Decline of exhaled isoprene in lung cancer patients correlates with immune activation

Human breath contains a variety of endogenous volatile organic compounds (VOCs). The origin and pathophysiological importance of these VOCs is still poorly investigated. The observation that dogs seem to possess the ability to sniff out early stage colorectal cancer [1] together with other observations initiated studies to examine VOCs in exhaled breath as non-invasive screening for cancer development [2–5]. Breath isoprene (2-methylbuta-1,3-diene; mass: 68D) concentrations have been reported to be altered in a number of clinical conditions [6] and therefore might be of diagnostic potential [7].

Isoprene represents a precursor molecule of isoprenoids (steroids, terpens), and available data suggest that isoprene is related to cholesterol biosynthesis [2]. However, the physiological meaning of isoprene changes has not been established and isoprene testing has not yet reached the level of routine clinical methods. Nevertheless, lower concentrations of isoprene have been reported recently in lung cancer patients as compared to healthy volunteers [8], and it is still of interest to show whether isoprene concentrations can reveal any prognostic information for the future course of the disease in patients.

Concentrations of inflammation markers such as C-reactive protein (CRP) [9–11] and of parameters indicating immune activation such as neopterin were observed to be elevated in a subgroup of patients suffering from malignant diseases including patients with bronchus carcinoma [12–15]. Also, the rate of tryptophan breakdown is found to be increased in patients with cancer including lung cancer and was found to be of predictive value [16–19]. Accelerated tryptophan breakdown, as is indicated by an increased kynurenine to tryptophan ratio (Kyn/Trp), is most probably due to enhanced activity of enzyme indoleamine 2,3-dioxygenase (IDO) [17]. Like neopterin production, IDO activity is primarily induced by pro-inflammatory stimuli such as Th1-type cytokine interferon-γ [20–22].

The sensitivity of these marker concentrations is usually not high enough to detect early cancer development, and the difference between distinct diagnostic categories such as staging or grading or between small cell cancer and non-small cell cancer is usually small, but higher concentrations of such markers are significantly associated with more rapid disease progression and earlier death of patients [13, 15–18]. Like the alterations of breath isoprene concentrations [7], also increased concentrations of markers of inflammation and immune activation are not specific to patients with lung cancer or cancer in general, rather patients suffering from a variety of diseases such as infections and autoimmune pathologies often present with elevated concentrations of CRP [23, 24] and/or neopterin [25]. Such findings render it likely that changes of breath isoprene may relate to the inflammation and immune activation status of patients.

The aim of this study was to investigate whether the lowered breath isoprene concentrations in lung cancer patients are associated with the inflammation and immune status in patients. For that purpose plasma specimens were collected in parallel from patients in whom breath analyses were performed [7].

This study includes 79 lung cancer patients (23 females, 56 males) with an average (+SEM) age of 62.9 + 9.3 years; more detailed demographic details are given in table 1. The group of this study represents a subgroup of patients of which the results of breath analyses have already been reported earlier [8].

The study was approved by the ethical committee, was in accordance with the Helsinki declaration, and all individuals included in the study gave informed consent that results of measurements of additional markers were solely used for scientific purposes in a blinded manner.

Samples of mixed breath gas were collected in Tedlar bags (SKC Inc., Eighty Four, PA) with parallel collection of ambient air (also in Tedlar bags) [8]. Breath gas samples were obtained after an ∼5 min sitting of a volunteer. Each subject provided one or two breath samples by use of a straw. All samples were processed within 3–6 h. Mixed alveolar breath instead of alveolar breath was collected in order to also detect isoprene directly released from the lungs. Before collection of breath, all bags were thoroughly cleaned to remove any residual contaminants by flushing with nitrogen gas (purity of 99.9999%), and then finally filled with nitrogen and heated at 85 °C for more than 8 h with a complete evacuation at the end. 18 ml of gas sample was transferred to 20 ml volume evacuated glass vials, and equilibrated with nitrogen gas.

Concentrations of isoprene (m/z 69, tentatively identified as protonated isoprene but also protonated 1-octen-3-ol, pentanal, octanal and nonanal appear at this mass-to-charge ratio) were measured in breath samples [8]. Since these other compounds are also measurable in exhaled breath, isoprene is not specifically detected. However, the concentration of isoprene in exhaled breath is at least one order of magnitude higher than the concentrations of the other compounds. For proton transfer reaction mass spectroscopy (PTR-MS) analysis, VOCs are ionized by proton transfer from H₃O⁺-ions produced in the ion source of the instrument with subsequent measurement using a quadrupol-mass-spectrometer. A high-sensitivity proton transfer reaction mass spectrometer (3 turbopumps; Ionicon Analytic GMBH, Innsbruck, Austria) with Teflon rings (instead of Viton rings) was used. Reagents and standard were bought from Sigma-Aldrich (Steinheim, Germany) and standard was prepared by evaporation of liquid in a glass gas bulb. Before using, all bulbs had been cleaned with methanol, dried in oven at 120 °C for at least 20 h, and then purged with ultra-clean nitrogen for at least 10 min. Afterwards, bulbs were evacuated by means of a vacuum pump for 10 min. Calibration standard was obtained by injection of 1–3 µl through membrane, using a GC syringe into a glass bulb. After evaporation, appropriate amount of vapor was removed using gas tight syringe and introduced into Tedlar® bags (SKC 232 Series, Eighty Four, PA, USA) with 0.5, 1 or 1.5 L of nitrogen. For additional details see [8].

At the time of breath sampling, parallel plasma specimens were collected and were kept frozen at –20 °C until analyzed. Concentrations of immune activation marker neopterin (ELISA, BRAHMS, Hennigsdorf, Germany), tryptophan and kynurenine (by HPLC [27]), lipid parameters (routine enzymology) and CRP (nephelometry) were measured.

Concentrations of all analytes are given as mean values + SD. Because not all the data sets showed normal distribution, non-parametric statistics were applied for data analyses. For comparison of grouped data Friedman test and Wilcoxon signed rank test were used. Associations between parameters were examined by the Spearman rank correlation technique. The Statistical Package for the Social Sciences (PASW Statistics 18, Chicago, IL, USA) was used. p-values below 0.05 were considered to indicate significant differences and correlations.

Isoprene concentrations in lung cancer patients were median (25th–75th percentile) 92.5 ppb (79–131 ppb; table 2) which was lower than those in healthy controls measured earlier (median: 105 ppb; p < 0.05) and this result agrees well with the earlier obtained data [8]. This earlier analysis also already showed a great overlap of isoprene concentrations in patients and controls which certainly hampers drastically its utility as a cancer screening strategy.

Forty-two patients (53.9%) presented with CRP concentrations above 10 mg l⁻¹, the upper limit of the normal [26]. Average neopterin concentrations were higher as compared with healthy controls of similar age, and again only a subgroup of 27 patients (34.6%) presented with neopterin levels higher than 8.7 nmol l⁻¹, the 95th percentile in healthy controls [27]. Eighteen patients (23%) presented with total cholesterol of < 160 mg dl⁻¹ and 17 patients (22%) had HDL cholesterol concentrations beyond 40 mg dl^–1. Females had lower HDL cholesterol (49.6 + 16.4 mg dl⁻¹) than males (56.4 + 14.0 mg dl⁻¹; U = 2.167, p < 0.05); none of the concentrations of the other analytes were influenced significantly by gender; in females median isoprene was 97 (25th–75th percentile: 81–114) ppb and in males 92 (76–125) ppb, not significant. Also there was no significant influence of age and disease stage or grade on breath isoprene concentrations, but isoprene concentrations correlated significantly with total cholesterol (rs = 0.281, p < 0.01), LDL cholesterol (rs = 0.236, p < 0.05) and triglycerides (rs = 0.164, p < 0.01) but not with HDL cholesterol (rs = 0.048, not significant; table 3). There was a significant correlation between isoprene and neopterin concentrations (rs = −0.215, p < 0.05; figure 1) but no significant associations existed between isoprene concentrations and changes of tryptophan metabolism. Aside from that, lipid metabolism was associated with markers of inflammation and immune activation (table 3): total cholesterol and HDL cholesterol correlated inversely with CRP and neopterin concentrations and with tryptophan breakdown Kyn/Trp (table 3). No such associations existed between these parameters and LDL cholesterol and triglyceride concentrations with the exception of an inverse relationship between neopterin and LDL cholesterol concentrations (table 3).

Zoom In Zoom Out Reset image size

Age and smoking habits had no influence on breath isoprene levels and concentrations of lipid metabolites and CRP, but age was significantly correlated with neopterin (rs = 0.461, p < 0.001) and Kyn/Trp (rs = 0.348, both p < 0.01). CRP concentrations correlated with tumor stage (rs = 0.326, p < 0.01) and kynurenine with grade (rs = 0.307, p = 0.01). Finally, neopterin concentrations correlated with tryptophan breakdown (kynurenine: rs = 0.451; Kyn/Trp: rs = 0.585, both p < 0.001).

Different multiple regression models were applied using isoprene as outcome but they did not reveal any other significant association with blood biomarkers or influence of the demographic characteristics. The population size was probably too small to find such associations.

Our investigation reveals significant associations between the decreased isoprene exhalation in lung cancer patients and the concentrations of immune activation marker neopterin and parameters of lipid metabolism. Results indicate a link between the decline of isoprene and of total and LDL cholesterol concentrations in the patients.

A rather weak but significant inverse correlation was observed between breath isoprene and serum neopterin concentrations. Also neopterin concentrations were related to lipid metabolic abnormalities. This was true also for the other parameters of inflammation (CRP) and immune activation (Kyn/Trp): the higher the immune activation status, the lower the total cholesterol and HDL cholesterol concentrations. Neopterin concentrations also correlated inversely with LDL cholesterol. Thus, the activated immune system is probably involved in the decline of circulating cholesterol. The impact of the immune system on cholesterol biosynthesis might relate to the decline of breath isoprene which correlated positively, albeit weakly, with total and LDL cholesterol.

Immune system activation and inflammation are associated with an overwhelming production of antimicrobial and cytocidal enzymes and chemicals, mostly of oxidizing nature such as reactive oxygen species (ROS) such as hydrogene peroxide (H₂O₂) and superoxide anion (O₂⁻) [28, 29]. High output production of ROS by stimulated immune effector cell-like macrophages could influence on the one hand cholesterol biosynthesis, e.g. by interfering with 3-hydroxy-3-methylglutaryl-(HMG) coenzyme-A-reductase (EC 1.1.1.88), the rate-controlling enzyme of the mevalonate pathway [30]. This metabolic pathway is the key element for the production of cholesterol and other isoprenoid compounds [2]. Any interference with this pathway would not only diminish biosynthesis of cholesterol but might also interfere with isoprene production. It has been demonstrated earlier that the inhibition of cholesterol biosynthesis by statins causes a decline of isoprene in parallel to cholesterol [2].

Isoprene itself, containing two double bonds in its chemical structure, can serve as target of oxidation and might thus suffer from oxidative stress, which is developing in the case of a chronic inflammatory process with excessive ROS production when antioxidant enzymes and compounds are overwhelmed by oxidizing chemicals. Notably, decline of several antioxidant compounds and vitamins has been observed to correlate inversely with the increase of neopterin levels in individuals at risk of atherosclerosis [31] or with cancer [32]. Data suggest that immune activation, oxidative stress and oxidation could play a role in the decline of isoprene and probably as a consequence also of lipid metabolic changes. Immune activation, oxidative stress and oxidation are closely linked together by the strong potential of interferon-γ to induce formation of ROS in target cells [28], which include not only macrophages but also neutrophils [33, 34]. Earlier studies already showed that the chosen immune activation marker neopterin correlates with the accumulation of oxidation products and also with the decline of antioxidants in patients which indicates a systemic oxidative state [35].

A potential influence of immune responses on isoprene levels is not a completely new aspect; it has already been reported that, e.g., vaccination is associated with a significant change of breath isoprene exhalation [36]. Notably also among the different diseases in which a decline of breath isoprene concentrations has been reported, a reasonable number of conditions with an inflammation and/or immune activation background is included [3, 5, 7]. Interestingly, similar relationships between elevated neopterin and decreased HDL and total cholesterol concentrations have been reported earlier in patients with HIV-1 infection [37]. Moreover, also in patients suffering from acute malaria, a drop of HDL cholesterol has been described [38, 39], a clinical condition which is well known to be associated with a rapid increase of immune activation cascades and high output neopterin production [40, 41]. On the one hand, the earlier analysis by Bajtarevic et al already showed the lower levels of isoprene in the lung cancer patients as compared with healthy controls [8], and on the other hand, there is ample literature available, the first one already published in 1975 [42], showing the decline of total cholesterol and also of LDL cholesterol in patients with cancer including lung cancer. Moreover, in some of these reports on large populations of patients with different types of cancer it was demonstrated that the decline of concentrations of the lipid parameters is a sign of poor prognosis [42–46]. Finally, immunostimulatory treatment of cancer patients with IL-2 was found to further the decline of HDL, LDL and total cholesterol levels [47]. This result further supports the view that activation of the immune system is most likely the reason for the decline of cholesterol concentrations in cancer patients.

This study confirms and extends the finding that lung cancer is associated with a decline of isoprene, which contrasts the usual consideration that a convenient tumor marker is produced and released from tumor cells at a higher rate. For isoprene the opposite is obviously the case: the underlying tumor disease slows down a biochemical process or leads to degradation of the compound, and probably the immune system is involved in the observed abnormality. From studies on tryptophan and iron metabolism it is well established that one part of the antiproliferative strategy of immunocompetent cells is the withdrawal of essential nutrients from the site of an infection or malignant development. To achieve this goal, immunoregulatory circuits lead to the breakdown of essential amino acid tryptophan by the specific IDO enzyme [17, 20]. In parallel, iron is moved from the circulation to storage sites [48]. With this in mind it is no surprise when immune cells also try to reduce availability of specific lipid metabolites such as cholesterol which are also of great importance for normal cellular development and growth.

With regards to dogs sniffing cancer [1] it has to be considered that these animals not necessarily need to detect an increased concentration of a compound in breath. It could also relate to the decrease of a specific compound, such as isoprene, the decline of which can probably be detected by the dog as well.

Our study shows that the decline of breath isoprene concentrations in patients with lung cancer is associated on the one hand with total and LDL cholesterol concentrations. On the other hand, lower isoprene concentrations were found to correlate, albeit weakly, with higher concentrations of immune activation marker neopterin supporting the view that chronic immune activation might underlie subtle changes of cholesterol biosynthesis which also influences isoprene metabolism. Further studies will be conducted to substantiate this idea.

The authors thank Mrs Maria Pfurtscheller for excellent technical assistance. AA greatly appreciates the generous support of the government of Vorarlberg and its governor Landeshauptmann Dr Herbert Sausgruber.