgery (SRS) in the modern management of patients with brain metastases.

It is time to move forward with Waldenström!

COPD: it is time to change!

STRICTA: is it time to do more?

Aprotinin: is it time to reconsider?

Is it time to measure apolipoprotein B?

Is it time to rethink syphilis control?