表 3.

患者体液及肺癌细胞系的VOCs研究总结

Studies on the VOCs analysis of patient-derived body fluids and lung cancer cell lines

Sources	Study	Collection	Control groups	Analysis methods	Efficiency				Summary
					Sensi- tivity (%)	Speci- ficity (%)	AUC	Other
a: lung cancer vs patients with benign nodules; b: lung cancer vs smokers; c: lung cancer vs non-smokers; AUC: area under the curve; PET: positron emission tomography.
Breathe	2021[33]	Tedlar bag	Healthy	E-nose	95.6	91.1	-		New technology and prospective way
	2019[39]	Tedlar bag	Healthy	E-nose	96.2	90.6	­-		New technology and prospective way
	2019[40]	Tedlar bag	Healthy	GC-MS	80	91.23	­-		Typical
	2019[41]	Tedlar bag	Healthy	GC-MS	75.4	85	­-		Typical
	2017[42]	Tedlar bag	Healthy	GC-MS	95	89	-		Typical
	2015[43]	Tedlar bag	Healthy	GC-MS	63.5	72.4	0.65		Typical
	2015[44]	Tedlar bag	Healthy benign	FT-ICR-MS GC-MS	96	64a 86b 100c	0.96		High sensitivity and mass accuracy
	2014[45]	Tedlar bag	Healthy benign	FT-ICR-MS	28	100	0.86		Higher specificity than PET in distinguishing benign disease
	2021[11]	Tedlar bag	Healthy	SIFT-MS	96	88	0.98		A persuasive platform for lung cancer prediction
	2016[46]	Tedlar bag	Healthy	PTR-MS	-	-	-		Cannot identify compounds with certainty
	2009[47]	Tedlar bag	Healthy	PTR-MS GC-MS	80	100	-		Does not need preconcentr-ation, more reliable quant- itative results
Urine	2017[20]	SPME	Healthy	HS-PTV-MS	100	100	-		High sensitivity and specificity, insufficient samples
Blood	2010[48]	Ultrasound-assisted HS-SDME	Healthy	HPLC		-	-	Some alkanes are high	Simplicity, low cost and short sample preparation time
Pleural effusions	2014[22]	HS-SPME	Healthy	GC-MS	-	-	-	P < 0.05	Simple, rapid, sensitive and solvent-free
Cancer cell line A549 Calu3	2018[49]	SPME	MCF7 WI38VA13	GC-MS E-nose	-	-	-	Some alkanes are high	Non-invasive diagnostic tool to classify lung cancer