Search bioRxiv⌕ Search

Biology subjects

Thu, K. L.

Publications and source records attributed to Thu, K. L..

2 recordsLinked to original sources

Identification of KIFC1 as a putative vulnerability in lung cancers with centrosome amplification

Centrosome amplification (CA), an abnormal increase in the number of centrosomes in the cell, is a recurrent phenomenon in lung and other malignancies. Although CA contributes to tumor development and progression by promoting genomic instability (GIN), it also induces mitotic stress that jeopardizes cellular integrity. The presence of extra centrosomes leads to the formation of multipolar mitotic spindles prone to causing lethal chromosome segregation errors during cell division. To sustain the benefits of CA, malignant cells are dependent on adaptive mechanisms to mitigate its detrimental consequences, and these mechanisms represent therapeutic vulnerabilities. We aimed to discover genetic dependencies associated with CA in lung cancer. Combining a CRISPR/Cas9 functional genomics screen with analyses of tumor genomic data, we identified the motor protein KIFC1 as a putative vulnerability specifically in lung cancers with CA. KIFC1 expression was positively correlated with CA in lung adenocarcinoma (LUAD) cell lines and with a gene expression signature predictive of CA in LUAD tumor tissues. High KIFC1 expression was associated with worse patient outcomes, smoking history, and indicators of GIN. KIFC1 loss-of-function sensitized LUAD cells to potentiation of CA and sensitization was associated with a diminished ability of KIFC1-depleted cells to cluster extra centrosomes into pseudo-bipolar mitotic spindles. Our work suggests that KIFC1 inhibition represents a novel approach for potentiating GIN to lethal levels in LC with CA by forcing cells to divide with multipolar spindles, rationalizing the clinical development of KIFC1 inhibitors and further studies to investigate its therapeutic potential.

cancer biology↗

Gene isoforms as expression-based biomarkers predictive of drug response in vitro

BackgroundOne of the main challenges in precision medicine is the identification of molecular features associated to drug response to provide clinicians with tools to select the best therapy for each individual cancer patient. The recent adoption of next-generation sequencing technologies enables accurate profiling of not only gene expression but also alternatively-spliced transcripts in large-scale pharmacogenomic studies. Given that altered mRNA splicing has been shown to be prominent in cancers, linking this feature to drug response will open new avenues of research in biomarker discovery.\n\nMethodsTo address the lack of reproducibility of drug sensitivity measurements across studies, we developed a meta-analytical framework combining the pharmacological data generated within the Cancer Cell Line Encyclopedia (CCLE) and the Genomics of Drug Sensitivity in Cancer (GDSC). Predictive models are fitted with CCLE RNA-seq data as predictor variables, controlled for tissue type, and combined GDSC and CCLE drug sensitivity values as dependent variables.\n\nResultsWe first validated the biomarkers identified from GDSC and CCLE using an existing pharmacogenomic dataset of 70 breast cancer cell lines. We further selected four drugs with the most promising biomarkers to test whether their predictive value is robust to change in pharmacological assay. We successfully validated 10 isoform-based biomarkers predictive of drug response in breast cancer, including TGFA-001 for the MEK tyrosine kinase inhibitor (TKI) AZD6244, DUOX-001 for the EGFR inhibitor erlotinib, and CPEB4-001 transcript expression associated with lack of sensitivity to paclitaxel.\n\nConclusionThe results of our meta-analysis of pharmacogenomic data suggest that isoforms represent a rich resource for biomarkers predictive of response to chemo- and targeted therapies. Our study also showed that the validation rate for this type of biomarkers is low (<50%) for most drugs, supporting the requirements for independent datasets to identify reproducible predictors of response to anticancer drugs.

bioinformatics↗