Molecular evaluation of biopsy and autopsy specimens from patients receiving in vivo retroviral gene therapy

We used the polymerase chain reaction (PCR) to assay for the presence of re troviral vector and replication-competent retrovirus (RCR) in autopsy and b iopsy specimens from patients who received inoculations of retroviral vecto r producer cells (VPCs) into brain tumors or apparently normal tissues surr ounding resected tumors. The PCR assays were capable of detecting 1 or more proviral copies of vector or RCR in 500,000 cells. Of 113 patients treated in clinical trials between 1994 and 1997, autopsy specimens were available from 32 patients. Brain tumor biopsies were also available from 24 patient s. A total of 346 specimens was analyzed. Vector DNA was detected in 55% of tumor samples and 22% of brain samples obtained from resection margins. In contrast, most of the nonbrain tissues were negative for vector DNA; only low levels (<0.03%) of vector sequence were detected in 6 of 240 (2.5%) non brain tissues. Vector DNA was not detected in gonadal tissues from 12 men a nd 10 women. More importantly, RCR was not detected in any of the 134 biops y and autopsy tissues tested, including all brain tumor, brain, and gonadal specimens. These results comprise the largest data set on molecular analys is of autopsy specimens from patients receiving retroviral gene therapy and indicate that distribution of retroviral vectors following injection of hi gh doses of VPCs is limited to the site of inoculation.