As cancer advances, cells often spread from the primary tumor to other parts of the body and form metastases. Here we investigate a conceptually elegant model of metastasis formation where each primary cell can initiate metastatic lesions which lesions then evolve as independent branching processes. We assume that the primary tumor is resected upon detection. Of fundamental importance is whether synchronous (detectable) or metachronous (undetectable) mets are present at this detection time, the distribution of their numbers and sizes. If there are only metachronous mets at detection, how long until these mets become detectable, leading to the relapse of the disease? We'll extend this model to cancers in which cells first need to evolve the ability to metastasize. Using sequence data from primary and corresponding met samples we'll propose that these intermediate cells are indeed present for certain primary met parings, why not there for others.