Key points are not available for this paper at this time.
By placing a transparent window (Sanders chamber) in the neck of a hamster and a light source in its mouth, one can microscopically observe vascularizing tumor fragments in the cheek pouch. When a microporous filter membrane with pore size 0.5 μ or less (ultrathin Millipore or Nuclepore) was interposed between explants of the transplantable Greene human choriocarcinoma and the underlying pouch tissue within a Sanders chamber, circumscribed dense networks of proliferated capillaries and venules formed within a week beneath the filter—under, alongside, or encircling the overlying tumor explants. For wholemounts, pouch vessels were filled with carbon black via aortic perfusion, and the pouch was fixed and stained. Wholemounts clearly showed the zones of vasoproliferation directly underneath growing sheets of tumor cells. Electron microscopy revealed particulate material of probable tumor origin, apparently passing from tumor pseudopods to pouch tissue through the filter membrane pores. Although this material appeared to consist of discrete particles, it could not be unquestionably proved they were not portions of narrow tumor pseudopods cut in different planes. Explants of hamster and human endometrium failed to elicit any vascular reaction across a Millipore filter. When considered with the work of others, the vasoproliferative response to choriocarcinoma explants is another example of the tumor cell's ability to rapidly establish its blood supply. Continued experimentation with microporous membranes may prove that the tumor cells are indeed releasing a vasoactive substance.
Ehrmann et al. (Sun,) studied this question.