Date of Award


Document Type


Degree Name

Master of Medicine (MMed)

First Supervisor/Advisor


Second Supervisor/Advisor


Third Supervisor/Advisor



Pathology (East Africa)


Introduction: Breast Cancer (BC) is the most diagnosed cancer among women worldwide and in Kenya. It is a highly heterogeneous disease requiring patient specific treatment and strategies to monitor response to therapy, especially in the neoadjuvant (NAC) setting. Currently, there are no reliable markers in use that can predict response to chemotherapy among patients receiving NAC. Tumour Infiltrating Lymphocytes (TILs) and Cell free DNA (cfDNA) are biomarkers that could possibly bridge this gap. However, limited studies have been carried out on the African population. With the increasing mortality rates attributable to BC in Sub-Saharan Africa, coupled with the growing preference for NAC use, there is need for a biomarker capable of predicting response to NAC in order to potentially improve clinical outcomes. The lack of soild evidence to support the use of TILs and cfDNA in the clinical setting has limited their application exclusively to the research setting.

Aims: This study aimed at determining the association of immune and genomic biomarkers and response to neoadjuvant chemotherapy in breast cancer.

Methodology: To evaluate TILs in response to NAC we conducted a retrospective study where we retrieved archived paraffin embedded tissue blocks from the Aga Khan University Hospital Nairobi (AKUHN). For this purpose, 99 cases diagnosed between January 2012 and March 2022 were utilised. The diagnostic core biopsies for these cases were evaluated visually for stromal TILs, compared to the resection specimen after NAC. To determine associations between TILs and response to chemotherapy we used Fisher’s Exact Test and unconditional logistic regression. To determine if %cfDNA from plasma samples was higher after NAC, we prospectively enrolled newly diagnosed BC patients from AKUHN, Kisii County Referral Hospital & Oasis Healthcare Group’s Oncology Clinics between March 2021 and March 2022. Their blood samples were drawn at baseline, at mid chemotherapy and before the last chemotherapy cycle. These samples were then evaluated for cfDNA. iv

Results: Ninety-nine cases were available for analysis to determine TILs association with NAC. Of these, 86.9% were Invasive Ductal Carcinoma (NST). Histologic Grade 2 and 3 tumours accounted for 93.9% of the total and 44.4% were TIL High. Those that has Extensive Residual Disease (RCB Class 3) after NAC accounted for 35.4% of all the cases. An association was found between TIL concentration and response to chemotherapy (p<0.001). Adjusted for age, sex, ER status, menopausal status, and smoking, TIL High patients had a 70.6% reduction in the odds of residual disease compared to TIL Low (p=0.036). No association was found between TIL concentration and clinicopathological and sociodemographic factors (age, pre-treatment tumour size, menopausal status, smoking and alcohol) (p>0.05). Nineteen patients were analysed to determine plasma cfDNA levels. We had a 100% success in detection of cfDNA. Median levels of %cfDNA were 39, 79 and 76 from timepoint 1, 2 and 3 respectively. The mean levels of %cfDNA were seen to steadily rise from timepoint 1 to timepoint 2 with a slight decrease at timepoint 3 but remaining higher than that at timepoint 1. cfDNA levels were lower in patients who attained pathologic complete response on resection compared to those who did not.

Conclusion: We observed an association between higher TIL levels at diagnosis and the response to NAC, supporting routine reporting of this marker in diagnostic core biopsies in breast cancer. cfDNA can be reliably extracted from plasma samples of breast cancer patients undergoing NAC and the levels are seen to rise with increased tumour death occasioned by chemotherapy. Patients with residual disease have higher levels of this biomarker at the end of NAC. Larger prospective studies investigating the association of this molecule to response to chemotherapy are necessary.

Included in

Pathology Commons