Unfortunately, relativistic mass does not directly determine how much a body gravitates, specifically plugging "relativistic mass" into Newton's gravity law doesn't give the correct answers.
There is a paper that addresses the effect of a moving body on a motionless dust field that gives the correct relativistic answer to how much velocity is imparted to a stationary dust field by a relativistic flyby, and compares it to the Newtonian answer. See Olson, D.W.; Guarino, R. C. (1985). "Measuring the active gravitational mass of a moving object". When compared in this particular manner, the relativistic answer is that you get more of an induced velocity due to a relativistic flyby than you would by (incorrectly) using Newton's laws and plugging in the "relativisitc mass" in place of the Newtonian M.
But I don't know of anyone who has derived a similar result for gravitational lensing. I suspect that the result will be similar to Olson's for dust, but I can't guarantee it . You may have some difficulty tracking the paper down, though. You can find the abstract of the paper at
http://adsabs.harvard.edu/abs/1985AmJPh..53..661O, I'll attempt to quote the abstract (but it's better to read the full paper if you can get it).
The abstract is a bit over-general, it's important to realize that the "active gravitational mass" is a term not (AFAIK) in general use, specific to this paper, that computes what one might term the "average" field due to a relativistic flyby. This is something that is explained well in the paper but isn't clear at all from the abstract. The more commonly used notions of mass in GR are such masses as the Komar mass, the Bondi mass, and the ADM mass, which behave differently than the sort of "mass" the author describes in this paper.
Something else that is interesting to note and may not be mentioned in this paper, though it's mentioned in other papers. If you look at the peak field (and not the average field) of an ultra-relativistic flyby, you get the Aichelberg sexl solution, which is an impulsive plane-polarized gravity wave. In the ultra-relativistic limit, the velocity of a test particle changes by an impulse function, not in a continuous manner.