Pilot risk score for lung cancer risk stratification in patients with intermediate bronchial cytology based on big clinical data modelling

This abstract has open access
Abstract Description
Submission ID :
HAC234
Submission Type
Authors (including presenting author) :
Joanna K. M. Ng (1), Alex H. Lin (2), Ka Pang Chan (3), Ka Tik Cheung (4), Celine S. L. Chui (5), Joshua J. X. Li (1)(2)
Affiliation :
(1) Department of Pathology, Queen Mary Hospital

(2) Department of Pathology, School of Clinical Medicine, Queen Mary Hospital, The University of Hong Kong, Hong Kong

(3) Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong

(4) Department of Applied Science, School of Science and Technology, Hong Kong Metropolitan University, Hong Kong

(5) School of Nursing, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China
Introduction :
Lung cancer is the leading malignancy in terms of incidence and mortality in Hong Kong. Bronchial cytology is a minimally invasive while able obtain tissue diagnosis. The limitation is that a significant percentage of patients will receive an intermediate (atypical) result. Risk stratification by pre-procedure / baseline investigations will guide timely follow-up and cost-efficient management for these cases.
Objectives :
1) To identify predictive demographical and clinical parameters for patients undergoing lung cancer investigation for a lung cancer diagnosis and 2) to propose a risk score based on such parameters that effectively stratifies high-risk and low-risk patients.
Methodology :
Patient demographics and clinical investigations including complete blood count, liver and renal function, clotting profile and c-reactive protein from patient episodes with bronchoscopy performed in Queen Mary Hospital and resulting in an “atypical” bronchial cytology diagnosis were retrieved. Diagnosis codes were reviewed determining the reference (lung cancer) status. Data was used to identifying correlations and producing a clinical risk score model.
Result & Outcome :
From the year 2011 to 2023, a total of 1934 intermediate bronchial cytology were reported, of which 452 (23.4%) were diagnosed with lung cancer. Sex (p< 0.001), albumin (p< 0.001), eosinophil count (p< 0.001), mean corpuscular hemoglobin concentration (MCHC) (p=0.015), neutrophil count (p< 0.001), platelet count (p< 0.001), serum protein (p=0.011), prothrombin time (p=0.025) and total white cell count (WCC) (p< 0.001) were significantly associated with a lung cancer diagnosis. A risk score = (albumin low: +2; MCHC high: -1; neutrophil high: +2, low: -1; platelet low: -1, protein high: +1, prothrombin time high: +2; WCC high: +1, low: -2) + 4, stratified high risk (score ≥ 7, n=182/579, 31.4% risk), intermediate (score 5 – 6, n=249/1141, 21.8%) and low (score 0 – 4, n=21/214,9.8%) cancer risk.



Demographical and clinical data is useful in risk stratification of intermediate bronchial cytology diagnoses. The risk score, after validation, can be practically applied by a clinician or reporting pathologist to guide follow-up and management.
The University of Hong Kong
Queen Mary Hospital
The University of Hong Kong
The Chinese University of Hong Kong
Hong Kong Metropolitan University
The University of Hong Kong
29 visits