[Skip to Content]
[Skip to Content Landing]
Figure 1.
CONSORT Diagram of the 88 Patients With Non–Small-Cell Lung Cancer (NSCLC)
CONSORT Diagram of the 88 Patients With Non–Small-Cell Lung Cancer (NSCLC)

BMs indicates brain metastases; ITT, intention-to-treat; SRS, stereotactic radiosurgery; and WBRT, whole-brain radiotherapy.

aPatients were stratified by number of BMs 1 vs 2-4); primary tumor (lung vs others); systemic disease (controlled vs uncontrolled).

Figure 2.
Overall Survival of Patients With Non–Small-Cell Lung Cancer, With DS-GPA Scores of 2.5 to 4.0 and 0.5 to 2.0
Overall Survival of Patients With Non–Small-Cell Lung Cancer, With DS-GPA Scores of 2.5 to 4.0 and 0.5 to 2.0

DS-GPA indicates diagnosis-specific Graded Prognostic Assessment; SRS, stereotactic radiosurgery; and WBRT, whole-brain radiotherapy.

Figure 3.
Brain Tumor Recurrence-Free Rate of Patients With Non–Small-Cell Lung Cancer, With DS-GPA Scores of 2.5 to 4.0 and 0.5 to 2.0
Brain Tumor Recurrence-Free Rate of Patients With Non–Small-Cell Lung Cancer, With DS-GPA Scores of 2.5 to 4.0 and 0.5 to 2.0

DS-GPA indicates diagnosis-specific Graded Prognostic Assessment; SRS, stereotactic radiosurgery; and WBRT, whole-brain radiotherapy.

Table 1.  
Distribution of Patients in Each Treatment Group According to the DS-GPA and Associated Factors
Distribution of Patients in Each Treatment Group According to the DS-GPA and Associated Factors
Table 2.  
Outcomes in Each Treatment Group According to the DS-GPA Score
Outcomes in Each Treatment Group According to the DS-GPA Score
1.
Lin  NU, Lee  EQ, Aoyama  H,  et al; Response Assessment in Neuro-Oncology (RANO) group.  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.PubMedGoogle ScholarCrossref
2.
Sperduto  PW, Berkey  B, Gaspar  LE, Mehta  M, Curran  W.  A new prognostic index and comparison to three other indices for patients with brain metastases: an analysis of 1,960 patients in the RTOG database.  Int J Radiat Oncol Biol Phys. 2008;70(2):510-514.PubMedGoogle ScholarCrossref
3.
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.PubMedGoogle ScholarCrossref
4.
Gaspar  L, Scott  C, Rotman  M,  et al.  Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials.  Int J Radiat Oncol Biol Phys. 1997;37(4):745-751.PubMedGoogle ScholarCrossref
5.
Sperduto  PW, Kased  N, Roberge  D,  et al.  Summary report on the graded prognostic assessment: an accurate and facile diagnosis-specific tool to estimate survival for patients with brain metastases.  J Clin Oncol. 2012;30(4):419-425.PubMedGoogle ScholarCrossref
6.
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.PubMedGoogle ScholarCrossref
7.
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.PubMedGoogle ScholarCrossref
8.
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.PubMedGoogle ScholarCrossref
9.
Piaggio  G, Elbourne  DR, Pocock  SJ, Evans  SJ, Altman  DG; CONSORT Group.  Reporting of noninferiority and equivalence randomized trials: extension of the CONSORT 2010 statement.  JAMA. 2012;308(24):2594-2604.PubMedGoogle ScholarCrossref
10.
Aupérin  A, Arriagada  R, Pignon  JP,  et al; Prophylactic Cranial Irradiation Overview Collaborative Group.  Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission.  N Engl J Med. 1999;341(7):476-484.PubMedGoogle ScholarCrossref
11.
Andrews  DW, Scott  CB, Sperduto  PW,  et al.  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 randomised trial.  Lancet. 2004;363(9422):1665-1672.PubMedGoogle ScholarCrossref
12.
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.PubMedGoogle ScholarCrossref
13.
Gore  EM, Bae  K, Wong  SJ,  et al.  Phase III comparison of prophylactic cranial irradiation versus observation in patients with locally advanced non-small-cell lung cancer: primary analysis of radiation therapy oncology group study RTOG 0214.  J Clin Oncol. 2011;29(3):272-278.PubMedGoogle ScholarCrossref
14.
Pirzkall  A, Debus  J, Lohr  F,  et al.  Radiosurgery alone or in combination with whole-brain radiotherapy for brain metastases.  J Clin Oncol. 1998;16(11):3563-3569.PubMedGoogle Scholar
15.
Sperduto  PW, Shanley  R, Luo  X,  et al.  Secondary analysis of RTOG 9508, a phase 3 randomized trial of whole-brain radiation therapy versus WBRT plus stereotactic radiosurgery in patients with 1-3 brain metastases; poststratified by the graded prognostic assessment (GPA).  Int J Radiat Oncol Biol Phys. 2014;90(3):526-531.PubMedGoogle ScholarCrossref
16.
Soon  YY, Tham  IW, Lim  KH, Koh  WY, Lu  JJ.  Surgery or radiosurgery plus whole brain radiotherapy versus surgery or radiosurgery alone for brain metastases.  Cochrane Database Syst Rev. 2014;3:CD009454.PubMedGoogle Scholar
17.
Onodera  S, Aoyama  H, Tha  KK,  et al.  The value of 4-month neurocognitive function as an endpoint in brain metastases trials.  J Neurooncol. 2014;120(2):311-319.PubMedGoogle ScholarCrossref
18.
McDuff  SG, Taich  ZJ, Lawson  JD,  et al.  Neurocognitive assessment following whole brain radiation therapy and radiosurgery for patients with cerebral metastases.  J Neurol Neurosurg Psychiatry. 2013;84(12):1384-1391.PubMedGoogle ScholarCrossref
19.
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.PubMedGoogle ScholarCrossref
20.
Aoyama  H, Tago  M, Kato  N,  et al.  Neurocognitive function of patients with brain metastasis who received either whole brain radiotherapy plus stereotactic radiosurgery or radiosurgery alone.  Int J Radiat Oncol Biol Phys. 2007;68(5):1388-1395.PubMedGoogle ScholarCrossref
21.
Shaw  EG, Rosdhal  R, D’Agostino  RB  Jr,  et al.  Phase II study of donepezil in irradiated brain tumor patients: effect on cognitive function, mood, and quality of life.  J Clin Oncol. 2006;24(9):1415-1420.PubMedGoogle ScholarCrossref
22.
Brown  PD, Pugh  S, Laack  NN,  et al; Radiation Therapy Oncology Group (RTOG).  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.PubMedGoogle ScholarCrossref
23.
Gondi  V, Pugh  SL, Tome  WA,  et al.  Preservation of memory with conformal avoidance of the hippocampal neural stem-cell compartment during whole-brain radiotherapy for brain metastases (RTOG 0933): a phase II multi-institutional trial.  J Clin Oncol. 2014;32(34):3810-3816.PubMedGoogle ScholarCrossref
24.
Grosshans  DR, Meyers  CA, Allen  PK, Davenport  SD, Komaki  R.  Neurocognitive function in patients with small cell lung cancer : effect of prophylactic cranial irradiation.  Cancer. 2008;112(3):589-595.PubMedGoogle ScholarCrossref
25.
Maemondo  M, Inoue  A, Kobayashi  K,  et al; North-East Japan Study Group.  Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR.  N Engl J Med. 2010;362(25):2380-2388.PubMedGoogle ScholarCrossref
26.
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.PubMedGoogle ScholarCrossref
27.
Gerber  NK, Yamada  Y, Rimner  A,  et al.  Erlotinib versus radiation therapy for brain metastases in patients with EGFR-mutant lung adenocarcinoma.  Int J Radiat Oncol Biol Phys. 2014;89(2):322-329.PubMedGoogle ScholarCrossref
28.
Owonikoko  TK, Arbiser  J, Zelnak  A,  et al.  Current approaches to the treatment of metastatic brain tumours.  Nat Rev Clin Oncol. 2014;11(4):203-222.PubMedGoogle ScholarCrossref
Original Investigation
July 2015

Stereotactic Radiosurgery With or Without Whole-Brain Radiotherapy for Brain Metastases: Secondary Analysis of the JROSG 99-1 Randomized Clinical Trial

Hidefumi Aoyama, MD, PhD1; Masao Tago, MD, PhD2; Hiroki Shirato, MD, PhD3; et al for the Japanese Radiation Oncology Study Group 99-1 (JROSG 99-1) Investigators
Author Affiliations
  • 1Department of Radiology, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan
  • 2Department of Radiology, Teikyo University Mizonokuchi Hospital, Kawasaki, Japan
  • 3Department of Radiation Medicine, Hokkaido University Graduate School of Medicine, Sapporo, Japan
JAMA Oncol. 2015;1(4):457-464. doi:10.1001/jamaoncol.2015.1145
Abstract

Importance  It remains uncertain whether treatment with stereotactic radiosurgery (SRS) alone can be safely applied to all patient populations with 1 to 4 brain metastases (BMs) exhibiting heterogeneous prognoses.

Objective  To investigate the feasibility of SRS alone for patients with different prognoses determined by the diagnosis-specific Graded Prognostic Assessment (DS-GPA).

Design, Setting, and Participants  A secondary analysis (performed in September 2014) of the Japanese Radiation Oncology Study Group (JROSG) 99-1, a phase 3 randomized trial, comparing SRS alone and whole-brain radiotherapy (WBRT) + SRS conducted in 1999 to 2003. Among a total of 132 patients, 88 with non–small-cell lung cancer (NSCLC) and 1 to 4 BMs were included and poststratified by DS-GPA scores to avoid potential bias from BMs from different primary cancer types. The median follow-up time was 8.05 months.

Interventions  The WBRT schedule was 30 Gy in 10 fractions over 2 to 2.5 weeks. The mean SRS dose was 21.9 Gy in SRS alone and 16.6 Gy in WBRT + SRS.

Main Outcomes and Measures  The primary end point was overall survival (OS), and the secondary end points included brain tumor recurrence (BTR), salvage treatment, and radiation toxic effects.

Results  Forty-seven patients had a favorable prognosis, with DS-GPA scores of 2.5 to 4.0 (26 SRS-alone and 21 WBRT + SRS [DS-GPA 2.5-4.0 group]), and 41 had an unfavorable prognosis, with DS-GPA scores of 0.5 to 2.0 (19 SRS-alone and 22 WBRT + SRS [DS-GPA 0.5-2.0 group]). Significantly better OS was observed in the DS-GPA 2.5-4.0 group in WBRT + SRS vs the SRS alone, with a median survival time of 16.7 (95% CI, 7.5-72.9) months vs 10.6 (95% CI, 7.7-15.5) months (P = .04) (hazard ratio [HR], 1.92; 95% CI, 1.01-3.78). However, no such difference was observed in the DS-GPA 0.5-2.0 group (HR, 1.05; 95% CI, 0.55-1.99) (P = .86). This benefit could be explained by the differing BTR rates, in that the prevention against BTR by WBRT had a more significant impact in the DS-GPA 2.5-4.0 group (HR, 8.31; 95% CI, 3.05-29.13) (P < .001) vs the DS-GPA 0.5-2.0 group (HR, 3.57; 95% CI, 1.02-16.49) (P = .04).

Conclusions and Relevance  Despite the current trend of using SRS alone, the important role of WBRT for patients with BMs from NSCLC with a favorable prognosis should be considered. Our findings should be validated through appropriately designed prospective studies.

Trial Registration  umin.ac.jp/ctr Identifier: C000000412

Introduction

Brain metastases (BMs) are a serious and increasingly common complication in patients with solid cancers. Lung cancer represents the most common primary tumor linked to BMs, accounting for 40% to 50% of BM cases, followed by breast cancer, which accounts for 10% to 20%.1 Historically, the prognosis of patients with BMs has been considered uniformly poor, with a median survival in the 2- to 4-month range. However, it has become evident that not all patients with BMs have the same poor prognosis, and the use of an identical management strategy for all patients is no longer appropriate.2

In the initial 2006 report of the Japanese Radiation Oncology Study Group (JROSG) 99-1 phase 3 randomized clinical trial comparing up-front whole-brain radiation therapy (WBRT) combined with stereotactic radiosurgery (SRS) (WBRT + SRS arm) and SRS without up-front WBRT (SRS-alone arm) for patients with 1 to 4 BMs, we reported that adding WBRT to SRS significantly reduced brain tumor recurrence (BTR) at both initial and distant sites in the brain.3 The impact on overall survival (OS) was not significant, but the trial was prematurely closed before reaching the accrual goal.3 When we designed that trial in the late 1990s, the Radiation Therapy Oncology Group (RTOG) Recursive Partitioning Analysis (RPA) was the only well-established prognostic index for patients with BMs.4

The RPA index divided patients into 3 classes using the following 4 factors: age, Karnofsky Performance Status (KPS), primary tumor status, and extracranial metastases. One of the weaknesses of this system was that the majority of patients eligible for the clinical trials (KPS score ≥70) tended to be classified into RPA class II; in fact, in JROSG 99-1, 86% of the patients were classified into RPA class II. Another weakness is that the RPA system is not diagnosis specific. In 2012, Sperduto et al5 refined the “original” RPA and proposed a new index, namely the diagnosis-specific Graded Prognostic Assessment (DS-GPA).5 In the DS-GPA index, different scoring systems were prepared for 6 different primary tumor sites. The significant factors used for scoring were KPS, age, the presence of extracranial metastases, and the number of BMs for non–small-cell lung cancer (NSCLC) and small-cell lung cancer (SCLC); tumor subtype, KPS, and age for breast cancer; KPS only for gastrointestinal (GI) cancer; and KPS and number of BMs for both renal cell cancer and melanoma. A score of 4.0 correlates with the best prognosis, whereas 0.0 correlates with the worst prognosis.

One of the realities of this index is that patients with different primary cancers and different pathological diagnoses have different prognoses even with the same DS-GPA score. For example, the median survival times (MSTs) of patients with NSCLC with GPA scores of 2.5 to 3.0 and 3.5 to 4.0 are 9.4 and 14.8 months, respectively, compared with 15.1 and 25.3 months, respectively, for breast cancer; 11.3 and 14.8 months, respectively, for renal cell cancer; and 6.9 and 13.5 months, respectively, for GI cancer.

After the initial report of JROSG 99-1,3 other groups reported results of similarly designed randomized clinical trials (RCTs).6,7 All 3 trials showed that the omission of up-front WBRT significantly increased the incidence of BTR at both the original and distant sites3,6,7; however, the impact on OS was not significant in the 2 largest trials,3,7 whereas in the smallest trial, the omission of WBRT was associated with improved OS.6 On the basis of these findings, the current trend for limited BMs has shifted toward SRS without up-front WBRT when the number of BMs is up to 4 (and increasingly for even a larger number of tumors).8 It remains uncertain, however, whether a policy of treatment with SRS alone could be safely applied to all patient populations with BMs, recognizing that they exhibit heterogeneous prognoses. Herein, we report the results of a secondary analysis of the JROSG 99-1 data poststratified by the patients’ DS-GPA scores.

Box Section Ref ID

At a Glance

  • To investigate the feasibility of stereotactic radiosurgery (SRS) alone compared with whole-brain radiotherapy (WBRT) + SRS for patients with brain metastases exhibiting different prognoses determined by the diagnosis-specific Graded Prognostic Assessment (DS-GPA).

  • A secondary analysis of a randomized trial comparing SRS alone with WBRT + SRS (Japanese Radiation Oncology Study Group [JROSG] 99-1) for 88 patients with non–small-cell lung cancer with 1 to 4 BMs.

  • In a favorable prognosis group, with DS-GPA score of 2.5 to 4.0 (DS-GPA 2.5-4.0 group), the median survival time of the WBRT + SRS arm was significantly longer (16.7 months) than that of the SRS-alone arm (10.6 months; P = .04). In an unfavorable prognosis group, with DS-GPA scores of 0.5 to 2.0 (DS-GPA 0.5-2.0 group), such a difference was not observed (P = .86).

  • This benefit could be explained by the difference in brain tumor recurrence rate, in that the prevention effect of brain tumor recurrence by WBRT had a greater impact in the DS-GPA 2.5-4.0 group (P < .001) than in the DS-GPA 0.5-2.0 group (P = .04).

  • Despite the current trend of preferring SRS alone, we need to carefully consider the important role of WBRT, especially for patients with BMs from non–small-cell lung cancer with a favorable prognosis.

Methods
Study Design

This secondary post hoc analysis was based on an RCT, the JROSG 99-1 trial comparing SRS alone with WBRT + SRS for up to 4 BMs. Eligible patients were required to have a KPS score of 70 or more, age 18 years or older, and 1 to 4 BMs with a maximum diameter of 3 cm or smaller on contrast-enhanced magnetic resonance images, derived from a histologically confirmed systemic cancer. Patients with BMs from small-cell carcinoma, lymphoma, germinoma, and multiple myeloma were excluded. Before randomization, the patients were stratified based on the following criteria: primary tumor site (lung vs other sites), number of BMs (single vs 2-4), and the status of extracranial disease (controlled vs uncontrolled).

The research protocol was approved by the relevant institutional review boards or ethics committees, and all participants gave written informed consent. The recruitment period was from October 1999 to December 2003. The data were updated in July 2014, and the secondary analysis was performed in September 2014. The median follow-up time of the 88 patients with NSCLC included in this analysis was 8.05 months (range, 0.5-163.8 months). There were 160 eligible patients, and after excluding 28 patient because of various reasons, as described in the original publication,3 132 (83%) were randomized (65 to the WBRT + SRS arm and 67 to the SRS-alone arm). Patient accrual was prematurely terminated before the planned accrual number was reached.

For DS-GPA-based secondary analysis in the present study, only patients with NSCLC were included because of the lack of the tumor subtype information for patients with breast cancer and also to diminish the potential bias caused by the BMs from different primary types.5 The details of the DS-GPA classification system for NSCLC are described in eTable 1 in the Supplement.

Treatments

The WBRT schedule was 30 Gy in 10 fractions over 2 to 2.5 weeks. For the patients assigned to the WBRT + SRS arm, WBRT was followed by SRS. The SRS dose was prescribed to the tumor margin. Metastatic tumors with a maximum diameter of up to 2 cm were treated with 22 to 25 Gy, and those diameters larger than 2 cm were treated with 18 to 20 Gy. The dose was reduced by 30% when the treatment was combined with WBRT.

Statistical Analyses

The primary end point of the original study was the patients’ OS. Secondary end points included BTR, salvage treatment, and radiation toxic effects. All analyses were conducted on an intention-to-treat basis. End points were measured from the date of randomization. For time-to-event outcomes, the Kaplan-Meier method was used to estimate the median time to the event, and the differences were compared using a log-rank test. To identify significant variables associated with OS, multivariate analysis by Cox proportional hazards model was performed to calculate hazard ratios (HRs) and 95% CIs. A forward stepwise regression procedure with a cutoff of P = .05 was used. Candidate variables included KPS score (70-80 vs 90-100), extracranial metastases (present vs absent), the status of primary tumor (uncontrolled vs controlled), and the DS-GPA score. The DS-GPA score was categorized to 4 groups (0-1.0, 1.5-2.0, 2.5-3.0, and 3.5-4.0) and then transformed to a design variable. The Fisher exact test was used for the comparison of categorical variables, and the Wilcoxon rank-sum test was used for the comparison of continuous variables. P < .05, 2 sided, was considered statistically significant. All statistical analyses were performed one of the authors (H.A.) using JMP 11 software (SAS Institute Inc) and verified by a statistician (K.A.) using SPSS version 20 (IBM Corp).

Results
Patients’ Distributions According to the DS-GPA

The CONSORT diagram of patients with NSCLC is provided in Figure 1.9 Among the 132 randomized patients, 88 (67%) had NSCLC (45 SRS-alone arm and 43 WBRT + SRS arm). Seventy-five patients (85%) were male and 13 were female (15%). Forty-seven patients had a favorable prognosis (defined as DS-GPA score of 2.5-4.0; 26 in the SRS-alone arm and 21 in the WBRT + SRS arm [DS-GPA 2.5-4.0 group]) and 41 had an unfavorable prognosis (DS-GPA score of 0.5-2.0; 19 patients in the SRS-alone arm and 22 in the WBRT + SRS arm [DS-GPA 0.5-2.0 group]). The baseline characteristics of the treatment arms were well balanced in regard to KPS, age, number of BMs, status of primary tumor, and the existence of extracranial metastases (Table 1).

OS, Brain Tumor Recurrence, and Toxic Effects

As a result of Cox analysis, it was suggested that a dose-dependent DS-GPA score (Reference: DS-GPA 3.5-4.0; DS-GPA 0.0-1.0: HR, 7.48 [95% CI, 2.13-26.33] P = .002; DS-GPA 1.5-2.0: HR, 3.04 [95% CI, 1.10-8.39] P = .03; DS-GPA 2.5-3.0: HR, 1.77 (95% CI, 0.65-4.77] P = .26) and the status of the primary tumor (HR, 1.90, 95% CI, 1.11-3.23; P = .02) were selected as the independent and significant variables for OS. The OS values of each treatment group according to the DS-GPA are summarized in Table 2. Significantly better OS was observed in the patients with a favorable prognosis (DS-GPA 2.5-4.0) in the WBRT + SRS arm vs the SRS-alone arm, with the MST values of 16.7 (95% CI, 7.5-72.9) months vs 10.6 (95% CI, 7.7-15.5) months (P = .04) with an HR of 1.92 (95% CI, 1.01-3.78) in favor of WBRT + SRS (Figure 2A). However, this survival differential was not observed in the patients with an unfavorable prognosis (DS-GPA 0.5-2.0) (HR, 1.05; 95% CI, 0.55-1.99) (P = .86) (Figure 2B).

The frequency and the pattern of BTR, median BTR-free time, and the number of patients requiring salvage brain therapy are summarized in Table 2. The omission of WBRT increased BTRs at both the initial and distant sites in the brain. The preventive effect of WBRT was most prominent in the DS-GPA 2.5-4.0 group (HR, 8.31; 95% CI, 3.05–29.13) (P < .001) (Figure 3A) compared with the DS-GPA 0.5-2.0 group (HR, 3.57; 95% CI, 1.02-16.49) (P = .04) (Figure 3B).

As a result, salvage brain treatment for BTR was more frequently required in the patients in the DS-GPA 2.5-4.0 stratum who received SRS alone (54%) compared with both the DS-GPA 2.5-4.0 patients who were allocated to the WBRT + SRS arm (19%) and the DS-GPA 0.5-2.0 patients with unfavorable prognoses (SRS alone, 21%; and WBRT + SRS arm, 9%). Regarding radiation-induced late toxic effects, all 3 of the patients who developed grade 3 or 4 toxic effects belonged to the DS-GPA 2.5-4.0 group.

Neurocognitive Function

Neurocognitive function was assessed by the Japanese version of the Mini-Mental State Examination (MMSE), and the results are summarized in eTable 2 in the Supplement. Baseline data were available in 70 patients. At baseline, the MMSE score in the GPA 2.5-4.0 group was significantly better than that in the GPA 0.5-2.0 group (28.0 vs 27.0; P = .01). When the 2 prognostic groups (DS-GPA 0.5-2.0 and 2.5-4.0) were considered separately, there was no significant difference in baseline MMSE scores between the 2 treatment arms in either group. Follow-up MMSE data were available in 57 patients (81%). Among the 24 patients in the DS-GPA 0.5-2.0 group, the median duration until the last follow-up MMSE was 3.6 (range, 1.3-14.5) months in the SRS-alone arm and 3.6 (range, 1.3-49.5) months in the WBRT + SRS arm (P = .86). Among the 33 patients in the DS-GPA 2.5-4.0 group, these values were 8.5 (range, 1.4-49.8) months and 9.5 (1.8-58.7) months, respectively (P = .81). Regarding the score at the last follow-up, no significant difference between the treatment arms was observed in either the DS-GPA 0.5-2.0 group (SRS-alone arm, 27.5; and WBRT + SRS arm 28.0; P = .77) or DS-GPA 2.5-4.0 group (SRS-alone arm, 28.0; and WBRT + SRS arm, 26.5; P = .40).

Discussion

It has been recognized that the addition of WBRT to SRS may increase the likelihood of cognitive adverse effects without increasing OS for patients with up to 3 or 4 BMs, although it significantly reduces BTR at both the initial and distant sites.6 As a result, over the last decade SRS without up-front WBRT has seen increasing use as a treatment for patients with up to 3 or 4 BMs; moreover, a recent report indicated that the SRS-alone strategy could be safely applied for up to 10 BMs.8 However, the major flaw in this approach is the failure to recognize that improved intracranial control could translate to improved survival in patients at preferential risk of dying from intracranial as opposed to extracranial progression. There are several examples of this in the radiotherapy literature, with the 2 most notable ones being improved survival through the control of intracranial disease, when WBRT is used prophylactically for either patients with limited SCLC who have experienced a good response to chemotherapy10 or patients with a single metastasis, for whom the only randomized data showing survival improvement were from trials that combined WBRT with SRS11 or resection.12 Because this survival benefit from WBRT is not observed uniformly, with the most notable examples of this being the use of prophylactic cranial irradiation in NSCLC13 and the addition of WBRT to unselected cohorts of patients with limited BMs managed with SRS,3,7 it has not been clear whether there are subsets embedded within these groups that may actually experience a survival benefit from WBRT.

An initial hint in this regard can be found in the study conducted by Pirzkall et al,14 who described 236 patients treated with SRS with or without the nonrandom use of WBRT. In the subset of patients without extracranial disease, ie, the group least likely to rapidly succumb to extracranial progression, SRS + WBRT resulted in a median survival of 15.4 months, compared with 8.3 months for those treated with SRS alone (P = .08). Although this observation was not significant, it provided the hypothesis that improved intracranial control resulting from WBRT could potentially affect overall survival in selected subsets of patients. The JROSG 99-1 trial was the first RCT comparing SRS alone with WBRT + SRS. In the initial analysis, we could not extract groups for whom the combination therapy conferred a survival benefit because there was no sufficiently sensitive prognostic index available in 2006. In the present secondary post hoc analysis, however, we were indeed able to identify such a group, ie, the NSCLC patients with a favorable prognosis (DS-GPA 2.5-4.0) appeared to benefit by the addition of WBRT to SRS. This survival benefit, which did not extend to the unfavorable prognostic group, may have been attributable to the prevention of BTR by WBRT, which had more impact in the favorable than in the unfavorable group.

In this context, we note that Sperduto et al15 recently published the results of a secondary analysis of RTOG 9508, which is an RCT comparing WBRT alone and WBRT + SRS for patients with up to 3 BMs. The initial report showed that patients had a survival benefit on post hoc analysis if they had NSCLC.11 In the secondary analysis, patients were poststratified according to the DS-GPA score. It is of note that patients with breast cancer were excluded because of the lack of HER2 status information; as a result, NSCLC became more dominant in the secondary report (84%) than in the initial report (64%). Sperduto et al15 found that there was no survival difference between treatments when they analyzed the overall group; however, patients with a DS-GPA score of 3.5 to 4.0 had better OS when treated with WBRT + SRS (MST, 21.0 months) than with WBRT alone (MST, 10.3 months) (P = .05).15

By combining the findings of the DS-GPA–based secondary analyses of JROSG 99-1 and RTOG 9508, it becomes clear that patients with BM from NSCLC with a favorable prognosis could have realized a survival benefit by the combination of WBRT and SRS compared with either SRS alone or WBRT alone; for this reason, an improved understanding of the long-term neurocognitive outcomes is increasingly important.

The impact of up-front WBRT on neurocognitive function and neurological adverse events has remained uncertain owing to the high risk of performance and detection bias and the lack of consistency in the instruments and methods used to measure and report results across studies.16 In a trial conducted by Chang and colleagues,6 the deterioration in learning and memory function at 4 months after treatment was significantly more frequent among the patients who received WBRT + SRS than among those who received SRS alone. However, the deterioration at 4 months is usually transient and could be reversed to baseline by 8 months17; therefore, neurocognitive function as measured by the Hopkins Verbal Learning Test–Total Recall (HVLT-TR) at 4 months might not be adequate for understanding the full trajectory of neurocognitive function following WBRT in patients with a favorable prognosis.18

It is important to note that neurocognitive function is also closely related to the brain tumor burden, and in one report, the preservation of neurocognitive function was better among the patients whose tumor burden became smaller after brain irradiation.19 In the initial analysis of neurocognitive function in JROSG 99-1, a trend of better preservation of neurocognitive function at 12 months was observed in the WBRT + SRS arm (76%) compared with the SRS-alone arm (59%) owing to the better brain tumor control in the WBRT + SRS arm.20 In the present analysis, no significant difference in the MMSE score at the last follow-up was observed when the patients were classified by their DS-GPA scores, implying no excess residual cognitive dysfunction in the WBRT + SRS arm. This may have been because the positive effect on neurocognitive function of the reduced frequency of BTR after WBRT and the negative impact of the late adverse effects of WBRT offset each other. Nevertheless, the deterioration of neurocognitive function as a consequence of late adverse events following WBRT is a real and serious matter of concern for long-term survivors. N-methyl-d-aspartate (NMDA)-receptor agonists used to treat Alzheimer disease have been shown to delay the progression of neurocognitive deterioration associated with WBRT.21,22

Hippocampal-avoidance WBRT was recently reported to preserve cognition compared with WBRT in a single-arm phase 2 trial, and phase 3 trials are under way.23 Modification of the dose-fractionation schedule of WBRT combined with SRS would be another approach. Hypofractionated WBRT regimens including 30 Gy in 10 fractions or 37.5 Gy in 15 fractions became standard through RTOG trials in the 1970s and 1980s, when SRS was not widely available. In addition, the early detection of BMs was usually not possible because of the limited availability of MRI. Today, such modalities have become a part of daily practice; therefore, the role of WBRT combined with SRS is different from that in the 1970s and 1980s.

We recently launched a clinical trial investigating the combination of reduced-dose as well as reduced dose-per-fraction WBRT (25 Gy in 10 fractions) combined with SRS (JROSG 13-1) for patients satisfying eligibility criteria similar to those used in the JROSG 99-1 trial. This dose-fractionation schedule has been commonly used in prophylactic cranial irradiation for patients with SCLC and those with NSCLC, and the degree of long-term toxic effects in neurocognitive function has been confirmed to be milder than that from the more conventionally hypofractionated schedules such as 30 Gy in 10 fractions or 37.5 Gy in 15 fractions.24 In light of the survival benefit of the combination of WBRT and SRS, intensive efforts to reduce the cognitive effects of WBRT are now warranted.

The present study has several limitations that are common to all secondary analyses. First, the patients were not stratified by DS-GPA scores; as a result, the imbalance of patient distribution within the DS-GPA 2.5-4.0 group could not be completely eliminated, although the differences were not significant. Second, over the last decade great progress has been made in systemic therapies, including molecularly targeted therapies.25 Epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors (TKIs) are now a standard treatment for patients with advanced NSCLC with EGFR mutations, and they may also be effective for controlling BMs in such patients.26 Up-front EGFR-TKIs could be one of the treatment choices for patients with EGFR-mutant NSCLC with asymptomatic BMs, but it remains unsolved whether up-front EGFR-TKI treatment or radiation therapy is more appropriate owing to the absence of head-to-head comparisons.27 Similar data and questions are now also emerging for NSCLC cases positive for echinoderm microtubule–associated protein-like 4-anaplastic lymphoma kinase.28

Conclusions

Despite the current trend of preferring SRS alone, we need to carefully consider the important role of WBRT, especially for patients with BMs from NSCLC who have a favorable prognosis. These findings should be validated through prospective studies, not only for NSCLC but also for other primary cancers. In addition, further investigations targeting WBRT methods that result in less cognitive impairment with a reliable and durable neurocognitive end point after treatment are warranted.

Back to top
Article Information

Accepted for Publication: March 30, 2015.

Corresponding Author: Hidefumi Aoyama, MD, PhD, Department of Radiology, Niigata University Graduate School of Medical and Dental Sciences, 1-757 Asahimachi-dori, Chuo-ku, Niigata 951-8510, Japan (h-aoyama@med.niigata-u.ac.jp).

Published Online: May 14, 2015. doi:10.1001/jamaoncol.2015.1145.

Author Contributions: Dr Aoyama had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Aoyama and Tago contributed equally to this study.

Study concept and design: All authors.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Aoyama.

Critical revision of the manuscript for important intellectual content: Tago, Shirato.

Statistical analysis: Aoyama.

Obtained funding: Aoyama, Shirato.

Administrative, technical, or material support: Shirato.

Study supervision: Shirato.

Conflict of Interest Disclosures: Dr Aoyama has a consulting role with Brainlab, AG, outside the submitted work. Dr Shirato has received grants from Hitachi Co, Ltd, Mitsubishi Heavy Industry, and Shimadzu Co, Ltd, outside the submitted work. No other disclosures are reported.

Funding/Support: This work was supported in part by the JROSG and the Niigata University Clinical Research Project.

Role of the Funder/Sponsor: The JROSG and the Niigata University Clinical Research Project had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Previous Presentations: This study was reported in part at the 27th Annual Meeting of the Japanese Society for Radiation Oncology December 13, 2014; Yokohama, Japan. The initial report was published in JAMA (Aoyama et al3), and a related report was published in Int J Radiat Oncol Biol Phys (Aoyama et al20).

Group Information: The JROSG 99-1 investigators were Norio Kato, MD, Department of Radiation Medicine, Hokkaido University, Sapporo, Japan; Toshihiko Iuchi, MD, Department of Neurosurgery, Chiba Cancer Center, Chiba, Japan; Keiichi Nakagawa, MD, Department of Radiology, University of Tokyo Hospital, Tokyo, Japan; Tatsuya Toyoda, MD, Department of Radiology, Kanto Medical Center NTT EC, Tokyo, Japan; Kazuo Hatano, MD, Department of Radiology, Chiba Cancer Center, Chiba, Japan; Masahiro Kenjyo, MD, Department of Radiology, Hiroshima University School of Medicine, Hiroshima, Japan; Natsuo Oya, MD, Department of Radiology, Kyoto University School of Medicine, Kyoto, Japan; Saeko Hirota, MD, Department of Radiology, The Hyogo Medical Center for Adults, Akashi, Japan; Hiroki Shioura, MD, Department of Radiology, Izumisano General Hospital, Izumisano, Japan; Etsuo Kunieda, MD, Department of Radiology, Keio University School of Medicine, Tokyo, Japan; Taisuke Inomata, MD, Department of Radiology, Osaka Medical College, Osaka, Japan; and Kazushige Hayakawa, MD, Department of Radiology, Kitazato Medical School, Sagamihara, Japan.

Additional Contributions: We thank Dr Kohei Akazawa, PhD of the Department of Medical Informatics, Niigata University Medical and Dental Hospital, Niigata, who contributed to the verification of statistical analyses, Dr Norio Kato, MD of the Department of Radiation Medicine, Hokkaido University, Sapporo, and Dr Toshihiko Iuchi, MD, of the Department of Neurosurgery, Chiba Cancer Center, Chiba who contributed to the acquisition of follow-up data, and Dr Minesh P. Mehta, MD, of the Maryland Proton Treatment Center, University of Maryland School of Medicine for reviewing initial drafts of the manuscript and suggesting changes, and the following colleagues for their contributions to the data collection in the initial work: Keiichi Nakagawa, MD, Department of Radiology, University of Tokyo Hospital, Tokyo, Japan; Tatsuya Toyoda, MD, Department of Radiology, Kanto Medical Center NTT EC, Tokyo; Kazuo Hatano, MD, Department of Radiology, Chiba Cancer Center, Chiba, Japan; Masahiro Kenjyo, MD, Department of Radiology, Hiroshima University School of Medicine, Hiroshima, Japan; Natsuo Oya, MD, Department of Radiology, Kyoto University School of Medicine, Kyoto, Japan; Saeko Hirota, MD, Department of Radiology, The Hyogo Medical Center for Adults, Akashi, Japan; Hiroki Shioura, MD, Department of Radiology, Izumisano General Hospital, Izumisano, Japan; Etsuo Kunieda, MD, Department of Radiology, Keio University School of Medicine, Tokyo; Taisuke Inomata, MD, Department of Radiology, Osaka Medical College, Osaka, Japan; Kazushige Hayakawa, MD, Department of Radiology, Kitazato Medical School, Sagamihara, Japan.

References
1.
Lin  NU, Lee  EQ, Aoyama  H,  et al; Response Assessment in Neuro-Oncology (RANO) group.  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.PubMedGoogle ScholarCrossref
2.
Sperduto  PW, Berkey  B, Gaspar  LE, Mehta  M, Curran  W.  A new prognostic index and comparison to three other indices for patients with brain metastases: an analysis of 1,960 patients in the RTOG database.  Int J Radiat Oncol Biol Phys. 2008;70(2):510-514.PubMedGoogle ScholarCrossref
3.
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.PubMedGoogle ScholarCrossref
4.
Gaspar  L, Scott  C, Rotman  M,  et al.  Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials.  Int J Radiat Oncol Biol Phys. 1997;37(4):745-751.PubMedGoogle ScholarCrossref
5.
Sperduto  PW, Kased  N, Roberge  D,  et al.  Summary report on the graded prognostic assessment: an accurate and facile diagnosis-specific tool to estimate survival for patients with brain metastases.  J Clin Oncol. 2012;30(4):419-425.PubMedGoogle ScholarCrossref
6.
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.PubMedGoogle ScholarCrossref
7.
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.PubMedGoogle ScholarCrossref
8.
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.PubMedGoogle ScholarCrossref
9.
Piaggio  G, Elbourne  DR, Pocock  SJ, Evans  SJ, Altman  DG; CONSORT Group.  Reporting of noninferiority and equivalence randomized trials: extension of the CONSORT 2010 statement.  JAMA. 2012;308(24):2594-2604.PubMedGoogle ScholarCrossref
10.
Aupérin  A, Arriagada  R, Pignon  JP,  et al; Prophylactic Cranial Irradiation Overview Collaborative Group.  Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission.  N Engl J Med. 1999;341(7):476-484.PubMedGoogle ScholarCrossref
11.
Andrews  DW, Scott  CB, Sperduto  PW,  et al.  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 randomised trial.  Lancet. 2004;363(9422):1665-1672.PubMedGoogle ScholarCrossref
12.
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.PubMedGoogle ScholarCrossref
13.
Gore  EM, Bae  K, Wong  SJ,  et al.  Phase III comparison of prophylactic cranial irradiation versus observation in patients with locally advanced non-small-cell lung cancer: primary analysis of radiation therapy oncology group study RTOG 0214.  J Clin Oncol. 2011;29(3):272-278.PubMedGoogle ScholarCrossref
14.
Pirzkall  A, Debus  J, Lohr  F,  et al.  Radiosurgery alone or in combination with whole-brain radiotherapy for brain metastases.  J Clin Oncol. 1998;16(11):3563-3569.PubMedGoogle Scholar
15.
Sperduto  PW, Shanley  R, Luo  X,  et al.  Secondary analysis of RTOG 9508, a phase 3 randomized trial of whole-brain radiation therapy versus WBRT plus stereotactic radiosurgery in patients with 1-3 brain metastases; poststratified by the graded prognostic assessment (GPA).  Int J Radiat Oncol Biol Phys. 2014;90(3):526-531.PubMedGoogle ScholarCrossref
16.
Soon  YY, Tham  IW, Lim  KH, Koh  WY, Lu  JJ.  Surgery or radiosurgery plus whole brain radiotherapy versus surgery or radiosurgery alone for brain metastases.  Cochrane Database Syst Rev. 2014;3:CD009454.PubMedGoogle Scholar
17.
Onodera  S, Aoyama  H, Tha  KK,  et al.  The value of 4-month neurocognitive function as an endpoint in brain metastases trials.  J Neurooncol. 2014;120(2):311-319.PubMedGoogle ScholarCrossref
18.
McDuff  SG, Taich  ZJ, Lawson  JD,  et al.  Neurocognitive assessment following whole brain radiation therapy and radiosurgery for patients with cerebral metastases.  J Neurol Neurosurg Psychiatry. 2013;84(12):1384-1391.PubMedGoogle ScholarCrossref
19.
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.PubMedGoogle ScholarCrossref
20.
Aoyama  H, Tago  M, Kato  N,  et al.  Neurocognitive function of patients with brain metastasis who received either whole brain radiotherapy plus stereotactic radiosurgery or radiosurgery alone.  Int J Radiat Oncol Biol Phys. 2007;68(5):1388-1395.PubMedGoogle ScholarCrossref
21.
Shaw  EG, Rosdhal  R, D’Agostino  RB  Jr,  et al.  Phase II study of donepezil in irradiated brain tumor patients: effect on cognitive function, mood, and quality of life.  J Clin Oncol. 2006;24(9):1415-1420.PubMedGoogle ScholarCrossref
22.
Brown  PD, Pugh  S, Laack  NN,  et al; Radiation Therapy Oncology Group (RTOG).  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.PubMedGoogle ScholarCrossref
23.
Gondi  V, Pugh  SL, Tome  WA,  et al.  Preservation of memory with conformal avoidance of the hippocampal neural stem-cell compartment during whole-brain radiotherapy for brain metastases (RTOG 0933): a phase II multi-institutional trial.  J Clin Oncol. 2014;32(34):3810-3816.PubMedGoogle ScholarCrossref
24.
Grosshans  DR, Meyers  CA, Allen  PK, Davenport  SD, Komaki  R.  Neurocognitive function in patients with small cell lung cancer : effect of prophylactic cranial irradiation.  Cancer. 2008;112(3):589-595.PubMedGoogle ScholarCrossref
25.
Maemondo  M, Inoue  A, Kobayashi  K,  et al; North-East Japan Study Group.  Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR.  N Engl J Med. 2010;362(25):2380-2388.PubMedGoogle ScholarCrossref
26.
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.PubMedGoogle ScholarCrossref
27.
Gerber  NK, Yamada  Y, Rimner  A,  et al.  Erlotinib versus radiation therapy for brain metastases in patients with EGFR-mutant lung adenocarcinoma.  Int J Radiat Oncol Biol Phys. 2014;89(2):322-329.PubMedGoogle ScholarCrossref
28.
Owonikoko  TK, Arbiser  J, Zelnak  A,  et al.  Current approaches to the treatment of metastatic brain tumours.  Nat Rev Clin Oncol. 2014;11(4):203-222.PubMedGoogle ScholarCrossref
×