Early Detection and Screening of Lung Cancer

Celine Mascaux; Nir Peled; Kavita Garg; Yasufumi Kato; Murry W Wynes; Fred R Hirsch

Expert Rev Mol Diagn. 2010;10(6):799-815.

Bronchoscopy

Bronchoscopy for Early Detection & Screening for Lung Cancer

CT scans are good at detecting small peripheral lesions, especially adenocarcinoma. However, CT scans are not suitable for detecting preinvasive lesions and early lung cancer in the central airways, specifically small-cell lung cancer (SCLC) and the early stages of squamous cell carcinoma, which comprise 17–29% of all lung cancers.^[34] Centrally located preinvasive and very early stages bronchial malignancies (e.g., severe dysplasia and CIS) are occult to the CT scans. McWilliams et al. found 28 lung cancers cases in a screened cohort of 561 high-risk patients (5%).^[35] Seven (25%) of the detected malignant cases were found by bronchoscopy, but not by image evaluation. However, this study used computer-generated cytologic evaluation for assessing sputum, again a technology not generally available. Centrally located tumors require direct inspection for diagnoses, namely bronchoscopy, with or without biomarker prescreening such as abnormal sputum cytology analysis. Thus, the main challenge is to detect pre-invasive or early invasive disease. White light bronchoscopy (WLB) is insufficient to detect such lesions; therefore autofluorescence bronchoscopy (AFB) was developed to address this limitation. AFB is now the gold standard for detecting pre-invasive lesions.^[36] Lam et al. were the first to use AFB in 1992 and the light-induced fluorescence endoscopy device became commercially available in 1998.^[37] The light-induced fluorescence endoscopy system uses a helium–cadmium laser to illuminate the bronchial mucosa with 442-nm light. The red and green auto-fluorescence emitted light is captured by a photo-amplifier camera and diplayed as green for normal areas and red-brown for abnormal areas.

In a European multicenter randomized controlled trial,^[38] 1173 high risk patients were randomized to undergo WLB or WLB plus AFB diagnosis. Overall, preinvasive lesions (moderate dysplasia or greater and CIS) were detected in 3.9% of the patients, while the prevalence with WLB was 2.7% and with AFB 5.1% (p = 0.037). Adding AFB increased the rate of detecting moderate to severe dysplasia and CIS by a factor of 2.1 and 1.25, respectively. The sensitivity to detect moderate dysplasia or greater increased from 57.9 to 82.3% by the use of AFB. Numerous other studies comparing the sensitivity of WLB to WLB plus AFB are summarized in Table 2. These studies show that adding AFB to WLB increases the sensitivity from 9 to 65% and 4 to 100% for moderate-to-severe dysplasia and CIS, respectively.^{[9,11,37–52]} However, AFB still hasn't been adopted in most of the clinics.

Within the past few years, a new bronchoscope, the narrow-band imaging bronchoscope (NBI), has been evaluated to detect bronchial dysplasia and CIS. While AFB uses blue light (390–440 nm wavelength), the NBI uses two bandwidths of light: 390–445 nm (blue) light that is absorbed by superficial capillaries and 530–550 nm (green) light that is absorbed by blood vessels below the mucosal capillaries. These narrow wavelengths reduce the scattering of light and enable enhanced visualization of blood vessels.^[53] Vincent et al. showed that NBI detected one cancer and four dysplastic cases in 22 patients that appeared normal by WLB.^[54] Peled et al. evaluated 104 postoperative patients 1 year after lung cancer surgery.^[55] All cases were evaluated by WLB and, of those, 47 cases were also evaluated by NBI. While they found 4% of early local recurrence with local stumpal polyp, adding NBI detected seven cases of bronchial dysplasia and metaplasia. A comparison between AFB and NBI has been reported recently by Herth et al..^[53] In 57 high-risk patients, the sensitivity of WLB to detect intra-epithelial neoplasia is 18%. The relative sensitivities (compared with WLB) of AFI and NBI were 3.7 (p = 0.005) and 3.0 (p = 0.03), respectively. Combining AFI and NBI did not increase diagnostic yield significantly.

Two other new tools have been developed: optical coherence tomography (OCT) and confocal fluorescence microscopy (CFM).

Optical coherence tomography is an optical imaging method that can offer microscopic resolution for visualizing cellular and extracellular structures at and below a tissue surface.^[56–60] OCT is similar to ultrasound imaging, but uses light rather than acoustical waves. In ultrasound, the imaging is accomplished by measuring the delay time (echo delay) for an incident ultrasonic pulse to be reflected back from structures within tissue. Lam reported that quantitative measurement of the epithelial thickness showed that invasive carcinoma was significantly different than CIS (p = 0.004) and dysplasia was significantly different than metaplasia or hyperplasia (p = 0.002).^[61] The autofluorescence endoscopy-guided OCT imaging of bronchial lesions is technically feasible. OCT appears to have great potential for management of early lung cancer and precancerous lesions.

Confocal fluorescence microscopy is a new technique that procedures microscopic imaging of a living tissue and that enables in vivo microscopic observation of the airways and alveoli. Similar to autofluorescence bronchoscopy, the device utilizes a blue laser. The small probe is integrated into the working channel of a bronchoscope into the distal airway and alveoli. The magnification and resolution of the images is such that alveolar structures and intra-alveolar cells can be clearly visible.^[62] Thiberville et al. reported early findings of fiberoptic CFM in 29 patients at high risk for lung cancer and healthy controls with premalignant and benign airway pathology.^[62] They found several microscopic patterns that may help in the recognition of dysplastic lesions. Then, they studied fiberoptic CFM in 41 healthy patients, including 17 current smokers.^[63] They reported a strong correlation between the number of cigarettes smoked per day and the amount of large and mobile macrophages observed in vivo, as well as with the intensity of the macrophage alveolitis. Fiberoptic CFM enables accurate exploration of the peripheral lung in vivo in both smokers and nonsmokers.

To summarize, development of new bronchoscopy techniques has improved the early detection of lung cancer, mainly for the central airways, and increased sensitivity compared with WLB, while more studies should be done to compare the yield of NBI to the AFB.

Sputum Cytology & AFB

Sputum cytology has been shown to be effective in detecting early squamous carcinomas of the major bronchi.^[64] In the update by the Colorado Lung Cancer group, the risk for lung cancer was increased when sputum atypia was graded moderate or greater (hazard ratio of 2.37 [1.68–3.34]), with a stronger association if the sputum sample was collected 5 months or less before the diagnosis of lung cancer (odds ratio of 10.32 [5.34–19.97]).^[10] Lam et al. recently reported that sputum cytology in high-risk patients has a sensitivity of 69% and specificity of 40% for lung cancer.^[65] The positive and negative predictive values of sputum atypia for lung cancer were 8 and 94%, respectively. They recommended sputum cytology followed by AFB as a screening regimen for high-risk groups.

Similarly to AFB, sputum cytology is rarely performed, especially not as a screening device, despite decades of reports regarding its efficacy.

1 2 3 4 5 6 7 8 9

Next Section

Cite this: Early Detection and Screening of Lung Cancer - Medscape - Sep 20, 2010.

Table 1. Recommended follow-up and management of nodules smaller than 8 mm detected incidentally at nonscreening computed tomography.
Nodule size (mm)	Low-risk patient^†	High-risk patient
≤4	No follow-up needed	Follow-up CT at 12 months; if unchanged, no further follow-up
>4–6	Follow-up CT at 12 months; if unchanged, no further follow-up	Initial follow-up CT at 6–12 months then at 18–24 months if no change
>6–8	Initial follow-up CT at 6–12 months then at 18–24 months if no change	Initial follow-up CT at 3–6 months then at 9–12 months and 24 months if no change
>8	Follow-up CT at around 3, 9 and 24 months, dynamic contrast-enhanced CT, PET and/or biopsy	Same as for low-risk patient

^†Low-risk patient = age ≤35 years, <12 pack-years smoking, no history of previous malignancy, no family history of aerodigestive tract malignancy.
CT: Computed tomography.
^‡Data taken from.^[159]

Table 2. The additive value of autofluorescence bronchoscopy to white light bronchoscopy to detect moderate/severe dysplasia and carcinoma *in situ*.
Study (year)	Patients (n)	Biopsies (n)	Sensitivity (%) - WLB	Sensitivity (%) - AFB ± WLB	Ref.
Lam et al. (1993)	94	328	48	73	^[39]
Lam et al. (1994)	223	451	39	79	^[40]
Lam et al. (1998)	173	700	9	56	^[37]
Kurie et al. (1998)	39	245	NA	43	^[11]
Venmans et al. (1999)	95	681	59	85	^[41]
Ikeda et al. (1999)	158	997	58	92	^[42]
Vermylen et al. (1999)	34	142	25	94	^[43]
van Rens et al. (2001)	72		20	100	^[44]
Hirsch et al. (2001)	55	391	18	79	^[9]
Shibuya et al. (2001)	64	212	69	91	^[45]
Sato et al. (2001)	63		61	89	^[46]
Moro-Sibilot et al. (2002)	244	354	36	86	^[47]
Beamis et al. (2004)	293	821	11	66	^[48]
Chhajed et al. (2005)	151	343	63	95	^[49]
Chiyo et al. (2005)	32	62	62	85	^[50]
Haussinger et al. (2005)	1173	1978	58	82	^[38]
Lam et al. (2006)	62	91	58	91	^[51]
Ikeda et al. (2006)	154	166	65	90	^[52]

AFB: Autofluorescence bronchoscopy; WLB: White light bronchoscopy.

Table 3. Studies assessing sensitivity and specificity of testing molecular markers for early detection of lung cancer.
Study (year)	Source	Biomarker tested	Technique	Sensitivity (%)	Specificity (%)	Ref.
Sozzi et al. (2003)	Plasma	Free circulating DNA	PCR	90^†	86^†	^[67]
Xie et al. (2004)	Plasma	Free circulating DNA	Pico green method	80.6^†	86.4^†	^[69]
Herrera et al. (2005)	Plasma	Free circulating DNA	PCR	48^†	100^†	^[70]
Ludovini et al. (2008)	Plasma	Free circulating DNA	PCR	80^†	61^†	^[74]
Ulivi et al. (2008)	Serum	Free circulating DNA	PCR	83^†	92^†	^[76]
Paci et al. (2009)	Plasma	Free circulating DNA	PCR	85.8^†	46.8^†	^[68]
Yoon et al. (2009)	Plasma	Free circulating DNA	PCR	91.2^†	57.1^†	^[73]
Zhong et al. (2006)	Plasma	Tumor associated antibodies	Customized diagnostic chips	82.6^‡	87.5^‡	^[83]
Yildiz et al. (2007)	Serum	Proteomic profiles	MALDI MS	67.4^† 58^‡	88.9^† 85.7^‡	^[86]
Patz et al. (2007)	Serum	Proteins	ELISA	89.3^† 77.8^‡	84.7^† 75.4^‡	^[87]
Belinsky et al. (2006)	Sputum	Promoter methylation of DNA	Methylation-specific PCR	64^†	64^†	^[97]
Varella-Garcia et al. (2010)	Sputum	Chromosomal aneusomy	FISH	76^†	88^†	^[98]
Liloglou et al. (2001)	BAL	LOH	PCR	73.9^†	76.5^†	^[99]
Fielding et al. (1999)	BAL	hnRNP A2/B1 antigen	IHC	96^†	82^†	^[100]
Voss et al. (2010)	BB and BW	Chromosomal aneusomy	Cytology + FISH	61^†	75^†	^[102]
Spira et al. (2007)	BB	Gene expression	Microarrays	80^‡	84^‡	^[103]
Beane et al. (2008)	BB	Clinicogenomic model	Microarrays	100^‡	91^‡	^[104]
Mascaux et al. (2009)	Bronchial biopsies	Gene expression	Microarrays	93^†	88^†	^[109]
Mascaux et al. (2009)	Bronchial biopsies	microRNAs	RT-PCR	83^†	100^†	^[108]

^†Test set.
^‡Validation set.
BAL: Bronchoalveolar lavage; BB: Bronchoscopic brushings; BW: Bronchoscopy washings; IHC: Immunohistochemistry; LOH: Loss of heterozygosity; MALDI MS: MALDI mass spectrometry; RT-PCR: Real-time PCR.