Abstract Microfluidic devices can offer a unique opportunity to more fully examine key factors of the tumor microenvironment that mediate metastasis. A microfluidic bioreactor was fabricated and cast in (poly)dimethylsiloxane (PDMS). Highly metastatic MDA MB 231-GFP+ tagged cells were allowed to aggregate into spheroids over a 24 hour period via a hanging drop method. The cancer cells were pipetted in droplets of media onto the inverted lid of a 100 mm well dish. The lid was then re-inverted onto its base and the dish was incubated overnight, where the cells in each drop would aggregate into a single spheroid. Afterward, the center channel of the microfluidic device was coated with an extra-cellular matrix (ECM), which consisted of 56% HEPES buffer, 24% Type 1 rat tail collagen, and 20% Matrigel. A single spheroid was then introduced into the center of the device by lowering the lid of the well dish onto the device such that the hanging drop was allowed to fall into an inlet hole, taking with it the spheroid. After a brief incubation, the device was submerged in media in a petri dish and monitored for 14 days. The device was used to study the metastatic potential of the cancer cells as a function of the microenvironment. In these experiments, NIH3T3-mcherry+-tagged fibroblasts were loaded into the channel along with the ECM. The spheroid remained fairly tightly aggregated during the first week, while the fibroblasts on either side of the channel grew an array of tubules within the ECM, migrating toward the spheroid. As the fibroblasts tubules approached the spheroid in the center of the device, the spheroid began branching out toward the incoming fibroblast tubules. The tumor cells used the fibroblast tubules as a scaffold to migrate outward in a single-file pattern. The cancer cells migrated a distance of roughly 0.55 mm/day. In other experiments, when a device containing a spheroid and fibroblasts was constantly exposed to 10 ng/mL CXCL12, there was an inhibition of fibroblast growth and tubule formation in the ECM, while the cancer cells migrated extensively even in the absence of fibroblast/matrix tubules, at a rate of roughly 0.37 mm/day. This contrasted with the control devices, where the cancer cells did not migrate until the fibroblast tubules reached the spheroid. In devices that consisted of a spheroid alone, the cells remained tightly aggregated and did not migrate into the surrounding environment. These data indicate that the microbioreactor utilized herein will be useful to dissect the interaction between cancer cell migration and the microenvironmental factors that facilitate this migration, leading to metastasis. When exposed to a pro-tumor chemokine environment or tubular fibrils laid down by fibroblasts, cancer cells can migrate freely. However, in a neutral microenvironment, the cancer cells may remain tightly aggregated until appropriate stimuli are provided, even though they have an intrinsic capability to metastasize. Citation Format: Matthew Rogers, Tammy Sobolik, David Schaffer, Philip Samson, John Wikswo, Ann Richmond. Examining the 3D tumor microenvironment via microbioreactors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4268.