Maria Dimitrova¹, Petya Milushewa¹, Elina Petrova¹, Desislava Mihaylova², Nezabravka Tzvetanova² Guenka Petrova¹, Ivaylo Petrov³, Jelena Duza³

1. Department of Organization and economics of pharmacy, Faculty of pharmacy, Medical University of Sofia
2. Sqilline, Danny Platform - analytics platform for real-world data, Sofia, Bulgaria www.sqilline.com
3. Novartis

Breast cancer is one of the most common oncologic diagnoses that women receive worldwide. Of these, triple negative breast cancer (TNBC) accounts for 12-15% of all breast cancers and is the most aggressive.[1] For women with TNBC, time to diagnosis and access to treatment are two crucial factors that affect survival rates. Access to treatment for TNBC patients has often been complicated by the fact that there are few treatments options available as the cancer is difficult to treat and with that, poor treatment outcomes.[2] Personalised approaches to the treatment of TNBC are beginning to bear promising results. Treatments are becoming more targeted and precise through the use of novel assays and therapies that target molecular and immune mechanisms. Individualised approaches to patients based on accurate staging, gene mutation identification, molecular subtype determination, treatment regimen choice and assessment of risk of recurrence or progression are increasing the chance of more promising treatments, especially for challenging types of cancer such as TNBC.

Behind much of the progress in developing treatment options for TNBC are digital technologies, such as artificial intelligence (AI) which provide essential real-world evidence to demonstrate the effectiveness of novel therapies. Their potential can often be difficult to prove if structured access to patient data is unavailable which is often the case. In Bulgaria, a one-year real-world retrospective study on the patho-histological status and treatment of a representative cohort of patients with TNBC was performed. It was based on data extracted from the electronic AI platform Sqilline - Danny Platform (www.sqilline.com) which collected massive amounts of real-world data and analysed unstructured information from patients’ records. The records were from all university hospitals and major oncology centres and covered almost all available cancer cases in the country. Within the system, complex Deep-learning Natural Language Processing (DLNLP) was developed in order to make data ready for analyses of patient treatment, drug efficacy and increase efficiency and value of patient recruitment for clinical trials.

During the observed period (January 2019 - December 2019), 6880 breast cancer patients from eight major oncology hospitals were included in the database. The average age of all women was 60 years. 234 (3.4%) of them were diagnosed with TNBC and 10% were at an unknown stage. Due to lack of tumor staging data in the primary documentation, the majority of the patients were assigned to chemotherapy (84%) of which 35% were on adjuvant. Most changes in the therapy were observed in the neo-adjuvant group. The results from this study provided evidence that the treatment patterns of TNBC, and changes in therapy were in compliance with international guidelines.[3] Importantly, it also identified less patients with TNBC than the frequencies reported in international epidemiological studies. This might be attributed to a lack of funding of necessary tests or insufficient data in patients record.

For healthcare systems, gains in efficiency and resources are realised when ineffective treatments are avoided in favour of innovative, precise treatments which reduce the disease burden on the entire health system. Dynamic patient registers are valuable tools in this effort, as they can be used to perform a real-world study of treatment patterns. They can also guide policy makers in improving, time to diagnosis with implementing screening practices, diagnostic guidelines and committing further resources, through identifying if the reported frequencies are far from the international average.  This study provides another example that demonstrates how a value-based healthcare system is a win-win for both patients and the health care system. The extraction and systematisation of information from medical records are of a great significance for the improvement of diagnosis, treatment, survival prediction, resource allocation and decision making. Personalised treatments, combined with the power of AI driven analytics of health data, have proven to have a transformative impact on both patients and health systems, calling for further investments in this type of synergy by public and private stakeholders.

Original article source: Maria Dimitrova, Petya Milushewa, Elina Petrova, Desislava Mihaylova, Nezabravka Tzvetanova & Guenka Petrova (2021) Triple negative breast cancer in Bulgaria: epidemiological data and treatment patterns based on real world evidence and patient registries, Biotechnology & Biotechnological Equipment, 35:1, 551-559, DOI: 10.1080/13102818.2021.1903338

[1] Maria Dimitrova, Petya Milushewa, Elina Petrova, Desislava Mihaylova, Nezabravka Tzvetanova & Guenka Petrova (2021) Triple negative breast cancer in Bulgaria: epidemiological data and treatment patterns based on real world evidence and patient registries, Biotechnology & Biotechnological Equipment, 35:1, 551-559, DOI: 10.1080/13102818.2021.1903338

[2] Houts AC, Olufade T, Shenolikar R, et al . Treatment patterns, clinical outcomes, health resource utilization, and cost in patients with BRCA-mutated metastatic breast cancer treated in community oncology settings. Cancer Treat Res Commun. 2019;19:100121.

 [Crossref], [PubMed], [Google Scholar]

[3] Dimitrova, M., Lakic, D., Petrova, G., Bešlija, S., & Culig, J. (2020). Comparative analysis of the access to health-care services and breast cancer therapy in 10 Eastern European countries. SAGE open medicine, 8, 2050312120922029.