Abstract
Accurate and early diagnosis is vital in healthcare, improving patient outcomes by enabling timely intervention and better disease management. Early detection is crucial for non-small cell lung cancer (NSCLC), the most common type of lung cancer, where it is closely linked to improved outcomes. Traditional tissue biopsies, though informative, are invasive and impractical for routine disease monitoring. Pulmonary hypertension (PH), with its complex subtypes, is diagnosed using invasive right heart catheterization (RHC), which may not fully capture disease variations. These gaps highlight the need for non-invasive biomarker-based diagnostics to improve disease understanding and patient care.
This thesis explores the potential of using platelet RNA profiles for the early, less invasive detection of lung diseases, particularly NSCLC and PH. The study centers on liquid biopsies, using tumor-educated platelets (TEPs) as a diagnostic tool. Initially, the research focuses on NSCLC, aiming to refine detection methods and investigate platelet RNA’s role in early diagnosis. Later, it addresses PH, showing that platelet RNA profiling can differentiate PH patients from asymptomatic controls. With further optimization, it can also distinguish between pre- and post-capillary PH, improving diagnosis accuracy. The research extends to chronic thromboembolic pulmonary hypertension (CTEPH), aiming to enhance early detection and disease management.
Chapter 2 investigates thromboSeq, a platelet RNA-sequencing protocol, to assess its diagnostic potential in cancer. An optimized support vector machine (SVM) algorithm, enhanced by particle swarm optimization (PSO), demonstrated thromboSeq’s ability to distinguish 18 cancer types from controls with high specificity. The study analyzed over 2,400 samples and identified a panel of 493 platelet RNA biomarkers. The algorithm achieved 99% specificity in asymptomatic individuals, detecting cancer in 64% of cases. Additionally, thromboSeq accurately predicted the primary tumor site with 85% accuracy, providing critical information for cancer treatment planning.
Chapter 3 narrows the focus to NSCLC, applying thromboSeq to develop two diagnostic tests: HighSens and HighSpec. HighSens maximizes sensitivity, detecting 95% of NSCLC cases, including 80% of early-stage cases. HighSpec prioritizes specificity, accurately identifying 94% of non-cancer controls. These tests offer a flexible framework for NSCLC screening, balancing sensitivity and specificity based on clinical needs.
Chapter 4 shifts the focus to PH, identifying a platelet RNA signature that differentiates PH patients from controls. The study developed an RNA panel that detected PH with 93% sensitivity and 77% accuracy. This RNA-based test holds promise as a non-invasive diagnostic tool, though further research is needed to confirm its clinical value.
Chapter 5 explores differentiating pre- and post-capillary PH using platelet RNA profiles. An SVM-PSO algorithm selected an RNA panel that distinguished between these subtypes with 100% sensitivity and 60% specificity. This method could reduce the need for invasive procedures like RHC, helping tailor treatment to individual patient profiles.
Finally, Chapter 6 examines CTEPH detection using platelet RNA profiles. A PSO-selected panel of 26 RNAs distinguished CTEPH patients from controls with 100% sensitivity and 50% specificity. While promising, these findings require further validation in larger patient cohorts.
In conclusion, platelet-derived RNA profiles offer a non-invasive alternative for diagnosing lung diseases like NSCLC and PH. Tests like HighSens and HighSpec contribute to more accurate diagnostic protocols, improving early detection and intervention. However, further validation and a deeper understanding of the biological mechanisms are needed to confirm these biomarkers’ clinical utility. Future studies, including broader analyses of platelet RNA, may refine these diagnostic tools, advancing early detection and management of lung diseases.
Original language | English |
---|---|
Qualification | PhD |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 27 Nov 2024 |
Print ISBNs | 9789465064819 |
DOIs | |
Publication status | Published - 27 Nov 2024 |