Joseph C.T. Chen

Asymmetric Correlations of Equity Portfolios

Correlations between US stocks and the aggregate US market are much greater for downside moves, especially for extreme downside moves, than for upside moves. We develop a new statistic for measuring, comparing and testing asymmetries in conditional correlations. Conditional on the downside, correlations in the data differ from the conditional correlations implied by a normal distribution by 11.6%. We find that conditional asymmetric correlations are fundamentally different from other measures of asymmetry like skewness and co-skewness. We find that small stocks, value stocks and past loser stocks have more asymmetric movements. Controlling for size, we find that stocks with lower betas exhibit greater correlation asymmetries and we find no relationship between leverage and correlation asymmetries. Correlation asymmetries in the data reject the null of multivariate normal distributions at daily, weekly and monthly frequencies. However, several empirical models with greater flexibility, particularly regime-switching models, perform much better at capturing correlation asymmetries.

Metastatic melanoma to the brain: prognostic factors after gamma knife radiosurgery.

PURPOSE To identify important prognostic factors predictive of survival and tumor control in patients with metastatic melanoma to the brain who underwent gamma knife radiosurgery. METHODS AND MATERIALS A total of 122 consecutive patients with 332 intracranial melanoma metastases underwent gamma knife radiosurgery over a 5-year period. Of these, 39 (32%) also received whole-brain irradiation (WBI). The median tumor volume was 0.8 cm(3) (range: 0.02-30.20 cm(3)), and the median prescribed dose was 20 Gy (range: 14-24 Gy). Median follow-up was 6.8 months. Univariate and multivariate analyses of survival and freedom from progression were performed using the following parameters: status of systemic disease, intracranial tumor volume, number of lesions, tumor location, Karnofsky performance status, gender, age, and WBI. RESULTS Overall median survival was 7.0 months from time of radiosurgery and 9.1 months from the onset of brain metastasis. In multivariate analysis, improved survival was noted in patients with total intracranial tumor volume <3 cm(3) (p = 0.003) and inactive systemic disease (p = 0.0065), whereas other parameters studied were of lesser importance (tumor location, p = 0.056, and Karnofsky performance status, p = 0.086), or of no significance (number of lesions, WBI, age, and gender). Freedom from subsequent brain metastasis depended on intracranial tumor volume (p = 0.0018) and status of systemic disease (p = 0.034). CONCLUSIONS Stereotactic radiosurgery is an effective treatment modality for patients with intracranial metastatic melanoma. Tumor volume and status of systemic disease are good independent predictors of survival and freedom from tumor progression.

Radiosurgical management of benign cavernous sinus tumors: dose profiles and acute complications.

OBJECTIVE Radiosurgery has emerged as an alternative treatment modality for cranial base tumors in patients deemed not suited for primary surgical extirpation, patients with recurrent or residual tumor after open surgery, or patients who refuse surgical treatment. We review our short-term experience with radiosurgical management of cavernous sinus region tumors with the Leksell gamma knife. METHODS From August 1994 to February 1999, 69 patients with cavernous sinus lesions were treated in 72 separate treatment sessions. The tumor type distribution was 29 pituitary adenomas, 35 meningiomas, 4 schwannomas, and 1 paraganglioma. The median follow-up was 122 weeks. Lesions were stratified according to a five-level surgical grade. The grade distribution of the tumors was as follows: Grade I, 13; Grade II, 21; Grade III, 19; Grade IV, 12; Grade V, 4. Median tumor volume was 4.7 cm3. The median radiation dose was 15 Gy to the 50% isodose line. Median maximal radiation dose was 30 Gy. RESULTS Analysis of tumor characteristics and radiation dose to optic nerve and pontine structures revealed a significant correlation between distance and dose. Much lower correlation coefficients were found between tumor volume and dose. One lesion in this series had evidence of transient progression and later regression on follow-up radiographic studies. No other lesions in this series were demonstrated to have exhibited progression. Complications after radiosurgical treatment were uncommon. Two patients had cranial nerve deficits after treatment. One patient with a surgical Grade III pituitary adenoma had VIth cranial nerve palsy 25 months after radiosurgical treatment that spontaneously resolved 10 months later. A patient with a bilateral pituitary adenoma experienced bilateral VIth cranial nerve palsy 3 months after treatment that had not resolved at 35 months after treatment. Six patients with preoperative cranial nerve deficits experienced resolution or improvement of their deficits after treatment. One patient with a prolactin-secreting adenoma experienced normalization of endocrine function with return of menses. CONCLUSION Radiosurgical treatment represents an important advance in the management of cavernous sinus tumors, with low risk of neurological deficit in comparison with open surgical treatment, even in patients with high surgical grades.

Gamma knife radiosurgery for benign cavernous sinus tumors: quantitative analysis of treatment outcomes.

OBJECTIVE:We review our 8-year experience with gamma knife radiosurgery (GKRS) for the treatment of patients with benign cavernous sinus tumors and present a quantitative analysis of factors relevant to treatment outcomes. METHODS:From 1994 to 2002, a total of 139 patients with benign cavernous sinus tumors were treated in 145 sessions. Their median age was 53 years, and the median follow-up was 3.5 years. The tumors included 57 meningiomas, 76 pituitary tumors (49 nonfunctional adenomas, 15 prolactinomas, 5 adrenocorticotropic hormone-secreting tumors, 6 growth hormone-secreting tumors, and 1 plurihormone-secreting tumor), 4 schwannomas, 1 hemangioma, and 1 paraganglioma. Sekhar tumor grades were as follows: I, n = 28 (20%); II, n =42 (30%); III, n =42 (30%); IV, n =19 (14%); and V, n =8 (6%). The median tumor volume was 3.4 cm3, and the median prescribed dose was 15 Gy defined to the 50% isodose line. RESULTS:A total of 136 treated tumors (97.8%) were well controlled by GKRS, with low morbidity. For meningiomas, 29 tumors (51%) were unchanged and 26 (46%) were smaller at a median of 15.2 months. For pituitary tumors, 50 (66%) were unchanged and 25 (33%) were smaller at a median of 20.6 months. Improvement in cranial nerve (CN) function was seen in 19 (36.5%) of 52 patients with pre-GKRS deficits, and 3 patients (2.2%) developed new stable CN deficits after GKRS: 1 patient developed IVth CN palsy at 9 months, and 2 developed persistent VIth CN palsies at 43 and 45 months, respectively, that required surgical correction. Two patients developed transient VIth CN palsies at 48 months that self-resolved after another year. Endocrine function normalized for all 6 treated patients with a growth hormone-secreting tumor at a median of 18 months. One of the 5 adrenocorticotropic hormone-secreting tumors required retreatment after 17 months because of continued cortisol elevation. Thirteen (87%) of 15 prolactinoma patients had normalized prolactin levels within 2 years of the procedure; 2 patients relapsed by endocrine criteria at 18 and 22 months after GKRS. Two patients with normalized prolactin levels completed three normal pregnancies within 3 years of treatment. Six patients (4.3%) with a median tumor volume of 8 cm3 developed radiation-induced injury at a median of 36 months after GKRS. Five of these patients also underwent external beam radiotherapy and received a median dose of 52.2 Gy in 30 fractions. Quantitative analysis revealed that the radiation dose to critical structures (optic apparatus and pons) is correlated with their distance from tumor margins. Underdosed tumor volume, tumor volume, and total treated volume are correlated with treatment outcomes. CONCLUSION:GKRS is a safe and effective treatment for selected patients with benign cavernous sinus tumors and is an important adjunct for treating postoperative tumor residual and/or recurrent tumor. Continued analysis of treated patients over an extended time is needed to evaluate long-term disease control and potential late GKRS complications.

Stereotactic Radiosurgery in the Treatment of Metastatic Disease to the Brain

We review 190 consecutive patients with 434 metastatic tumors treated by gamma knife stereotactic radiosurgery, from August 1994 to February 1999. Median actuarial survival for all patients was 34 weeks. Factors correlated with significantly improved survival included controlled systemic disease and nonmelanoma histology. We found that no significant survival benefit could be discerned from adjuvant whole brain radiotherapy in this patient group. Survival was not statistically different for patients initially presenting with 1-4 metastases at initial treatment.

Radiosurgical salvage therapy for patients presenting with recurrence of metastatic disease to the brain.

OBJECTIVE Radiosurgery has emerged as an important modality in the management of metastatic disease to the brain. A number of groups have published results suggesting that high local control rates can be achieved, with improvements in overall survival that rival the results of open surgical treatment. Typically, however, whole-brain radiotherapy has been used in the salvage therapy of patients who have undergone previous craniotomy or radiosurgery. We describe our experience with radiosurgical salvage in this group of patients. METHODS From August 1994 to February 1999, 190 patients with brain metastasis were treated with gamma unit radiosurgery at our institution. A subset of 45 patients, who underwent radiosurgical salvage for new tumors in a region remote from an initially treated tumor, form the population base for this study. The usual criteria for repeat treatment were recurrence with five or fewer discrete lesions outside of the previously treated radiosurgical volume and Karnofsky Performance Scale score of at least 70. Survival and freedom from progression were measured from the time of radiosurgical treatment and were computed by the Kaplan-Meier product-limit method. Two or more curves were compared using the log-rank method. RESULTS In this subgroup of patients, a total of 176 tumors were treated. The median time from first radiosurgical procedure to first salvage was 17.4 weeks. Median survival from the second radiosurgical intervention was 28 weeks. Of the 45 study patients, 34 patients underwent a single salvage procedure, 10 patients underwent two salvage procedures, and 1 patient had three salvage procedures. The actuarial freedom from progression for treated tumors at 52 weeks was 92.4%. Patients undergoing upfront whole-brain irradiation were less likely to require salvage therapy (P = 0.008). There were 33 deaths after salvage radiosurgery during the reporting period. Central nervous system causes accounted for 13 deaths, whereas 19 deaths resulted from systemic disease. The cause of death in one patient could not be determined. No statistically significant advantage in overall survival could be demonstrated in patients treated with whole-brain irradiation. CONCLUSION Radiosurgical salvage represents a valuable means of treatment for central nervous system recurrence for patients who have undergone previous treatment for metastatic disease to the brain. Whole-brain irradiation may reduce the need for salvage therapy, but no advantage in overall survival could be demonstrated in this subgroup.

Stereoscopic Head-mounted Display Incorporated into Microsurgical Procedures: Technical Note

OBJECTIVE We have previously evaluated video technologies that have allowed for the use of electronic imaging during microneurosurgical and endoscopic procedures. This stereoscopic camera set and monitor/recording system allows for the use of stereoscopic images during surgery by the primary surgical assistant and ancillary personnel. It also allows for stereoscopic recording and playback using a wide video home system format. We describe a novel prototype of a head-mounted display (HMD) that allows the surgical team to simultaneously visualize the surgical field stereoscopically and includes picture-in-picture, voice control, and stereoscopic recording capabilities. INSTRUMENTATION A stereoscopic HMD with 640 x 480 (video graphics array) pixel resolution and the ability to display 24-bit images has been designed. This device weighs 900 g. It is interfaceable with common video display formats. RESULTS This stereoscopic HMD is being evaluated in a prospective multicenter trial of open microsurgical and endoscopic minimally invasive procedures. In our experience to date, there have been no equipment failures or complications attributable to the use of the display system. The equipment was well accepted by users who reported significant benefits in visualization. CONCLUSION The use of a stereoscopic HMD may result in improved efficiency and safety in both endoscopic and open microsurgical procedures. We have verified that the HMD is comfortable during the course of a surgical procedure, is reliable, and allows for accessibility to the operative field with an excellent field of view and three-dimensional perception. Positioning and dexterity within the operative field are also enhanced. Additional uses relate to surgical training, multimodal information display, and operative rehearsals.

Photon-based Fractionated Stereotactic Radiotherapy for Postoperative Treatment of Skull Base Chordomas

Objectives: We report our series of skull base chordoma patients who underwent surgical resection followed by high-dose fractionated stereotactic radiotherapy (FSRT) as an alternative to proton radiotherapy (RT). Methods: Between 2002 and 2009, 12 patients with skull base chordomas without prior radiation history were treated with adjuvant or salvage RT. FSRT with dynamic conformal arcs and intensity-modulated radiation therapy boost was used until 2006 when image-guided intensity-modulated FSRT was instituted. Median dose of 66.6 Gy (range, 48.6 to 68.4 Gy) was delivered in 180 cGy fractions prescribed to the 90% isodose line that covered the target volume to achieve a median isocenter dose of 74 Gy (range, 54 to 76 Gy). Results: Median follow-up was 42 months. Median time from surgery to initiation of RT was 3.6 months. Overall survival was 76.4% at 5 years, and 46.9% and 37.5% of patients were free of progression at 24 and 60 months, respectively. Six patients had disease progression after radiation with a median time to progression of 17.3 months. One patient was salvaged with radiosurgery and surgical resection, with stable disease almost 7 years since diagnosis. Two patients were salvaged with molecular targeted therapy with stable disease at 20 and 23 months. At last follow-up, 9 patients had stable or reduced disease. Conclusions: FSRT as postoperative treatment of skull base chordomas resulted in promising overall survival results comparable with the published literature of particle therapy without significant complications. Our technique for treating skull base chordomas can be considered a safe and less costly alternative to proton RT.

Contemporary methods of radiosurgery treatment with the Novalis linear accelerator system

Radiosurgery has emerged as an indispensable component of the multidisciplinary approach to neoplastic, functional, and vascular diseases of the central nervous system. In recent years, a number of newly developed integrated systems have been introduced for radiosurgery and fractionated stereotactic radiotherapy treatments. These modern systems extend the flexibility of radiosurgical treatment in allowing the use of frameless image-guided radiation delivery as well as high-precision fractionated treatments. The Novalis linear accelerator system demonstrates adequate precision and reliability for cranial and extracranial radiosurgery, including functional treatments utilizing either frame-based or frameless image-guided methods.

Prognostic factors for radiosurgery treatment of trigeminal neuralgia.

OBJECTIVETrigeminal neuralgia treatment results are thought to be highly dependent upon selection criteria. We retrospectively analyzed a series of patients to determine the likelihood of treatment success for patients treated with radiosurgery. METHODSA retrospective analysis of 82 patients treated with linear accelerator radiosurgery was undertaken with a median follow-up period of 18 months. Patients were evaluated with a standard inventory using the Barrow Neurological Institute pain scale as the primary means of outcome measurement. Patients were treated with a linear accelerator using a single isocenter plan delivered via a 4-mm collimator, typically with seven noncoplanar arcs to a peak dose of 85 or 90 Gy in primary treatments and 60 Gy in retreatments. The primary target was the cisternal component of the trigeminal nerve. Posttreatment outcomes were analyzed in light of pretreatment patient characteristics, including age, sex, anticonvulsant responsiveness, quality and pattern of pain, length of disease, number of previous procedures, and radiation dose exposure to the root entry zone. Univariate analysis and multivariate logistic regression analysis were used to determine the prognostic significance of various pretreatment variables. RESULTSGood results as defined by a Barrow Neurological Institute outcome score of IIIb or better were seen in 85.3% of patients. Excellent results as defined by a Barrow Neurological Institute outcome score of I were seen in 49% of patients. The median time to satisfactory improvement of pain was 4 weeks. Only one variable, sensitivity to anticonvulsant medication, was found to be statistically significant in both univariate (P = 0.003) and multivariate analysis (P = 0.025). All other variables analyzed failed to reach statistical significance. Complications were not common, with seven patients (8.5%) developing new-onset hypoesthesia and two patients (2%) developing dry eye symptoms. CONCLUSIONAnticonvulsant responsiveness is the single most important prognostic indicator of treatment success for patients presenting with facial pain. Other predictive factors generally failed to reach statistical significance. Linear accelerator radiosurgery for trigeminal neuralgia is a safe and effective treatment for well-selected patients, with results similar to those obtained with gamma unit radiosurgery.