The objectives of this study were to measure semi-quantitatively uptake of 99Tcm-sestamibi (99Tcm-MIBI) by tumour tissue in patients with lung cancer and to investigate its relationship with clinical response to chemotherapy. 99Tcm-MIBI single photon emission tomography was performed at the time of diagnosis in 31 patients with biopsy-proven lung cancer (19 small cell carcinomas, 12 non-small cell carcinomas), all of whom were undergoing chemotherapy. Fifteen patients were also investigated 2 weeks after the first and third cycles of chemotherapy. To quantify 99Tcm-MIBI uptake, a tumour/lung (T/L) ratio was calculated for the tomographic slices. The response to chemotherapy was rated as complete remission, partial remission or no remission using dimensional criteria. The results were expressed as the median and inter-quartile range; non-parametric statistical analyses were used. Forty one neoplastic localizations (31 primary tumours and 10 hilar or mediastinal lymph node masses) were assessed. The median T/L ratio of the primary tumours was 1.85 (range 1.7-2.4). Patients with a different response to chemotherapy had a significantly different median T/L ratio before chemotherapy: complete remission (n = 8), T/L ratio = 2.95 (range 2.20-3.25); partial remission (n = 10), 2.15 (range 1.77-2.40); no remission (n = 13), 1.70 (range 1.47-1.75) (Kruskal-Wallis, P < 0.0001). A T/L ratio of 1.80 gave sensitivity of 83%, specificity of 85% and accuracy of 84% in the prediction of the response to chemotherapy. The patients with small cell carcinomas demonstrated greater 99Tcm-MIBI uptake than those with non-small cell carcinomas: T/L ratio, median 2.30 (range 1.76-3.00) vs 1.70 (range 1.50-1.78) (Mann-Whitney U-test, P = 0.001). No significant difference in 99Tcm-MIBI uptake was observed between the 10 lymph node metastases and the corresponding primary tumours: T/L ratio, median 2.30 (range 1.75-2.50) vs 2.15 (1.77-3.00) (Wilcoxon's paired samples rank test, N.S.). Of the 15 patients who were monitored with scintigraphy during chemotherapy, 10 showed complete or partial remission and a parallel reduction in their T/L ratio. The other five patients showed no response to chemotherapy and their T/L ratio was either unaffected or increased. We conclude that the semi-quantitative assessment of 99Tcm-MIBI uptake may have a significant role to play in the management of lung cancer, providing an effective means of predicting the efficacy of chemotherapy and of selecting subgroups of patients requiring radiotherapy or combined protocols before the start of treatment. 99Tcm-MIBI imaging may also be of use in monitoring clinical response to chemotherapy.