Using PCR-based microsatellite DNA analysis with 48 markers we examined sputum and bronchial washing for genetic alterations compared with lymphocyte extracted DNA from 124 lung cancer patients and 36 healthy donors as normal control. Microsatellite alterations (MA) in at least one locus were detected in all cancer patient-derived specimens but only in 22. 2% of the healthy donors. Loss of heterozygosity (LOH) was detected in bronchial washings from 101 non-small cell lung cancer (NSCLC) predominantly on 17p13. 1-p13. 3 (69. 7%), 9p13. 3-p24. 1 (63. 3%), 1p34 VSports手机版. 2-p36. 22 (48. 5%), 13q12. 1-q13. 1 (47. 7%) and 3p22. 3-p23 (42. 7%). In bronchial washings from 23 small cell lung cancer (SCLC) LOH was detected mostly on 3p22. 3-p23 (88. 6%), 17p13. 1-p13. 2 (82. 3%), 5q32-q33. 1 (66. 6%), 13q12. 2-q13. 1 (65. 6%) and 9q22. 33-q31. 3 (52. 9%). The different LOH patterns indicate that different genetic background may be responsible for the different physiology of NSCLC and SCLC. The fractional allele loss (FAL) mean value of all cancer specimens was 0. 243+/-0. 021 compared with 0. 007+/-0. 008 of healthy donors with a confidence interval (CI) 99. 5%. Only seven out of 124 lung cancer specimens (5. 2%) exhibited FAL value less than 0. 083, the highest was observed in the healthy donors group. FAL appears to be a likely indicator for lung cancer detection. Microsatellite instability (MIN) was detected in 8. 7% of SCLC and 4. 0% of NSCLC bronchial washings in at least three loci tested. LOH and MIN detection in sputum and bronchial washing from the same patient was 77. 6%. Calculation of these indexes per marker exhibits significant variations that could be attributed to diffuse lung disorders or non-cancer specific genetic alterations. .