Photodynamic therapy of brain tumours: evaluation of porphyrin uptake versus clinical outcome.

The objective of this study was to investigate whether the level of the photosensitizer haematoporphyrin derivative (HpD) uptake measured in tissue samples taken from brain tumour patients was associated with survival post-treatment with photodynamic therapy (PDT). The mean HpD uptake in tumour tissue was significantly higher in glioblastoma multiforme than anaplastic astrocytoma. Recurrent tumours had a higher mean uptake compared to primary tumours, which was evident in all grades of tumour. Among patients with GBM, there was a significant association between greater HpD uptake and survival (HR = 0.26 [0.12, 0.59], p = 0.001). There was also some evidence of a weak association between greater HpD uptake and survival among patients with AA, although the result was inconclusive (HR = 0.73 [0.32, 1.71], p = 0.472).     

Photodynamic therapy of malignant brain tumours

Fifty patients with malignant supratentorial tumours were treated with intra-operative photodynamic therapy (PDT); in 33 cases the tumour was recurrent. In 45 patients the tumour was a cerebral glioma and in 5 cases a solitary cerebral metastasis. All patients received a porphyrin photosensitizer 18-24 hours pre-operatively. Photoillumination was carried out at 630 nm to a tumour cavity created by radical tumour resection and/or tumour cyst drainage. The light energy density ranged from 8 to 175 J/cm2. In 8 patients additional interstitial light was administered. The operative mortality was 4%. Follow up has ranged from 1 to 30 months. The median survival for the 45 primary malignant tumours was 8.6 months with a 1 and 2 year actuarial survival rate of 32% and 18%, respectively. In 12 patients a complete or near complete CT scan response was identified post PDT. These patients tended to have a tumour geometry (eg. cystic) that allowed complete or near complete light distribution to the tumour. The median survival for this group was 17.1 months with a 1 and 2 year actuarial survival of 62% and 38%, respectively. In the 33 cases who did not have a complete response the median survival was 6.5 months with a 1 and 2 year actuarial survival of 22% and 11%, respectively. Photodynamic therapy of malignant brain tumours can be carried out with acceptable risk. Good responses appear to be related to adequate light delivery to the tumour.     

Promising survival in patients with high-grade gliomas following therapy with a novel boronated porphyrin.

Current treatment strategies for high-grade gliomas are inadequate with high rates of disease recurrence and poor overall survival. Photodynamic therapy has been extensively investigated for a variety of tumours including high-grade gliomas. We have previously described a novel boronated porphyrin (BOPP) which has highly selective uptake by tumour cells. A Phase I study documented that BOPP at a dose of 4mg/kg was well tolerated. We report here the survival data arising from this Phase I study. Overall 28 patients (pts) with high-grade gliomas were treated with BOPP photodynamic therapy with a median overall survival (OS) of 14 months (2-48+ months). Sixteen pts had glioblastoma multiforme (GBM) with a median OS of 8 months (2-38+). Of 9 pts with recurrent GBM, the median OS was 11 months (3-38+), with 56% surviving at least 12 months. In conclusion, BOPP photodynamic therapy was associated with encouraging survival particularly in the recurrent GBM setting.     

Carbonic anhydrase IX is a prognostic biomarker in glioblastoma multiforme

The identification of prognostic factors in patients with glioblastoma multiforme (GBM) represents an area of increasing interest. Carbonic anhydrase IX (CA‐IX), a hypoxia marker, correlates with tumor progression in a variety of human cancers. However, the role of CA‐IX in GBM remains largely unknown. In the present study, we evaluated the prognostic role of CA‐IX in GBM patients. In total, 66 consecutive patients with GBM who received concomitant chemoradiotherapy and adjuvant chemotherapy with temozolomide were retrospectively reviewed, and all patients received temozolomide chemotherapy for at least 3 months. Kaplan–Meier curves and log‐rank tests were used for analysis of progression‐free survival (PFS) and overall survival (OS), and a multivariate Cox proportional hazard model was employed to identify factors with an independent effect on survival. The median OS was longer in patients with low levels of CA‐IX expression (18 months) compared to patients overexpressing CA‐IX (9 months) (P = 0.004). There was not a statistically significant difference in median PFS (3.5 vs. 8 months, P = 0.054) between patients with high or low levels of CA‐IX expression. In multivariate analysis, the variables that were identified as significant prognostic factors for OS were preoperative Karnofsky performance scale score (KPS) (hazard ratio (HR), 3.703; P = 0.001), CA‐IX overexpression (HR, 1.967; P = 0.019), and incomplete adjuvant temozolomide treatment (HR, 2.241; P = 0.003) and gross‐total resection (HR, 1.956; P = 0.034). Our findings indicated that CA‐IX may be a potential prognostic biomarker in the treatment of GBM.     

Extent of resection affects prognosis for patients with glioblastoma in non-eloquent regions

Glioblastoma (GBM) is a malignant cerebral neoplasm carrying poor prognosis. The importance of extent of resection (EoR) in GBM patient outcomes has been argued in the literature. Previous studies included tumors in eloquent regions of the brain. This confounds the role of EoR by including patients with intrinsically worse outcomes but will be over-represented in the reduced EoR category. In a homogenous group of patients in whom GTR was considered achievable, we investigated the effect of increasing EoR on survival. A retrospective review of 51 patients was undertaken. Quantitative, volumetric analysis of pre-operative and post-operative magnetic resonance image was compared with corresponding clinical details. The primary outcome measured was post-operative overall survival. Median overall survival was 18.3 months for GTR patients compared to 11.6 months for non-GTR (p = 0.025). Median pre-operative contrast-enhancing tumor volume for GTR patients was 54.7 cm³ and 24.9 cm³ for non-GTR. Post-operative median residual tumor volume was 1.1 cm³ in the non-GTR cohort. In multivariate analyses, GTR (HR [95% CI] = 0.973 [0.954–0.994], p = 0.00559) and increasing EoR (HR [95% CI] = 0.964 [0.944–0.985], p = 0.000665) remained predictors of survival. Centile dichotomization of EoR revealed 74% (HR [95% CI] = 0.351 [0.128–0.958], p = 0.0409) as the lowest threshold conferring statistically significant survival benefit. Where technically feasible, both GTR and EoR remained as independent prognostic factors for survival. GTR remains the gold standard for surgical treatment of GBM in patients, 74% being the minimum EoR required to confer survival benefit.     

光敏素Photofrin光动力学治疗恶性脑胶质瘤

目的 探讨术中及术后对18例恶性脑胶质瘤患实行Photofrin(光敏素）光动力学疗法的疗效。方法 于术前24h静滴血Photofrin，术中大体全切或尽可能多的切除肿瘤并且在残留瘤腔内激光照射。术后24h通过光纤追加照射一次。结果 随访3-14月。16例病人仍处于影象学无瘤生存状态，1例明确复发，1例可疑复发。无术后持久神经功能缺失和死亡。结论 光敏素Photofrin激光光动力学治疗脑胶质瘤能明显延长患的无瘤生存期和改善生存质量，是一种治疗恶性脑质瘤的有效方法。     

Continuous Low-Dose Temozolomide Chemotherapy and Microvessel Density in Recurrent Glioblastoma

Objective

The purpose of this study was to evaluate the clinical efficacy of continuous low-dose temozolomide (TMZ) chemotherapy for recurrent and TMZ-refractory glioblastoma multiforme (GBM) and to study the relationship between its efficacy and microvessel density within the tumor.

Methods

Thirty patients who had recurrent GBM following Stupp''s regimen received TMZ daily at 50 mg/m²/day until tumor progression between 2007 and 2013. The median duration of continuous low-dose TMZ administration was 8 weeks (range, 2-64).

Results

The median progression-free survival (PFS) of continuous low-dose TMZ therapy was 2 months (range, 0.5-16). At 6 months, PFS was 20%. The median overall survival (OS) from the start of this therapy to death was 6 months (95% CI : 5.1-6.9). Microvessel density of recurrent tumor tissues obtained by reoperation of 17 patients was 22.7±24.1/mm² (mean±standard deviation), and this was lower than that of the initial tumor (61.4±32.7/mm²) (p-value=0.001). It suggests that standard TMZ-chemoradiotherapy reduces the microvessel density within GBM and that recurrences develop in tumor cells with low metabolic burden. The efficacy of continuous low-dose TMZ could not be expected in recurrent GBM cells in poor angiogenic environments.

Conclusion

The efficacy of continuous low-dose TMZ chemotherapy is marginal. This study suggests the need to develop further treatment strategies for recurrent and TMZ-refractory GBM.     

Clinical management and survival outcomes of gliosarcomas in the era of multimodality therapy

Gliosarcoma (GSM) is a rare primary malignant brain tumour accounting for less than 0.5% of all intracranial tumours. It has a biphasic histological composition, demonstrating both gliomatous and sarcomatous elements. In clinical practice GSM are generally managed similarly to glioblastoma multiforme (GBM). However, unique features including its clinical propensity for extra-cranial metastasis, distinct radiological features and possible worse prognosis than GBM suggest that GSM may be a distinct clinico-pathological entity. Hence we reviewed patterns of care and outcomes for a series of Australian patients diagnosed with GSM in the era of combined chemo-radiotherapy. Patients were identified by searching the Australian Genomics and Clinical Outcomes of Glioma (AGOG) database and the Western Australian Interhospital Neurosurgical database. Nineteen patients with GSM were identified. Of these, 15 patients were diagnosed with primary GSM and four patients developed secondary GSM after radiation therapy for primary GBM. For comparative purposes, 408 primary GBM patients were identified from the AGOG database during the same study period. The overall median survival for all primary GSM patients was 9.7 months. In comparison the overall median survival for GBM patients recruited to the AGOG database over the same period was 12.2 months. The median survival for secondary GSM patients from the time of diagnosis was 5 months. Primary and secondary GSM pose a great clinical challenge due to their rarity. Our study adds further evidence to support GSM as a unique clinical entity with a likely worse prognosis than GBM.     

Photodynamic therapy of cerebral glioma--a review Part I--a biological basis.

Photodynamic therapy (PDT) has been investigated extensively in the laboratory for decades, and for over 25 years in the clinical environment, establishing it as a useful adjuvant to standard treatments for many cancers. A combination of both photochemical and photobiological processes occur that lead to the eventual selective destruction of the tumour cells. It is a potentially valuable adjuvant therapy that can be used in conjunction with other conventional therapies for the treatment of cerebral glioma. PDT has undergone extensive laboratory studies and clinical trials with a variety of photosensitizers (PS) and tumour models of cerebral glioma. Many environmental and genetically based factors influence the outcome of the PDT response. The biological basis of PDT is discussed with reference to laboratory and preclinical studies.     

A phase I trial of surgical resection with Gliadel Wafer placement followed by vaccination with dendritic cells pulsed with tumor lysate for patients with malignant glioma

High grade gliomas are associated with poor prognosis and high mortality. Conventional treatments and management of high grade gliomas have shown little improvement in 5-year overall survival. This phase I trial evaluated the safety, immunogenicity, and potential synergy of surgical resection with Gliadel Wafer implantation, followed by autologous tumor lysate-pulsed dendritic cell (DC) vaccine in patients with malignant glioma. Primary end points of this study were safety and surrogate markers of immunogenicity, overall survival, and progression free survival. Following surgical resection, Gliadel Wafers were placed along the resection cavity. Patients subsequently received intradermal injections of autologous tumor lysate-pulsed DC vaccines 3 times at 2 week intervals. Treatment response was evaluated clinically and through MRI at regular intervals. Twenty-eight patients received Gliadel Wafers and DC vaccination: 11 newly diagnosed (8 glioblastoma [GBM], 2 anaplastic astrocytoma [AA], and 1 anaplastic oligodendroglioma [AO]) and 17 recurrent (15 GBMs, 1 AA, and 1 AO) high grade gliomas. Immunogenicity data was collected for 20 of the 28 patients. Five of 20 patients showed elevated IFN-γ responses following vaccination. Median progression-free survival and overall survival for all GBM patients in the trial from the start of vaccination were 3.6 months and 16.9 months respectively. Comparisons between vaccine responders and non-vaccine responders were not statistically significant. Adjuvant autologous dendritic cells pulsed with tumor-lysate following resection and Gliadel Wafer placement is safe, elicits modest immunogenicity and shows similar clinical outcomes in patients who had DC vaccination in previous studies.     

Efficacy of temozolomide as adjuvant chemotherapy after postsurgical radiotherapy alone for glioblastomas

Objectives

The aim of this study was to assess the efficacy of adjuvant TMZ chemotherapy for newly diagnosed GBM patients who were treated with surgery followed by radiotherapy alone.

Material and methods

Between January 2003 and April 2005, 59 consecutive GBM patients underwent radiation therapy after surgical resection and subsequently received TMZ chemotherapy. For the comparative analysis, we selected 60 clinically matched GBM patients who underwent radiotherapy followed by nitrosourea-based chemotherapy (NUBC), at the same institution between June 1995 and April 2005. The study cohort was divided into two groups, those with adjuvant TMZ treatment and with NUBC.

Results

59 patients with adjuvant TMZ treatment were assigned to the treatment group and 60 patients with NUBC to the control group. The median overall survival for the treatment group was 18.2 months (95% CI, 11.7–24.7 months), compared with the survival of 14.5 months (95% CI, 11.2–17.7 months) for the control group (p = 0.019). The progression-free survival for the treatment group was 5.6 months (95% CI, 4.4–6.7 months), while the control group showed progression-free survival of 3.3 months (95% CT, 3.2–6.0 months) (p = 0.030). Uni- and multivariate analysis revealed that extent of surgical resection, age ≥55 years and postoperative KPS were significantly associated with survival.

Conclusion

Adjuvant TMZ chemotherapy provided a clinically relevant benefit of survival, as compared with NUBC. Thus, we suggest that adjuvant TMZ chemotherapy may be effective even for patients who did not receive concomitant chemoradiotherapy for GBM.     

Photodynamic therapy: cavitary photoillumination of malignant cerebral tumours using a laser coupled inflatable balloon

Interest in photodynamic therapy of malignant brain tumours has been growing in recent years as intra-operative laser applications become more available. We have developed an inflatable balloon which can be coupled to an argon dye pump laser in order to deliver light to a brain tumour cavity created by the subtotal resection of tumour. Eight patients with primary malignant brain tumors have been treated with photodynamic therapy (PDT) using this device. The 8 patients tolerated the treatment well; morbidity attributable to the PDT was acceptably low.     

Hypo-fractionated IMRT for patients with newly diagnosed glioblastoma multiforme: A 6 year single institutional experience

Objectives

Glioblastoma (GBM) is the most common malignant primary brain tumour in adults. Surgery and radiotherapy constitute the cornerstones for the therapeutic management of GBM. The standard treatment today is maximal surgical resection followed by concomitant chemo-radiation therapy followed by adjuvant TMZ according to Stupp protocol. Despite the progress in neurosurgery, radiotherapy and oncology, the prognosis still results poor.In order to reduce the long time of standard treatment, maintaining or improving the clinical results, in our institute we have investigated the effects of hypo-fractionated radiation therapy for patients with GBM.

Patients and methods

Sixty-seven patients affected by GBM who had previously undergone surgical resection (total, subtotal or biopsy) were enrolled between October 2005 and December 2011 in a single institutional study of hypo-fractionated intensity modulated radiation therapy (IMRT) followed or not by adjuvant chemotherapy with TMZ (6–12 cycles). The most important eligibility criteria were: biopsy-proven GBM, KPS ≥ 60, age ≥ 18 years, no previous brain irradiation, informed consensus. Hypo-fractionated IMRT was delivered to a total dose of 25 Gy in 5 fractions prescribed to 70% isodose. Response to treatment, OS, PFS, toxicity and patterns of recurrence were evaluated, and sex, age, type of surgery, Karnofsky performance status, Recursive Partitioning Analysis (RPA) classification, time between surgery and initiation of radiotherapy were evaluated as potential prognostic factors for survival.

Results

All patients have completed the treatment protocol. Median age was 64.5 years (range 41–82 years) with 31 females (46%) and 36 males (54%). Median KPS at time of treatment was 80. The surgery was gross total in 38 patients and subtotal in 14 patients; 15 patients underwent only biopsy.No grade 3–4 acute or late neurotoxicity was observed. With median follow-up of 14.9 months, the median OS and PFS were 13.4 and 7.9 months, respectively.

Conclusions

The hypo-fractionated radiation therapy can be used for patients with GBM, resulting in favourable overall survival, low rates of toxicity and satisfying QoL. Future investigations are needed to determine the optimal fractionation for GBM.     

Geminin表达水平与脑胶质瘤病人预后的关系

目的 探讨Geminin表达水平与脑胶质瘤病人预后的关系。方法 收集2014年1月~2017年1月手术切除并经术后病理证实的脑胶质瘤标本90例,以及因颅脑损伤内减压术切除的非肿瘤脑组织50,采用PCR检测Geminin mRNA表达水平。90例脑胶质瘤根据Geminin mRNA表达水平中位数分为高表达组和低表达组。90例胶质瘤病人随访时间截止2020年1月1日或者病人死亡,随访32~55个月,中位随访时间41个月。结果 随访期间,死亡80例,存活10例。脑胶质瘤组织Geminin mRNA表达水平明显高于非肿瘤脑组织（P<0.05）。多因素Cox比例回归风险模型分析结果显示,Geminin mRNA过表达是脑胶质瘤病人死亡的独立危险因素（HR=1.874;95% CI 1.323-2.295;P=0.043）。Kaplan-Meier生存曲线显示,低表达组中位生存期较高表达组明显延长（P<0.05）。结论 脑胶质瘤Geminin呈高表达,是病人生存预后不良的独立危险因素。     

Epilepsy and primary cerebral tumours

Although the association of epilepsy with cerebral tumours is well recognized, the reported incidence of seizures and relationship to tumour pathology varies significantly. This study assessed retrospectively the incidence of seizures, relationship to tumour pathology, natural history of epilepsy and prognostic significance of presentation with a seizure in 120 consecutive adults with histologically proven primary cerebral hemisphere tumours including meningiomas. 52% had a seizure and most were at presentation. Seizures were more common with anaplastic astrocytoma (AA) (18 23 ) than glioblastoma multiforme (21 56 ) (p = 0.001) and seizure occurrence was associated with cortical invasion. 52% of meningioma patients had a seizure. Seizures recurred in 34%, more frequently with glioma (19 of 46) than meningioma (1 of 15) (p < 0.05). Patients with AA presenting with a seizure had a longer survival (28 months) than patients without seizure (8 months) (p = 0.05 one sided). In conclusion, seizures are a common complication of cerebral tumours, usually at presentation and correlate with tumour pathology. A seizure at presentation in AA correlates with longer survival.     

An investigation of boron neutron capture therapy for recurrent glioblastoma multiforme

Objectives –  To explore the use of boron neutron capture therapy (BNCT) for patients with glioblastoma multiforme (GBM), recurring after surgery and conventional radiotherapy (photon radiotherapy).
Materials and methods –  Boron uptake in recurrent GBM was measured for four patients. Twelve patients were subsequently treated by BNCT with boronophenylalanine-fructose (900 mg/kg body weight), administered by intravenous infusion for 6 h.
Results –  Median survival time from initial diagnosis was 22.2 months. Comparison with other BNCT studies indicates a clinical advantage of the prolonged infusion. BNCT was well tolerated and quality of life remained stable until tumor progression for all 12 patients. No correlation was found between survival times and minimum tumor dose and number of radiation fields.
Conclusions –  Boron neutron capture therapy, with the prolonged procedure for infusion, is at least as effective as other radiation therapies for recurrent GBM and is delivered in one treatment session, with low radiation dose to the healthy brain. Survival from diagnosis compares favorably with that obtained with conventional radiotherapy plus concomitant and adjuvant temozolomide (TMZ) and survival from recurrence compares favorably with that obtained with TMZ at first relapse. The results of the present investigation are encouraging and should be confirmed in a randomized trial.     

Hypericin uptake: a prognostic marker for survival in high-grade glioma.

Currently adjuvant chemotherapy for glioblastoma patients can prolong survival time relative to patients who receive only surgery and radiotherapy. Despite these improvements and experimental and clinical efforts the prognosis for glioblastoma patients remains poor. At present, interest is focused on individual prognostic factors influencing patient responses to therapy. Photodynamic therapy may be a promising therapeutic option in the treatment of glioblastoma. In this investigation we examined whether uptake of hypericin (HY), a fluorescent photosensitization agent, by ex vivo glioblastoma cell lines correlates with prognosis of the individual from which the cell lines were derived. Twelve primary human glioma cell cultures were incubated with 20 micromol HY. Fluorescence intensity was measured using fluorescence microscopy. Three patients suffered from an anaplastic astrocytoma, WHO grade III, nine had a glioblastoma, WHO grade IV. In 6/12 patients complete tumour resection was possible. The mean survival time of the six patients in whom complete tumour resection was performed was 26 months, compared with 5 months for those who underwent incomplete resection. Eleven patients received radiation therapy. The five patients who received chemotherapy survived for a mean duration of 26 months, compared with the seven patients who survived for a mean duration of 5 months without chemotherapy. Statistical analysis using a parametric survival model based on the Weibull distribution showed that fluorescence intensity was the variable with the lowest p-value associated with survival (p=0.0225). An increase of 553 arbitrary units of fluorescence intensity is predicted to double survival time. Uptake of HY, a lipophilic molecule, is assumed to be related to low-density lipoprotein (LDL) uptake and metabolism. Cell proliferation is associated with a high turnover of cholesterol and membrane growth, which is related to cholesterol uptake by LDL. In summary, HY uptake by ex vivo glioblastoma cell cultures seems to be positively associated with survival of patients with malignant glioma.     

Combined modality treatment of newly diagnosed glioblastoma multiforme in a regional neurosurgical centre

The “local experience” of the Stupp protocol was examined in the treatment of patients with newly diagnosed glioblastoma multiforme (GBM) with particular emphasis given to the extent of surgical resection and its effect on survival. Thirty-one patients with newly diagnosed GBM who underwent combined modality treatment according to the Stupp protocol were assessed retrospectively. Variables assessed were the extent of surgery, size and site of the tumour, age and performance status. Primary end points were overall survival (OS) and progression-free survival (PFS). Median OS was 33 months for macroscopic tumour resection (9 patients; 29%), 15 months for debulking (15; 48%) and 9 months for biopsy (7; 22%). Macroscopic tumour resection resulted in significantly improved OS and PFS compared to the two less radical surgical options (p < 0.001). Patients with GBM undergoing maximal resection of the tumour followed by adjuvant radiotherapy and chemotherapy have an improved survival compared to patients undergoing either subtotal resection or biopsy alone. This statistically significant survival benefit was achieved in a regional neurosurgical centre with minimal additional toxicity.     

The survival impact of significant delays between surgery and radiochemotherapy in glioblastoma patients: A retrospective analysis from a large tertiary center

The optimal timing of adjuvant radiochemotherapy (RCT) in glioblastoma (GBM) patients remains unknown and the paradigm of ‘the sooner, the better’ has been challenged by many recent publications. In this study, we present unique data on the outcomes of patients with significant treatment delays. The study group consisted of 346 GBM patients (median age 56.8 years) who received surgical treatment (total or subtotal resection) and then underwent adjuvant concurrent RCT at one institution. The main endpoint was overall survival (OS). The Univariate and multivariate Cox Proportional-Hazard Model, log-rank test, and Kaplan-Meier method were used for the analysis. The median OS was 18.7 months and the 5-year overall survival was 8.5%. The median time interval from surgery to RCT was 9.8 weeks. The Cox regression showed that the time interval had no statistically significant impact on OS both in uni- and multivariate analysis. The explorative analysis suggested a positive trend for improved survival for patients in the 1st quartile of the time interval, especially for patients with residual disease or local recurrence prior to RCT, However, considering the 6.9 weeks median interval in the 1st quartile, this subgroup should still be regarded as 'moderate delay' compared with other literature data. The results indicate that the time interval is not a clear prognostic factor in the treatment of GBM. Prospective trials are highly warranted, as data suggest that moderate delays in the initiation of adjuvant treatment might be associated with survival benefit.     

Symptomatic intracranial atherosclerosis: outcome of patients who fail antithrombotic therapy

OBJECTIVE: To determine the prognosis of patients with symptomatic intracranial atherosclerosis who fail antithrombotic therapy. BACKGROUND: The outcome of patients with symptomatic intracranial atherosclerosis who fail antithrombotic therapy is unknown. These patients may represent the target group for investigation of more aggressive therapies such as intracranial angioplasty. METHODS: The authors performed a chart review and telephone interview of patients with symptomatic intracranial atherosclerosis identified in the Stanford Stroke Center clinical database. A Cox regression model was created to identify factors predictive of failure of antithrombotic therapy. The authors generated Kaplan-Meier survival curves to determine the timing of recurrent TIA, stroke, or death after failure of antithrombotic therapy. RESULTS: Fifty-two patients had symptomatic intracranial atherosclerosis and fulfilled entry criteria. Twenty-nine of the 52 patients (55.8%) had cerebral ischemic events while receiving an antithrombotic agent (antiplatelet agents [55%], warfarin [31%], or heparin [14%]). In a Cox regression model, older age was an independent predictor of failure of antithrombotic therapy, and warfarin use was associated with a decrease in risk. Recurrent TIA (n = 7), nonfatal/fatal stroke (n = 6/1), or death (n = 1) occurred in 15 of 29 (51.7%) of the patients who failed antithrombotic therapy. The median time to recurrent TIA, stroke, or death was 36 days (95% CI 13 to 59). CONCLUSIONS: Patients with symptomatic intracranial atherosclerosis who fail antithrombotic therapy have extremely high rates of recurrent TIA/stroke or death. Recurrent ischemic events typically occur within a few months after failure of standard medical therapy. The high recurrence risk observed warrants testing of alternative treatment strategies such as intracranial angioplasty.