Integrative Clinical, Molecular, and AI-Guided Analysis of Treatment Patterns and Druggable Targets in Breast Cancer: A Real-World Study from Eastern Saudi Arabia.
Year: 2026
Type: report
Study design: Observational Study
Abstract: Tahani Al-Qurashi 1,2*, Mashail Ahmad Albargi1,2*, Kanza Rashid1*, Abdulkareem AlGarni1,2, Yaqob Taleb2,5,6*, Nasser Mohammed AlQahtani7, Muhammad Farooq Sabar4, Muhammad Alshuaibi7, Essa Al Mansour7, Mashael AlShuker7, Azfar Athar Ishaqui8, Nawaf, Alanazi1,2, Rizwan Naeem9, Masood A. Shammas10, Sohail Rao11@, Zafar Iqbal1,2,3*@ 1Genomic & Experimental Medicine Group (GEM), Quality Assurance & Accreditation Unit (QAAA) & Clinical Laboratory Sciences Program (CLSP), College of Applied Medical Sciences (CoAMS-A), King Saud Bin Abdulaziz University for Health Sciences (KSAU-HS)/ King Abdullah International Medical Research Centre-Eastern Region (KAIMRC-ER)/ King Abdulaziz Medical City (KAMC), National Guard Health Affairs (NGHA), Al-Ahsa, Kingdom of Saudi Arabia (KSA); 2King Abdulaziz Hospital, KAMC, NGHA, Al-Ahsa, KSA; 3Center of Excellence for Research in Oncology & Personalized Medicine and Biotechnology, INNOVACORETM Center for Research & Biotechnology, 6918 Camp Bullis Rd. San Antonio, TX 78256, USA; 4KAIMRC, Al-Ahsa, Saudi Arabia. 5College of Applied Medical Sciences (CoAMS-A), King Saud Bin Abdulaziz University for Health Sciences (KSAU-HS); KAIMRC-ER, KAMC, NGHA, Al-Ahsa, Saudi Arabia; 7Division of Adult Oncology, Department of Medicine, King Abdulaziz Hospital / KAIMRC-ER, Al-Ahsa, Saudi Arabia,8Department of Clinical Pharmacy, College of Pharmacy, King Khalid University, Abha, Saudi Arabia; 9Montefiore Medical Centre & Albert Einstein College of Medicine, New York, NY,10 Department of Medical Oncology, Harvard Dana Farbar / Harvard Cancer Institute (DFCI), University of Harvard, Boston, Massachusetts, USA; 11President, Chief Executive Officer & Founding Member, INNOVACORETM Center for Research & Biotechnology, 6918 Camp Bullis Rd. San Antonio, TX 78256, USA. Correspondence: Prof. Dr Zafar Iqbal, E-mail: [email protected], [email protected]; Prof. Dr. Sohail Rao, E-mail:[email protected] ; *Equal contribution. Abstract Background Breast cancer treatment selection increasingly relies on integration of clinicopathological characteristics, molecular profiling, and advanced analytical approaches; however, translation of precision oncology into real-world practice remains inconsistent, particularly in non-Western populations. Aim To characterize real-world breast cancer treatment patterns and actionable molecular targets in Saudi Arabia using an integrative clinical, molecular, and artificial intelligence (AI)–guided analytical framework. Methods This was a retrospective cross-sectional study of 221 breast cancer patients treated at King Abdulaziz Medical City, Al-Ahsa. Demographic, clinical, treatment, hormone receptor status, molecular subtype, and genetic testing data were extracted from electronic medical records. Associations between estrogen receptor (ER), progesterone receptor (PR), molecular subtypes, and treatment modalities were evaluated using traditional statistical methods. In order to complement classical statistics and improve the interpretability of treatment–biology relationships, analyses of effect size, unsupervised learning of patterns, and AI-guided analyses were used. Results Chemotherapy was the most common treatment modality, followed by radiotherapy and hormonal therapy, with targeted therapy using Olaparib administered to a smaller group of patients. ER and PR positivity were significantly related to the use of hormonal therapy (P < 0.05), thus confirming a strong receptor-driven treatment paradigm, whereas for chemotherapy and radiotherapy, their use was not statistically significantly dependent on receptor status. Molecular testing in a clinically selected subset revealed PIK3CA to be the most common actionable mutation, followed by alterations in the breast cancer susceptibility genes BRCA1 and BRCA2. Luminal A was the most common molecular subtype, with a significant presence of Luminal B, HER2-enriched, and basal-like tumors, indicating subtle biological heterogeneity. AI-guided effect-size analyses in breast cancer demonstrated the magnitude of receptor–treatment associations identified and revealed different association profiles for targeted treatment, highlighting the complexity and underutilization of precision treatment allocation within real-world clinical practice. Conclusions This integrative real-world analysis shows that breast cancer treatment patterns in Saudi Arabia are still largely determined by hormone receptor status and molecular subtype, while actionable genomic alterations are underutilized. Additional insights into real-world oncology data and the growing role of information flow in advancing precision breast cancer care are emerging through the use of AI-guided analytical approaches. Keywords: Breast cancer; Real-world evidence; Artificial intelligence; Druggable mutations; Precision oncology.
Language: English
Subfield: Cancer Research
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7134832588