Objective:To explore the aerobic exercise tolerance and ventilatory efficiency during cardiopulmonary exercise testing (CPET) of persons with non-small cell lung cancer (NSCLC) complicated by type 2 diabetes mellitus (T2DM).Methods:Forty-eight persons with NSCLC and T2DM formed an NSCLC-T2DM group while another 48 persons with NSCLC but not T2DM formed an NSCLC-non T2DM group. Another 24 healthy counterparts were enrolled into the control group. All completed CPET before pneumonectomies were performed on those with NSCLC. Indexes of static pulmonary function, exercise tolerance, heart rate recovery, ventilation efficiency and gas exchange were computed.Results:Compared with the control group, both NSCLC groups had, on average, lower peak oxygen uptake (VO 2peak), lower anaerobic thresholds (ATs) and lower peak O 2 pulse rates. They also had higher average VE/VCO 2 slopes and VE/VCO 2 nadirs. Compared with the NSCLC-non T2DM group, those with T2DM had a significantly lower average VO 2peak and WRpeak, as well as significantly higher average VE/VCO 2 slope and VE/VCO 2 nadir. Compared with the control group, the average VO 2 and VCO 2 of both NSCLC groups was lower at the AT and during peak exercise, with the NSCLC-T2DM group′s averages significantly lower than those of the NSCLC-non T2DM group during peak exercise. During warm-up and at the AT, the NSCLC groups had a significantly higher average heart rate than the control group. Then, compared with the control group and the NSCLC-non T2DM group, the average heart rate in the NSCLC-T2DM group decreased significantly more slowly during the first three minutes of the recovery period. Compared with the control group, the VE/VCO 2 values of the NSCLC groups were significantly higher at the AT and during peak exercise. During the warm-up and at the AT, the average partial pressures of end-tidal carbon dioxide in the NSCLC groups were significantly lower than among control group, and during peak exercise the NSCLC-T2DM group′s average value was significantly lower than the control group′s. Compared with the control group and the NSCLC-non T2DM group, the NSCLC-T2DM group′s average forced expiratory volume in one second, forced vital capacity, peak expiratory flow rate and maximum voluntary ventilation were all significantly lower. Conclusions:Diabetes impairs the exercise tolerance and ventilation efficiency of persons with NSCLC. Without diabetes their exercise tolerance and ventilation efficiency would be impaired only slightly. CPET can provide a basis for risk assessment before pneumonectomy.