This group from the US and Switzerland performed a retrospective study comparing FISH with conventional cytology on 61 stored bronchial washing samples. Twenty two of the samples were negative for lung cancer and thus acted as controls. Of the 39 samples with histologically proven lung cancer, only 18 were positive at cytological examination. A multi-probe FISH assay was used against four commonly mutated chromosomal regions in lung cancer with at least two positive probes in six or more cells needed for diagnosis. The FISH system had a significantly higher sensitivity—32 (82%) v 18 (54%) patients (p=0.007)—but lower specificity—18 (82%) v 22 (100%) patients (p=NS). Of the four false positive samples, one had metastatic colorectal carcinoma. The authors suggest that these false positive results may be an early detector of malignant change.

Although the small numbers and its retrospective nature limit this study, it does suggest that DNA probe techniques have some advantage over conventional cytopathology. This is an interesting idea that needs large prospective studies to confirm the usefulness of FISH in day to day clinical practice.