In early phases of the software cycle, requirements prioritization necessarily relies on the specified requirements and on predictions of benefit and cost of individual requirements. This paper presents results of a systematic review of literature, which investigates how existing methods approach the problem of requirements prioritization based on benefit and cost. From this review, it derives a set of under-researched issues which warrant future efforts and sketches an agenda for future research in this area. Key words: Requirements Prioritization, Software Economics, Non-Functional Requirements, Quantitative Prioritization, Systematic Review