Back to results

University of Exeter

Evaluating the Diagnostic Value and Molecular Mechanisms of Thrombocytosis in Lung Cancer

Abstract

dc:description

This multidisciplinary thesis investigates the role of thrombocytosis as an indicator of lung cancer in primary care, integrating electronic health record analysis with mechanistic in vitro investigations to evaluate its diagnostic value and underlying biology. The first and largest phase, evaluates the diagnostic value of thrombocytosis for lung cancer detection in primary care, using large-scale data from the Clinical Practice Research Datalink, Hospital Episode Statistics Admitted Patient Care, the National Cancer Registration and Analysis Service, and the UK Biobank. The second phase investigates the cellular processes underpinning the association between thrombocytosis and lung cancer in vitro. A systematic review and meta-analysis of 36 studies identified thrombocytosis as a common feature across histological subtypes (22 to 36% prevalence), although substantial inter-study heterogeneity (I² = 93.5%) limited statistical inferences across groups. Building on these findings, large primary care analyses quantified the prevalence, timing, and clinical relevance of pre-diagnostic thrombocytosis. In the year before diagnosis, thrombocytosis was observed in 13.3% of lung cancer cases compared with 1.4% of age-, sex-, and general practice-matched controls. The highest prevalence occurred in patients with squamous cell carcinoma (SCC, 16.9%) relative to adenocarcinoma (ADC, 10.7%, p<0.001) or small cell lung cancer (SCLC, 11.8%. p=0.001). Although smoking and chronic obstructive pulmonary disease were independently associated with thrombocytosis, adjustment for these factors did not attenuate the strong association with lung cancer (adjusted odds ratio (OR) 8.9). Pre-diagnostic thrombocytosis was associated with reduced likelihood of early-stage diagnoses for patients with ADC and SCC, but not SCLC. Longitudinal analyses indicated that platelet counts begin 359 days before diagnosis (95% CI 331 to 388 days) for all lung cancer patients, with minimal variation by histological subtype. Ethnicity influenced baseline platelet distributions and thrombocytosis prevalence, with Black patients exhibiting the lowest median platelet counts and prevalence. However, the diagnostic association remained broadly consistent across White, Asian, and Black populations, with ORs of 5.0 (95% CI 4.7 to 5.2), 4.5 (95% CI 2.9 to 6.9), and 7.3 (95% CI 4.6 to 11.7), respectively (interaction p-value=0.44). A lung cancer prediction model incorporating platelet count alongside other full blood count parameters, including neutrophil and lymphocyte counts, demonstrated substantially improved discrimination compared with models based on age and smoking status alone, achieving AUCs of 0.86 (95% CI 0.85 to 0.86) in development and internal validation, and 0.83 (95% CI 0.80 to 0.86) in external validation. Mechanistic in vitro experiments demonstrated that lung cancer cells can directly induce proplatelet formation in megakaryocytes, the precursors to platelets. This effect was significantly greater in megakaryocytes exposed to lung cancer conditioned medium than those cultured in supplemented medium containing established drivers of megakaryopoiesis and thrombopoiesis (p<0.001). These findings provide evidence of a direct mechanistic link between lung cancer cells and platelet biogenesis and suggest activation of a non-canonical pathway thrombopoietic pathway. Collectively, this thesis establishes thrombocytosis as a clinically informative and broadly generalisable indicator of lung cancer across histological subtypes and different population groups in primary care. It provides mechanistic evidence supporting a direct biological link between lung cancer activity and platelet production and further demonstrates that diagnostic accuracy can be enhanced through integration of multiple full blood count parameters, highlighting the potential of combined full blood count markers to support earlier lung cancer diagnosis.<p></p>

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Melissa Barlow (21058934)

Subjects

dc:subject × 4

Rights

dc:rights
Statement dc:rights
  • All rights reserved

Identifiers

dc:identifier.*
Identifier
10779/exe.32616006.v1
OAI identifier oai:identifier
oai:figshare.com:article/32616006

Chain of custody

source
Harvested from
University of Exeter
Base URL
api.figshare.com/v2/oai
Last updated
2026-07-27
Source record
OAI-PMH GetRecord
citation

Melissa Barlow (21058934). Evaluating the Diagnostic Value and Molecular Mechanisms of Thrombocytosis in Lung Cancer. 2026.