Protein signatures in uterine aspirates to improve diagnosis of endometrial cancer. The CEMARK project

Elena Martínez García

Dr. Jaume Reventós Puigjaner Dr. Eva Colás Ortega Dr. Antonio Gil Moreno

Biomedical Research in Gynecology

Endometrial cancer (EC) is the fourth most common cancer in women in developed countries, with 61,380 new cases and 10,920 deaths estimated for 2017 in the United States 1. Unfortunately, death rates for EC have increased in recent years, mainly as a consequence of the growth and aging of the population. Currently, EC is the sixth leading cause of cancer death in women. Patients diagnosed at early stages of the disease present an overall 5-year survival rate of 95%. However, 30% of EC patients are still diagnosed at an advanced stage of the disease, which is associated with a drastic decrease in the 5-year survival rate 1. Consequently, early diagnosis is a major issue to appropriately manage EC and to decrease mortality associated with this disease. Abnormal uterine bleeding (AUB) is one of the most common symptoms for gynecological consultations. Although only 5-10% of women with AUB will have an EC, women with this symptom are alerted to consult a specialist since AUB occurs in 90% of EC cases 2. This represents that a large number of women with benign disorders presenting AUB need to undergo a diagnostic process to rule out EC. Current diagnosis is achieved by the histopathological examination of an endometrial biopsy, which is preferably obtained by aspiration from the uterine cavity (i.e., uterine aspirate). This procedure can be performed in the clinician's office and it is minimally invasive, cost-effective and well tolerated by patients. Unfortunately, the limited number of cells available for examination in these biopsies is associated with two important limitations: i) an average of 22% of patients do not receive a final diagnosis and need to undergo more invasive sampling methods 3; and ii) up to 50% of EC patients receive an incorrect assignment of the EC histotype and grade, which can compromise the optimal surgical treatment 4. In this context, the main goal of the doctoral thesis "Protein signatures in uterine aspirates to improve diagnosis of endometrial cancer. The CEMARK project" is the identification of protein biomarkers in the fluid fraction of those uterine aspirates, independent of the cellular content of these samples, to overcome those limitations. In order to achieve this main objective, the doctoral thesis followed the sequential phases of the biomarker identification pipeline. First, an extensive literature review was performed and a list of 506 proteins associated with EC was generated. Then, we demonstrated that those proteins, mainly studied by immunohistochemistry at the tissue level, could be measured in minimally-invasive uterine aspirate samples by highly multiplexing targeted mass spectrometry (MS)-based proteomics. Afterwards, we designed a stepwise workflow to reduce the initial list of 506 candidate biomarkers down to 52 proteins. These proteins were analyzed in the fluid fraction of uterine aspirates from 20 EC patients and 18 non-EC controls with AUB by the parallel reaction monitoring acquisition (LC-PRM), the latest generation of targeted acquisition method used on a high resolution accurate mass spectrometer. The differential expression of 26 proteins was observed, and among them ten biomarkers showed a high sensitivity and specificity to discriminate between EC and non-EC patients (AUC > 0.9). Furthermore, we demonstrated the benefits of LC-PRM acquisition for the accurate quantification of a high number of candidate biomarkers in complex clinical samples. The same 52 proteins were then measured in the uterine aspirates from 116 patients (69 EC patients and 47 controls with EC suspicion) by LC-PRM. The differential level of the previous 26 proteins was validated in this independent cohort of patients. The combination of MMP9 and KPYM exhibited 94% sensitivity and 87% specificity for detecting EC cases. This panel perfectly complemented the standard histopathological diagnosis, achieving 100% of correct diagnosis in this dataset. Nine proteins were significantly increased in endometrioid EC (n=49) compared to less common but more aggressive serous EC (n=20). The combination of CTNB1, XPO2 and CAPG achieved 95% sensitivity and 96% specificity for the discrimination of these EC histological subtypes. In order to facilitate the implementation of these biomarker signatures in clinical practice, the translation of the LC-PRM results to commercially available ELISA kits, and the simplification of the analytical assay and the sample preparation was then evaluated. Our next objective was to bring these promising results from the bench to the clinic as an in vitro diagnostics (IVD) test. In this regard, apart from the scientific work, other fields are equally important in order to successfully introduce a product in the market. During the last year of my doctoral thesis I participated in several valorization activities in parallel to the experimental work. As shown in Figure 1, the results generated in this thesis have been protected with two European patents (PCT/EP2017/057635 for the diagnostic biomarkers and EP17382483.0 for the prognostic biomarkers) and a "freedom to operate" analysis of the proteins that form the panels has been carried out to evaluate if an IVD test including those proteins can be produced and commercialized without infringing third party intellectual property rights Importantly, I have participated in several mentoring and networking activities offered by several grants that the project has received, including the "CaixaImpulse program" promoted by the Caixa Foundation and Caixa Capital Risc, and the "IDEA2 global program" organized by the Massachusetts Institute of Technology (MIT) in Boston (USA). We have also initiated market research activities such as interviews with stakeholders (patients, gynecologists, hospital managers, pharmaceutical companies, etc.) and pharmaco-economic studies in the different countries where the product would be introduced. Moreover, we were selected to participate in the MAP EADA program organized by the EADA business school and "Agencia para la Competitividad de la Empresa" that, together with other grants that the project received, helped us in the design of the business plan to be presented to investors and possible licensees. These activities and grants also helped us in the planning of the next steps to follow.

Cancer is today one of the most common causes of death in the world, with over 8 million annual deaths 5. Early detection, i.e., finding tumors when they are still confined to the organ of origin, is crucial to favor complete resection of the tumor and to ensure a more effective treatment. The World Health Organization (WHO) has emphasized that the combination of current therapies with effective early detection would significantly impact survival 6. In this context, biomarkers have become a keystone of medical care to favor early detection and IVD testing is now an indispensable tool in the clinical environment 7. IVD is the largest sector of medical technology worldwide followed by cardiology and orthopedics, and it is expected to undergo the greatest sales growth of the medical technology sector during the period 2014-2020 8. EC is the fourth most common cancer in women in developed countries and its incidence and mortality rates are increasing annually worldwide. Moreover, around 94 million women in the world have an AUB annually, a symptom present in 90% of EC cases and thus, a large number of women with benign disorders presenting AUB need to undergo a diagnostic process to rule out EC. Despite the enormous efforts and investments made to identify protein biomarker signatures to improve early diagnosis of this disease, no protein has yet reached clinical use. The two proteomic signatures identified in the fluid fraction of uterine aspirates in this doctoral thesis are expected to have a greater impact on EC diagnosis. From one side, the diagnostic signature would perfectly complement the current diagnostic procedure based on the histological examination of the limited cellular fraction of these samples, precluding the use of subsequent invasive sampling methods such as dilatation and curettage or hysteroscopy. On the other hand, the prognostic signature would help to assist in the prediction of the optimal surgical treatment of EC patients. Altogether, the implementation of these signatures is expected to improve the management of EC patients and substantially reduce disease-related healthcare costs. Beyond these important achievements in the field of EC diagnosis, we believe that uterine aspirates might also be a novel promising source of protein biomarkers for other gynecological diseases such as endometriosis and ovarian cancer, two diseases that lack an effective diagnostic procedure. The optimization of the sample preparation and MS analysis, together with the stepwise proteomic workflow described in this thesis, will translate into a more efficient identification of clinically useful biomarker signatures in these complex uterine aspirate samples.

Endometrial Cancer (EC); Abnormal Uterine Bleeding (AUB); Uterine Aspirate; Protein Biomarkers; Mass Spectrometry (MS)-based Proteomics; LC-PRM; Diagnostic Biomarkers; Prognostic Biomarkers; In Vitro Diagnostics (IVD) Test; Patent Applications; Technology Transfer; Clinical Implementation; Multidisciplinary Collaboration; Market Research; Business Development; MMP9; KPYM; CTNB1; XPO2; CAPG