A cyclic peptide, in essence, is no different from any other peptide. You just have a peptide bond between the first and last residues. The bonded parameters for this linkage (bond, angles, dihedrals, and impropers if the force field uses them) are identical to any other peptide bond. Run settings are also the same; cutoffs are set based on the chosen force field.
You have NH3+ and COO- termini defined. This is the default behavior in pdb2gmx unless you use the -ter option and choose "None" along with -missing to prevent a fatal error. You then have to correct the topology to add bonds, angles, and dihedrals to define the linked connectivity.