-KIDS: A Novel Feature Evaluation in the Ultrahigh-dimensional Right-censored Setting, with Application to Head and Neck Cancer

Overview

Journal medRxiv

Date 2024 Aug 26

PMID 39185537

Authors

Atika FArzana Urmi

Chenlu Ke

Dipankar Bandyopadhyay

Affiliations

Soon will be listed here.

Abstract

Recent advances in sequencing technologies have allowed collection of massive genome-wide information that substantially enhances the diagnosis and prognosis of head and neck cancer. Identifying predictive markers for survival time is crucial for devising prognostic systems, and learning the underlying molecular driver of the cancer course. In this paper, we introduce -KIDS, a model-free feature screening procedure with false discovery rate (FDR) control for ultrahigh dimensional right-censored data, which is robust against unknown censoring mechanisms. Specifically, our two-stage procedure initially selects a set of important features with a dual screening mechanism using nonparametric reproducing-kernel-based ANOVA statistics, followed by identifying a refined set (of features) under directional FDR control through a unified knockoff procedure. The finite sample properties of our method, and its novelty (in light of existing alternatives) are evaluated via simulation studies. Furthermore, we illustrate our methodology via application to a motivating right-censored head and neck (HN) cancer survival data derived from The Cancer Genome Atlas, with further validation on a similar HN cancer data from the Gene Expression Omnibus database. The methodology can be implemented via the R package DSFDRC, available in GitHub.

References

Wichmann G, Rosolowski M, Krohn K, Kreuz M, Boehm A, Reiche A . The role of HPV RNA transcription, immune response-related gene expression and disruptive TP53 mutations in diagnostic and prognostic profiling of head and neck cancer. Int J Cancer. 2015; 137(12):2846-57. DOI: 10.1002/ijc.29649. View

Heagerty P, Zheng Y . Survival model predictive accuracy and ROC curves. Biometrics. 2005; 61(1):92-105. DOI: 10.1111/j.0006-341X.2005.030814.x. View

Li R, Zhong W, Zhu L . Feature Screening via Distance Correlation Learning. J Am Stat Assoc. 2014; 107(499):1129-1139. PMC: 4170057. DOI: 10.1080/01621459.2012.695654. View

Kosorok M . On Brownian Distance Covariance and High Dimensional Data. Ann Appl Stat. 2010; 3(4):1266-1269. PMC: 2889501. DOI: 10.1214/09-AOAS312. View

Liu Y, Chen X, Li G . A new joint screening method for right-censored time-to-event data with ultra-high dimensional covariates. Stat Methods Med Res. 2019; 29(6):1499-1513. PMC: 8285086. DOI: 10.1177/0962280219864710. View

Fan J, Lv J . A Selective Overview of Variable Selection in High Dimensional Feature Space. Stat Sin. 2011; 20(1):101-148. PMC: 3092303. View

Fan L, Jeng Y, Lu Y, Lien H . SPOCK1 Is a Novel Transforming Growth Factor-β-Induced Myoepithelial Marker That Enhances Invasion and Correlates with Poor Prognosis in Breast Cancer. PLoS One. 2016; 11(9):e0162933. PMC: 5023187. DOI: 10.1371/journal.pone.0162933. View

Song R, Lu W, Ma S, Jeng X . Censored Rank Independence Screening for High-dimensional Survival Data. Biometrika. 2015; 101(4):799-814. PMC: 4318124. DOI: 10.1093/biomet/asu047. View

Kimple R, Harari P . The prognostic value of HPV in head and neck cancer patients undergoing postoperative chemoradiotherapy. Ann Transl Med. 2015; 3(Suppl 1):S14. PMC: 4437932. DOI: 10.3978/j.issn.2305-5839.2015.01.37. View

10.

Su K, Wang Y, Gu H, Ma L, Xuan G . Overexpression of fatty acid desaturase 3 predicts poor prognosis in head and neck squamous cell carcinoma. Medicine (Baltimore). 2023; 101(49):e32119. PMC: 9750517. DOI: 10.1097/MD.0000000000032119. View

11.

Fadista J, Vikman P, Ottosson Laakso E, Mollet I, Esguerra J, Taneera J . Global genomic and transcriptomic analysis of human pancreatic islets reveals novel genes influencing glucose metabolism. Proc Natl Acad Sci U S A. 2014; 111(38):13924-9. PMC: 4183326. DOI: 10.1073/pnas.1402665111. View

12.

Li J, Zheng Q, Peng L, Huang Z . Survival impact index and ultrahigh-dimensional model-free screening with survival outcomes. Biometrics. 2016; 72(4):1145-1154. PMC: 4993699. DOI: 10.1111/biom.12499. View

13.

Hong H, Chen X, Christiani D, Li Y . Integrated powered density: Screening ultrahigh dimensional covariates with survival outcomes. Biometrics. 2017; 74(2):421-429. PMC: 6495533. DOI: 10.1111/biom.12820. View

14.

Ribeiro I, Esteves L, Caramelo F, Carreira I, Melo J . Integrated Multi-Omics Signature Predicts Survival in Head and Neck Cancer. Cells. 2022; 11(16). PMC: 9406438. DOI: 10.3390/cells11162536. View

15.

Leemans C, Braakhuis B, Brakenhoff R . The molecular biology of head and neck cancer. Nat Rev Cancer. 2010; 11(1):9-22. DOI: 10.1038/nrc2982. View

16.

Fan J, Feng Y, Song R . Nonparametric Independence Screening in Sparse Ultra-High Dimensional Additive Models. J Am Stat Assoc. 2012; 106(494):544-557. PMC: 3263471. DOI: 10.1198/jasa.2011.tm09779. View

17.

Zhang P, Zhao Y, Xia X, Mei S, Huang Y, Zhu Y . Expression of gene on tumor-associated macrophages of head and neck squamous cell carcinoma, and its correlation with clinical outcome. Oncoimmunology. 2023; 12(1):2203073. PMC: 10120517. DOI: 10.1080/2162402X.2023.2203073. View

18.

Zhang H . Discussion of "Sure Independence Screening for Ultra-High Dimensional Feature Space. J R Stat Soc Series B Stat Methodol. 2009; 70(5):903. PMC: 2709408. DOI: 10.1111/j.1467-9868.2008.00674.x. View

19.

Shu Y, Weng H, Ye Y, Hu Y, Bao R, Cao Y . SPOCK1 as a potential cancer prognostic marker promotes the proliferation and metastasis of gallbladder cancer cells by activating the PI3K/AKT pathway. Mol Cancer. 2015; 14:12. PMC: 4320842. DOI: 10.1186/s12943-014-0276-y. View

20.

Argiris A, Karamouzis M, Raben D, Ferris R . Head and neck cancer. Lancet. 2008; 371(9625):1695-709. PMC: 7720415. DOI: 10.1016/S0140-6736(08)60728-X. View