I have several hundred NEC EB2-12NU (12 is the operating voltage, also available in other voltages) in customers hands where the relay dry switches microvolt level RF and have not had a single failure to date. Many of my preampifier products use it, as does my active antenna. And yes, I know they are rated at 10mV minimum DC voltage, but all I can say is they have proven to be extremely reliable switching radio signals.
These are inexpensive parts, available in surface mount and through hole, designed for the wire-line telephone industry. The through hole parts are the EA series and I have not checked recently to see if they are still available.
The main problem with relays not specifically designed for RF is leakage through the open contacts - a good RF relay will have some internal shielding to minimize leakage.
But at the frequency range you are interested in, a pF or two leakage may not be important.
Plots below show the performance I measured of the EB2-12NU relay when used in the Z10080A RF switching board I sell. The first three plots show isolation of various contact configurations (the contacts are open) and the fourth plot shows the through loss when installed on a Z10080A board. The data is taken in a 50 ohm environment, so if you are switching high impedance circuits, the effect of a couple pF leakage capacitance will become more important.
http://www.cliftonlaboratories.com/z10080a_bypass_relay.htm for more information on the switching board.