A randomized phase 2 study of MK-2206 versus everolimus in refractory renal cell carcinoma.

Background: Activation of the phosphoinisitide-3 kinase (PI3K) pathway through mutation and constitutive upregulation has been described in renal cell carcinoma (RCC), making it an attractive target for therapeutic intervention. We performed a randomized phase II study in vascular endothelial growth factor (VEGF) therapy refractory patients to determine whether MK-2206, an allosteric inhibitor of AKT, was more efficacious than the mammalian target of rapamycin inhibitor everolimus. Patients and methods: A total of 43 patients were randomized in a 2:1 distribution, with 29 patients assigned to the MK-2206 arm and 14 to the everolimus arm. Progression-free survival (PFS) was the primary endpoint. Results: The trial was closed at the first futility analysis with an observed PFS of 3.68 months in the MK-2206 arm and 5.98 months in the everolimus arm. Dichotomous response rate profiles were seen in the MK-2206 arm with one complete response and three partial responses in the MK-2206 arm versus none in the everolimus arm. On the other hand, progressive disease was best response in 44.8% of MK2206 versus 14.3% of everolimus-treated patients. MK-2206 induced significantly more rash and pruritis than everolimus, and dose reduction occurred in 37.9% of MK-2206 versus 21.4% of everolimus-treated patients. Genomic analysis revealed that 57.1% of the patients in the PD group had either deleterious TP53 mutations or ATM mutations or deletions. In contrast, none of the patients in the non-PD group had TP53 or ATM defects. No predictive marker for response was observed in this small dataset. Conclusions: Dichotomous outcomes are observed when VEGF therapy refractory patients are treated with MK-2206, and MK-2206 does not demonstrate superiority to everolimus. Additionally, mutations in DNA repair genes are associated with early disease progression, indicating that dysregulation of DNA repair is associated with a more aggressive tumor phenotype in RCC.

Open-label randomized clinical trial of standard neoadjuvant chemotherapy with paclitaxel followed by FEC versus the combination of paclitaxel and everolimus followed by FEC in women with triple receptor-negative breast cancer†.

BACKGROUND: Everolimus synergistically enhances taxane-induced cytotoxicity in breast cancer cells in vitro and in vivo in addition to demonstrating a direct antiproliferative activity. We aim to determine pharmacodynamics changes and response of adding everolimus to standard neoadjuvant chemotherapy in triple-negative breast cancer (TNBC). PATIENTS AND METHODS: Phase II study in patients with primary TNBC randomized to T-FEC (paclitaxel 80 mg/m(2) i.v. weekly for 12 weeks, followed by 5-fluorouracil 500 mg/m(2), epirubicin 100 mg/m(2), and cyclophosphamide 500 mg/m(2) every 3 weeks for four cycles) versus TR-FEC (paclitaxel 80 mg/m(2) i.v. and everolimus 30 mg PO weekly for 12 weeks, followed by FEC). Tumor samples were collected to assess molecular changes in the PI3K/AKT/mTOR pathway, at baseline, 48 h, 12 weeks, and at surgery by reverse phase protein arrays (RPPA). Clinical end points included 12-week clinical response rate (12-week RR), pathological complete response (pCR), and toxicity. RESULTS: Sixty-two patients were registered, and 50 were randomized, 27 received T-FEC, and 23 received TR-FEC. Median age was 48 (range 31-75). There was downregulation of the mTOR pathway at 48 h in the TR-FEC arm. Twelve-week RR by ultrasound were 29.6% versus 47.8%, (P = 0.075), and pCR were 25.9% versus 30.4% (P = 0.76) for T-FEC and TR-FEC, respectively. mTOR downregulation at 48 h did not correlate with 12-week RR in the TR-FEC group (P = 0.58). Main NCI grade 3/4 toxicities included anemia, neutropenia, rash/desquamation, and vomiting in both arms. There was one case of grade 3 pneumonitis in the TR-FEC arm. No grade 3/4 stomatitis occurred. CONCLUSION: The addition of everolimus to paclitaxel was well tolerated. Everolimus downregulated mTOR signaling but downregulation of mTOR at 48 h did not correlate with 12-week RR in the TR-FEC group. CLINICAL TRIAL NUMBER: NCT00499603.

Functional proteomics characterization of residual triple-negative breast cancer after standard neoadjuvant chemotherapy.

BACKGROUND: In this study, we used functional proteomics to determine the molecular characteristics of residual triple receptor-negative breast cancer (TNBC) patients after neoadjuvant systemic chemotherapy (NCT) and their relationship with patient outcomes in order to identify potential targets for therapy. PATIENTS AND METHODS: Protein was extracted from 54 residual TNBCs, and 76 proteins related to breast cancer signaling were measured by reverse phase protein arrays (RPPAs). Univariable and multivariable Cox proportional hazard models were fitted for each protein. Survival outcomes were estimated by the Kaplan-Meier product limit method. Training and cross validation were carried out. The coefficients estimated from the multivariable Cox model were used to calculate a risk score (RS) for each sample. RESULTS: Multivariable analysis using the top 25 proteins from univariable analysis at a false discovery rate (FDR) of 0.3 showed that AKT, IGFBP2, LKB1, S6 and Stathmin were predictors of recurrence-free survival (RFS). The cross-validation model was reproducible. The RS model calculated based on the multivariable analysis was -1.1086 × AKT + 0.2501 × IGFBP2 - 0.6745 × LKB1+1.0692 × S6 + 1.4086 × stathmin with a corresponding area under the curve, AUC = 0.856. The RS was an independent predictor of RFS (HR = 3.28, 95%CI = 2.07-5.20, P < 0.001). CONCLUSIONS: We found a five-protein model that independently predicted RFS risk in patients with residual TNBC disease. The PI3 K pathway may represent potential therapeutic targets in this resistant disease.

Functional proteomics characterization of residual breast cancer after neoadjuvant systemic chemotherapy.

BACKGROUND: The purpose of this study was to determine the functional proteomic characteristics of residual breast cancer and hormone receptor (HR)-positive breast cancer after neoadjuvant systemic chemotherapy, and their relationship with patient outcomes. METHODS: Reverse phase protein arrays of 76 proteins were carried out. A boosting approach in conjunction with a Cox proportional hazard model defined relapse predictors. A risk score (RS) was calculated with the sum of the coefficients from the final model. Survival outcomes and associations of the RS with relapse were estimated. An independent test set was used to validate the results. RESULTS: Test (n = 99) and validation sets (n = 79) were comparable. CoxBoost revealed a three-biomarker (CHK1pS345, Caveolin1, and RAB25) and a two-biomarker (CD31 and Cyclin E1) model that correlated with recurrence-free survival (RFS) in all residual breast cancers and in HR-positive disease, respectively. Unsupervised clustering split patients into high- and low risk of relapse groups with different 3-year RFS (P ≤ 0.001 both). RS was a substantial predictor of RFS (P = 0.0008 and 0.0083) after adjustment for other substantial characteristics. Similar results were found in validation sets. CONCLUSIONS: We found models that independently predicted RFS in all residual breast cancer and in residual HR-positive disease that may represent potential targets of therapy in this resistant disease.

Cytokines and angiogenic factors in patients with metastatic renal cell carcinoma treated with interferon-alpha: association of pretreatment serum levels with survival.

BACKGROUND: To correlate serum cytokine and angiogenic factor (CAF) levels with overall survival (OS) in metastatic renal cell carcinoma (mRCC) treated with interferon-alpha (IFN-alpha). PATIENTS AND METHODS: Serum CAF levels were measured in 103 patients treated on a randomized trial with IFN-alpha 0.5 million units (MU) twice daily or 5 MU daily. Concentrations of 17 analytes were determined by multiplex bead immunoassays [vascular endothelial growth factor A (VEGF(A)) and several cytokines] or enzyme-linked immunosorbent assay (basic fibroblast growth factor). We used proportional hazards models to evaluate the effect of CAF levels and clinical factors on OS. RESULTS: Pretreatment serum interleukin (IL) 5, IL-12 p40, VEGF(A), and IL-6 levels and Memorial Sloan-Kettering Cancer Center risk grouping independently correlated with OS, with hazard ratios of 2.33, 2.00, 2.07, 1.82, and 0.39, respectively (concordance index = 0.69 for the combined model versus 0.60 for the CAF model versus 0.52 for the clinical model). Based on an index derived from these five risk factors (RFs), patients with 0-2 RF had a median OS time of 32 months versus 9 months for patients with 3-5 RF (P < 0.0001). CONCLUSIONS: Serum CAF profiling contributes to prognostic evaluation in mRCC and helps to identify a subset of patients with 20% 5-year OS.

Levels of soluble HLA-I and beta2M in patients with acute myeloid leukemia and advanced myelodysplastic syndrome: association with clinical behavior and outcome of induction therapy.

beta-2 Microglobulin (beta2M), a subunit of human leukocyte antigen-class I (HLA-I), is well established as a marker of prognosis in various solid tumors and hematologic malignancies. The prognostic role of intact free-circulating HLA-I (sHLA-I) is less well understood. We compared the clinical relevance of plasma levels of sHLA-I and beta2M in patients with acute myeloid leukemia (AML; n=209) or advanced myelodysplastic syndrome (MDS; n=98). sHLA-1 and beta2M levels were significantly higher in AML and MDS patients than in control subjects, but did not differ significantly between the two disease groups. In AML patients, multivariate analysis showed both sHLA-1 and beta2-M to be highly predictive of complete remission (CR), survival and duration of complete response (CRD). In MDS, the predictive value of the two markers differed substantially from one another: beta2M was associated with survival, CR and CRD, whereas sHLA-I was not. These findings not only establish the role of sHLA-I as a tumor marker in AML but also support that MDS is clinically and biologically distinct from AML. sHLA-I has been reported to be an immunomodulator inhibiting the cytotoxic effects of T-lymphocytes, which may offset its predictive value for disease aggressiveness in patients with MDS.

ATM function and its relationship with ATM gene mutations in chronic lymphocytic leukemia with the recurrent deletion (11q22.3-23.2).

Approximately 10-20% of chronic lymphocytic leukemia (CLL) patients exhibit del(11q22-23) before treatment, this cohort increases to over 40% upon progression following chemoimmunotherapy. The coding sequence of the DNA damage response gene, ataxia-telangiectasia-mutated (ATM), is contained in this deletion. The residual ATM allele is frequently mutated, suggesting a relationship between gene function and clinical response. To investigate this possibility, we sought to develop and validate an assay for the function of ATM protein in these patients. SMC1 (structural maintenance of chromosomes 1) and KAP1 (KRAB-associated protein 1) were found to be unique substrates of ATM kinase by immunoblot detection following ionizing radiation. Using a pool of eight fluorescence in situ hybridization-negative CLL samples as a standard, the phosphorylation of SMC1 and KAP1 from 46 del (11q22-23) samples was analyzed using normal mixture model-based clustering. This identified 13 samples (28%) that were deficient in ATM function. Targeted sequencing of the ATM gene of these samples, with reference to genomic DNA, revealed 12 somatic mutations and 15 germline mutations in these samples. No strong correlation was observed between ATM mutation and function. Therefore, mutation status may not be taken as an indicator of ATM function. Rather, a direct assay of the kinase activity should be used in the development of therapies.

Outcomes for patients with chronic lymphocytic leukemia and acute leukemia or myelodysplastic syndrome.

Acute leukemia (AL) and myelodysplastic syndrome (MDS) are uncommon in chronic lymphocytic leukemia (CLL). We retrospectively identified 95 patients with CLL, also diagnosed with AL (n=38) or MDS (n=57), either concurrently (n=5) or subsequent (n=90) to CLL diagnosis and report their outcomes. Median number of CLL treatments prior to AL and MDS was 2 (0-9) and 1 (0-8), respectively; the most common regimen was purine analog combined with alkylating agent±CD20 monoclonal antibody. Twelve cases had no prior CLL treatment. Among 38 cases with AL, 33 had acute myelogenous leukemia (AML), 3 had acute lymphoid leukemia (ALL; 1 Philadelphia chromosome positive), 1 had biphenotypic and 1 had extramedullary (bladder) AML. Unfavorable AML karyotype was noted in 26, and intermediate risk in 7 patients. There was no association between survival from AL and number of prior CLL regimens or karyotype. Expression of CD7 on blasts was associated with shorter survival. Among MDS cases, all International Prognostic Scoring System (IPSS) were represented; karyotype was unfavorable in 36, intermediate in 6 and favorable in 12 patients; 10 experienced transformation to AML. Shorter survival from MDS correlated with higher risk IPSS, poor-risk karyotype and increased number of prior CLL treatments. Overall, outcomes for patients with CLL subsequently diagnosed with AL or MDS were very poor; AL/MDS occurred without prior CLL treatment. Effective therapies for these patients are desperately needed.

Expression profile of 11 proteins and their prognostic significance in patients with chronic lymphocytic leukemia (CLL).

It has been suggested that the expansion of the leukemic cells in chronic lymphocytic leukemia (CLL) is due to dysregulation of pathways of programmed cell death (apoptosis) rather than cell proliferation, although differences may exist in early vs late and treated vs untreated patients. In the present study, we analyzed the expression of 11 proteins in CLL cells that are implicated in the control of apoptosis, proliferation, and differentiation, and correlated this expression profile with survival. Using a quantitative solid-phase radioimmunoassay (RIA), we measured the cellular protein levels of Bcl-2, cyclin D1, PCNA, ATM, Fas, Bax, retinoic acid receptor alpha (RARalpha), retinoic acid receptor beta (RXRbeta), Flt1, VEGF, and cellular beta2-microglobulin in 230 samples of CLL. Univariate analysis using the Cox proportional hazard model showed a correlation with survival of only the following proteins: Bcl-2 (P < 0.001), cyclin D1 (P = 0.027), Fas (P = 0.055), PCNA (P < 0.001), and ATM (P = 0.028). In a multivariate analysis using classification and regression tree analysis (CART), five groups of patients (nodes) could be generated with significant differences of survival expectation (P < 0.0001) based on levels of expression of the above proteins. Based on CART analysis, Bcl-2 levels emerge as the most important protein in predicting survival between all 11 proteins studied. Patients with marked elevation in Bcl-2 levels had the worst outcome while patients with intermediate levels, but with high levels of PCNA and cyclin D1 or abnormal ATM expression had intermediate survival. These data indicate that intracellular levels of proteins such as Bcl-2, ATM, cyclin D1, and PCNA can be used as markers to predict clinical behavior and survival in patients with CLL. The pathways in which these proteins are involved may also represent possible targets for future therapeutic trials in CLL.

Correlation of the proliferative index of residual leukemia with outcome in patients treated with sequential high dose ara-C and asparaginase.

Therapy of acute myelogenous leukemia (AML) with sequential high-dose ara-C and asparaginase (HiDAC----ASNase) on a day 1 and 8 schedule was designed to exploit potential recruitment of residual leukemia cells following initial cytoreduction from day 1 treatment. DNA flow cytometry was used to evaluate the proliferative index (%S + G2M) of bone marrow leukemia cells from pretreatment and day 8 marrow samples. The proliferative index on day 1, day 8, and incremental change (day 8 minus day 1) were analyzed for their correlation with bone marrow aplasia on day 15 and with the attainment of subsequent complete remission. Pretreatment (day 1) and the change in proliferative index did not correlate (p greater than 0.10) with day 15 marrow aplasia or with clinical outcome. However, the magnitude of the day 8 proliferative index did relate to the attainment of bone marrow aplasia on day 15 (p = 0.05) and the attainment of complete remission (p = 0.002). Recruitment of residual leukemia cells into the proliferative phases of the cell cycle may contribute to the unique efficacy of the day 1 and 8 schedule of HIDAC----ASNase. Additionally, the cytokinetics of residual leukemia after initial chemotherapy may be predictive of outcome and could be useful as a marker for the design of optimal therapeutic regimens.

Interaction of granulocytopenia and construction activity as risk factors for nosocomial invasive filamentous fungal disease in patients with hematologic disorders.

A clinicoepidemiologic study was undertaken to investigate an apparent increase in frequency of nosocomial invasive filamentous fungal disease (NIFFD) in adult patients with acute leukemia hospitalized during a period of hospital construction, and to determine if a relationship existed between the construction activity and the acquisition of NIFFD. The first study goal, to determine the incidence of NIFFD before and during construction, was approached by calculation of incidence rates of NIFFD in patients with acute leukemia, comparing 1982 and 1983 (a baseline period free of construction) to 1986 (a year when construction activity was at its peak). The second study goal, to identify risk factors for the development of NIFFD, was accomplished by reviewing the autopsy records of all patients with underlying hematologic disorders accompanied by granulocytopenia who died in our hospital from 1982 through 1986. Patients with and without autopsy evidence of NIFFD were then compared by univariate and multivariate (logistic regression) analysis to identify potential risk factors for the acquisition of NIFFD. The incidence of NIFFD in patients with acute leukemia hospitalized during the period of hospital construction was significantly increased when compared to a baseline period without construction (11 per 139 versus 4 per 333, p less than .001). Review of all granulocytopenic patients autopsied over the five-year interval 1982 through 1986 revealed duration of granulocytopenia and hospitalization during construction to be risk factors for NIFFD by univariate analysis (p less than .005). Logistic regression showed duration of granulocytopenia to be highly significant independent risk factor (p less than .01) and construction activity to be a probable independent risk factor (p = .09).(ABSTRACT TRUNCATED AT 250 WORDS)

Standardized measurement of the future liver remnant prior to extended liver resection: methodology and clinical associations.

BACKGROUND: There is no agreement regarding the preoperative measurement of liver volumes and the minimal safe size of the liver remnant after extended hepatectomy. METHODS: In 20 patients with hepatobiliary malignancy and no underlying chronic liver disease, volumetric measurements of the liver remnant (segments 2 and 3 +/- 1) were obtained before extended right lobectomy (right trisegmentectomy). The ratios of future liver remnant to total liver volume were calculated by using a formula based on body surface area. In 12 patients, response to preoperative right trisectoral portal vein embolization was evaluated. In 15 patients who underwent the planned resection, preoperative volumes were correlated with biochemical and clinical outcome parameters. RESULTS: The future liver remnants increased after portal vein embolization (26% versus 36%, P < .01). Smaller size liver remnants were associated with an increase in postoperative liver function tests (P < .05) and longer lengths of hospital stay (P < .02). Preliminary data indicates an increase in major complications for liver volumes < or = 25% (P = .02). CONCLUSIONS: A simple method of measurement provides an assessment of the liver remnant before resection. It is useful in evaluating response to portal vein embolization and in predicating the outcome before extended liver resections.

Underlying liver disease, not tumor factors, predicts long-term survival after resection of hepatocellular carcinoma.

HYPOTHESIS: A subset of patients can be identified who will survive without recurrence beyond 5 years after hepatic resection for hepatocellular carcinoma (HCC). DESIGN: A retrospective review of a multi-institutional database of 591 patients who had undergone hepatic resection for HCC and on-site reviews of clinical records and pathology slides. SETTING: All patients had been treated in academic referral centers within university-based hospitals. PATIENTS: We identified 145 patients who had survived for 5 years or longer after hepatic resection for HCC. MAIN OUTCOME MEASURES: Clinical and pathologic factors, as well as scoring of hepatitis and fibrosis in the surrounding liver parenchyma, were assessed for possible association with survival beyond 5 years and cause of death among the 145 five-year survivors. RESULTS: Median additional survival duration longer than 5 years was 4.1 years. Women had significantly longer median additional survival durations than did men (81 months vs 38 months, respectively, after the 5-year mark) (P =.008). Surgical margins, type of resection, an elevated preoperative alpha-fetoprotein level, and the presence of multiple tumors or microscopic vascular invasion had no bearing on survival longer than 5 years. However, patients who survived for 5 years who also had normal underlying liver or minimal fibrosis (score, 0-2) at surgery had significantly longer additional survival than did patients with moderate fibrosis (score, 3-4) or severe fibrosis/cirrhosis (score, 5-6) (P<.001). CONCLUSIONS: Death caused by HCC is rare beyond 5 years after resection of HCC in the absence of fibrosis or cirrhosis. The data suggest that chronic liver disease acts as a field of cancerization contributing to new HCC. These patients may benefit from therapies directed at the underlying liver disease.

Diagnostic value of p53 immunohistochemistry in Barrett's esophagus: an endoscopic study.

The aim of this study was to establish the diagnostic utility of p53 immunohistochemistry in Barrett's esophagus. Three pathologists reviewed endoscopic biopsy specimens and one surgical specimen derived from 102 subjects in a prospective follow-up series. Dysplasia was graded as negative, indefinite, low grade and high grade. p53 staining was assessed as negative, weak or patchy and strong. Kappa coefficients for interobserver agreement for the three combinations of observer pairs were moderate to substantial (0.48, 0.55 and 0.68) for dysplasia and substantial to near perfect (0.77, 0.82 and 0.89) for p53 immunostaining. In the consensus grading achieved for dysplasia, strong p53 staining was recorded in 0/79 samples that were negative or indefinite for dysplasia, 5/15 (33%) examples of low grade dysplasia and 7/8 (87%) examples of high grade dysplasia. Non-dysplastic p53 positive glands were seen in specimens from two subjects harbouring dysplasia or cancer elsewhere. Two p53 positive specimens were, upon review and discussion, re-assigned from low- to high-grade dysplasia. It is concluded that p53 immunohistochemistry facilitates the interpretation of Barrett's epithelium but need only be employed to confirm a suspected diagnosis of dysplasia and assist with the distinction between low- and high-grade dysplasia.

Definition of Ross River virus vectors at Maroochy Shire, Australia.

Evidence of Ross River (RR) virus infection in field-collected mosquitoes and data from laboratory vector competence experiments incriminated a range of mosquito species as important vectors of RR virus in Maroochy Shire, Queensland, Australia. Nine RR and 2 Barmah Forest virus isolates were recovered from 27,529 mosquitoes collected in light traps from Maroochy Shire during 1996. Nine of the 10 most abundant mosquito species collected in light traps were fed on blood containing the B94/20 strain RR isolated from Queensland in 1994. All species except for Culex sitiens Wiedemann were susceptible to experimental infection. Evidence of RR virus transmission to mice was found with Aedes vigilax (Skuse), Aedes funereus (Theobald), Aedes procax (Skuse), Culex annulirostris Skuse, Mansonia uniformis (Theobald) and Culex australicus Dobrotworsky & Drummond. Aedes notoscriptus (Skuse) and Aedes multiplex (Theobald) were susceptible to RR virus infection, although there was no evidence of virus transmission. Based on adult abundance and vector competence results, freshwater species such as Cx. annulirostris, Ae. procax, and Ae. funereus, and saltmarsh Ae. vigilax, appear to be important vectors of RR virus in Maroochy Shire and control programs should be revised to include these species.

Spatial and temporal analysis of Ross River virus disease patterns at Maroochy Shire, Australia: association between human morbidity and mosquito (Diptera: Culicidae) abundance.

Notifications of Ross River (RR) virus disease in Maroochy Shire were mapped according to the patient's place of residence, and standardized morbidity ratios were calculated for each of 11 census districts, with 4 areas having higher than average overall rates of RR virus disease notification. Temporal analysis of RR virus disease notifications from each of the 11 areas indicated that epidemics of RR virus disease either were widespread, resulting in higher than average numbers of cases from the majority of areas, or limited to only certain areas. Mosquitoes were collected in CO2 light traps, and the numbers analyzed to determine whether there was any association between mosquito abundance and disease incidence. Seasonal light trap indices for the fresh water-breeding Culex annulirostris Skuse and brackish water-breeding Aedes funereus (Theobald) were correlated positively with disease incidence. There was no apparent association between Aedes vigilax (Skuse) or Aedes procax (Skuse) light trap indices and disease incidence. Identification of human subpopulations with higher than average rates of RR virus disease, and the mosquito vectors associated with virus transmission, should provide a framework for the development of focused and therefore more effective control programs.

Reconstructive management of cranial base defects after tumor ablation.

Successful reconstruction after cranial base tumor ablation is paramount in preventing potentially life-threatening complications. The purpose of this study was to evaluate experiences of cranial base reconstruction and to identify reconstructive management principles that may assist in achieving successful cranial base reconstruction. All cranial base reconstructions performed by the Department of Plastic Surgery at the University of Texas M. D. Anderson Cancer Center between January of 1993 and September of 1999 were reviewed. Analyses were performed to assess the impact of location of defect, type of reconstruction, type of dural repair, and history of preoperative radiation and chemotherapy on rates of complications, and patient survival. The 77 patients who underwent cranial base reconstruction after tumor ablation during the study period had a mean age of 52 years (6 to 84 years). The mean follow-up period was 28.7 months (1 to 76 months). Squamous cell carcinoma, the most common histopathologic type, was present in 24 patients (31 percent), and 35 patients (45 percent) presented with recurrent disease. Location of defects involved region I (anterior) in 31 patients (40 percent), region II (anterior-lateral) in 18 (23 percent), region III (lateral-posterior) in six (8 percent), and more than one region in 22 (29 percent). Reconstructive methods included free flaps in 52 patients (68 percent), temporalis muscle flaps in 14 (18 percent), pericranial flaps in eight (10 percent), and other local flaps (two galeal, one scalp) in three (4 percent). Of the 52 free flaps, 18 (35 percent) were used in region I, 14 (27 percent) in region II, six (12 percent) in region III, and 14 (27 percent) in defects involving more than one region. Of the 14 temporalis muscle flaps, 13 (93 percent) were used for defects involving regions I or II and one (7 percent) was used for a defect involving region III. Of the 11 pericranial and other local flaps, nine (82 percent) were used in region I, one (9 percent) in region II, and one (9 percent) in a combination of regions II and III. Complications occurred in 21 patients (27 percent): three total flap losses (4 percent), three partial flap losses (4 percent), two cerebrospinal fluid leaks (3 percent), two cases of meningitis (3 percent), two abscesses (3 percent), five cases of delayed wound healing (6 percent), two hematomas (3 percent), one wound infection (1 percent), and one cerebrovascular accident (1 percent). Overall survival was 77 percent at 2 years and 58 percent at 4 years. The type of reconstruction, location of defect, type of dural repair, and history of preoperative radiation and chemotherapy had no significant association with the incidence of complications. Neither the type of reconstruction nor the location of defect showed a significant effect on patient survival. In this experience, local flaps, such as pericranial or temporalis muscle flaps, are good choices for reconstruction of smaller anterior or lateral cranial base defects. For defects that require larger amounts of soft tissue, free flaps are appropriate. With proper patient selection, successful cranial base reconstruction can be performed with either local or free flaps with a low incidence of complications.

Estimating Shared Copy Number Aberrations for Array CGH Data: The Linear-Median Method.

MOTIVATION: Existing methods for estimating copy number variations in array comparative genomic hybridization (aCGH) data are limited to estimations of the gain/loss of chromosome regions for single sample analysis. We propose the linear-median method for estimating shared copy numbers in DNA sequences across multiple samples, demonstrate its operating characteristics through simulations and applications to real cancer data, and compare it to two existing methods. RESULTS: Our proposed linear-median method has the power to estimate common changes that appear at isolated single probe positions or very short regions. Such changes are hard to detect by current methods. This new method shows a higher rate of true positives and a lower rate of false positives. The linear-median method is non-parametric and hence is more robust in estimating copy number. Additionally the linear-median method is easily computable for practical aCGH data sets compared to other copy number estimation methods.

Application of gene shaving and mixture models to cluster microarray gene expression data.

Researchers are frequently faced with the analysis of microarray data of a relatively large number of genes using a small number of tissue samples. We examine the application of two statistical methods for clustering such microarray expression data: EMMIX-GENE and GeneClust. EMMIX-GENE is a mixture-model based clustering approach, designed primarily to cluster tissue samples on the basis of the genes. GeneClust is an implementation of the gene shaving methodology, motivated by research to identify distinct sets of genes for which variation in expression could be related to a biological property of the tissue samples. We illustrate the use of these two methods in the analysis of Affymetrix oligonucleotide arrays of well-known data sets from colon tissue samples with and without tumors, and of tumor tissue samples from patients with leukemia. Although the two approaches have been developed from different perspectives, the results demonstrate a clear correspondence between gene clusters produced by GeneClust and EMMIX-GENE for the colon tissue data. It is demonstrated, for the case of ribosomal proteins and smooth muscle genes in the colon data set, that both methods can classify genes into co-regulated families. It is further demonstrated that tissue types (tumor and normal) can be separated on the basis of subtle distributed patterns of genes. Application to the leukemia tissue data produces a division of tissues corresponding closely to the external classification, acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL), for both methods. In addition, we also identify genes specific for the subgroup of ALL-Tcell samples. Overall, we find that the gene shaving method produces gene clusters at great speed; allows variable cluster sizes and can incorporate partial or full supervision; and finds clusters of genes in which the gene expression varies greatly over the tissue samples while maintaining a high level of coherence between the gene expression profiles. The intent of the EMMIX-GENE method is to cluster the tissue samples. It performs a filtering step that results in a subset of relevant genes, followed by gene clustering, and then tissue clustering, and is favorable in its accuracy of ranking the clusters produced.

Discriminant analysis of event-related potential curves using smoothed principal components.

Principal component analysis enhanced by the use of smoothing is used in conjunction with discriminant analysis techniques to devise a statistical classification method for the analysis of event-related potential data. A training set of premedication potentials collected from adolescents with attention-deficit hyperactive disorder (ADHD) is used to predict whether adolescents from an independent subject group will respond to long-term medication. Comparison of outcome prediction rates demonstrates that this method, which uses information from the whole ERP curve, is superior to the classification technique currently used by clinicians, which is based on a single ERP curve feature. The need to administer an initial dose of medication to classify patients is also eliminated.