Abstract PS3-09-17: De-escalation of systemic therapy for breast cancer in the Indian Setting: A Proteomics and artificial intelligence based Approach to Risk Stratification beyond Ki67, Tumour Grade, and NPI for HR+/HER2-Negative Breast Cancer

Abstract Objective Determining the need for adjuvant chemotherapy in early-stage hormone receptor-positive (HR+), HER2-negative breast cancer remains a clinical challenge, particularly in stage I/II disease. This study evaluates the prognostic performance of the CanAssist Breast (CAB) test a proteomics-based risk stratification tool against conventional markers such as tumour grade, Ki67 index, and Nottingham Prognostic Index (NPI), with a focus on validation in the Indian population. Notably, the CAB test is cost-effective, priced below USD 1000. Materials and Methods A retrospective observational analysis was conducted on over 160 patients diagnosed with stage I/II HR+/HER2-negative invasive breast carcinoma, who underwent the CAB test at HCG Cancer Centre, Ahmedabad up to June 2025. Patients were classified into low- or high-risk groups using a CAB score cut-off of 15.5. Risk categorization was compared with histopathological parameters: tumour grade (1, 2, 3), NPI (low, intermediate, high), and Ki67 index (low, moderate, high). Associations between CAB scores and traditional prognostic markers were assessed using Chi-square statistical analysis. Ongoing follow-up is being carried out to update patient outcomes and recurrence data. Results Over 60% of patients were categorized as low-risk based on the CAB score. A statistically significant association was observed between CAB scores and tumour grade: most grade 3 tumours exhibited high CAB scores, whereas grade 2 tumours predominantly showed low CAB scores. Similarly, a significant correlation was found between NPI and CAB scores, with most intermediate NPI cases falling into the low-risk CAB group. Interestingly, a substantial proportion (>50%) of patients with moderate or high Ki67 indices also exhibited low CAB scores. Additionally, a subset of node-positive (N1) patients fell into the CAB low-risk category. Importantly, none of the low-risk patients who were spared chemotherapy had reported recurrence at the time of the last follow-up. Follow up ananlysis is ongoing currently. Final results will be further substantiated upon completion of follow-up data analysis. Conclusion The CanAssist Breast test effectively refines risk stratification in early-stage HR+/HER2-negative breast cancer, especially in patients with ambiguous traditional markers (e.g., grade 2, intermediate NPI, moderate Ki67). By reducing the reliance on adjuvant chemotherapy in low-risk groups, the CAB test helps avoid unnecessary treatment-related toxicities. These findings support the utility of CAB as a cost-efficient, clinically meaningful decision-making tool in the Indian oncological landscape. Citation Format: T. N. Shah, V. Devanhalli, M. Shah, M. Bakre. De-escalation of systemic therapy for breast cancer in the Indian Setting: A Proteomics and artificial intelligence based Approach to Risk Stratification beyond Ki67, Tumour Grade, and NPI for HR+/HER2-Negative Breast Cancer [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS3-09-17.