Search

Motivation The principle of Breiman's random forest (RF) is to build and assemble complementary classification trees in a way that maximizes their variability. We propose a new type of random forest that disobeys Breiman's principles and involves building trees with no classification errors in very large quantities. We used a new type of decision tree that uses a neuron at each node as well as an in-innovative half Christmas tree structure. With these new RFs, we developed a score, based on a family of ten new statistical information criteria, called Nguyen information criteria (NICs), to evaluate the predictive qualities of features in three dimensions. Results The first NIC allowed the Akaike information criterion to be minimized more quickly than data obtained with the Gini index when the features were introduced in a logistic regression model. The selected features based on the NICScore showed a slight advantage compared to the support vector machines—recursive feature elimination (SVM-RFE) method. We demonstrate that the inclusion of artificial neurons in tree nodes allows a large number of classifiers in the same node to be taken into account simultaneously and results in perfect trees without classification errors. Availability and implementation The methods used to build the perfect trees in this article were implemented in the 'ROP' R package, archived at https://cran.r-project.org/web/packages/ROP/index.html. Supplementary information Supplementary data are available at Bioinformatics online. [ABSTRACT FROM AUTHOR]

'Breast cancer gene-expression miner' (bc-GenExMiner) is a breast cancer–associated web portal (http://bcgenex.ico.unicancer.fr). Here, we describe the development of a new statistical mining module, which permits several differential gene expression analyses, i.e. 'Expression' module. Sixty-two breast cancer cohorts and one healthy breast cohort with their corresponding clinicopathological information are included in bc-GenExMiner v4.5 version. Analyses are based on microarray or RNAseq transcriptomic data. Thirty-nine differential gene expression analyses, grouped into 13 categories, according to clinicopathological and molecular characteristics ('Targeted' and 'Exhaustive') and gene expression ('Customized'), have been developed. Output results are visualized in four forms of plots. This new statistical mining module offers, among other things, the possibility to compare gene expression in healthy (cancer-free), tumour-adjacent and tumour tissues at once and in three triple-negative breast cancer subtypes (i.e. C1: molecular apocrine tumours; C2: basal-like tumours infiltrated by immune suppressive cells and C3: basal-like tumours triggering an ineffective immune response). Several validation tests showed that bioinformatics process did not alter the pathobiological information contained in the source data. In this work, we developed and demonstrated that bc-GenExMiner 'Expression' module can be used for exploratory and validation purposes. Database URL : http://bcgenex.ico.unicancer.fr [ABSTRACT FROM AUTHOR]

Background. This randomized phase II trial was designed to compare the rate of pathologic complete response (pCR) induced by neoadjuvant cyclophosphamide plus doxorubicin (AC) followed by ixabepilone or paclitaxel in women with early stage breast cancer (BC). Expression of βIII-tubulin as a predictive marker was also evaluated. Patients and Methods. Women with untreated, histologically confirmed primary invasive breast adenocarcinoma received four cycles of AC followed by 1:1 randomization to either ixabepilone 40 mg/m2 (3-hour infusion) every 3 weeks for four cycles (n = 148) or weekly paclitaxel 80 mg/m2 (1-hour infusion) for 12 weeks (n = 147). All patients underwent a core needle biopsy of the primary cancer for molecular marker analysis prior to chemotherapy. βIII-Tubulin expression was assessed using immunohistochemistry. Results. There was no significant difference in the rate of pCR in the ixabepilone treatment arm (24.3%; 90% confidence interval [CI],18.6-30.8)and the paclitaxel treatment arm (25.2%;90%CI, 19.4-31.7). βIII-Tubulin-positive patients obtained higher pCR Rates compared with βIII-tubulin-negative patients in both treatment arms; however, βIII-tubulin expression was not significantly associated with a differential response to ixabepilone or paclitaxel. The safety profiles of both regimens were generally similar, although neutropenia occurred more frequently in the ixabepilone arm (grade 3/4: 41.3% vs. 8.4%). The most common nonhematologic toxicity was peripheral neuropathy. Conclusions. Neoadjuvant treatment of early stage BC with AC followed by ixabepilone every 3 weeks or weekly paclitaxel was well tolerated with no significant difference in efficacy. Higher response rates were observed among βIII-tubulin-positive patients. [ABSTRACT FROM AUTHOR]

Background Breast Cancer Trials of Oral Everolimus 2 (BOLERO-2), a phase III study in postmenopausal women with estrogen receptor–positive breast cancer progressing despite nonsteroidal aromatase inhibitor therapy, showed statistically significant benefits with adding everolimus to exemestane. Moreover, in preclinical studies, mammalian target of rapamycin inhibition was associated with decreased osteoclast survival and activity. Exploratory analyses in BOLERO-2 evaluated the effect of everolimus on bone marker levels and progressive disease in bone. Methods Patients were treated with exemestane (25mg/day) and randomized (2:1) to everolimus (10mg/day; combination) or placebo (exemestane only). Exploratory endpoints included changes in bone turnover marker levels vs baseline and progressive disease in bone, defined as unequivocal progression of a preexisting bone lesion or the appearance of a new bone lesion. Results Baseline disease characteristics were well balanced between arms (N = 724); baseline bisphosphonate use was not (43.9% combination vs 54.0% exemestane only). At a median of 18 months of follow-up, median progression-free survival (primary endpoint) was statistically significantly longer with the combination vs exemestane only (Cox proportional hazard ratio = 0.45, 95% confidence interval = 0.38 to 0.54; log-rank, 1-sided P < .0001). Bone marker levels at 6 and 12 weeks increased with exemestane only, as expected, but decreased with the combination. The cumulative incidence rate of progressive disease in bone was lower in the combination arm. Bone-related adverse events occurred with similar frequency in both arms (3.3% combination vs 4.2% exemestane only). Conclusion These exploratory analyses suggest that everolimus has beneficial effects on bone turnover and progressive disease in bone in patients receiving exemestane for hormone receptor–positive breast cancer progressing during/after nonsteroidal aromatase inhibitor therapy. [ABSTRACT FROM PUBLISHER]

We recently developed a user-friendly web-based application called bc-GenExMiner (http://bcgenex.centregauducheau.fr), which offered the possibility to evaluate prognostic informativity of genes in breast cancer by means of a 'prognostic module'. In this study, we develop a new module called 'correlation module', which includes three kinds of gene expression correlation analyses. The first one computes correlation coefficient between 2 or more (up to 10) chosen genes. The second one produces two lists of genes that are most correlated (positively and negatively) to a 'tested' gene. A gene ontology (GO) mining function is also proposed to explore GO 'biological process', 'molecular function' and 'cellular component' terms enrichment for the output lists of most correlated genes. The third one explores gene expression correlation between the 15 telomeric and 15 centromeric genes surrounding a 'tested' gene. These correlation analyses can be performed in different groups of patients: all patients (without any subtyping), in molecular subtypes (basal-like, HER2+, luminal A and luminal B) and according to oestrogen receptor status. Validation tests based on published data showed that these automatized analyses lead to results consistent with studies' conclusions. In brief, this new module has been developed to help basic researchers explore molecular mechanisms of breast cancer. [ABSTRACT FROM AUTHOR]

Patients and Methods. Between June 1997 and March 2000, a total of 1,999 patients (age <65) with localized, resectable, non-pretreated, unilateral breast cancer were randomly assigned to receive either standard FEC100 for 6 cycles or 3 cycles of FEC100 followed by 3 cycles of 100 mg/m2 docetaxel (FEC-D), both given every 21 days. Radiotherapy was mandatory after conservative surgery and tamoxifen was given for 5 years to hormone receptor (HR)- positive patients. Five-year DFS was the trial's main endpoint. Updated 8-year survival data are presented. Results. With a median follow-up of 92.8 months, 639 patients experienced at least one event. A total number of 383 deaths were registered. Eight-year DFS rates were 65.8% with FEC alone and 70.2% with FEC-D. OS rates at 8 years were 78% with FEC alone and 83.2% with FEC-D. Cox regression analysis adjusted for age and number of positive nodes showed a 15% reduction in the relative risk of relapse and a 25% reduction in the relative risk of death in favor of FEC-D. Significant relative risk reductions were observed in the HR-positive, HER2-positive, and Ki67≥20% subpopulations. [ABSTRACT FROM AUTHOR]

Background. The Hermine study observed the use of trastuzumab for metastatic breast cancer (MBC) in routine practice, including patients who received trastuzumab treatment beyond progression (TBP). Patients and Methods. The study observed 623 patients for ≥2 years. Treatment was given according to oncologists' normal clinical practices. Endpoints included duration of treatment, efficacy, and cardiac safety. The TBP subanalysis compared overall survival (OS) in 177 patients who received first-line trastuzumab and either continued trastuzumab for ≥30 days following progression or stopped at or before progression. Results. The median treatment duration was 13.3 months. In the first-, second-, and third-line or beyond treatment groups, the median time to progression (TTP) were 10.3 months, 9.0 months, and 6.3 months, and the median OS times were 30.3 months, 27.1 months, and 23.2 months, respectively. Heart failure was observed in 2.6% of patients, although no cardiac-associated deaths occurred. In the TBP subanalysis, the median OS duration from treatment initiation and time of disease progression were longer in patients who continued receiving trastuzumab TBP (>27.8 months and 21.3 months, respectively) than in those who stopped (16.8 months and 4.6 months, respectively). However, the groups were not completely comparable, because patients who continued trastuzumab TBP had better prognoses at treatment initiation. The median TTP was longer in patients who continued trastuzumab TBP (10.2 months) than in those who stopped (7.1 months). Conclusion. The Hermine findings confirm that the pivotal trials of first-line trastuzumab treatment in MBC patients are applicable in clinical practice. The subanalysis suggests that trastuzumab TBP offers a survival benefit to MBC patients treated with first-line trastuzumab. [ABSTRACT FROM AUTHOR]

Background: Peripheral neuropathy (PN) is a recognized side effect of microtubule-targeting agents and the most clinically relevant toxicity observed with the epothilone sagopilone (SAG). Studies suggest that acetyl-L-carnitine (ALC) may prevent chemotherapy-induced PN. We conducted a prospective, placebo (PBO)-controlled, doubleblind, randomized trial to investigate the safety and efficacy of ALC for the prevention of SAG-induced PN. Methods: Patients with ovarian cancer (OC) or castrationresistant prostate cancer (CRPC) and no evidence of neuropathy received SAG (16 mg/m² intravenously over 3 hours every 3 weeks) with ALC (1,000 mg every 3 days) or placebo (PBO). The primary endpoint was incidence of PN within six or fewer cycles in both treatment groups. Results: Overall, 150 patients enrolled (98 OC patients, 52 CRPC patients), with 75 per treatment arm. No significant difference in overall PN incidence was observed between treatment arms. The incidence of grade ±3 PN was significantly lower in the ALC arm in OC patients. Median duration of neuropathy was similar between treatment arms. The best overall response (according to the modified Response Evaluation Criteria in Solid Tumors), response according to tumor markers, time-to-event variables, and discontinuations because of adverse events (AEs) were comparable between treatment arms. Conclusion: Administration of ALC with SAG did not result in a significant difference in overall PN incidence compared with a PBO. OC patients in the SAG/ALC arm had a significantly lower incidence of grade 3 or 4 PN compared with OC patients in the SAG/PBO arm. [ABSTRACT FROM AUTHOR]