Xenografts of primary human prostatic carcinoma

Background: Prostatic carcinoma is both the most common invasive cancer and the second most common cause of cancer deaths in men in the United States. Before 1991, attempts to propagate prostatic carcinoma from primary tumors for periods longer than 3 months were unsuccessful in vivo and in vitro with rare exceptions. In 1991, we reported establishment of slowly growing tumors for six of 10 human primary prostatic carcinomas approximately 2-6 months after transplantation. However, none of the tumors were larger than 5 mm or serially transplantable. Purpose: Our purpose in this study was to determine whether human primary prostatic carcinoma could be grown as serially transplantable xenografts. Methods: Cells from primary prostatic carcinomas obtained from transurethral prostatic resections or total prostatectomies in 20 patients were injected subcutaneously into male nude mice on the day of surgery. Sustained-release testosterone pellets were placed subcutaneously in the mice 2-24 days before transplantation of tumors and at intervals of 10-12 weeks. Serial transplantations in subsequent generations of mice were carried out by similar methods. Chromosome analysis was performed on six tumors. Results: Six of 20 primary prostatic carcinomas have grown sufficiently to permit serial transplantation into second mice; four have been documented histopathologically in the second mouse and serially transplanted into three or more successive mice. When a single primary tumor was injected into several mice by the same procedure, tumors failed to grow in some recipients but became serially transplantable in others. Growth of these tumors is slow and irregular, with frequent regressions. Short-term cultures of 10 tumors, eight of which were injected into mice in parallel, were initiated on the day of surgery; CWR31, which was successfully transplanted serially, exhibited only aberrant metaphases and showed clonal, chromosomal changes in culture. Including CWR31, three of the six tumors for which chromosomal analysis was successful contained clonal aberrations. Preliminary studies of SCID (severe combined immunodeficient) mice suggest that they are not superior to nude mice for establishment of serially transplantable prostatic carcinoma xenografts. Conclusions: A proportion of human primary prostatic carcinomas can be grown as xenografts. Four new serially transplantable xenografts (CWR21, CWR31, CWR91, and CWR22) are currently propagated in our laboratory, a resource that was not previously available. Implications: Our experience suggests that the most important factor in serial transplantation is the collaboration of urologists and pathologists in expediting placement of the tumor in cold saline, examination of the frozen section, and transplantation. [J Natl Cancer Inst 85:394-398, 1993]