Skip to main content
SSO 2026 Annual Meeting Main banner

P9. Discovery of novel genetic drivers of Cowden Syndrome using genome-wide CRISPR/Cas9 screening

Thursday, March 5, 2026
11:28 AM - 11:35 AM MST
Location: 301A
Introduction:

Variants in the tumor suppressor gene PTEN drive the cancer predisposition syndrome Cowden syndrome (CS), characterized by an increased risk of both benign and malignant breast, thyroid, endometrial, and other cancers. Only about 25% of patients who meet clinical criteria for CS have a germline PTEN variant to explain their phenotype. We used genome-wide CRISPR screening to identify novel candidate genetic variants that may drive differential PTEN expression in patients with PTEN-wildtype CS.


Methods:

We performed fluorescence-activated cell sorting based genome-wide CRISPR/Cas9 screens in four cancer cell lines (HCT116 (colorectal), PC9 (lung), MCF7 (breast), and OCI-AML2 (leukemia)) to identify genes whose knock-out resulted in either a decrease or increase in PTEN expression. We paired these screen results with whole-exome sequencing data from 87 patients with PTEN wildtype CS to identify candidate driver gene mutations. Allele frequencies were compared between our patient cohort and population controls from the Genome Aggregation Database (gnomAD) using Fisher’s Exact methods.


Results:

4 genes scored as strong positive PTEN regulators in all four screens (EIF4G2, STRAP, ZNF236, and MTOR), 5 genes scored as strong positive regulators in 3 out of 4 screens (EIF4A1, DDX3X, CSDE1, DYRK1A, and ATXN7L3), and 2 genes scored as strong negative PTEN regulators in 3 of 4 screens (DUT, RTEL1). Within the Cowden-like syndrome patient cohort, we found a total of 28 potentially deleterious mutations of 9 of these 11 candidate genes. 18/28 (64%) had a higher prevalence in the Cowden-like syndrome cohort than in gnomAD (Fisher’s exact, p< 0.05), so we focused our subsequent validation on this subset. Using ChIP-seq and a reporter assay, we demonstrated that the two patient-identified variants in ZNF236 disrupt binding of ZNF236 to the PTEN promoter (Figure 1 A-C).


Conclusions:

We have identified novel candidate PTEN regulators, variants of which may explain the clinical phenotype of Cowden Syndrome in the absence of a germline PTEN mutation. We demonstrate the feasibility of combining in vitro screening techniques with patient sequencing data to identify pathogenic gene variants. Further characterization of these new regulatory pathways and mutations will provide novel insights into disease pathogenesis and also improve molecular diagnosis, especially predictive testing, cancer risk assessment, genetic counseling, and clinical management in PTEN wildtype CS.

Learning Objectives:

  • Understand how in vitro genome-wide screening can be applied to understanding patient sequencing data.
  • Identify new potential regulators of PTEN.
  • Understand how ZNF236 mutations change PTEN expression.