||5R03CA235122-02 Interpret this number
||Fred Hutchinson Cancer Research Center
||Novel Methods for Missing Subtype Data in Colorectal Cancer
The proposed project is in response to PAR-18-021, NCI Small Grants Program for Cancer Research (NCI
Omnibus R03). We are motivated by the resources in the Genetics and Epidemiology of Colorectal Cancer
Consortium (GECCO). We are primarily interested in developing and applying innovative statistical methods
for missing cancer subtype data. Many diseases, such as colorectal cancer (CRC), are heterogeneous.
Molecular characterization of tumors has provided evidence of multiple tumor subtypes that develop through
activation of diverse neoplastic pathways. Important CRC tumor subtypes include microsatellite instability
(MSI) status, somatic mutations in BRAF and KRAS, and CpG island methylator phenotype. For instance,
MSI status is associated with survival outcomes and treatment response. However, some individuals may
have unknown MSI status, and other tumor biomarkers. Regression analysis may encounter a challenge
due to missing data in some individuals of the study cohort. Methodology for missing data is often required
to address the issue on bias in effect estimation and efﬁciency. Speciﬁc aims of this proposal include:
(i) To develop and apply methods to take into account missing cancer subtype data in multinomial logistic
regression. (ii) To develop and apply methods to adjust for survival analysis among cancer cases in which
multiple tumor biomarkers may be missing. The methods developed in the proposal are applicable to the
GECCO and other studies where cancer subtype data may be unknown among some study individuals. The
methods can be applied to other common study designs such as nested case-control designs and
Cox regression with competing risks.
Multinomial logistic regression with missing outcome data: An application to cancer subtypes.
, Hsu L.
Statistics in medicine, 2020-10-30; 39(24), p. 3299-3312.
Semiparametric regression calibration for general hazard models in survival analysis with covariate measurement error; surprising performance under linear hazard.
, Song X.
Biometrics, 2020-06-18; , .
Methods for generalized change-point models: with applications to human immunodeficiency virus surveillance and diabetes data.
, Wang C.Y.
, Zangeneh S.
, Chen Y.Q.
Statistics in medicine, 2020-04-15; 39(8), p. 1167-1182.