Digistain Prognostic Score Development: Infrared Spectroscopy to Guide Chemotherapy Decisions
File(s)
Author(s)
Al-Khalili, Zamzam
Type
Thesis
Abstract
Digistain is a mid-infrared technique that measures absorption at wavelengths corresponding to phosphate moieties in the nucleus and amide moieties in the cytoplasm. It produces an objective metric, named Digistain Index (DI), that inherently correlates with Nuclear-to-Cytoplasmic Ratio (NCR), a well-recognized measure of solid malignancies including breast cancer. The aim of this thesis is to develop a prognostic score that evaluates cancer risk and guides adjuvant chemotherapy treatment by incorporating DI with histological and pathological factors.
A cohort of 1184 breast cancer patients, who did not receive adjuvant chemotherapy, was selected for this study. Spectral measurements were conducted using a Fourier-Transform Infrared (FTIR) spectrometer on tissue samples collected from the identified patient cohort. Then, the DI was extracted from the spectral data using a computational code. Subsequently, the Digistain Prognostic Score (DPS), incorporating DI with tumour grade and lymph node status, was developed and validated using survival analysis tools, mainly Cox proportional hazards regression and Kaplan-Meier curves, in addition to Receiver Operating Characteristic (ROC) curves.
The resulting DPS showed statistically significant prognostic power and stratification capability for Overall Survival (OS) and Recurrence-Free Survival (RFS) at 5- and 10- years. The prognostic power was evidenced by the strong hazard ratios (> 2) and significant p-values ( 0.0001) obtained from Cox proportional hazards regression, in addition to the consistently high Area Under the Curve (AUC) values from ROC curves (0.7 - 0.8). The stratification capability was demonstrated by the clear distinction between high- and low- risk groups represented by Kaplen-Meier curves.
These results suggest that DPS holds substantial clinical utility as it facilitates individualized patient care by guiding treatment plans, particularly in regards to adjuvant chemotherapy.
A cohort of 1184 breast cancer patients, who did not receive adjuvant chemotherapy, was selected for this study. Spectral measurements were conducted using a Fourier-Transform Infrared (FTIR) spectrometer on tissue samples collected from the identified patient cohort. Then, the DI was extracted from the spectral data using a computational code. Subsequently, the Digistain Prognostic Score (DPS), incorporating DI with tumour grade and lymph node status, was developed and validated using survival analysis tools, mainly Cox proportional hazards regression and Kaplan-Meier curves, in addition to Receiver Operating Characteristic (ROC) curves.
The resulting DPS showed statistically significant prognostic power and stratification capability for Overall Survival (OS) and Recurrence-Free Survival (RFS) at 5- and 10- years. The prognostic power was evidenced by the strong hazard ratios (> 2) and significant p-values ( 0.0001) obtained from Cox proportional hazards regression, in addition to the consistently high Area Under the Curve (AUC) values from ROC curves (0.7 - 0.8). The stratification capability was demonstrated by the clear distinction between high- and low- risk groups represented by Kaplen-Meier curves.
These results suggest that DPS holds substantial clinical utility as it facilitates individualized patient care by guiding treatment plans, particularly in regards to adjuvant chemotherapy.
Version
Open Access
Editor(s)
Phillips, Chris
Date Issued
2025-08-08
Date Awarded
01/02/2025
Citation
2025
License URL
Advisor
Phillips, Chris
Sponsor
Sultan Qaboos University
Publisher Department
Department of Physics
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)