Rape isn't just wrong, but it is historically among the handful of the very worst types of crimes, sharing that unhappy position with murder, torture and treason. I take it that every instance of rape is very seriously wrong.
But now consider a spectrum of sexual acts: on one of the spectrum is a sexual act motivated by the man threatening the woman with death; on the other end of the spectrum is a sexual act motivated by the man threatening the woman that if she doesn't have sex with him, he won't take her out to the movie that they planned to go to. In the death case, the sex is non-consensual and hence rape. In the no-movie case, the sex is consensual and hence not rape: the threat is way too mild to invalidate consent.
Somewhere in that spectrum is a transition--be it vague or not--between non-consent and consent and hence between rape and non-rape. But every instance of rape is very seriously wrong. When we have continuously transitioned away from a very serious wrong, we shouldn't expect to immediately land in the territory of moral innocence. Rather, we should expect to land either in the territory of another wrong, either another very serious wrong or a "merely" serious wrong. If we start with an act of torture and continuously reduce the degree of pain, eventually we will get an act that isn't torture--but an act that falls somewhat short of the amount of pain needed for it to count as torture is still a serious battery.
Thus we should expect there to be sexual acts that are consensual, but seriously wrong because they are neighbors to rape. Moreover, we should expect that these acts will still wrong for reasons connected to their sexual nature, just as rape is very seriously wrong for reasons connected to its sexual nature. Consent, thus, is necessary but not sufficient for sexual integrity.
Here's a different way to put the argument. If one thinks that consent is the only condition needed for the permissibility of sex (with respect to sexual integrity--of course, there are other conditions, such as whether promises are broken, etc., but they aren't properly sexual), then one has to think either that (a) we have a transition from a very seriously wrong act to a completely innocent act in the above spectrum without any intermediate cases that are wrong but not seriously so, or (b) there are cases of rape that are non-seriously wrong. I think (a) is implausible and (b) is clearly false.
The spectrum I generated above was based on a spectrum of threats. But one can also generate a spectrum based on degrees of sobriety, degrees of understanding, clarity of expression (consent is a speech act), etc.
This has an important consequence particularly relevant to college judicial policies: If acts that aren't rape but are close to rape are seriously wrong, then in cases where it cannot be shown that a rape occurred, but it can be shown that either a rape occurred or a serious wrong close to rape occurred, it can still be just to levy serious punishment. Of course, this would require due process, and hence a way to operationalize the notion of such acts close to rape.
Note 1: None of my argument is meant to give aid or comfort to those who want to narrow the definition of rape. Rather the point is to widen the scope of wrong acts, for instance in the way that the "enthusiastic consent" movement does.
Note 2: The argument I am giving is not a sorites. Vagueness complicates the argument, but does not, I think, destroy it.