The effect of direct referral for fast CT scan in early lung cancer detection in general practice. A clinical, cluster-randomised trial.

UNLABELLED: This PhD thesis is based on the project "The effect of direct referral for fast CT scan in early lung cancer detection in general practice. A clinical, cluster-randomised trial", performed in Denmark in 2010-2013. The thesis includes four papers and focuses on early lung cancer diagnostics in general practice.

INTRODUCTION: A total of 4200 new cases of lung cancer are diagnosed in Denmark annually. The stage of the disease is an important prognostic factor; thus, the opportunity for curative treatment declines with more advanced tumour stage. Lung cancer patients in Denmark (like in the UK) have a poorer prognosis than lung cancer patients in other European countries. One explanation could be delayed diagnosis. A fast-track pathway was therefore introduced in an attempt to expedite the diagnosis of cancer. However, it seems that not all patients can be diagnosed through this pathway. In order to ensure fast and early lung cancer diagnosis, it is crucial to examine the initial diagnostic process in general and the role general practice plays in lung cancer diagnostics in particular. The specific areas of investigation include the pathways to diagnosis, the characteristics of patients who are at special risk of delayed diagnosis and the level of prediagnostic activity in general practice. A chest radiograph is often the first choice in the investigation of lung cancer. Unfortunately, radiographs are less suitable for central and small tumours. Low-dose computer tomography (LDCT), however, has a high sensitivity for lung cancer which implies that it can be used to detect patients with localised, potentially curable disease.

AIM: The aim of this thesis was to increase our knowledge of the initial stages of lung cancer diagnostics in general practice. The thesis also examined the effect of a direct referral from general practice to an additional diagnostic test, the LDCT. The aims of this thesis were: 1) To describe Danish patients' pathways to the diagnosis of lung cancer in general and the prediagnostic activity leading up to diagnosis in particular. An additional aim was to explore the diagnostic intervals for specific patient groups (Paper I). 2) In a randomised, controlled trial including all patients referred for the existing fast-track scheme to either direct chest and upper abdomen CE-MDCT or to evaluation by the chest physician, (i) to test: Fast-track performance measured by the number of CE-MDCT scans and chest physician specialist time per diagnosis  (Paper II) 3) In a two-arm, clinical, controlled, cluster-randomised trial where direct referral to CT together with a lung cancer update is compared with usual practice, (i) to test how CT is used in this group of patients and the outcome of CT (Paper III); and (ii) to test the effect of either modality on the time to lung cancer diagnosis, the TNM stage and the use of the fast-track pathway for lung cancer (Paper IV).

METHODS: Study I was a national registry-based cohort study of 971 consecutive, incident lung cancer patients in 2010 Data were derived from national registries and questionnaires filled in by general practitioners (GPs). Study II was a randomised, controlled trial enrolling 493 patients referred from general practice to a fast-track evaluation. Half of the patients were randomly assigned to the intervention and went straight to a chest CT before a chest physician evaluation. Studies III and IV were a cluster-randomised, controlled trial (IV) and a cohort study nested in the trial (III). A total of 199 general practices with 266 GPs were randomised into two groups. Intervention GPs were offered direct access to a low-dose chest CT combined with a meeting on early lung cancer detection. Study III concerned the intervention arm solely and reported uses and outcomes of the scans. Study IV evaluated the effect of direct low-dose CT on the time to diagnosis and stage at diagnoses for patients from intervention and control GPs.

RESULTS: In Study I, we found that GPs were involved in 2/3 of all lung cancer diagnostic pathways. One quarter of the patients followed the obvious pathway from general practice to fast-track detection. At least one radiograph was performed in 85.6% of patients, whereas 1/3 of all patients had two or more radiograph performed during the 90 days preceding diagnosis. Patients with co-morbidity or unspecific symptoms more often had two or more X-rays performed than patients without these characteristics. In Study II, there was no difference between the groups in the number of CTs performed. In the intervention group, chest physicians spent mean 13.3 minutes less per referred patient than in the control group. In Study III, we found that 648 patients were referred to low-dose CT during a 19-month period. Half of the referred patients needed further work-up, and 15 (2.3%) of the patients had lung cancer, 60% in a localised stage. For all patients, 6.8% were diagnosed with a severe lung disease. In all, 2/3 of the GPs used the CT opportunity; and the referral rate was 61% higher for GPs participating in the lung cancer meeting than for GPs who did not participate in such meetings. In Study IV, we found that direct, low-dose CT from primary care did not significantly influence stage at diagnosis and had only a limited impact on time to diagnosis.

CONCLUSION AND PERSPECTIVES: This thesis contributes to the knowledge of the early diagnosis of lung cancer in Denmark. General practice was found to play an important role, but only a small part of Danish lung cancer patients were diagnosed from general practice through the fast-track pathway. This together with the fact that a high proportion of patients had two or more radiographs within the 90 days preceding the diagnosis indicate that other diagnostic strategies should be tested in an attempt to provide GPs with the best opportunity for early diagnosis. This thesis provides evidence that GPs are, indeed, able to refer patients straight-to-test in the fast-track pathway. This knowledge may be used when organising other fast tracks. Furthermore, GPs participating in education about early lung cancer diagnosis were willing to refer patients direct to low-dose CT (LDCT) from primary care. Half of the patients needed further diagnostic work-up, and 2.3% of all patients referred were diagnosed with lung cancer. In addition, many lung diseases were diagnosed by LDCT. No effect on time to diagnosis or stage at diagnosis was found when patients from intervention GPs were compared with patients from control GPs. The effect of combining direct access to LDCT with referral to the existing fast-track pathway should be analysed as it may ensure earlier and faster lung cancer detection in primary care. Direct access to LDCT scan may also be an alternative to lung cancer screening. Furthermore, if a LDCT screening program is going to be implemented, it should be considered to supplement the program with access to CT directly from primary care for the symptomatic, not-screened patients.