Lung cancer is the leading cause of cancer-specific mortality in North America with over 158,000 estimated deaths in the USA and about 20,000 deaths in Canada [1, 2]. The main reason for the high mortality is the advanced stage at diagnosis in the majority of patients. Early detection of lung cancer by screening high-risk individuals has been proposed to improve survival. An effective screening program may prevent up to 80% of the deaths from lung cancer [3]. We undertook this study to determine if ¹H Magnetic Resonance Spectroscopy (MRS) of sputum and exhaled breath condensate samples can complement and/or provide an alternate tool to screen and detect lung cancer early among high-risk population.

Sputum and Exhaled Breath Condensate (EBC) samples were collected from 15 patients. Pathologically confirmed non-small cell lung cancer (NSCLC, n=8), and patients with respiratory conditions other than lung cancer (controls, n=7) were enrolled. Most were not current smokers. Both sputum and EBC samples were collected in thirteen patients; two patients provided EBC samples only. Sputum samples were thawed and 250 mL of each sample was dispersed in 2M NaCl (prepared in PBS buffer) solution in the ratio 1:1 (v/v) and vortexed to get homogenous clear suspension; frozen EBC samples were thawed and 500 mL neat sample was transferred into the NMR tube (with a reusable co-axial capillary tube containing standard TSP solution in D₂O) for analysis. All MRS experiments were performed on a Bruker Avance 400 MHz NMR spectrometer at the University of Winnipeg. In order to detect metabolite signals, the water signals in both EBC and sputum samples were suppressed using PRESAT or excitation sculpting (ES) pulse sequence. The MRS lab staff was blind to the diagnoses until all the samples were analyzed.

In EBC samples, propionate was detected in 43% of the patients in both groups; with similar median concentration of the metabolite. Ethanol and acetate were detected in all patients, with relatively higher median concentration of the metabolites in the cancer group. Acetone and formate were detected in 75% and 50% of the cancer group and 43% and 57% of the control group, with higher median concentration of the metabolites in cancer group. Methanol was detected in 75% and 100% of the cancer and control group respectively, with median lower concentration of the metabolite in the cancer group (P= 0.040), (Fig. 1, Table 1).

In sputum samples, propionate and N-acetyl sugars were detected in all patients with lower median concentration of the metabolite in cancer group. Glycoprotein was detected in all patients; median concentration of the metabolite was lower in the cancer group (P= 0.047). Lactate was detected in all cancer patients and 71% of the control group, with minimal difference in the concentration of the metabolite. Glucose was absent in all patients with cancer and present in 57% of the control group (Fig. 2, Table 2).

In our preliminary study, we analyzed sputum samples from lung cancer patients and observed the absence of glucose in lung cancer patients [4]. Given the usefulness and ease of obtaining EBC [5, 6], we analyzed EBC samples in addition to sputum in the current study. We have detected additional biomarker in sputum (decreased Glycoproteins) and in EBC (decreased methanol) samples which may be metabolic biomarkers of lung cancer. It was intriguing to note that one patient originally assigned to the cancer group did not have a confirmed diagnosis of lung cancer but had suspicious cancer cells only. This patient had high concentration of glycoprotein and presence of glucose in the sputum and high methanol in EBC.

Table 1: Median values of metabolite concentrations (mM) in EBC samples along with their P-values (Wilcoxon 2-samples test)

Table 2: Median values of metabolite concentrations (mM) in Sputum samples along with their P-values (Wilcoxon 2-sample test)

Lower concentrations of methanol (EBC), glycoprotein (sputum) and absence of glucose in sputum of the cancer patients suggest that MRS may provide a screening tool for lung cancer in high-risk patients and requires validation of these observations in a larger study.