Dataset Information

A comparative study of survival models for breast cancer prognostication revisited: the benefits of multi-gene models.

ABSTRACT: BACKGROUND:The development of clinical -omic biomarkers for predicting patient prognosis has mostly focused on multi-gene models. However, several studies have described significant weaknesses of multi-gene biomarkers. Indeed, some high-profile reports have even indicated that multi-gene biomarkers fail to consistently outperform simple single-gene ones. Given the continual improvements in -omics technologies and the availability of larger, better-powered datasets, we revisited this "single-gene hypothesis" using new techniques and datasets. RESULTS:By deeply sampling the population of available gene sets, we compare the intrinsic properties of single-gene biomarkers to multi-gene biomarkers in twelve different partitions of a large breast cancer meta-dataset. We show that simple multi-gene models consistently outperformed single-gene biomarkers in all twelve partitions. We found 270 multi-gene biomarkers (one per ~11,111 sampled) that always made better predictions than the best single-gene model. CONCLUSIONS:The single-gene hypothesis for breast cancer does not appear to retain its validity in the face of improved statistical models, lower-noise genomic technology and better-powered patient cohorts. These results highlight that it is critical to revisit older hypotheses in the light of newer techniques and datasets.

SUBMITTER: Grzadkowski MR

PROVIDER: S-EPMC6215649 | biostudies-literature | 2018 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A comparative study of survival models for breast cancer prognostication revisited: the benefits of multi-gene models.

Grzadkowski Michal R MR Sendorek Dorota H DH P'ng Christine C Huang Vincent V Boutros Paul C PC

BMC bioinformatics 20181103 1

<h4>Background</h4>The development of clinical -omic biomarkers for predicting patient prognosis has mostly focused on multi-gene models. However, several studies have described significant weaknesses of multi-gene biomarkers. Indeed, some high-profile reports have even indicated that multi-gene biomarkers fail to consistently outperform simple single-gene ones. Given the continual improvements in -omics technologies and the availability of larger, better-powered datasets, we revisited this "sin ...[more]

PMID: 30390622

Similar Datasets

Project description:BACKGROUND:The mass extirpation of the island of Moorea's endemic partulid tree snail fauna, following the deliberate introduction of the alien predator Euglandina rosea, represents one of the highest profile conservation crises of the past thirty years. All of the island's partulids were thought to be extirpated by 1987, with five species persisting in zoos, but intensive field surveys have recently detected a number of surviving wild populations. We report here a mitochondrial (mt) phylogenetic estimate of Moorean partulid wild and captive lineage survival calibrated with a reference museum collection that pre-dates the predator's introduction and that also includes a parallel dataset from the neighboring island of Tahiti. RESULTS:Although severe winnowing of Moorea's mt lineage diversity has occurred, seven of eight (six Partula; two Samoana) partulid tip clades remain extant. The extinct mt clade occurred predominantly in the P. suturalis species complex and it represented a major component of Moorea's endemic partulid treespace. Extant Moorean mt clades exhibited a complex spectrum of persistence on Moorea, in captivity, and (in the form of five phylogenetically distinct sister lineages) on Tahiti. Most notably, three Partula taxa, bearing two multi-island mt lineages, have survived decades of E. rosea predation on Moorea (P. taeniata) and in the valleys of Tahiti (P. hyalina and P. clara). Their differential persistence was correlated with intrinsic attributes, such as taxonomy and mt lineages, rather than with their respective within-island distribution patterns. CONCLUSION:Conservation efforts directed toward Moorean and Tahitian partulids have typically operated within a single island frame of reference, but our discovery of robust genealogical ties among survivors on both islands implies that a multi-island perspective is required. Understanding what genetic and/or ecological factors have enabled Partula taeniata, P. hyalina and P. clara to differentially survive long-term direct exposure to the predator may provide important clues toward developing a viable long term conservation plan for Society Island partulid tree snails.

Project description:BACKGROUND:Non-small-cell lung cancer (NSCLC) patients often demonstrate varying clinical courses and outcomes, even within the same tumor stage. This study explores deep learning applications in medical imaging allowing for the automated quantification of radiographic characteristics and potentially improving patient stratification. METHODS AND FINDINGS:We performed an integrative analysis on 7 independent datasets across 5 institutions totaling 1,194 NSCLC patients (age median = 68.3 years [range 32.5-93.3], survival median = 1.7 years [range 0.0-11.7]). Using external validation in computed tomography (CT) data, we identified prognostic signatures using a 3D convolutional neural network (CNN) for patients treated with radiotherapy (n = 771, age median = 68.0 years [range 32.5-93.3], survival median = 1.3 years [range 0.0-11.7]). We then employed a transfer learning approach to achieve the same for surgery patients (n = 391, age median = 69.1 years [range 37.2-88.0], survival median = 3.1 years [range 0.0-8.8]). We found that the CNN predictions were significantly associated with 2-year overall survival from the start of respective treatment for radiotherapy (area under the receiver operating characteristic curve [AUC] = 0.70 [95% CI 0.63-0.78], p < 0.001) and surgery (AUC = 0.71 [95% CI 0.60-0.82], p < 0.001) patients. The CNN was also able to significantly stratify patients into low and high mortality risk groups in both the radiotherapy (p < 0.001) and surgery (p = 0.03) datasets. Additionally, the CNN was found to significantly outperform random forest models built on clinical parameters-including age, sex, and tumor node metastasis stage-as well as demonstrate high robustness against test-retest (intraclass correlation coefficient = 0.91) and inter-reader (Spearman's rank-order correlation = 0.88) variations. To gain a better understanding of the characteristics captured by the CNN, we identified regions with the most contribution towards predictions and highlighted the importance of tumor-surrounding tissue in patient stratification. We also present preliminary findings on the biological basis of the captured phenotypes as being linked to cell cycle and transcriptional processes. Limitations include the retrospective nature of this study as well as the opaque black box nature of deep learning networks. CONCLUSIONS:Our results provide evidence that deep learning networks may be used for mortality risk stratification based on standard-of-care CT images from NSCLC patients. This evidence motivates future research into better deciphering the clinical and biological basis of deep learning networks as well as validation in prospective data.

Project description:Triple-negative breast cancer (TNBC) is an aggressive breast cancer subtype lacking expression of estrogen and progesterone receptors (ER/PR) and HER2, thus limiting therapy options. We hypothesized that meta-analysis of TNBC gene expression profiles would illuminate mechanisms underlying the aggressive nature of this disease and identify therapeutic targets. Meta-analysis in the Oncomine database identified 206 genes that were recurrently deregulated in TNBC compared with non-TNBC and in tumors that metastasized or led to death within 5 years. This 'aggressiveness gene list' was enriched for two core functions/metagenes: chromosomal instability (CIN) and ER signaling metagenes. We calculated an 'aggressiveness score' as the ratio of the CIN metagene to the ER metagene, which identified aggressive tumors in breast cancer data sets regardless of subtype or other clinico-pathological indicators. A score calculated from six genes from the CIN metagene and two genes from the ER metagene recapitulated the aggressiveness score. By multivariate survival analysis, we show that our aggressiveness scores (from 206 genes or the 8 representative genes) outperformed several published prognostic signatures. Small interfering RNA screen revealed that the CIN metagene holds therapeutic targets against TNBC. Particularly, the inhibition of TTK significantly reduced the survival of TNBC cells and synergized with docetaxel in vitro. Importantly, mitosis-independent expression of TTK protein was associated with aggressive subgroups, poor survival and further stratified outcome within grade 3, lymph node-positive, HER2-positive and TNBC patients. In conclusion, we identified the core components of CIN and ER metagenes that identify aggressive breast tumors and have therapeutic potential in TNBC and aggressive breast tumors. Prognostication from these metagenes at the mRNA level was limited to ER-positive tumors. However, we provide evidence that mitosis-independent expression of TTK protein was prognostic in TNBC and other aggressive breast cancer subgroups, suggesting that protection of CIN/aneuploidy drives aggressiveness and treatment resistance.

Dataset Information

A comparative study of survival models for breast cancer prognostication revisited: the benefits of multi-gene models.

Publications

A comparative study of survival models for breast cancer prognostication revisited: the benefits of multi-gene models.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure