Github is an open-source platform which focuses on collaboration based on full-request. As an important task of the Github collaboration platform, a code review is critically important, as it plays a key role in the collaboration ecosystem as well as the software quality improvement. Therefore, there have been several works to recommend proper reviewers to improve the Github collaboration ecosystem. Despite the fact, however, previous studies have some common limitations. What is at the core in previous works is to focus on the recommendation of the existing network only, not considering the chances of extending the network in order to consider more proper candidates. For this purpose, this study suggests a GCN-based link prediction based on the network extension, which expands the boundaries of potential reviewer candidates in the network. As a result, the model based on the network extension shows a good performance compared to the existing network. In addition, we conducted a comparison for the existing reviewers and new reviewers, and identified the new and potential reviewers as having a positive impact compared to the existing reviewers.