Bierta E. Barfod

Subclassification of Recursive Partitioning Analysis Class II Patients With Brain Metastases Treated Radiosurgically

PURPOSE Although the recursive partitioning analysis (RPA) class is generally used for predicting survival periods of patients with brain metastases (METs), the majority of such patients are Class II and clinical factors vary quite widely within this category. This prompted us to divide RPA Class II patients into three subclasses. METHODS AND MATERIALS This was a two-institution, institutional review board-approved, retrospective cohort study using two databases: the Mito series (2,000 consecutive patients, comprising 787 women and 1,213 men; mean age, 65 years [range, 19-96 years]) and the Chiba series (1,753 patients, comprising 673 female and 1,080 male patients; mean age, 65 years [range, 7-94 years]). Both patient series underwent Gamma Knife radiosurgery alone, without whole-brain radiotherapy, for brain METs during the same 10-year period, July 1998 through June 2008. The Cox proportional hazard model with a step-wise selection procedure was used for multivariate analysis. RESULTS In the Mito series, four factors were identified as favoring longer survival: Karnofsky Performance Status (90% to 100% vs. 70% to 80%), tumor numbers (solitary vs. multiple), primary tumor status (controlled vs. not controlled), and non-brain METs (no vs. yes). This new index is the sum of scores (0 and 1) of these four factors: RPA Class II-a, score of 0 or 1; RPA Class II-b, score of 2; and RPA Class II-c, score of 3 or 4. Next, using the Chiba series, we tested whether our index is valid for a different patient group. This new system showed highly statistically significant differences among subclasses in both the Mito series and the Chiba series (p < 0.001 for all subclasses). In addition, this new index was confirmed to be applicable to Class II patients with four major primary tumor sites, that is, lung, breast, alimentary tract, and urogenital organs. CONCLUSIONS Our new grading system should be considered when designing future clinical trials involving brain MET patients.

Treatment of brain metastasis from lung cancer.

Brain metastasis from lung cancer occupies a significant portion of all brain metastases. About 15-20% of patients with non-small cell lung cancer (NSCLC) develop brain metastasis during the course of the disease. The prognosis of brain metastasis is poor with median survival of less than 1 year. Whole-brain radiation therapy (WBRT) is widely used for the treatment of brain metastasis. WBRT can also be used as adjuvant treatment along with surgery and stereotactic radiosurgery (SRS).Surgery provides a rapid relief of mass effects and may be the best choice for a large single metastasis. SRS confers local control rates comparable to those for surgery with minimal toxicities and versatility that makes it applicable to multiple lesions, deep-seated lesions, and to patients with poor medical conditions. Recursive partitioning analysis (RPA) classes are widely used for prognostic stratification. However, the validity of RPA classes, especially for NSCLC, has been questioned and other scoring systems are being developed. Synchronous presentation of primary NSCLC and brain metastases is a special situation in which surgery for the lung lesion and surgery or SRS for brain lesions are recommended if the thoracic disease is in early stages. Small cell lung cancer (SCLC) has a higher likelihood for brain metastasis than NSCLC and prophylactic cranial irradiation and subsequent WBRT are usually recommended. Recently, SRS for brain metastasis from SCLC has been tried, but requires further verification.

Gamma Knife surgery for patients with brainstem metastases

OBJECT Because brainstem metastases are not deemed resectable, stereotactic radiosurgery (SRS) is the only treatment modality expected to achieve a radical cure. The authors describe their treatment results, focusing particularly on how long patients can survive without neurological deterioration following SRS for brainstem metastases. METHODS This was an institutional review board-approved, retrospective cohort study in which the authors pulled from their database information on 2553 consecutive patients with brain metastases who underwent Gamma Knife surgery (GKS) at the Mito GammaHouse between July 1998 and July 2011. Among the 2553 patients, excluding cases in which there was meningeal dissemination, 200 cases of brainstem metastases (78 women and 122 men with a mean age of 64 years [range 36-86 years]) were identified and analyzed. The most common primary site was the lung (137 patients) followed by the gastrointestinal tract (24 patients), breast (17 patients), kidney (12 patients), and others (10 patients). Among the 200 patients, 15 patients (7.5%) harbored at least 2 tumors in the brainstem: 11 patients had 2 tumors, 2 patients had 3 tumors, and 1 patient each had 4 or 5 tumors. Therefore, a total of 222 tumors were irradiated. These 222 tumors were located in the pons (121 lesions), the midbrain (65 lesions), and the medulla oblongata (36 lesions). The mean and median tumor volumes were 1.3 and 0.2 cm(3) (range 0.005-10.7 cm(3)), and the median peripheral radiation dose was 18.0 Gy (range 12.0-25.0 Gy). RESULTS The overall median survival time (MST) was 6.0 months. Distribution of MSTs across Recursive Partitioning Analysis (RPA) classes showed that the MSTs were 9.4 months in Class I (20 patients), 6.0 months in Class II (171 patients), and 1.9 months in Class III (9 patients). Better Karnofsky Performance Scale score, single metastasis, and well-controlled primary tumor were significant predictive factors for longer survival. The neurological and qualitative survival rates were 90.8% and 89.2%, respectively, at 24 months post-GKS. Better KPS score and smaller tumor volume tended to be associated with prolonged qualitative survival. Follow-up imaging studies were available for 129 patients (64.5%). The tumor control rate was 81.8% at 24 months post-GKS. Smaller tumor volume tended to contribute to tumor control. CONCLUSIONS The present results indicate that GKS is effective in the treatment of brainstem metastases, particularly from the viewpoint of maintaining a good neurological condition in the patient.

Delayed Complications in Patients Surviving at Least 3 Years After Stereotactic Radiosurgery for Brain Metastases

PURPOSE Little is known about delayed complications after stereotactic radiosurgery in long-surviving patients with brain metastases. We studied the actual incidence and predictors of delayed complications. PATIENTS AND METHODS This was an institutional review board-approved, retrospective cohort study that used our database. Among our consecutive series of 2000 patients with brain metastases who underwent Gamma Knife radiosurgery (GKRS) from 1991-2008, 167 patients (8.4%, 89 women, 78 men, mean age 62 years [range, 19-88 years]) who survived at least 3 years after GKRS were studied. RESULTS Among the 167 patients, 17 (10.2%, 18 lesions) experienced delayed complications (mass lesions with or without cyst in 8, cyst alone in 8, edema in 2) occurring 24.0-121.0 months (median, 57.5 months) after GKRS. The actuarial incidences of delayed complications estimated by competing risk analysis were 4.2% and 21.2% at the 60th month and 120th month, respectively, after GKRS. Among various pre-GKRS clinical factors, univariate analysis demonstrated tumor volume-related factors: largest tumor volume (hazard ratio [HR], 1.091; 95% confidence interval [CI], 1.018-1.154; P=.0174) and tumor volume≤10 cc vs >10 cc (HR, 4.343; 95% CI, 1.444-12.14; P=.0108) to be the only significant predictors of delayed complications. Univariate analysis revealed no correlations between delayed complications and radiosurgical parameters (ie, radiosurgical doses, conformity and gradient indexes, and brain volumes receiving >5 Gy and >12 Gy). After GKRS, an area of prolonged enhancement at the irradiated lesion was shown to be a possible risk factor for the development of delayed complications (HR, 8.751; 95% CI, 1.785-157.9; P=.0037). Neurosurgical interventions were performed in 13 patients (14 lesions) and mass removal for 6 lesions and Ommaya reservoir placement for the other 8. The results were favorable. CONCLUSIONS Long-term follow-up is crucial for patients with brain metastases treated with GKRS because the risk of complications long after treatment is not insignificant. However, even when delayed complications occur, favorable outcomes can be expected with timely neurosurgical intervention.

How Many Metastases Can Be Treated with Radiosurgery

We describe postradiosurgical treatment outcomes of our consecutive series of 1,676 patients (654 females and 1,022 males, mean age 63 years, range 19-92 years) who underwent Gamma Knife radiosurgery (GKRS) for brain metastases, focusing particularly on GKRS for multiple lesions. The most common primary cancer was lung (1,057; 63.1%), followed by alimentary tract (198; 11.8%), breast (180; 10.7%), uro-genital (113; 6.7%) and others (128; 7.6%). Mean and median lesion numbers were 7 and 3, respectively, range 1-85. The overall median survival times were 9.0 months in females and 5.9 in males after GKRS (p < 0.0001). The Kaplan-Meier method was used to assess tumor numbers by group: 1-4, 5-9, 10-14, 15-19, 20-24, 25-29, 30-39 and ≥40. The post-GKRS median survival times were 8.3, 5.3, 6.9, 5.2, 5.6, 3.0, 5.3 and 4.3 months, respectively (p < 0.0001). Also, the Kaplan-Meier method was used to compare 15 pairs of groups based on tumor numbers: 1 vs. ≥2, ≤2 vs. ≥3, ≤3 vs. ≥4, ---, and ≤15 vs. ≥16. In each of the 15 pairs, the median survival times of patients with lower tumor numbers were significantly longer than those of patients with higher tumor numbers (p < 0.0001). Furthermore, 14 other pairs of groups, based on tumor numbers, were also assessed by this method: 1 vs. 2, 2 vs. 3, 3 vs. 4, ---, and 14 vs. 15. Among the 14 pairs, only the 1 vs. 2 pair showed a significant median survival time difference (p = 0.0002); no significant differences were detected for the other 13 pairs. Although tumor number was demonstrated to have a significant impact on the duration of survival, approximately 85% of patients with brain metastasis died of causes other than brain disease progression, regardless of tumor number.

Validity of Three Recently Proposed Prognostic Grading Indexes for Breast Cancer Patients With Radiosurgically Treated Brain Metastases

PURPOSE We tested the validity of 3 recently proposed prognostic indexes for breast cancer patients with brain metastases (METs) treated radiosurgically. The 3 indexes are Diagnosis-Specific Graded Prognostic Assessment (DS-GPA), New Breast Cancer (NBC)-Recursive Partitioning Analysis (RPA), and our index, sub-classification of RPA class II patients into 3 sub-classes (RPA class II-a, II-b and II-c) based on Karnofsky performance status, tumor number, original tumor status, and non-brain METs. METHODS AND MATERIALS This was an institutional review board-approved, retrospective cohort study using our database of 269 consecutive female breast cancer patients (mean age, 55 years; range, 26-86 years) who underwent Gamma Knife radiosurgery (GKRS) alone, without whole-brain radiation therapy, for brain METs during the 15-year period between 1996 and 2011. The Kaplan-Meier method was used to estimate the absolute risk of each event. RESULTS Kaplan-Meier plots of our patient series showed statistically significant survival differences among patients stratified into 3, 4, or 5 groups based on the 3 systems (P<.001). However, the mean survival time (MST) differences between some pairs of groups failed to reach statistical significance with all 3 systems. Thus, we attempted to regrade our 269 breast cancer patients into 3 groups by modifying our aforementioned index along with the original RPA class I and III, (ie, RPA I+II-a, II-b, and II-c+III). There were statistically significant MST differences among these 3 groups without overlap of 95% confidence intervals (CIs) between any 2 pairs of groups: 18.4 (95% CI = 14.0-29.5) months in I+II-a, 9.2 in II-b (95% CI = 6.8-12.9, P<.001 vs I+II-a) and 5.0 in II-c+III (95% CI = 4.2-6.8, P<.001 vs II-b). CONCLUSIONS As none of the new grading systems, DS-GPS, BC-RPA and our system, was applicable to our set of radiosurgically treated patients for comparing survivals after GKRS, we slightly modified our system for breast cancer patients.

Gamma Knife Radiosurgery for Brain Metastases of Non-Lung Cancer Origin: Focusing on Multiple Brain Lesions

We describe postradiosurgical treatment outcomes of our consecutive series of 456 patients (220 females, 236 males, mean age; 60.5 years, range 19-86 years) who underwent gamma knife (GK) treatment for brain metastases originating from non-lung cancers, focusing particularly on GK treatment for multiple lesions. The most common primary cancers were breast (122; 26.8%), followed by lower alimentary tract (105; 23.0%), uro-genital (100; 21.9%), upper alimentary tract (56; 12.3%), others (41; 9.0%) and unknown (32; 7.0%). Mean and median tumor numbers were 6 and 2, respectively, range 1-55. The mean and median survival times were 12.7 and 7.0 months after GK radiosurgery. Postradiosurgical survival rates were 52.7% at 6, 29.0% at 12, 19.1% at 18, 13.5% at 24, 6.5% at 36 and 5.0% at 60 months. Number of lesions, maximum and cumulative tumor volumes, non-symptomatic, well-controlled primary tumors, no non-brain metastatic lesions, Karnofsky performance status better than 80%, having prior surgery and having at least two procedures were significant predictive factors for survival. Although tumor number was demonstrated to have a significant impact on the duration of survival, approximately 85% of patients with brain metastases died of causes other than brain disease progression, regardless of tumor number.

Stereotactic radiosurgery for brain metastases: a case-matched study comparing treatment results for patients 80 years of age or older versus patients 65–79 years of age

OBJECT Recently, an increasing number of patients with brain metastases, even patients over 80 years of age, have been treated with stereotactic radiosurgery (SRS). However, there is little information on SRS treatment results for patients with brain metastases 80 years of age and older. The authors undertook this study to reappraise whether SRS treatment results for patients 80 years of age or older differ from those of patients who are 65-79 years old. METHODS This was an institutional review board-approved, retrospective cohort study. Among 2552 consecutive brain metastasis patients who underwent SRS during the 1998-2011 period, we studied 165 who were 80 years of age or older (Group A) and 1181 who were age 65-79 years old (Group B). Because of the remarkable disproportion in patient numbers between the 2 groups and considerable differences in pre-SRS clinical factors, the authors conducted a case-matched study using the propensity score matching method. Ultimately, 330 patients (165 from each group, A and B) were selected. For time-to-event outcomes, the Kaplan-Meier method was used to estimate overall survival and competing risk analysis was used to estimate other study end points, as appropriate. RESULTS Although the case-matched study showed that post-SRS median survival time (MST, months) was shorter in Group A patients (5.3 months, 95% CI 3.9-7.0 months) than in Group B patients (6.9 months, 95% CI 5.0-8.1 months), this difference was not statistically significant (HR 1.147, 95% CI 0.921-1.429, p = 0.22). Incidences of neurological death and deterioration were slightly lower in Group A than in Group B patients (6.3% vs 11.8% and 8.5% vs 13.9%), but these differences did not reach statistical significance (p = 0.11 and p = 0.16). Furthermore, competing risk analyses showed that the 2 groups did not differ significantly in cumulative incidence of local recurrence (HR 0.830, 95% CI 0.268-2.573, p = 0.75), rates of repeat SRS (HR 0.738, 95% CI 0.438-1.242, p = 0.25), or incidence of SRS-related complications (HR 0.616, 95% CI 0.152-2.495, p = 0.49). Among the Group A patients, post-SRS MSTs were 11.6 months (95% CI 7.8-19.6 months), 7.9 months (95% CI 5.2-10.9 months), and 2.8 months (95% CI; 2.4-4.6 months) in patients whose disease status was modified-recursive partitioning analysis (RPA) Class(es) I+IIa, IIb, and IIc+III, respectively (p < 0.001). CONCLUSIONS Our results suggest that patients 80 years of age or older are not unfavorable candidates for SRS as compared with those 65-79 years old. Particularly, even among patients 80 years and older, those with modified-RPA Class I+IIa or IIb disease are considered to be favorable candidates for more aggressive treatment of brain metastases.

Validity of prognostic grading indices for brain metastasis patients undergoing repeat radiosurgery.

OBJECTIVES We tested the validity of 5 prognostic indices, Recursive Partitioning Analysis (RPA), Score Index for Radiosurgery (SIR), Basic Score for Brain Metastases (BSBM), Graded Prognostic Assessment (GPA), and Modified-RPA, for patients who underwent repeat stereotactic radiosurgery (re-SRS). METHODS For this study, we used our database, which included 804 patients who underwent gamma knife re-SRS during the period 1998-2013. RESULTS There were statistically significant survival differences among patients stratified into 3 or 4 groups based on the 5 systems (P < 0.001). With RPA, SIR, BSBM, and the Modified-RPA, there were statistically significant median survival time (MST) differences between any 2 pairs within the 3/4 groups. With the GPA system, however, the MST difference between the GPA 3.5-4.0 and GPA 3.0 groups did not reach statistical significance (P = 0.48). There were large patient number discrepancies among the 3/4 groups in the RPA, SIR, BSBM, and GPA whereas patient numbers were very similar among the 3 Modified-RPA system groups. Our present results show the RPA and BSBM systems to reflect changes less well, with 86%-95% of patients remaining in the same categories between the first and second SRS procedures. However, with SIR, GPA, and the Modified-RPA, 25%-31% of patients were categorized into different subclasses, either better or worse. With the modified-RPA system, such categorical change correlated well with post-re-SRS MSTs. CONCLUSIONS Among the 5 systems, based on patient number proportions, MST separation among the 3/4 groups, and/or detailed reflection of status changes, the Modified-RPA system was shown to be most applicable to re-SRS patients.

Long-Term Side Effects of Radiosurgery for Arteriovenous Malformations

In this report, our experiences with 181 consecutive patients who underwent Gamma Knife radiosurgery (GKRS) for arteriovenous malformations (AVMs) during an approximately quarter century period from 1978 to 2002 will be summarized along with much of what we have learned from a wealth of already published data. During the mean post-GKRS follow-up period of 10.2 years (range, 5.4-30.6), 15 (8.3%) of our 181 patients experienced stereotactic radiosurgery-related, symptomatic complications. Among these 15 patients, 12 manifested complications 5 years or more after GKRS. Furthermore, in 5 of these 12, the complications were seen 10 or more years after GKRS. In the present series, the actuarial complication rates computed using the Kaplan-Meier method were 2.3% at the fifth, 8.2% at the tenth, 15.2% at the fifteenth and 31.1% at the twentieth post-GKRS year. AVM volumes, Pollock-Flickinger AVM scores, the initial presentation of bleeding and centrally located AVMs were demonstrated to be significantly associated with the risk of delayed complications after GKRS (p<0.05). There was a significant difference in complication rates between two patient groups, based on whether dose planning was performed using the older (Kula) or the modern (Gamma Plan) system (18.0 vs. 4.6%, χ2 p=0.0002). Although GKRS is undoubtedly an alternative to microsurgical resection for appropriately selected AVMs, we must weigh treatment results against complication risks which are not negligibly low.