This paper investigates the interaction of a wide range of electricity and fuel prices and technical factors of combined heat and power production in a district heating system. A linear programming-based optimization model with the objective to minimize system cost was used to study the energy systems in the cities of Gavle and Sandviken in Sweden. The comprehensive outcomes from optimization and parametric studies have been analyzed using a polynomial-based metamodel. System costs include variable costs for the production and revenues for sale of heat and electricity. The metamodel is used as an analytical and explanatory tool to interpret input-output relationships. Municipal district heating systems of Gavle and Sandviken in Sweden are studied as an interconnected regional system with improved and new combined heat and power plants. The results show that effects from electricity and fuel prices are important, but that variations in energy system cost may also be caused by many cross-factor interactions with technical factors. A comparative system performance analysis with defined cases and optimal factor setting shows a substantial increase in the electricity production, here by up to 650 GWh annually. The profitability of investing in a new plant depends highly on the considered investment risk and electricity and fuel market prices. CO2 emission savings by up to 466 kton annually can be accomplished if marginal electricity production from coal-condensing power plants is avoided and biofuel is released at the same time.