Multi-institutional competing risks analysis of distant brain failure and salvage patterns after upfront radiosurgery without whole brain radiotherapy for brain metastasis

Multi-institutional competing risks analysis of distant brain failure and salvage patterns after... Abstract Background In this study, we use a competing risks analysis to assess factors predictive of early-salvage whole brain radiotherapy (WBRT) and early death after upfront stereotactic radiosurgery (SRS) alone for brain metastases in an attempt to identify populations that benefit less from upfront SRS. Patients and methods Patients from eight academic centers were treated with SRS for brain metastasis. Competing risks analysis was carried out for distant brain failure (DBF) versus death prior to DBF as well as for salvage SRS versus salvage WBRT versus death prior to salvage. Linear regression was used to determine predictors of the number of brain metastases at initial DBF (nDBF). Results A total of 2657 patients were treated with upfront SRS alone. Multivariate analysis (MVA) identified an increased hazard of DBF associated with increasing number of brain metastases (P < 0.001), lowest SRS dose received (P < 0.001), and melanoma histology (P < 0.001), while there was a decreased hazard of DBF associated with increasing age (P < 0.001), KPS < 70 (P < 0.001), and progressive systemic disease (P = 0.004). MVA for first salvage SRS versus WBRT versus death prior to salvage revealed an increased hazard of first salvage WBRT seen with increasing number of brain metastases (P < 0.001) and a decreased hazard with widespread systemic disease (P = 0.002) and increasing age (P < 0.001). Variables associated with nDBF included age (P = 0.02), systemic disease status (P = 0.03), melanoma histology (P = 0.05), and initial number of brain metastases (P < 0.001). Conclusions Patients with a higher initial number of brain metastases were more likely to experience DBF, have a higher nDBF, and receive early-salvage WBRT, while patients who were older, had lower KPS, or had more systemic disease were more likely to experience death prior to DBF or salvage WBRT. radiosurgery, brain metastasis, distant brain failure, competing risks, whole brain radiotherapy Key Message This study assessed the factors predictive of the competing risks of early-salvage whole brain radiotherapy (WBRT) versus early death for patients with brain metastasis treated upfront with radiosurgery. A higher number of initial brain metastases were predictive of early-salvage WBRT, whereas older age, lower performance status, and extracranial disease were predictive of death prior to salvage. Introduction The use of stereotactic radiosurgery (SRS) for the treatment of brain metastases has increased in recent years due both to the increasing availability of technology [1] as well as publication of randomized trials showing improvements in cognition over combined-modality therapy based on whole brain radiotherapy (WBRT) [2, 3]. Previously published randomized trials comparing SRS alone to SRS + WBRT were limited to patients having four or fewer brain metastases [2–5]; however, a recent multi-institutional prospective observational study from Japan has reported that patients with up to 10 brain metastases can be treated with SRS alone, and that these patients have survival similar to patients with 2–4 metastases [6]. As such, it has become less clear as to which patients should be treated with upfront SRS without WBRT. The cost of SRS is approximately three- to sixfold higher than WBRT in the United States [7, 8], with overall differences in cost exacerbated by the fact that SRS patients also require more salvage procedures than those receiving WBRT [4]. With ∼170 000 patients in the United States diagnosed with brain metastases each year [9], the appropriation of resources for brain metastasis management is a large-scale clinical dilemma. The use of SRS monotherapy appears to improve quality of life [10] and cognition [2, 3] over the use of WBRT, and although neurocognitive symptoms often begin to manifest within 3 months following WBRT [11], there is no apparent plateau in incidence or severity of symptoms [12]. Based upon cognitive outcomes, it would therefore seem reasonable to treat with SRS rather than WBRT when possible, however, patients who have poor life expectancy or rapid development of multiple new brain metastases may not derive significant benefit from SRS as they are unlikely to live long enough to experience WBRT neurocognitive sequelae. The goal of the present study is to assess, in patients treated upfront with SRS alone, the competing risks of distant brain failure (DBF) and the need for salvage therapy versus the risk of death prior to DBF or salvage therapy. A total of eight academic medical centers with high-volume SRS programs contributed data to the current analysis. Methods Data acquisition This study was approved by the Institutional Review Board at all eight participating institutions. Data were reviewed and collected at all institutions for patients treated between 1 January 2000 and 1 January 2014. Patients were included in the study if they underwent SRS following initial diagnosis with brain metastasis. Electronic medical records were reviewed, and putative predictors of DBF and salvage therapy were recorded for each patient [13]. Extent of extracranial disease was defined based upon prior publications [14]. SRS doses varied by institution, but were generally determined based upon the guidelines published by Shaw et al. for single fraction radiosurgical treatment of brain metastases [15]. As SRS dose is inversely proportional to tumor volume, the lowest SRS dose prescribed to any lesion was recorded for each patient to serve as a surrogate for tumor volume. The majority of patients underwent follow-up with magnetic resonance imaging (MRI) 1–3 months after their SRS procedures, with subsequent MRIs every 3 months thereafter. DBF events were defined as the development of one or more new brain metastases outside of the previous SRS target volume(s). DBF events and the number of brain metastases at time of DBF were determined based upon MRI results. Statistical analysis Median follow-up and time-to-event outcomes were defined beginning at the time of SRS and extending to the time of most recent follow-up or to the event of interest. Time-to-event outcomes were summarized using the Kaplan–Meier method. Cumulative incidences were estimated for DBF, death prior to DBF/salvage, first salvage SRS, and first salvage WBRT [16]. Competing risks models were developed to determine the single variable subdistribution hazard ratios (HR) associated with each predictor for each of these events. Statistically significant (P < 0.05) variables identified on univariate analysis were included in forward stepwise regressions to identify the multivariable models that minimized the Akaike information criterion (AIC) [17]. These results were used to guide purposeful development of multivariate competing risks models for DBF versus death prior to DBF as well as first salvage with SRS versus first salvage with WBRT versus death prior to salvage. Linear regression was carried out for predictor variables of interest for the outcome of number of brain metastases at time of DBF (nDBF). The linear regression analysis was limited to patients who experienced DBF and who had a quantifiable number of metastases at time of failure. Stepwise regression was used to identify the nDBF multiple regression model with the lowest AIC [17]. Statistical analysis was carried out using R version 3.2.1 software (R Foundation for Statistical Computing, Vienna, Austria). Results Clinical outcomes Of 2657 patients treated with upfront SRS alone, 2208 (83.1%) patients were deceased at the time of analysis, with DBF occurring in 1244 (46.8%) of all patients. Patient characteristics can be found in Table 1. Median overall survival (OS) was 9.8 months (95% CI: 9.3–10.4 months). After initial SRS, a total of 619 (23.3%) patients received salvage SRS and 447 (16.8%) of patients received salvage WBRT after initial SRS. Of the 447 patients who received salvage WBRT, 336 (75.2%) received salvage WBRT as their first salvage treatment. Table 1. Patient characteristics   n/median (%/IQR)  Total patients  2657  Age, year  61 (53, 70)  Gender     Female  1331 (50.1)   Male  1326 (49.9)  Ethnicity     Asian  65 (2.5)   African-American  181 (7.0)   Latino  46 (1.8)   White  2232 (86.0)   Other  72 (2.7)  Primary malignancy     Breast  373 (14.0)    Her2+  159 (43.6)    Her2−  159 (43.6)    Her2 unknown  55 (14.7)   Lung  1279 (48.1)    Adeno  649 (50.7)    SqCC  111 (8.7)    NSCLC NOS  392 (30.6)    Other  127 (9.9)   RCC  272 (10.2)   Melanoma  479 (18.0)   GI  121 (5.2)   GU  9 (0.4)   Gyn  41 (1.8)   H/N  19 (0.8)   Sarcoma  14 (0.6)   Other  50 (1.9)  Systemic disease burden     None  376 (14.2)   Oligometastatic (≤5 extracranial metastases)  995 (37.4)   Widespread (>5 extracranial metastases)  1199 (45.1)   Unknown  87 (3.3)  Systemic disease status     Stable  1224 (46.1)   Progressive  1254 (47.2)   Unknown  179 (6.7)  KPS     <70  199 (7.9)   ≥70  2326 (92.1)  RPA     1  391 (14.7)   2  1935 (72.9)   3  199 (7.5)  No. of initial brain metastases     1  1295 (48.7)   2  584 (22.0)   3  350 (13.2)   4  176 (6.6)   5+  252 (9.5)  Lowest SRS margin dose  18.5 (17.0, 21.0)    n/median (%/IQR)  Total patients  2657  Age, year  61 (53, 70)  Gender     Female  1331 (50.1)   Male  1326 (49.9)  Ethnicity     Asian  65 (2.5)   African-American  181 (7.0)   Latino  46 (1.8)   White  2232 (86.0)   Other  72 (2.7)  Primary malignancy     Breast  373 (14.0)    Her2+  159 (43.6)    Her2−  159 (43.6)    Her2 unknown  55 (14.7)   Lung  1279 (48.1)    Adeno  649 (50.7)    SqCC  111 (8.7)    NSCLC NOS  392 (30.6)    Other  127 (9.9)   RCC  272 (10.2)   Melanoma  479 (18.0)   GI  121 (5.2)   GU  9 (0.4)   Gyn  41 (1.8)   H/N  19 (0.8)   Sarcoma  14 (0.6)   Other  50 (1.9)  Systemic disease burden     None  376 (14.2)   Oligometastatic (≤5 extracranial metastases)  995 (37.4)   Widespread (>5 extracranial metastases)  1199 (45.1)   Unknown  87 (3.3)  Systemic disease status     Stable  1224 (46.1)   Progressive  1254 (47.2)   Unknown  179 (6.7)  KPS     <70  199 (7.9)   ≥70  2326 (92.1)  RPA     1  391 (14.7)   2  1935 (72.9)   3  199 (7.5)  No. of initial brain metastases     1  1295 (48.7)   2  584 (22.0)   3  350 (13.2)   4  176 (6.6)   5+  252 (9.5)  Lowest SRS margin dose  18.5 (17.0, 21.0)  Adeno, adenocarcinoma; GI, gastrointestinal; GU, genitourinary; Gyn, gynecologic; H/N, head and neck; IQR, interquartile range; KPS, Karnofsky performance status; NSCLC, non-small-cell lung cancer; RPA, recursive partitioning analysis; SqCC, squamous cell carcinoma; SRS, stereotactic radiosurgery. DBF versus death prior to DBF The estimated cumulative incidences of DBF at 6, 12, and 24 months were 27.7%, 38.5%, and 46.0%, respectively (Figure 1). Predictors of DBF identified on univariate analysis included age (P < 0.001), KPS (P < 0.001), systemic disease status (P < 0.001), number of brain metastases (P < 0.001), SRS dose (P < 0.001), and melanoma histology (P < 0.001). Cumulative incidences of DBF as stratified by variables included in the final multivariate analysis (MVA) model can be found in Figure 2. Cumulative incidences of DBF at 1 and 2 years as stratified by all clinical variables of interest can be found in supplementary Table S1, available at Annals of Oncology online. MVA identified an increased hazard of DBF associated with increasing number of brain metastases (HR = 1.09, P < 0.001), lowest SRS dose (HR = 1.04, P <0.001), and melanoma histology (HR = 1.37, P < 0.001), while there was a decreased hazard of DBF associated with increasing age (HR = 0.99, P < 0.001), KPS < 70 (HR = 0.59, P < 0.001), and progressive systemic disease (HR = 0.84, P = 0.004) (Table 2). Table 2. Multivariate proportional subdistribution hazard models for DBF, first salvage WBRT, and first salvage SRS   HR  95% CI  P-value  Distant brain failure (DBF)         Age (continuous)  0.99  0.98–0.99  <0.001   KPS <70  0.59  0.44–0.79  <0.001   Melanoma histology  1.37  1.17–1.60  <0.001   Progressive systemic disease  0.84  0.74–0.94  0.004   No. of brain metastases (continuous)  1.09  1.05–1.13  <0.001   Lowest SRS dose (continuous)  1.04  1.02–1.06  <0.001  First salvage WBRT   Age (continuous)  0.98  0.97–0.99  <0.001   Widespread extracranial disease  0.67  0.52–0.87  0.002   Progressive systemic disease  0.65  0.50–0.83  <0.001   No. of brain metastases (continuous)  1.09  1.05–1.13  <0.001  First salvage SRS   KPS <70  0.30  0.17–0.53  <0.001   Progressive systemic disease  0.69  0.58–0.83  <0.001   Lowest SRS dose (continuous)  1.12  1.08–1.15  <0.001    HR  95% CI  P-value  Distant brain failure (DBF)         Age (continuous)  0.99  0.98–0.99  <0.001   KPS <70  0.59  0.44–0.79  <0.001   Melanoma histology  1.37  1.17–1.60  <0.001   Progressive systemic disease  0.84  0.74–0.94  0.004   No. of brain metastases (continuous)  1.09  1.05–1.13  <0.001   Lowest SRS dose (continuous)  1.04  1.02–1.06  <0.001  First salvage WBRT   Age (continuous)  0.98  0.97–0.99  <0.001   Widespread extracranial disease  0.67  0.52–0.87  0.002   Progressive systemic disease  0.65  0.50–0.83  <0.001   No. of brain metastases (continuous)  1.09  1.05–1.13  <0.001  First salvage SRS   KPS <70  0.30  0.17–0.53  <0.001   Progressive systemic disease  0.69  0.58–0.83  <0.001   Lowest SRS dose (continuous)  1.12  1.08–1.15  <0.001  CI, confidence interval; HR, hazard ratio; KPS, Karnofsky performance status; SRS, stereotactic radiosurgery; WBRT, whole brain radiotherapy. Figure 1. View largeDownload slide Cumulative incidence of DBF versus death prior to DBF (A) and first salvage versus death prior to salvage (B). DBF, distant brain failure; SRS, stereotactic radiosurgery; WBRT, whole brain radiotherapy. Figure 1. View largeDownload slide Cumulative incidence of DBF versus death prior to DBF (A) and first salvage versus death prior to salvage (B). DBF, distant brain failure; SRS, stereotactic radiosurgery; WBRT, whole brain radiotherapy. Figure 2. View largeDownload slide Cumulative incidences of DBF as stratified by histology (A), number of initial brain metastases (B), age (C), KPS (D), lowest initial SRS dose (E), and systemic disease status (F). DBF, distant brain failure; KPS, Karnofsky performance status; met, brain metastasis; RCC, renal cell carcinoma; SRS, stereotactic radiosurgery. Figure 2. View largeDownload slide Cumulative incidences of DBF as stratified by histology (A), number of initial brain metastases (B), age (C), KPS (D), lowest initial SRS dose (E), and systemic disease status (F). DBF, distant brain failure; KPS, Karnofsky performance status; met, brain metastasis; RCC, renal cell carcinoma; SRS, stereotactic radiosurgery. First salvage: SRS versus WBRT versus death prior to salvage The estimated cumulative incidences of first salvage with SRS at 6, 12, and 24 months were 10.1%, 15.4%, and 19.4%, respectively (Figure 1). The estimated cumulative incidences of first salvage with WBRT at 6, 12, and 24 months were 8.5%, 11.1%, and 12.9%, respectively. MVA for first salvage WBRT revealed an increased hazard with increasing number of brain metastases (HR = 1.09, P < 0.001) and a decreased hazard with widespread systemic disease (HR = 0.67, P = 0.002) and increasing age (HR = 0.98, P < 0.001). MVA for first salvage SRS revealed an increased hazard with increasing lowest SRS dose (HR = 1.12, P <0.001), and a decreased hazard with KPS < 70 (HR = 0.30, P <0.001). Progressive systemic disease was associated with an increased hazard of death prior to salvage with either SRS or WBRT (HR = 1.69, P < 0.001), and a decreased hazard of both first salvage WBRT (HR = 0.65, P <0.001) and first salvage SRS (HR = 0.69, P < 0.001) on respective MVAs. MVA results for first salvage WBRT and first salvage SRS can be found in Table 2. Cumulative incidences of first salvage WBRT as stratified by variables included in the final MVA model can be found in supplementary Figure S1, available at Annals of Oncology online. Cumulative incidences of first salvage WBRT at 1 and 2 years as stratified by all clinical variables of interest can be found in supplementary Table S1, available at Annals of Oncology online. Number of brain metastases at time of DBF Of the 1244 patients with DBF, there were 924 patients (74.3%) who had a quantifiable number of brain metastases at time of initial DBF (nDBF). On multiple regression, increasing age was associated with lower nDBF (P = 0.02), while systemic disease status (P = 0.03), melanoma histology (P = 0.05), and higher number of initial brain metastases (P < 0.001) were associated with higher nDBF, as shown in supplementary Table S2, available at Annals of Oncology online. When patients were analyzed by histologic subset, initial number of brain metastases was predictive of nDBF for lung (P = 0.001), RCC (P = 0.006), and melanoma (P < 0.001), but was not predictive of nDBF for patients with breast cancer (P = 0.29) (supplementary Figure S2, available at Annals of Oncology online). Discussion Identification of populations that do and do not benefit from SRS will be critical for the proper appropriation of a costly resource. Treating patients with a severely limited life expectancy is unlikely to be a cost-effective use of SRS given the high cost relative to a short-term benefit. As the cognitive toxicity of WBRT is permanent, those with a longer life expectancy may experience the greatest benefit from postponement or complete avoidance of WBRT [12]. The median life expectancy of patients with brain metastases in the modern era is ∼8–10 months, though the range is wide [4–6]. The tail of this range will likely broaden further with the increasing use and availability of novel systemic treatment options such as checkpoint inhibitors that seem to be significantly prolonging survival for some patients, as notably shown recently with advanced melanoma [18]. As such, prognostic indices that help to identify populations with particularly poor prognoses will be helpful for clinical decision-making [19]. In the current analysis, poor performance status, progressive systemic disease and larger brain metastases (lower SRS dose) predicted for early death after SRS alone. Those with a low burden of brain disease and short life expectancy may be best treated with supportive care or observed until symptomatic [20]. Those with a high total volume of metastatic brain disease have a poorer prognosis and greater likelihood of dying of brain metastases, and may warrant consideration of upfront WBRT [21], as large brain metastases treated with SRS alone have previously been associated with a higher likelihood of neurologic death [22]. These patients may be best treated with fractionated SRS or a multi-modality approach such as WBRT followed by SRS boost, SRS followed by surgery, or surgery followed by radiotherapy. The threshold number of brain metastases that should be treated with SRS monotherapy remains controversial. Published randomized trials have used a limit of 3–4 brain metastases [2–4], though a prospective study from Japan suggests that up to 10 can be treated with equivalent survival [6]. Traditionally, such a limit was posed because treating larger numbers of metastases would overwhelm the capacity of many linear accelerator-based radiosurgery platforms. With recent improvements, these technical limitations are no longer as relevant [1]. Multiple series have suggested that the total volume of brain metastases is likely more important than the number of metastases at time of SRS with regards to OS [6, 23]. However, other series have implied that a greater number of metastases predicts for earlier DBF [13, 24], as seen in the present series. While the optimal upfront modality for brain metastasis management has been heavily investigated, the criteria for optimal use of each salvage modality remain undetermined. As previously mentioned, the number of new metastases at time of DBF is a dominant determinant of whether WBRT or SRS will be used in the salvage setting [25, 26]. Patients who receive multiple salvage SRS procedures for a limited number of brain metastases may represent a population who have survived the competing risks of early death and early WBRT. Conversely, those who experience an acceleration of intracranial progression likely require WBRT at the point of acceleration. The recently introduced concept of brain metastasis velocity is one in which the re-seeding rate of the brain is used to triage patients to salvage modality [25]. While this concept remains to be validated, it has potential in aiding practice decisions and is supported by the finding in the current study that patients with greater number of metastases at time of diagnosis also have a greater number at time of treatment failure. Brain metastasis patients have a large competing risk of dying of their extracranial disease, with only 20% of patients actually dying of brain metastases [5]. The results of the current study support this notion, as patients with widespread extracranial disease were less likely to receive salvage WBRT. Previous reports have suggested, though, that those with early CNS failure are at higher risk of dying of neurologic death [27]. Patients who require early WBRT often do so because of rapid multifocal DBF, leptomeningeal seeding or large volume local failure. Should these events occur within several weeks after brain metastasis diagnosis, patients may be less likely to benefit from SRS because of the lack of meaningful postponement of the toxicities of WBRT. There have been recent attempts to predict patients who will have rapid DBF and require early WBRT [13, 28, 29]. A limitation of these prior models is that they do not predict the actual number of lesions at time of DBF. There are several limitations to the current series. While the number of patients was large, the dataset was derived from several high volume academic centers in the United States and Canada. Given the retrospective nature of the dataset and the patient selection bias of North American academic centers, the results will need to be validated prospectively. Although the variables that were systematically collected by every institution were fairly comprehensive, systemic treatment details were not routinely collected by all institutions, and as such were not able to be reported in this study. Variation in the salvage philosophies at each of the centers may have influenced the decision to treat patients with salvage WBRT, as there were no standard guidelines across institutions regarding the use of SRS versus WBRT for salvage, with most institutions generally favoring salvage with SRS rather than WBRT when feasible. There were also no universal guidelines or protocol for MRI acquisition across institutions, in either the initial pre-treatment or follow-up setting. One implication of this is that similarly sized, small brain metastases may have potentially been identified and treated at different times across institutions, based upon differences in institutional MRI capabilities and frequency of follow-up imaging. Funding None declared. Disclosure MA: consulting and grant from Elekta; grant support from Boehringer Ingelheim, Bristol-Myers Squibb, Novartis, Spectrum Pharmaceuticals, Tracon Pharmaceuticals, Novocure; Consultant for Merck, Genentech/Roche, Incyte, Caris Lifesciences, Monteris Medical, MRI interventions, Inc. SC: honorarium from Varian Medical Systems, Abbvie, and Zeiss. All remaining authors have declared no conflicts of interest. References 1 Thomas EM, Popple RA, Wu X et al.   Comparison of plan quality and delivery time between volumetric arc therapy (RapidArc) and Gamma Knife radiosurgery for multiple cranial metastases. Neurosurgery  2014; 75( 4): 409– 417; discussion 417–418. Google Scholar CrossRef Search ADS PubMed  2 Chang EL, Wefel JS, Hess KR et al.   Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain irradiation: a randomised controlled trial. Lancet Oncol  2009; 10( 11): 1037– 1044. Google Scholar CrossRef Search ADS PubMed  3 Brown PD, Jaeckle K, Ballman KV et al.   Effect of radiosurgery alone vs radiosurgery with whole brain radiation therapy on cognitive function in patients with 1 to 3 brain metastases: a randomized clinical trial. JAMA  2016; 316( 4): 401– 409. Google Scholar CrossRef Search ADS PubMed  4 Aoyama H, Shirato H, Tago M et al.   Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA  2006; 295( 21): 2483– 2491. Google Scholar CrossRef Search ADS PubMed  5 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. JCO  2011; 29( 2): 134– 141. Google Scholar CrossRef Search ADS   6 Yamamoto M, Serizawa T, Shuto T et al.   Stereotactic radiosurgery for patients with multiple brain metastases (JLGK0901): a multi-institutional prospective observational study. Lancet Oncol  2014; 15( 4): 387– 395. Google Scholar CrossRef Search ADS PubMed  7 Lester SC, Taksler GB, Kuremsky JG et al.   Clinical and economic outcomes of patients with brain metastases based on symptoms: an argument for routine brain screening of those treated with upfront radiosurgery. Cancer  2014; 120( 3): 433– 441. Google Scholar CrossRef Search ADS PubMed  8 Brown PD, Kee AY, Eshleman JS, Fiveash JB, Adjuvant whole brain radiotherapy: strong emotions decide but rationale studies are needed: in regard to Brown et al. (Int J Radiat Oncol Biol Phys 2008; 70: 1305–1309). In reply to Drs. Larson and Sahgal. Int J Radiat Oncol Biol Phys  2009; 75( 1): 316– 317. Google Scholar CrossRef Search ADS PubMed  9 Ellis TL, Neal MT, Chan MD. The role of surgery, radiosurgery and whole brain radiation therapy in the management of patients with metastatic brain tumors. Int J Surg Oncol  2012; 2012: 952345. Google Scholar PubMed  10 Soffietti R, Kocher M, Abacioglu UM et al.   A European Organisation for research and treatment of cancer phase III trial of adjuvant whole-brain radiotherapy versus observation in patients with one to three brain metastases from solid tumors after surgical resection or radiosurgery: quality-of-life results. J Clin Oncol  2013; 31( 1): 65– 72. Google Scholar CrossRef Search ADS PubMed  11 Brown PD, Pugh S, Laack NN et al.   Memantine for the prevention of cognitive dysfunction in patients receiving whole-brain radiotherapy: a randomized, double-blind, placebo-controlled trial. Neuro Oncol  2013; 15( 10): 1429– 1437. Google Scholar CrossRef Search ADS PubMed  12 Greene-Schloesser D, Robbins ME, Peiffer AM et al.   Radiation-induced brain injury: a review. Front Oncol  2012; 2: 73. Google Scholar CrossRef Search ADS PubMed  13 Ayala-Peacock DN, Peiffer AM, Lucas JT et al.   A nomogram for predicting distant brain failure in patients treated with gamma knife stereotactic radiosurgery without whole brain radiotherapy. Neuro Oncol  2014; 16( 9): 1283– 1288. Google Scholar CrossRef Search ADS PubMed  14 Harris S, Chan MD, Lovato JF et al.   Gamma knife stereotactic radiosurgery as salvage therapy after failure of whole-brain radiotherapy in patients with small-cell lung cancer. Int J Radiat Oncol Biol Phys  2012; 83( 1): e53– e59. Google Scholar CrossRef Search ADS PubMed  15 Shaw E, Scott C, Souhami L et al.   Single dose radiosurgical treatment of recurrent previously irradiated primary brain tumors and brain metastases: final report of RTOG protocol 90-05. Int J Radiat Oncol Biol Phys  2000; 47( 2): 291– 298. Google Scholar CrossRef Search ADS PubMed  16 Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc  1999; 94( 446): 496– 509. Google Scholar CrossRef Search ADS   17 Akaike H. A new look at the statistical model identification. IEEE Trans Automat Contr  1974; 19( 6): 716– 723. Google Scholar CrossRef Search ADS   18 Wolchok JD, Chiarion-Sileni V, Gonzalez R et al.   Overall survival with combined nivolumab and ipilimumab in advanced melanoma. N Engl J Med  2017; 377( 14): 1345– 1356. Google Scholar CrossRef Search ADS PubMed  19 Sperduto PW, Chao ST, Sneed PK et al.   Diagnosis-specific prognostic factors, indexes, and treatment outcomes for patients with newly diagnosed brain metastases: a multi-institutional analysis of 4, 259 patients. Int J Radiat Oncol Biol Phys  2010; 77( 3): 655– 661. Google Scholar CrossRef Search ADS PubMed  20 Mulvenna P, Nankivell M, Barton R et al.   Dexamethasone and supportive care with or without whole brain radiotherapy in treating patients with non-small cell lung cancer with brain metastases unsuitable for resection or stereotactic radiotherapy (QUARTZ): results from a phase 3, non-inferiority, randomised trial. Lancet  2016; 388( 10055): 2004– 2014. Google Scholar CrossRef Search ADS PubMed  21 McTyre E, Scott J, Chinnaiyan P. Whole brain radiotherapy for brain metastasis. Surg Neurol Int  2013; 4( 5): S236– S244. Google Scholar CrossRef Search ADS PubMed  22 Lucas JTJr, Colmer HGIV, White L et al.   Competing risk analysis of neurologic versus nonneurologic death in patients undergoing radiosurgical salvage after whole-brain radiation therapy failure: who actually dies of their brain metastases? Int J Radiat Oncol Biol Phys  2015; 92( 5): 1008– 1015. Google Scholar CrossRef Search ADS PubMed  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 surgery. J Neurosurg  2013; 119( 5): 1139– 1144. Google Scholar CrossRef Search ADS PubMed  24 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. Google Scholar CrossRef Search ADS PubMed  25 Farris M, McTyre ER, Cramer CK et al.   Brain metastasis velocity: a novel prognostic metric predictive of overall survival and freedom from whole brain radiotherapy after distant brain failure following upfront radiosurgery alone. Int J Radiat Oncol Biol Phys  2017; 98( 1): 131– 141. Google Scholar CrossRef Search ADS PubMed  26 Gorovets D, Rava P, Ebner DK et al.   Predictors for long-term survival free from whole brain radiation therapy in patients treated with radiosurgery for limited brain metastases. Front Oncol  2015; 5: 110. Google Scholar CrossRef Search ADS PubMed  27 McTyre ER, Johnson AG, Ruiz J et al.   Predictors of neurologic and nonneurologic death in patients with brain metastasis initially treated with upfront stereotactic radiosurgery without whole-brain radiation therapy. Neuro Oncol  August 2017; 19( 4): 558– 566. Google Scholar PubMed  28 Press RH, Prabhu RS, Nickleach DC et al.   Novel risk stratification score for predicting early distant brain failure and salvage whole-brain radiotherapy after stereotactic radiosurgery for brain metastases. Cancer  2015; 121( 21): 3836– 3843. Google Scholar CrossRef Search ADS PubMed  29 Gorovets D, Ayala-Peacock D, Tybor DJ et al.   Multi-institutional nomogram predicting survival free from salvage whole brain radiation after radiosurgery in patients with brain metastases. Int J Radiat Oncol Biol Phys  2017; 97( 2): 246– 253. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For Permissions, please email: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Oncology Oxford University Press

Multi-institutional competing risks analysis of distant brain failure and salvage patterns after upfront radiosurgery without whole brain radiotherapy for brain metastasis

Loading next page...
 
/lp/ou_press/multi-institutional-competing-risks-analysis-of-distant-brain-failure-2n4gBU6Ykc
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For Permissions, please email: journals.permissions@oup.com.
ISSN
0923-7534
eISSN
1569-8041
D.O.I.
10.1093/annonc/mdx740
Publisher site
See Article on Publisher Site

Abstract

Abstract Background In this study, we use a competing risks analysis to assess factors predictive of early-salvage whole brain radiotherapy (WBRT) and early death after upfront stereotactic radiosurgery (SRS) alone for brain metastases in an attempt to identify populations that benefit less from upfront SRS. Patients and methods Patients from eight academic centers were treated with SRS for brain metastasis. Competing risks analysis was carried out for distant brain failure (DBF) versus death prior to DBF as well as for salvage SRS versus salvage WBRT versus death prior to salvage. Linear regression was used to determine predictors of the number of brain metastases at initial DBF (nDBF). Results A total of 2657 patients were treated with upfront SRS alone. Multivariate analysis (MVA) identified an increased hazard of DBF associated with increasing number of brain metastases (P < 0.001), lowest SRS dose received (P < 0.001), and melanoma histology (P < 0.001), while there was a decreased hazard of DBF associated with increasing age (P < 0.001), KPS < 70 (P < 0.001), and progressive systemic disease (P = 0.004). MVA for first salvage SRS versus WBRT versus death prior to salvage revealed an increased hazard of first salvage WBRT seen with increasing number of brain metastases (P < 0.001) and a decreased hazard with widespread systemic disease (P = 0.002) and increasing age (P < 0.001). Variables associated with nDBF included age (P = 0.02), systemic disease status (P = 0.03), melanoma histology (P = 0.05), and initial number of brain metastases (P < 0.001). Conclusions Patients with a higher initial number of brain metastases were more likely to experience DBF, have a higher nDBF, and receive early-salvage WBRT, while patients who were older, had lower KPS, or had more systemic disease were more likely to experience death prior to DBF or salvage WBRT. radiosurgery, brain metastasis, distant brain failure, competing risks, whole brain radiotherapy Key Message This study assessed the factors predictive of the competing risks of early-salvage whole brain radiotherapy (WBRT) versus early death for patients with brain metastasis treated upfront with radiosurgery. A higher number of initial brain metastases were predictive of early-salvage WBRT, whereas older age, lower performance status, and extracranial disease were predictive of death prior to salvage. Introduction The use of stereotactic radiosurgery (SRS) for the treatment of brain metastases has increased in recent years due both to the increasing availability of technology [1] as well as publication of randomized trials showing improvements in cognition over combined-modality therapy based on whole brain radiotherapy (WBRT) [2, 3]. Previously published randomized trials comparing SRS alone to SRS + WBRT were limited to patients having four or fewer brain metastases [2–5]; however, a recent multi-institutional prospective observational study from Japan has reported that patients with up to 10 brain metastases can be treated with SRS alone, and that these patients have survival similar to patients with 2–4 metastases [6]. As such, it has become less clear as to which patients should be treated with upfront SRS without WBRT. The cost of SRS is approximately three- to sixfold higher than WBRT in the United States [7, 8], with overall differences in cost exacerbated by the fact that SRS patients also require more salvage procedures than those receiving WBRT [4]. With ∼170 000 patients in the United States diagnosed with brain metastases each year [9], the appropriation of resources for brain metastasis management is a large-scale clinical dilemma. The use of SRS monotherapy appears to improve quality of life [10] and cognition [2, 3] over the use of WBRT, and although neurocognitive symptoms often begin to manifest within 3 months following WBRT [11], there is no apparent plateau in incidence or severity of symptoms [12]. Based upon cognitive outcomes, it would therefore seem reasonable to treat with SRS rather than WBRT when possible, however, patients who have poor life expectancy or rapid development of multiple new brain metastases may not derive significant benefit from SRS as they are unlikely to live long enough to experience WBRT neurocognitive sequelae. The goal of the present study is to assess, in patients treated upfront with SRS alone, the competing risks of distant brain failure (DBF) and the need for salvage therapy versus the risk of death prior to DBF or salvage therapy. A total of eight academic medical centers with high-volume SRS programs contributed data to the current analysis. Methods Data acquisition This study was approved by the Institutional Review Board at all eight participating institutions. Data were reviewed and collected at all institutions for patients treated between 1 January 2000 and 1 January 2014. Patients were included in the study if they underwent SRS following initial diagnosis with brain metastasis. Electronic medical records were reviewed, and putative predictors of DBF and salvage therapy were recorded for each patient [13]. Extent of extracranial disease was defined based upon prior publications [14]. SRS doses varied by institution, but were generally determined based upon the guidelines published by Shaw et al. for single fraction radiosurgical treatment of brain metastases [15]. As SRS dose is inversely proportional to tumor volume, the lowest SRS dose prescribed to any lesion was recorded for each patient to serve as a surrogate for tumor volume. The majority of patients underwent follow-up with magnetic resonance imaging (MRI) 1–3 months after their SRS procedures, with subsequent MRIs every 3 months thereafter. DBF events were defined as the development of one or more new brain metastases outside of the previous SRS target volume(s). DBF events and the number of brain metastases at time of DBF were determined based upon MRI results. Statistical analysis Median follow-up and time-to-event outcomes were defined beginning at the time of SRS and extending to the time of most recent follow-up or to the event of interest. Time-to-event outcomes were summarized using the Kaplan–Meier method. Cumulative incidences were estimated for DBF, death prior to DBF/salvage, first salvage SRS, and first salvage WBRT [16]. Competing risks models were developed to determine the single variable subdistribution hazard ratios (HR) associated with each predictor for each of these events. Statistically significant (P < 0.05) variables identified on univariate analysis were included in forward stepwise regressions to identify the multivariable models that minimized the Akaike information criterion (AIC) [17]. These results were used to guide purposeful development of multivariate competing risks models for DBF versus death prior to DBF as well as first salvage with SRS versus first salvage with WBRT versus death prior to salvage. Linear regression was carried out for predictor variables of interest for the outcome of number of brain metastases at time of DBF (nDBF). The linear regression analysis was limited to patients who experienced DBF and who had a quantifiable number of metastases at time of failure. Stepwise regression was used to identify the nDBF multiple regression model with the lowest AIC [17]. Statistical analysis was carried out using R version 3.2.1 software (R Foundation for Statistical Computing, Vienna, Austria). Results Clinical outcomes Of 2657 patients treated with upfront SRS alone, 2208 (83.1%) patients were deceased at the time of analysis, with DBF occurring in 1244 (46.8%) of all patients. Patient characteristics can be found in Table 1. Median overall survival (OS) was 9.8 months (95% CI: 9.3–10.4 months). After initial SRS, a total of 619 (23.3%) patients received salvage SRS and 447 (16.8%) of patients received salvage WBRT after initial SRS. Of the 447 patients who received salvage WBRT, 336 (75.2%) received salvage WBRT as their first salvage treatment. Table 1. Patient characteristics   n/median (%/IQR)  Total patients  2657  Age, year  61 (53, 70)  Gender     Female  1331 (50.1)   Male  1326 (49.9)  Ethnicity     Asian  65 (2.5)   African-American  181 (7.0)   Latino  46 (1.8)   White  2232 (86.0)   Other  72 (2.7)  Primary malignancy     Breast  373 (14.0)    Her2+  159 (43.6)    Her2−  159 (43.6)    Her2 unknown  55 (14.7)   Lung  1279 (48.1)    Adeno  649 (50.7)    SqCC  111 (8.7)    NSCLC NOS  392 (30.6)    Other  127 (9.9)   RCC  272 (10.2)   Melanoma  479 (18.0)   GI  121 (5.2)   GU  9 (0.4)   Gyn  41 (1.8)   H/N  19 (0.8)   Sarcoma  14 (0.6)   Other  50 (1.9)  Systemic disease burden     None  376 (14.2)   Oligometastatic (≤5 extracranial metastases)  995 (37.4)   Widespread (>5 extracranial metastases)  1199 (45.1)   Unknown  87 (3.3)  Systemic disease status     Stable  1224 (46.1)   Progressive  1254 (47.2)   Unknown  179 (6.7)  KPS     <70  199 (7.9)   ≥70  2326 (92.1)  RPA     1  391 (14.7)   2  1935 (72.9)   3  199 (7.5)  No. of initial brain metastases     1  1295 (48.7)   2  584 (22.0)   3  350 (13.2)   4  176 (6.6)   5+  252 (9.5)  Lowest SRS margin dose  18.5 (17.0, 21.0)    n/median (%/IQR)  Total patients  2657  Age, year  61 (53, 70)  Gender     Female  1331 (50.1)   Male  1326 (49.9)  Ethnicity     Asian  65 (2.5)   African-American  181 (7.0)   Latino  46 (1.8)   White  2232 (86.0)   Other  72 (2.7)  Primary malignancy     Breast  373 (14.0)    Her2+  159 (43.6)    Her2−  159 (43.6)    Her2 unknown  55 (14.7)   Lung  1279 (48.1)    Adeno  649 (50.7)    SqCC  111 (8.7)    NSCLC NOS  392 (30.6)    Other  127 (9.9)   RCC  272 (10.2)   Melanoma  479 (18.0)   GI  121 (5.2)   GU  9 (0.4)   Gyn  41 (1.8)   H/N  19 (0.8)   Sarcoma  14 (0.6)   Other  50 (1.9)  Systemic disease burden     None  376 (14.2)   Oligometastatic (≤5 extracranial metastases)  995 (37.4)   Widespread (>5 extracranial metastases)  1199 (45.1)   Unknown  87 (3.3)  Systemic disease status     Stable  1224 (46.1)   Progressive  1254 (47.2)   Unknown  179 (6.7)  KPS     <70  199 (7.9)   ≥70  2326 (92.1)  RPA     1  391 (14.7)   2  1935 (72.9)   3  199 (7.5)  No. of initial brain metastases     1  1295 (48.7)   2  584 (22.0)   3  350 (13.2)   4  176 (6.6)   5+  252 (9.5)  Lowest SRS margin dose  18.5 (17.0, 21.0)  Adeno, adenocarcinoma; GI, gastrointestinal; GU, genitourinary; Gyn, gynecologic; H/N, head and neck; IQR, interquartile range; KPS, Karnofsky performance status; NSCLC, non-small-cell lung cancer; RPA, recursive partitioning analysis; SqCC, squamous cell carcinoma; SRS, stereotactic radiosurgery. DBF versus death prior to DBF The estimated cumulative incidences of DBF at 6, 12, and 24 months were 27.7%, 38.5%, and 46.0%, respectively (Figure 1). Predictors of DBF identified on univariate analysis included age (P < 0.001), KPS (P < 0.001), systemic disease status (P < 0.001), number of brain metastases (P < 0.001), SRS dose (P < 0.001), and melanoma histology (P < 0.001). Cumulative incidences of DBF as stratified by variables included in the final multivariate analysis (MVA) model can be found in Figure 2. Cumulative incidences of DBF at 1 and 2 years as stratified by all clinical variables of interest can be found in supplementary Table S1, available at Annals of Oncology online. MVA identified an increased hazard of DBF associated with increasing number of brain metastases (HR = 1.09, P < 0.001), lowest SRS dose (HR = 1.04, P <0.001), and melanoma histology (HR = 1.37, P < 0.001), while there was a decreased hazard of DBF associated with increasing age (HR = 0.99, P < 0.001), KPS < 70 (HR = 0.59, P < 0.001), and progressive systemic disease (HR = 0.84, P = 0.004) (Table 2). Table 2. Multivariate proportional subdistribution hazard models for DBF, first salvage WBRT, and first salvage SRS   HR  95% CI  P-value  Distant brain failure (DBF)         Age (continuous)  0.99  0.98–0.99  <0.001   KPS <70  0.59  0.44–0.79  <0.001   Melanoma histology  1.37  1.17–1.60  <0.001   Progressive systemic disease  0.84  0.74–0.94  0.004   No. of brain metastases (continuous)  1.09  1.05–1.13  <0.001   Lowest SRS dose (continuous)  1.04  1.02–1.06  <0.001  First salvage WBRT   Age (continuous)  0.98  0.97–0.99  <0.001   Widespread extracranial disease  0.67  0.52–0.87  0.002   Progressive systemic disease  0.65  0.50–0.83  <0.001   No. of brain metastases (continuous)  1.09  1.05–1.13  <0.001  First salvage SRS   KPS <70  0.30  0.17–0.53  <0.001   Progressive systemic disease  0.69  0.58–0.83  <0.001   Lowest SRS dose (continuous)  1.12  1.08–1.15  <0.001    HR  95% CI  P-value  Distant brain failure (DBF)         Age (continuous)  0.99  0.98–0.99  <0.001   KPS <70  0.59  0.44–0.79  <0.001   Melanoma histology  1.37  1.17–1.60  <0.001   Progressive systemic disease  0.84  0.74–0.94  0.004   No. of brain metastases (continuous)  1.09  1.05–1.13  <0.001   Lowest SRS dose (continuous)  1.04  1.02–1.06  <0.001  First salvage WBRT   Age (continuous)  0.98  0.97–0.99  <0.001   Widespread extracranial disease  0.67  0.52–0.87  0.002   Progressive systemic disease  0.65  0.50–0.83  <0.001   No. of brain metastases (continuous)  1.09  1.05–1.13  <0.001  First salvage SRS   KPS <70  0.30  0.17–0.53  <0.001   Progressive systemic disease  0.69  0.58–0.83  <0.001   Lowest SRS dose (continuous)  1.12  1.08–1.15  <0.001  CI, confidence interval; HR, hazard ratio; KPS, Karnofsky performance status; SRS, stereotactic radiosurgery; WBRT, whole brain radiotherapy. Figure 1. View largeDownload slide Cumulative incidence of DBF versus death prior to DBF (A) and first salvage versus death prior to salvage (B). DBF, distant brain failure; SRS, stereotactic radiosurgery; WBRT, whole brain radiotherapy. Figure 1. View largeDownload slide Cumulative incidence of DBF versus death prior to DBF (A) and first salvage versus death prior to salvage (B). DBF, distant brain failure; SRS, stereotactic radiosurgery; WBRT, whole brain radiotherapy. Figure 2. View largeDownload slide Cumulative incidences of DBF as stratified by histology (A), number of initial brain metastases (B), age (C), KPS (D), lowest initial SRS dose (E), and systemic disease status (F). DBF, distant brain failure; KPS, Karnofsky performance status; met, brain metastasis; RCC, renal cell carcinoma; SRS, stereotactic radiosurgery. Figure 2. View largeDownload slide Cumulative incidences of DBF as stratified by histology (A), number of initial brain metastases (B), age (C), KPS (D), lowest initial SRS dose (E), and systemic disease status (F). DBF, distant brain failure; KPS, Karnofsky performance status; met, brain metastasis; RCC, renal cell carcinoma; SRS, stereotactic radiosurgery. First salvage: SRS versus WBRT versus death prior to salvage The estimated cumulative incidences of first salvage with SRS at 6, 12, and 24 months were 10.1%, 15.4%, and 19.4%, respectively (Figure 1). The estimated cumulative incidences of first salvage with WBRT at 6, 12, and 24 months were 8.5%, 11.1%, and 12.9%, respectively. MVA for first salvage WBRT revealed an increased hazard with increasing number of brain metastases (HR = 1.09, P < 0.001) and a decreased hazard with widespread systemic disease (HR = 0.67, P = 0.002) and increasing age (HR = 0.98, P < 0.001). MVA for first salvage SRS revealed an increased hazard with increasing lowest SRS dose (HR = 1.12, P <0.001), and a decreased hazard with KPS < 70 (HR = 0.30, P <0.001). Progressive systemic disease was associated with an increased hazard of death prior to salvage with either SRS or WBRT (HR = 1.69, P < 0.001), and a decreased hazard of both first salvage WBRT (HR = 0.65, P <0.001) and first salvage SRS (HR = 0.69, P < 0.001) on respective MVAs. MVA results for first salvage WBRT and first salvage SRS can be found in Table 2. Cumulative incidences of first salvage WBRT as stratified by variables included in the final MVA model can be found in supplementary Figure S1, available at Annals of Oncology online. Cumulative incidences of first salvage WBRT at 1 and 2 years as stratified by all clinical variables of interest can be found in supplementary Table S1, available at Annals of Oncology online. Number of brain metastases at time of DBF Of the 1244 patients with DBF, there were 924 patients (74.3%) who had a quantifiable number of brain metastases at time of initial DBF (nDBF). On multiple regression, increasing age was associated with lower nDBF (P = 0.02), while systemic disease status (P = 0.03), melanoma histology (P = 0.05), and higher number of initial brain metastases (P < 0.001) were associated with higher nDBF, as shown in supplementary Table S2, available at Annals of Oncology online. When patients were analyzed by histologic subset, initial number of brain metastases was predictive of nDBF for lung (P = 0.001), RCC (P = 0.006), and melanoma (P < 0.001), but was not predictive of nDBF for patients with breast cancer (P = 0.29) (supplementary Figure S2, available at Annals of Oncology online). Discussion Identification of populations that do and do not benefit from SRS will be critical for the proper appropriation of a costly resource. Treating patients with a severely limited life expectancy is unlikely to be a cost-effective use of SRS given the high cost relative to a short-term benefit. As the cognitive toxicity of WBRT is permanent, those with a longer life expectancy may experience the greatest benefit from postponement or complete avoidance of WBRT [12]. The median life expectancy of patients with brain metastases in the modern era is ∼8–10 months, though the range is wide [4–6]. The tail of this range will likely broaden further with the increasing use and availability of novel systemic treatment options such as checkpoint inhibitors that seem to be significantly prolonging survival for some patients, as notably shown recently with advanced melanoma [18]. As such, prognostic indices that help to identify populations with particularly poor prognoses will be helpful for clinical decision-making [19]. In the current analysis, poor performance status, progressive systemic disease and larger brain metastases (lower SRS dose) predicted for early death after SRS alone. Those with a low burden of brain disease and short life expectancy may be best treated with supportive care or observed until symptomatic [20]. Those with a high total volume of metastatic brain disease have a poorer prognosis and greater likelihood of dying of brain metastases, and may warrant consideration of upfront WBRT [21], as large brain metastases treated with SRS alone have previously been associated with a higher likelihood of neurologic death [22]. These patients may be best treated with fractionated SRS or a multi-modality approach such as WBRT followed by SRS boost, SRS followed by surgery, or surgery followed by radiotherapy. The threshold number of brain metastases that should be treated with SRS monotherapy remains controversial. Published randomized trials have used a limit of 3–4 brain metastases [2–4], though a prospective study from Japan suggests that up to 10 can be treated with equivalent survival [6]. Traditionally, such a limit was posed because treating larger numbers of metastases would overwhelm the capacity of many linear accelerator-based radiosurgery platforms. With recent improvements, these technical limitations are no longer as relevant [1]. Multiple series have suggested that the total volume of brain metastases is likely more important than the number of metastases at time of SRS with regards to OS [6, 23]. However, other series have implied that a greater number of metastases predicts for earlier DBF [13, 24], as seen in the present series. While the optimal upfront modality for brain metastasis management has been heavily investigated, the criteria for optimal use of each salvage modality remain undetermined. As previously mentioned, the number of new metastases at time of DBF is a dominant determinant of whether WBRT or SRS will be used in the salvage setting [25, 26]. Patients who receive multiple salvage SRS procedures for a limited number of brain metastases may represent a population who have survived the competing risks of early death and early WBRT. Conversely, those who experience an acceleration of intracranial progression likely require WBRT at the point of acceleration. The recently introduced concept of brain metastasis velocity is one in which the re-seeding rate of the brain is used to triage patients to salvage modality [25]. While this concept remains to be validated, it has potential in aiding practice decisions and is supported by the finding in the current study that patients with greater number of metastases at time of diagnosis also have a greater number at time of treatment failure. Brain metastasis patients have a large competing risk of dying of their extracranial disease, with only 20% of patients actually dying of brain metastases [5]. The results of the current study support this notion, as patients with widespread extracranial disease were less likely to receive salvage WBRT. Previous reports have suggested, though, that those with early CNS failure are at higher risk of dying of neurologic death [27]. Patients who require early WBRT often do so because of rapid multifocal DBF, leptomeningeal seeding or large volume local failure. Should these events occur within several weeks after brain metastasis diagnosis, patients may be less likely to benefit from SRS because of the lack of meaningful postponement of the toxicities of WBRT. There have been recent attempts to predict patients who will have rapid DBF and require early WBRT [13, 28, 29]. A limitation of these prior models is that they do not predict the actual number of lesions at time of DBF. There are several limitations to the current series. While the number of patients was large, the dataset was derived from several high volume academic centers in the United States and Canada. Given the retrospective nature of the dataset and the patient selection bias of North American academic centers, the results will need to be validated prospectively. Although the variables that were systematically collected by every institution were fairly comprehensive, systemic treatment details were not routinely collected by all institutions, and as such were not able to be reported in this study. Variation in the salvage philosophies at each of the centers may have influenced the decision to treat patients with salvage WBRT, as there were no standard guidelines across institutions regarding the use of SRS versus WBRT for salvage, with most institutions generally favoring salvage with SRS rather than WBRT when feasible. There were also no universal guidelines or protocol for MRI acquisition across institutions, in either the initial pre-treatment or follow-up setting. One implication of this is that similarly sized, small brain metastases may have potentially been identified and treated at different times across institutions, based upon differences in institutional MRI capabilities and frequency of follow-up imaging. Funding None declared. Disclosure MA: consulting and grant from Elekta; grant support from Boehringer Ingelheim, Bristol-Myers Squibb, Novartis, Spectrum Pharmaceuticals, Tracon Pharmaceuticals, Novocure; Consultant for Merck, Genentech/Roche, Incyte, Caris Lifesciences, Monteris Medical, MRI interventions, Inc. SC: honorarium from Varian Medical Systems, Abbvie, and Zeiss. All remaining authors have declared no conflicts of interest. References 1 Thomas EM, Popple RA, Wu X et al.   Comparison of plan quality and delivery time between volumetric arc therapy (RapidArc) and Gamma Knife radiosurgery for multiple cranial metastases. Neurosurgery  2014; 75( 4): 409– 417; discussion 417–418. Google Scholar CrossRef Search ADS PubMed  2 Chang EL, Wefel JS, Hess KR et al.   Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain irradiation: a randomised controlled trial. Lancet Oncol  2009; 10( 11): 1037– 1044. Google Scholar CrossRef Search ADS PubMed  3 Brown PD, Jaeckle K, Ballman KV et al.   Effect of radiosurgery alone vs radiosurgery with whole brain radiation therapy on cognitive function in patients with 1 to 3 brain metastases: a randomized clinical trial. JAMA  2016; 316( 4): 401– 409. Google Scholar CrossRef Search ADS PubMed  4 Aoyama H, Shirato H, Tago M et al.   Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA  2006; 295( 21): 2483– 2491. Google Scholar CrossRef Search ADS PubMed  5 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. JCO  2011; 29( 2): 134– 141. Google Scholar CrossRef Search ADS   6 Yamamoto M, Serizawa T, Shuto T et al.   Stereotactic radiosurgery for patients with multiple brain metastases (JLGK0901): a multi-institutional prospective observational study. Lancet Oncol  2014; 15( 4): 387– 395. Google Scholar CrossRef Search ADS PubMed  7 Lester SC, Taksler GB, Kuremsky JG et al.   Clinical and economic outcomes of patients with brain metastases based on symptoms: an argument for routine brain screening of those treated with upfront radiosurgery. Cancer  2014; 120( 3): 433– 441. Google Scholar CrossRef Search ADS PubMed  8 Brown PD, Kee AY, Eshleman JS, Fiveash JB, Adjuvant whole brain radiotherapy: strong emotions decide but rationale studies are needed: in regard to Brown et al. (Int J Radiat Oncol Biol Phys 2008; 70: 1305–1309). In reply to Drs. Larson and Sahgal. Int J Radiat Oncol Biol Phys  2009; 75( 1): 316– 317. Google Scholar CrossRef Search ADS PubMed  9 Ellis TL, Neal MT, Chan MD. The role of surgery, radiosurgery and whole brain radiation therapy in the management of patients with metastatic brain tumors. Int J Surg Oncol  2012; 2012: 952345. Google Scholar PubMed  10 Soffietti R, Kocher M, Abacioglu UM et al.   A European Organisation for research and treatment of cancer phase III trial of adjuvant whole-brain radiotherapy versus observation in patients with one to three brain metastases from solid tumors after surgical resection or radiosurgery: quality-of-life results. J Clin Oncol  2013; 31( 1): 65– 72. Google Scholar CrossRef Search ADS PubMed  11 Brown PD, Pugh S, Laack NN et al.   Memantine for the prevention of cognitive dysfunction in patients receiving whole-brain radiotherapy: a randomized, double-blind, placebo-controlled trial. Neuro Oncol  2013; 15( 10): 1429– 1437. Google Scholar CrossRef Search ADS PubMed  12 Greene-Schloesser D, Robbins ME, Peiffer AM et al.   Radiation-induced brain injury: a review. Front Oncol  2012; 2: 73. Google Scholar CrossRef Search ADS PubMed  13 Ayala-Peacock DN, Peiffer AM, Lucas JT et al.   A nomogram for predicting distant brain failure in patients treated with gamma knife stereotactic radiosurgery without whole brain radiotherapy. Neuro Oncol  2014; 16( 9): 1283– 1288. Google Scholar CrossRef Search ADS PubMed  14 Harris S, Chan MD, Lovato JF et al.   Gamma knife stereotactic radiosurgery as salvage therapy after failure of whole-brain radiotherapy in patients with small-cell lung cancer. Int J Radiat Oncol Biol Phys  2012; 83( 1): e53– e59. Google Scholar CrossRef Search ADS PubMed  15 Shaw E, Scott C, Souhami L et al.   Single dose radiosurgical treatment of recurrent previously irradiated primary brain tumors and brain metastases: final report of RTOG protocol 90-05. Int J Radiat Oncol Biol Phys  2000; 47( 2): 291– 298. Google Scholar CrossRef Search ADS PubMed  16 Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc  1999; 94( 446): 496– 509. Google Scholar CrossRef Search ADS   17 Akaike H. A new look at the statistical model identification. IEEE Trans Automat Contr  1974; 19( 6): 716– 723. Google Scholar CrossRef Search ADS   18 Wolchok JD, Chiarion-Sileni V, Gonzalez R et al.   Overall survival with combined nivolumab and ipilimumab in advanced melanoma. N Engl J Med  2017; 377( 14): 1345– 1356. Google Scholar CrossRef Search ADS PubMed  19 Sperduto PW, Chao ST, Sneed PK et al.   Diagnosis-specific prognostic factors, indexes, and treatment outcomes for patients with newly diagnosed brain metastases: a multi-institutional analysis of 4, 259 patients. Int J Radiat Oncol Biol Phys  2010; 77( 3): 655– 661. Google Scholar CrossRef Search ADS PubMed  20 Mulvenna P, Nankivell M, Barton R et al.   Dexamethasone and supportive care with or without whole brain radiotherapy in treating patients with non-small cell lung cancer with brain metastases unsuitable for resection or stereotactic radiotherapy (QUARTZ): results from a phase 3, non-inferiority, randomised trial. Lancet  2016; 388( 10055): 2004– 2014. Google Scholar CrossRef Search ADS PubMed  21 McTyre E, Scott J, Chinnaiyan P. Whole brain radiotherapy for brain metastasis. Surg Neurol Int  2013; 4( 5): S236– S244. Google Scholar CrossRef Search ADS PubMed  22 Lucas JTJr, Colmer HGIV, White L et al.   Competing risk analysis of neurologic versus nonneurologic death in patients undergoing radiosurgical salvage after whole-brain radiation therapy failure: who actually dies of their brain metastases? Int J Radiat Oncol Biol Phys  2015; 92( 5): 1008– 1015. Google Scholar CrossRef Search ADS PubMed  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 surgery. J Neurosurg  2013; 119( 5): 1139– 1144. Google Scholar CrossRef Search ADS PubMed  24 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. Google Scholar CrossRef Search ADS PubMed  25 Farris M, McTyre ER, Cramer CK et al.   Brain metastasis velocity: a novel prognostic metric predictive of overall survival and freedom from whole brain radiotherapy after distant brain failure following upfront radiosurgery alone. Int J Radiat Oncol Biol Phys  2017; 98( 1): 131– 141. Google Scholar CrossRef Search ADS PubMed  26 Gorovets D, Rava P, Ebner DK et al.   Predictors for long-term survival free from whole brain radiation therapy in patients treated with radiosurgery for limited brain metastases. Front Oncol  2015; 5: 110. Google Scholar CrossRef Search ADS PubMed  27 McTyre ER, Johnson AG, Ruiz J et al.   Predictors of neurologic and nonneurologic death in patients with brain metastasis initially treated with upfront stereotactic radiosurgery without whole-brain radiation therapy. Neuro Oncol  August 2017; 19( 4): 558– 566. Google Scholar PubMed  28 Press RH, Prabhu RS, Nickleach DC et al.   Novel risk stratification score for predicting early distant brain failure and salvage whole-brain radiotherapy after stereotactic radiosurgery for brain metastases. Cancer  2015; 121( 21): 3836– 3843. Google Scholar CrossRef Search ADS PubMed  29 Gorovets D, Ayala-Peacock D, Tybor DJ et al.   Multi-institutional nomogram predicting survival free from salvage whole brain radiation after radiosurgery in patients with brain metastases. Int J Radiat Oncol Biol Phys  2017; 97( 2): 246– 253. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Journal

Annals of OncologyOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial