Tessera Biomedical, a mid-tier pharmaceutical subsidiary of Palladian, conducted a Phase III clinical trial for an experimental oncological therapy designated TB-7714 in late 2222. The therapy used programmable viral vectors to deliver targeted apoptosis instructions to malignant cells — instructing tumors to destroy themselves. In twelve of forty-seven trial participants, the therapy did not work as designed. The tumors did not die. They stopped being cancer. They became something else. Tessera's internal documentation uses the phrase "aberrant therapeutic response." The oncologists on the trial used different language. Dr. Yuki Tanaka, the lead researcher, wrote in her personal notes: "The cancer learned architecture."

The twelve affected patients presented with tumor tissue that had ceased malignant replication and begun structured growth. Not random — structured. The tissue formed lattices with consistent spacing. Channels that appeared to serve circulatory functions, moving fluid through the tumor structure in patterns that one histologist described as "disturbingly efficient." In three patients, the tumor tissue developed formations that, under electron microscopy, resembled circuitry — parallel lines of differentiated cells with consistent conductivity properties, junction points where different tissue types met at precise angles. The structures were growing. They were not killing the host. The cancer had stopped being a disease and started being an organ.

The trial was terminated immediately. Official records cite "adverse events requiring extended observation." The twelve patients were transferred from Tessera's clinical facility to a private research installation whose location is not filed with any regulatory body. Their families were told the patients required specialized follow-up care. Communication was restricted. Legal agreements were signed. The patients, as far as public record is concerned, completed the trial and returned to normal life. They did not. They are somewhere in a facility that draws power equivalent to a small hospital and exists on no map. Tessera's parent company, Palladian, saw its stock price triple in the quarter following the trial termination. No explanation was offered. Analysts attributed the rise to "strong pipeline confidence."

One nurse who worked the trial before the transfer — Elise Moreau — has spoken to the documentation project on condition of anonymity. Her account is clinical and precise and deeply disturbing. The structures growing in the patients were not independent. In the weeks before the transfer, Moreau observed that the tumor architectures in different patients were growing toward each other. Not metaphorically — when two affected patients were placed in adjacent beds, imaging showed their respective structures orienting toward the nearest wall, as if reaching. When separated by greater distances, the growth pattern returned to standard expansion. But in proximity, the tumors seemed to be building toward connection. "Like roots looking for other roots," Moreau said. "Like they were trying to become a network."

The GLMZ Anomaly Documentation Project has been unable to locate the facility where the twelve patients are held. Tessera Biomedical's corporate records show no expenditure consistent with a long-term residential research installation. Palladian's records are similarly clean. But power grid analysis of the greater GLMZ area shows an unaccounted load in the city's southeastern industrial zone that matches the profile Moreau described: hospital-scale, constant, with regular spikes that could indicate imaging equipment or — and the documentation project notes this without endorsement — bioelectric stimulation. The cancer that learned to build is still building. Whatever it is constructing, it is doing so with the cooperation of the bodies it inhabits. And if Moreau is right about the network, it is not building twelve separate structures. It is building one.