DNA heterogeneity in metastasizing squamous cell head and neck cancer.

This study was carried out to determine whether analysis of DNA content of tumour cells (expressed as DNA-index: DI) from patients with head and neck squamous cell cancer (HNSCC) and lung squamous cell cancer (LSCC) could be helpful in distinguishing HNSCC patients with LSCC as a second primary from those in which LSCC represents a distant metastasis. Based on the assumption that metastasizing tumours retain their original DNA content, the same DI at both locations would suggest LSCC to be a distant metastasis from HNSCC, whereas a difference in DI at both tumour locations makes LSCC to be a second primary more likely. The study comprised 21 cases with HNSCC as well as LSCC. However, the basic assumption that the identity of a tumour can be inferred from its DI proved to be false as 7 of the 21 cases were characterized by more than one DI signifying several tumour cell populations. This DNA heterogeneity was further substantiated by differences in DI between the primary tumour and cervical lymph node metastasis in a second series composed of 16 HNSCC patients. These data indicate that due to variation in DI within one and the same tumour, DNA-analysis does not offer reliable information when trying to differentiate between lung cancer as a second primary or a distant metastasis in HNSCC patients.

Is preoperative transrectal ultrasonography of value in localised prostatic carcinoma? A blind comparative study between preoperative transrectal ultrasonography and the histopathological radical prostatectomy specimen.

Digital rectal examination (DRE) for staging is subjective and unreliable. Understaging has been reported in 25-72% and clinical overstaging of T3 tumors varies from 24-50%. In the present study of 15 patients, transrectal ultrasound (TRUS) staging was compared, in a blind comparative fashion, with pathological staging of the surgical specimen. Multifocal lesions were present in 8/15 patients (53%). A distinction was made between capsular involvement and 'clear' capsular penetration. TRUS was more sensitive in predicting capsular involvement than DRE (83% vs 17%), but the specificity was low (67% vs 100%). If capsular perforation was considered, the sensitivity and specificity of TRUS are 43% and 91%, respectively. Sensitivity and specificity for seminal vesicle tumor involvement by TRUS was 63% and 86%. Using TRUS the overall staging was improved by 33% compared with DRE and therefore TRUS is considered to be a valuable acquisition in localising and staging prostate cancer.