Identification of plasma proteins associated with ovarian cancer risk

By Pooja Toshniwal PahariaAug 8 2023Reviewed by Sophia Coveney

In a recent study posted with Preprints with The Lancet to the SSRN server, researchers performed proteomic Mendelian randomization (MR) as well as co-localization analyses to identify drug targets for ovarian cancer.

Study: Potential Drug Targets for Ovarian Cancer Identified Through Mendelian Randomization and Colocalization Analysis. Image Credit: Chinnapong/Shutterstock.com

Ovarian cancer is a common cancer among postmenopausal women, with various subtypes, and is often diagnosed in the advanced stage. Treatment options include platinum-based chemotherapy and debulking surgery. Immunotherapy and targeted therapy offer a new approach, but relapse and low five-year survival rates hinder the development of new drugs.

*Important notice: Preprints with The Lancet / SSRN publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

About the study

In the present study, researchers genetically estimated plasma proteins associated with ovarian cancer risk.

Protein quantitative trait loci (pQTL) of cis-type with genome-wide statistical significance and independent associations for 734 serological proteins [738.0 cis-acting single-nucleotide polymorphisms (SNPs)] were obtained from a recent large-scale genome-wide association study (GWAS) to explore potential drug targets for ovarian cancer.

Summary-level information on genomic links with ovarian cancer was provided by the Ovarian Cancer Association Consortium (OCAC), which comprises 66,450 European individuals, including 25,509 and 40,941 cases and controls, respectively, from 14 nations. Mendelian randomization was carried out to assess the relationship between plasma proteins and ovarian cancer risk.

Co-localization analyses were performed to investigate whether the proteins and ovarian cancer shared incidental and causal variants. Additionally, the team performed phenome-wide Mendelian randomization to evaluate protein relationships by phenotype.

Target drug databases (Clinical Trial, Therapeutic Target Database, and Drugbank) and previously published druggable gene lists were searched to assess the druggability of the identified proteins and validate drugs.

To assess drug formability potentials, the proteins were categorized as follows: (i) approved (≥1.0 drugs that target particular proteins authorized for use); (ii) in clinical studies (target drug under research in clinical trials); (iii) pre-clinical (target drug in preclinical research); and (iv) druggable (proteins not identified in drug databases, but listed as medicinal targets).

Results

In total, 44 genetically estimated proteins were related to ovarian cancer risk, of which 17 elevated ovarian cancer risk, such as microfibril-associated protein 2 (MFAP2), C-type lectin domain containing 11A (CLEC11A), protein disulphide isomerase family A member 3 (PDIA3), thymidine phosphorylase (TYMP), serine peptidase inhibitor Kazal type 1 (SPINK1), and angiopoietin-like 1 (ANGPTL1).

Other proteins related to increased ovarian cancer risk were interleukin 1 receptor type 1 (IL1R1), Dickkopf WNT signaling pathway inhibitor 2 (DKK2), plasminogen activator, urokinase (PLAU), delta-like non-canonical Notch ligand 1 (DLK1), interleukin 6 cytokine family signal transducer (IL6ST), cell adhesion associated, oncogene regulated (CDON), leucine-rich repeat containing 15 (LRRC15), semaphorin 4D (SEMA4DTNF alpha-induced protein 6 (TNFAIP6), Fc gamma receptor IIb (FCGR2B), and aldo-keto reductase family 1 member A1 (AKR1A1).

Among the proteins investigated, 27 lowered ovarian cancer risk, including retinoic acid receptor responder 1 (RARRES1), sialic acid-binding Ig-like lectin 9 (SIGLEC9), transmembrane protein 132A (TMEM132A), serine peptidase inhibitor, Kunitz type 3 (SPINT3), contactin 2 (CNTN2), transforming growth factor β-induced (TGFBI), hepatitis A virus-cell receptor 2 (HAVCR2), and HGF activator (HGFAC).

The list included glutathione S-transferase alpha 1 (GSTA1), GRAM domain containing 1C (GRAMD1C), triggering receptor expressed on myeloid cells like 2 (TREML2), copine 1 (CPNE1), N-acyl sphingosine amidohydrolase 2 (ASAH2), MAPK activated protein kinase 2 (MAPKAPK2), C-C motif chemokine ligand 25 (CCL25), prolyl endopeptidase (PREP), and protein O-fucosyltransferase 1 (POFUT1).

In addition, proteins such as carbonic anhydrase 10 (CA10), netrin G1 (NTNG1), carbonic anhydrase 8 (CA8), calcium voltage-gated channel auxiliary subunit alpha2delta 3 (CACNA2D3), mannose receptor C type 2 (MRC2), mannosidase alpha class 1C member 1 (MAN1C1), retinol-binding protein 4 (RBP4), interleukin 10 receptor subunit beta (IL10RB), calcium binding and coiled-coil domain 2 (CALCOCO2), and glycoprotein V platelet (GP5) also lowered ovarian cancer risks.

Bayesian co-localization showed that RBP4, GRAMD1C, PDIA3, PLAU, POFUT1, MFAP2, DKK2, and MAN1C1 shared variants with ovarian cancer.

Most proteins showed causal influences on ovarian cancer risk factors, such as anthropometric characteristics (height, weight, body mass index, fat mass, and waist or hip circumference), metabolic characteristics (cholesterol, triglycerides, calcium, and urate), and female reproductive characteristics (sex hormones, menstrual cycle). In addition, a few proteins impacted the basal metabolic rate, birth weight, blood cell counts, and breast cancer.

Further, genetically determined PLAU, MAPKAPK2, and MFAP2 elevated rheumatoid arthritis risk in diseases involving the breast and non-female genital tissues. Genetically determined DLK1 and CALCOCO2 elevated diabetes type 2 risk.

TGFBI and CPNE1 increased breast cancer risk, whereas PDIA3 and MAN1C1 decreased it. GRAMD1C lowered pharyngeal cancer risk, CA8 lowered the risk of skin cancer, and CALCOCO2 increased uterine cervical carcinoma risk. CA8 lowered ovarian cancer risk but increased varicose vein risk.

MAPKAPK2 lowered ovarian cancer risk but increased that of ulcerative colitis and inflammatory bowel syndrome, and CPNE1 increased asthma risk. Among the proteins, nine (21%) were identified as drug targets under clinical trial research (HAVCR2, SEMA4D, MAPKAPK2, CACNA2D3, FCGR2B, PREP, PLAU, DLK1, and TYMP).

Nine proteins have already been approved as drug targets, including CACNA2D3, RARRES1, and GSTA1; however, none were approved as drug targets for ovarian cancer.

Overall, the study findings showed causal associations between genetically estimated proteins and ovarian cancer risk, which could be promising therapeutic targets for ovarian cancer.

*Important notice: Preprints with The Lancet / SSRN publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.