• 
• 

Three Part Question

In [asymptomatic patients with risk factors for lung cancer] is the use of [chest X-ray] of benefit in terms of [improved disease-free survival]

Clinical Scenario

You are a chest registrar seeing a 55-year-old patient in a rapid access out-patient clinic who has recently presented with cough and hemopytsis. He is a smoker and had these symptoms for just a few weeks before being sent for a chest X-ray. It shows a large lesion in the right upper zone. The patient suspects he has lung cancer, which he probably does. He wants to know why he could not have had a chest X-ray before he was sick to pick up his lung cancer.

Search Strategy

Medline 1966 – Feb 2006 and Embase 1980 – Feb 2006 using the Dialog Datastar interface
[Lung-Neoplasms#.DE. OR Lung-Tumor#.DE. OR (Lung NEAR (Neoplasm$ OR Cancer$ OR Carcinoma$ OR Adenocarcinoma$ OR Angiosarcoma$ OR Chrondosarcoma$ OR Sarcoma$ OR Teratoma$ OR Lymphoma$ OR Blastoma$ OR Microcytic$ OR Carcinogenesis OR Tumor$ OR Tumour$ OR Metast$4)). TI,AB. OR NSCLC.TI,AB. OR SCLC.TI,AB.] AND [Mass-Screening.DE. OR Cancer-Screening.DE. OR (Screen$3 OR Case ADJ Finding OR Casefinding OR Case-Finding).TI,AB.] AND [Radiography-Toracic.DE. OR Mass-Chest-X-Ray.DE. OR Tomography-X-Ray.DE. OR Thorax-Radiography.DE. OR X-Ray.DE.] OR ((Chest OR Thoracic) NEAR (X ADJ Ray$ OR X-Ray$)).TI,AB.] limit to English. This search was repeated in Cochrane Central Register of Controlled Trials.

Search Outcome

A total of 136 papers were found of which 10 were deemed to be relevant. Only Randomised Control Trials (RCTs) or reviews of RCTs were included. Several systematic reviews and Guidelines for screening were reviewed including the most recent and only meta-analysis on chest X-ray screening. The same group has subsequently updated its previous Cochrane review. The individual randomised trials are presented

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Brett GZ 1960-1962 UK	N=55034 Males 40 Smokers and non-smokers	Cluster RCT	Resectability of patients with lung cancer	6 monthly CXR (I) 29% versus CXR at entry and exit of study (C) 44%	Limitation of screening biases Comparison is between intensive versus less intensive screening patients
			Disease specific 5-year survival	5-year survival of patients with lung cancer for intervention 15% over control group 6%
			Lung cancer mortality in population per 1000 patients/year	No benefit shown between intervention (0.7) over control group (0.8) RR (screen group/control) was 1.03 (95% CI, 0.74–1.42)
Wilde J 1972-1977 Germany	N=104880 Males aged 40-65 years Smokers and non-smokers	Cluster RCT	Resectability of patients with lung cancer	6 monthly CXR (I) 28% vs. 18 monthly CXR (C) 19%	Limitation of screening biases Compliance with scheduled screening was not described in detail
			Disease specific 5-year survival	5-year survival of patients with lung cancer for intervention 14% over control group 8%
			Lung cancer mortality in population per 1000 patients/year	No benefit shown between intervention (0.6) over control group (0.8) RR (screen group/control) was 1.34 (95% CI, 0.94–1.98)
Frost JK 1973-1978 USA	N=10384 Males 45	RCT	Lung cancer detection rate in population per 1000 patients/year	Annual CXR and 4 monthly sputum 4.8 versus annual CXR 5.5	Comparison is between intensive versus less intensive screening patients Adherence to strict protocol was poor
			Resectability of patients with lung cancer	Annual CXR & 4-monthly sputum (I) 4.8% versus annual CXR (C) 5.5%
			Disease specific 5-year survival	5-year survival of patients with lung cancer for intervention 47% over control group 44%
			Lung cancer mortality in population per 1000 patients	No benefit shown between intervention (0.6) over control group (0.8) RR (screen group/control) was 0.80 (95% CI, 0.65–1.00)
Fontana RS 1971-1976	N=10933 Males 45 Heavy smokers	RCT	Lung cancer detection rate in population per 1000 patients/year	4 monthly CXR and sputum 4.5 versus advised annual CXR and sputum 3.5	Lung cancer mortality is higher in intervention group compared to control group due to possible over diagnosis bias Comparison is between intensive versus less intensive screening patients
			Resectability of patients with lung cancer	4 monthly CXR and sputum (I) 4.5% versus annual CXR and sputum (C) 3.5%
			Disease specific 5-year survival	5-year survival of patients with lung cancer for intervention 46% over control group 32%
Berlin and Melamed - Memorial Sloan-Kettering Study 1974-1978 USA	N=10040 Age >45 years old Smokers	RCT	Resectability of patients with lung cancer	Annual CXR and sputum (I) 53% versus annual CXR (C) 51%
			Disease specific 5-year survival	5-year survival of patients with lung cancer for intervention 37% over control group 33%
			Lung cancer mortality in population per 1000 patients/year	No benefit shown between intervention (2.7) over control group (2.7) RR (screen group/control) was 0.98 (95% CI, 0.76–1.26)
Kubik A 1976-1982 Czechoslovakia	N=6364 Males aged 40-64 years Current heavy smokers	RCT	Resectability of patients with lung cancer	6-monthly CXR years 1, 2, 4 and annual CXR years 4, 5, 6 (I) 25% versus CXR years 4, 5, 6 (C) 16%	No unscreened control group
			Disease specific 5-year survival	5-year survival of patients with lung cancer for intervention 26% over control group 0%
			Lung cancer mortality in population per 1000 patients/year	No benefit shown between intervention (1.7) over control group (1.5) RR (screen group/control) was 1.14 (95% CI, 0.96–1.36)
Manser RL 2003 and 2004 Australia	N=245 610 Aged > 40 years old	A systematic review and meta-analysis of controlled trials	Lung cancer mortality was significantly greater in the group undergoing more frequent CXR than in those receiving less frequent screening (p=0.05)	More frequent CXR screening was associated with an 11% relative increase in mortality over less frequent screening (RR 1.11, 95% CI, 1.00 to 1.23) A non-statistically trend to reduced mortality from lung cancer was observed when screening with CXR and sputum cytology was compared to CXR alone (RR 0.88, 95% CI 0.74 to 1.03)	Most of the trials reviewed excluded women, young patients <45 years old and ex-smokers
Bach and Bach 2003 and 2003 USA	Review of 5 RCTs comparing CXR±sputum cytology versus control	Non-systematic review	Prolonged life expectancy of individual with disease	Neither CXR and/or sputum was of benefit	Review not systematic with no further statistical analysis
			Test not harmful or painful	Not addressed in sufficient detail in any of the studies reviewed
US Preventative Services Task Force 2004 USA	Systematic review and guideline of 6 RCTs and 1 non-RCT comparing CXR ± sputum cytology versus control having searched Medline from 1966 to 2003	Systematic review and guideline	Studies were graded according to criteria developed by USPSTF	None of the 6 CXR ± sputum cytology RCTs showed benefit among those screened All studies were limited because some level of screening occurred in control group 4 control-studies from Japan suggested benefit to both high and low risk participants, with screening using CXR ± sputum cytology occurring within 1 year of diagnosis, OR range 0.4–0.72
Oken MM 2005 USA	154942 participants aged 55-74 years with no history of PLCO cancer, randomly assigned to an intervention arm (77465)	RCT, baseline screen report	Number of initial suspicious CXR	8.9% (9.5% CI = 8.7%–9.2%) (N=5991) CXRS suspicious for Lung cancer, 206 (3.4%, 95% CI=3–3.9%) biopsies, 126 (61.2%, 95% CI=54.5–67.8%) diagnosed with lung cancer
			Number of lung cancers detected per 1000 screens	1.9 lung cancers were detected per 1000 screens, with positive predictive value 2.1% (95% CI=1.7–2.5%)
			Number of lung cancers detected per 1000 screens of smokers or ex-smokers	6.3 lung cancers were detected per 1000 screens, amongst current smoker; and 4.9 per 1000 screens amongst ex-smokers (less than 15 years)
			Detection of early stage lung cancer	Among cancers diagnosed, 44% (95% CI=35–52%) were stage I NSCLC

Comment(s)

The trials reviewed included only male current smokers over 40–45 years of age, and generally assessed more intense screening with chest X-ray±sputum cytology versus less intense chest X-ray screening. Typically the studies tended to show a higher incidence of lung cancer, a higher rate of surgical resection and a better survival in the more intensely screened groups. However, overall there appeared to be no significant reduction in mortality from lung cancer in the intense screening group compared to the less intense screened group. In fact, the subsequent meta-analysis [Manser] demonstrated that more frequent chest X-ray screening was associated with an 11% relative increase in mortality over less frequent screening. A non-statistically trend to reduced mortality from lung cancer was observed when screening with chest X-ray and sputum cytology was compared to chest X-ray alone (RR 0.88, 95% CI 0.74 to 1.03) [Manser, U.S. Preventive Services Task Force].

Editor Comment

(RCT=randomised controlled trial; I=intervention; C=Control group; RR=relative risks; CI=confidence interval; OR=Odds Ratio; Chest X-ray=CXR; NSCLC=Non-Small Cell Lung Carcinoma).

Clinical Bottom Line

The current evidence does not support the use of chest X-ray (with or without sputum cytology) as a screening test for lung cancer.

References

1. Brett GZ. The value of lung cancer detection by six-monthly chest radiographs. Thorax 1968; 23:414–420.
2. Wilde J. A 10 year follow-up of semi-annual screening for early detection of lung cancer in the Erfurt County, GDR. Eur Respir J 1989; 2:656–662.
3. Frost JK, Ball WC Jr, Levin ML, Tockman MS, Baker RR, Carter D, Eggleston JC, Erozan YS, Gupta PK, Khouri NF. Early lung cancer detection: results of the initial (prevalence) radiologic and cytologic screening in the Johns Hopkins study. Am Rev Respir Dis 1984; 130:549–554.
4. Fontana RS, Sanderson DR, Taylor WF, Woolner LB, Miller WE, Muhm JR, Uhlenhopp MA. Early lung cancer detection: results of the initial (prevalence) radiologic and cytologic screening in the Mayo Clinic study. Am Rev Respir Dis 1984; 130:561–565.
5. Fontana RS, Sanderson DR, Woolner LB, Taylor WF, Miller WE, Muhm JR, Bernatz PE, Payne WS, Pairolero PC, Bergstralh EJ. Screening for lung cancer. A critique of the Mayo Lung Project. Cancer 1991; 67:1155–1164.
6. Berlin NI. Overview of the NCI Cooperative Early Lung Cancer Detection Program. Cancer 2000; 89:2349–2351.
7. Melamed MR. Lung cancer screening results in the National Cancer Institute New York study. Cancer 2000; 89:2356–2362.
8. Kubik A, Polak J. Lung cancer detection. Results of a randomised prospective study in Czechoslovakia. Cancer 1986; 57:2427–2437.
9. Manser RL, Irving LB, Byrnes G, Abramson MJ, Stone CA, Campbell DA. Screening for lung cancer: a systematic review and meta-analysis of controlled trials. Thorax 2003; 58:784–789.
10. Manser RL, Irving LB, Stone C, Byrnes G, Abramson M, Campbell D. Screening for lung cancer. 2004 Cochrane Database Syst Rev CD001991
11. Bach PB, Kelley MJ, Tate RC, McCrory DC. Screening for lung cancer: a review of the current literature. Chest 2003; 123:72S–82S.
12. Bach PB, Niewoehner DE, Black WC. Screening for lung cancer: the guidelines. Chest 2003; 123:83S–88S.
13. U.S. Preventive Services Task Force, Agency for Healthcare Research and Quality Lung cancer screening:recommendation statement. Ann Intern Med 2004; 140:738–739.
14. Oken MM, Marcus PM, Hu P, Beck TM, Hocking W, Kvale PA, Cordes J, Riley TL, Winslow SD, Peace S, Levin DL, Prorok PC, Gohagan JK. Baseline chest radiograph for lung cancer detection in the randomised Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial. J Natl Cancer Inst 2005; 97:1832–1839.