Dataset Information

A two-stage approach for combining gene expression and mutation with clinical data improves survival prediction in myelodysplastic syndromes and ovarian cancer.

ABSTRACT:

Motivation

Many traditional clinical prognostic factors have been known for cancer for years, but usually provide poor survival prediction. Genomic information is more easily available now which offers opportunities to build more accurate prognostic models. The challenge is how to integrate them to improve survival prediction. The common approach of jointly analyzing all type of covariates directly in one single model may not improve the prediction due to increased model complexity and cannot be easily applied to different datasets.

Results

We proposed a two-stage procedure to better combine different sources of information for survival prediction, and applied the two-stage procedure in two cancer datasets: myelodysplastic syndromes (MDS) and ovarian cancer. Our analysis suggests that the prediction performance of different data types are very different, and combining clinical, gene expression and mutation data using the two-stage procedure improves survival prediction in terms of improved concordance index and reduced prediction error.

Availability and implementation

The two-stage procedure can be implemented in BhGLM package which is freely available at http://www.ssg.uab.edu/bhglm/.

Contact

nyi@uab.edu.

SUBMITTER: Li Y

PROVIDER: S-EPMC8351588 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundSF3B1 mutations (SF3B1mut ) in myelodysplastic syndromes (MDS) frequently involve codon K700E and have a favorable prognosis. The prognostic effect of non-K700E SF3B1mut is uncertain.MethodsThe authors analyzed the clinicopathological features and outcomes of a single-institution series of 94 treatment-naive SF3B1mut MDS patients (18%) and 415 treatment-naive SF3B1wt MDS patients and explored the differences between K700E and non-K700E SF3B1mut MDS.ResultsFifty-five patients (59%) carried K700E. Recurrent non-K700E mutations (39 [41%]) included R625, H662, and K666. Compared with SF3B1mut K700E patients, non-K700E patients had a lower median absolute neutrophil count (1.8 vs 2.4; P = .005) and were frequently "high" according to the Revised International Prognostic Scoring System (19% vs 4%; P = .031). Non-K700E MDS was associated frequently with RUNX1 (26% vs 7%; P = .012) and exclusively with BCOR, IDH2, and SRSF2 mutations. A splicing analysis showed the differential distribution of alternatively spliced events and gene expression profiles between K700 and non-K700E MDS patients. The majority (at least 80%) of SF3B1mut K700E, SF3B1mut non-K700E, and SF3B1wt patients were treated with hypomethylating agents. Over a median follow-up of 16 months, SF3B1mut had superior overall survival (OS) in comparison with SF3B1wt in all MDS patients (not reached vs 25.2 months; P = .0003), in patients with low-grade MDS, and in patients with myelodysplastic syndromes with ring sideroblasts (MDS-RS). Compared with SF3B1wt , SF3B1mut K700E had superior outcomes in all MDS (median OS, 25 months vs not reached; P = .0001), in low-grade MDS (median OS, 41.3 months vs not reached; P = .0015), and in MDS-RS (median OS, 22.3 months vs not reached; P = .0001), but no significant difference was seen between non-K700E and SF3B1wt MDS. By multivariable analysis, the absence of SF3B1mut K700E mutations was independently associated with the prognosis.ConclusionsThis study highlights the importance of the SF3B1 mutation subtype in MDS risk assessment.Lay summaryMyelodysplastic syndromes (MDS) with SF3B1 mutations are regarded as having a favorable prognosis by both the World Health Organization and the International Working Group for the Prognosis of Myelodysplastic Syndromes. However, this article shows that only MDS patients with SF3B1 K700E mutations have a favorable prognosis (and not MDS patients with SF3B1 mutations involving other codons). This has important implications for refining future MDS subclassification and risk assessment criteria.

Project description:BackgroundSeveral gene sets for prediction of breast cancer survival have been derived from whole-genome mRNA expression profiles. Here, we develop a statistical framework to explore whether combination of the information from such sets may improve prediction of recurrence and breast cancer specific death in early-stage breast cancers. Microarray data from two clinically similar cohorts of breast cancer patients are used as training (n?=?123) and test set (n?=?81), respectively. Gene sets from eleven previously published gene signatures are included in the study.Principal findingsTo investigate the relationship between breast cancer survival and gene expression on a particular gene set, a Cox proportional hazards model is applied using partial likelihood regression with an L2 penalty to avoid overfitting and using cross-validation to determine the penalty weight. The fitted models are applied to an independent test set to obtain a predicted risk for each individual and each gene set. Hierarchical clustering of the test individuals on the basis of the vector of predicted risks results in two clusters with distinct clinical characteristics in terms of the distribution of molecular subtypes, ER, PR status, TP53 mutation status and histological grade category, and associated with significantly different survival probabilities (recurrence: p?=?0.005; breast cancer death: p?=?0.014). Finally, principal components analysis of the gene signatures is used to derive combined predictors used to fit a new Cox model. This model classifies test individuals into two risk groups with distinct survival characteristics (recurrence: p?=?0.003; breast cancer death: p?=?0.001). The latter classifier outperforms all the individual gene signatures, as well as Cox models based on traditional clinical parameters and the Adjuvant! Online for survival prediction.ConclusionCombining the predictive strength of multiple gene signatures improves prediction of breast cancer survival. The presented methodology is broadly applicable to breast cancer risk assessment using any new identified gene set.